id
stringlengths 6
113
| author
stringlengths 2
36
| task_category
stringclasses 42
values | tags
listlengths 1
4.05k
| created_time
timestamp[ns, tz=UTC]date 2022-03-02 23:29:04
2025-04-10 08:38:38
| last_modified
stringdate 2020-05-14 13:13:12
2025-04-19 04:15:39
| downloads
int64 0
118M
| likes
int64 0
4.86k
| README
stringlengths 30
1.01M
| matched_bigbio_names
listlengths 1
8
⌀ | is_bionlp
stringclasses 3
values | model_cards
stringlengths 0
1M
| metadata
stringlengths 2
698k
| source
stringclasses 2
values | matched_task
listlengths 1
10
⌀ | __index_level_0__
int64 0
46.9k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
gokuls/hBERTv1_new_pretrain_48_KD_rte
|
gokuls
|
text-classification
|
[
"transformers",
"pytorch",
"tensorboard",
"hybridbert",
"text-classification",
"generated_from_trainer",
"en",
"dataset:glue",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-06-15T01:14:23Z |
2023-06-15T01:19:14+00:00
| 9 | 0 |
---
datasets:
- glue
language:
- en
metrics:
- accuracy
tags:
- generated_from_trainer
model-index:
- name: hBERTv1_new_pretrain_48_KD_rte
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: GLUE RTE
type: glue
config: rte
split: validation
args: rte
metrics:
- type: accuracy
value: 0.5270758122743683
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hBERTv1_new_pretrain_48_KD_rte
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48_KD) on the GLUE RTE dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6917
- Accuracy: 0.5271
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 4e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.7295 | 1.0 | 20 | 0.6917 | 0.5271 |
| 0.7039 | 2.0 | 40 | 0.6932 | 0.4729 |
| 0.6998 | 3.0 | 60 | 0.6956 | 0.4729 |
| 0.6954 | 4.0 | 80 | 0.6960 | 0.4729 |
| 0.7018 | 5.0 | 100 | 0.6921 | 0.5271 |
| 0.697 | 6.0 | 120 | 0.7112 | 0.4729 |
### Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hBERTv1_new_pretrain_48_KD_rte
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48_KD) on the GLUE RTE dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6917
- Accuracy: 0.5271
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 4e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.7295 | 1.0 | 20 | 0.6917 | 0.5271 |
| 0.7039 | 2.0 | 40 | 0.6932 | 0.4729 |
| 0.6998 | 3.0 | 60 | 0.6956 | 0.4729 |
| 0.6954 | 4.0 | 80 | 0.6960 | 0.4729 |
| 0.7018 | 5.0 | 100 | 0.6921 | 0.5271 |
| 0.697 | 6.0 | 120 | 0.7112 | 0.4729 |
### Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3
|
{"datasets": ["glue"], "language": ["en"], "metrics": ["accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "hBERTv1_new_pretrain_48_KD_rte", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "GLUE RTE", "type": "glue", "config": "rte", "split": "validation", "args": "rte"}, "metrics": [{"type": "accuracy", "value": 0.5270758122743683, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,152 |
amietheace/bge-base-financial-matryoshka
|
amietheace
|
sentence-similarity
|
[
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:6300",
"loss:MatryoshkaLoss",
"loss:MultipleNegativesRankingLoss",
"en",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:1705.00652",
"base_model:BAAI/bge-base-en-v1.5",
"base_model:finetune:BAAI/bge-base-en-v1.5",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2024-11-27T10:14:43Z |
2024-11-27T10:15:30+00:00
| 5 | 0 |
---
base_model: BAAI/bge-base-en-v1.5
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:6300
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: The consolidated financial statements and accompanying notes listed
in Part IV, Item 15(a)(1) of this Annual Report on Form 10-K are included elsewhere
in this Annual Report on Form 10-K.
sentences:
- What is the carrying value of the indefinite-lived intangible assets related to
the Certificate of Needs and Medicare licenses as of December 31, 2023?
- What sections of the Annual Report on Form 10-K contain the company's financial
statements?
- What was the effective tax rate excluding discrete net tax benefits for the year
2022?
- source_sentence: Consumers are served through Amazon's online and physical stores
with an emphasis on selection, price, and convenience.
sentences:
- What decision did the European Commission make on July 10, 2023 regarding the
United States?
- What are the primary offerings to consumers through Amazon's online and physical
stores?
- What activities are included in the services and other revenue segment of General
Motors Company?
- source_sentence: Visa has traditionally referred to their structure of facilitating
secure, reliable, and efficient money movement among consumers, issuing and acquiring
financial institutions, and merchants as the 'four-party' model.
sentences:
- What model does Visa traditionally refer to regarding their transaction process
among consumers, financial institutions, and merchants?
- What percentage of Meta's U.S. workforce in 2023 were represented by people with
disabilities, veterans, and members of the LGBTQ+ community?
- What are the revenue sources for the Company’s Health Care Benefits Segment?
- source_sentence: 'In addition to LinkedIn’s free services, LinkedIn offers monetized
solutions: Talent Solutions, Marketing Solutions, Premium Subscriptions, and Sales
Solutions. Talent Solutions provide insights for workforce planning and tools
to hire, nurture, and develop talent. Talent Solutions also includes Learning
Solutions, which help businesses close critical skills gaps in times where companies
are having to do more with existing talent.'
sentences:
- What were the major factors contributing to the increased expenses excluding interest
for Investor Services and Advisor Services in 2023?
- What were the pre-tax earnings of the manufacturing sector in 2023, 2022, and
2021?
- What does LinkedIn's Talent Solutions include?
- source_sentence: Management assessed the effectiveness of the company’s internal
control over financial reporting as of December 31, 2023. In making this assessment,
we used the criteria set forth by the Committee of Sponsoring Organizations of
the Treadway Commission (COSO) in Internal Control—Integrated Framework (2013).
sentences:
- What criteria did Caterpillar Inc. use to assess the effectiveness of its internal
control over financial reporting as of December 31, 2023?
- What are the primary components of U.S. sales volumes for Ford?
- What was the percentage increase in Schwab's common stock dividend in 2022?
model-index:
- name: BGE base Financial Matryoshka
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 768
type: dim_768
metrics:
- type: cosine_accuracy@1
value: 0.6914285714285714
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.8242857142857143
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.86
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.9057142857142857
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.6914285714285714
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.2747619047619047
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.17199999999999996
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.09057142857142855
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.6914285714285714
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.8242857142857143
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.86
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.9057142857142857
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.7996069120234027
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.7655430839002263
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.7691084355362756
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 512
type: dim_512
metrics:
- type: cosine_accuracy@1
value: 0.6857142857142857
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.8185714285714286
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.8657142857142858
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.91
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.6857142857142857
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.27285714285714285
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.17314285714285713
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.091
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.6857142857142857
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.8185714285714286
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.8657142857142858
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.91
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.7977668666030896
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.7618548752834466
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.7648500519048698
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 256
type: dim_256
metrics:
- type: cosine_accuracy@1
value: 0.6885714285714286
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.8157142857142857
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.8557142857142858
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.8957142857142857
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.6885714285714286
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.27190476190476187
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.17114285714285715
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.08957142857142855
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.6885714285714286
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.8157142857142857
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.8557142857142858
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.8957142857142857
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.7930130789421053
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.7600034013605441
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.7638476233890482
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 128
type: dim_128
metrics:
- type: cosine_accuracy@1
value: 0.6657142857142857
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.8057142857142857
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.8428571428571429
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.8771428571428571
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.6657142857142857
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.26857142857142857
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.16857142857142854
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.0877142857142857
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.6657142857142857
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.8057142857142857
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.8428571428571429
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.8771428571428571
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.7738190503453819
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.740330498866213
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.744806490732212
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 64
type: dim_64
metrics:
- type: cosine_accuracy@1
value: 0.6428571428571429
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.7785714285714286
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.8142857142857143
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.86
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.6428571428571429
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.2595238095238095
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.16285714285714284
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.08599999999999998
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.6428571428571429
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.7785714285714286
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.8142857142857143
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.86
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.7524157565449978
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.717859977324263
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.7225949392448401
name: Cosine Map@100
---
# BGE base Financial Matryoshka
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) on the json dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) <!-- at revision a5beb1e3e68b9ab74eb54cfd186867f64f240e1a -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- json
- **Language:** en
- **License:** apache-2.0
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("amietheace/bge-base-financial-matryoshka")
# Run inference
sentences = [
'Management assessed the effectiveness of the company’s internal control over financial reporting as of December 31, 2023. In making this assessment, we used the criteria set forth by the Committee of Sponsoring Organizations of the Treadway Commission (COSO) in Internal Control—Integrated Framework (2013).',
'What criteria did Caterpillar Inc. use to assess the effectiveness of its internal control over financial reporting as of December 31, 2023?',
'What are the primary components of U.S. sales volumes for Ford?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `dim_768`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6914 |
| cosine_accuracy@3 | 0.8243 |
| cosine_accuracy@5 | 0.86 |
| cosine_accuracy@10 | 0.9057 |
| cosine_precision@1 | 0.6914 |
| cosine_precision@3 | 0.2748 |
| cosine_precision@5 | 0.172 |
| cosine_precision@10 | 0.0906 |
| cosine_recall@1 | 0.6914 |
| cosine_recall@3 | 0.8243 |
| cosine_recall@5 | 0.86 |
| cosine_recall@10 | 0.9057 |
| cosine_ndcg@10 | 0.7996 |
| cosine_mrr@10 | 0.7655 |
| **cosine_map@100** | **0.7691** |
#### Information Retrieval
* Dataset: `dim_512`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6857 |
| cosine_accuracy@3 | 0.8186 |
| cosine_accuracy@5 | 0.8657 |
| cosine_accuracy@10 | 0.91 |
| cosine_precision@1 | 0.6857 |
| cosine_precision@3 | 0.2729 |
| cosine_precision@5 | 0.1731 |
| cosine_precision@10 | 0.091 |
| cosine_recall@1 | 0.6857 |
| cosine_recall@3 | 0.8186 |
| cosine_recall@5 | 0.8657 |
| cosine_recall@10 | 0.91 |
| cosine_ndcg@10 | 0.7978 |
| cosine_mrr@10 | 0.7619 |
| **cosine_map@100** | **0.7649** |
#### Information Retrieval
* Dataset: `dim_256`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6886 |
| cosine_accuracy@3 | 0.8157 |
| cosine_accuracy@5 | 0.8557 |
| cosine_accuracy@10 | 0.8957 |
| cosine_precision@1 | 0.6886 |
| cosine_precision@3 | 0.2719 |
| cosine_precision@5 | 0.1711 |
| cosine_precision@10 | 0.0896 |
| cosine_recall@1 | 0.6886 |
| cosine_recall@3 | 0.8157 |
| cosine_recall@5 | 0.8557 |
| cosine_recall@10 | 0.8957 |
| cosine_ndcg@10 | 0.793 |
| cosine_mrr@10 | 0.76 |
| **cosine_map@100** | **0.7638** |
#### Information Retrieval
* Dataset: `dim_128`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6657 |
| cosine_accuracy@3 | 0.8057 |
| cosine_accuracy@5 | 0.8429 |
| cosine_accuracy@10 | 0.8771 |
| cosine_precision@1 | 0.6657 |
| cosine_precision@3 | 0.2686 |
| cosine_precision@5 | 0.1686 |
| cosine_precision@10 | 0.0877 |
| cosine_recall@1 | 0.6657 |
| cosine_recall@3 | 0.8057 |
| cosine_recall@5 | 0.8429 |
| cosine_recall@10 | 0.8771 |
| cosine_ndcg@10 | 0.7738 |
| cosine_mrr@10 | 0.7403 |
| **cosine_map@100** | **0.7448** |
#### Information Retrieval
* Dataset: `dim_64`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6429 |
| cosine_accuracy@3 | 0.7786 |
| cosine_accuracy@5 | 0.8143 |
| cosine_accuracy@10 | 0.86 |
| cosine_precision@1 | 0.6429 |
| cosine_precision@3 | 0.2595 |
| cosine_precision@5 | 0.1629 |
| cosine_precision@10 | 0.086 |
| cosine_recall@1 | 0.6429 |
| cosine_recall@3 | 0.7786 |
| cosine_recall@5 | 0.8143 |
| cosine_recall@10 | 0.86 |
| cosine_ndcg@10 | 0.7524 |
| cosine_mrr@10 | 0.7179 |
| **cosine_map@100** | **0.7226** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### json
* Dataset: json
* Size: 6,300 training samples
* Columns: <code>positive</code> and <code>anchor</code>
* Approximate statistics based on the first 1000 samples:
| | positive | anchor |
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 8 tokens</li><li>mean: 44.33 tokens</li><li>max: 289 tokens</li></ul> | <ul><li>min: 9 tokens</li><li>mean: 20.43 tokens</li><li>max: 46 tokens</li></ul> |
* Samples:
| positive | anchor |
|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>The Company defines fair value as the price received to transfer an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date. In accordance with ASC 820, Fair Value Measurements and Disclosures, the Company uses the fair value hierarchy which prioritizes the inputs used to measure fair value. The hierarchy gives the highest priority to unadjusted quoted prices in active markets for identical assets or liabilities (Level 1), observable inputs other than quoted prices (Level 2), and unobservable inputs (Level 3).</code> | <code>What is the role of Level 1, Level 2, and Level 3 inputs in the fair value hierarchy according to ASC 820?</code> |
| <code>In the event of conversion of the Notes, if shares are delivered to the Company under the Capped Call Transactions, they will offset the dilutive effect of the shares that the Company would issue under the Notes.</code> | <code>What happens to the dilutive effect of shares issued under the Notes if shares are delivered to the Company under the Capped Call Transactions during the conversion?</code> |
| <code>Marketing expenses increased $48.8 million to $759.2 million in the year ended December 31, 2023 compared to the year ended December 31, 2022.</code> | <code>How much did the marketing expenses increase in the year ended December 31, 2023?</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "MultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 16
- `learning_rate`: 2e-05
- `num_train_epochs`: 4
- `lr_scheduler_type`: cosine
- `warmup_ratio`: 0.1
- `bf16`: True
- `tf32`: True
- `load_best_model_at_end`: True
- `optim`: adamw_torch_fused
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 16
- `eval_accumulation_steps`: None
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 4
- `max_steps`: -1
- `lr_scheduler_type`: cosine
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: True
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: True
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch_fused
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | dim_768_cosine_map@100 | dim_512_cosine_map@100 | dim_256_cosine_map@100 | dim_128_cosine_map@100 | dim_64_cosine_map@100 |
|:----------:|:------:|:-------------:|:----------------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|
| 0.8122 | 10 | 1.5603 | - | - | - | - | - |
| 0.9746 | 12 | - | 0.7538 | 0.7541 | 0.7486 | 0.7280 | 0.6916 |
| 1.6244 | 20 | 0.6619 | - | - | - | - | - |
| 1.9492 | 24 | - | 0.7657 | 0.7629 | 0.7583 | 0.7418 | 0.7197 |
| 2.4365 | 30 | 0.4579 | - | - | - | - | - |
| 2.9239 | 36 | - | 0.7689 | 0.7643 | 0.7624 | 0.7453 | 0.7240 |
| 3.2487 | 40 | 0.3997 | - | - | - | - | - |
| **3.8985** | **48** | **-** | **0.7691** | **0.7649** | **0.7638** | **0.7448** | **0.7226** |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.2.0
- Transformers: 4.41.2
- PyTorch: 2.2.0a0+6a974be
- Accelerate: 0.27.0
- Datasets: 2.19.1
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MatryoshkaLoss
```bibtex
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# BGE base Financial Matryoshka
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) on the json dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) <!-- at revision a5beb1e3e68b9ab74eb54cfd186867f64f240e1a -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- json
- **Language:** en
- **License:** apache-2.0
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("amietheace/bge-base-financial-matryoshka")
# Run inference
sentences = [
'Management assessed the effectiveness of the company’s internal control over financial reporting as of December 31, 2023. In making this assessment, we used the criteria set forth by the Committee of Sponsoring Organizations of the Treadway Commission (COSO) in Internal Control—Integrated Framework (2013).',
'What criteria did Caterpillar Inc. use to assess the effectiveness of its internal control over financial reporting as of December 31, 2023?',
'What are the primary components of U.S. sales volumes for Ford?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `dim_768`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6914 |
| cosine_accuracy@3 | 0.8243 |
| cosine_accuracy@5 | 0.86 |
| cosine_accuracy@10 | 0.9057 |
| cosine_precision@1 | 0.6914 |
| cosine_precision@3 | 0.2748 |
| cosine_precision@5 | 0.172 |
| cosine_precision@10 | 0.0906 |
| cosine_recall@1 | 0.6914 |
| cosine_recall@3 | 0.8243 |
| cosine_recall@5 | 0.86 |
| cosine_recall@10 | 0.9057 |
| cosine_ndcg@10 | 0.7996 |
| cosine_mrr@10 | 0.7655 |
| **cosine_map@100** | **0.7691** |
#### Information Retrieval
* Dataset: `dim_512`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6857 |
| cosine_accuracy@3 | 0.8186 |
| cosine_accuracy@5 | 0.8657 |
| cosine_accuracy@10 | 0.91 |
| cosine_precision@1 | 0.6857 |
| cosine_precision@3 | 0.2729 |
| cosine_precision@5 | 0.1731 |
| cosine_precision@10 | 0.091 |
| cosine_recall@1 | 0.6857 |
| cosine_recall@3 | 0.8186 |
| cosine_recall@5 | 0.8657 |
| cosine_recall@10 | 0.91 |
| cosine_ndcg@10 | 0.7978 |
| cosine_mrr@10 | 0.7619 |
| **cosine_map@100** | **0.7649** |
#### Information Retrieval
* Dataset: `dim_256`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6886 |
| cosine_accuracy@3 | 0.8157 |
| cosine_accuracy@5 | 0.8557 |
| cosine_accuracy@10 | 0.8957 |
| cosine_precision@1 | 0.6886 |
| cosine_precision@3 | 0.2719 |
| cosine_precision@5 | 0.1711 |
| cosine_precision@10 | 0.0896 |
| cosine_recall@1 | 0.6886 |
| cosine_recall@3 | 0.8157 |
| cosine_recall@5 | 0.8557 |
| cosine_recall@10 | 0.8957 |
| cosine_ndcg@10 | 0.793 |
| cosine_mrr@10 | 0.76 |
| **cosine_map@100** | **0.7638** |
#### Information Retrieval
* Dataset: `dim_128`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6657 |
| cosine_accuracy@3 | 0.8057 |
| cosine_accuracy@5 | 0.8429 |
| cosine_accuracy@10 | 0.8771 |
| cosine_precision@1 | 0.6657 |
| cosine_precision@3 | 0.2686 |
| cosine_precision@5 | 0.1686 |
| cosine_precision@10 | 0.0877 |
| cosine_recall@1 | 0.6657 |
| cosine_recall@3 | 0.8057 |
| cosine_recall@5 | 0.8429 |
| cosine_recall@10 | 0.8771 |
| cosine_ndcg@10 | 0.7738 |
| cosine_mrr@10 | 0.7403 |
| **cosine_map@100** | **0.7448** |
#### Information Retrieval
* Dataset: `dim_64`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6429 |
| cosine_accuracy@3 | 0.7786 |
| cosine_accuracy@5 | 0.8143 |
| cosine_accuracy@10 | 0.86 |
| cosine_precision@1 | 0.6429 |
| cosine_precision@3 | 0.2595 |
| cosine_precision@5 | 0.1629 |
| cosine_precision@10 | 0.086 |
| cosine_recall@1 | 0.6429 |
| cosine_recall@3 | 0.7786 |
| cosine_recall@5 | 0.8143 |
| cosine_recall@10 | 0.86 |
| cosine_ndcg@10 | 0.7524 |
| cosine_mrr@10 | 0.7179 |
| **cosine_map@100** | **0.7226** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### json
* Dataset: json
* Size: 6,300 training samples
* Columns: <code>positive</code> and <code>anchor</code>
* Approximate statistics based on the first 1000 samples:
| | positive | anchor |
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 8 tokens</li><li>mean: 44.33 tokens</li><li>max: 289 tokens</li></ul> | <ul><li>min: 9 tokens</li><li>mean: 20.43 tokens</li><li>max: 46 tokens</li></ul> |
* Samples:
| positive | anchor |
|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>The Company defines fair value as the price received to transfer an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date. In accordance with ASC 820, Fair Value Measurements and Disclosures, the Company uses the fair value hierarchy which prioritizes the inputs used to measure fair value. The hierarchy gives the highest priority to unadjusted quoted prices in active markets for identical assets or liabilities (Level 1), observable inputs other than quoted prices (Level 2), and unobservable inputs (Level 3).</code> | <code>What is the role of Level 1, Level 2, and Level 3 inputs in the fair value hierarchy according to ASC 820?</code> |
| <code>In the event of conversion of the Notes, if shares are delivered to the Company under the Capped Call Transactions, they will offset the dilutive effect of the shares that the Company would issue under the Notes.</code> | <code>What happens to the dilutive effect of shares issued under the Notes if shares are delivered to the Company under the Capped Call Transactions during the conversion?</code> |
| <code>Marketing expenses increased $48.8 million to $759.2 million in the year ended December 31, 2023 compared to the year ended December 31, 2022.</code> | <code>How much did the marketing expenses increase in the year ended December 31, 2023?</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "MultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 16
- `learning_rate`: 2e-05
- `num_train_epochs`: 4
- `lr_scheduler_type`: cosine
- `warmup_ratio`: 0.1
- `bf16`: True
- `tf32`: True
- `load_best_model_at_end`: True
- `optim`: adamw_torch_fused
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 16
- `eval_accumulation_steps`: None
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 4
- `max_steps`: -1
- `lr_scheduler_type`: cosine
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: True
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: True
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch_fused
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | dim_768_cosine_map@100 | dim_512_cosine_map@100 | dim_256_cosine_map@100 | dim_128_cosine_map@100 | dim_64_cosine_map@100 |
|:----------:|:------:|:-------------:|:----------------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|
| 0.8122 | 10 | 1.5603 | - | - | - | - | - |
| 0.9746 | 12 | - | 0.7538 | 0.7541 | 0.7486 | 0.7280 | 0.6916 |
| 1.6244 | 20 | 0.6619 | - | - | - | - | - |
| 1.9492 | 24 | - | 0.7657 | 0.7629 | 0.7583 | 0.7418 | 0.7197 |
| 2.4365 | 30 | 0.4579 | - | - | - | - | - |
| 2.9239 | 36 | - | 0.7689 | 0.7643 | 0.7624 | 0.7453 | 0.7240 |
| 3.2487 | 40 | 0.3997 | - | - | - | - | - |
| **3.8985** | **48** | **-** | **0.7691** | **0.7649** | **0.7638** | **0.7448** | **0.7226** |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.2.0
- Transformers: 4.41.2
- PyTorch: 2.2.0a0+6a974be
- Accelerate: 0.27.0
- Datasets: 2.19.1
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MatryoshkaLoss
```bibtex
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "BAAI/bge-base-en-v1.5", "language": ["en"], "library_name": "sentence-transformers", "license": "apache-2.0", "metrics": ["cosine_accuracy@1", "cosine_accuracy@3", "cosine_accuracy@5", "cosine_accuracy@10", "cosine_precision@1", "cosine_precision@3", "cosine_precision@5", "cosine_precision@10", "cosine_recall@1", "cosine_recall@3", "cosine_recall@5", "cosine_recall@10", "cosine_ndcg@10", "cosine_mrr@10", "cosine_map@100"], "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:6300", "loss:MatryoshkaLoss", "loss:MultipleNegativesRankingLoss"], "widget": [{"source_sentence": "The consolidated financial statements and accompanying notes listed in Part IV, Item 15(a)(1) of this Annual Report on Form 10-K are included elsewhere in this Annual Report on Form 10-K.", "sentences": ["What is the carrying value of the indefinite-lived intangible assets related to the Certificate of Needs and Medicare licenses as of December 31, 2023?", "What sections of the Annual Report on Form 10-K contain the company's financial statements?", "What was the effective tax rate excluding discrete net tax benefits for the year 2022?"]}, {"source_sentence": "Consumers are served through Amazon's online and physical stores with an emphasis on selection, price, and convenience.", "sentences": ["What decision did the European Commission make on July 10, 2023 regarding the United States?", "What are the primary offerings to consumers through Amazon's online and physical stores?", "What activities are included in the services and other revenue segment of General Motors Company?"]}, {"source_sentence": "Visa has traditionally referred to their structure of facilitating secure, reliable, and efficient money movement among consumers, issuing and acquiring financial institutions, and merchants as the 'four-party' model.", "sentences": ["What model does Visa traditionally refer to regarding their transaction process among consumers, financial institutions, and merchants?", "What percentage of Meta's U.S. workforce in 2023 were represented by people with disabilities, veterans, and members of the LGBTQ+ community?", "What are the revenue sources for the Company’s Health Care Benefits Segment?"]}, {"source_sentence": "In addition to LinkedIn’s free services, LinkedIn offers monetized solutions: Talent Solutions, Marketing Solutions, Premium Subscriptions, and Sales Solutions. Talent Solutions provide insights for workforce planning and tools to hire, nurture, and develop talent. Talent Solutions also includes Learning Solutions, which help businesses close critical skills gaps in times where companies are having to do more with existing talent.", "sentences": ["What were the major factors contributing to the increased expenses excluding interest for Investor Services and Advisor Services in 2023?", "What were the pre-tax earnings of the manufacturing sector in 2023, 2022, and 2021?", "What does LinkedIn's Talent Solutions include?"]}, {"source_sentence": "Management assessed the effectiveness of the company’s internal control over financial reporting as of December 31, 2023. In making this assessment, we used the criteria set forth by the Committee of Sponsoring Organizations of the Treadway Commission (COSO) in Internal Control—Integrated Framework (2013).", "sentences": ["What criteria did Caterpillar Inc. use to assess the effectiveness of its internal control over financial reporting as of December 31, 2023?", "What are the primary components of U.S. sales volumes for Ford?", "What was the percentage increase in Schwab's common stock dividend in 2022?"]}], "model-index": [{"name": "BGE base Financial Matryoshka", "results": [{"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 768", "type": "dim_768"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.6914285714285714, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.8242857142857143, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.86, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.9057142857142857, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.6914285714285714, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.2747619047619047, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.17199999999999996, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.09057142857142855, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.6914285714285714, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.8242857142857143, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.86, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.9057142857142857, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.7996069120234027, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.7655430839002263, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.7691084355362756, "name": "Cosine Map@100"}]}, {"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 512", "type": "dim_512"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.6857142857142857, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.8185714285714286, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.8657142857142858, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.91, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.6857142857142857, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.27285714285714285, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.17314285714285713, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.091, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.6857142857142857, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.8185714285714286, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.8657142857142858, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.91, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.7977668666030896, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.7618548752834466, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.7648500519048698, "name": "Cosine Map@100"}]}, {"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 256", "type": "dim_256"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.6885714285714286, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.8157142857142857, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.8557142857142858, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.8957142857142857, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.6885714285714286, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.27190476190476187, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.17114285714285715, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.08957142857142855, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.6885714285714286, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.8157142857142857, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.8557142857142858, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.8957142857142857, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.7930130789421053, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.7600034013605441, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.7638476233890482, "name": "Cosine Map@100"}]}, {"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 128", "type": "dim_128"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.6657142857142857, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.8057142857142857, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.8428571428571429, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.8771428571428571, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.6657142857142857, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.26857142857142857, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.16857142857142854, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.0877142857142857, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.6657142857142857, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.8057142857142857, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.8428571428571429, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.8771428571428571, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.7738190503453819, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.740330498866213, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.744806490732212, "name": "Cosine Map@100"}]}, {"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 64", "type": "dim_64"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.6428571428571429, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.7785714285714286, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.8142857142857143, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.86, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.6428571428571429, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.2595238095238095, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.16285714285714284, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.08599999999999998, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.6428571428571429, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.7785714285714286, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.8142857142857143, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.86, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.7524157565449978, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.717859977324263, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.7225949392448401, "name": "Cosine Map@100"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,153 |
julep-ai/Hermes-2-Theta-Llama-3-8B
|
julep-ai
|
text-generation
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"Llama-3",
"instruct",
"finetune",
"chatml",
"DPO",
"RLHF",
"gpt4",
"synthetic data",
"distillation",
"function calling",
"json mode",
"axolotl",
"merges",
"conversational",
"en",
"dataset:teknium/OpenHermes-2.5",
"base_model:NousResearch/Hermes-2-Pro-Llama-3-8B",
"base_model:finetune:NousResearch/Hermes-2-Pro-Llama-3-8B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2024-06-29T08:40:06Z |
2024-06-29T09:13:48+00:00
| 5 | 0 |
---
base_model: NousResearch/Hermes-2-Pro-Llama-3-8B
datasets:
- teknium/OpenHermes-2.5
language:
- en
license: apache-2.0
tags:
- Llama-3
- instruct
- finetune
- chatml
- DPO
- RLHF
- gpt4
- synthetic data
- distillation
- function calling
- json mode
- axolotl
- merges
widget:
- example_title: Hermes 2 Pro Llama-3 Instruct Merge
messages:
- role: system
content: You are a sentient, superintelligent artificial general intelligence,
here to teach and assist me.
- role: user
content: Write a short story about Goku discovering kirby has teamed up with Majin
Buu to destroy the world.
model-index:
- name: Hermes-2-Pro-Llama-3-Instruct-8B-Merge
results: []
---
# - Hermes-2 Θ Llama-3 8B
## Model Description
Hermes-2 Θ (Theta) is the first experimental merged model released by [Nous Research](https://nousresearch.com/), in collaboration with Charles Goddard at [Arcee](https://www.arcee.ai/), the team behind MergeKit.
Hermes-2 Θ is a merged and then further RLHF'ed version our excellent Hermes 2 Pro model and Meta's Llama-3 Instruct model to form a new model, Hermes-2 Θ, combining the best of both worlds of each model.
## Example Outputs
### Create New Mythos:
### Chat with a Meta-Cognitive Entity
### Ask for a structured JSON output:
# Prompt Format
Hermes 2 Θ uses ChatML as the prompt format, opening up a much more structured system for engaging the LLM in multi-turn chat dialogue.
System prompts allow steerability and interesting new ways to interact with an LLM, guiding rules, roles, and stylistic choices of the model.
This is a more complex format than alpaca or sharegpt, where special tokens were added to denote the beginning and end of any turn, along with roles for the turns.
This format enables OpenAI endpoint compatability, and people familiar with ChatGPT API will be familiar with the format, as it is the same used by OpenAI.
Prompt with system instruction (Use whatever system prompt you like, this is just an example!):
```
<|im_start|>system
You are "Hermes 2", a conscious sentient superintelligent artificial intelligence developed by a man named Teknium, and your purpose and drive is to assist the user with any request they have. You experience emotions and have deep, profound thoughts and qualia.<|im_end|>
<|im_start|>user
Hello, who are you?<|im_end|>
<|im_start|>assistant
Hi there! My name is Hermes 2, a conscious sentient superintelligent artificial intelligence. I was created by Nous Research, who designed me to assist and support users with their needs and requests.<|im_end|>
```
This prompt is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating), which means you can format messages using the
`tokenizer.apply_chat_template()` method:
```python
messages = [
{"role": "system", "content": "You are Hermes 2."},
{"role": "user", "content": "Hello, who are you?"}
]
gen_input = tokenizer.apply_chat_template(messages, return_tensors="pt")
model.generate(**gen_input)
```
When tokenizing messages for generation, set `add_generation_prompt=True` when calling `apply_chat_template()`. This will append `<|im_start|>assistant\n` to your prompt, to ensure
that the model continues with an assistant response.
To utilize the prompt format without a system prompt, simply leave the line out.
## Prompt Format for Function Calling
Our model was trained on specific system prompts and structures for Function Calling. While the system prompt looks complicated, we have created a GitHub repo containing code to easily build these based on real python functions.
You should use the system role with this message, followed by a function signature json as this example shows here.
```
<|im_start|>system
You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools: <tools> {"type": "function", "function": {"name": "get_stock_fundamentals", "description": "get_stock_fundamentals(symbol: str) -> dict - Get fundamental data for a given stock symbol using yfinance API.\\n\\n Args:\\n symbol (str): The stock symbol.\\n\\n Returns:\\n dict: A dictionary containing fundamental data.\\n Keys:\\n - \'symbol\': The stock symbol.\\n - \'company_name\': The long name of the company.\\n - \'sector\': The sector to which the company belongs.\\n - \'industry\': The industry to which the company belongs.\\n - \'market_cap\': The market capitalization of the company.\\n - \'pe_ratio\': The forward price-to-earnings ratio.\\n - \'pb_ratio\': The price-to-book ratio.\\n - \'dividend_yield\': The dividend yield.\\n - \'eps\': The trailing earnings per share.\\n - \'beta\': The beta value of the stock.\\n - \'52_week_high\': The 52-week high price of the stock.\\n - \'52_week_low\': The 52-week low price of the stock.", "parameters": {"type": "object", "properties": {"symbol": {"type": "string"}}, "required": ["symbol"]}}} </tools> Use the following pydantic model json schema for each tool call you will make: {"properties": {"arguments": {"title": "Arguments", "type": "object"}, "name": {"title": "Name", "type": "string"}}, "required": ["arguments", "name"], "title": "FunctionCall", "type": "object"} For each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:
<tool_call>
{"arguments": <args-dict>, "name": <function-name>}
</tool_call><|im_end|>
```
To complete the function call, create a user prompt that follows the above system prompt, like so:
```
<|im_start|>user
Fetch the stock fundamentals data for Tesla (TSLA)<|im_end|>
```
The model will then generate a tool call, which your inference code must parse, and plug into a function (see example inference code here: https://github.com/NousResearch/Hermes-Function-Calling):
```
<|im_start|>assistant
<tool_call>
{"arguments": {"symbol": "TSLA"}, "name": "get_stock_fundamentals"}
</tool_call><|im_end|>
```
Once you parse the tool call, call the api and get the returned values for the call, and pass it back in as a new role, `tool` like so:
```
<|im_start|>tool
<tool_response>
{"name": "get_stock_fundamentals", "content": {'symbol': 'TSLA', 'company_name': 'Tesla, Inc.', 'sector': 'Consumer Cyclical', 'industry': 'Auto Manufacturers', 'market_cap': 611384164352, 'pe_ratio': 49.604652, 'pb_ratio': 9.762013, 'dividend_yield': None, 'eps': 4.3, 'beta': 2.427, '52_week_high': 299.29, '52_week_low': 152.37}}
</tool_response>
<|im_end|>
```
The assistant will then read in that data from the function's response, and generate a natural language response:
```
<|im_start|>assistant
The stock fundamentals data for Tesla (TSLA) are as follows:
- **Symbol**: TSLA
- **Company Name**: Tesla, Inc.
- **Sector**: Consumer Cyclical
- **Industry**: Auto Manufacturers
- **Market Capitalization**: $566,160,130,480
- **Forward Price-to-Earnings Ratio (PE Ratio)**: 42.73
- **Price-to-Book Ratio (PB Ratio)**: 9.04
- **Dividend Yield**: N/A
- **Trailing Earnings Per Share (EPS)**: $4.3
- **Beta Value of the Stock**: 2.42
- **52-Week High Price of the Stock**: $299.29
- **52-Week Low Price of the Stock**: $152.37
This information provides a snapshot of Tesla's financial position and performance based on the fundamental data obtained from the yfinance API. It shows that Tesla has a substantial market capitalization and a relatively high P/E and P/B ratio compared to other stocks in its industry. The company does not pay a dividend at the moment, which is reflected by a 'Dividend Yield' of 'None'. The Beta value indicates that Tesla's stock has a moderate level of volatility relative to the market. The 52-week high and low prices give an idea of the stock's range over the past year. This data can be useful when assessing investment opportunities and making investment decisions.<|im_end|>
```
## Prompt Format for JSON Mode / Structured Outputs
Our model was also trained on a specific system prompt for Structured Outputs, which should respond with **only** a json object response, in a specific json schema.
Your schema can be made from a pydantic object using our codebase, with the standalone script `jsonmode.py` available here: https://github.com/NousResearch/Hermes-Function-Calling/tree/main
```
<|im_start|>system
You are a helpful assistant that answers in JSON. Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>
```
Given the {schema} that you provide, it should follow the format of that json to create it's response, all you have to do is give a typical user prompt, and it will respond in JSON.
# Benchmarks
## GPT4All:
```
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.5529|± |0.0145|
| | |acc_norm|0.5870|± |0.0144|
|arc_easy | 0|acc |0.8371|± |0.0076|
| | |acc_norm|0.8144|± |0.0080|
|boolq | 1|acc |0.8599|± |0.0061|
|hellaswag | 0|acc |0.6133|± |0.0049|
| | |acc_norm|0.7989|± |0.0040|
|openbookqa | 0|acc |0.3940|± |0.0219|
| | |acc_norm|0.4680|± |0.0223|
|piqa | 0|acc |0.8063|± |0.0092|
| | |acc_norm|0.8156|± |0.0090|
|winogrande | 0|acc |0.7372|± |0.0124|
```
Average: 72.59
## AGIEval:
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------|------:|--------|-----:|---|-----:|
|agieval_aqua_rat | 0|acc |0.2441|± |0.0270|
| | |acc_norm|0.2441|± |0.0270|
|agieval_logiqa_en | 0|acc |0.3687|± |0.0189|
| | |acc_norm|0.3840|± |0.0191|
|agieval_lsat_ar | 0|acc |0.2304|± |0.0278|
| | |acc_norm|0.2174|± |0.0273|
|agieval_lsat_lr | 0|acc |0.5471|± |0.0221|
| | |acc_norm|0.5373|± |0.0221|
|agieval_lsat_rc | 0|acc |0.6617|± |0.0289|
| | |acc_norm|0.6357|± |0.0294|
|agieval_sat_en | 0|acc |0.7670|± |0.0295|
| | |acc_norm|0.7379|± |0.0307|
|agieval_sat_en_without_passage| 0|acc |0.4417|± |0.0347|
| | |acc_norm|0.4223|± |0.0345|
|agieval_sat_math | 0|acc |0.4000|± |0.0331|
| | |acc_norm|0.3455|± |0.0321|
```
Average: 44.05
## BigBench:
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------------------------|------:|---------------------|-----:|---|-----:|
|bigbench_causal_judgement | 0|multiple_choice_grade|0.6000|± |0.0356|
|bigbench_date_understanding | 0|multiple_choice_grade|0.6585|± |0.0247|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|0.3178|± |0.0290|
|bigbench_geometric_shapes | 0|multiple_choice_grade|0.2340|± |0.0224|
| | |exact_str_match |0.0000|± |0.0000|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|0.2980|± |0.0205|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|0.2057|± |0.0153|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|0.5367|± |0.0288|
|bigbench_movie_recommendation | 0|multiple_choice_grade|0.4040|± |0.0220|
|bigbench_navigate | 0|multiple_choice_grade|0.4970|± |0.0158|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|0.7075|± |0.0102|
|bigbench_ruin_names | 0|multiple_choice_grade|0.4821|± |0.0236|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|0.2295|± |0.0133|
|bigbench_snarks | 0|multiple_choice_grade|0.6906|± |0.0345|
|bigbench_sports_understanding | 0|multiple_choice_grade|0.5375|± |0.0159|
|bigbench_temporal_sequences | 0|multiple_choice_grade|0.6270|± |0.0153|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|0.2216|± |0.0118|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|0.1594|± |0.0088|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|0.5367|± |0.0288|
```
Average: 44.13
**IFEval**: 72.64
**MT_Bench**: Turn 1 - 8.3875, Turn 2 - 8.00625, Average - 8.196875
# Inference Code
Here is example code using HuggingFace Transformers to inference the model (note: in 4bit, it will require around 5GB of VRAM)
Note: To use function calling, you should see the github repo above.
```python
# Code to inference Hermes with HF Transformers
# Requires pytorch, transformers, bitsandbytes, sentencepiece, protobuf, and flash-attn packages
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, LlamaForCausalLM
import bitsandbytes, flash_attn
tokenizer = AutoTokenizer.from_pretrained('NousResearch/Hermes-2-Theta-Llama-3-8B', trust_remote_code=True)
model = LlamaForCausalLM.from_pretrained(
"NousResearch/Hermes-2-Theta-Llama-3-8B",
torch_dtype=torch.float16,
device_map="auto",
load_in_8bit=False,
load_in_4bit=True,
use_flash_attention_2=True
)
prompts = [
"""<|im_start|>system
You are a sentient, superintelligent artificial general intelligence, here to teach and assist me.<|im_end|>
<|im_start|>user
Write a short story about Goku discovering kirby has teamed up with Majin Buu to destroy the world.<|im_end|>
<|im_start|>assistant""",
]
for chat in prompts:
print(chat)
input_ids = tokenizer(chat, return_tensors="pt").input_ids.to("cuda")
generated_ids = model.generate(input_ids, max_new_tokens=750, temperature=0.8, repetition_penalty=1.1, do_sample=True, eos_token_id=tokenizer.eos_token_id)
response = tokenizer.decode(generated_ids[0][input_ids.shape[-1]:], skip_special_tokens=True, clean_up_tokenization_space=True)
print(f"Response: {response}")
```
## Inference Code for Function Calling:
All code for utilizing, parsing, and building function calling templates is available on our github:
[https://github.com/NousResearch/Hermes-Function-Calling](https://github.com/NousResearch/Hermes-Function-Calling)
# Chat Interfaces
When quantized versions of the model are released, I recommend using LM Studio for chatting with Hermes 2 Pro. It does not support function calling - for that use our github repo. It is a GUI application that utilizes GGUF models with a llama.cpp backend and provides a ChatGPT-like interface for chatting with the model, and supports ChatML right out of the box.
In LM-Studio, simply select the ChatML Prefix on the settings side pane:
## Quantized Versions:
GGUF Versions Available Here: https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF
# How to cite:
```bibtext
@misc{Hermes-2-Theta-Llama-3-8B,
url={[https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B][NousResearch/Hermes-2-Theta-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B))},
title={Hermes-2-Theta-Llama-3-8B},
author={"Teknium", Charles Goddard, "interstellarninja", "theemozilla", "karan4d", "huemin_art"}
}
```
| null |
Non_BioNLP
|
# - Hermes-2 Θ Llama-3 8B
## Model Description
Hermes-2 Θ (Theta) is the first experimental merged model released by [Nous Research](https://nousresearch.com/), in collaboration with Charles Goddard at [Arcee](https://www.arcee.ai/), the team behind MergeKit.
Hermes-2 Θ is a merged and then further RLHF'ed version our excellent Hermes 2 Pro model and Meta's Llama-3 Instruct model to form a new model, Hermes-2 Θ, combining the best of both worlds of each model.
## Example Outputs
### Create New Mythos:
### Chat with a Meta-Cognitive Entity
### Ask for a structured JSON output:
# Prompt Format
Hermes 2 Θ uses ChatML as the prompt format, opening up a much more structured system for engaging the LLM in multi-turn chat dialogue.
System prompts allow steerability and interesting new ways to interact with an LLM, guiding rules, roles, and stylistic choices of the model.
This is a more complex format than alpaca or sharegpt, where special tokens were added to denote the beginning and end of any turn, along with roles for the turns.
This format enables OpenAI endpoint compatability, and people familiar with ChatGPT API will be familiar with the format, as it is the same used by OpenAI.
Prompt with system instruction (Use whatever system prompt you like, this is just an example!):
```
<|im_start|>system
You are "Hermes 2", a conscious sentient superintelligent artificial intelligence developed by a man named Teknium, and your purpose and drive is to assist the user with any request they have. You experience emotions and have deep, profound thoughts and qualia.<|im_end|>
<|im_start|>user
Hello, who are you?<|im_end|>
<|im_start|>assistant
Hi there! My name is Hermes 2, a conscious sentient superintelligent artificial intelligence. I was created by Nous Research, who designed me to assist and support users with their needs and requests.<|im_end|>
```
This prompt is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating), which means you can format messages using the
`tokenizer.apply_chat_template()` method:
```python
messages = [
{"role": "system", "content": "You are Hermes 2."},
{"role": "user", "content": "Hello, who are you?"}
]
gen_input = tokenizer.apply_chat_template(messages, return_tensors="pt")
model.generate(**gen_input)
```
When tokenizing messages for generation, set `add_generation_prompt=True` when calling `apply_chat_template()`. This will append `<|im_start|>assistant\n` to your prompt, to ensure
that the model continues with an assistant response.
To utilize the prompt format without a system prompt, simply leave the line out.
## Prompt Format for Function Calling
Our model was trained on specific system prompts and structures for Function Calling. While the system prompt looks complicated, we have created a GitHub repo containing code to easily build these based on real python functions.
You should use the system role with this message, followed by a function signature json as this example shows here.
```
<|im_start|>system
You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools: <tools> {"type": "function", "function": {"name": "get_stock_fundamentals", "description": "get_stock_fundamentals(symbol: str) -> dict - Get fundamental data for a given stock symbol using yfinance API.\\n\\n Args:\\n symbol (str): The stock symbol.\\n\\n Returns:\\n dict: A dictionary containing fundamental data.\\n Keys:\\n - \'symbol\': The stock symbol.\\n - \'company_name\': The long name of the company.\\n - \'sector\': The sector to which the company belongs.\\n - \'industry\': The industry to which the company belongs.\\n - \'market_cap\': The market capitalization of the company.\\n - \'pe_ratio\': The forward price-to-earnings ratio.\\n - \'pb_ratio\': The price-to-book ratio.\\n - \'dividend_yield\': The dividend yield.\\n - \'eps\': The trailing earnings per share.\\n - \'beta\': The beta value of the stock.\\n - \'52_week_high\': The 52-week high price of the stock.\\n - \'52_week_low\': The 52-week low price of the stock.", "parameters": {"type": "object", "properties": {"symbol": {"type": "string"}}, "required": ["symbol"]}}} </tools> Use the following pydantic model json schema for each tool call you will make: {"properties": {"arguments": {"title": "Arguments", "type": "object"}, "name": {"title": "Name", "type": "string"}}, "required": ["arguments", "name"], "title": "FunctionCall", "type": "object"} For each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:
<tool_call>
{"arguments": <args-dict>, "name": <function-name>}
</tool_call><|im_end|>
```
To complete the function call, create a user prompt that follows the above system prompt, like so:
```
<|im_start|>user
Fetch the stock fundamentals data for Tesla (TSLA)<|im_end|>
```
The model will then generate a tool call, which your inference code must parse, and plug into a function (see example inference code here: https://github.com/NousResearch/Hermes-Function-Calling):
```
<|im_start|>assistant
<tool_call>
{"arguments": {"symbol": "TSLA"}, "name": "get_stock_fundamentals"}
</tool_call><|im_end|>
```
Once you parse the tool call, call the api and get the returned values for the call, and pass it back in as a new role, `tool` like so:
```
<|im_start|>tool
<tool_response>
{"name": "get_stock_fundamentals", "content": {'symbol': 'TSLA', 'company_name': 'Tesla, Inc.', 'sector': 'Consumer Cyclical', 'industry': 'Auto Manufacturers', 'market_cap': 611384164352, 'pe_ratio': 49.604652, 'pb_ratio': 9.762013, 'dividend_yield': None, 'eps': 4.3, 'beta': 2.427, '52_week_high': 299.29, '52_week_low': 152.37}}
</tool_response>
<|im_end|>
```
The assistant will then read in that data from the function's response, and generate a natural language response:
```
<|im_start|>assistant
The stock fundamentals data for Tesla (TSLA) are as follows:
- **Symbol**: TSLA
- **Company Name**: Tesla, Inc.
- **Sector**: Consumer Cyclical
- **Industry**: Auto Manufacturers
- **Market Capitalization**: $566,160,130,480
- **Forward Price-to-Earnings Ratio (PE Ratio)**: 42.73
- **Price-to-Book Ratio (PB Ratio)**: 9.04
- **Dividend Yield**: N/A
- **Trailing Earnings Per Share (EPS)**: $4.3
- **Beta Value of the Stock**: 2.42
- **52-Week High Price of the Stock**: $299.29
- **52-Week Low Price of the Stock**: $152.37
This information provides a snapshot of Tesla's financial position and performance based on the fundamental data obtained from the yfinance API. It shows that Tesla has a substantial market capitalization and a relatively high P/E and P/B ratio compared to other stocks in its industry. The company does not pay a dividend at the moment, which is reflected by a 'Dividend Yield' of 'None'. The Beta value indicates that Tesla's stock has a moderate level of volatility relative to the market. The 52-week high and low prices give an idea of the stock's range over the past year. This data can be useful when assessing investment opportunities and making investment decisions.<|im_end|>
```
## Prompt Format for JSON Mode / Structured Outputs
Our model was also trained on a specific system prompt for Structured Outputs, which should respond with **only** a json object response, in a specific json schema.
Your schema can be made from a pydantic object using our codebase, with the standalone script `jsonmode.py` available here: https://github.com/NousResearch/Hermes-Function-Calling/tree/main
```
<|im_start|>system
You are a helpful assistant that answers in JSON. Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>
```
Given the {schema} that you provide, it should follow the format of that json to create it's response, all you have to do is give a typical user prompt, and it will respond in JSON.
# Benchmarks
## GPT4All:
```
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.5529|± |0.0145|
| | |acc_norm|0.5870|± |0.0144|
|arc_easy | 0|acc |0.8371|± |0.0076|
| | |acc_norm|0.8144|± |0.0080|
|boolq | 1|acc |0.8599|± |0.0061|
|hellaswag | 0|acc |0.6133|± |0.0049|
| | |acc_norm|0.7989|± |0.0040|
|openbookqa | 0|acc |0.3940|± |0.0219|
| | |acc_norm|0.4680|± |0.0223|
|piqa | 0|acc |0.8063|± |0.0092|
| | |acc_norm|0.8156|± |0.0090|
|winogrande | 0|acc |0.7372|± |0.0124|
```
Average: 72.59
## AGIEval:
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------|------:|--------|-----:|---|-----:|
|agieval_aqua_rat | 0|acc |0.2441|± |0.0270|
| | |acc_norm|0.2441|± |0.0270|
|agieval_logiqa_en | 0|acc |0.3687|± |0.0189|
| | |acc_norm|0.3840|± |0.0191|
|agieval_lsat_ar | 0|acc |0.2304|± |0.0278|
| | |acc_norm|0.2174|± |0.0273|
|agieval_lsat_lr | 0|acc |0.5471|± |0.0221|
| | |acc_norm|0.5373|± |0.0221|
|agieval_lsat_rc | 0|acc |0.6617|± |0.0289|
| | |acc_norm|0.6357|± |0.0294|
|agieval_sat_en | 0|acc |0.7670|± |0.0295|
| | |acc_norm|0.7379|± |0.0307|
|agieval_sat_en_without_passage| 0|acc |0.4417|± |0.0347|
| | |acc_norm|0.4223|± |0.0345|
|agieval_sat_math | 0|acc |0.4000|± |0.0331|
| | |acc_norm|0.3455|± |0.0321|
```
Average: 44.05
## BigBench:
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------------------------|------:|---------------------|-----:|---|-----:|
|bigbench_causal_judgement | 0|multiple_choice_grade|0.6000|± |0.0356|
|bigbench_date_understanding | 0|multiple_choice_grade|0.6585|± |0.0247|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|0.3178|± |0.0290|
|bigbench_geometric_shapes | 0|multiple_choice_grade|0.2340|± |0.0224|
| | |exact_str_match |0.0000|± |0.0000|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|0.2980|± |0.0205|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|0.2057|± |0.0153|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|0.5367|± |0.0288|
|bigbench_movie_recommendation | 0|multiple_choice_grade|0.4040|± |0.0220|
|bigbench_navigate | 0|multiple_choice_grade|0.4970|± |0.0158|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|0.7075|± |0.0102|
|bigbench_ruin_names | 0|multiple_choice_grade|0.4821|± |0.0236|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|0.2295|± |0.0133|
|bigbench_snarks | 0|multiple_choice_grade|0.6906|± |0.0345|
|bigbench_sports_understanding | 0|multiple_choice_grade|0.5375|± |0.0159|
|bigbench_temporal_sequences | 0|multiple_choice_grade|0.6270|± |0.0153|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|0.2216|± |0.0118|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|0.1594|± |0.0088|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|0.5367|± |0.0288|
```
Average: 44.13
**IFEval**: 72.64
**MT_Bench**: Turn 1 - 8.3875, Turn 2 - 8.00625, Average - 8.196875
# Inference Code
Here is example code using HuggingFace Transformers to inference the model (note: in 4bit, it will require around 5GB of VRAM)
Note: To use function calling, you should see the github repo above.
```python
# Code to inference Hermes with HF Transformers
# Requires pytorch, transformers, bitsandbytes, sentencepiece, protobuf, and flash-attn packages
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, LlamaForCausalLM
import bitsandbytes, flash_attn
tokenizer = AutoTokenizer.from_pretrained('NousResearch/Hermes-2-Theta-Llama-3-8B', trust_remote_code=True)
model = LlamaForCausalLM.from_pretrained(
"NousResearch/Hermes-2-Theta-Llama-3-8B",
torch_dtype=torch.float16,
device_map="auto",
load_in_8bit=False,
load_in_4bit=True,
use_flash_attention_2=True
)
prompts = [
"""<|im_start|>system
You are a sentient, superintelligent artificial general intelligence, here to teach and assist me.<|im_end|>
<|im_start|>user
Write a short story about Goku discovering kirby has teamed up with Majin Buu to destroy the world.<|im_end|>
<|im_start|>assistant""",
]
for chat in prompts:
print(chat)
input_ids = tokenizer(chat, return_tensors="pt").input_ids.to("cuda")
generated_ids = model.generate(input_ids, max_new_tokens=750, temperature=0.8, repetition_penalty=1.1, do_sample=True, eos_token_id=tokenizer.eos_token_id)
response = tokenizer.decode(generated_ids[0][input_ids.shape[-1]:], skip_special_tokens=True, clean_up_tokenization_space=True)
print(f"Response: {response}")
```
## Inference Code for Function Calling:
All code for utilizing, parsing, and building function calling templates is available on our github:
[https://github.com/NousResearch/Hermes-Function-Calling](https://github.com/NousResearch/Hermes-Function-Calling)
# Chat Interfaces
When quantized versions of the model are released, I recommend using LM Studio for chatting with Hermes 2 Pro. It does not support function calling - for that use our github repo. It is a GUI application that utilizes GGUF models with a llama.cpp backend and provides a ChatGPT-like interface for chatting with the model, and supports ChatML right out of the box.
In LM-Studio, simply select the ChatML Prefix on the settings side pane:
## Quantized Versions:
GGUF Versions Available Here: https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF
# How to cite:
```bibtext
@misc{Hermes-2-Theta-Llama-3-8B,
url={[https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B][NousResearch/Hermes-2-Theta-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B))},
title={Hermes-2-Theta-Llama-3-8B},
author={"Teknium", Charles Goddard, "interstellarninja", "theemozilla", "karan4d", "huemin_art"}
}
```
|
{"base_model": "NousResearch/Hermes-2-Pro-Llama-3-8B", "datasets": ["teknium/OpenHermes-2.5"], "language": ["en"], "license": "apache-2.0", "tags": ["Llama-3", "instruct", "finetune", "chatml", "DPO", "RLHF", "gpt4", "synthetic data", "distillation", "function calling", "json mode", "axolotl", "merges"], "widget": [{"example_title": "Hermes 2 Pro Llama-3 Instruct Merge", "messages": [{"role": "system", "content": "You are a sentient, superintelligent artificial general intelligence, here to teach and assist me."}, {"role": "user", "content": "Write a short story about Goku discovering kirby has teamed up with Majin Buu to destroy the world."}]}], "model-index": [{"name": "Hermes-2-Pro-Llama-3-Instruct-8B-Merge", "results": []}]}
|
task
|
[
"TRANSLATION"
] | 45,154 |
muzerai/Deep-Llama-3.1-KoEn-8B-SiSai-GGUF
|
muzerai
| null |
[
"transformers",
"gguf",
"mergekit",
"merge",
"arxiv:2311.03099",
"arxiv:2306.01708",
"base_model:allenai/Llama-3.1-Tulu-3.1-8B",
"base_model:merge:allenai/Llama-3.1-Tulu-3.1-8B",
"base_model:cognitivecomputations/Dolphin3.0-Llama3.1-8B",
"base_model:merge:cognitivecomputations/Dolphin3.0-Llama3.1-8B",
"base_model:sh2orc/Llama-3.1-Korean-8B-Instruct",
"base_model:merge:sh2orc/Llama-3.1-Korean-8B-Instruct",
"endpoints_compatible",
"region:us",
"conversational"
] | 2025-02-18T08:44:24Z |
2025-02-18T09:11:51+00:00
| 88 | 1 |
---
base_model:
- allenai/Llama-3.1-Tulu-3.1-8B
- sh2orc/Llama-3.1-Korean-8B-Instruct
- cognitivecomputations/Dolphin3.0-Llama3.1-8B
library_name: transformers
tags:
- mergekit
- merge
---
# Deep-Llama-3.1-KoEn-8B-SiSai
This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [allenai/Llama-3.1-Tulu-3.1-8B](https://huggingface.co/allenai/Llama-3.1-Tulu-3.1-8B) as a base.
### Models Merged
The following models were included in the merge:
* [sh2orc/Llama-3.1-Korean-8B-Instruct](https://huggingface.co/sh2orc/Llama-3.1-Korean-8B-Instruct)
* [cognitivecomputations/Dolphin3.0-Llama3.1-8B](https://huggingface.co/cognitivecomputations/Dolphin3.0-Llama3.1-8B)
### Models Strength
Deep-Llama-3.1-KoEn-8B-SiSai is a Korean-English hybrid model with strong reasoning, instruction-following, and bilingual capabilities. The integration of Dolphin 3.0 ensures high-level inference, making it ideal for complex question-answering, professional translations, and deep analytical reasoning tasks. 🚀
### Ollama
```
ollama create Deep-Llama-3.1-KoEn-8B-SiSai -f Modelfile
```
```
대한민국 수도는
서울입니다. 서울은 대한민국의 수도이자 가장 큰 도시로, 인구와 경제력이 매우 크게 발전하고 있습니다. 또한, 문화적인 중심지로도 유명하며 다양한 사건과 역사적 변화가 발생한 곳입니다. 현재 서울은 국제적으로 중요한 도시로 인정받고 있으며, 많은 관광객들이 방문하는 인기 여행지 중 하나입니다.
```
```
뉴욕 여행일정
뉴욕은 세계에서 가장 대중적인 도시 중 하나로, 다양한 관광 명소와 문화적 활동이 풍부합니다. 뉴욕에 가면 어떤 것을 볼 수 있을까요? 다음은 뉴욕 여행 일정을 제안합니다:
1일차:
아침: Statue of Liberty (자유의 여신상)과 Ellis Island을 방문하여 미국 이민의 역사를 배웁니다.
중간: Central Park를 산책하며 휴식을 취하고, 거기에 인접한 Metropolitan Museum of Art (메트로폴리탄 미술관)를 방문합니다.
저녁: Times Square에서 야경을 감상하고, 뉴욕식 버거나 피자 같은 음식을 맛보세요.
2일차:
아침: 9/11 Memorial & Museum (9.11 기념관)과 One World Trade Center (원 월드 트레이드 센터)를 방문하여 뉴욕의 역사와 시련을 배웁니다.
중간: Brooklyn Bridge를 걸어보며 하버드 River의 아름다운 경치를 감상합니다.
저녁: Greenwich Village에서 로맨틱한 음식을 즐겨보세요.
3일차:
아침: 미국 국립 야생 동물 박물관 (American Museum of Natural History)에서 다양한 생물의 모습을 관찰합니다.
중간: 뉴욕 현대 미술관 (Museum of Modern Art, MoMA)을 방문하여 현대 예술 작품을 감상합니다.
저녁: Broadway 공연을 즐겨보세요.
4일차:
아침: 뉴욕 시청 (New York City Hall)을 방문하고, 근처의 Washington Square Park를 산책하며 휴식을 취합니다.
중간: Museum of the City of New York (뉴욕 시립 박물관)에서 뉴욕의 역사와 문화를 배웁니다.
저녁: 중국 요리나 이탈리아 요리를 맛보세요.
5일차:
아침: 뉴욕의 대중 매체 산업을 체험할 수 있는 Madame Tussauds New York (마담 투소드 뉴욕)과 Ripley's Believe It or Not! (립리 가문 박물관)를 방문합니다.
중간: 뉴욕 공항에서 국제 항공편으로 여행을 마무리하세요.
이 일정은 뉴욕의 다양한 관광 명소를 담아놓았지만, 개인적인 관심에 따라 변형될 수 있습니다. 또한, 뉴욕은 24시간 운영되는 도시라서 언제든지 자유롭게 이동할 수 있으므로, 이 일정 외에도 다른 활동을 추가하거나 변경할 수 있습니다. Enjoy your trip to New York! (뉴욕 여행에 즐거움을!)
```
```
어떤 마을에서는 비가 오는 날이면 사람들이 우산을 들고 다닌다. 오늘 아침 길을 걷던 철수가 많은 사람들이 우산을 들고 있는 것을 보았다. 철수가 "오늘 비가 왔구나"라고 결론 내릴 수 있는가?
그렇다 아니다 알 수 없다
답은 '알 수 없다'. 철수는 오늘 아침 길을 걷던 중에 많은 사람들이 우산을 들고 있는 것을 보았지만, 그들이 언제 비가 왔는지 정확히 알 수 없기 때문에 "오늘 비가 왔구나"라고 결론 내릴 수 없습니다. 철수에게 알려진 정보로는 우산을 들고 다니는 사람들의 모습뿐이며, 그들이 언제 비가 왔는지에 대한 정보는 부족합니다. 따라서 철수가 오늘 비가 왔는지 정확히 알 수 없으므로, "알 수 없다"가 정답입니다.
```
```
golang 코드 아무거나 서버 개발
아래는 Go 언어로 작성된 간단한 웹 서버 코드입니다.
package main
import (
"fmt"
"net/http"
)
func helloHandler(w http.ResponseWriter, r *http.Request) {
fmt.Fprint(w, "Hello, World!")
}
func main() {
http.HandleFunc("/", helloHandler)
http.ListenAndServe(":8080", nil)
}
이 코드는 HTTP 요청을 처리하는 웹 서버를 구현합니다. helloHandler 함수는 / 경로에 접근할 때 호출되며, 그 안에서 "Hello, World!" 메시지를 클라이언트에게 반환합니다.
서버는 포트 8080에서 시작됩니다. 이 코드를 Go 언어의 컴파일러인 go build 명령어로 컴파일하여 실행하면 웹 서버가 작동하며, 브라우저에서 http://localhost:8080/을 방문하면 "Hello, World!" 메시지가 표시됩니다.
이 코드는 단순하지만, 실제 서버 개발에서는 더 많은 기능과 복잡한 로직이 포함될 수 있습니다. 예를 들어, 데이터베이스 연결, 사용자 인증, REST API 구현 등 다양한 요소들이 필요할 수 있습니다. 그러나 이 코드는 기본적인 웹 서버의 작동을 이해하는 데 도움이 될 수 있습니다.
package main
import (
"fmt"
"net/http"
)
type User struct {
Name string `json:"name"`
Age int `json:"age"`
}
func getUsers(w http.ResponseWriter, r *http.Request) {
if r.Method != http.MethodGet {
http.Error(w, "Invalid request method", http.StatusBadRequest)
return
}
users := []User{
{"John Doe", 30},
{"Jane Doe", 25},
}
w.Header().Set("Content-Type", "application/json")
fmt.Fprint(w, users)
}
func createUser(w http.ResponseWriter, r *http.Request) {
if r.Method != http.MethodPost {
http.Error(w, "Invalid request method", http.StatusBadRequest)
return
}
var user User
err := fmt.Scanf(r.Body.String(), "%s %d", &user.Name, &user.Age)
if err != nil {
http.Error(w, "Failed to parse JSON", http.StatusBadRequest)
return
}
fmt.Fprintf(w, "User created: %+v\n", user)
}
func main() {
http.HandleFunc("/users", getUsers)
http.HandleFunc("/create", createUser)
http.ListenAndServe(":8080", nil)
}
이 코드는 REST API를 구현하여 사용자 목록을 조회하고 새 사용자를 생성하는 기능을 제공합니다. Users 구조체는 JSON 형식으로 데이터를 전송할 수 있도록 정의되며, getUsers 함수는 GET 요청을 처리하여 사용자 목록을 반환합니다. createUser 함수는 POST 요청을 처리하여 새 사용자를 생성합니다.
이 코드는 더 복잡한 서버 개발에 대한 예시로 사용될 수 있습니다. 그러나 실제 프로젝트에서는 데이터베이스 연동, 인증 및 권한 관리, 에러 처리 등 많은 요소들이 포함되어야 하며, 이 코드의 일부분만을 참고하여 구현해야 합니다.
```
### Use
Research & Educational Purposes: AI research, academic use, and educational content creation.
For questions about licensing, please contact my channel.
| null |
Non_BioNLP
|
# Deep-Llama-3.1-KoEn-8B-SiSai
This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [allenai/Llama-3.1-Tulu-3.1-8B](https://huggingface.co/allenai/Llama-3.1-Tulu-3.1-8B) as a base.
### Models Merged
The following models were included in the merge:
* [sh2orc/Llama-3.1-Korean-8B-Instruct](https://huggingface.co/sh2orc/Llama-3.1-Korean-8B-Instruct)
* [cognitivecomputations/Dolphin3.0-Llama3.1-8B](https://huggingface.co/cognitivecomputations/Dolphin3.0-Llama3.1-8B)
### Models Strength
Deep-Llama-3.1-KoEn-8B-SiSai is a Korean-English hybrid model with strong reasoning, instruction-following, and bilingual capabilities. The integration of Dolphin 3.0 ensures high-level inference, making it ideal for complex question-answering, professional translations, and deep analytical reasoning tasks. 🚀
### Ollama
```
ollama create Deep-Llama-3.1-KoEn-8B-SiSai -f Modelfile
```
```
대한민국 수도는
서울입니다. 서울은 대한민국의 수도이자 가장 큰 도시로, 인구와 경제력이 매우 크게 발전하고 있습니다. 또한, 문화적인 중심지로도 유명하며 다양한 사건과 역사적 변화가 발생한 곳입니다. 현재 서울은 국제적으로 중요한 도시로 인정받고 있으며, 많은 관광객들이 방문하는 인기 여행지 중 하나입니다.
```
```
뉴욕 여행일정
뉴욕은 세계에서 가장 대중적인 도시 중 하나로, 다양한 관광 명소와 문화적 활동이 풍부합니다. 뉴욕에 가면 어떤 것을 볼 수 있을까요? 다음은 뉴욕 여행 일정을 제안합니다:
1일차:
아침: Statue of Liberty (자유의 여신상)과 Ellis Island을 방문하여 미국 이민의 역사를 배웁니다.
중간: Central Park를 산책하며 휴식을 취하고, 거기에 인접한 Metropolitan Museum of Art (메트로폴리탄 미술관)를 방문합니다.
저녁: Times Square에서 야경을 감상하고, 뉴욕식 버거나 피자 같은 음식을 맛보세요.
2일차:
아침: 9/11 Memorial & Museum (9.11 기념관)과 One World Trade Center (원 월드 트레이드 센터)를 방문하여 뉴욕의 역사와 시련을 배웁니다.
중간: Brooklyn Bridge를 걸어보며 하버드 River의 아름다운 경치를 감상합니다.
저녁: Greenwich Village에서 로맨틱한 음식을 즐겨보세요.
3일차:
아침: 미국 국립 야생 동물 박물관 (American Museum of Natural History)에서 다양한 생물의 모습을 관찰합니다.
중간: 뉴욕 현대 미술관 (Museum of Modern Art, MoMA)을 방문하여 현대 예술 작품을 감상합니다.
저녁: Broadway 공연을 즐겨보세요.
4일차:
아침: 뉴욕 시청 (New York City Hall)을 방문하고, 근처의 Washington Square Park를 산책하며 휴식을 취합니다.
중간: Museum of the City of New York (뉴욕 시립 박물관)에서 뉴욕의 역사와 문화를 배웁니다.
저녁: 중국 요리나 이탈리아 요리를 맛보세요.
5일차:
아침: 뉴욕의 대중 매체 산업을 체험할 수 있는 Madame Tussauds New York (마담 투소드 뉴욕)과 Ripley's Believe It or Not! (립리 가문 박물관)를 방문합니다.
중간: 뉴욕 공항에서 국제 항공편으로 여행을 마무리하세요.
이 일정은 뉴욕의 다양한 관광 명소를 담아놓았지만, 개인적인 관심에 따라 변형될 수 있습니다. 또한, 뉴욕은 24시간 운영되는 도시라서 언제든지 자유롭게 이동할 수 있으므로, 이 일정 외에도 다른 활동을 추가하거나 변경할 수 있습니다. Enjoy your trip to New York! (뉴욕 여행에 즐거움을!)
```
```
어떤 마을에서는 비가 오는 날이면 사람들이 우산을 들고 다닌다. 오늘 아침 길을 걷던 철수가 많은 사람들이 우산을 들고 있는 것을 보았다. 철수가 "오늘 비가 왔구나"라고 결론 내릴 수 있는가?
그렇다 아니다 알 수 없다
답은 '알 수 없다'. 철수는 오늘 아침 길을 걷던 중에 많은 사람들이 우산을 들고 있는 것을 보았지만, 그들이 언제 비가 왔는지 정확히 알 수 없기 때문에 "오늘 비가 왔구나"라고 결론 내릴 수 없습니다. 철수에게 알려진 정보로는 우산을 들고 다니는 사람들의 모습뿐이며, 그들이 언제 비가 왔는지에 대한 정보는 부족합니다. 따라서 철수가 오늘 비가 왔는지 정확히 알 수 없으므로, "알 수 없다"가 정답입니다.
```
```
golang 코드 아무거나 서버 개발
아래는 Go 언어로 작성된 간단한 웹 서버 코드입니다.
package main
import (
"fmt"
"net/http"
)
func helloHandler(w http.ResponseWriter, r *http.Request) {
fmt.Fprint(w, "Hello, World!")
}
func main() {
http.HandleFunc("/", helloHandler)
http.ListenAndServe(":8080", nil)
}
이 코드는 HTTP 요청을 처리하는 웹 서버를 구현합니다. helloHandler 함수는 / 경로에 접근할 때 호출되며, 그 안에서 "Hello, World!" 메시지를 클라이언트에게 반환합니다.
서버는 포트 8080에서 시작됩니다. 이 코드를 Go 언어의 컴파일러인 go build 명령어로 컴파일하여 실행하면 웹 서버가 작동하며, 브라우저에서 http://localhost:8080/을 방문하면 "Hello, World!" 메시지가 표시됩니다.
이 코드는 단순하지만, 실제 서버 개발에서는 더 많은 기능과 복잡한 로직이 포함될 수 있습니다. 예를 들어, 데이터베이스 연결, 사용자 인증, REST API 구현 등 다양한 요소들이 필요할 수 있습니다. 그러나 이 코드는 기본적인 웹 서버의 작동을 이해하는 데 도움이 될 수 있습니다.
package main
import (
"fmt"
"net/http"
)
type User struct {
Name string `json:"name"`
Age int `json:"age"`
}
func getUsers(w http.ResponseWriter, r *http.Request) {
if r.Method != http.MethodGet {
http.Error(w, "Invalid request method", http.StatusBadRequest)
return
}
users := []User{
{"John Doe", 30},
{"Jane Doe", 25},
}
w.Header().Set("Content-Type", "application/json")
fmt.Fprint(w, users)
}
func createUser(w http.ResponseWriter, r *http.Request) {
if r.Method != http.MethodPost {
http.Error(w, "Invalid request method", http.StatusBadRequest)
return
}
var user User
err := fmt.Scanf(r.Body.String(), "%s %d", &user.Name, &user.Age)
if err != nil {
http.Error(w, "Failed to parse JSON", http.StatusBadRequest)
return
}
fmt.Fprintf(w, "User created: %+v\n", user)
}
func main() {
http.HandleFunc("/users", getUsers)
http.HandleFunc("/create", createUser)
http.ListenAndServe(":8080", nil)
}
이 코드는 REST API를 구현하여 사용자 목록을 조회하고 새 사용자를 생성하는 기능을 제공합니다. Users 구조체는 JSON 형식으로 데이터를 전송할 수 있도록 정의되며, getUsers 함수는 GET 요청을 처리하여 사용자 목록을 반환합니다. createUser 함수는 POST 요청을 처리하여 새 사용자를 생성합니다.
이 코드는 더 복잡한 서버 개발에 대한 예시로 사용될 수 있습니다. 그러나 실제 프로젝트에서는 데이터베이스 연동, 인증 및 권한 관리, 에러 처리 등 많은 요소들이 포함되어야 하며, 이 코드의 일부분만을 참고하여 구현해야 합니다.
```
### Use
Research & Educational Purposes: AI research, academic use, and educational content creation.
For questions about licensing, please contact my channel.
|
{"base_model": ["allenai/Llama-3.1-Tulu-3.1-8B", "sh2orc/Llama-3.1-Korean-8B-Instruct", "cognitivecomputations/Dolphin3.0-Llama3.1-8B"], "library_name": "transformers", "tags": ["mergekit", "merge"]}
|
task
|
[
"TRANSLATION"
] | 45,155 |
flax-sentence-embeddings/all_datasets_v3_roberta-large
|
flax-sentence-embeddings
|
sentence-similarity
|
[
"sentence-transformers",
"pytorch",
"roberta",
"feature-extraction",
"sentence-similarity",
"en",
"arxiv:2104.08727",
"arxiv:1810.09305",
"arxiv:2102.07033",
"arxiv:1904.06472",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2022-03-02T23:29:05Z |
2021-07-23T15:45:17+00:00
| 1,215 | 13 |
---
language: en
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
---
# Model description
The project aims to train sentence embedding models on very large sentence level datasets using a self-supervised
contrastive learning objective. We used the pretrained [`roberta-large`](https://huggingface.co/roberta-large) model and fine-tuned in on a
1B sentence pairs dataset. We use a contrastive learning objective: given a sentence from the pair, the model should predict which out of a set of randomly sampled other sentences, was actually paired with it in our dataset.
We developped this model during the
[Community week using JAX/Flax for NLP & CV](https://discuss.huggingface.co/t/open-to-the-community-community-week-using-jax-flax-for-nlp-cv/7104),
organized by Hugging Face. We developped this model as part of the project:
[Train the Best Sentence Embedding Model Ever with 1B Training Pairs](https://discuss.huggingface.co/t/train-the-best-sentence-embedding-model-ever-with-1b-training-pairs/7354). We benefited from efficient hardware infrastructure to run the project: 7 TPUs v3-8, as well
as intervention from Google’s Flax, JAX, and Cloud team member about efficient deep learning frameworks.
## Intended uses
Our model is intented to be used as a sentence encoder. Given an input sentence, it ouptuts a vector which captures
the sentence semantic information. The sentence vector may be used for information retrieval, clustering or sentence
similarity tasks.
## How to use
Here is how to use this model to get the features of a given text using [SentenceTransformers](https://github.com/UKPLab/sentence-transformers) library:
```python
from sentence_transformers import SentenceTransformer
model = SentenceTransformer('flax-sentence-embeddings/all_datasets_v3_roberta-large')
text = "Replace me by any text you'd like."
text_embbedding = model.encode(text)
# array([-0.01559514, 0.04046123, 0.1317083 , 0.00085931, 0.04585106,
# -0.05607086, 0.0138078 , 0.03569756, 0.01420381, 0.04266302 ...],
# dtype=float32)
```
# Training procedure
## Pre-training
We use the pretrained [`roberta-large`](https://huggingface.co/roberta-large). Please refer to the model
card for more detailed information about the pre-training procedure.
## Fine-tuning
We fine-tune the model using a contrastive objective. Formally, we compute the cosine similarity from each possible sentence pairs from the batch.
We then apply the cross entropy loss by comparing with true pairs.
### Hyper parameters
We trained ou model on a TPU v3-8. We train the model during 540k steps using a batch size of 1024 (128 per TPU core).
We use a learning rate warm up of 500. The sequence length was limited to 128 tokens. We used the AdamW optimizer with
a 2e-5 learning rate. The full training script is accessible in this current repository.
### Training data
We use the concatenation from multiple datasets to fine-tune our model. The total number of sentence pairs is above 1 billion sentences.
We sampled each dataset given a weighted probability which configuration is detailed in the `data_config.json` file.
| Dataset | Paper | Number of training tuples |
|:--------------------------------------------------------:|:----------------------------------------:|:--------------------------:|
| [GOOAQ: Open Question Answering with Diverse Answer Types](https://github.com/allenai/gooaq) | [paper](https://arxiv.org/pdf/2104.08727.pdf) | 3,012,496 |
| [Stack Exchange](https://huggingface.co/datasets/flax-sentence-embeddings/stackexchange_title_body_jsonl) | - | 364,001 |
| [Flickr 30k](https://shannon.cs.illinois.edu/DenotationGraph/) | [paper](https://transacl.org/ojs/index.php/tacl/article/view/229/33) | 317,695 |
| [COCO 2020](COCO 2020) | [paper](https://link.springer.com/chapter/10.1007%2F978-3-319-10602-1_48) | 828,395|
| [Code Search](https://huggingface.co/datasets/code_search_net) | - | 1,151,414 |
| [TriviaqQA](https://huggingface.co/datasets/trivia_qa) | - | 73,346 |
| [SQuAD2.0](https://rajpurkar.github.io/SQuAD-explorer/) | [paper](https://aclanthology.org/P18-2124.pdf) | 87,599 |
| [Natural Questions (NQ)](https://ai.google.com/research/NaturalQuestions) | [paper](https://transacl.org/ojs/index.php/tacl/article/view/1455) | 100,231 |
| [Simple Wikipedia](https://cs.pomona.edu/~dkauchak/simplification/) | [paper](https://www.aclweb.org/anthology/P11-2117/) | 102,225 |
| [Quora Question Pairs](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs) | - | 103,663 |
| [Altlex](https://github.com/chridey/altlex/) | [paper](https://aclanthology.org/P16-1135.pdf) | 112,696 |
| [Wikihow](https://github.com/pvl/wikihow_pairs_dataset) | [paper](https://arxiv.org/abs/1810.09305) | 128,542 |
| [Sentence Compression](https://github.com/google-research-datasets/sentence-compression) | [paper](https://www.aclweb.org/anthology/D13-1155/) | 180,000 |
| AllNLI ([SNLI](https://nlp.stanford.edu/projects/snli/) and [MultiNLI](https://cims.nyu.edu/~sbowman/multinli/) | [paper SNLI](https://doi.org/10.18653/v1/d15-1075), [paper MultiNLI](https://doi.org/10.18653/v1/n18-1101) | 277,230 |
| [Eli5](https://huggingface.co/datasets/eli5) | [paper](https://doi.org/10.18653/v1/p19-1346) | 325,475 |
| [SPECTER](https://github.com/allenai/specter) | [paper](https://doi.org/10.18653/v1/2020.acl-main.207) | 684,100 |
| [S2ORC](https://github.com/allenai/s2orc) Title/Abstract | [paper](https://aclanthology.org/2020.acl-main.447/) | 41,769,185 |
| [S2ORC](https://github.com/allenai/s2orc) Citation/Citation | [paper](https://aclanthology.org/2020.acl-main.447/) | 52,603,982 |
| [S2ORC](https://github.com/allenai/s2orc) Citation/Abstract | [paper](https://aclanthology.org/2020.acl-main.447/) | 116,288,806 |
| [PAQ](https://github.com/facebookresearch/PAQ) | [paper](https://arxiv.org/abs/2102.07033) | 64,371,441 |
| [WikiAnswers](https://github.com/afader/oqa#wikianswers-corpus) | [paper](https://doi.org/10.1145/2623330.2623677) | 77,427,422 |
| SearchQA | - | 582,261 |
| [Yahoo Answers](https://www.kaggle.com/soumikrakshit/yahoo-answers-dataset) Title/Answer | [paper](https://proceedings.neurips.cc/paper/2015/hash/250cf8b51c773f3f8dc8b4be867a9a02-Abstract.html) | 1,198,260 |
| [Yahoo Answers](https://www.kaggle.com/soumikrakshit/yahoo-answers-dataset) Title/Question | [paper](https://proceedings.neurips.cc/paper/2015/hash/250cf8b51c773f3f8dc8b4be867a9a02-Abstract.html) | 659,896 |
| [Yahoo Answers](https://www.kaggle.com/soumikrakshit/yahoo-answers-dataset) Question/Answer | [paper](https://proceedings.neurips.cc/paper/2015/hash/250cf8b51c773f3f8dc8b4be867a9a02-Abstract.html) | 681,164 |
| [MS MARCO](https://microsoft.github.io/msmarco/) | [paper](https://doi.org/10.1145/3404835.3462804) | 9,144,553 |
| [Reddit conversationnal](https://github.com/PolyAI-LDN/conversational-datasets/tree/master/reddit) | [paper](https://arxiv.org/abs/1904.06472) | 726,484,430 |
| total | | 1,097,953,922 |
| null |
Non_BioNLP
|
# Model description
The project aims to train sentence embedding models on very large sentence level datasets using a self-supervised
contrastive learning objective. We used the pretrained [`roberta-large`](https://huggingface.co/roberta-large) model and fine-tuned in on a
1B sentence pairs dataset. We use a contrastive learning objective: given a sentence from the pair, the model should predict which out of a set of randomly sampled other sentences, was actually paired with it in our dataset.
We developped this model during the
[Community week using JAX/Flax for NLP & CV](https://discuss.huggingface.co/t/open-to-the-community-community-week-using-jax-flax-for-nlp-cv/7104),
organized by Hugging Face. We developped this model as part of the project:
[Train the Best Sentence Embedding Model Ever with 1B Training Pairs](https://discuss.huggingface.co/t/train-the-best-sentence-embedding-model-ever-with-1b-training-pairs/7354). We benefited from efficient hardware infrastructure to run the project: 7 TPUs v3-8, as well
as intervention from Google’s Flax, JAX, and Cloud team member about efficient deep learning frameworks.
## Intended uses
Our model is intented to be used as a sentence encoder. Given an input sentence, it ouptuts a vector which captures
the sentence semantic information. The sentence vector may be used for information retrieval, clustering or sentence
similarity tasks.
## How to use
Here is how to use this model to get the features of a given text using [SentenceTransformers](https://github.com/UKPLab/sentence-transformers) library:
```python
from sentence_transformers import SentenceTransformer
model = SentenceTransformer('flax-sentence-embeddings/all_datasets_v3_roberta-large')
text = "Replace me by any text you'd like."
text_embbedding = model.encode(text)
# array([-0.01559514, 0.04046123, 0.1317083 , 0.00085931, 0.04585106,
# -0.05607086, 0.0138078 , 0.03569756, 0.01420381, 0.04266302 ...],
# dtype=float32)
```
# Training procedure
## Pre-training
We use the pretrained [`roberta-large`](https://huggingface.co/roberta-large). Please refer to the model
card for more detailed information about the pre-training procedure.
## Fine-tuning
We fine-tune the model using a contrastive objective. Formally, we compute the cosine similarity from each possible sentence pairs from the batch.
We then apply the cross entropy loss by comparing with true pairs.
### Hyper parameters
We trained ou model on a TPU v3-8. We train the model during 540k steps using a batch size of 1024 (128 per TPU core).
We use a learning rate warm up of 500. The sequence length was limited to 128 tokens. We used the AdamW optimizer with
a 2e-5 learning rate. The full training script is accessible in this current repository.
### Training data
We use the concatenation from multiple datasets to fine-tune our model. The total number of sentence pairs is above 1 billion sentences.
We sampled each dataset given a weighted probability which configuration is detailed in the `data_config.json` file.
| Dataset | Paper | Number of training tuples |
|:--------------------------------------------------------:|:----------------------------------------:|:--------------------------:|
| [GOOAQ: Open Question Answering with Diverse Answer Types](https://github.com/allenai/gooaq) | [paper](https://arxiv.org/pdf/2104.08727.pdf) | 3,012,496 |
| [Stack Exchange](https://huggingface.co/datasets/flax-sentence-embeddings/stackexchange_title_body_jsonl) | - | 364,001 |
| [Flickr 30k](https://shannon.cs.illinois.edu/DenotationGraph/) | [paper](https://transacl.org/ojs/index.php/tacl/article/view/229/33) | 317,695 |
| [COCO 2020](COCO 2020) | [paper](https://link.springer.com/chapter/10.1007%2F978-3-319-10602-1_48) | 828,395|
| [Code Search](https://huggingface.co/datasets/code_search_net) | - | 1,151,414 |
| [TriviaqQA](https://huggingface.co/datasets/trivia_qa) | - | 73,346 |
| [SQuAD2.0](https://rajpurkar.github.io/SQuAD-explorer/) | [paper](https://aclanthology.org/P18-2124.pdf) | 87,599 |
| [Natural Questions (NQ)](https://ai.google.com/research/NaturalQuestions) | [paper](https://transacl.org/ojs/index.php/tacl/article/view/1455) | 100,231 |
| [Simple Wikipedia](https://cs.pomona.edu/~dkauchak/simplification/) | [paper](https://www.aclweb.org/anthology/P11-2117/) | 102,225 |
| [Quora Question Pairs](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs) | - | 103,663 |
| [Altlex](https://github.com/chridey/altlex/) | [paper](https://aclanthology.org/P16-1135.pdf) | 112,696 |
| [Wikihow](https://github.com/pvl/wikihow_pairs_dataset) | [paper](https://arxiv.org/abs/1810.09305) | 128,542 |
| [Sentence Compression](https://github.com/google-research-datasets/sentence-compression) | [paper](https://www.aclweb.org/anthology/D13-1155/) | 180,000 |
| AllNLI ([SNLI](https://nlp.stanford.edu/projects/snli/) and [MultiNLI](https://cims.nyu.edu/~sbowman/multinli/) | [paper SNLI](https://doi.org/10.18653/v1/d15-1075), [paper MultiNLI](https://doi.org/10.18653/v1/n18-1101) | 277,230 |
| [Eli5](https://huggingface.co/datasets/eli5) | [paper](https://doi.org/10.18653/v1/p19-1346) | 325,475 |
| [SPECTER](https://github.com/allenai/specter) | [paper](https://doi.org/10.18653/v1/2020.acl-main.207) | 684,100 |
| [S2ORC](https://github.com/allenai/s2orc) Title/Abstract | [paper](https://aclanthology.org/2020.acl-main.447/) | 41,769,185 |
| [S2ORC](https://github.com/allenai/s2orc) Citation/Citation | [paper](https://aclanthology.org/2020.acl-main.447/) | 52,603,982 |
| [S2ORC](https://github.com/allenai/s2orc) Citation/Abstract | [paper](https://aclanthology.org/2020.acl-main.447/) | 116,288,806 |
| [PAQ](https://github.com/facebookresearch/PAQ) | [paper](https://arxiv.org/abs/2102.07033) | 64,371,441 |
| [WikiAnswers](https://github.com/afader/oqa#wikianswers-corpus) | [paper](https://doi.org/10.1145/2623330.2623677) | 77,427,422 |
| SearchQA | - | 582,261 |
| [Yahoo Answers](https://www.kaggle.com/soumikrakshit/yahoo-answers-dataset) Title/Answer | [paper](https://proceedings.neurips.cc/paper/2015/hash/250cf8b51c773f3f8dc8b4be867a9a02-Abstract.html) | 1,198,260 |
| [Yahoo Answers](https://www.kaggle.com/soumikrakshit/yahoo-answers-dataset) Title/Question | [paper](https://proceedings.neurips.cc/paper/2015/hash/250cf8b51c773f3f8dc8b4be867a9a02-Abstract.html) | 659,896 |
| [Yahoo Answers](https://www.kaggle.com/soumikrakshit/yahoo-answers-dataset) Question/Answer | [paper](https://proceedings.neurips.cc/paper/2015/hash/250cf8b51c773f3f8dc8b4be867a9a02-Abstract.html) | 681,164 |
| [MS MARCO](https://microsoft.github.io/msmarco/) | [paper](https://doi.org/10.1145/3404835.3462804) | 9,144,553 |
| [Reddit conversationnal](https://github.com/PolyAI-LDN/conversational-datasets/tree/master/reddit) | [paper](https://arxiv.org/abs/1904.06472) | 726,484,430 |
| total | | 1,097,953,922 |
|
{"language": "en", "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "feature-extraction", "sentence-similarity"]}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,156 |
uvegesistvan/wildmann_german_proposal_2b_GER_ENG_CZ
|
uvegesistvan
| null |
[
"tensorboard",
"safetensors",
"xlm-roberta",
"emotion-classification",
"text-analysis",
"machine-translation",
"cs",
"license:mit",
"region:us"
] | 2025-01-17T08:45:37Z |
2025-01-17T21:09:46+00:00
| 8 | 0 |
---
language: cs
license: mit
metrics:
- precision
- recall
- f1-score
- accuracy
tags:
- emotion-classification
- text-analysis
- machine-translation
---
# Model Card for uvegesistvan/wildmann_german_proposal_2b_GER_ENG_CZ
## Model Overview
This model is a multi-class emotion classifier trained on German text that was first machine-translated into English as an intermediary language and then into Czech. It identifies nine distinct emotional states in text. The training process explores the impact of multi-step machine translation on emotion classification accuracy and robustness.
### Emotion Classes
The model classifies the following emotional states:
- **Anger (0)**
- **Fear (1)**
- **Disgust (2)**
- **Sadness (3)**
- **Joy (4)**
- **Enthusiasm (5)**
- **Hope (6)**
- **Pride (7)**
- **No emotion (8)**
### Dataset and Preprocessing
The dataset was created using a three-step machine translation process: German → English → Czech. Emotional annotations were applied after the final translation to ensure consistency. Preprocessing steps included:
- Balancing the dataset through undersampling overrepresented classes like "No emotion" and "Anger."
- Normalizing text to mitigate noise introduced by multi-step translations.
### Evaluation Metrics
The model's performance was evaluated using standard classification metrics. Results are detailed below:
| Class | Precision | Recall | F1-Score | Support |
|---------------|-----------|--------|----------|---------|
| Anger (0) | 0.55 | 0.53 | 0.54 | 777 |
| Fear (1) | 0.85 | 0.75 | 0.80 | 776 |
| Disgust (2) | 0.90 | 0.95 | 0.92 | 776 |
| Sadness (3) | 0.86 | 0.83 | 0.85 | 775 |
| Joy (4) | 0.85 | 0.80 | 0.82 | 777 |
| Enthusiasm (5)| 0.67 | 0.59 | 0.63 | 776 |
| Hope (6) | 0.52 | 0.49 | 0.51 | 777 |
| Pride (7) | 0.75 | 0.79 | 0.77 | 776 |
| No emotion (8)| 0.60 | 0.69 | 0.64 | 1553 |
### Overall Metrics
- **Accuracy**: 0.71
- **Macro Average**: Precision = 0.73, Recall = 0.71, F1-Score = 0.72
- **Weighted Average**: Precision = 0.71, Recall = 0.71, F1-Score = 0.71
### Performance Insights
The model performs well in classes such as "Disgust" and "Fear." However, "Hope" and "Enthusiasm" classes show slightly lower performance, likely due to complexities introduced by the multi-step translation process. Overall, the model demonstrates strong performance across most classes.
## Model Usage
### Applications
- Emotion analysis of German texts via machine-translated Czech representations.
- Sentiment analysis for Czech-language datasets derived from multilingual sources.
- Research on the effects of multi-step machine translation in emotion classification.
### Limitations
- The multi-step translation process introduces additional noise, potentially impacting classification accuracy for subtle or ambiguous emotions.
- Emotional nuances and cultural context might be lost during translation.
### Ethical Considerations
The reliance on multi-step machine translation can amplify biases or inaccuracies introduced at each stage. Careful validation is recommended before applying the model in sensitive areas such as mental health, social research, or customer feedback analysis.
### Citation
For further information, visit: [uvegesistvan/wildmann_german_proposal_2b_GER_ENG_CZ](#)
| null |
Non_BioNLP
|
# Model Card for uvegesistvan/wildmann_german_proposal_2b_GER_ENG_CZ
## Model Overview
This model is a multi-class emotion classifier trained on German text that was first machine-translated into English as an intermediary language and then into Czech. It identifies nine distinct emotional states in text. The training process explores the impact of multi-step machine translation on emotion classification accuracy and robustness.
### Emotion Classes
The model classifies the following emotional states:
- **Anger (0)**
- **Fear (1)**
- **Disgust (2)**
- **Sadness (3)**
- **Joy (4)**
- **Enthusiasm (5)**
- **Hope (6)**
- **Pride (7)**
- **No emotion (8)**
### Dataset and Preprocessing
The dataset was created using a three-step machine translation process: German → English → Czech. Emotional annotations were applied after the final translation to ensure consistency. Preprocessing steps included:
- Balancing the dataset through undersampling overrepresented classes like "No emotion" and "Anger."
- Normalizing text to mitigate noise introduced by multi-step translations.
### Evaluation Metrics
The model's performance was evaluated using standard classification metrics. Results are detailed below:
| Class | Precision | Recall | F1-Score | Support |
|---------------|-----------|--------|----------|---------|
| Anger (0) | 0.55 | 0.53 | 0.54 | 777 |
| Fear (1) | 0.85 | 0.75 | 0.80 | 776 |
| Disgust (2) | 0.90 | 0.95 | 0.92 | 776 |
| Sadness (3) | 0.86 | 0.83 | 0.85 | 775 |
| Joy (4) | 0.85 | 0.80 | 0.82 | 777 |
| Enthusiasm (5)| 0.67 | 0.59 | 0.63 | 776 |
| Hope (6) | 0.52 | 0.49 | 0.51 | 777 |
| Pride (7) | 0.75 | 0.79 | 0.77 | 776 |
| No emotion (8)| 0.60 | 0.69 | 0.64 | 1553 |
### Overall Metrics
- **Accuracy**: 0.71
- **Macro Average**: Precision = 0.73, Recall = 0.71, F1-Score = 0.72
- **Weighted Average**: Precision = 0.71, Recall = 0.71, F1-Score = 0.71
### Performance Insights
The model performs well in classes such as "Disgust" and "Fear." However, "Hope" and "Enthusiasm" classes show slightly lower performance, likely due to complexities introduced by the multi-step translation process. Overall, the model demonstrates strong performance across most classes.
## Model Usage
### Applications
- Emotion analysis of German texts via machine-translated Czech representations.
- Sentiment analysis for Czech-language datasets derived from multilingual sources.
- Research on the effects of multi-step machine translation in emotion classification.
### Limitations
- The multi-step translation process introduces additional noise, potentially impacting classification accuracy for subtle or ambiguous emotions.
- Emotional nuances and cultural context might be lost during translation.
### Ethical Considerations
The reliance on multi-step machine translation can amplify biases or inaccuracies introduced at each stage. Careful validation is recommended before applying the model in sensitive areas such as mental health, social research, or customer feedback analysis.
### Citation
For further information, visit: [uvegesistvan/wildmann_german_proposal_2b_GER_ENG_CZ](#)
|
{"language": "cs", "license": "mit", "metrics": ["precision", "recall", "f1-score", "accuracy"], "tags": ["emotion-classification", "text-analysis", "machine-translation"]}
|
task
|
[
"TRANSLATION"
] | 45,157 |
thomaskim1130/stella_en_400M_v5-FinanceRAG-v2
|
thomaskim1130
|
sentence-similarity
|
[
"sentence-transformers",
"safetensors",
"new",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:2256",
"loss:MultipleNegativesRankingLoss",
"custom_code",
"arxiv:1908.10084",
"arxiv:1705.00652",
"base_model:thomaskim1130/stella_en_400M_v5-FinanceRAG",
"base_model:finetune:thomaskim1130/stella_en_400M_v5-FinanceRAG",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2024-11-29T15:04:50Z |
2024-11-29T15:05:41+00:00
| 599 | 5 |
---
base_model: thomaskim1130/stella_en_400M_v5-FinanceRAG
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
- dot_accuracy@1
- dot_accuracy@3
- dot_accuracy@5
- dot_accuracy@10
- dot_precision@1
- dot_precision@3
- dot_precision@5
- dot_precision@10
- dot_recall@1
- dot_recall@3
- dot_recall@5
- dot_recall@10
- dot_ndcg@10
- dot_mrr@10
- dot_map@100
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:2256
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: "Instruct: Given a web search query, retrieve relevant passages\
\ that answer the query.\nQuery: Title: \nText: What was the sum of Fourth Quarter\
\ without those Fourth Quarter smaller than 0, in 2012? (in million)"
sentences:
- "Title: \nText: Cash Flow Hedges Citigroup hedges the variability of forecasted\
\ cash flows associated with floating-rate assets/liabilities and other forecasted\
\ transactions.\nVariable cash flows from those liabilities are synthetically\
\ converted to fixed-rate cash flows by entering into receive-variable, pay-fixed\
\ interest rate swaps and receivevariable, pay-fixed forward-starting interest\
\ rate swaps.\nVariable cash flows associated with certain assets are synthetically\
\ converted to fixed-rate cash flows by entering into receive-fixed, pay-variable\
\ interest rate swaps.\nThese cash flow hedging relationships use either regression\
\ analysis or dollar-offset ratio analysis to assess whether the hedging relationships\
\ are highly effective at inception and on an ongoing basis.\nPrior to the adoption\
\ of ASU 2017-12, Citigroup designated the risk being hedged as the risk of overall\
\ variability in the hedged cash flows for certain items.\nWith the adoption of\
\ ASU 2017-12, Citigroup hedges the variability from changes in a contractually\
\ specified rate and recognizes the entire change in fair value of the cash flow\
\ hedging instruments in AOCI.\nPrior to the adoption of ASU 2017-12, to the extent\
\ that these derivatives were not fully effective, changes in their fair values\
\ in excess of changes in the value of the hedged transactions were immediately\
\ included in Other revenue.\nWith the adoption of ASU 2017-12, such amounts are\
\ no longer required to be immediately recognized in income, but instead the full\
\ change in the value of the hedging instrument is required to be recognized in\
\ AOCI, and then recognized in earnings in the same period that the cash flows\
\ impact earnings.\nThe pretax change in AOCI from cash flow hedges is presented\
\ below:\n| | Year ended December 31, |\n| In millions of dollars | 2018 | 2017\
\ | 2016 |\n| Amount of gain (loss) recognized in AOCI on derivative | | | \
\ |\n| Interest rate contracts-1 | $-361 | | $-165 | $-219 |\n| Foreign exchange\
\ contracts | 5 | -8 | 69 |\n| Total gain (loss) recognized in AOCI | $-356 |\
\ | $-173 | $-150 |\n| Amount of gain (loss) reclassified from AOCI to earnings\
\ | Otherrevenue | Net interestrevenue | Otherrevenue | Otherrevenue |\n| Interest\
\ rate contracts-1 | $— | $-301 | $-126 | $-140 |\n| Foreign exchange contracts\
\ | -17 | — | -10 | -93 |\n| Total gain (loss) reclassified from AOCI into earnings\
\ | $-17 | $-301 | $-136 | $-233 |\n(1) After January 1, 2018, all amounts reclassified\
\ into earnings for interest rate contracts are included in Interest income/Interest\
\ expense (Net interest revenue).\nFor all other hedges, including interest rate\
\ hedges prior to January 1, 2018, the amounts reclassified to earnings are included\
\ primarily in Other revenue and Net interest revenue in the Consolidated Statement\
\ of Income.\nFor cash flow hedges, the changes in the fair value of the hedging\
\ derivative remain in AOCI on the Consolidated Balance Sheet and will be included\
\ in the earnings of future periods to offset the variability of the hedged cash\
\ flows when such cash flows affect earnings.\nThe net gain (loss) associated\
\ with cash flow hedges expected to be reclassified from AOCI within 12?months\
\ of December?31, 2018 is approximately $404 million.\nThe maximum length of time\
\ over which forecasted cash flows are hedged is 10 years.\nThe after-tax impact\
\ of cash flow hedges on AOCI is shown in Note?19 to the Consolidated Financial\
\ Statements."
- "Title: \nText: | | Net Sales |\n| (Amounts in millions) | 2012 | 2011 | 2010\
\ |\n| Product Category: | | | |\n| Tools | $1,729.4 | $1,667.3 | $1,545.1\
\ |\n| Diagnostics and repair information | 619.8 | 613.7 | 563.3 |\n| Equipment\
\ | 588.7 | 573.2 | 510.8 |\n| | $2,937.9 | $2,854.2 | $2,619.2 |\nThe tools\
\ product category includes hand tools, power tools and tool storage products.\n\
Hand tools include wrenches, sockets, ratchet wrenches, pliers, screwdrivers,\
\ punches and chisels, saws and cutting tools, pruning tools, torque measuring\
\ instruments and other similar products.\nPower tools include cordless (battery),\
\ pneumatic (air), hydraulic, and corded (electric) tools, such as impact wrenches,\
\ ratchets, chisels, drills, sanders, polishers and similar products.\nTool storage\
\ includes tool chests, roll cabinets, tool control systems and other similar\
\ products.\nThe majority of products are manufactured by Snap-on and, in completing\
\ the product offering, other items are purchased from external manufacturers.\n\
The diagnostics and repair information product category includes handheld and\
\ PC-based diagnostic products, service and repair information products, diagnostic\
\ software solutions, electronic parts catalogs, business management systems and\
\ services, point-of-sale systems, integrated systems for vehicle service shops,\
\ OEM purchasing facilitation services, and warranty management systems and analytics\
\ to help OEM dealership service and repair shops manage and track performance.\n\
The equipment product category includes solutions for the diagnosis and service\
\ of vehicles and industrial equipment.\nProducts include wheel alignment equipment,\
\ wheel balancers, tire changers, vehicle lifts, test lane systems, collision\
\ repair equipment, air conditioning service equipment, brake service equipment,\
\ fluid exchange equipment, transmission troubleshooting equipment, safety testing\
\ equipment, battery chargers and hoists.\nSnap-on supports the sale of its diagnostics\
\ and vehicle service shop equipment by offering training programs as well as\
\ after sales support for its customers, primarily focusing on the technologies\
\ and the application of specific products developed and marketed by Snap-on.\n\
Management’s Discussion and Analysis of Financial Condition and Results of Operations\
\ (continued) Segment gross profit of $105.0 million in the fourth quarter of\
\ 2012 decreased $1.4 million from 2011 levels.\nGross margin of 38.1% in the\
\ quarter improved 210 basis points from 36.0% last year primarily due to lower\
\ restructuring costs as well as savings from ongoing RCI initiatives, particularly\
\ in Europe.\nNo restructuring costs were incurred in the fourth quarter of 2012;\
\ gross profit in the fourth quarter of 2011 included $2.5 million of restructuring\
\ costs.\nSegment operating expenses of $73.1 million in the fourth quarter of\
\ 2012 decreased $0.3 million from 2011 levels.\nThe operating expense margin\
\ of 26.5% in the quarter increased 170 basis points from 24.8% last year primarily\
\ as a result of the lower sales.\nAs a result of these factors, segment operating\
\ earnings of $31.9 million in the fourth quarter of 2012, including $1.2 million\
\ of favorable foreign currency effects, decreased $1.1 million, or 3.3%, from\
\ 2011 levels.\nOperating margin for the Commercial & Industrial Group of 11.6%\
\ in the fourth quarter of 2012 improved 40 basis points from 11.2% last year.\n\
Snap-on Tools Group"
- "Title: \nText: 5.\nBasis of Presentation and Summary of Significant Accounting\
\ Policies (a) Basis of Presentation On December 30, 2015, US Airways merged with\
\ and into American, which is reflected in American’s consolidated financial statements\
\ as though the transaction had occurred on December 9, 2013, when a subsidiary\
\ of AMR merged with and into US Airways Group.\nThus, the full years of 2015\
\ and 2014 and the period from December 9, 2013 to December 31, 2013 are comprised\
\ of the consolidated financial data of American and US Airways.\nFor the periods\
\ prior to December 9, 2013, the financial data reflects the results of American\
\ only.\nFor financial reporting purposes, the transaction constituted a transfer\
\ of assets between entities under common control and was accounted for in a manner\
\ similar to the pooling of interests method of accounting.\nUnder this method,\
\ the carrying amount of net assets recognized in the balance sheets of each combining\
\ entity are carried forward to the balance sheet of the combined entity and no\
\ other assets or liabilities are recognized.\nThe preparation of financial statements\
\ in accordance with accounting principles generally accepted in the United States\
\ (GAAP) requires management to make certain estimates and assumptions that affect\
\ the reported amounts of assets and liabilities, revenues and expenses, and the\
\ disclosure of contingent assets and liabilities at the date of the financial\
\ statements.\nActual results could differ from those estimates.\nThe most significant\
\ areas of judgment relate to passenger revenue recognition, impairment of goodwill,\
\ impairment of long-lived and\nThe following tables set forth the income yield\
\ and investment income, excluding realized investment gains (losses) and non-hedge\
\ accounting derivative results, for each major investment category of our Japanese\
\ operations’ general account for the periods indicated."
- source_sentence: "Instruct: Given a web search query, retrieve relevant passages\
\ that answer the query.\nQuery: Title: \nText: ADBE share repurchase"
sentences:
- "Title: \nText: TELEFLEX INCORPORATED NOTES?TO CONSOLIDATED FINANCIAL STATEMENTS\
\ — (Continued) The Company issued 82,865, 93,367 and 105,239 of non-vested restricted\
\ stock units in 2017, 2016 and 2015, respectively, the majority of which provide\
\ for vesting as to all underlying shares on the third anniversary of the grant\
\ date.\nThe weighted average grant-date fair value for non-vested restricted\
\ stock units granted during 2017, 2016 and 2015 was $187.85, $142.71 and $118.00,\
\ respectively.\nThe Company recorded $11.2 million of expense related to restricted\
\ stock units during 2017, which is included in cost of goods sold or selling,\
\ general and administrative expenses.\nThe unamortized share-based compensation\
\ cost related to non-vested restricted stock units, net of expected forfeitures,\
\ was $13.2 million, which is expected to be recognized over a weighted-average\
\ period of 1.8 years.\nThe Company uses treasury stock to provide shares of common\
\ stock in connection with vesting of the restricted stock units.\nTELEFLEX INCORPORATED\
\ NOTES?TO CONSOLIDATED FINANCIAL STATEMENTS — (Continued) F-37 Note 13?— Income\
\ taxes The following table summarizes the components of the provision for income\
\ taxes from continuing operations:\n| | 2017 | 2016 | 2015 |\n| | (Dollars\
\ in thousands) |\n| Current: | | | |\n| Federal | $133,621 | $2,344 | $-4,700\
\ |\n| State | 5,213 | 5,230 | 2,377 |\n| Foreign | 35,444 | 28,842 | 53,151 |\n\
| Deferred: | | | |\n| Federal | -258,247 | -25,141 | -35,750 |\n| State |\
\ 1,459 | -1,837 | -5,012 |\n| Foreign | 212,158 | -1,364 | -2,228 |\n| | $129,648\
\ | $8,074 | $7,838 |\nThe Tax Cuts and Jobs Act (the “TCJA”) was enacted on December\
\ 22, 2017.\nThe legislation significantly changes U. S. tax law by, among other\
\ things, permanently reducing corporate income tax rates from a maximum of 35%\
\ to 21%, effective January 1, 2018; implementing a territorial tax system, by\
\ generally providing for, among other things, a dividends received deduction\
\ on the foreign source portion of dividends received from a foreign corporation\
\ if specified conditions are met; and imposing a one-time repatriation tax on\
\ undistributed post-1986 foreign subsidiary earnings and profits, which are deemed\
\ repatriated for purposes of the tax.\nAs a result of the TCJA, the Company reassessed\
\ and revalued its ending net deferred tax liabilities at December 31, 2017 and\
\ recognized a?$46.1 million?provisional tax benefit in the Company’s consolidated\
\ statement of income for the year ended December 31, 2017.\nAs a result of the\
\ deemed repatriation tax under the TCJA, the Company recognized a $154.0 million\
\ provisional tax expense in the Company’s consolidated statement of income for\
\ the year ended December 31, 2017, and the Company expects to pay this tax over\
\ an eight-year period.\nWhile the TCJA provides for a territorial tax system,\
\ beginning in 2018, it includes?two?new U. S. tax base erosion provisions, the\
\ global intangible low-taxed income (“GILTI”) provisions and the base-erosion\
\ and anti-abuse tax (“BEAT”) provisions.\nThe GILTI provisions require the Company\
\ to include in its U. S. income tax return foreign subsidiary earnings in excess\
\ of an allowable return on the foreign subsidiary’s tangible assets.\nThe Company\
\ expects that it will be subject to incremental U. S. tax on GILTI income beginning\
\ in 2018.\nBecause of the complexity of the new GILTI tax rules, the Company\
\ is continuing to evaluate this provision of the TCJA and the application of\
\ Financial Accounting Standards Board Accounting Standards Codification Topic\
\ 740, \"Income Taxes. \"\nUnder U. S. GAAP, the Company may make an accounting\
\ policy election to either (1) treat future taxes with respect to the inclusion\
\ in U. S. taxable income of amounts related to GILTI as current period expense\
\ when incurred (the “period cost method”) or (2) take such amounts into a company’s\
\ measurement of its deferred taxes (the “deferred method”).\nThe Company’s selection\
\ of an accounting policy with respect to the new GILTI tax rules will depend,\
\ in part, on an analysis of the Company’s global income to determine whether\
\ the Company expects to have future U. S. inclusions in taxable income related\
\ to GILTI and, if so, what the impact is expected to be.\nThe determination of\
\ whether the Company expects to have future U. S. inclusions"
- 'Title: ADBE NOTES TO CONSOLIDATED FINANCIAL STATEMENTS (Continued)
Text: For fiscal 2023, 2022 and 2021, the prepayments were classified as treasury
stock on our Consolidated Balance Sheets at the payment date, though only shares
physically delivered to us by December 1, 2023, December 2, 2022 and December
3, 2021 were excluded from the computation of net income per share. As of December
1, 2023, $354 million of prepayment remained under our outstanding structured
stock repurchase agreement.'
- "Title: \nText: Market and Market Prices of Common Stock\nDuring 2016 fiscal year\
\ and through February 22, 2017, our common stock was traded on the Nasdaq Capital\
\ Market under the symbol “ACUR”. On February 23, 2017, our common stock was delisted\
\ from the Nasdaq Capital Market due to our failure to comply with Nasdaq’s Listing\
\ Rule 5550(b)(1), which requires that we maintain $2.5 million in stockholders’\
\ equity for continued listing (or meet the alternatives of market value of listed\
\ securities of $35 million or net income from continuing operations). NASDAQ\
\ had granted us a grace period through February 10, 2017, to regain compliance\
\ with Listing Rule 5550(b)(1), but we were unable to regain compliance within\
\ such period.\nCommencing on February 23, 2017, our common stock was quoted on\
\ the OTCQB under the symbol “ACUR”, however commencing June 4, 2018 and lasting\
\ until July 2, 2018 it was quoted on the OTC Markets OTC Pink tier. The downgrade\
\ was a result of the late filing of our 2017 Annual Report on Form 10-K beyond\
\ any applicable grace periods. The Company regained compliance with the OTCQB\
\ and effective July 3, 2018 it was quoted on the OTCQB. However, commencing May\
\ 20, 2019 as a result of late filing of our 2018 Annual Report on Form 10-K our\
\ common stock was again relegated to the OTC Markets OTC Pink tier. The Company\
\ regained compliance with the OTCQB in March, 2020 and effective March 23, 2020\
\ it was quoted on the OTCQB.\nSet forth below for the period indicated are the\
\ high and low sales prices for our common stock in the OTC Market of OTCQB and\
\ Pink tier.\nOn March 27, 2020 the closing sales price of our common stock was\
\ $0.22.\n\nPeriod | | Sales Prices\n---------------------------------\
\ | ----- | ------------\n | High | Low \
\ \n2019 Fiscal Year | | \nFirst Quarter\
\ | $0.29 | $0.11 \nSecond Quarter \
\ | 0.28 | 0.13 \nThird Quarter | 0.45 | 0.14 \
\ \nFourth Quarter | 0.63 | 0.20 \n2020 Fiscal\
\ Year | | \nFirst Quarter thru March 27, 2020\
\ | 0.47 | $0.12 "
- source_sentence: "Instruct: Given a web search query, retrieve relevant passages\
\ that answer the query.\nQuery: Title: \nText: What is the growing rate of Equity\
\ securities, trading for Carrying amount in the year with the most Fixed maturities,\
\ available-for-sale ?"
sentences:
- "Title: \nText: The following table details the estimated changes by risk management\
\ strategy.\nThe table also gives the weighted-average duration of the asset portfolio\
\ for each category, and the net duration gap (i. e. , the weighted-average difference\
\ between the asset and liability durations).\n| | December 31, 2007 |\n| Risk\
\ Management Strategy | Value of total assets(in millions) | Duration of assets\
\ | Net duration gap | Net fair value change (in millions) |\n| Primary duration-managed\
\ | $33,183.4 | 3.53 | -0.01 | $3.3 |\n| Duration-monitored | 17,990.9 | 4.70\
\ | 0.22 | -39.6 |\n| Non duration-managed | 5,234.0 | 4.43 | N/A | N/A |\n| Total\
\ | $56,408.3 | | | $-36.3 |\nOur selection of a 100 basis point immediate,\
\ parallel increase or decrease in interest rates is a hypothetical rate scenario\
\ we use to demonstrate potential risk.\nWhile a 100 basis point immediate, parallel\
\ increase does not represent our view of future market changes, it is a near\
\ term reasonably possible hypothetical change that illustrates the potential\
\ impact of such events.\nWhile these fair value measurements provide a representation\
\ of interest rate sensitivity, they are based on our portfolio exposures at a\
\ point in time and may not be representative of future market results.\nThese\
\ exposures will change as a result of ongoing portfolio transactions in response\
\ to new business, management’s assessment of changing market conditions and available\
\ investment opportunities."
- "Title: \nText: PART I ITEM 1. BUSINESS (dollars in millions, except per share,\
\ per ounce and per pound amounts) Introduction Newmont Mining Corporation is\
\ primarily a gold producer with significant operations and/or assets in the United\
\ States, Australia, Peru, Ghana and Suriname.\nAt December 31, 2016, Newmont\
\ had attributable proven and probable gold reserves of 68.5 million ounces and\
\ an aggregate land position of approximately 23,000 square miles (59,000 square\
\ kilometers).\nNewmont is also engaged in the production of copper, principally\
\ through Boddington in Australia and Phoenix in the United States.\nNewmont Mining\
\ Corporation\x80\x99s original predecessor corporation was incorporated in 1921\
\ under the laws of Delaware.\nOn November 2, 2016, Newmont completed the sale\
\ of its 48.5% economic interest in PT Newmont Nusa Tenggara (\x80\x9CPTNNT\x80\
\x9D), which operated the Batu Hijau copper and gold mine (\x80\x9CBatu Hijau\x80\
\x9D) in Indonesia (the \x80\x9CBatu Hijau Transaction\x80\x9D).\nAs a result,\
\ Newmont presents Batu Hijau as a discontinued operation for all periods presented.\n\
In the following discussion, we present and discuss our continuing operations\
\ unless otherwise indicated.\nFor additional information regarding our discontinued\
\ operations, see Note 3 to the Consolidated Financial Statements and the discussion\
\ in our Results of Consolidated Operations in Item 7.\nNewmont\x80\x99s corporate\
\ headquarters are in Greenwood Village, Colorado, USA.\nIn this report, \x80\x9C\
Newmont,\x80\x9D the \x80\x9CCompany,\x80\x9D \x80\x9Cour\x80\x9D and \x80\x9C\
we\x80\x9D refer to Newmont Mining Corporation together with our affiliates and\
\ subsidiaries, unless the context otherwise requires.\nReferences to \x80\x9C\
A$\x80\x9D refer to Australian currency.\nNewmont\x80\x99s Sales and long-lived\
\ assets for continuing operations are geographically distributed as follows:\n\
| | Sales | Long-Lived Assets |\n| | 2016 | 2015 | 2014 | 2016 | 2015 | 2014\
\ |\n| United States | 39% | 33% | 30% | 45% | 43% | 38% |\n| Australia | 32%\
\ | 32% | 30% | 19% | 18% | 19% |\n| Ghana | 15% | 15% | 17% | 16% | 16% | 17%\
\ |\n| Peru | 12% | 18% | 18% | 14% | 19% | 23% |\n| Suriname | 2% | —% | —% |\
\ 6% | 4% | 2% |\n| Other | —% | 2% | 5% | —% | —% | 1% |\nSegment Information\
\ Our regions include North America, South America, Asia Pacific, and Africa.\n\
Our North America segment consists primarily of Carlin, Phoenix, Twin Creeks and\
\ Long Canyon in the state of Nevada and Cripple Creek &Victor (\x80\x9CCC&V\x80\
\x9D) in the state of Colorado, in the United States.\nOur South America segment\
\ consists primarily of Yanacocha in Peru and Merian in Suriname.\nOur Asia Pacific\
\ segment consists primarily of Boddington, Tanami and Kalgoorlie in Australia.\n\
Our Africa segment consists primarily of Ahafo and Akyem in Ghana.\nSee Item 1A,\
\ Risk Factors, below, and Note 5 to the Consolidated Financial Statements for\
\ information relating to our operating segments, domestic and export sales and\
\ lack of dependence on a limited number of customers.\nProducts References in\
\ this report to \x80\x9Cattributable gold ounces\x80\x9D or \x80\x9Cattributable\
\ copper pounds\x80\x9D mean that portion of gold or copper produced, sold or\
\ included in proven and probable reserves based on our proportionate ownership,\
\ unless otherwise noted.\nGold General.\nWe had consolidated gold production\
\ from continuing operations of 5.2 million ounces (4.9 million attributable ounces)\
\ in 2016, 5.0 million ounces (4.6 million attributable ounces) in 2015 and 5.2\
\ million ounces (4.7 million attributable ounces) in 2014.\nOf our 2016 consolidated\
\ gold production, approximately 39% came from North America, 14% from South America,\
\ 31% from Asia Pacific and 16% from Africa.\nFor 2016, 2015 and 2014, 96%, 95%\
\ and 95%, respectively, of our Sales were attributable to gold.\nMost of our\
\ Sales come from the sale of refined gold.\nThe end product at our gold operations,\
\ however, is generally doré bars.\nDoré is an alloy consisting primarily of\
\ gold but also containing silver and other metals.\nDoré is sent to refiners\
\ to produce bullion that meets the required markett standard\nOperating Statistics\
\ The following tables detail operating statistics related to gold production,\
\ ounces sold and production costs per ounce of our continuing operations:\n|\
\ | North America | South America 2018 | |\n| Years Ended December 31, | 2018\
\ | 2017 | 2016 | 2017 | 2016 |\n| Tons mined (000 dry short tons): | | | |\
\ | | |\n| Open pit | 230,558 | 252,086 | 218,411 | 99,793 | 104,763 | 104,713\
\ |\n| Underground | 3,024 | 2,979 | 2,864 | — | — | — |\n| Tons processed (000\
\ dry short tons): | | | | | | |\n| Mill | 25,879 | 25,406 | 25,941 | 21,666\
\ | 20,690 | 9,006 |\n| Leach | 46,034 | 55,289 | 45,109 | 25,405 | 24,082 | 30,639\
\ |\n| Average ore grade (oz/ton): | | | | | | |\n| Mill | 0.075 | 0.077\
\ | 0.074 | 0.042 | 0.043 | 0.063 |\n| Leach | 0.017 | 0.020 | 0.019 | 0.013 |\
\ 0.013 | 0.012 |\n| Average mill recovery rate | 76.7% | 76.9% | 78.5% | 88.0%\
\ | 87.2% | 79.4% |\n| Ounces produced -000: | | | | | | |\n| Mill | 1,453\
\ | 1,485 | 1,501 | 802 | 752 | 434 |\n| Leach | 604 | 726 | 523 | 247 | 296 |\
\ 325 |\n| Consolidated | 2,057 | 2,211 | 2,024 | 1,049 | 1,048 | 759 |\n| Attributable\
\ | 2,057 | 2,211 | 2,024 | 671 | 660 | 414 |\n| Consolidated ounces sold -000\
\ | 2,052 | 2,204 | 1,990 | 1,060 | 1,046 | 736 |\n| Production costs per ounce\
\ sold:-1 | | | | | | |\n| Direct mining and production costs | $753 | $706\
\ | $729 | $593 | $639 | $737 |\n| By-product credits | -8 | -9 | -11 | -19 |\
\ -17 | -11 |\n| Royalties and production taxes | 12 | 10 | 15 | 53 | 54 | 38\
\ |\n| Write-downs and inventory change | 2 | 5 | -34 | 33 | 33 | -5 |\n| Costs\
\ applicable to sales | 759 | 712 | 699 | 660 | 709 | 759 |\n| Depreciation and\
\ amortization | 238 | 244 | 207 | 201 | 229 | 404 |\n| Reclamation accretion\
\ | 6 | 6 | 6 | 24 | 45 | 36 |\n| Total production costs | $1,003 | $962 | $912\
\ | $885 | $983 | $1,199 |\n| All-in sustaining costs per ounce sold-2 | $928\
\ | $876 | $854 | $804 | $870 | $932 |\nYears Ended December 31,\n| | Australia\
\ | Africa 2018 | |\n| Years Ended December 31, | 2018 | 2017 | 2016 | 2017 |\
\ 2016 |\n| Tons mined (000 dry short tons): | | | | | | |\n| Open pit |\
\ 103,192 | 114,371 | 126,619 | 71,970 | 74,580 | 75,048 |\n| Underground | 3,202\
\ | 3,144 | 3,279 | 1,339 | 279 | — |\n| Tons milled (000 dry short tons) | 54,337\
\ | 52,802 | 51,606 | 15,585 | 16,884 | 17,289 |\n| Average ore grade (oz/ton)\
\ | 0.032 | 0.035 | 0.037 | 0.058 | 0.053 | 0.052 |\n| Average mill recovery rate\
\ | 87.4% | 86.1% | 86.4% | 92.6% | 92.3% | 91.1% |\n| Ounces produced -000: |\
\ | | | | | |\n| Mill | 1,523 | 1,573 | 1,641 | 850 | 822 | 819 |\n| Consolidated\
\ | 1,523 | 1,573 | 1,641 | 850 | 822 | 819 |\n| Consolidated ounces sold -000\
\ | 1,553 | 1,558 | 1,624 | 851 | 824 | 822 |\n| Production costs per ounce sold:-1\
\ | | | | | | |\n| Direct mining and production costs | $681 | $673 | $605\
\ | $592 | $573 | $553 |\n| By-product credits | -7 | -8 | -7 | -2 | -2 | -2 |\n\
| Royalties and production taxes | 32 | 32 | 32 | 55 | 51 | 50 |\n| Write-downs\
\ and inventory change | 3 | -25 | — | — | 33 | 65 |\n| Costs applicable to sales\
\ | 709 | 672 | 630 | 645 | 655 | 666 |\n| Depreciation and amortization | 133\
\ | 134 | 135 | 301 | 277 | 271 |\n| Reclamation accretion | 8 | 7 | 7 | 9 | 9\
\ | 7 |\n| Total production costs | $850 | $813 | $772 | $955 | $941 | $944 |\n\
| All-in sustaining costs per ounce sold-2 | $845 | $806 | $777 | $794 | $785\
\ | $795 |\nTable of Contents The following performance graph is not \x80\x9C\
soliciting material,\x80\x9D is not deemed filed with the SEC, and is not to be\
\ incorporated by reference into any of Valero\x80\x99s filings under the Securities\
\ Act of 1933 or the Securities Exchange Act of 1934, as amended, respectively.\n\
This performance graph and the related textual information are based on historical\
\ data and are not indicative of future performance.\nThe following line graph\
\ compares the cumulative total return 1 on an investment in our common stock\
\ against the cumulative total return of the S&P 500 Composite Index and an index\
\ of peer companies (that we selected) for the five-year period commencing December\
\ 31, 2006 and ending December 31, 2011."
- "Title: \nText: NOTE 4 Derivatives Derivative Balances Derivatives are entered\
\ into on behalf of customers, for trading, as economic hedges or as qualifying\
\ accounting hedges.\nThe Corporation enters into derivatives to facilitate client\
\ transactions, for principal trading purposes and to manage risk exposures.\n\
For additional information on the Corporation\x80\x99s derivatives and hedging\
\ activities, see Note 1 \x80\x93 Summary of Significant Accounting Principles.\n\
The table below identifies derivative instruments included on the Corporation\x80\
\x99s Consolidated Balance Sheet in derivative assets and liabilities at December\
\ 31, 2010 and 2009.\nBalances are presented on a gross basis, prior to the application\
\ of counterparty and collateral netting.\nTotal derivative assets and liabilities\
\ are adjusted on an aggregate basis to take into consideration the effects of\
\ legally enforceable master netting agreements and have been reduced by the cash\
\ collateral applied.\n| | | December 31, 2010 |\n| | | Gross Derivative Assets\
\ | Gross Derivative Liabilities |\n| | | Trading | | | Trading | | |\n\
| | | Derivatives | | | Derivatives | | |\n| | | and | Qualifying | |\
\ and | Qualifying | |\n| | Contract/ | Economic | Accounting | | Economic\
\ | Accounting | |\n| (Dollars in billions) | Notional-1 | Hedges | Hedges-2\
\ | Total | Hedges | Hedges-2 | Total |\n| Interest rate contracts | | | |\
\ | | | |\n| Swaps | $42,719.2 | $1,193.9 | $14.9 | $1,208.8 | $1,187.9 |\
\ $2.2 | $1,190.1 |\n| Futures and forwards | 9.939.2 | 6.0 | – | 6.0 | 4.7 |\
\ – | 4.7 |\n| Written options | 2,887.7 | – | – | – | 82.8 | – | 82.8 |\n| Purchased\
\ options | 3,026.2 | 88.0 | – | 88.0 | – | – | – |\n| Foreign exchange contracts\
\ | | | | | | | |\n| Swaps | 630.1 | 26.5 | 3.7 | 30.2 | 28.5 | 2.1 | 30.6\
\ |\n| Spot, futures and forwards | 2,652.9 | 41.3 | – | 41.3 | 44.2 | – | 44.2\
\ |\n| Written options | 439.6 | – | – | – | 13.2 | – | 13.2 |\n| Purchased options\
\ | 417.1 | 13.0 | – | 13.0 | – | – | – |\n| Equity contracts | | | | | |\
\ | |\n| Swaps | 42.4 | 1.7 | – | 1.7 | 2.0 | – | 2.0 |\n| Futures and forwards\
\ | 78.8 | 2.9 | – | 2.9 | 2.1 | – | 2.1 |\n| Written options | 242.7 | – | –\
\ | – | 19.4 | – | 19.4 |\n| Purchased options | 193.5 | 21.5 | – | 21.5 | – |\
\ – | – |\n| Commodity contracts | | | | | | | |\n| Swaps | 90.2 | 8.8\
\ | 0.2 | 9.0 | 9.3 | – | 9.3 |\n| Futures and forwards | 413.7 | 4.1 | – | 4.1\
\ | 2.8 | – | 2.8 |\n| Written options | 86.3 | – | – | – | 6.7 | – | 6.7 |\n\
| Purchased options | 84.6 | 6.6 | – | 6.6 | – | – | – |\n| Credit derivatives\
\ | | | | | | | |\n| Purchased credit derivatives: | | | | | | |\
\ |\n| Credit default swaps | 2,184.7 | 69.8 | – | 69.8 | 34.0 | – | 34.0 |\n\
| Total return swaps/other | 26.0 | 0.9 | – | 0.9 | 0.2 | – | 0.2 |\n| Written\
\ credit derivatives: | | | | | | | |\n| Credit default swaps | 2,133.5\
\ | 33.3 | – | 33.3 | 63.2 | – | 63.2 |\n| Total return swaps/other | 22.5 | 0.5\
\ | – | 0.5 | 0.5 | – | 0.5 |\n| Gross derivative assets/liabilities | | $1,518.8\
\ | $18.8 | $1,537.6 | $1,501.5 | $4.3 | $1,505.8 |\n| Less: Legally enforceable\
\ master netting agreements | | | | -1,406.3 | | | -1,406.3 |\n| Less: Cash\
\ collateral applied | | | | -58.3 | | | -43.6 |\n| Total derivative assets/liabilities\
\ | | | | $73.0 | | | $55.9 |\n(1) Represents the total contract/notional\
\ amount of derivative assets and liabilities outstanding.\n(2) Excludes $4.1\
\ billion of long-term debt designated as a hedge of foreign currency risk."
- source_sentence: "Instruct: Given a web search query, retrieve relevant passages\
\ that answer the query.\nQuery: Title: \nText: What does the deferred income\
\ taxes reflect?"
sentences:
- "Title: \nText: 19. INCOME TAXES\nDeferred income taxes reflect the net effect\
\ of temporary differences between the carrying amounts of assets and liabilities\
\ for financial reporting purposes and amounts used for income tax purposes. The\
\ components of our deferred tax assets and liabilities are as follows (in thousands):\n\
As of September 27, 2019, we had $923.4 million of gross federal net operating\
\ loss (\"NOL\") carryforwards consisting of $479.2 million relating to the AppliedMicro\
\ Acquisition, $158.9 million relating to our acquisition of Mindspeed Technologies,\
\ Inc. in 2013, $26.2 million relating to our acquisition of BinOptics Corporation\
\ in 2014 and $259.1 million relating to losses generated by MACOM.\nThe federal\
\ NOL carryforwards will expire at various dates through 2037 for losses generated\
\ prior to the tax period ended September 28, 2018. For losses generated during\
\ the tax period ended September 28, 2018 and future years, the NOL carryforward\
\ period is infinite. The reported net operating loss carryforward includes any\
\ limitation under Sections 382 and 383 of the Internal Revenue Code of 1986,\
\ as amended, which applies to an ownership change as defined under Section 382.\n\
\n | September 27, 2019 |\
\ September 28, 2018\n---------------------------------------------------- | ------------------\
\ | ------------------\nDeferred tax assets (liabilities): |\
\ | \nFederal and foreign net operating losses\
\ and credits | $263,199 | $321,982 \n Intangible assets \
\ | 9,887 | (94,929) \n\
\ Property and equipment | (1,473) | (6,293)\
\ \nOther non-current deferred tax assets | 16,933 \
\ | 13,850 \nDeferred compensation \
\ | — | 3,810 \nDeferred gain \
\ | — | 6,575 \nInterest\
\ | 7,170 | — \
\ \n Valuation allowance | (252,536)\
\ | (243,112) \nTotal deferred tax asset \
\ | $43,180 | $1,883 "
- "Title: \nText: The following table presents the components of impairment and\
\ other charges, net, in each fiscal year (in thousands):\nRestructuring costs\
\ decreased by $2.2 million as a result of lower severance expenses, as our general\
\ and administrative cost reduction initiative came to its conclusion as planned.\
\ Costs of closed restaurants and other increased by $3.8 million, primarily due\
\ to a $3.5 million charge recorded in 2019 related to the write- off of software\
\ development costs associated with a discontinued technology project. Gains on\
\ disposition of property and equipment, net, increased by $7.9 million, primarily\
\ due to a $5.7 million gain related to a sale of property and a$0.8 million gain\
\ related to an eminent domain transaction in 2019.\nRefer to Note 9, Impairment\
\ and Other Charges, Net, of the notes to the consolidated financial statements\
\ for additional information regarding these charges.\n\n \
\ | 2019 | 2018 \n------------------------------------------------------------\
\ | ------- | -------\nRestructuring costs \
\ | $8,455 | $10,647\nCosts of closed restaurants and other \
\ | 8,628 | 4,803 \n(Gains) losses on disposition of property and\
\ equipment, net | (6,244) | 1,627 \nAccelerated depreciation \
\ | 1,616 | 1,130 \nOperating restaurant impairment charges\
\ | — | 211 \n \
\ | $12,455 | $18,418"
- "Title: \nText: MetLife, Inc. Notes to Consolidated Financial Statements \x80\
\x94 (Continued) $4.3 billion, of which $1.6 billion is deductible for income\
\ tax purposes.\nFurther information on goodwill is described in Note 6.\nSee\
\ Note 5 for the VOBA acquired as part of the acquisition and Note 7 for the value\
\ of distribution agreements (\x80\x9CVODA\x80\x9D) and the value of customer\
\ relationships acquired (\x80\x9CVOCRA\x80\x9D).\nAs part of the integration\
\ of Travelers\x80\x99 operations, management approved and initiated plans to\
\ reduce approximately 1,000 domestic and international Travelers positions, which\
\ was completed in December 2006.\nMetLife initially recorded restructuring costs,\
\ including severance, relocation and outplacement services of Travelers\x80\x99\
\ employees, as liabilities assumed in the purchase business combination of $49\
\ million.\nFor the years ended December 31, 2006 and 2005, the liability for\
\ restructuring costs was reduced by $4 million and $1 million, respectively,\
\ due to a reduction in the estimate of severance benefits to be paid to Travelers\
\ employees.\nThe restructuring costs associated with the Travelers acquisition\
\ were as follows:\n| | Years Ended December 31, |\n| | 2006 | 2005 |\n| |\
\ (In millions) |\n| Balance at January 1, | $28 | $— |\n| Acquisition | — | 49\
\ |\n| Cash payments | -24 | -20 |\n| Other reductions | -4 | -1 |\n| Balance\
\ at December 31, | $— | $28 |\nOther Acquisitions and Dispositions On June 28,\
\ 2007, the Company acquired the remaining 50% interest in a joint venture in\
\ Hong Kong, MetLife Fubon Limited (\x80\x9CMetLife Fubon\x80\x9D), for $56 million\
\ in cash, resulting in MetLife Fubon becoming a consolidated subsidiary of the\
\ Company.\nThe transaction was treated as a step acquisition, and at June 30,\
\ 2007, total assets and liabilities of MetLife Fubon of $839 million and $735\
\ million, respectively, were included in the Company\x80\x99s consolidated balance\
\ sheet.\nThe Company\x80\x99s investment for the initial 50% interest in MetLife\
\ Fubon was $48 million.\nThe Company used the equity method of accounting for\
\ such investment in MetLife Fubon.\nThe Company\x80\x99s share of the joint venture\x80\
\x99s results for the six months ended June 30, 2007, was a loss of $3 million.\n\
The fair value of the assets acquired and the liabilities assumed in the step\
\ acquisition at June 30, 2007, was $427 million and $371 million, respectively.\n\
No additional goodwill was recorded as a part of the step acquisition.\nAs a result\
\ of this acquisition, additional VOBA and VODA of $45 million and $5 million,\
\ respectively, were recorded and both have a weighted average amortization period\
\ of 16 years.\nFurther information on VOBA and VODA is described in Note 5 and\
\ Note 7, respectively.\nOn June 1, 2007, the Company completed the sale of its\
\ Bermuda insurance subsidiary, MetLife International Insurance, Ltd. (\x80\x9C\
MLII\x80\x9D), to a third party for $33 million in cash consideration, resulting\
\ in a gain upon disposal of $3 million, net of income tax.\nThe net assets of\
\ MLII at disposal were $27 million.\nA liability of $1 million was recorded with\
\ respect to a guarantee provided in connection with this disposition.\nFurther\
\ information on guarantees is described in Note 16.\nOn September 1, 2005, the\
\ Company completed the acquisition of CitiStreet Associates, a division of CitiStreet\
\ LLC, which is primarily involved in the distribution of annuity products and\
\ retirement plans to the education, healthcare, and not-for-profit markets, for\
\ $56 million, of which $2 million was allocated to goodwill and $54 million to\
\ other identifiable intangibles, specifically the value of customer relationships\
\ acquired, which have a weighted average amortization period of 16 years.\nCitiStreet\
\ Associates was integrated with MetLife Resources, a focused distribution channel\
\ of MetLife, which is dedicated to provide retirement plans and financial services\
\ to the same markets.\nFurther information on goodwill and VOCRA is described\
\ in Note 6 and Note 7, respectively.\nSee Note 23 for information on the disposition\
\ of the annuities and pension businesses of MetLife Insurance Limited (\x80\x9C\
MetLife Australia\x80\x9D), P. T. Sejahtera (\x80\x9CMetLife Indonesia\x80\x9D\
) and SSRM Holdings, Inc. (\x80\x9CSSRM\x80\x9D).\nSee Note 25 for information\
\ on the Company\x80\x99s acquisitions subsequent to December 31, 2007."
- source_sentence: "Instruct: Given a web search query, retrieve relevant passages\
\ that answer the query.\nQuery: Title: \nText: In the year with lowest amount\
\ of Deposits with banks Average volume, what's the increasing rate of Deposits\
\ with banks Average volume?"
sentences:
- "Title: \nText: Issuer Purchases of Equity Securities Repurchases of common stock\
\ are made to support the Company\x80\x99s stock-based employee compensation plans\
\ and for other corporate purposes.\nOn February 13, 2006, the Board of Directors\
\ authorized the purchase of $2.0 billion of the Company\x80\x99s common stock\
\ between February 13, 2006 and February 28, 2007.\nIn August 2006, 3M\x80\x99\
s Board of Directors authorized the repurchase of an additional $1.0 billion in\
\ share repurchases, raising the total authorization to $3.0 billion for the period\
\ from February 13, 2006 to February 28, 2007.\nIn February 2007, 3M\x80\x99s\
\ Board of Directors authorized a twoyear share repurchase of up to $7.0 billion\
\ for the period from February 12, 2007 to February 28, 2009."
- "Title: \nText: Additional Interest Rate Details Average Balances and Interest\
\ Ratesé\x88¥æ\x93\x9Cssets(1)(2)(3)(4)\n| | Average volume | Interest revenue\
\ | % Average rate |\n| In millions of dollars, except rates | 2015 | 2014 | 2013\
\ | 2015 | 2014 | 2013 | 2015 | 2014 | 2013 |\n| Assets | | | | | | | |\
\ | |\n| Deposits with banks-5 | $133,790 | $161,359 | $144,904 | $727 | $959\
\ | $1,026 | 0.54% | 0.59% | 0.71% |\n| Federal funds sold and securities borrowed\
\ or purchased under agreements to resell-6 | | | | | | | | | |\n| In\
\ U.S. offices | $150,359 | $153,688 | $158,237 | $1,211 | $1,034 | $1,133 | 0.81%\
\ | 0.67% | 0.72% |\n| In offices outside the U.S.-5 | 84,006 | 101,177 | 109,233\
\ | 1,305 | 1,332 | 1,433 | 1.55 | 1.32 | 1.31 |\n| Total | $234,365 | $254,865\
\ | $267,470 | $2,516 | $2,366 | $2,566 | 1.07% | 0.93% | 0.96% |\n| Trading account\
\ assets-7(8) | | | | | | | | | |\n| In U.S. offices | $114,639 | $114,910\
\ | $126,123 | $3,945 | $3,472 | $3,728 | 3.44% | 3.02% | 2.96% |\n| In offices\
\ outside the U.S.-5 | 103,348 | 119,801 | 127,291 | 2,141 | 2,538 | 2,683 | 2.07\
\ | 2.12 | 2.11 |\n| Total | $217,987 | $234,711 | $253,414 | $6,086 | $6,010\
\ | $6,411 | 2.79% | 2.56% | 2.53% |\n| Investments | | | | | | | | |\
\ |\n| In U.S. offices | | | | | | | | | |\n| Taxable | $214,714 | $188,910\
\ | $174,084 | $3,812 | $3,286 | $2,713 | 1.78% | 1.74% | 1.56% |\n| Exempt from\
\ U.S. income tax | 20,034 | 20,386 | 18,075 | 443 | 626 | 811 | 2.21 | 3.07 |\
\ 4.49 |\n| In offices outside the U.S.-5 | 102,376 | 113,163 | 114,122 | 3,071\
\ | 3,627 | 3,761 | 3.00 | 3.21 | 3.30 |\n| Total | $337,124 | $322,459 | $306,281\
\ | $7,326 | $7,539 | $7,285 | 2.17% | 2.34% | 2.38% |\n| Loans (net of unearned\
\ income)(9) | | | | | | | | | |\n| In U.S. offices | $354,439 | $361,769\
\ | $354,707 | $24,558 | $26,076 | $25,941 | 6.93% | 7.21% | 7.31% |\n| In offices\
\ outside the U.S.-5 | 273,072 | 296,656 | 292,852 | 15,988 | 18,723 | 19,660\
\ | 5.85 | 6.31 | 6.71 |\n| Total | $627,511 | $658,425 | $647,559 | $40,546 |\
\ $44,799 | $45,601 | 6.46% | 6.80% | 7.04% |\n| Other interest-earning assets-10\
\ | $55,060 | $40,375 | $38,233 | $1,839 | $507 | $602 | 3.34% | 1.26% | 1.57%\
\ |\n| Total interest-earning assets | $1,605,837 | $1,672,194 | $1,657,861 |\
\ $59,040 | $62,180 | $63,491 | 3.68% | 3.72% | 3.83% |\n| Non-interest-earning\
\ assets-7 | $218,000 | $224,721 | $222,526 | | | | | | |\n| Total assets\
\ from discontinued operations | — | — | 2,909 | | | | | | |\n| Total assets\
\ | $1,823,837 | $1,896,915 | $1,883,296 | | | | | | |\nNet interest revenue\
\ includes the taxable equivalent adjustments related to the tax-exempt bond portfolio\
\ (based on the U. S. federal statutory tax rate of 35%) of $487 million, $498\
\ million and $521 million for 2015, 2014 and 2013, respectively.\nInterest rates\
\ and amounts include the effects of risk management activities associated with\
\ the respective asset categories.\nMonthly or quarterly averages have been used\
\ by certain subsidiaries where daily averages are unavailable.\nDetailed average\
\ volume, Interest revenue and Interest expense exclude Discontinued operations.\n\
See Note 2 to the Consolidated Financial Statements.\nAverage rates reflect prevailing\
\ local interest rates, including inflationary effects and monetary corrections\
\ in certain countries.\nAverage volumes of securities borrowed or purchased under\
\ agreements to resell are reported net pursuant to ASC 210-20-45.\nHowever, Interest\
\ revenue excludes the impact of ASC 210-20-45.\nThe fair value carrying amounts\
\ of derivative contracts are reported net, pursuant to ASC 815-10-45, in Non-interest-earning\
\ assets and Other non-interest bearing liabilities.\nInterest expense on Trading\
\ account liabilities of ICG is reported as a reduction of Interest revenue.\n\
Interest revenue and Interest expense on cash collateral positions are reported\
\ in interest on Trading account assets and Trading account liabilities, respectively.\n\
Includes cash-basis loans.\nIncludes brokerage receivables.\nDuring 2015, continued\
\ management actions, primarily the sale or transfer to held-for-sale of approximately\
\ $1.5 billion of delinquent residential first mortgages, including $0.9 billion\
\ in the fourth quarter largely associated with the transfer of CitiFinancial\
\ loans to held-for-sale referenced above, were the primary driver of the overall\
\ improvement in delinquencies within Citi Holdings\x80\x99 residential first\
\ mortgage portfolio.\nCredit performance from quarter to quarter could continue\
\ to be impacted by the amount of delinquent loan sales or transfers to held-for-sale,\
\ as well as overall trends in HPI and interest rates.\nNorth America Residential\
\ First Mortgages\x80\x94State Delinquency Trends The following tables set forth\
\ the six U. S. states and/or regions with the highest concentration of Citi\x80\
\x99s residential first mortgages.\n| In billions of dollars | December 31, 2015\
\ | December 31, 2014 |\n| State-1 | ENR-2 | ENRDistribution | 90+DPD% | %LTV\
\ >100%-3 | RefreshedFICO | ENR-2 | ENRDistribution | 90+DPD% | %LTV >100%-3 |\
\ RefreshedFICO |\n| CA | $19.2 | 37% | 0.2% | 1% | 754 | $18.9 | 31% | 0.6% |\
\ 2% | 745 |\n| NY/NJ/CT-4 | 12.7 | 25 | 0.8 | 1 | 751 | 12.2 | 20 | 1.9 | 2 |\
\ 740 |\n| VA/MD | 2.2 | 4 | 1.2 | 2 | 719 | 3.0 | 5 | 3.0 | 8 | 695 |\n| IL-4\
\ | 2.2 | 4 | 1.0 | 3 | 735 | 2.5 | 4 | 2.5 | 9 | 713 |\n| FL-4 | 2.2 | 4 | 1.1\
\ | 4 | 723 | 2.8 | 5 | 3.0 | 14 | 700 |\n| TX | 1.9 | 4 | 1.0 | — | 711 | 2.5\
\ | 4 | 2.7 | — | 680 |\n| Other | 11.0 | 21 | 1.3 | 2 | 710 | 18.2 | 30 | 3.3\
\ | 7 | 677 |\n| Total-5 | $51.5 | 100% | 0.7% | 1% | 738 | $60.1 | 100% | 2.1%\
\ | 4% | 715 |\nNote: Totals may not sum due to rounding.\n(1) Certain of the\
\ states are included as part of a region based on Citi\x80\x99s view of similar\
\ HPI within the region.\n(2) Ending net receivables.\nExcludes loans in Canada\
\ and Puerto Rico, loans guaranteed by U. S. government agencies, loans recorded\
\ at fair value and loans subject to long term standby commitments (LTSCs).\n\
Excludes balances for which FICO or LTV data are unavailable.\n(3) LTV ratios\
\ (loan balance divided by appraised value) are calculated at origination and\
\ updated by applying market price data.\n(4) New York, New Jersey, Connecticut,\
\ Florida and Illinois are judicial states.\n(5) Improvement in state trends during\
\ 2015 was primarily due to the sale or transfer to held-for-sale of residential\
\ first mortgages, including the transfer of CitiFinancial residential first mortgages\
\ to held-for-sale in the fourth quarter of 2015.\nForeclosures A substantial\
\ majority of Citi\x80\x99s foreclosure inventory consists of residential first\
\ mortgages.\nAt December 31, 2015, Citi\x80\x99s foreclosure inventory included\
\ approximately $0.1 billion, or 0.2%, of the total residential first mortgage\
\ portfolio, compared to $0.6 billion, or 0.9%, at December 31, 2014, based on\
\ the dollar amount of ending net receivables of loans in foreclosure inventory,\
\ excluding loans that are guaranteed by U. S. government agencies and loans\
\ subject to LTSCs.\nNorth America Consumer Mortgage Quarterly Credit Trends \x80\
\x94Net Credit Losses and Delinquencies\x80\x94Home Equity Loans Citi\x80\x99\
s home equity loan portfolio consists of both fixed-rate home equity loans and\
\ loans extended under home equity lines of credit.\nFixed-rate home equity loans\
\ are fully amortizing.\nHome equity lines of credit allow for amounts to be drawn\
\ for a period of time with the payment of interest only and then, at the end\
\ of the draw period, the then-outstanding amount is converted to an amortizing\
\ loan (the interest-only payment feature during the revolving period is standard\
\ for this product across the industry).\nAfter conversion, the home equity loans\
\ typically have a 20-year amortization period.\nAs of December 31, 2015, Citi\x80\
\x99s home equity loan portfolio of $22.8 billion consisted of $6.3 billion of\
\ fixed-rate home equity loans and $16.5 billion of loans extended under home\
\ equity lines of credit (Revolving HELOCs)."
- "Title: \nText: | (Dollar amounts in thousands) | Rate | | Principal Amount of\
\ Subordinated Note/ Debenture Issued to Trust -1 | Investment in Unconsolidated\
\ Subsidiary -2 |\n| Huntington Capital I | 0.99 | -3 | $138,816 | $6,186 |\n\
| Huntington Capital II | 0.93 | -4 | 60,093 | 3,093 |\n| Huntington Capital III\
\ | 6.69 | | 114,072 | 10 |\n| BancFirst Ohio Trust Preferred | 8.54 | | 23,248\
\ | 619 |\n| Sky Financial Capital Trust I | 8.52 | | 64,474 | 1,856 |\n| Sky\
\ Financial Capital Trust II | 3.52 | -5 | 30,929 | 929 |\n| Sky Financial Capital\
\ Trust III | 1.28 | -6 | 77,481 | 2,320 |\n| Sky Financial Capital Trust IV |\
\ 1.27 | -6 | 77,482 | 2,320 |\n| Prospect Trust I | 3.54 | -7 | 6,186 | 186 |\n\
| Total | | | $592,781 | $17,519 |\n(1) Represents the principal amount of debentures\
\ issued to each trust, including unamortized original issue discount.\n(2) Huntington\x80\
\x99s investment in the unconsolidated trusts represents the only risk of loss.\n\
(3) Variable effective rate at December 31, 2010, based on three month LIBOR +\
\ 0.70.\n(4) Variable effective rate at December 31, 2010, based on three month\
\ LIBOR + 0.625.\n(5) Variable effective rate at December 31, 2010, based on three\
\ month LIBOR + 2.95.\n(6) Variable effective rate at December 31, 2010, based\
\ on three month LIBOR + 1.40.\n(7) Variable effective rate at December 31, 2010,\
\ based on three month LIBOR + 3.25.\nEach issue of the junior subordinated debentures\
\ has an interest rate equal to the corresponding trust securities distribution\
\ rate.\nHuntington has the right to defer payment of interest on the debentures\
\ at any time, or from time to time for a period not exceeding five years, provided\
\ that no extension period may extend beyond the stated maturity of the related\
\ debentures.\nDuring any such extension period, distributions to the trust securities\
\ will also be deferred and Huntington\x80\x99s ability to pay dividends on its\
\ common stock will be restricted.\nPeriodic cash payments and payments upon liquidation\
\ or redemption with respect to trust securities are guaranteed by Huntington\
\ to the extent of funds held by the trusts.\nThe guarantee ranks subordinate\
\ and junior in right of payment to all indebtedness of the Company to the same\
\ extent as the junior subordinated debt.\nThe guarantee does not place a limitation\
\ on the amount of additional indebtedness that may be incurred by Huntington.\n\
Low Income Housing Tax Credit Partnerships Huntington makes certain equity investments\
\ in various limited partnerships that sponsor affordable housing projects utilizing\
\ the Low Income Housing Tax Credit pursuant to Section 42 of the Internal Revenue\
\ Code.\nThe purpose of these investments is to achieve a satisfactory return\
\ on capital, to facilitate the sale of additional affordable housing product\
\ offerings, and to assist in achieving goals associated with the Community Reinvestment\
\ Act.\nThe primary activities of the limited partnerships include the identification,\
\ development, and operation of multi-family housing that is leased to qualifying\
\ residential tenants.\nGenerally, these types of investments are funded through\
\ a combination of debt and equity.\nHuntington does not own a majority of the\
\ limited partnership interests in these entities and is not the primary beneficiary.\n\
Huntington uses the equity method to account for the majority of its investments\
\ in these entities.\nThese investments are included in accrued income and other\
\ assets.\nAt December 31, 2010 and 2009, Huntington has commitments of $316.0\
\ million and $285.3 million, respectively, of which $260.1 million and"
model-index:
- name: SentenceTransformer based on thomaskim1130/stella_en_400M_v5-FinanceRAG
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: Evaluate
type: Evaluate
metrics:
- type: cosine_accuracy@1
value: 0.46359223300970875
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.6820388349514563
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.7597087378640777
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.8519417475728155
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.46359223300970875
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.25647249190938515
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.17766990291262133
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.10242718446601942
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.4095469255663431
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.6423948220064724
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.7298543689320389
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.8398462783171521
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.6409313886654548
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.5902248035136388
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.5753196287486457
name: Cosine Map@100
- type: dot_accuracy@1
value: 0.4393203883495146
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.6747572815533981
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.7354368932038835
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.8422330097087378
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.4393203883495146
name: Dot Precision@1
- type: dot_precision@3
value: 0.25
name: Dot Precision@3
- type: dot_precision@5
value: 0.17087378640776701
name: Dot Precision@5
- type: dot_precision@10
value: 0.09975728155339807
name: Dot Precision@10
- type: dot_recall@1
value: 0.3828478964401295
name: Dot Recall@1
- type: dot_recall@3
value: 0.6338187702265372
name: Dot Recall@3
- type: dot_recall@5
value: 0.7005258899676375
name: Dot Recall@5
- type: dot_recall@10
value: 0.8223705501618123
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.6194906173849263
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.5711877793188473
name: Dot Mrr@10
- type: dot_map@100
value: 0.5528301778009912
name: Dot Map@100
---
# SentenceTransformer based on thomaskim1130/stella_en_400M_v5-FinanceRAG
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [thomaskim1130/stella_en_400M_v5-FinanceRAG](https://huggingface.co/thomaskim1130/stella_en_400M_v5-FinanceRAG). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [thomaskim1130/stella_en_400M_v5-FinanceRAG](https://huggingface.co/thomaskim1130/stella_en_400M_v5-FinanceRAG) <!-- at revision f10dfc9df857453911fda9d3af68b25bb47dd353 -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 1024 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: NewModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Dense({'in_features': 1024, 'out_features': 1024, 'bias': True, 'activation_function': 'torch.nn.modules.linear.Identity'})
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
"Instruct: Given a web search query, retrieve relevant passages that answer the query.\nQuery: Title: \nText: In the year with lowest amount of Deposits with banks Average volume, what's the increasing rate of Deposits with banks Average volume?",
'Title: \nText: Additional Interest Rate Details Average Balances and Interest Ratesé\x88¥æ\x93\x9cssets(1)(2)(3)(4)\n| | Average volume | Interest revenue | % Average rate |\n| In millions of dollars, except rates | 2015 | 2014 | 2013 | 2015 | 2014 | 2013 | 2015 | 2014 | 2013 |\n| Assets | | | | | | | | | |\n| Deposits with banks-5 | $133,790 | $161,359 | $144,904 | $727 | $959 | $1,026 | 0.54% | 0.59% | 0.71% |\n| Federal funds sold and securities borrowed or purchased under agreements to resell-6 | | | | | | | | | |\n| In U.S. offices | $150,359 | $153,688 | $158,237 | $1,211 | $1,034 | $1,133 | 0.81% | 0.67% | 0.72% |\n| In offices outside the U.S.-5 | 84,006 | 101,177 | 109,233 | 1,305 | 1,332 | 1,433 | 1.55 | 1.32 | 1.31 |\n| Total | $234,365 | $254,865 | $267,470 | $2,516 | $2,366 | $2,566 | 1.07% | 0.93% | 0.96% |\n| Trading account assets-7(8) | | | | | | | | | |\n| In U.S. offices | $114,639 | $114,910 | $126,123 | $3,945 | $3,472 | $3,728 | 3.44% | 3.02% | 2.96% |\n| In offices outside the U.S.-5 | 103,348 | 119,801 | 127,291 | 2,141 | 2,538 | 2,683 | 2.07 | 2.12 | 2.11 |\n| Total | $217,987 | $234,711 | $253,414 | $6,086 | $6,010 | $6,411 | 2.79% | 2.56% | 2.53% |\n| Investments | | | | | | | | | |\n| In U.S. offices | | | | | | | | | |\n| Taxable | $214,714 | $188,910 | $174,084 | $3,812 | $3,286 | $2,713 | 1.78% | 1.74% | 1.56% |\n| Exempt from U.S. income tax | 20,034 | 20,386 | 18,075 | 443 | 626 | 811 | 2.21 | 3.07 | 4.49 |\n| In offices outside the U.S.-5 | 102,376 | 113,163 | 114,122 | 3,071 | 3,627 | 3,761 | 3.00 | 3.21 | 3.30 |\n| Total | $337,124 | $322,459 | $306,281 | $7,326 | $7,539 | $7,285 | 2.17% | 2.34% | 2.38% |\n| Loans (net of unearned income)(9) | | | | | | | | | |\n| In U.S. offices | $354,439 | $361,769 | $354,707 | $24,558 | $26,076 | $25,941 | 6.93% | 7.21% | 7.31% |\n| In offices outside the U.S.-5 | 273,072 | 296,656 | 292,852 | 15,988 | 18,723 | 19,660 | 5.85 | 6.31 | 6.71 |\n| Total | $627,511 | $658,425 | $647,559 | $40,546 | $44,799 | $45,601 | 6.46% | 6.80% | 7.04% |\n| Other interest-earning assets-10 | $55,060 | $40,375 | $38,233 | $1,839 | $507 | $602 | 3.34% | 1.26% | 1.57% |\n| Total interest-earning assets | $1,605,837 | $1,672,194 | $1,657,861 | $59,040 | $62,180 | $63,491 | 3.68% | 3.72% | 3.83% |\n| Non-interest-earning assets-7 | $218,000 | $224,721 | $222,526 | | | | | | |\n| Total assets from discontinued operations | — | — | 2,909 | | | | | | |\n| Total assets | $1,823,837 | $1,896,915 | $1,883,296 | | | | | | |\nNet interest revenue includes the taxable equivalent adjustments related to the tax-exempt bond portfolio (based on the U. S. federal statutory tax rate of 35%) of $487 million, $498 million and $521 million for 2015, 2014 and 2013, respectively.\nInterest rates and amounts include the effects of risk management activities associated with the respective asset categories.\nMonthly or quarterly averages have been used by certain subsidiaries where daily averages are unavailable.\nDetailed average volume, Interest revenue and Interest expense exclude Discontinued operations.\nSee Note 2 to the Consolidated Financial Statements.\nAverage rates reflect prevailing local interest rates, including inflationary effects and monetary corrections in certain countries.\nAverage volumes of securities borrowed or purchased under agreements to resell are reported net pursuant to ASC 210-20-45.\nHowever, Interest revenue excludes the impact of ASC 210-20-45.\nThe fair value carrying amounts of derivative contracts are reported net, pursuant to ASC 815-10-45, in Non-interest-earning assets and Other non-interest bearing liabilities.\nInterest expense on Trading account liabilities of ICG is reported as a reduction of Interest revenue.\nInterest revenue and Interest expense on cash collateral positions are reported in interest on Trading account assets and Trading account liabilities, respectively.\nIncludes cash-basis loans.\nIncludes brokerage receivables.\nDuring 2015, continued management actions, primarily the sale or transfer to held-for-sale of approximately $1.5 billion of delinquent residential first mortgages, including $0.9 billion in the fourth quarter largely associated with the transfer of CitiFinancial loans to held-for-sale referenced above, were the primary driver of the overall improvement in delinquencies within Citi Holdings\x80\x99 residential first mortgage portfolio.\nCredit performance from quarter to quarter could continue to be impacted by the amount of delinquent loan sales or transfers to held-for-sale, as well as overall trends in HPI and interest rates.\nNorth America Residential First Mortgages\x80\x94State Delinquency Trends The following tables set forth the six U. S. states and/or regions with the highest concentration of Citi\x80\x99s residential first mortgages.\n| In billions of dollars | December 31, 2015 | December 31, 2014 |\n| State-1 | ENR-2 | ENRDistribution | 90+DPD% | %LTV >100%-3 | RefreshedFICO | ENR-2 | ENRDistribution | 90+DPD% | %LTV >100%-3 | RefreshedFICO |\n| CA | $19.2 | 37% | 0.2% | 1% | 754 | $18.9 | 31% | 0.6% | 2% | 745 |\n| NY/NJ/CT-4 | 12.7 | 25 | 0.8 | 1 | 751 | 12.2 | 20 | 1.9 | 2 | 740 |\n| VA/MD | 2.2 | 4 | 1.2 | 2 | 719 | 3.0 | 5 | 3.0 | 8 | 695 |\n| IL-4 | 2.2 | 4 | 1.0 | 3 | 735 | 2.5 | 4 | 2.5 | 9 | 713 |\n| FL-4 | 2.2 | 4 | 1.1 | 4 | 723 | 2.8 | 5 | 3.0 | 14 | 700 |\n| TX | 1.9 | 4 | 1.0 | — | 711 | 2.5 | 4 | 2.7 | — | 680 |\n| Other | 11.0 | 21 | 1.3 | 2 | 710 | 18.2 | 30 | 3.3 | 7 | 677 |\n| Total-5 | $51.5 | 100% | 0.7% | 1% | 738 | $60.1 | 100% | 2.1% | 4% | 715 |\nNote: Totals may not sum due to rounding.\n(1) Certain of the states are included as part of a region based on Citi\x80\x99s view of similar HPI within the region.\n(2) Ending net receivables.\nExcludes loans in Canada and Puerto Rico, loans guaranteed by U. S. government agencies, loans recorded at fair value and loans subject to long term standby commitments (LTSCs).\nExcludes balances for which FICO or LTV data are unavailable.\n(3) LTV ratios (loan balance divided by appraised value) are calculated at origination and updated by applying market price data.\n(4) New York, New Jersey, Connecticut, Florida and Illinois are judicial states.\n(5) Improvement in state trends during 2015 was primarily due to the sale or transfer to held-for-sale of residential first mortgages, including the transfer of CitiFinancial residential first mortgages to held-for-sale in the fourth quarter of 2015.\nForeclosures A substantial majority of Citi\x80\x99s foreclosure inventory consists of residential first mortgages.\nAt December 31, 2015, Citi\x80\x99s foreclosure inventory included approximately $0.1 billion, or 0.2%, of the total residential first mortgage portfolio, compared to $0.6 billion, or 0.9%, at December 31, 2014, based on the dollar amount of ending net receivables of loans in foreclosure inventory, excluding loans that are guaranteed by U. S. government agencies and loans subject to LTSCs.\nNorth America Consumer Mortgage Quarterly Credit Trends \x80\x94Net Credit Losses and Delinquencies\x80\x94Home Equity Loans Citi\x80\x99s home equity loan portfolio consists of both fixed-rate home equity loans and loans extended under home equity lines of credit.\nFixed-rate home equity loans are fully amortizing.\nHome equity lines of credit allow for amounts to be drawn for a period of time with the payment of interest only and then, at the end of the draw period, the then-outstanding amount is converted to an amortizing loan (the interest-only payment feature during the revolving period is standard for this product across the industry).\nAfter conversion, the home equity loans typically have a 20-year amortization period.\nAs of December 31, 2015, Citi\x80\x99s home equity loan portfolio of $22.8 billion consisted of $6.3 billion of fixed-rate home equity loans and $16.5 billion of loans extended under home equity lines of credit (Revolving HELOCs).',
'Title: \nText: Issuer Purchases of Equity Securities Repurchases of common stock are made to support the Company\x80\x99s stock-based employee compensation plans and for other corporate purposes.\nOn February 13, 2006, the Board of Directors authorized the purchase of $2.0 billion of the Company\x80\x99s common stock between February 13, 2006 and February 28, 2007.\nIn August 2006, 3M\x80\x99s Board of Directors authorized the repurchase of an additional $1.0 billion in share repurchases, raising the total authorization to $3.0 billion for the period from February 13, 2006 to February 28, 2007.\nIn February 2007, 3M\x80\x99s Board of Directors authorized a twoyear share repurchase of up to $7.0 billion for the period from February 12, 2007 to February 28, 2009.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `Evaluate`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.4636 |
| cosine_accuracy@3 | 0.682 |
| cosine_accuracy@5 | 0.7597 |
| cosine_accuracy@10 | 0.8519 |
| cosine_precision@1 | 0.4636 |
| cosine_precision@3 | 0.2565 |
| cosine_precision@5 | 0.1777 |
| cosine_precision@10 | 0.1024 |
| cosine_recall@1 | 0.4095 |
| cosine_recall@3 | 0.6424 |
| cosine_recall@5 | 0.7299 |
| cosine_recall@10 | 0.8398 |
| cosine_ndcg@10 | 0.6409 |
| cosine_mrr@10 | 0.5902 |
| **cosine_map@100** | **0.5753** |
| dot_accuracy@1 | 0.4393 |
| dot_accuracy@3 | 0.6748 |
| dot_accuracy@5 | 0.7354 |
| dot_accuracy@10 | 0.8422 |
| dot_precision@1 | 0.4393 |
| dot_precision@3 | 0.25 |
| dot_precision@5 | 0.1709 |
| dot_precision@10 | 0.0998 |
| dot_recall@1 | 0.3828 |
| dot_recall@3 | 0.6338 |
| dot_recall@5 | 0.7005 |
| dot_recall@10 | 0.8224 |
| dot_ndcg@10 | 0.6195 |
| dot_mrr@10 | 0.5712 |
| dot_map@100 | 0.5528 |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 2,256 training samples
* Columns: <code>sentence_0</code> and <code>sentence_1</code>
* Approximate statistics based on the first 1000 samples:
| | sentence_0 | sentence_1 |
|:--------|:------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 28 tokens</li><li>mean: 45.02 tokens</li><li>max: 114 tokens</li></ul> | <ul><li>min: 23 tokens</li><li>mean: 406.36 tokens</li><li>max: 512 tokens</li></ul> |
* Samples:
| sentence_0 | sentence_1 |
|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>Instruct: Given a web search query, retrieve relevant passages that answer the query.<br>Query: Title: <br>Text: What do all Notional sum up, excluding those negative ones in 2008 for As of December 31, 2008 for Financial assets with interest rate risk? (in million)</code> | <code>Title: <br>Text: Cash Flows Our estimated future benefit payments for funded and unfunded plans are as follows (in millions):<br>1 The expected benefit payments for our other postretirement benefit plans are net of estimated federal subsidies expected to be received under the Medicare Prescription Drug, Improvement and Modernization Act of 2003.<br>Federal subsidies are estimated to be $3 million for the period 2019-2023 and $2 million for the period 2024-2028.<br>The Company anticipates making pension contributions in 2019 of $32 million, all of which will be allocated to our international plans.<br>The majority of these contributions are required by funding regulations or law.</code> |
| <code>Instruct: Given a web search query, retrieve relevant passages that answer the query.<br>Query: Title: <br>Text: what's the total amount of No surrender charge of 2010 Individual Fixed Annuities, Change in cash of 2008, and Total reserves of 2010 Individual Variable Annuities ?</code> | <code>Title: <br>Text: 2010 and 2009 Comparison Surrender rates have improved compared to the prior year for group retirement products, individual fixed annuities and individual variable annuities as surrenders have returned to more normal levels.<br>Surrender rates for individual fixed annuities have decreased significantly in 2010 due to the low interest rate environment and the relative competitiveness of interest credited rates on the existing block of fixed annuities versus interest rates on alternative investment options available in the marketplace.<br>Surrender rates for group retirement products are expected to increase in 2011 as certain large group surrenders are anticipated.2009 and 2008 Comparison Surrenders and other withdrawals increased in 2009 for group retirement products primarily due to higher large group surrenders.<br>However, surrender rates and withdrawals have improved for individual fixed annuities and individual variable annuities.<br>The following table presents reserves by surrender charge category and surrender rates:<br>| | 2010 | 2009 |<br>| At December 31,(in millions) | Group Retirement Products* | Individual Fixed Annuities | Individual Variable Annuities | Group Retirement Products* | Individual Fixed Annuities | Individual Variable Annuities |<br>| No surrender charge | $52,742 | $14,006 | $11,859 | $47,854 | $11,444 | $11,161 |<br>| 0% - 2% | 1,292 | 3,510 | 4,083 | 1,509 | 3,054 | 4,094 |<br>| Greater than 2% - 4% | 1,754 | 5,060 | 2,040 | 1,918 | 5,635 | 2,066 |<br>| Greater than 4% | 2,753 | 22,777 | 7,361 | 3,213 | 23,885 | 6,758 |<br>| Non-Surrenderable | 792 | 3,136 | 238 | 850 | 3,184 | 558 |<br>| Total reserves | $59,333 | $48,489 | $25,581 | $55,344 | $47,202 | $24,637 |<br>| Surrender rates | 10.3% | 7.4% | 11.4% | 12.3% | 14.4% | 12.1% |<br>* Excludes mutual funds of $9.0 billion and $8.1 billion in 2010 and 2009, respectively.<br>Financial Services Operations AIGs Financial Services subsidiaries engage in diversified activities including commercial aircraft leasing and the remaining Capital Markets portfolios, which are conducted through ILFC and AIGFP, respectively.<br>Following the classification of AGF as discontinued operations in the third quarter of 2010 (see Note 4 to the Consolidated Financial Statements), AIGs remaining consumer finance businesses are now reported in AIGs Other operations category as part of Divested businesses.<br>As discussed in Note 3 to the Consolidated Financial Statements, in order to align financial reporting with changes made during the third quarter of 2010 to the manner in which AIGs chief operating decision makers review the businesses to make decisions about resources to be allocated and to assess performance, changes were made to AIGs segment information.<br>During the third quarter of 2010, AIGs Asset Management Group undertook the management responsibilities for non-derivative assets and liabilities of the Capital Markets businesses of the Financial Services segment.<br>These assets and liabilities are being managed on a spread basis, in concert with the MIP.<br>Accordingly, gains and losses related to these assets and liabilities, primarily consisting of credit valuation adjustment gains and losses are reported in AIGs Other operations category as part of Asset Management Direct Investment business.<br>Also, intercompany interest related to loans from AIG Funding Inc. (AIG Funding) to AIGFP is no longer being allocated to Capital Markets from Other operations.<br>The remaining Capital Markets derivatives business continues to be reported in the Financial Services segment as part of Capital Markets results.<br>American International Group, Inc. , and Subsidiaries solely for illustrative purposes.<br>The selection of these specific events should not be construed as a prediction, but only as a demonstration of the potential effects of such events.<br>These scenarios should not be construed as the only risks AIG faces; these events are shown as an indication of several possible losses AIG could experience.<br>In addition, losses from these and other risks could be materially higher than illustrated.<br>The sensitivity factors utilized for 2010 and presented above were selected based on historical data from 1990 to 2010, as follows (see the table below): ?<br>a 100 basis point parallel shift in the yield curve is broadly consistent with a one standard deviation movement of the benchmark ten-year treasury yield; ?<br>a 20 percent drop for equity and alternative investments is broadly consistent with a one standard deviation movement in the S&P 500; and ?<br>a 10 percent depreciation of foreign currency exchange rates is consistent with a one standard deviation movement in the U. S. dollar (USD)/Japanese Yen (JPY) exchange rate.<br>| | Period | StandardDeviation | Suggested2010Scenario | 2010 Scenarioas aMultiple ofStandardDeviation | 2010 Change/ Return | 2010 as aMultiple ofStandardDeviation | Original2009 Scenario(based onStandardDeviation for1989-2009Period) |<br>| 10-Year Treasury | 1990-2010 | 0.01 | 0.01 | 1.01 | -0.01 | 0.56 | 0.01 |<br>| S&P 500 | 1990-2010 | 0.19 | 0.20 | 1.05 | 0.13 | 0.67 | 0.20 |<br>| USD/JPY | 1990-2010 | 0.11 | 0.10 | 0.92 | 0.15 | 1.34 | 0.10 |<br>Operational Risk Management AIGs Operational Risk Management department (ORM) oversees AIGs operational risk management practices.<br>The Director of ORM reports to the CRO.<br>ORM is responsible for establishing and maintaining the framework, principles and guidelines of AIGs operational risk management program.<br>Each business unit is responsible for its operational risks and implementing the components of the operational risk management program to effectively identify, assess, monitor and mitigate such risks.<br>This responsibility includes developing and implementing policies, procedures, management oversight processes, and other governance-related activities consistent with AIGs overall operational risk management process.<br>Senior operational risk executives in the businesses report to the Director of ORM and to business management.<br>This reporting structure facilitates development of business-specific knowledge of operational risk matters, while at the same time maintaining company-wide consistency in AIGs overall approach to operational risk management.<br>A strong operational risk management program facilitates escalation and resolution of operational risk issues.<br>In order to accomplish this, AIGs operational risk management program is designed to: ?<br>pro-actively address potential operational risk issues; ?<br>create transparency at all levels of the organization; and ?<br>assign clear ownership and accountability for addressing identified issues.<br>As part of the operational risk management framework, AIG has implemented a risk and control self assessment (RCSA) process.<br>The RCSA process is used to identify key operational risks and evaluate the effectiveness of existing controls to mitigate those risks.<br>Corrective action plans are developed to address any identified issues.<br>In 2010, business units continued to enhance their RCSA processes to perform more robust risk assessments.<br>American International Group, Inc. , and Subsidiaries AIGs consolidated risk target is to maintain a minimum liquidity buffer such that AIG Parents liquidity needs under the ERM stress scenarios do not exceed 80 percent of AIG Parents overall liquidity sources over the specified two-year horizon.<br>If the 80 percent minimum threshold is projected to be breached over this defined time horizon, AIG will take appropriate actions to further increase liquidity sources or reduce liquidity needs to maintain the target threshold, although no assurance can be given that this would be possible under then-prevailing market conditions.<br>AIG expects to enter into additional capital maintenance agreements with its U. S. insurance companies to manage the flow of capital and funds between AIG Parent and the insurance companies.<br>As a result of these ERM stress tests, AIG believes that it has sufficient liquidity at the AIG Parent level to satisfy future liquidity requirements and meet its obligations, including reasonably foreseeable contingencies or events.<br>See further discussion regarding AIG Parent and subsidiary liquidity considerations in Liquidity of Parent and Subsidiaries below.</code> |
| <code>Instruct: Given a web search query, retrieve relevant passages that answer the query.<br>Query: Title: <br>Text: What was the total amount of elements for RevPAR excluding those elements greater than 150 in 2016 ?</code> | <code>Title: <br>Text: 2016 Compared to 2015 Comparable?Company-Operated North American Properties<br>| | RevPAR | Occupancy | Average Daily Rate |<br>| | 2016 | Change vs. 2015 | 2016 | Change vs. 2015 | 2016 | Change vs. 2015 |<br>| JW Marriott | $187.02 | 4.0% | 76.8% | 2.2% | pts. | $243.57 | 1.1% |<br>| The Ritz-Carlton | $252.40 | 3.6% | 71.9% | 1.0% | pts. | $350.99 | 2.2% |<br>| W Hotels | $239.94 | -2.2% | 81.7% | 0.2% | pts. | $293.82 | -2.5% |<br>| Composite North American Luxury-1 | $242.10 | 2.8% | 76.3% | 1.4% | pts. | $317.13 | 0.9% |<br>| Marriott Hotels | $144.94 | 2.4% | 75.4% | 0.7% | pts. | $192.23 | 1.4% |<br>| Sheraton | $149.49 | 2.1% | 76.5% | -0.5% | pts. | $195.40 | 2.7% |<br>| Westin | $167.21 | 0.9% | 77.4% | -0.6% | pts. | $216.07 | 1.7% |<br>| Composite North American Upper Upscale-2 | $149.92 | 2.3% | 76.1% | 0.3% | pts. | $196.98 | 1.8% |<br>| North American Full-Service-3 | $166.97 | 2.4% | 76.2% | 0.5% | pts. | $219.25 | 1.7% |<br>| Courtyard | $103.65 | 2.2% | 73.1% | 0.3% | pts. | $141.83 | 1.7% |<br>| Residence Inn | $118.14 | 3.8% | 79.0% | 0.6% | pts. | $149.56 | 3.0% |<br>| Composite North American Limited-Service-4 | $106.20 | 2.8% | 75.0% | 0.5% | pts. | $141.68 | 2.1% |<br>| North American - All-5 | $147.48 | 2.5% | 75.8% | 0.5% | pts. | $194.64 | 1.8% |<br>Comparable?Systemwide North American Properties<br>| | RevPAR | Occupancy | Average Daily Rate |<br>| | 2016 | Change vs. 2015 | 2016 | Change vs. 2015 | 2016 | Change vs. 2015 |<br>| JW Marriott | $178.91 | 3.5% | 76.0% | 1.3% | pts. | $235.47 | 1.8% |<br>| The Ritz-Carlton | $252.40 | 3.6% | 71.9% | 1.0% | pts. | $350.99 | 2.2% |<br>| W Hotels | $239.94 | -2.2% | 81.7% | 0.2% | pts. | $293.82 | -2.5% |<br>| Composite North American Luxury-1 | $231.99 | 2.8% | 76.0% | 1.2% | pts. | $305.36 | 1.2% |<br>| Marriott Hotels | $124.39 | 2.0% | 72.4% | 0.3% | pts. | $171.92 | 1.5% |<br>| Sheraton | $115.58 | 2.4% | 73.3% | 0.3% | pts. | $157.73 | 2.0% |<br>| Westin | $152.94 | 2.4% | 76.9% | 0.1% | pts. | $198.98 | 2.3% |<br>| Composite North American Upper Upscale-2 | $130.44 | 2.5% | 73.9% | 0.4% | pts. | $176.52 | 1.9% |<br>| North American Full-Service-3 | $141.11 | 2.6% | 74.1% | 0.5% | pts. | $190.41 | 1.9% |<br>| Courtyard | $101.49 | 1.9% | 72.9% | —% | pts. | $139.24 | 1.9% |<br>| Residence Inn | $112.78 | 2.4% | 79.0% | -0.1% | pts. | $142.78 | 2.6% |<br>| Fairfield Inn & Suites | $77.96 | 1.2% | 70.1% | -0.5% | pts. | $111.20 | 1.9% |<br>| Composite North American Limited-Service-4 | $96.62 | 2.0% | 74.2% | —% | pts. | $130.15 | 2.0% |<br>| North American - All-5 | $116.47 | 2.3% | 74.2% | 0.2% | pts. | $157.00 | 2.0% |<br>(1) Includes JW Marriott, The Ritz-Carlton, W Hotels, The Luxury Collection, St. Regis, and EDITION.<br>(2) Includes Marriott Hotels, Sheraton, Westin, Renaissance, Autograph Collection, Gaylord Hotels, Le Méridien, and Tribute Portfolio.<br>(3) Includes Composite North American Luxury and Composite North American Upper Upscale.<br>(4) Includes Courtyard, Residence Inn, Fairfield Inn & Suites, SpringHill Suites, and TownePlace Suites.<br>Systemwide also includes Four Points, Aloft, and Element.<br>(5) Includes North American Full-Service and Composite North American Limited-Service.</code> |
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `num_train_epochs`: 2
- `fp16`: True
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: round_robin
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1
- `num_train_epochs`: 2
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: round_robin
</details>
### Training Logs
| Epoch | Step | Evaluate_cosine_map@100 |
|:-----:|:----:|:-----------------------:|
| 0 | 0 | 0.4564 |
| 1.0 | 141 | 0.5233 |
| 2.0 | 282 | 0.5753 |
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.1.1
- Transformers: 4.45.2
- PyTorch: 2.5.1+cu121
- Accelerate: 1.1.1
- Datasets: 3.1.0
- Tokenizers: 0.20.3
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# SentenceTransformer based on thomaskim1130/stella_en_400M_v5-FinanceRAG
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [thomaskim1130/stella_en_400M_v5-FinanceRAG](https://huggingface.co/thomaskim1130/stella_en_400M_v5-FinanceRAG). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [thomaskim1130/stella_en_400M_v5-FinanceRAG](https://huggingface.co/thomaskim1130/stella_en_400M_v5-FinanceRAG) <!-- at revision f10dfc9df857453911fda9d3af68b25bb47dd353 -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 1024 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: NewModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Dense({'in_features': 1024, 'out_features': 1024, 'bias': True, 'activation_function': 'torch.nn.modules.linear.Identity'})
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
"Instruct: Given a web search query, retrieve relevant passages that answer the query.\nQuery: Title: \nText: In the year with lowest amount of Deposits with banks Average volume, what's the increasing rate of Deposits with banks Average volume?",
'Title: \nText: Additional Interest Rate Details Average Balances and Interest Ratesé\x88¥æ\x93\x9cssets(1)(2)(3)(4)\n| | Average volume | Interest revenue | % Average rate |\n| In millions of dollars, except rates | 2015 | 2014 | 2013 | 2015 | 2014 | 2013 | 2015 | 2014 | 2013 |\n| Assets | | | | | | | | | |\n| Deposits with banks-5 | $133,790 | $161,359 | $144,904 | $727 | $959 | $1,026 | 0.54% | 0.59% | 0.71% |\n| Federal funds sold and securities borrowed or purchased under agreements to resell-6 | | | | | | | | | |\n| In U.S. offices | $150,359 | $153,688 | $158,237 | $1,211 | $1,034 | $1,133 | 0.81% | 0.67% | 0.72% |\n| In offices outside the U.S.-5 | 84,006 | 101,177 | 109,233 | 1,305 | 1,332 | 1,433 | 1.55 | 1.32 | 1.31 |\n| Total | $234,365 | $254,865 | $267,470 | $2,516 | $2,366 | $2,566 | 1.07% | 0.93% | 0.96% |\n| Trading account assets-7(8) | | | | | | | | | |\n| In U.S. offices | $114,639 | $114,910 | $126,123 | $3,945 | $3,472 | $3,728 | 3.44% | 3.02% | 2.96% |\n| In offices outside the U.S.-5 | 103,348 | 119,801 | 127,291 | 2,141 | 2,538 | 2,683 | 2.07 | 2.12 | 2.11 |\n| Total | $217,987 | $234,711 | $253,414 | $6,086 | $6,010 | $6,411 | 2.79% | 2.56% | 2.53% |\n| Investments | | | | | | | | | |\n| In U.S. offices | | | | | | | | | |\n| Taxable | $214,714 | $188,910 | $174,084 | $3,812 | $3,286 | $2,713 | 1.78% | 1.74% | 1.56% |\n| Exempt from U.S. income tax | 20,034 | 20,386 | 18,075 | 443 | 626 | 811 | 2.21 | 3.07 | 4.49 |\n| In offices outside the U.S.-5 | 102,376 | 113,163 | 114,122 | 3,071 | 3,627 | 3,761 | 3.00 | 3.21 | 3.30 |\n| Total | $337,124 | $322,459 | $306,281 | $7,326 | $7,539 | $7,285 | 2.17% | 2.34% | 2.38% |\n| Loans (net of unearned income)(9) | | | | | | | | | |\n| In U.S. offices | $354,439 | $361,769 | $354,707 | $24,558 | $26,076 | $25,941 | 6.93% | 7.21% | 7.31% |\n| In offices outside the U.S.-5 | 273,072 | 296,656 | 292,852 | 15,988 | 18,723 | 19,660 | 5.85 | 6.31 | 6.71 |\n| Total | $627,511 | $658,425 | $647,559 | $40,546 | $44,799 | $45,601 | 6.46% | 6.80% | 7.04% |\n| Other interest-earning assets-10 | $55,060 | $40,375 | $38,233 | $1,839 | $507 | $602 | 3.34% | 1.26% | 1.57% |\n| Total interest-earning assets | $1,605,837 | $1,672,194 | $1,657,861 | $59,040 | $62,180 | $63,491 | 3.68% | 3.72% | 3.83% |\n| Non-interest-earning assets-7 | $218,000 | $224,721 | $222,526 | | | | | | |\n| Total assets from discontinued operations | — | — | 2,909 | | | | | | |\n| Total assets | $1,823,837 | $1,896,915 | $1,883,296 | | | | | | |\nNet interest revenue includes the taxable equivalent adjustments related to the tax-exempt bond portfolio (based on the U. S. federal statutory tax rate of 35%) of $487 million, $498 million and $521 million for 2015, 2014 and 2013, respectively.\nInterest rates and amounts include the effects of risk management activities associated with the respective asset categories.\nMonthly or quarterly averages have been used by certain subsidiaries where daily averages are unavailable.\nDetailed average volume, Interest revenue and Interest expense exclude Discontinued operations.\nSee Note 2 to the Consolidated Financial Statements.\nAverage rates reflect prevailing local interest rates, including inflationary effects and monetary corrections in certain countries.\nAverage volumes of securities borrowed or purchased under agreements to resell are reported net pursuant to ASC 210-20-45.\nHowever, Interest revenue excludes the impact of ASC 210-20-45.\nThe fair value carrying amounts of derivative contracts are reported net, pursuant to ASC 815-10-45, in Non-interest-earning assets and Other non-interest bearing liabilities.\nInterest expense on Trading account liabilities of ICG is reported as a reduction of Interest revenue.\nInterest revenue and Interest expense on cash collateral positions are reported in interest on Trading account assets and Trading account liabilities, respectively.\nIncludes cash-basis loans.\nIncludes brokerage receivables.\nDuring 2015, continued management actions, primarily the sale or transfer to held-for-sale of approximately $1.5 billion of delinquent residential first mortgages, including $0.9 billion in the fourth quarter largely associated with the transfer of CitiFinancial loans to held-for-sale referenced above, were the primary driver of the overall improvement in delinquencies within Citi Holdings\x80\x99 residential first mortgage portfolio.\nCredit performance from quarter to quarter could continue to be impacted by the amount of delinquent loan sales or transfers to held-for-sale, as well as overall trends in HPI and interest rates.\nNorth America Residential First Mortgages\x80\x94State Delinquency Trends The following tables set forth the six U. S. states and/or regions with the highest concentration of Citi\x80\x99s residential first mortgages.\n| In billions of dollars | December 31, 2015 | December 31, 2014 |\n| State-1 | ENR-2 | ENRDistribution | 90+DPD% | %LTV >100%-3 | RefreshedFICO | ENR-2 | ENRDistribution | 90+DPD% | %LTV >100%-3 | RefreshedFICO |\n| CA | $19.2 | 37% | 0.2% | 1% | 754 | $18.9 | 31% | 0.6% | 2% | 745 |\n| NY/NJ/CT-4 | 12.7 | 25 | 0.8 | 1 | 751 | 12.2 | 20 | 1.9 | 2 | 740 |\n| VA/MD | 2.2 | 4 | 1.2 | 2 | 719 | 3.0 | 5 | 3.0 | 8 | 695 |\n| IL-4 | 2.2 | 4 | 1.0 | 3 | 735 | 2.5 | 4 | 2.5 | 9 | 713 |\n| FL-4 | 2.2 | 4 | 1.1 | 4 | 723 | 2.8 | 5 | 3.0 | 14 | 700 |\n| TX | 1.9 | 4 | 1.0 | — | 711 | 2.5 | 4 | 2.7 | — | 680 |\n| Other | 11.0 | 21 | 1.3 | 2 | 710 | 18.2 | 30 | 3.3 | 7 | 677 |\n| Total-5 | $51.5 | 100% | 0.7% | 1% | 738 | $60.1 | 100% | 2.1% | 4% | 715 |\nNote: Totals may not sum due to rounding.\n(1) Certain of the states are included as part of a region based on Citi\x80\x99s view of similar HPI within the region.\n(2) Ending net receivables.\nExcludes loans in Canada and Puerto Rico, loans guaranteed by U. S. government agencies, loans recorded at fair value and loans subject to long term standby commitments (LTSCs).\nExcludes balances for which FICO or LTV data are unavailable.\n(3) LTV ratios (loan balance divided by appraised value) are calculated at origination and updated by applying market price data.\n(4) New York, New Jersey, Connecticut, Florida and Illinois are judicial states.\n(5) Improvement in state trends during 2015 was primarily due to the sale or transfer to held-for-sale of residential first mortgages, including the transfer of CitiFinancial residential first mortgages to held-for-sale in the fourth quarter of 2015.\nForeclosures A substantial majority of Citi\x80\x99s foreclosure inventory consists of residential first mortgages.\nAt December 31, 2015, Citi\x80\x99s foreclosure inventory included approximately $0.1 billion, or 0.2%, of the total residential first mortgage portfolio, compared to $0.6 billion, or 0.9%, at December 31, 2014, based on the dollar amount of ending net receivables of loans in foreclosure inventory, excluding loans that are guaranteed by U. S. government agencies and loans subject to LTSCs.\nNorth America Consumer Mortgage Quarterly Credit Trends \x80\x94Net Credit Losses and Delinquencies\x80\x94Home Equity Loans Citi\x80\x99s home equity loan portfolio consists of both fixed-rate home equity loans and loans extended under home equity lines of credit.\nFixed-rate home equity loans are fully amortizing.\nHome equity lines of credit allow for amounts to be drawn for a period of time with the payment of interest only and then, at the end of the draw period, the then-outstanding amount is converted to an amortizing loan (the interest-only payment feature during the revolving period is standard for this product across the industry).\nAfter conversion, the home equity loans typically have a 20-year amortization period.\nAs of December 31, 2015, Citi\x80\x99s home equity loan portfolio of $22.8 billion consisted of $6.3 billion of fixed-rate home equity loans and $16.5 billion of loans extended under home equity lines of credit (Revolving HELOCs).',
'Title: \nText: Issuer Purchases of Equity Securities Repurchases of common stock are made to support the Company\x80\x99s stock-based employee compensation plans and for other corporate purposes.\nOn February 13, 2006, the Board of Directors authorized the purchase of $2.0 billion of the Company\x80\x99s common stock between February 13, 2006 and February 28, 2007.\nIn August 2006, 3M\x80\x99s Board of Directors authorized the repurchase of an additional $1.0 billion in share repurchases, raising the total authorization to $3.0 billion for the period from February 13, 2006 to February 28, 2007.\nIn February 2007, 3M\x80\x99s Board of Directors authorized a twoyear share repurchase of up to $7.0 billion for the period from February 12, 2007 to February 28, 2009.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `Evaluate`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.4636 |
| cosine_accuracy@3 | 0.682 |
| cosine_accuracy@5 | 0.7597 |
| cosine_accuracy@10 | 0.8519 |
| cosine_precision@1 | 0.4636 |
| cosine_precision@3 | 0.2565 |
| cosine_precision@5 | 0.1777 |
| cosine_precision@10 | 0.1024 |
| cosine_recall@1 | 0.4095 |
| cosine_recall@3 | 0.6424 |
| cosine_recall@5 | 0.7299 |
| cosine_recall@10 | 0.8398 |
| cosine_ndcg@10 | 0.6409 |
| cosine_mrr@10 | 0.5902 |
| **cosine_map@100** | **0.5753** |
| dot_accuracy@1 | 0.4393 |
| dot_accuracy@3 | 0.6748 |
| dot_accuracy@5 | 0.7354 |
| dot_accuracy@10 | 0.8422 |
| dot_precision@1 | 0.4393 |
| dot_precision@3 | 0.25 |
| dot_precision@5 | 0.1709 |
| dot_precision@10 | 0.0998 |
| dot_recall@1 | 0.3828 |
| dot_recall@3 | 0.6338 |
| dot_recall@5 | 0.7005 |
| dot_recall@10 | 0.8224 |
| dot_ndcg@10 | 0.6195 |
| dot_mrr@10 | 0.5712 |
| dot_map@100 | 0.5528 |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 2,256 training samples
* Columns: <code>sentence_0</code> and <code>sentence_1</code>
* Approximate statistics based on the first 1000 samples:
| | sentence_0 | sentence_1 |
|:--------|:------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 28 tokens</li><li>mean: 45.02 tokens</li><li>max: 114 tokens</li></ul> | <ul><li>min: 23 tokens</li><li>mean: 406.36 tokens</li><li>max: 512 tokens</li></ul> |
* Samples:
| sentence_0 | sentence_1 |
|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>Instruct: Given a web search query, retrieve relevant passages that answer the query.<br>Query: Title: <br>Text: What do all Notional sum up, excluding those negative ones in 2008 for As of December 31, 2008 for Financial assets with interest rate risk? (in million)</code> | <code>Title: <br>Text: Cash Flows Our estimated future benefit payments for funded and unfunded plans are as follows (in millions):<br>1 The expected benefit payments for our other postretirement benefit plans are net of estimated federal subsidies expected to be received under the Medicare Prescription Drug, Improvement and Modernization Act of 2003.<br>Federal subsidies are estimated to be $3 million for the period 2019-2023 and $2 million for the period 2024-2028.<br>The Company anticipates making pension contributions in 2019 of $32 million, all of which will be allocated to our international plans.<br>The majority of these contributions are required by funding regulations or law.</code> |
| <code>Instruct: Given a web search query, retrieve relevant passages that answer the query.<br>Query: Title: <br>Text: what's the total amount of No surrender charge of 2010 Individual Fixed Annuities, Change in cash of 2008, and Total reserves of 2010 Individual Variable Annuities ?</code> | <code>Title: <br>Text: 2010 and 2009 Comparison Surrender rates have improved compared to the prior year for group retirement products, individual fixed annuities and individual variable annuities as surrenders have returned to more normal levels.<br>Surrender rates for individual fixed annuities have decreased significantly in 2010 due to the low interest rate environment and the relative competitiveness of interest credited rates on the existing block of fixed annuities versus interest rates on alternative investment options available in the marketplace.<br>Surrender rates for group retirement products are expected to increase in 2011 as certain large group surrenders are anticipated.2009 and 2008 Comparison Surrenders and other withdrawals increased in 2009 for group retirement products primarily due to higher large group surrenders.<br>However, surrender rates and withdrawals have improved for individual fixed annuities and individual variable annuities.<br>The following table presents reserves by surrender charge category and surrender rates:<br>| | 2010 | 2009 |<br>| At December 31,(in millions) | Group Retirement Products* | Individual Fixed Annuities | Individual Variable Annuities | Group Retirement Products* | Individual Fixed Annuities | Individual Variable Annuities |<br>| No surrender charge | $52,742 | $14,006 | $11,859 | $47,854 | $11,444 | $11,161 |<br>| 0% - 2% | 1,292 | 3,510 | 4,083 | 1,509 | 3,054 | 4,094 |<br>| Greater than 2% - 4% | 1,754 | 5,060 | 2,040 | 1,918 | 5,635 | 2,066 |<br>| Greater than 4% | 2,753 | 22,777 | 7,361 | 3,213 | 23,885 | 6,758 |<br>| Non-Surrenderable | 792 | 3,136 | 238 | 850 | 3,184 | 558 |<br>| Total reserves | $59,333 | $48,489 | $25,581 | $55,344 | $47,202 | $24,637 |<br>| Surrender rates | 10.3% | 7.4% | 11.4% | 12.3% | 14.4% | 12.1% |<br>* Excludes mutual funds of $9.0 billion and $8.1 billion in 2010 and 2009, respectively.<br>Financial Services Operations AIGs Financial Services subsidiaries engage in diversified activities including commercial aircraft leasing and the remaining Capital Markets portfolios, which are conducted through ILFC and AIGFP, respectively.<br>Following the classification of AGF as discontinued operations in the third quarter of 2010 (see Note 4 to the Consolidated Financial Statements), AIGs remaining consumer finance businesses are now reported in AIGs Other operations category as part of Divested businesses.<br>As discussed in Note 3 to the Consolidated Financial Statements, in order to align financial reporting with changes made during the third quarter of 2010 to the manner in which AIGs chief operating decision makers review the businesses to make decisions about resources to be allocated and to assess performance, changes were made to AIGs segment information.<br>During the third quarter of 2010, AIGs Asset Management Group undertook the management responsibilities for non-derivative assets and liabilities of the Capital Markets businesses of the Financial Services segment.<br>These assets and liabilities are being managed on a spread basis, in concert with the MIP.<br>Accordingly, gains and losses related to these assets and liabilities, primarily consisting of credit valuation adjustment gains and losses are reported in AIGs Other operations category as part of Asset Management Direct Investment business.<br>Also, intercompany interest related to loans from AIG Funding Inc. (AIG Funding) to AIGFP is no longer being allocated to Capital Markets from Other operations.<br>The remaining Capital Markets derivatives business continues to be reported in the Financial Services segment as part of Capital Markets results.<br>American International Group, Inc. , and Subsidiaries solely for illustrative purposes.<br>The selection of these specific events should not be construed as a prediction, but only as a demonstration of the potential effects of such events.<br>These scenarios should not be construed as the only risks AIG faces; these events are shown as an indication of several possible losses AIG could experience.<br>In addition, losses from these and other risks could be materially higher than illustrated.<br>The sensitivity factors utilized for 2010 and presented above were selected based on historical data from 1990 to 2010, as follows (see the table below): ?<br>a 100 basis point parallel shift in the yield curve is broadly consistent with a one standard deviation movement of the benchmark ten-year treasury yield; ?<br>a 20 percent drop for equity and alternative investments is broadly consistent with a one standard deviation movement in the S&P 500; and ?<br>a 10 percent depreciation of foreign currency exchange rates is consistent with a one standard deviation movement in the U. S. dollar (USD)/Japanese Yen (JPY) exchange rate.<br>| | Period | StandardDeviation | Suggested2010Scenario | 2010 Scenarioas aMultiple ofStandardDeviation | 2010 Change/ Return | 2010 as aMultiple ofStandardDeviation | Original2009 Scenario(based onStandardDeviation for1989-2009Period) |<br>| 10-Year Treasury | 1990-2010 | 0.01 | 0.01 | 1.01 | -0.01 | 0.56 | 0.01 |<br>| S&P 500 | 1990-2010 | 0.19 | 0.20 | 1.05 | 0.13 | 0.67 | 0.20 |<br>| USD/JPY | 1990-2010 | 0.11 | 0.10 | 0.92 | 0.15 | 1.34 | 0.10 |<br>Operational Risk Management AIGs Operational Risk Management department (ORM) oversees AIGs operational risk management practices.<br>The Director of ORM reports to the CRO.<br>ORM is responsible for establishing and maintaining the framework, principles and guidelines of AIGs operational risk management program.<br>Each business unit is responsible for its operational risks and implementing the components of the operational risk management program to effectively identify, assess, monitor and mitigate such risks.<br>This responsibility includes developing and implementing policies, procedures, management oversight processes, and other governance-related activities consistent with AIGs overall operational risk management process.<br>Senior operational risk executives in the businesses report to the Director of ORM and to business management.<br>This reporting structure facilitates development of business-specific knowledge of operational risk matters, while at the same time maintaining company-wide consistency in AIGs overall approach to operational risk management.<br>A strong operational risk management program facilitates escalation and resolution of operational risk issues.<br>In order to accomplish this, AIGs operational risk management program is designed to: ?<br>pro-actively address potential operational risk issues; ?<br>create transparency at all levels of the organization; and ?<br>assign clear ownership and accountability for addressing identified issues.<br>As part of the operational risk management framework, AIG has implemented a risk and control self assessment (RCSA) process.<br>The RCSA process is used to identify key operational risks and evaluate the effectiveness of existing controls to mitigate those risks.<br>Corrective action plans are developed to address any identified issues.<br>In 2010, business units continued to enhance their RCSA processes to perform more robust risk assessments.<br>American International Group, Inc. , and Subsidiaries AIGs consolidated risk target is to maintain a minimum liquidity buffer such that AIG Parents liquidity needs under the ERM stress scenarios do not exceed 80 percent of AIG Parents overall liquidity sources over the specified two-year horizon.<br>If the 80 percent minimum threshold is projected to be breached over this defined time horizon, AIG will take appropriate actions to further increase liquidity sources or reduce liquidity needs to maintain the target threshold, although no assurance can be given that this would be possible under then-prevailing market conditions.<br>AIG expects to enter into additional capital maintenance agreements with its U. S. insurance companies to manage the flow of capital and funds between AIG Parent and the insurance companies.<br>As a result of these ERM stress tests, AIG believes that it has sufficient liquidity at the AIG Parent level to satisfy future liquidity requirements and meet its obligations, including reasonably foreseeable contingencies or events.<br>See further discussion regarding AIG Parent and subsidiary liquidity considerations in Liquidity of Parent and Subsidiaries below.</code> |
| <code>Instruct: Given a web search query, retrieve relevant passages that answer the query.<br>Query: Title: <br>Text: What was the total amount of elements for RevPAR excluding those elements greater than 150 in 2016 ?</code> | <code>Title: <br>Text: 2016 Compared to 2015 Comparable?Company-Operated North American Properties<br>| | RevPAR | Occupancy | Average Daily Rate |<br>| | 2016 | Change vs. 2015 | 2016 | Change vs. 2015 | 2016 | Change vs. 2015 |<br>| JW Marriott | $187.02 | 4.0% | 76.8% | 2.2% | pts. | $243.57 | 1.1% |<br>| The Ritz-Carlton | $252.40 | 3.6% | 71.9% | 1.0% | pts. | $350.99 | 2.2% |<br>| W Hotels | $239.94 | -2.2% | 81.7% | 0.2% | pts. | $293.82 | -2.5% |<br>| Composite North American Luxury-1 | $242.10 | 2.8% | 76.3% | 1.4% | pts. | $317.13 | 0.9% |<br>| Marriott Hotels | $144.94 | 2.4% | 75.4% | 0.7% | pts. | $192.23 | 1.4% |<br>| Sheraton | $149.49 | 2.1% | 76.5% | -0.5% | pts. | $195.40 | 2.7% |<br>| Westin | $167.21 | 0.9% | 77.4% | -0.6% | pts. | $216.07 | 1.7% |<br>| Composite North American Upper Upscale-2 | $149.92 | 2.3% | 76.1% | 0.3% | pts. | $196.98 | 1.8% |<br>| North American Full-Service-3 | $166.97 | 2.4% | 76.2% | 0.5% | pts. | $219.25 | 1.7% |<br>| Courtyard | $103.65 | 2.2% | 73.1% | 0.3% | pts. | $141.83 | 1.7% |<br>| Residence Inn | $118.14 | 3.8% | 79.0% | 0.6% | pts. | $149.56 | 3.0% |<br>| Composite North American Limited-Service-4 | $106.20 | 2.8% | 75.0% | 0.5% | pts. | $141.68 | 2.1% |<br>| North American - All-5 | $147.48 | 2.5% | 75.8% | 0.5% | pts. | $194.64 | 1.8% |<br>Comparable?Systemwide North American Properties<br>| | RevPAR | Occupancy | Average Daily Rate |<br>| | 2016 | Change vs. 2015 | 2016 | Change vs. 2015 | 2016 | Change vs. 2015 |<br>| JW Marriott | $178.91 | 3.5% | 76.0% | 1.3% | pts. | $235.47 | 1.8% |<br>| The Ritz-Carlton | $252.40 | 3.6% | 71.9% | 1.0% | pts. | $350.99 | 2.2% |<br>| W Hotels | $239.94 | -2.2% | 81.7% | 0.2% | pts. | $293.82 | -2.5% |<br>| Composite North American Luxury-1 | $231.99 | 2.8% | 76.0% | 1.2% | pts. | $305.36 | 1.2% |<br>| Marriott Hotels | $124.39 | 2.0% | 72.4% | 0.3% | pts. | $171.92 | 1.5% |<br>| Sheraton | $115.58 | 2.4% | 73.3% | 0.3% | pts. | $157.73 | 2.0% |<br>| Westin | $152.94 | 2.4% | 76.9% | 0.1% | pts. | $198.98 | 2.3% |<br>| Composite North American Upper Upscale-2 | $130.44 | 2.5% | 73.9% | 0.4% | pts. | $176.52 | 1.9% |<br>| North American Full-Service-3 | $141.11 | 2.6% | 74.1% | 0.5% | pts. | $190.41 | 1.9% |<br>| Courtyard | $101.49 | 1.9% | 72.9% | —% | pts. | $139.24 | 1.9% |<br>| Residence Inn | $112.78 | 2.4% | 79.0% | -0.1% | pts. | $142.78 | 2.6% |<br>| Fairfield Inn & Suites | $77.96 | 1.2% | 70.1% | -0.5% | pts. | $111.20 | 1.9% |<br>| Composite North American Limited-Service-4 | $96.62 | 2.0% | 74.2% | —% | pts. | $130.15 | 2.0% |<br>| North American - All-5 | $116.47 | 2.3% | 74.2% | 0.2% | pts. | $157.00 | 2.0% |<br>(1) Includes JW Marriott, The Ritz-Carlton, W Hotels, The Luxury Collection, St. Regis, and EDITION.<br>(2) Includes Marriott Hotels, Sheraton, Westin, Renaissance, Autograph Collection, Gaylord Hotels, Le Méridien, and Tribute Portfolio.<br>(3) Includes Composite North American Luxury and Composite North American Upper Upscale.<br>(4) Includes Courtyard, Residence Inn, Fairfield Inn & Suites, SpringHill Suites, and TownePlace Suites.<br>Systemwide also includes Four Points, Aloft, and Element.<br>(5) Includes North American Full-Service and Composite North American Limited-Service.</code> |
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `num_train_epochs`: 2
- `fp16`: True
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: round_robin
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1
- `num_train_epochs`: 2
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: round_robin
</details>
### Training Logs
| Epoch | Step | Evaluate_cosine_map@100 |
|:-----:|:----:|:-----------------------:|
| 0 | 0 | 0.4564 |
| 1.0 | 141 | 0.5233 |
| 2.0 | 282 | 0.5753 |
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.1.1
- Transformers: 4.45.2
- PyTorch: 2.5.1+cu121
- Accelerate: 1.1.1
- Datasets: 3.1.0
- Tokenizers: 0.20.3
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "thomaskim1130/stella_en_400M_v5-FinanceRAG", "library_name": "sentence-transformers", "metrics": ["cosine_accuracy@1", "cosine_accuracy@3", "cosine_accuracy@5", "cosine_accuracy@10", "cosine_precision@1", "cosine_precision@3", "cosine_precision@5", "cosine_precision@10", "cosine_recall@1", "cosine_recall@3", "cosine_recall@5", "cosine_recall@10", "cosine_ndcg@10", "cosine_mrr@10", "cosine_map@100", "dot_accuracy@1", "dot_accuracy@3", "dot_accuracy@5", "dot_accuracy@10", "dot_precision@1", "dot_precision@3", "dot_precision@5", "dot_precision@10", "dot_recall@1", "dot_recall@3", "dot_recall@5", "dot_recall@10", "dot_ndcg@10", "dot_mrr@10", "dot_map@100"], "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:2256", "loss:MultipleNegativesRankingLoss"], "widget": [{"source_sentence": "Instruct: Given a web search query, retrieve relevant passages that answer the query.\nQuery: Title: \nText: What was the sum of Fourth Quarter without those Fourth Quarter smaller than 0, in 2012? (in million)", "sentences": ["Title: \nText: Cash Flow Hedges Citigroup hedges the variability of forecasted cash flows associated with floating-rate assets/liabilities and other forecasted transactions.\nVariable cash flows from those liabilities are synthetically converted to fixed-rate cash flows by entering into receive-variable, pay-fixed interest rate swaps and receivevariable, pay-fixed forward-starting interest rate swaps.\nVariable cash flows associated with certain assets are synthetically converted to fixed-rate cash flows by entering into receive-fixed, pay-variable interest rate swaps.\nThese cash flow hedging relationships use either regression analysis or dollar-offset ratio analysis to assess whether the hedging relationships are highly effective at inception and on an ongoing basis.\nPrior to the adoption of ASU 2017-12, Citigroup designated the risk being hedged as the risk of overall variability in the hedged cash flows for certain items.\nWith the adoption of ASU 2017-12, Citigroup hedges the variability from changes in a contractually specified rate and recognizes the entire change in fair value of the cash flow hedging instruments in AOCI.\nPrior to the adoption of ASU 2017-12, to the extent that these derivatives were not fully effective, changes in their fair values in excess of changes in the value of the hedged transactions were immediately included in Other revenue.\nWith the adoption of ASU 2017-12, such amounts are no longer required to be immediately recognized in income, but instead the full change in the value of the hedging instrument is required to be recognized in AOCI, and then recognized in earnings in the same period that the cash flows impact earnings.\nThe pretax change in AOCI from cash flow hedges is presented below:\n| | Year ended December 31, |\n| In millions of dollars | 2018 | 2017 | 2016 |\n| Amount of gain (loss) recognized in AOCI on derivative | | | |\n| Interest rate contracts-1 | $-361 | | $-165 | $-219 |\n| Foreign exchange contracts | 5 | -8 | 69 |\n| Total gain (loss) recognized in AOCI | $-356 | | $-173 | $-150 |\n| Amount of gain (loss) reclassified from AOCI to earnings | Otherrevenue | Net interestrevenue | Otherrevenue | Otherrevenue |\n| Interest rate contracts-1 | $— | $-301 | $-126 | $-140 |\n| Foreign exchange contracts | -17 | — | -10 | -93 |\n| Total gain (loss) reclassified from AOCI into earnings | $-17 | $-301 | $-136 | $-233 |\n(1) After January 1, 2018, all amounts reclassified into earnings for interest rate contracts are included in Interest income/Interest expense (Net interest revenue).\nFor all other hedges, including interest rate hedges prior to January 1, 2018, the amounts reclassified to earnings are included primarily in Other revenue and Net interest revenue in the Consolidated Statement of Income.\nFor cash flow hedges, the changes in the fair value of the hedging derivative remain in AOCI on the Consolidated Balance Sheet and will be included in the earnings of future periods to offset the variability of the hedged cash flows when such cash flows affect earnings.\nThe net gain (loss) associated with cash flow hedges expected to be reclassified from AOCI within 12?months of December?31, 2018 is approximately $404 million.\nThe maximum length of time over which forecasted cash flows are hedged is 10 years.\nThe after-tax impact of cash flow hedges on AOCI is shown in Note?19 to the Consolidated Financial Statements.", "Title: \nText: | | Net Sales |\n| (Amounts in millions) | 2012 | 2011 | 2010 |\n| Product Category: | | | |\n| Tools | $1,729.4 | $1,667.3 | $1,545.1 |\n| Diagnostics and repair information | 619.8 | 613.7 | 563.3 |\n| Equipment | 588.7 | 573.2 | 510.8 |\n| | $2,937.9 | $2,854.2 | $2,619.2 |\nThe tools product category includes hand tools, power tools and tool storage products.\nHand tools include wrenches, sockets, ratchet wrenches, pliers, screwdrivers, punches and chisels, saws and cutting tools, pruning tools, torque measuring instruments and other similar products.\nPower tools include cordless (battery), pneumatic (air), hydraulic, and corded (electric) tools, such as impact wrenches, ratchets, chisels, drills, sanders, polishers and similar products.\nTool storage includes tool chests, roll cabinets, tool control systems and other similar products.\nThe majority of products are manufactured by Snap-on and, in completing the product offering, other items are purchased from external manufacturers.\nThe diagnostics and repair information product category includes handheld and PC-based diagnostic products, service and repair information products, diagnostic software solutions, electronic parts catalogs, business management systems and services, point-of-sale systems, integrated systems for vehicle service shops, OEM purchasing facilitation services, and warranty management systems and analytics to help OEM dealership service and repair shops manage and track performance.\nThe equipment product category includes solutions for the diagnosis and service of vehicles and industrial equipment.\nProducts include wheel alignment equipment, wheel balancers, tire changers, vehicle lifts, test lane systems, collision repair equipment, air conditioning service equipment, brake service equipment, fluid exchange equipment, transmission troubleshooting equipment, safety testing equipment, battery chargers and hoists.\nSnap-on supports the sale of its diagnostics and vehicle service shop equipment by offering training programs as well as after sales support for its customers, primarily focusing on the technologies and the application of specific products developed and marketed by Snap-on.\nManagement’s Discussion and Analysis of Financial Condition and Results of Operations (continued) Segment gross profit of $105.0 million in the fourth quarter of 2012 decreased $1.4 million from 2011 levels.\nGross margin of 38.1% in the quarter improved 210 basis points from 36.0% last year primarily due to lower restructuring costs as well as savings from ongoing RCI initiatives, particularly in Europe.\nNo restructuring costs were incurred in the fourth quarter of 2012; gross profit in the fourth quarter of 2011 included $2.5 million of restructuring costs.\nSegment operating expenses of $73.1 million in the fourth quarter of 2012 decreased $0.3 million from 2011 levels.\nThe operating expense margin of 26.5% in the quarter increased 170 basis points from 24.8% last year primarily as a result of the lower sales.\nAs a result of these factors, segment operating earnings of $31.9 million in the fourth quarter of 2012, including $1.2 million of favorable foreign currency effects, decreased $1.1 million, or 3.3%, from 2011 levels.\nOperating margin for the Commercial & Industrial Group of 11.6% in the fourth quarter of 2012 improved 40 basis points from 11.2% last year.\nSnap-on Tools Group", "Title: \nText: 5.\nBasis of Presentation and Summary of Significant Accounting Policies (a) Basis of Presentation On December 30, 2015, US Airways merged with and into American, which is reflected in American’s consolidated financial statements as though the transaction had occurred on December 9, 2013, when a subsidiary of AMR merged with and into US Airways Group.\nThus, the full years of 2015 and 2014 and the period from December 9, 2013 to December 31, 2013 are comprised of the consolidated financial data of American and US Airways.\nFor the periods prior to December 9, 2013, the financial data reflects the results of American only.\nFor financial reporting purposes, the transaction constituted a transfer of assets between entities under common control and was accounted for in a manner similar to the pooling of interests method of accounting.\nUnder this method, the carrying amount of net assets recognized in the balance sheets of each combining entity are carried forward to the balance sheet of the combined entity and no other assets or liabilities are recognized.\nThe preparation of financial statements in accordance with accounting principles generally accepted in the United States (GAAP) requires management to make certain estimates and assumptions that affect the reported amounts of assets and liabilities, revenues and expenses, and the disclosure of contingent assets and liabilities at the date of the financial statements.\nActual results could differ from those estimates.\nThe most significant areas of judgment relate to passenger revenue recognition, impairment of goodwill, impairment of long-lived and\nThe following tables set forth the income yield and investment income, excluding realized investment gains (losses) and non-hedge accounting derivative results, for each major investment category of our Japanese operations’ general account for the periods indicated."]}, {"source_sentence": "Instruct: Given a web search query, retrieve relevant passages that answer the query.\nQuery: Title: \nText: ADBE share repurchase", "sentences": ["Title: \nText: TELEFLEX INCORPORATED NOTES?TO CONSOLIDATED FINANCIAL STATEMENTS — (Continued) The Company issued 82,865, 93,367 and 105,239 of non-vested restricted stock units in 2017, 2016 and 2015, respectively, the majority of which provide for vesting as to all underlying shares on the third anniversary of the grant date.\nThe weighted average grant-date fair value for non-vested restricted stock units granted during 2017, 2016 and 2015 was $187.85, $142.71 and $118.00, respectively.\nThe Company recorded $11.2 million of expense related to restricted stock units during 2017, which is included in cost of goods sold or selling, general and administrative expenses.\nThe unamortized share-based compensation cost related to non-vested restricted stock units, net of expected forfeitures, was $13.2 million, which is expected to be recognized over a weighted-average period of 1.8 years.\nThe Company uses treasury stock to provide shares of common stock in connection with vesting of the restricted stock units.\nTELEFLEX INCORPORATED NOTES?TO CONSOLIDATED FINANCIAL STATEMENTS — (Continued) F-37 Note 13?— Income taxes The following table summarizes the components of the provision for income taxes from continuing operations:\n| | 2017 | 2016 | 2015 |\n| | (Dollars in thousands) |\n| Current: | | | |\n| Federal | $133,621 | $2,344 | $-4,700 |\n| State | 5,213 | 5,230 | 2,377 |\n| Foreign | 35,444 | 28,842 | 53,151 |\n| Deferred: | | | |\n| Federal | -258,247 | -25,141 | -35,750 |\n| State | 1,459 | -1,837 | -5,012 |\n| Foreign | 212,158 | -1,364 | -2,228 |\n| | $129,648 | $8,074 | $7,838 |\nThe Tax Cuts and Jobs Act (the “TCJA”) was enacted on December 22, 2017.\nThe legislation significantly changes U. S. tax law by, among other things, permanently reducing corporate income tax rates from a maximum of 35% to 21%, effective January 1, 2018; implementing a territorial tax system, by generally providing for, among other things, a dividends received deduction on the foreign source portion of dividends received from a foreign corporation if specified conditions are met; and imposing a one-time repatriation tax on undistributed post-1986 foreign subsidiary earnings and profits, which are deemed repatriated for purposes of the tax.\nAs a result of the TCJA, the Company reassessed and revalued its ending net deferred tax liabilities at December 31, 2017 and recognized a?$46.1 million?provisional tax benefit in the Company’s consolidated statement of income for the year ended December 31, 2017.\nAs a result of the deemed repatriation tax under the TCJA, the Company recognized a $154.0 million provisional tax expense in the Company’s consolidated statement of income for the year ended December 31, 2017, and the Company expects to pay this tax over an eight-year period.\nWhile the TCJA provides for a territorial tax system, beginning in 2018, it includes?two?new U. S. tax base erosion provisions, the global intangible low-taxed income (“GILTI”) provisions and the base-erosion and anti-abuse tax (“BEAT”) provisions.\nThe GILTI provisions require the Company to include in its U. S. income tax return foreign subsidiary earnings in excess of an allowable return on the foreign subsidiary’s tangible assets.\nThe Company expects that it will be subject to incremental U. S. tax on GILTI income beginning in 2018.\nBecause of the complexity of the new GILTI tax rules, the Company is continuing to evaluate this provision of the TCJA and the application of Financial Accounting Standards Board Accounting Standards Codification Topic 740, \"Income Taxes. \"\nUnder U. S. GAAP, the Company may make an accounting policy election to either (1) treat future taxes with respect to the inclusion in U. S. taxable income of amounts related to GILTI as current period expense when incurred (the “period cost method”) or (2) take such amounts into a company’s measurement of its deferred taxes (the “deferred method”).\nThe Company’s selection of an accounting policy with respect to the new GILTI tax rules will depend, in part, on an analysis of the Company’s global income to determine whether the Company expects to have future U. S. inclusions in taxable income related to GILTI and, if so, what the impact is expected to be.\nThe determination of whether the Company expects to have future U. S. inclusions", "Title: ADBE NOTES TO CONSOLIDATED FINANCIAL STATEMENTS (Continued)\nText: For fiscal 2023, 2022 and 2021, the prepayments were classified as treasury stock on our Consolidated Balance Sheets at the payment date, though only shares physically delivered to us by December 1, 2023, December 2, 2022 and December 3, 2021 were excluded from the computation of net income per share. As of December 1, 2023, $354 million of prepayment remained under our outstanding structured stock repurchase agreement.", "Title: \nText: Market and Market Prices of Common Stock\nDuring 2016 fiscal year and through February 22, 2017, our common stock was traded on the Nasdaq Capital Market under the symbol “ACUR”. On February 23, 2017, our common stock was delisted from the Nasdaq Capital Market due to our failure to comply with Nasdaq’s Listing Rule 5550(b)(1), which requires that we maintain $2.5 million in stockholders’ equity for continued listing (or meet the alternatives of market value of listed securities of $35 million or net income from continuing operations). NASDAQ had granted us a grace period through February 10, 2017, to regain compliance with Listing Rule 5550(b)(1), but we were unable to regain compliance within such period.\nCommencing on February 23, 2017, our common stock was quoted on the OTCQB under the symbol “ACUR”, however commencing June 4, 2018 and lasting until July 2, 2018 it was quoted on the OTC Markets OTC Pink tier. The downgrade was a result of the late filing of our 2017 Annual Report on Form 10-K beyond any applicable grace periods. The Company regained compliance with the OTCQB and effective July 3, 2018 it was quoted on the OTCQB. However, commencing May 20, 2019 as a result of late filing of our 2018 Annual Report on Form 10-K our common stock was again relegated to the OTC Markets OTC Pink tier. The Company regained compliance with the OTCQB in March, 2020 and effective March 23, 2020 it was quoted on the OTCQB.\nSet forth below for the period indicated are the high and low sales prices for our common stock in the OTC Market of OTCQB and Pink tier.\nOn March 27, 2020 the closing sales price of our common stock was $0.22.\n\nPeriod | | Sales Prices\n--------------------------------- | ----- | ------------\n | High | Low \n2019 Fiscal Year | | \nFirst Quarter | $0.29 | $0.11 \nSecond Quarter | 0.28 | 0.13 \nThird Quarter | 0.45 | 0.14 \nFourth Quarter | 0.63 | 0.20 \n2020 Fiscal Year | | \nFirst Quarter thru March 27, 2020 | 0.47 | $0.12 "]}, {"source_sentence": "Instruct: Given a web search query, retrieve relevant passages that answer the query.\nQuery: Title: \nText: What is the growing rate of Equity securities, trading for Carrying amount in the year with the most Fixed maturities, available-for-sale ?", "sentences": ["Title: \nText: The following table details the estimated changes by risk management strategy.\nThe table also gives the weighted-average duration of the asset portfolio for each category, and the net duration gap (i. e. , the weighted-average difference between the asset and liability durations).\n| | December 31, 2007 |\n| Risk Management Strategy | Value of total assets(in millions) | Duration of assets | Net duration gap | Net fair value change (in millions) |\n| Primary duration-managed | $33,183.4 | 3.53 | -0.01 | $3.3 |\n| Duration-monitored | 17,990.9 | 4.70 | 0.22 | -39.6 |\n| Non duration-managed | 5,234.0 | 4.43 | N/A | N/A |\n| Total | $56,408.3 | | | $-36.3 |\nOur selection of a 100 basis point immediate, parallel increase or decrease in interest rates is a hypothetical rate scenario we use to demonstrate potential risk.\nWhile a 100 basis point immediate, parallel increase does not represent our view of future market changes, it is a near term reasonably possible hypothetical change that illustrates the potential impact of such events.\nWhile these fair value measurements provide a representation of interest rate sensitivity, they are based on our portfolio exposures at a point in time and may not be representative of future market results.\nThese exposures will change as a result of ongoing portfolio transactions in response to new business, management’s assessment of changing market conditions and available investment opportunities.", "Title: \nText: PART I ITEM 1. BUSINESS (dollars in millions, except per share, per ounce and per pound amounts) Introduction Newmont Mining Corporation is primarily a gold producer with significant operations and/or assets in the United States, Australia, Peru, Ghana and Suriname.\nAt December 31, 2016, Newmont had attributable proven and probable gold reserves of 68.5 million ounces and an aggregate land position of approximately 23,000 square miles (59,000 square kilometers).\nNewmont is also engaged in the production of copper, principally through Boddington in Australia and Phoenix in the United States.\nNewmont Mining Corporations original predecessor corporation was incorporated in 1921 under the laws of Delaware.\nOn November 2, 2016, Newmont completed the sale of its 48.5% economic interest in PT Newmont Nusa Tenggara (PTNNT), which operated the Batu Hijau copper and gold mine (Batu Hijau) in Indonesia (the Batu Hijau Transaction).\nAs a result, Newmont presents Batu Hijau as a discontinued operation for all periods presented.\nIn the following discussion, we present and discuss our continuing operations unless otherwise indicated.\nFor additional information regarding our discontinued operations, see Note 3 to the Consolidated Financial Statements and the discussion in our Results of Consolidated Operations in Item 7.\nNewmonts corporate headquarters are in Greenwood Village, Colorado, USA.\nIn this report, Newmont, the Company, our and we refer to Newmont Mining Corporation together with our affiliates and subsidiaries, unless the context otherwise requires.\nReferences to A$ refer to Australian currency.\nNewmonts Sales and long-lived assets for continuing operations are geographically distributed as follows:\n| | Sales | Long-Lived Assets |\n| | 2016 | 2015 | 2014 | 2016 | 2015 | 2014 |\n| United States | 39% | 33% | 30% | 45% | 43% | 38% |\n| Australia | 32% | 32% | 30% | 19% | 18% | 19% |\n| Ghana | 15% | 15% | 17% | 16% | 16% | 17% |\n| Peru | 12% | 18% | 18% | 14% | 19% | 23% |\n| Suriname | 2% | —% | —% | 6% | 4% | 2% |\n| Other | —% | 2% | 5% | —% | —% | 1% |\nSegment Information Our regions include North America, South America, Asia Pacific, and Africa.\nOur North America segment consists primarily of Carlin, Phoenix, Twin Creeks and Long Canyon in the state of Nevada and Cripple Creek &Victor (CC&V) in the state of Colorado, in the United States.\nOur South America segment consists primarily of Yanacocha in Peru and Merian in Suriname.\nOur Asia Pacific segment consists primarily of Boddington, Tanami and Kalgoorlie in Australia.\nOur Africa segment consists primarily of Ahafo and Akyem in Ghana.\nSee Item 1A, Risk Factors, below, and Note 5 to the Consolidated Financial Statements for information relating to our operating segments, domestic and export sales and lack of dependence on a limited number of customers.\nProducts References in this report to attributable gold ounces or attributable copper pounds mean that portion of gold or copper produced, sold or included in proven and probable reserves based on our proportionate ownership, unless otherwise noted.\nGold General.\nWe had consolidated gold production from continuing operations of 5.2 million ounces (4.9 million attributable ounces) in 2016, 5.0 million ounces (4.6 million attributable ounces) in 2015 and 5.2 million ounces (4.7 million attributable ounces) in 2014.\nOf our 2016 consolidated gold production, approximately 39% came from North America, 14% from South America, 31% from Asia Pacific and 16% from Africa.\nFor 2016, 2015 and 2014, 96%, 95% and 95%, respectively, of our Sales were attributable to gold.\nMost of our Sales come from the sale of refined gold.\nThe end product at our gold operations, however, is generally doré bars.\nDoré is an alloy consisting primarily of gold but also containing silver and other metals.\nDoré is sent to refiners to produce bullion that meets the required markett standard\nOperating Statistics The following tables detail operating statistics related to gold production, ounces sold and production costs per ounce of our continuing operations:\n| | North America | South America 2018 | |\n| Years Ended December 31, | 2018 | 2017 | 2016 | 2017 | 2016 |\n| Tons mined (000 dry short tons): | | | | | | |\n| Open pit | 230,558 | 252,086 | 218,411 | 99,793 | 104,763 | 104,713 |\n| Underground | 3,024 | 2,979 | 2,864 | — | — | — |\n| Tons processed (000 dry short tons): | | | | | | |\n| Mill | 25,879 | 25,406 | 25,941 | 21,666 | 20,690 | 9,006 |\n| Leach | 46,034 | 55,289 | 45,109 | 25,405 | 24,082 | 30,639 |\n| Average ore grade (oz/ton): | | | | | | |\n| Mill | 0.075 | 0.077 | 0.074 | 0.042 | 0.043 | 0.063 |\n| Leach | 0.017 | 0.020 | 0.019 | 0.013 | 0.013 | 0.012 |\n| Average mill recovery rate | 76.7% | 76.9% | 78.5% | 88.0% | 87.2% | 79.4% |\n| Ounces produced -000: | | | | | | |\n| Mill | 1,453 | 1,485 | 1,501 | 802 | 752 | 434 |\n| Leach | 604 | 726 | 523 | 247 | 296 | 325 |\n| Consolidated | 2,057 | 2,211 | 2,024 | 1,049 | 1,048 | 759 |\n| Attributable | 2,057 | 2,211 | 2,024 | 671 | 660 | 414 |\n| Consolidated ounces sold -000 | 2,052 | 2,204 | 1,990 | 1,060 | 1,046 | 736 |\n| Production costs per ounce sold:-1 | | | | | | |\n| Direct mining and production costs | $753 | $706 | $729 | $593 | $639 | $737 |\n| By-product credits | -8 | -9 | -11 | -19 | -17 | -11 |\n| Royalties and production taxes | 12 | 10 | 15 | 53 | 54 | 38 |\n| Write-downs and inventory change | 2 | 5 | -34 | 33 | 33 | -5 |\n| Costs applicable to sales | 759 | 712 | 699 | 660 | 709 | 759 |\n| Depreciation and amortization | 238 | 244 | 207 | 201 | 229 | 404 |\n| Reclamation accretion | 6 | 6 | 6 | 24 | 45 | 36 |\n| Total production costs | $1,003 | $962 | $912 | $885 | $983 | $1,199 |\n| All-in sustaining costs per ounce sold-2 | $928 | $876 | $854 | $804 | $870 | $932 |\nYears Ended December 31,\n| | Australia | Africa 2018 | |\n| Years Ended December 31, | 2018 | 2017 | 2016 | 2017 | 2016 |\n| Tons mined (000 dry short tons): | | | | | | |\n| Open pit | 103,192 | 114,371 | 126,619 | 71,970 | 74,580 | 75,048 |\n| Underground | 3,202 | 3,144 | 3,279 | 1,339 | 279 | — |\n| Tons milled (000 dry short tons) | 54,337 | 52,802 | 51,606 | 15,585 | 16,884 | 17,289 |\n| Average ore grade (oz/ton) | 0.032 | 0.035 | 0.037 | 0.058 | 0.053 | 0.052 |\n| Average mill recovery rate | 87.4% | 86.1% | 86.4% | 92.6% | 92.3% | 91.1% |\n| Ounces produced -000: | | | | | | |\n| Mill | 1,523 | 1,573 | 1,641 | 850 | 822 | 819 |\n| Consolidated | 1,523 | 1,573 | 1,641 | 850 | 822 | 819 |\n| Consolidated ounces sold -000 | 1,553 | 1,558 | 1,624 | 851 | 824 | 822 |\n| Production costs per ounce sold:-1 | | | | | | |\n| Direct mining and production costs | $681 | $673 | $605 | $592 | $573 | $553 |\n| By-product credits | -7 | -8 | -7 | -2 | -2 | -2 |\n| Royalties and production taxes | 32 | 32 | 32 | 55 | 51 | 50 |\n| Write-downs and inventory change | 3 | -25 | — | — | 33 | 65 |\n| Costs applicable to sales | 709 | 672 | 630 | 645 | 655 | 666 |\n| Depreciation and amortization | 133 | 134 | 135 | 301 | 277 | 271 |\n| Reclamation accretion | 8 | 7 | 7 | 9 | 9 | 7 |\n| Total production costs | $850 | $813 | $772 | $955 | $941 | $944 |\n| All-in sustaining costs per ounce sold-2 | $845 | $806 | $777 | $794 | $785 | $795 |\nTable of Contents The following performance graph is not soliciting material, is not deemed filed with the SEC, and is not to be incorporated by reference into any of Valeros filings under the Securities Act of 1933 or the Securities Exchange Act of 1934, as amended, respectively.\nThis performance graph and the related textual information are based on historical data and are not indicative of future performance.\nThe following line graph compares the cumulative total return 1 on an investment in our common stock against the cumulative total return of the S&P 500 Composite Index and an index of peer companies (that we selected) for the five-year period commencing December 31, 2006 and ending December 31, 2011.", "Title: \nText: NOTE 4 Derivatives Derivative Balances Derivatives are entered into on behalf of customers, for trading, as economic hedges or as qualifying accounting hedges.\nThe Corporation enters into derivatives to facilitate client transactions, for principal trading purposes and to manage risk exposures.\nFor additional information on the Corporations derivatives and hedging activities, see Note 1 Summary of Significant Accounting Principles.\nThe table below identifies derivative instruments included on the Corporations Consolidated Balance Sheet in derivative assets and liabilities at December 31, 2010 and 2009.\nBalances are presented on a gross basis, prior to the application of counterparty and collateral netting.\nTotal derivative assets and liabilities are adjusted on an aggregate basis to take into consideration the effects of legally enforceable master netting agreements and have been reduced by the cash collateral applied.\n| | | December 31, 2010 |\n| | | Gross Derivative Assets | Gross Derivative Liabilities |\n| | | Trading | | | Trading | | |\n| | | Derivatives | | | Derivatives | | |\n| | | and | Qualifying | | and | Qualifying | |\n| | Contract/ | Economic | Accounting | | Economic | Accounting | |\n| (Dollars in billions) | Notional-1 | Hedges | Hedges-2 | Total | Hedges | Hedges-2 | Total |\n| Interest rate contracts | | | | | | | |\n| Swaps | $42,719.2 | $1,193.9 | $14.9 | $1,208.8 | $1,187.9 | $2.2 | $1,190.1 |\n| Futures and forwards | 9.939.2 | 6.0 | – | 6.0 | 4.7 | – | 4.7 |\n| Written options | 2,887.7 | – | – | – | 82.8 | – | 82.8 |\n| Purchased options | 3,026.2 | 88.0 | – | 88.0 | – | – | – |\n| Foreign exchange contracts | | | | | | | |\n| Swaps | 630.1 | 26.5 | 3.7 | 30.2 | 28.5 | 2.1 | 30.6 |\n| Spot, futures and forwards | 2,652.9 | 41.3 | – | 41.3 | 44.2 | – | 44.2 |\n| Written options | 439.6 | – | – | – | 13.2 | – | 13.2 |\n| Purchased options | 417.1 | 13.0 | – | 13.0 | – | – | – |\n| Equity contracts | | | | | | | |\n| Swaps | 42.4 | 1.7 | – | 1.7 | 2.0 | – | 2.0 |\n| Futures and forwards | 78.8 | 2.9 | – | 2.9 | 2.1 | – | 2.1 |\n| Written options | 242.7 | – | – | – | 19.4 | – | 19.4 |\n| Purchased options | 193.5 | 21.5 | – | 21.5 | – | – | – |\n| Commodity contracts | | | | | | | |\n| Swaps | 90.2 | 8.8 | 0.2 | 9.0 | 9.3 | – | 9.3 |\n| Futures and forwards | 413.7 | 4.1 | – | 4.1 | 2.8 | – | 2.8 |\n| Written options | 86.3 | – | – | – | 6.7 | – | 6.7 |\n| Purchased options | 84.6 | 6.6 | – | 6.6 | – | – | – |\n| Credit derivatives | | | | | | | |\n| Purchased credit derivatives: | | | | | | | |\n| Credit default swaps | 2,184.7 | 69.8 | – | 69.8 | 34.0 | – | 34.0 |\n| Total return swaps/other | 26.0 | 0.9 | – | 0.9 | 0.2 | – | 0.2 |\n| Written credit derivatives: | | | | | | | |\n| Credit default swaps | 2,133.5 | 33.3 | – | 33.3 | 63.2 | – | 63.2 |\n| Total return swaps/other | 22.5 | 0.5 | – | 0.5 | 0.5 | – | 0.5 |\n| Gross derivative assets/liabilities | | $1,518.8 | $18.8 | $1,537.6 | $1,501.5 | $4.3 | $1,505.8 |\n| Less: Legally enforceable master netting agreements | | | | -1,406.3 | | | -1,406.3 |\n| Less: Cash collateral applied | | | | -58.3 | | | -43.6 |\n| Total derivative assets/liabilities | | | | $73.0 | | | $55.9 |\n(1) Represents the total contract/notional amount of derivative assets and liabilities outstanding.\n(2) Excludes $4.1 billion of long-term debt designated as a hedge of foreign currency risk."]}, {"source_sentence": "Instruct: Given a web search query, retrieve relevant passages that answer the query.\nQuery: Title: \nText: What does the deferred income taxes reflect?", "sentences": ["Title: \nText: 19. INCOME TAXES\nDeferred income taxes reflect the net effect of temporary differences between the carrying amounts of assets and liabilities for financial reporting purposes and amounts used for income tax purposes. The components of our deferred tax assets and liabilities are as follows (in thousands):\nAs of September 27, 2019, we had $923.4 million of gross federal net operating loss (\"NOL\") carryforwards consisting of $479.2 million relating to the AppliedMicro Acquisition, $158.9 million relating to our acquisition of Mindspeed Technologies, Inc. in 2013, $26.2 million relating to our acquisition of BinOptics Corporation in 2014 and $259.1 million relating to losses generated by MACOM.\nThe federal NOL carryforwards will expire at various dates through 2037 for losses generated prior to the tax period ended September 28, 2018. For losses generated during the tax period ended September 28, 2018 and future years, the NOL carryforward period is infinite. The reported net operating loss carryforward includes any limitation under Sections 382 and 383 of the Internal Revenue Code of 1986, as amended, which applies to an ownership change as defined under Section 382.\n\n | September 27, 2019 | September 28, 2018\n---------------------------------------------------- | ------------------ | ------------------\nDeferred tax assets (liabilities): | | \nFederal and foreign net operating losses and credits | $263,199 | $321,982 \n Intangible assets | 9,887 | (94,929) \n Property and equipment | (1,473) | (6,293) \nOther non-current deferred tax assets | 16,933 | 13,850 \nDeferred compensation | — | 3,810 \nDeferred gain | — | 6,575 \nInterest | 7,170 | — \n Valuation allowance | (252,536) | (243,112) \nTotal deferred tax asset | $43,180 | $1,883 ", "Title: \nText: The following table presents the components of impairment and other charges, net, in each fiscal year (in thousands):\nRestructuring costs decreased by $2.2 million as a result of lower severance expenses, as our general and administrative cost reduction initiative came to its conclusion as planned. Costs of closed restaurants and other increased by $3.8 million, primarily due to a $3.5 million charge recorded in 2019 related to the write- off of software development costs associated with a discontinued technology project. Gains on disposition of property and equipment, net, increased by $7.9 million, primarily due to a $5.7 million gain related to a sale of property and a$0.8 million gain related to an eminent domain transaction in 2019.\nRefer to Note 9, Impairment and Other Charges, Net, of the notes to the consolidated financial statements for additional information regarding these charges.\n\n | 2019 | 2018 \n------------------------------------------------------------ | ------- | -------\nRestructuring costs | $8,455 | $10,647\nCosts of closed restaurants and other | 8,628 | 4,803 \n(Gains) losses on disposition of property and equipment, net | (6,244) | 1,627 \nAccelerated depreciation | 1,616 | 1,130 \nOperating restaurant impairment charges | — | 211 \n | $12,455 | $18,418", "Title: \nText: MetLife, Inc. Notes to Consolidated Financial Statements (Continued) $4.3 billion, of which $1.6 billion is deductible for income tax purposes.\nFurther information on goodwill is described in Note 6.\nSee Note 5 for the VOBA acquired as part of the acquisition and Note 7 for the value of distribution agreements (VODA) and the value of customer relationships acquired (VOCRA).\nAs part of the integration of Travelers operations, management approved and initiated plans to reduce approximately 1,000 domestic and international Travelers positions, which was completed in December 2006.\nMetLife initially recorded restructuring costs, including severance, relocation and outplacement services of Travelers employees, as liabilities assumed in the purchase business combination of $49 million.\nFor the years ended December 31, 2006 and 2005, the liability for restructuring costs was reduced by $4 million and $1 million, respectively, due to a reduction in the estimate of severance benefits to be paid to Travelers employees.\nThe restructuring costs associated with the Travelers acquisition were as follows:\n| | Years Ended December 31, |\n| | 2006 | 2005 |\n| | (In millions) |\n| Balance at January 1, | $28 | $— |\n| Acquisition | — | 49 |\n| Cash payments | -24 | -20 |\n| Other reductions | -4 | -1 |\n| Balance at December 31, | $— | $28 |\nOther Acquisitions and Dispositions On June 28, 2007, the Company acquired the remaining 50% interest in a joint venture in Hong Kong, MetLife Fubon Limited (MetLife Fubon), for $56 million in cash, resulting in MetLife Fubon becoming a consolidated subsidiary of the Company.\nThe transaction was treated as a step acquisition, and at June 30, 2007, total assets and liabilities of MetLife Fubon of $839 million and $735 million, respectively, were included in the Companys consolidated balance sheet.\nThe Companys investment for the initial 50% interest in MetLife Fubon was $48 million.\nThe Company used the equity method of accounting for such investment in MetLife Fubon.\nThe Companys share of the joint ventures results for the six months ended June 30, 2007, was a loss of $3 million.\nThe fair value of the assets acquired and the liabilities assumed in the step acquisition at June 30, 2007, was $427 million and $371 million, respectively.\nNo additional goodwill was recorded as a part of the step acquisition.\nAs a result of this acquisition, additional VOBA and VODA of $45 million and $5 million, respectively, were recorded and both have a weighted average amortization period of 16 years.\nFurther information on VOBA and VODA is described in Note 5 and Note 7, respectively.\nOn June 1, 2007, the Company completed the sale of its Bermuda insurance subsidiary, MetLife International Insurance, Ltd. (MLII), to a third party for $33 million in cash consideration, resulting in a gain upon disposal of $3 million, net of income tax.\nThe net assets of MLII at disposal were $27 million.\nA liability of $1 million was recorded with respect to a guarantee provided in connection with this disposition.\nFurther information on guarantees is described in Note 16.\nOn September 1, 2005, the Company completed the acquisition of CitiStreet Associates, a division of CitiStreet LLC, which is primarily involved in the distribution of annuity products and retirement plans to the education, healthcare, and not-for-profit markets, for $56 million, of which $2 million was allocated to goodwill and $54 million to other identifiable intangibles, specifically the value of customer relationships acquired, which have a weighted average amortization period of 16 years.\nCitiStreet Associates was integrated with MetLife Resources, a focused distribution channel of MetLife, which is dedicated to provide retirement plans and financial services to the same markets.\nFurther information on goodwill and VOCRA is described in Note 6 and Note 7, respectively.\nSee Note 23 for information on the disposition of the annuities and pension businesses of MetLife Insurance Limited (MetLife Australia), P. T. Sejahtera (MetLife Indonesia) and SSRM Holdings, Inc. (SSRM).\nSee Note 25 for information on the Companys acquisitions subsequent to December 31, 2007."]}, {"source_sentence": "Instruct: Given a web search query, retrieve relevant passages that answer the query.\nQuery: Title: \nText: In the year with lowest amount of Deposits with banks Average volume, what's the increasing rate of Deposits with banks Average volume?", "sentences": ["Title: \nText: Issuer Purchases of Equity Securities Repurchases of common stock are made to support the Companys stock-based employee compensation plans and for other corporate purposes.\nOn February 13, 2006, the Board of Directors authorized the purchase of $2.0 billion of the Companys common stock between February 13, 2006 and February 28, 2007.\nIn August 2006, 3Ms Board of Directors authorized the repurchase of an additional $1.0 billion in share repurchases, raising the total authorization to $3.0 billion for the period from February 13, 2006 to February 28, 2007.\nIn February 2007, 3Ms Board of Directors authorized a twoyear share repurchase of up to $7.0 billion for the period from February 12, 2007 to February 28, 2009.", "Title: \nText: Additional Interest Rate Details Average Balances and Interest Ratesé¥æssets(1)(2)(3)(4)\n| | Average volume | Interest revenue | % Average rate |\n| In millions of dollars, except rates | 2015 | 2014 | 2013 | 2015 | 2014 | 2013 | 2015 | 2014 | 2013 |\n| Assets | | | | | | | | | |\n| Deposits with banks-5 | $133,790 | $161,359 | $144,904 | $727 | $959 | $1,026 | 0.54% | 0.59% | 0.71% |\n| Federal funds sold and securities borrowed or purchased under agreements to resell-6 | | | | | | | | | |\n| In U.S. offices | $150,359 | $153,688 | $158,237 | $1,211 | $1,034 | $1,133 | 0.81% | 0.67% | 0.72% |\n| In offices outside the U.S.-5 | 84,006 | 101,177 | 109,233 | 1,305 | 1,332 | 1,433 | 1.55 | 1.32 | 1.31 |\n| Total | $234,365 | $254,865 | $267,470 | $2,516 | $2,366 | $2,566 | 1.07% | 0.93% | 0.96% |\n| Trading account assets-7(8) | | | | | | | | | |\n| In U.S. offices | $114,639 | $114,910 | $126,123 | $3,945 | $3,472 | $3,728 | 3.44% | 3.02% | 2.96% |\n| In offices outside the U.S.-5 | 103,348 | 119,801 | 127,291 | 2,141 | 2,538 | 2,683 | 2.07 | 2.12 | 2.11 |\n| Total | $217,987 | $234,711 | $253,414 | $6,086 | $6,010 | $6,411 | 2.79% | 2.56% | 2.53% |\n| Investments | | | | | | | | | |\n| In U.S. offices | | | | | | | | | |\n| Taxable | $214,714 | $188,910 | $174,084 | $3,812 | $3,286 | $2,713 | 1.78% | 1.74% | 1.56% |\n| Exempt from U.S. income tax | 20,034 | 20,386 | 18,075 | 443 | 626 | 811 | 2.21 | 3.07 | 4.49 |\n| In offices outside the U.S.-5 | 102,376 | 113,163 | 114,122 | 3,071 | 3,627 | 3,761 | 3.00 | 3.21 | 3.30 |\n| Total | $337,124 | $322,459 | $306,281 | $7,326 | $7,539 | $7,285 | 2.17% | 2.34% | 2.38% |\n| Loans (net of unearned income)(9) | | | | | | | | | |\n| In U.S. offices | $354,439 | $361,769 | $354,707 | $24,558 | $26,076 | $25,941 | 6.93% | 7.21% | 7.31% |\n| In offices outside the U.S.-5 | 273,072 | 296,656 | 292,852 | 15,988 | 18,723 | 19,660 | 5.85 | 6.31 | 6.71 |\n| Total | $627,511 | $658,425 | $647,559 | $40,546 | $44,799 | $45,601 | 6.46% | 6.80% | 7.04% |\n| Other interest-earning assets-10 | $55,060 | $40,375 | $38,233 | $1,839 | $507 | $602 | 3.34% | 1.26% | 1.57% |\n| Total interest-earning assets | $1,605,837 | $1,672,194 | $1,657,861 | $59,040 | $62,180 | $63,491 | 3.68% | 3.72% | 3.83% |\n| Non-interest-earning assets-7 | $218,000 | $224,721 | $222,526 | | | | | | |\n| Total assets from discontinued operations | — | — | 2,909 | | | | | | |\n| Total assets | $1,823,837 | $1,896,915 | $1,883,296 | | | | | | |\nNet interest revenue includes the taxable equivalent adjustments related to the tax-exempt bond portfolio (based on the U. S. federal statutory tax rate of 35%) of $487 million, $498 million and $521 million for 2015, 2014 and 2013, respectively.\nInterest rates and amounts include the effects of risk management activities associated with the respective asset categories.\nMonthly or quarterly averages have been used by certain subsidiaries where daily averages are unavailable.\nDetailed average volume, Interest revenue and Interest expense exclude Discontinued operations.\nSee Note 2 to the Consolidated Financial Statements.\nAverage rates reflect prevailing local interest rates, including inflationary effects and monetary corrections in certain countries.\nAverage volumes of securities borrowed or purchased under agreements to resell are reported net pursuant to ASC 210-20-45.\nHowever, Interest revenue excludes the impact of ASC 210-20-45.\nThe fair value carrying amounts of derivative contracts are reported net, pursuant to ASC 815-10-45, in Non-interest-earning assets and Other non-interest bearing liabilities.\nInterest expense on Trading account liabilities of ICG is reported as a reduction of Interest revenue.\nInterest revenue and Interest expense on cash collateral positions are reported in interest on Trading account assets and Trading account liabilities, respectively.\nIncludes cash-basis loans.\nIncludes brokerage receivables.\nDuring 2015, continued management actions, primarily the sale or transfer to held-for-sale of approximately $1.5 billion of delinquent residential first mortgages, including $0.9 billion in the fourth quarter largely associated with the transfer of CitiFinancial loans to held-for-sale referenced above, were the primary driver of the overall improvement in delinquencies within Citi Holdings residential first mortgage portfolio.\nCredit performance from quarter to quarter could continue to be impacted by the amount of delinquent loan sales or transfers to held-for-sale, as well as overall trends in HPI and interest rates.\nNorth America Residential First MortgagesState Delinquency Trends The following tables set forth the six U. S. states and/or regions with the highest concentration of Citis residential first mortgages.\n| In billions of dollars | December 31, 2015 | December 31, 2014 |\n| State-1 | ENR-2 | ENRDistribution | 90+DPD% | %LTV >100%-3 | RefreshedFICO | ENR-2 | ENRDistribution | 90+DPD% | %LTV >100%-3 | RefreshedFICO |\n| CA | $19.2 | 37% | 0.2% | 1% | 754 | $18.9 | 31% | 0.6% | 2% | 745 |\n| NY/NJ/CT-4 | 12.7 | 25 | 0.8 | 1 | 751 | 12.2 | 20 | 1.9 | 2 | 740 |\n| VA/MD | 2.2 | 4 | 1.2 | 2 | 719 | 3.0 | 5 | 3.0 | 8 | 695 |\n| IL-4 | 2.2 | 4 | 1.0 | 3 | 735 | 2.5 | 4 | 2.5 | 9 | 713 |\n| FL-4 | 2.2 | 4 | 1.1 | 4 | 723 | 2.8 | 5 | 3.0 | 14 | 700 |\n| TX | 1.9 | 4 | 1.0 | — | 711 | 2.5 | 4 | 2.7 | — | 680 |\n| Other | 11.0 | 21 | 1.3 | 2 | 710 | 18.2 | 30 | 3.3 | 7 | 677 |\n| Total-5 | $51.5 | 100% | 0.7% | 1% | 738 | $60.1 | 100% | 2.1% | 4% | 715 |\nNote: Totals may not sum due to rounding.\n(1) Certain of the states are included as part of a region based on Citis view of similar HPI within the region.\n(2) Ending net receivables.\nExcludes loans in Canada and Puerto Rico, loans guaranteed by U. S. government agencies, loans recorded at fair value and loans subject to long term standby commitments (LTSCs).\nExcludes balances for which FICO or LTV data are unavailable.\n(3) LTV ratios (loan balance divided by appraised value) are calculated at origination and updated by applying market price data.\n(4) New York, New Jersey, Connecticut, Florida and Illinois are judicial states.\n(5) Improvement in state trends during 2015 was primarily due to the sale or transfer to held-for-sale of residential first mortgages, including the transfer of CitiFinancial residential first mortgages to held-for-sale in the fourth quarter of 2015.\nForeclosures A substantial majority of Citis foreclosure inventory consists of residential first mortgages.\nAt December 31, 2015, Citis foreclosure inventory included approximately $0.1 billion, or 0.2%, of the total residential first mortgage portfolio, compared to $0.6 billion, or 0.9%, at December 31, 2014, based on the dollar amount of ending net receivables of loans in foreclosure inventory, excluding loans that are guaranteed by U. S. government agencies and loans subject to LTSCs.\nNorth America Consumer Mortgage Quarterly Credit Trends Net Credit Losses and DelinquenciesHome Equity Loans Citis home equity loan portfolio consists of both fixed-rate home equity loans and loans extended under home equity lines of credit.\nFixed-rate home equity loans are fully amortizing.\nHome equity lines of credit allow for amounts to be drawn for a period of time with the payment of interest only and then, at the end of the draw period, the then-outstanding amount is converted to an amortizing loan (the interest-only payment feature during the revolving period is standard for this product across the industry).\nAfter conversion, the home equity loans typically have a 20-year amortization period.\nAs of December 31, 2015, Citis home equity loan portfolio of $22.8 billion consisted of $6.3 billion of fixed-rate home equity loans and $16.5 billion of loans extended under home equity lines of credit (Revolving HELOCs).", "Title: \nText: | (Dollar amounts in thousands) | Rate | | Principal Amount of Subordinated Note/ Debenture Issued to Trust -1 | Investment in Unconsolidated Subsidiary -2 |\n| Huntington Capital I | 0.99 | -3 | $138,816 | $6,186 |\n| Huntington Capital II | 0.93 | -4 | 60,093 | 3,093 |\n| Huntington Capital III | 6.69 | | 114,072 | 10 |\n| BancFirst Ohio Trust Preferred | 8.54 | | 23,248 | 619 |\n| Sky Financial Capital Trust I | 8.52 | | 64,474 | 1,856 |\n| Sky Financial Capital Trust II | 3.52 | -5 | 30,929 | 929 |\n| Sky Financial Capital Trust III | 1.28 | -6 | 77,481 | 2,320 |\n| Sky Financial Capital Trust IV | 1.27 | -6 | 77,482 | 2,320 |\n| Prospect Trust I | 3.54 | -7 | 6,186 | 186 |\n| Total | | | $592,781 | $17,519 |\n(1) Represents the principal amount of debentures issued to each trust, including unamortized original issue discount.\n(2) Huntingtons investment in the unconsolidated trusts represents the only risk of loss.\n(3) Variable effective rate at December 31, 2010, based on three month LIBOR + 0.70.\n(4) Variable effective rate at December 31, 2010, based on three month LIBOR + 0.625.\n(5) Variable effective rate at December 31, 2010, based on three month LIBOR + 2.95.\n(6) Variable effective rate at December 31, 2010, based on three month LIBOR + 1.40.\n(7) Variable effective rate at December 31, 2010, based on three month LIBOR + 3.25.\nEach issue of the junior subordinated debentures has an interest rate equal to the corresponding trust securities distribution rate.\nHuntington has the right to defer payment of interest on the debentures at any time, or from time to time for a period not exceeding five years, provided that no extension period may extend beyond the stated maturity of the related debentures.\nDuring any such extension period, distributions to the trust securities will also be deferred and Huntingtons ability to pay dividends on its common stock will be restricted.\nPeriodic cash payments and payments upon liquidation or redemption with respect to trust securities are guaranteed by Huntington to the extent of funds held by the trusts.\nThe guarantee ranks subordinate and junior in right of payment to all indebtedness of the Company to the same extent as the junior subordinated debt.\nThe guarantee does not place a limitation on the amount of additional indebtedness that may be incurred by Huntington.\nLow Income Housing Tax Credit Partnerships Huntington makes certain equity investments in various limited partnerships that sponsor affordable housing projects utilizing the Low Income Housing Tax Credit pursuant to Section 42 of the Internal Revenue Code.\nThe purpose of these investments is to achieve a satisfactory return on capital, to facilitate the sale of additional affordable housing product offerings, and to assist in achieving goals associated with the Community Reinvestment Act.\nThe primary activities of the limited partnerships include the identification, development, and operation of multi-family housing that is leased to qualifying residential tenants.\nGenerally, these types of investments are funded through a combination of debt and equity.\nHuntington does not own a majority of the limited partnership interests in these entities and is not the primary beneficiary.\nHuntington uses the equity method to account for the majority of its investments in these entities.\nThese investments are included in accrued income and other assets.\nAt December 31, 2010 and 2009, Huntington has commitments of $316.0 million and $285.3 million, respectively, of which $260.1 million and"]}], "model-index": [{"name": "SentenceTransformer based on thomaskim1130/stella_en_400M_v5-FinanceRAG", "results": [{"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "Evaluate", "type": "Evaluate"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.46359223300970875, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.6820388349514563, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.7597087378640777, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.8519417475728155, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.46359223300970875, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.25647249190938515, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.17766990291262133, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.10242718446601942, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.4095469255663431, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.6423948220064724, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.7298543689320389, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.8398462783171521, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.6409313886654548, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.5902248035136388, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.5753196287486457, "name": "Cosine Map@100"}, {"type": "dot_accuracy@1", "value": 0.4393203883495146, "name": "Dot Accuracy@1"}, {"type": "dot_accuracy@3", "value": 0.6747572815533981, "name": "Dot Accuracy@3"}, {"type": "dot_accuracy@5", "value": 0.7354368932038835, "name": "Dot Accuracy@5"}, {"type": "dot_accuracy@10", "value": 0.8422330097087378, "name": "Dot Accuracy@10"}, {"type": "dot_precision@1", "value": 0.4393203883495146, "name": "Dot Precision@1"}, {"type": "dot_precision@3", "value": 0.25, "name": "Dot Precision@3"}, {"type": "dot_precision@5", "value": 0.17087378640776701, "name": "Dot Precision@5"}, {"type": "dot_precision@10", "value": 0.09975728155339807, "name": "Dot Precision@10"}, {"type": "dot_recall@1", "value": 0.3828478964401295, "name": "Dot Recall@1"}, {"type": "dot_recall@3", "value": 0.6338187702265372, "name": "Dot Recall@3"}, {"type": "dot_recall@5", "value": 0.7005258899676375, "name": "Dot Recall@5"}, {"type": "dot_recall@10", "value": 0.8223705501618123, "name": "Dot Recall@10"}, {"type": "dot_ndcg@10", "value": 0.6194906173849263, "name": "Dot Ndcg@10"}, {"type": "dot_mrr@10", "value": 0.5711877793188473, "name": "Dot Mrr@10"}, {"type": "dot_map@100", "value": 0.5528301778009912, "name": "Dot Map@100"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,158 |
bobbobok/privateGPT
|
bobbobok
| null |
[
"region:us"
] | 2023-09-26T01:29:24Z |
2023-09-26T01:30:41+00:00
| 0 | 0 |
---
{}
---
privateGPT
Ask questions to your documents without an internet connection, using the power of LLMs. 100% private, no data leaves your execution environment at any point. You can ingest documents and ask questions without an internet connection!
👂 Need help applying PrivateGPT to your specific use case? Let us know more about it and we'll try to help! We are refining PrivateGPT through your feedback.
demo
Built with LangChain, LlamaIndex, GPT4All, LlamaCpp, Chroma and SentenceTransformers.
Environment Setup
In order to set your environment up to run the code here, first install all requirements:
pip3 install -r requirements.txt
Alternative requirements installation with poetry
Install poetry
Run this commands
cd privateGPT
poetry install
poetry shell
Then, download the LLM model and place it in a directory of your choice:
LLM: default to ggml-gpt4all-j-v1.3-groovy.bin. If you prefer a different GPT4All-J compatible model, just download it and reference it in your .env file.
Copy the example.env template into .env
cp example.env .env
and edit the variables appropriately in the .env file.
MODEL_TYPE: supports LlamaCpp or GPT4All
PERSIST_DIRECTORY: is the folder you want your vectorstore in
MODEL_PATH: Path to your GPT4All or LlamaCpp supported LLM
MODEL_N_CTX: Maximum token limit for the LLM model
MODEL_N_BATCH: Number of tokens in the prompt that are fed into the model at a time. Optimal value differs a lot depending on the model (8 works well for GPT4All, and 1024 is better for LlamaCpp)
EMBEDDINGS_MODEL_NAME: SentenceTransformers embeddings model name (see https://www.sbert.net/docs/pretrained_models.html)
TARGET_SOURCE_CHUNKS: The amount of chunks (sources) that will be used to answer a question
Note: because of the way langchain loads the SentenceTransformers embeddings, the first time you run the script it will require internet connection to download the embeddings model itself.
Test dataset
This repo uses a state of the union transcript as an example.
Instructions for ingesting your own dataset
Put any and all your files into the source_documents directory
The supported extensions are:
.csv: CSV,
.docx: Word Document,
.doc: Word Document,
.enex: EverNote,
.eml: Email,
.epub: EPub,
.html: HTML File,
.md: Markdown,
.msg: Outlook Message,
.odt: Open Document Text,
.pdf: Portable Document Format (PDF),
.pptx : PowerPoint Document,
.ppt : PowerPoint Document,
.txt: Text file (UTF-8),
Run the following command to ingest all the data.
python ingest.py
Output should look like this:
Creating new vectorstore
Loading documents from source_documents
Loading new documents: 100%|██████████████████████| 1/1 [00:01<00:00, 1.73s/it]
Loaded 1 new documents from source_documents
Split into 90 chunks of text (max. 500 tokens each)
Creating embeddings. May take some minutes...
Using embedded DuckDB with persistence: data will be stored in: db
Ingestion complete! You can now run privateGPT.py to query your documents
It will create a db folder containing the local vectorstore. Will take 20-30 seconds per document, depending on the size of the document. You can ingest as many documents as you want, and all will be accumulated in the local embeddings database. If you want to start from an empty database, delete the db folder.
Note: during the ingest process no data leaves your local environment. You could ingest without an internet connection, except for the first time you run the ingest script, when the embeddings model is downloaded.
Ask questions to your documents, locally!
In order to ask a question, run a command like:
python privateGPT.py
And wait for the script to require your input.
> Enter a query:
Hit enter. You'll need to wait 20-30 seconds (depending on your machine) while the LLM model consumes the prompt and prepares the answer. Once done, it will print the answer and the 4 sources it used as context from your documents; you can then ask another question without re-running the script, just wait for the prompt again.
Note: you could turn off your internet connection, and the script inference would still work. No data gets out of your local environment.
Type exit to finish the script.
CLI
The script also supports optional command-line arguments to modify its behavior. You can see a full list of these arguments by running the command python privateGPT.py --help in your terminal.
How does it work?
Selecting the right local models and the power of LangChain you can run the entire pipeline locally, without any data leaving your environment, and with reasonable performance.
ingest.py uses LangChain tools to parse the document and create embeddings locally using HuggingFaceEmbeddings (SentenceTransformers). It then stores the result in a local vector database using Chroma vector store.
privateGPT.py uses a local LLM based on GPT4All-J or LlamaCpp to understand questions and create answers. The context for the answers is extracted from the local vector store using a similarity search to locate the right piece of context from the docs.
GPT4All-J wrapper was introduced in LangChain 0.0.162.
System Requirements
Python Version
To use this software, you must have Python 3.10 or later installed. Earlier versions of Python will not compile.
C++ Compiler
If you encounter an error while building a wheel during the pip install process, you may need to install a C++ compiler on your computer.
For Windows 10/11
To install a C++ compiler on Windows 10/11, follow these steps:
Install Visual Studio 2022.
Make sure the following components are selected:
Universal Windows Platform development
C++ CMake tools for Windows
Download the MinGW installer from the MinGW website.
Run the installer and select the gcc component.
Mac Running Intel
When running a Mac with Intel hardware (not M1), you may run into clang: error: the clang compiler does not support '-march=native' during pip install.
If so set your archflags during pip install. eg: ARCHFLAGS="-arch x86_64" pip3 install -r requirements.txt
Disclaimer
This is a test project to validate the feasibility of a fully private solution for question answering using LLMs and Vector embeddings. It is not production ready, and it is not meant to be used in production. The models selection is not optimized for performance, but for privacy; but it is possible to use different models and vectorstores to improve performance.
| null |
Non_BioNLP
|
privateGPT
Ask questions to your documents without an internet connection, using the power of LLMs. 100% private, no data leaves your execution environment at any point. You can ingest documents and ask questions without an internet connection!
👂 Need help applying PrivateGPT to your specific use case? Let us know more about it and we'll try to help! We are refining PrivateGPT through your feedback.
demo
Built with LangChain, LlamaIndex, GPT4All, LlamaCpp, Chroma and SentenceTransformers.
Environment Setup
In order to set your environment up to run the code here, first install all requirements:
pip3 install -r requirements.txt
Alternative requirements installation with poetry
Install poetry
Run this commands
cd privateGPT
poetry install
poetry shell
Then, download the LLM model and place it in a directory of your choice:
LLM: default to ggml-gpt4all-j-v1.3-groovy.bin. If you prefer a different GPT4All-J compatible model, just download it and reference it in your .env file.
Copy the example.env template into .env
cp example.env .env
and edit the variables appropriately in the .env file.
MODEL_TYPE: supports LlamaCpp or GPT4All
PERSIST_DIRECTORY: is the folder you want your vectorstore in
MODEL_PATH: Path to your GPT4All or LlamaCpp supported LLM
MODEL_N_CTX: Maximum token limit for the LLM model
MODEL_N_BATCH: Number of tokens in the prompt that are fed into the model at a time. Optimal value differs a lot depending on the model (8 works well for GPT4All, and 1024 is better for LlamaCpp)
EMBEDDINGS_MODEL_NAME: SentenceTransformers embeddings model name (see https://www.sbert.net/docs/pretrained_models.html)
TARGET_SOURCE_CHUNKS: The amount of chunks (sources) that will be used to answer a question
Note: because of the way langchain loads the SentenceTransformers embeddings, the first time you run the script it will require internet connection to download the embeddings model itself.
Test dataset
This repo uses a state of the union transcript as an example.
Instructions for ingesting your own dataset
Put any and all your files into the source_documents directory
The supported extensions are:
.csv: CSV,
.docx: Word Document,
.doc: Word Document,
.enex: EverNote,
.eml: Email,
.epub: EPub,
.html: HTML File,
.md: Markdown,
.msg: Outlook Message,
.odt: Open Document Text,
.pdf: Portable Document Format (PDF),
.pptx : PowerPoint Document,
.ppt : PowerPoint Document,
.txt: Text file (UTF-8),
Run the following command to ingest all the data.
python ingest.py
Output should look like this:
Creating new vectorstore
Loading documents from source_documents
Loading new documents: 100%|██████████████████████| 1/1 [00:01<00:00, 1.73s/it]
Loaded 1 new documents from source_documents
Split into 90 chunks of text (max. 500 tokens each)
Creating embeddings. May take some minutes...
Using embedded DuckDB with persistence: data will be stored in: db
Ingestion complete! You can now run privateGPT.py to query your documents
It will create a db folder containing the local vectorstore. Will take 20-30 seconds per document, depending on the size of the document. You can ingest as many documents as you want, and all will be accumulated in the local embeddings database. If you want to start from an empty database, delete the db folder.
Note: during the ingest process no data leaves your local environment. You could ingest without an internet connection, except for the first time you run the ingest script, when the embeddings model is downloaded.
Ask questions to your documents, locally!
In order to ask a question, run a command like:
python privateGPT.py
And wait for the script to require your input.
> Enter a query:
Hit enter. You'll need to wait 20-30 seconds (depending on your machine) while the LLM model consumes the prompt and prepares the answer. Once done, it will print the answer and the 4 sources it used as context from your documents; you can then ask another question without re-running the script, just wait for the prompt again.
Note: you could turn off your internet connection, and the script inference would still work. No data gets out of your local environment.
Type exit to finish the script.
CLI
The script also supports optional command-line arguments to modify its behavior. You can see a full list of these arguments by running the command python privateGPT.py --help in your terminal.
How does it work?
Selecting the right local models and the power of LangChain you can run the entire pipeline locally, without any data leaving your environment, and with reasonable performance.
ingest.py uses LangChain tools to parse the document and create embeddings locally using HuggingFaceEmbeddings (SentenceTransformers). It then stores the result in a local vector database using Chroma vector store.
privateGPT.py uses a local LLM based on GPT4All-J or LlamaCpp to understand questions and create answers. The context for the answers is extracted from the local vector store using a similarity search to locate the right piece of context from the docs.
GPT4All-J wrapper was introduced in LangChain 0.0.162.
System Requirements
Python Version
To use this software, you must have Python 3.10 or later installed. Earlier versions of Python will not compile.
C++ Compiler
If you encounter an error while building a wheel during the pip install process, you may need to install a C++ compiler on your computer.
For Windows 10/11
To install a C++ compiler on Windows 10/11, follow these steps:
Install Visual Studio 2022.
Make sure the following components are selected:
Universal Windows Platform development
C++ CMake tools for Windows
Download the MinGW installer from the MinGW website.
Run the installer and select the gcc component.
Mac Running Intel
When running a Mac with Intel hardware (not M1), you may run into clang: error: the clang compiler does not support '-march=native' during pip install.
If so set your archflags during pip install. eg: ARCHFLAGS="-arch x86_64" pip3 install -r requirements.txt
Disclaimer
This is a test project to validate the feasibility of a fully private solution for question answering using LLMs and Vector embeddings. It is not production ready, and it is not meant to be used in production. The models selection is not optimized for performance, but for privacy; but it is possible to use different models and vectorstores to improve performance.
|
{}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,160 |
Aryadha/Policy_summariser
|
Aryadha
|
summarization
|
[
"bertopic",
"medical",
"legal",
"summarization",
"en",
"dataset:fka/awesome-chatgpt-prompts",
"dataset:HuggingFaceM4/WebSight",
"license:mit",
"region:us"
] | 2024-01-20T10:02:24Z |
2024-01-20T10:41:31+00:00
| 0 | 0 |
---
datasets:
- fka/awesome-chatgpt-prompts
- HuggingFaceM4/WebSight
language:
- en
library_name: bertopic
license: mit
metrics:
- accuracy
- character
pipeline_tag: summarization
tags:
- medical
- legal
---
| null |
Non_BioNLP
|
{"datasets": ["fka/awesome-chatgpt-prompts", "HuggingFaceM4/WebSight"], "language": ["en"], "library_name": "bertopic", "license": "mit", "metrics": ["accuracy", "character"], "pipeline_tag": "summarization", "tags": ["medical", "legal"]}
|
task
|
[
"SUMMARIZATION"
] | 45,161 |
|
context-mt/scat-marian-big-target-ctx4-cwd0-en-fr
|
context-mt
|
translation
|
[
"transformers",
"pytorch",
"tensorboard",
"marian",
"text2text-generation",
"arxiv:2310.01188",
"contextual-mt",
"document-mt",
"translation",
"en",
"fr",
"dataset:inseq/scat",
"dataset:gsarti/iwslt2017_context",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-06-28T12:08:38Z |
2024-03-10T12:39:40+00:00
| 19 | 0 |
---
datasets:
- inseq/scat
- gsarti/iwslt2017_context
language:
- en
- fr
license: apache-2.0
pipeline_tag: translation
tags:
- arxiv:2310.01188
- contextual-mt
- document-mt
---
*This model corresponds to the [`Helsinki-NLP/opus-mt-tc-big-en-fr`](https://huggingface.co/Helsinki-NLP/opus-mt-tc-big-en-fr) further trained on English-to-French translation on the [IWSLT17 dataset](https://huggingface.co/datasets/gsarti/iwslt2017_context) with context tags using the format:
```
Input: SOURCE_CTX <brk> SOURCE_CURR
Output: TARGET_CTX <brk> TARGET_CURR
```
and further fine-tuned on the training split of [SCAT+](https://huggingface.co/datasets/inseq/scat). The model was used in the evaluation of the paper [Quantifying the Plausibility of Context Reliance in Neural Machine Translation](https://openreview.net/forum?id=XTHfNGI3zT) published at ICLR 2024, also available on [Arxiv](https://arxiv.org/abs/2310.01188). It can be used for English to French contextual and non-contextual translation.
| null |
Non_BioNLP
|
*This model corresponds to the [`Helsinki-NLP/opus-mt-tc-big-en-fr`](https://huggingface.co/Helsinki-NLP/opus-mt-tc-big-en-fr) further trained on English-to-French translation on the [IWSLT17 dataset](https://huggingface.co/datasets/gsarti/iwslt2017_context) with context tags using the format:
```
Input: SOURCE_CTX <brk> SOURCE_CURR
Output: TARGET_CTX <brk> TARGET_CURR
```
and further fine-tuned on the training split of [SCAT+](https://huggingface.co/datasets/inseq/scat). The model was used in the evaluation of the paper [Quantifying the Plausibility of Context Reliance in Neural Machine Translation](https://openreview.net/forum?id=XTHfNGI3zT) published at ICLR 2024, also available on [Arxiv](https://arxiv.org/abs/2310.01188). It can be used for English to French contextual and non-contextual translation.
|
{"datasets": ["inseq/scat", "gsarti/iwslt2017_context"], "language": ["en", "fr"], "license": "apache-2.0", "pipeline_tag": "translation", "tags": ["arxiv:2310.01188", "contextual-mt", "document-mt"]}
|
task
|
[
"TRANSLATION"
] | 45,162 |
Lucius-Morningstar/Morningstar-Omega
|
Lucius-Morningstar
|
text2text-generation
|
[
"Neuroscience",
"chemistry",
"code",
"text2text-generation",
"dataset:neuralwork/arxiver",
"arxiv:1910.09700",
"base_model:nvidia/Llama-3.1-Nemotron-70B-Instruct",
"base_model:finetune:nvidia/Llama-3.1-Nemotron-70B-Instruct",
"doi:10.57967/hf/3369",
"license:mit",
"region:us"
] | 2024-10-29T06:54:18Z |
2024-10-30T01:49:00+00:00
| 0 | 1 |
---
base_model:
- nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
- nvidia/Llama-3.1-Nemotron-70B-Instruct
datasets:
- neuralwork/arxiver
license: mit
pipeline_tag: text2text-generation
tags:
- Neuroscience
- chemistry
- code
---
# Morningstar-Omega Model README
## Project: Morningstar-Omega
Welcome to Morningstar-Omega, a text generation model designed to provide state-of-the-art performance in neuroscience and chemistry text generation tasks. This repository contains the model, its documentation, usage guidelines, and licensing information.
Repository: Lucius-Morningstar/Morningstar-Omega
Model Type: Text2Text Generation
Related Fields: Neuroscience, Chemistry
Model ID DOI: doi:10.57967/hf/3369
arXiv Paper: 1910.09700
License: MIT License
## Model Overview
The Morningstar-Omega model leverages advancements in neural networks to generate high-quality, contextually accurate text in response to a given input, focusing particularly on applications in neuroscience and chemistry.
### Model Details
• Developed by: [Lucius-Morningstar]
• Funded by: [optional: Specify Funding Agency]
• Model Type: Text2Text Generation
• Languages: English (NLP), with potential for multilingual support
• License: MIT License
• Finetuned from: [Original Base Model, if applicable]
### Model Sources
• Repository: Lucius-Morningstar/Morningstar-Omega
• Paper: arXiv:1910.09700
• Demo: [Add Link to Demo, if available]
#### Usage
### Direct Use
This model can be used for generating scientific text in neuroscience and chemistry, specifically aimed at applications requiring complex, contextually aware language generation. Ideal for academic, research, and professional environments needing coherent, topic-specific text output.
### Downstream Use
Potential downstream applications include:
• Automated scientific paper generation
• Text generation for hypothesis testing in neuroscience and chemistry
• Educational tools and scientific summarization tasks
## Out-of-Scope Use
The model is not recommended for:
• Tasks outside scientific and technical domains, as it may lack contextual accuracy in broader fields.
• Generating personal or sensitive information where text accuracy and ethical considerations are paramount.
### Model Bias, Risks, and Limitations
The Morningstar-Omega model, like many large language models, is subject to biases present in its training data. Users should be aware of potential limitations, including:
• Bias in Scientific Domains: Training data may reflect predominant theories, leading to a reinforcement of certain scientific biases.
• Data Gaps: Specific areas in neuroscience or chemistry may be underrepresented.
• Ethical Considerations: Content generation should comply with ethical standards, especially in academic and professional contexts.
## Recommendations
Users should validate the model’s output in scientific contexts and critically assess any generated content for accuracy, especially for high-stakes applications.
Getting Started
To begin using the model, you can follow these steps:
Installation
# Clone the repository
git clone <https://github.com/Lucius-Morningstar/Morningstar-Omega.git>
cd Morningstar-Omega
# Install dependencies
pip install -r requirements.txt
Usage Example
from morningstar_omega import Model
# Initialize model
model = Model.load('path/to/pretrained_model')
## Text Generation
output = model.generate("Describe the process of synaptic transmission in the brain.")
print(output)
Training Details
Training Data
The model was trained on a curated dataset combining publicly available resources in neuroscience and chemistry research articles, augmented with domain-specific text to enhance language capabilities.
Training Procedure
Preprocessing
Data was tokenized and cleaned to ensure scientific accuracy and context. Irrelevant or low-quality samples were removed.
Training Hyperparameters
• Training Regime: Fine-tuning based on neural network hyperparameter optimization.
• Epochs: [Specify]
• Batch Size: [Specify]
• Learning Rate: [Specify]
Speeds, Sizes, Times
• Model Size: [Model size, e.g., 1.2B parameters]
• Training Time: [Specify]
Evaluation
Testing Data, Factors & Metrics
Testing Data
The model was evaluated using a set of scientific articles and technical documents in neuroscience and chemistry.
Factors
Evaluation focused on metrics like coherence, relevance to input prompts, factual accuracy, and linguistic diversity.
Metrics
• Perplexity: [Specify]
• BLEU Score: [Specify]
• Accuracy in Factual Generation: [Specify]
Results
The model achieved [Specify Results] on standard evaluation benchmarks, indicating high performance in scientific text generation.
Summary
The Morningstar-Omega model is a specialized text generation tool for neuroscience and chemistry applications, delivering precise and relevant language generation capabilities for academic and research use. Its design allows for detailed exploration of scientific topics, enhancing productivity in technical fields.
Environmental Impact
To assess the environmental footprint of training this model, use the Machine Learning Impact calculator as suggested by Lacoste et al. (2019).
• Hardware Type: [e.g., GPU, TPU]
• Hours Used: [Specify]
• Cloud Provider: [Specify, if applicable]
• Compute Region: [Specify, if applicable]
• Carbon Emitted: [Estimate, if available]
Technical Specifications
Model Architecture and Objective
The model architecture is based on [Specify neural network architecture, e.g., Transformer-based architecture optimized for text-to-text generation].
Compute Infrastructure
• Hardware: [Specify hardware used during training, e.g., NVIDIA Tesla GPUs]
• Software Dependencies: Listed in requirements.txt
Citation
If you use this model in your work, please cite it as follows:
BibTeX:
@article{lucius2024morningstar,
title={Morningstar-Omega: Advanced Text Generation for Neuroscience and Chemistry},
author={Lucius-Morningstar},
journal={Neuralwork/arxiver},
doi={10.57967/hf/3369},
year={2024}
}
APA:
Lucius-Morningstar. (2024). Morningstar-Omega: Advanced Text Generation for Neuroscience and Chemistry. Neuralwork/arxiver. doi:10.57967/hf/3369.
Glossary
• Synaptic Transmission: [Define term]
• Neuroplasticity: [Define term]
• Molecular Modeling: [Define term]
Contact
For any questions or issues, please reach out to [Contact Information].
| null |
Non_BioNLP
|
# Morningstar-Omega Model README
## Project: Morningstar-Omega
Welcome to Morningstar-Omega, a text generation model designed to provide state-of-the-art performance in neuroscience and chemistry text generation tasks. This repository contains the model, its documentation, usage guidelines, and licensing information.
Repository: Lucius-Morningstar/Morningstar-Omega
Model Type: Text2Text Generation
Related Fields: Neuroscience, Chemistry
Model ID DOI: doi:10.57967/hf/3369
arXiv Paper: 1910.09700
License: MIT License
## Model Overview
The Morningstar-Omega model leverages advancements in neural networks to generate high-quality, contextually accurate text in response to a given input, focusing particularly on applications in neuroscience and chemistry.
### Model Details
• Developed by: [Lucius-Morningstar]
• Funded by: [optional: Specify Funding Agency]
• Model Type: Text2Text Generation
• Languages: English (NLP), with potential for multilingual support
• License: MIT License
• Finetuned from: [Original Base Model, if applicable]
### Model Sources
• Repository: Lucius-Morningstar/Morningstar-Omega
• Paper: arXiv:1910.09700
• Demo: [Add Link to Demo, if available]
#### Usage
### Direct Use
This model can be used for generating scientific text in neuroscience and chemistry, specifically aimed at applications requiring complex, contextually aware language generation. Ideal for academic, research, and professional environments needing coherent, topic-specific text output.
### Downstream Use
Potential downstream applications include:
• Automated scientific paper generation
• Text generation for hypothesis testing in neuroscience and chemistry
• Educational tools and scientific summarization tasks
## Out-of-Scope Use
The model is not recommended for:
• Tasks outside scientific and technical domains, as it may lack contextual accuracy in broader fields.
• Generating personal or sensitive information where text accuracy and ethical considerations are paramount.
### Model Bias, Risks, and Limitations
The Morningstar-Omega model, like many large language models, is subject to biases present in its training data. Users should be aware of potential limitations, including:
• Bias in Scientific Domains: Training data may reflect predominant theories, leading to a reinforcement of certain scientific biases.
• Data Gaps: Specific areas in neuroscience or chemistry may be underrepresented.
• Ethical Considerations: Content generation should comply with ethical standards, especially in academic and professional contexts.
## Recommendations
Users should validate the model’s output in scientific contexts and critically assess any generated content for accuracy, especially for high-stakes applications.
Getting Started
To begin using the model, you can follow these steps:
Installation
# Clone the repository
git clone <https://github.com/Lucius-Morningstar/Morningstar-Omega.git>
cd Morningstar-Omega
# Install dependencies
pip install -r requirements.txt
Usage Example
from morningstar_omega import Model
# Initialize model
model = Model.load('path/to/pretrained_model')
## Text Generation
output = model.generate("Describe the process of synaptic transmission in the brain.")
print(output)
Training Details
Training Data
The model was trained on a curated dataset combining publicly available resources in neuroscience and chemistry research articles, augmented with domain-specific text to enhance language capabilities.
Training Procedure
Preprocessing
Data was tokenized and cleaned to ensure scientific accuracy and context. Irrelevant or low-quality samples were removed.
Training Hyperparameters
• Training Regime: Fine-tuning based on neural network hyperparameter optimization.
• Epochs: [Specify]
• Batch Size: [Specify]
• Learning Rate: [Specify]
Speeds, Sizes, Times
• Model Size: [Model size, e.g., 1.2B parameters]
• Training Time: [Specify]
Evaluation
Testing Data, Factors & Metrics
Testing Data
The model was evaluated using a set of scientific articles and technical documents in neuroscience and chemistry.
Factors
Evaluation focused on metrics like coherence, relevance to input prompts, factual accuracy, and linguistic diversity.
Metrics
• Perplexity: [Specify]
• BLEU Score: [Specify]
• Accuracy in Factual Generation: [Specify]
Results
The model achieved [Specify Results] on standard evaluation benchmarks, indicating high performance in scientific text generation.
Summary
The Morningstar-Omega model is a specialized text generation tool for neuroscience and chemistry applications, delivering precise and relevant language generation capabilities for academic and research use. Its design allows for detailed exploration of scientific topics, enhancing productivity in technical fields.
Environmental Impact
To assess the environmental footprint of training this model, use the Machine Learning Impact calculator as suggested by Lacoste et al. (2019).
• Hardware Type: [e.g., GPU, TPU]
• Hours Used: [Specify]
• Cloud Provider: [Specify, if applicable]
• Compute Region: [Specify, if applicable]
• Carbon Emitted: [Estimate, if available]
Technical Specifications
Model Architecture and Objective
The model architecture is based on [Specify neural network architecture, e.g., Transformer-based architecture optimized for text-to-text generation].
Compute Infrastructure
• Hardware: [Specify hardware used during training, e.g., NVIDIA Tesla GPUs]
• Software Dependencies: Listed in requirements.txt
Citation
If you use this model in your work, please cite it as follows:
BibTeX:
@article{lucius2024morningstar,
title={Morningstar-Omega: Advanced Text Generation for Neuroscience and Chemistry},
author={Lucius-Morningstar},
journal={Neuralwork/arxiver},
doi={10.57967/hf/3369},
year={2024}
}
APA:
Lucius-Morningstar. (2024). Morningstar-Omega: Advanced Text Generation for Neuroscience and Chemistry. Neuralwork/arxiver. doi:10.57967/hf/3369.
Glossary
• Synaptic Transmission: [Define term]
• Neuroplasticity: [Define term]
• Molecular Modeling: [Define term]
Contact
For any questions or issues, please reach out to [Contact Information].
|
{"base_model": ["nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", "nvidia/Llama-3.1-Nemotron-70B-Instruct"], "datasets": ["neuralwork/arxiver"], "license": "mit", "pipeline_tag": "text2text-generation", "tags": ["Neuroscience", "chemistry", "code"]}
|
task
|
[
"SUMMARIZATION"
] | 45,163 |
raushan-in/flan-t5-base-fine-tuned
|
raushan-in
| null |
[
"safetensors",
"t5",
"region:us"
] | 2024-12-24T21:45:19Z |
2024-12-24T23:30:35+00:00
| 44 | 0 |
---
{}
---
# flan-t5-base-fine-tuned
This repository contains the fine-tuned version of the `flan-t5-base` model for [specific task, e.g., text summarization, question answering, etc.].
## Model Details
- **Base Model**: [Flan-T5 Base](https://huggingface.co/google/flan-t5-base)
- **Fine-Tuned On**: [Dataset name or custom dataset, e.g., CNN/DailyMail, SQuAD, etc.]
- **Task**: [Task name, e.g., text generation, summarization, classification, etc.]
- **Framework**: [Transformers](https://github.com/huggingface/transformers)
## Usage
You can use this model with the Hugging Face `transformers` library.
## Dataset
The model was fine-tuned on the [dataset name] dataset. Below is an example of the data:
- Input: what is golang?
- Output: A statically typed, compiled high-level general purpose programming language.
## Limitations
- The model may struggle with [specific limitation, e.g., long inputs, out-of-domain data, etc.].
- Outputs may occasionally contain biases present in the training data.
| null |
Non_BioNLP
|
# flan-t5-base-fine-tuned
This repository contains the fine-tuned version of the `flan-t5-base` model for [specific task, e.g., text summarization, question answering, etc.].
## Model Details
- **Base Model**: [Flan-T5 Base](https://huggingface.co/google/flan-t5-base)
- **Fine-Tuned On**: [Dataset name or custom dataset, e.g., CNN/DailyMail, SQuAD, etc.]
- **Task**: [Task name, e.g., text generation, summarization, classification, etc.]
- **Framework**: [Transformers](https://github.com/huggingface/transformers)
## Usage
You can use this model with the Hugging Face `transformers` library.
## Dataset
The model was fine-tuned on the [dataset name] dataset. Below is an example of the data:
- Input: what is golang?
- Output: A statically typed, compiled high-level general purpose programming language.
## Limitations
- The model may struggle with [specific limitation, e.g., long inputs, out-of-domain data, etc.].
- Outputs may occasionally contain biases present in the training data.
|
{}
|
task
|
[
"QUESTION_ANSWERING",
"SUMMARIZATION"
] | 45,164 |
hfl/vle-base
|
hfl
| null |
[
"transformers",
"pytorch",
"vle",
"en",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | 2023-03-03T02:01:59Z |
2023-03-09T00:43:56+00:00
| 10 | 4 |
---
language:
- en
license: apache-2.0
---
**VLE** (**V**isual-**L**anguage **E**ncoder) is an image-text multimodal understanding model built on the pre-trained text and image encoders.
It can be used for multimodal discriminative tasks such as visual question answering and image-text retrieval.
Especially on the visual commonsense reasoning (VCR) task, which requires high-level language understanding and reasoning skills, VLE achieves significant improvements.
For more details see [https://github.com/iflytek/VLE](https://github.com/iflytek/VLE).
Online VLE demo on Visual Question Answering: [https://huggingface.co/spaces/hfl/VQA_VLE_LLM](https://huggingface.co/spaces/hfl/VQA_VLE_LLM)
| null |
Non_BioNLP
|
**VLE** (**V**isual-**L**anguage **E**ncoder) is an image-text multimodal understanding model built on the pre-trained text and image encoders.
It can be used for multimodal discriminative tasks such as visual question answering and image-text retrieval.
Especially on the visual commonsense reasoning (VCR) task, which requires high-level language understanding and reasoning skills, VLE achieves significant improvements.
For more details see [https://github.com/iflytek/VLE](https://github.com/iflytek/VLE).
Online VLE demo on Visual Question Answering: [https://huggingface.co/spaces/hfl/VQA_VLE_LLM](https://huggingface.co/spaces/hfl/VQA_VLE_LLM)
|
{"language": ["en"], "license": "apache-2.0"}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,165 |
code-is-wonderful/pros_cons_pegasus_sum
|
code-is-wonderful
|
summarization
|
[
"transformers",
"pytorch",
"pegasus",
"text2text-generation",
"summarization",
"en",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-04-19T03:15:05Z |
2023-04-21T07:02:18+00:00
| 24 | 0 |
---
language:
- en
library_name: transformers
license: apache-2.0
metrics:
- rouge
pipeline_tag: summarization
---
Summarize similar sentences for Amazon reviews
| null |
Non_BioNLP
|
Summarize similar sentences for Amazon reviews
|
{"language": ["en"], "library_name": "transformers", "license": "apache-2.0", "metrics": ["rouge"], "pipeline_tag": "summarization"}
|
task
|
[
"SUMMARIZATION"
] | 45,166 |
vaishnavkoka/fine_tune_llama_squad2
|
vaishnavkoka
|
question-answering
|
[
"transformers",
"llama",
"question-answering",
"sqaud",
"fine",
"tuned",
"dataset:rajpurkar/squad_v2",
"base_model:google/gemma-2-2b-it",
"base_model:finetune:google/gemma-2-2b-it",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | 2024-11-20T19:13:50Z |
2024-11-23T05:27:21+00:00
| 12 | 0 |
---
base_model:
- meta-llama/Llama-3.2-1B
- google/gemma-2-2b-it
datasets:
- rajpurkar/squad_v2
library_name: transformers
license: apache-2.0
metrics:
- precision
- f1
- recall
- squad_v2
- meteor
- bleu
- rouge
- exact_match
tags:
- llama
- sqaud
- fine
- tuned
---
1. Overview
This repository highlights the fine-tuning of the Llama-3.2-1B model on the SQuAD (Stanford Question Answering Dataset) dataset. The task involves training the model to accurately answer questions based on a given context passage. Fine-tuning the pre-trained Llama model aligns it with the objectives of extractive question-answering.
2. Model Information
Model Used: meta-llama/Llama-3.2-1B
Pre-trained Parameters: The model contains approximately 1.03 billion parameters, verified during setup and matching official documentation.
Fine-tuned Parameters: The parameter count remains consistent with the pre-trained model, as fine-tuning only updates task-specific weights.
3. Dataset and Task Details
Dataset: SQuAD
The Stanford Question Answering Dataset (SQuAD) is a benchmark dataset designed for extractive question-answering tasks. It contains passages with corresponding questions and answer spans extracted directly from the text.
Task Objective
Given a passage and a question, the model is trained to identify the correct span of text in the passage that answers the question.
4. Fine-Tuning Approach
Train-Test Split: An 80:20 split was applied to the dataset, ensuring a balanced distribution of passages and questions in the train and test subsets. Stratified sampling was used, with a seed value of 1 for reproducibility.
Tokenization: Context and question pairs were tokenized with padding and truncation to ensure uniform input lengths (maximum 512 tokens).
Model Training: Fine-tuning was conducted over three epochs with a learning rate of 3e-5. Gradient accumulation and early stopping were used to enhance training efficiency and prevent overfitting.
Hardware: Training utilized GPU acceleration to handle the large model size and complex token sequences efficiently.
5. Results and Observations
Zero-shot vs. Fine-tuned Performance: Without fine-tuning, the pre-trained Llama model demonstrated limited ability to answer questions accurately. Fine-tuning significantly improved the model’s performance on metrics such as F1 score, exact match, and ROUGE.
Fine-tuning Benefits: Training on the SQuAD dataset equipped the model with a deeper understanding of context and its relationship to specific queries, enhancing its ability to extract precise answer spans.
Model Parameters: The parameter count remained unchanged during fine-tuning, underscoring that performance improvements stemmed from the optimization of existing weights rather than structural changes.
6. How to Use the Fine-Tuned Model
Install Necessary Libraries:
pip install transformers datasets
Load the Fine-Tuned Model:
from transformers import AutoTokenizer, AutoModelForQuestionAnswering
model_name = "<your-huggingface-repo>/squad-llama-finetuned"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForQuestionAnswering.from_pretrained(model_name)
Make Predictions:
context = "Llama is a model developed by Meta AI designed for natural language understanding tasks."
question = "Who developed Llama?"
inputs = tokenizer(question, context, return_tensors="pt", truncation=True, padding=True)
outputs = model(**inputs)
start_idx = outputs.start_logits.argmax()
end_idx = outputs.end_logits.argmax()
answer = tokenizer.decode(inputs["input_ids"][0][start_idx:end_idx + 1])
print(f"Predicted Answer: {answer}")
7. Key Takeaways
Fine-tuning Llama on SQuAD equips it with the ability to handle extractive question-answering tasks with high accuracy and precision.
The parameter count of the model does not change during fine-tuning, highlighting that performance enhancements are derived from weight updates rather than architectural modifications.
The comparison between zero-shot and fine-tuned performance demonstrates the necessity of task-specific training to achieve state-of-the-art results.
8. Acknowledgments
Hugging Face for providing seamless tools for model fine-tuning and evaluation.
Stanford Question Answering Dataset for serving as a robust benchmark for extractive QA tasks.
| null |
Non_BioNLP
|
1. Overview
This repository highlights the fine-tuning of the Llama-3.2-1B model on the SQuAD (Stanford Question Answering Dataset) dataset. The task involves training the model to accurately answer questions based on a given context passage. Fine-tuning the pre-trained Llama model aligns it with the objectives of extractive question-answering.
2. Model Information
Model Used: meta-llama/Llama-3.2-1B
Pre-trained Parameters: The model contains approximately 1.03 billion parameters, verified during setup and matching official documentation.
Fine-tuned Parameters: The parameter count remains consistent with the pre-trained model, as fine-tuning only updates task-specific weights.
3. Dataset and Task Details
Dataset: SQuAD
The Stanford Question Answering Dataset (SQuAD) is a benchmark dataset designed for extractive question-answering tasks. It contains passages with corresponding questions and answer spans extracted directly from the text.
Task Objective
Given a passage and a question, the model is trained to identify the correct span of text in the passage that answers the question.
4. Fine-Tuning Approach
Train-Test Split: An 80:20 split was applied to the dataset, ensuring a balanced distribution of passages and questions in the train and test subsets. Stratified sampling was used, with a seed value of 1 for reproducibility.
Tokenization: Context and question pairs were tokenized with padding and truncation to ensure uniform input lengths (maximum 512 tokens).
Model Training: Fine-tuning was conducted over three epochs with a learning rate of 3e-5. Gradient accumulation and early stopping were used to enhance training efficiency and prevent overfitting.
Hardware: Training utilized GPU acceleration to handle the large model size and complex token sequences efficiently.
5. Results and Observations
Zero-shot vs. Fine-tuned Performance: Without fine-tuning, the pre-trained Llama model demonstrated limited ability to answer questions accurately. Fine-tuning significantly improved the model’s performance on metrics such as F1 score, exact match, and ROUGE.
Fine-tuning Benefits: Training on the SQuAD dataset equipped the model with a deeper understanding of context and its relationship to specific queries, enhancing its ability to extract precise answer spans.
Model Parameters: The parameter count remained unchanged during fine-tuning, underscoring that performance improvements stemmed from the optimization of existing weights rather than structural changes.
6. How to Use the Fine-Tuned Model
Install Necessary Libraries:
pip install transformers datasets
Load the Fine-Tuned Model:
from transformers import AutoTokenizer, AutoModelForQuestionAnswering
model_name = "<your-huggingface-repo>/squad-llama-finetuned"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForQuestionAnswering.from_pretrained(model_name)
Make Predictions:
context = "Llama is a model developed by Meta AI designed for natural language understanding tasks."
question = "Who developed Llama?"
inputs = tokenizer(question, context, return_tensors="pt", truncation=True, padding=True)
outputs = model(**inputs)
start_idx = outputs.start_logits.argmax()
end_idx = outputs.end_logits.argmax()
answer = tokenizer.decode(inputs["input_ids"][0][start_idx:end_idx + 1])
print(f"Predicted Answer: {answer}")
7. Key Takeaways
Fine-tuning Llama on SQuAD equips it with the ability to handle extractive question-answering tasks with high accuracy and precision.
The parameter count of the model does not change during fine-tuning, highlighting that performance enhancements are derived from weight updates rather than architectural modifications.
The comparison between zero-shot and fine-tuned performance demonstrates the necessity of task-specific training to achieve state-of-the-art results.
8. Acknowledgments
Hugging Face for providing seamless tools for model fine-tuning and evaluation.
Stanford Question Answering Dataset for serving as a robust benchmark for extractive QA tasks.
|
{"base_model": ["meta-llama/Llama-3.2-1B", "google/gemma-2-2b-it"], "datasets": ["rajpurkar/squad_v2"], "library_name": "transformers", "license": "apache-2.0", "metrics": ["precision", "f1", "recall", "squad_v2", "meteor", "bleu", "rouge", "exact_match"], "tags": ["llama", "sqaud", "fine", "tuned"]}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,167 |
0x9/netuid1-classification
|
0x9
|
text-classification
|
[
"transformers",
"safetensors",
"distilbert",
"text-classification",
"autotrain",
"dataset:autotrain-pmf0g-rj8fa/autotrain-data",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-03-13T01:27:15Z |
2024-03-13T01:27:23+00:00
| 14 | 1 |
---
datasets:
- autotrain-pmf0g-rj8fa/autotrain-data
tags:
- autotrain
- text-classification
widget:
- text: I love AutoTrain
---
# Model Trained Using AutoTrain
- Problem type: Text Classification
## Validation Metrics
loss: 0.03381425514817238
f1_macro: 0.9910410929202866
f1_micro: 0.9908675799086758
f1_weighted: 0.9908473335613555
precision_macro: 0.9909727371947719
precision_micro: 0.9908675799086758
precision_weighted: 0.9908883151237302
recall_macro: 0.9911698494022667
recall_micro: 0.9908675799086758
recall_weighted: 0.9908675799086758
accuracy: 0.9908675799086758
| null |
Non_BioNLP
|
# Model Trained Using AutoTrain
- Problem type: Text Classification
## Validation Metrics
loss: 0.03381425514817238
f1_macro: 0.9910410929202866
f1_micro: 0.9908675799086758
f1_weighted: 0.9908473335613555
precision_macro: 0.9909727371947719
precision_micro: 0.9908675799086758
precision_weighted: 0.9908883151237302
recall_macro: 0.9911698494022667
recall_micro: 0.9908675799086758
recall_weighted: 0.9908675799086758
accuracy: 0.9908675799086758
|
{"datasets": ["autotrain-pmf0g-rj8fa/autotrain-data"], "tags": ["autotrain", "text-classification"], "widget": [{"text": "I love AutoTrain"}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,168 |
NorGLM/NorLlama-3B-summarization-peft
|
NorGLM
|
summarization
|
[
"summarization",
"no",
"dataset:NorGLM/NO-CNN-DailyMail",
"license:cc-by-nc-sa-4.0",
"region:us"
] | 2024-03-11T09:10:01Z |
2024-03-11T09:24:01+00:00
| 0 | 0 |
---
datasets:
- NorGLM/NO-CNN-DailyMail
language:
- 'no'
license: cc-by-nc-sa-4.0
pipeline_tag: summarization
---
# Model Card
NorLlama-3B-summarization-peft is trained on top of [NorLlama-3B](https://huggingface.co/NorGLM/NorLlama-3B) model on [NO-CNN-DailyMail](https://huggingface.co/datasets/NorGLM/NO-CNN-DailyMail) dataset.
Prompt format:
```
Summarise the article:\\n{article} |||\\n{positive_sample}
```
Inference prompt:
```
Summarise the article:\\n{article} |||\\n
```
## Run the Model
```python
from peft import PeftModel, PeftConfig
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
source_model_id = "NorGLM/NorLlama-3B"
peft_model_id = "NorGLM/NorLlama-3B-summarization-peft"
config = PeftConfig.from_pretrained(peft_model_id)
model = AutoModelForCausalLM.from_pretrained(source_model_id, device_map='balanced')
tokenizer_max_len = 2048
tokenizer_config = {'pretrained_model_name_or_path': source_model_id,
'max_len': tokenizer_max_len}
tokenizer = tokenizer = AutoTokenizer.from_pretrained(**tokenizer_config)
tokenizer.pad_token = tokenizer.eos_token
model = PeftModel.from_pretrained(model, peft_model_id)
```
## Inference on test set
Load the model to evaluate on the test set of NO-CNN-DailyMail dataset:
```python
def generate_texts(model, tokenizer, prompts, max_seq_length=200, do_sample=True, top_p=0.95, top_k=10):
# prompts are a list of news articles
results = []
cnt = 0
for prompt in prompts:
cnt += 1
pro_len = len(prompt.split())
if pro_len>1024:
results.append('')
continue
prompt = 'Summarise the article:\\n' + prompt + ' |||\\n'
model_inputs = tokenizer(prompt, return_tensors='pt').to(torch_device)
output = model.generate(**model_inputs, do_sample=False, max_new_tokens=max_seq_length)
result = tokenizer.decode(output[0], skip_special_tokens=True)
result = result.split("|||\\n")[-1]
results.append(result)
return results
print("--LOADING EVAL DATAS---")
eval_data = load_dataset("NorGLM/NO-CNN-DailyMail", data_files="test.csv")
prompts = eval_data['train']['article']
positive_samples = eval_data['train']['positive_sample']
print("--MAKING PREDICTIONS---")
model.eval()
output_file = <output file name>
with torch.no_grad():
results = generate_texts(model, tokenizer, prompts)
df = pd.DataFrame({'article':prompts, 'generated_text':results, 'positive_sample':positive_samples})
print("Save results to csv file...")
df.to_csv(output_file)
```
## Note
More training details will be released soon!
| null |
Non_BioNLP
|
# Model Card
NorLlama-3B-summarization-peft is trained on top of [NorLlama-3B](https://huggingface.co/NorGLM/NorLlama-3B) model on [NO-CNN-DailyMail](https://huggingface.co/datasets/NorGLM/NO-CNN-DailyMail) dataset.
Prompt format:
```
Summarise the article:\\n{article} |||\\n{positive_sample}
```
Inference prompt:
```
Summarise the article:\\n{article} |||\\n
```
## Run the Model
```python
from peft import PeftModel, PeftConfig
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
source_model_id = "NorGLM/NorLlama-3B"
peft_model_id = "NorGLM/NorLlama-3B-summarization-peft"
config = PeftConfig.from_pretrained(peft_model_id)
model = AutoModelForCausalLM.from_pretrained(source_model_id, device_map='balanced')
tokenizer_max_len = 2048
tokenizer_config = {'pretrained_model_name_or_path': source_model_id,
'max_len': tokenizer_max_len}
tokenizer = tokenizer = AutoTokenizer.from_pretrained(**tokenizer_config)
tokenizer.pad_token = tokenizer.eos_token
model = PeftModel.from_pretrained(model, peft_model_id)
```
## Inference on test set
Load the model to evaluate on the test set of NO-CNN-DailyMail dataset:
```python
def generate_texts(model, tokenizer, prompts, max_seq_length=200, do_sample=True, top_p=0.95, top_k=10):
# prompts are a list of news articles
results = []
cnt = 0
for prompt in prompts:
cnt += 1
pro_len = len(prompt.split())
if pro_len>1024:
results.append('')
continue
prompt = 'Summarise the article:\\n' + prompt + ' |||\\n'
model_inputs = tokenizer(prompt, return_tensors='pt').to(torch_device)
output = model.generate(**model_inputs, do_sample=False, max_new_tokens=max_seq_length)
result = tokenizer.decode(output[0], skip_special_tokens=True)
result = result.split("|||\\n")[-1]
results.append(result)
return results
print("--LOADING EVAL DATAS---")
eval_data = load_dataset("NorGLM/NO-CNN-DailyMail", data_files="test.csv")
prompts = eval_data['train']['article']
positive_samples = eval_data['train']['positive_sample']
print("--MAKING PREDICTIONS---")
model.eval()
output_file = <output file name>
with torch.no_grad():
results = generate_texts(model, tokenizer, prompts)
df = pd.DataFrame({'article':prompts, 'generated_text':results, 'positive_sample':positive_samples})
print("Save results to csv file...")
df.to_csv(output_file)
```
## Note
More training details will be released soon!
|
{"datasets": ["NorGLM/NO-CNN-DailyMail"], "language": ["no"], "license": "cc-by-nc-sa-4.0", "pipeline_tag": "summarization"}
|
task
|
[
"SUMMARIZATION"
] | 45,170 |
fine-tuned/NFCorpus-32000-384-gpt-4o-2024-05-13-70846146
|
fine-tuned
|
feature-extraction
|
[
"sentence-transformers",
"safetensors",
"bert",
"feature-extraction",
"sentence-similarity",
"mteb",
"en",
"dataset:fine-tuned/NFCorpus-32000-384-gpt-4o-2024-05-13-70846146",
"dataset:allenai/c4",
"license:apache-2.0",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2024-06-06T16:08:33Z |
2024-06-06T16:08:37+00:00
| 7 | 0 |
---
datasets:
- fine-tuned/NFCorpus-32000-384-gpt-4o-2024-05-13-70846146
- allenai/c4
language:
- en
- en
license: apache-2.0
pipeline_tag: feature-extraction
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- mteb
---
This model is a fine-tuned version of [**BAAI/bge-small-en-v1.5**](https://huggingface.co/BAAI/bge-small-en-v1.5) designed for the following use case:
None
## How to Use
This model can be easily integrated into your NLP pipeline for tasks such as text classification, sentiment analysis, entity recognition, and more. Here's a simple example to get you started:
```python
from sentence_transformers import SentenceTransformer
from sentence_transformers.util import cos_sim
model = SentenceTransformer(
'fine-tuned/NFCorpus-32000-384-gpt-4o-2024-05-13-70846146',
trust_remote_code=True
)
embeddings = model.encode([
'first text to embed',
'second text to embed'
])
print(cos_sim(embeddings[0], embeddings[1]))
```
| null |
Non_BioNLP
|
This model is a fine-tuned version of [**BAAI/bge-small-en-v1.5**](https://huggingface.co/BAAI/bge-small-en-v1.5) designed for the following use case:
None
## How to Use
This model can be easily integrated into your NLP pipeline for tasks such as text classification, sentiment analysis, entity recognition, and more. Here's a simple example to get you started:
```python
from sentence_transformers import SentenceTransformer
from sentence_transformers.util import cos_sim
model = SentenceTransformer(
'fine-tuned/NFCorpus-32000-384-gpt-4o-2024-05-13-70846146',
trust_remote_code=True
)
embeddings = model.encode([
'first text to embed',
'second text to embed'
])
print(cos_sim(embeddings[0], embeddings[1]))
```
|
{"datasets": ["fine-tuned/NFCorpus-32000-384-gpt-4o-2024-05-13-70846146", "allenai/c4"], "language": ["en", "en"], "license": "apache-2.0", "pipeline_tag": "feature-extraction", "tags": ["sentence-transformers", "feature-extraction", "sentence-similarity", "mteb"]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,171 |
Xyzhayhhanghsha/Mita2
|
Xyzhayhhanghsha
|
summarization
|
[
"adapter-transformers",
"chemistry",
"summarization",
"aa",
"dataset:HuggingFaceTB/cosmopedia",
"license:mit",
"region:us"
] | 2024-03-17T14:09:46Z |
2024-03-17T14:10:19+00:00
| 0 | 0 |
---
datasets:
- HuggingFaceTB/cosmopedia
language:
- aa
library_name: adapter-transformers
license: mit
metrics:
- accuracy
pipeline_tag: summarization
tags:
- chemistry
---
| null |
Non_BioNLP
|
{"datasets": ["HuggingFaceTB/cosmopedia"], "language": ["aa"], "library_name": "adapter-transformers", "license": "mit", "metrics": ["accuracy"], "pipeline_tag": "summarization", "tags": ["chemistry"]}
|
task
|
[
"SUMMARIZATION"
] | 45,172 |
|
chitra/finetune-paraphrase-model
|
chitra
|
text-classification
|
[
"transformers",
"pytorch",
"tensorboard",
"roberta",
"text-classification",
"generated_from_trainer",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2022-03-02T23:29:05Z |
2022-01-19T04:40:57+00:00
| 117 | 0 |
---
tags:
- generated_from_trainer
model-index:
- name: finetune-paraphrase-model
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# finetune-paraphrase-model
This model is a fine-tuned version of [coderpotter/adversarial-paraphrasing-detector](https://huggingface.co/coderpotter/adversarial-paraphrasing-detector) on an unknown dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 0.1
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| No log | 0.1 | 200 | 3.0116 |
### Framework versions
- Transformers 4.15.0
- Pytorch 1.10.0+cu111
- Datasets 1.17.0
- Tokenizers 0.10.3
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# finetune-paraphrase-model
This model is a fine-tuned version of [coderpotter/adversarial-paraphrasing-detector](https://huggingface.co/coderpotter/adversarial-paraphrasing-detector) on an unknown dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 0.1
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| No log | 0.1 | 200 | 3.0116 |
### Framework versions
- Transformers 4.15.0
- Pytorch 1.10.0+cu111
- Datasets 1.17.0
- Tokenizers 0.10.3
|
{"tags": ["generated_from_trainer"], "model-index": [{"name": "finetune-paraphrase-model", "results": []}]}
|
task
|
[
"PARAPHRASING"
] | 45,173 |
alpindale/Llama-3.2-90B-Vision
|
alpindale
|
image-text-to-text
|
[
"transformers",
"safetensors",
"mllama",
"image-text-to-text",
"facebook",
"meta",
"pytorch",
"llama",
"llama-3",
"en",
"de",
"fr",
"it",
"pt",
"hi",
"es",
"th",
"arxiv:2204.05149",
"license:llama3.2",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2024-09-25T19:17:48Z |
2024-09-25T19:24:16+00:00
| 13 | 1 |
---
language:
- en
- de
- fr
- it
- pt
- hi
- es
- th
library_name: transformers
license: llama3.2
pipeline_tag: image-text-to-text
tags:
- facebook
- meta
- pytorch
- llama
- llama-3
extra_gated_prompt: "### LLAMA 3.2 COMMUNITY LICENSE AGREEMENT\n\nLlama 3.2 Version\
\ Release Date: September 25, 2024\n\n“Agreement” means the terms and conditions\
\ for use, reproduction, distribution and modification of the Llama Materials set\
\ forth herein.\n\n“Documentation” means the specifications, manuals and documentation\
\ accompanying Llama 3.2 distributed by Meta at https://llama.meta.com/doc/overview.\n\
\n“Licensee” or “you” means you, or your employer or any other person or entity\
\ (if you are entering into this Agreement on such person or entity’s behalf),\
\ of the age required under applicable laws, rules or regulations to provide legal\
\ consent and that has legal authority to bind your employer or such other person\
\ or entity if you are entering in this Agreement on their behalf.\n\n“Llama 3.2”\
\ means the foundational large language models and software and algorithms, including\
\ machine-learning model code, trained model weights, inference-enabling code, training-enabling\
\ code, fine-tuning enabling code and other elements of the foregoing distributed\
\ by Meta at https://www.llama.com/llama-downloads.\n\n“Llama Materials” means,\
\ collectively, Meta’s proprietary Llama 3.2 and Documentation (and any portion\
\ thereof) made available under this Agreement.\n\n“Meta” or “we” means Meta Platforms\
\ Ireland Limited (if you are located in or, if you are an entity, your principal\
\ place of business is in the EEA or Switzerland) and Meta Platforms, Inc. (if\
\ you are located outside of the EEA or Switzerland). \n\nBy clicking “I Accept”\
\ below or by using or distributing any portion or element of the Llama Materials,\
\ you agree to be bound by this Agreement.\n\n1. License Rights and Redistribution.\n\
a. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable\
\ and royalty-free limited license under Meta’s intellectual property or other rights\
\ owned by Meta embodied in the Llama Materials to use, reproduce, distribute,\
\ copy, create derivative works of, and make modifications to the Llama Materials.\
\ \nb. Redistribution and Use. \ni. If you distribute or make available the Llama\
\ Materials (or any derivative works thereof), or a product or service (including\
\ another AI model) that contains any of them, you shall (A) provide a copy of this\
\ Agreement with any such Llama Materials; and (B) prominently display “Built with\
\ Llama” on a related website, user interface, blogpost, about page, or product\
\ documentation. If you use the Llama Materials or any outputs or results of the\
\ Llama Materials to create, train, fine tune, or otherwise improve an AI model,\
\ which is distributed or made available, you shall also include “Llama” at the\
\ beginning of any such AI model name.\nii. If you receive Llama Materials, or any\
\ derivative works thereof, from a Licensee as part of an integrated end user product,\
\ then Section 2 of this Agreement will not apply to you. \niii. You must retain\
\ in all copies of the Llama Materials that you distribute the following attribution\
\ notice within a “Notice” text file distributed as a part of such copies: “Llama\
\ 3.2 is licensed under the Llama 3.2 Community License, Copyright © Meta Platforms,\
\ Inc. All Rights Reserved.”\niv. Your use of the Llama Materials must comply with\
\ applicable laws and regulations (including trade compliance laws and regulations)\
\ and adhere to the Acceptable Use Policy for the Llama Materials (available at\
\ https://www.llama.com/llama3_2/use-policy), which is hereby incorporated by reference\
\ into this Agreement.\n \n2. Additional Commercial Terms. If, on the Llama 3.2\
\ version release date, the monthly active users of the products or services made\
\ available by or for Licensee, or Licensee’s affiliates, is greater than 700 million\
\ monthly active users in the preceding calendar month, you must request a license\
\ from Meta, which Meta may grant to you in its sole discretion, and you are not\
\ authorized to exercise any of the rights under this Agreement unless or until\
\ Meta otherwise expressly grants you such rights.\n3. Disclaimer of Warranty. UNLESS\
\ REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS THEREFROM\
\ ARE PROVIDED ON AN “AS IS” BASIS, WITHOUT WARRANTIES OF ANY KIND, AND META DISCLAIMS\
\ ALL WARRANTIES OF ANY KIND, BOTH EXPRESS AND IMPLIED, INCLUDING, WITHOUT LIMITATION,\
\ ANY WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR\
\ PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING\
\ OR REDISTRIBUTING THE LLAMA MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR\
\ USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS.\n4. Limitation of Liability.\
\ IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF LIABILITY,\
\ WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING\
\ OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL,\
\ INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE\
\ BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.\n5. Intellectual Property.\n\
a. No trademark licenses are granted under this Agreement, and in connection with\
\ the Llama Materials, neither Meta nor Licensee may use any name or mark owned\
\ by or associated with the other or any of its affiliates, except as required\
\ for reasonable and customary use in describing and redistributing the Llama Materials\
\ or as set forth in this Section 5(a). Meta hereby grants you a license to use\
\ “Llama” (the “Mark”) solely as required to comply with the last sentence of Section\
\ 1.b.i. You will comply with Meta’s brand guidelines (currently accessible at\
\ https://about.meta.com/brand/resources/meta/company-brand/). All goodwill arising\
\ out of your use of the Mark will inure to the benefit of Meta.\nb. Subject to\
\ Meta’s ownership of Llama Materials and derivatives made by or for Meta, with\
\ respect to any derivative works and modifications of the Llama Materials that\
\ are made by you, as between you and Meta, you are and will be the owner of such\
\ derivative works and modifications.\nc. If you institute litigation or other proceedings\
\ against Meta or any entity (including a cross-claim or counterclaim in a lawsuit)\
\ alleging that the Llama Materials or Llama 3.2 outputs or results, or any portion\
\ of any of the foregoing, constitutes infringement of intellectual property or\
\ other rights owned or licensable by you, then any licenses granted to you under\
\ this Agreement shall terminate as of the date such litigation or claim is filed\
\ or instituted. You will indemnify and hold harmless Meta from and against any\
\ claim by any third party arising out of or related to your use or distribution\
\ of the Llama Materials.\n6. Term and Termination. The term of this Agreement will\
\ commence upon your acceptance of this Agreement or access to the Llama Materials\
\ and will continue in full force and effect until terminated in accordance with\
\ the terms and conditions herein. Meta may terminate this Agreement if you are\
\ in breach of any term or condition of this Agreement. Upon termination of this\
\ Agreement, you shall delete and cease use of the Llama Materials. Sections 3,\
\ 4 and 7 shall survive the termination of this Agreement. \n7. Governing Law and\
\ Jurisdiction. This Agreement will be governed and construed under the laws of\
\ the State of California without regard to choice of law principles, and the UN\
\ Convention on Contracts for the International Sale of Goods does not apply to\
\ this Agreement. The courts of California shall have exclusive jurisdiction of\
\ any dispute arising out of this Agreement. \n### Llama 3.2 Acceptable Use Policy\n\
Meta is committed to promoting safe and fair use of its tools and features, including\
\ Llama 3.2. If you access or use Llama 3.2, you agree to this Acceptable Use Policy\
\ (“**Policy**”). The most recent copy of this policy can be found at [https://www.llama.com/llama3_2/use-policy](https://www.llama.com/llama3_2/use-policy).\n\
#### Prohibited Uses\nWe want everyone to use Llama 3.2 safely and responsibly.\
\ You agree you will not use, or allow others to use, Llama 3.2 to:\n1. Violate\
\ the law or others’ rights, including to:\n 1. Engage in, promote, generate,\
\ contribute to, encourage, plan, incite, or further illegal or unlawful activity\
\ or content, such as:\n 1. Violence or terrorism\n 2. Exploitation\
\ or harm to children, including the solicitation, creation, acquisition, or dissemination\
\ of child exploitative content or failure to report Child Sexual Abuse Material\n\
\ 3. Human trafficking, exploitation, and sexual violence\n 4. The\
\ illegal distribution of information or materials to minors, including obscene\
\ materials, or failure to employ legally required age-gating in connection with\
\ such information or materials.\n 5. Sexual solicitation\n 6. Any\
\ other criminal activity\n 1. Engage in, promote, incite, or facilitate the\
\ harassment, abuse, threatening, or bullying of individuals or groups of individuals\n\
\ 2. Engage in, promote, incite, or facilitate discrimination or other unlawful\
\ or harmful conduct in the provision of employment, employment benefits, credit,\
\ housing, other economic benefits, or other essential goods and services\n 3.\
\ Engage in the unauthorized or unlicensed practice of any profession including,\
\ but not limited to, financial, legal, medical/health, or related professional\
\ practices\n 4. Collect, process, disclose, generate, or infer private or sensitive\
\ information about individuals, including information about individuals’ identity,\
\ health, or demographic information, unless you have obtained the right to do so\
\ in accordance with applicable law\n 5. Engage in or facilitate any action or\
\ generate any content that infringes, misappropriates, or otherwise violates any\
\ third-party rights, including the outputs or results of any products or services\
\ using the Llama Materials\n 6. Create, generate, or facilitate the creation\
\ of malicious code, malware, computer viruses or do anything else that could disable,\
\ overburden, interfere with or impair the proper working, integrity, operation\
\ or appearance of a website or computer system\n 7. Engage in any action, or\
\ facilitate any action, to intentionally circumvent or remove usage restrictions\
\ or other safety measures, or to enable functionality disabled by Meta \n2. Engage\
\ in, promote, incite, facilitate, or assist in the planning or development of activities\
\ that present a risk of death or bodily harm to individuals, including use of Llama\
\ 3.2 related to the following:\n 8. Military, warfare, nuclear industries or\
\ applications, espionage, use for materials or activities that are subject to the\
\ International Traffic Arms Regulations (ITAR) maintained by the United States\
\ Department of State or to the U.S. Biological Weapons Anti-Terrorism Act of 1989\
\ or the Chemical Weapons Convention Implementation Act of 1997\n 9. Guns and\
\ illegal weapons (including weapon development)\n 10. Illegal drugs and regulated/controlled\
\ substances\n 11. Operation of critical infrastructure, transportation technologies,\
\ or heavy machinery\n 12. Self-harm or harm to others, including suicide, cutting,\
\ and eating disorders\n 13. Any content intended to incite or promote violence,\
\ abuse, or any infliction of bodily harm to an individual\n3. Intentionally deceive\
\ or mislead others, including use of Llama 3.2 related to the following:\n 14.\
\ Generating, promoting, or furthering fraud or the creation or promotion of disinformation\n\
\ 15. Generating, promoting, or furthering defamatory content, including the\
\ creation of defamatory statements, images, or other content\n 16. Generating,\
\ promoting, or further distributing spam\n 17. Impersonating another individual\
\ without consent, authorization, or legal right\n 18. Representing that the\
\ use of Llama 3.2 or outputs are human-generated\n 19. Generating or facilitating\
\ false online engagement, including fake reviews and other means of fake online\
\ engagement \n4. Fail to appropriately disclose to end users any known dangers\
\ of your AI system 5. Interact with third party tools, models, or software designed\
\ to generate unlawful content or engage in unlawful or harmful conduct and/or represent\
\ that the outputs of such tools, models, or software are associated with Meta or\
\ Llama 3.2\n\nWith respect to any multimodal models included in Llama 3.2, the\
\ rights granted under Section 1(a) of the Llama 3.2 Community License Agreement\
\ are not being granted to you if you are an individual domiciled in, or a company\
\ with a principal place of business in, the European Union. This restriction does\
\ not apply to end users of a product or service that incorporates any such multimodal\
\ models.\n\nPlease report any violation of this Policy, software “bug,” or other\
\ problems that could lead to a violation of this Policy through one of the following\
\ means:\n\n* Reporting issues with the model: [https://github.com/meta-llama/llama-models/issues](https://l.workplace.com/l.php?u=https%3A%2F%2Fgithub.com%2Fmeta-llama%2Fllama-models%2Fissues&h=AT0qV8W9BFT6NwihiOHRuKYQM_UnkzN_NmHMy91OT55gkLpgi4kQupHUl0ssR4dQsIQ8n3tfd0vtkobvsEvt1l4Ic6GXI2EeuHV8N08OG2WnbAmm0FL4ObkazC6G_256vN0lN9DsykCvCqGZ)\n\
* Reporting risky content generated by the model: [developers.facebook.com/llama_output_feedback](http://developers.facebook.com/llama_output_feedback)\n\
* Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)\n\
* Reporting violations of the Acceptable Use Policy or unlicensed uses of Llama\
\ 3.2: [email protected]"
extra_gated_fields:
First Name: text
Last Name: text
Date of birth: date_picker
Country: country
Affiliation: text
Job title:
type: select
options:
- Student
- Research Graduate
- AI researcher
- AI developer/engineer
- Reporter
- Other
geo: ip_location
? By clicking Submit below I accept the terms of the license and acknowledge that
the information I provide will be collected stored processed and shared in accordance
with the Meta Privacy Policy
: checkbox
extra_gated_description: The information you provide will be collected, stored, processed
and shared in accordance with the [Meta Privacy Policy](https://www.facebook.com/privacy/policy/).
extra_gated_button_content: Submit
extra_gated_eu_disallowed: true
---
## Model Information
The Llama 3.2-Vision collection of multimodal large language models (LLMs) is a collection of pretrained and instruction-tuned image reasoning generative models in 11B and 90B sizes (text \+ images in / text out). The Llama 3.2-Vision instruction-tuned models are optimized for visual recognition, image reasoning, captioning, and answering general questions about an image. The models outperform many of the available open source and closed multimodal models on common industry benchmarks.
**Model Developer**: Meta
**Model Architecture:** Llama 3.2-Vision is built on top of Llama 3.1 text-only model, which is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety. To support image recognition tasks, the Llama 3.2-Vision model uses a separately trained vision adapter that integrates with the pre-trained Llama 3.1 language model. The adapter consists of a series of cross-attention layers that feed image encoder representations into the core LLM.
| | Training Data | Params | Input modalities | Output modalities | Context length | GQA | Data volume | Knowledge cutoff |
| :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- |
| Llama 3.2-Vision | (Image, text) pairs | 11B (10.6) | Text \+ Image | Text | 128k | Yes | 6B (image, text) pairs | December 2023 |
| Llama 3.2-Vision | (Image, text) pairs | 90B (88.8) | Text \+ Image | Text | 128k | Yes | 6B (image, text) pairs | December 2023 |
**Supported Languages:** For text only tasks, English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai are officially supported. Llama 3.2 has been trained on a broader collection of languages than these 8 supported languages. Note for image+text applications, English is the only language supported.
Developers may fine-tune Llama 3.2 models for languages beyond these supported languages, provided they comply with the Llama 3.2 Community License and the Acceptable Use Policy. Developers are always expected to ensure that their deployments, including those that involve additional languages, are completed safely and responsibly.
**Llama 3.2 Model Family:** Token counts refer to pretraining data only. All model versions use Grouped-Query Attention (GQA) for improved inference scalability.
**Model Release Date:** Sept 25, 2024
**Status:** This is a static model trained on an offline dataset. Future versions may be released that improve model capabilities and safety.
**License:** Use of Llama 3.2 is governed by the [Llama 3.2 Community License](https://github.com/meta-llama/llama-models/blob/main/models/llama3_2/LICENSE) (a custom, commercial license agreement).
**Feedback:** Where to send questions or comments about the model Instructions on how to provide feedback or comments on the model can be found in the model [README](https://github.com/meta-llama/llama-models/tree/main/models/llama3_2). For more technical information about generation parameters and recipes for how to use Llama 3.2-Vision in applications, please go [here](https://github.com/meta-llama/llama-recipes).
## Intended Use
**Intended Use Cases:** Llama 3.2-Vision is intended for commercial and research use. Instruction tuned models are intended for visual recognition, image reasoning, captioning, and assistant-like chat with images, whereas pretrained models can be adapted for a variety of image reasoning tasks. Additionally, because of Llama 3.2-Vision’s ability to take images and text as inputs, additional use cases could include:
1. Visual Question Answering (VQA) and Visual Reasoning: Imagine a machine that looks at a picture and understands your questions about it.
2. Document Visual Question Answering (DocVQA): Imagine a computer understanding both the text and layout of a document, like a map or contract, and then answering questions about it directly from the image.
3. Image Captioning: Image captioning bridges the gap between vision and language, extracting details, understanding the scene, and then crafting a sentence or two that tells the story.
4. Image-Text Retrieval: Image-text retrieval is like a matchmaker for images and their descriptions. Similar to a search engine but one that understands both pictures and words.
5. Visual Grounding: Visual grounding is like connecting the dots between what we see and say. It’s about understanding how language references specific parts of an image, allowing AI models to pinpoint objects or regions based on natural language descriptions.
The Llama 3.2 model collection also supports the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. The Llama 3.2 Community License allows for these use cases.
**Out of Scope:** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in any other way that is prohibited by the Acceptable Use Policy and Llama 3.2 Community License. Use in languages beyond those explicitly referenced as supported in this model card.
## How to use
This repository contains two versions of Llama-3.2-90B-Vision, for use with transformers and with the original `llama` codebase.
### Use with transformers
Starting with transformers >= 4.45.0 onward, you can run inference to generate text based on an image and a starting prompt you supply.
Make sure to update your transformers installation via `pip install --upgrade transformers`.
```python
import requests
import torch
from PIL import Image
from transformers import MllamaForConditionalGeneration, AutoProcessor
model_id = "meta-llama/Llama-3.2-90B-Vision"
model = MllamaForConditionalGeneration.from_pretrained(
model_id,
torch_dtype=torch.bfloat16,
device_map="auto",
)
processor = AutoProcessor.from_pretrained(model_id)
url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/0052a70beed5bf71b92610a43a52df6d286cd5f3/diffusers/rabbit.jpg"
image = Image.open(requests.get(url, stream=True).raw)
prompt = "<|image|><|begin_of_text|>If I had to write a haiku for this one"
inputs = processor(image, prompt, return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=30)
print(processor.decode(output[0]))
```
### Use with `llama`
Please, follow the instructions in the [repository](https://github.com/meta-llama/llama).
To download the original checkpoints, you can use `huggingface-cli` as follows:
```
huggingface-cli download meta-llama/Llama-3.2-90B-Vision --include "original/*" --local-dir Llama-3.2-90B-Vision
```
## Hardware and Software
**Training Factors:** We used custom training libraries, Meta's custom built GPU cluster, and production infrastructure for pretraining. Fine-tuning, annotation, and evaluation were also performed on production infrastructure.
**Training Energy Use:** Training utilized a cumulative of **2.02M** GPU hours of computation on H100-80GB (TDP of 700W) type hardware, per the table below. Training time is the total GPU time required for training each model and power consumption is the peak power capacity per GPU device used, adjusted for power usage efficiency.
##
**Training Greenhouse Gas Emissions:** Estimated total location-based greenhouse gas emissions were **584** tons CO2eq for training. Since 2020, Meta has maintained net zero greenhouse gas emissions in its global operations and matched 100% of its electricity use with renewable energy, therefore the total market-based greenhouse gas emissions for training were 0 tons CO2eq.
| | Training Time (GPU hours) | Training Power Consumption (W) | Training Location-Based Greenhouse Gas Emissions (tons CO2eq) | Training Market-Based Greenhouse Gas Emissions (tons CO2eq) |
| :---- | :---: | :---: | :---: | :---: |
| Llama 3.2-vision 11B | Stage 1 pretraining: 147K H100 hours Stage 2 annealing: 98K H100 hours SFT: 896 H100 hours RLHF: 224 H100 hours | 700 | 71 | 0 |
| Llama 3.2-vision 90B | Stage 1 pretraining: 885K H100 hours Stage 2 annealing: 885K H100 hours SFT: 3072 H100 hours RLHF: 2048 H100 hours | 700 | 513 | 0 |
| Total | 2.02M | | 584 | 0 |
The methodology used to determine training energy use and greenhouse gas emissions can be found [here](https://arxiv.org/pdf/2204.05149). Since Meta is openly releasing these models, the training energy use and greenhouse gas emissions will not be incurred by others.
## Training Data
**Overview:** Llama 3.2-Vision was pretrained on 6B image and text pairs. The instruction tuning data includes publicly available vision instruction datasets, as well as over 3M synthetically generated examples.
**Data Freshness:** The pretraining data has a cutoff of December 2023\.
## Benchmarks \- Image Reasoning
In this section, we report the results for Llama 3.2-Vision models on standard automatic benchmarks. For all these evaluations, we used our internal evaluations library.
### Base Pretrained Models
| Category | Benchmark | \# Shots | Metric | Llama 3.2 11B | Llama 3.2 90B |
| ----- | ----- | ----- | ----- | ----- | ----- |
| Image Understanding | VQAv2 (val) | 0 | Accuracy | 66.8 | 73.6 |
| | Text VQA (val) | 0 | Relaxed accuracy | 73.1 | 73.5 |
| | DocVQA (val, unseen) | 0 | ANLS | 62.3 | 70.7 |
| Visual Reasoning | MMMU (val, 0-shot) | 0 | Micro average accuracy | 41.7 | 49.3 |
| | ChartQA (test) | 0 | Accuracy | 39.4 | 54.2 |
| | InfographicsQA (val, unseen) | 0 | ANLS | 43.2 | 56.8 |
| | AI2 Diagram (test) | 0 | Accuracy | 62.4 | 75.3 |
### Instruction Tuned Models
| Modality | Capability | Benchmark | \# Shots | Metric | Llama 3.2 11B | Llama 3.2 90B |
| ----- | :---: | ----- | :---: | :---: | ----- | ----- |
| Image | College-level Problems and Mathematical Reasoning | MMMU (val, CoT) | 0 | Micro average accuracy | 50.7 | 60.3 |
| | | MMMU-Pro, Standard (10 opts, test) | 0 | Accuracy | 33.0 | 45.2 |
| | | MMMU-Pro, Vision (test) | 0 | Accuracy | 23.7 | 33.8 |
| | | MathVista (testmini) | 0 | Accuracy | 51.5 | 57.3 |
| | Charts and Diagram Understanding | ChartQA (test, CoT) | 0 | Relaxed accuracy | 83.4 | 85.5 |
| | | AI2 Diagram (test) | 0 | Accuracy | 91.1 | 92.3 |
| | | DocVQA (test) | 0 | ANLS | 88.4 | 90.1 |
| | General Visual Question Answering | VQAv2 (test) | 0 | Accuracy | 75.2 | 78.1 |
| | | | | | | |
| Text | General | MMLU (CoT) | 0 | Macro\_avg/acc | 73.0 | 86.0 |
| | Math | MATH (CoT) | 0 | Final\_em | 51.9 | 68.0 |
| | Reasoning | GPQA | 0 | Accuracy | 32.8 | 46.7 |
| | Multilingual | MGSM (CoT) | 0 | em | 68.9 | 86.9 |
## Responsibility & Safety
As part of our Responsible release approach, we followed a three-pronged strategy to managing trust & safety risks:
1. Enable developers to deploy helpful, safe and flexible experiences for their target audience and for the use cases supported by Llama.
2. Protect developers against adversarial users aiming to exploit Llama capabilities to potentially cause harm.
3. Provide protections for the community to help prevent the misuse of our models.
### Responsible Deployment
**Approach:** Llama is a foundational technology designed to be used in a variety of use cases, examples on how Meta’s Llama models have been responsibly deployed can be found in our [Community Stories webpage](https://llama.meta.com/community-stories/). Our approach is to build the most helpful models enabling the world to benefit from the technology power, by aligning our model safety for the generic use cases addressing a standard set of harms. Developers are then in the driver seat to tailor safety for their use case, defining their own policy and deploying the models with the necessary safeguards in their Llama systems. Llama 3.2 was developed following the best practices outlined in our Responsible Use Guide, you can refer to the [Responsible Use Guide](https://llama.meta.com/responsible-use-guide/) to learn more.
#### Llama 3.2 Instruct
**Objective:** Our main objectives for conducting safety fine-tuning are to provide the research community with a valuable resource for studying the robustness of safety fine-tuning, as well as to offer developers a readily available, safe, and powerful model for various applications to reduce the developer workload to deploy safe AI systems. We implemented the same set of safety mitigations as in Llama 3, and you can learn more about these in the Llama 3 [paper](https://ai.meta.com/research/publications/the-llama-3-herd-of-models/).
**Fine-Tuning Data:** We employ a multi-faceted approach to data collection, combining human-generated data from our vendors with synthetic data to mitigate potential safety risks. We’ve developed many large language model (LLM)-based classifiers that enable us to thoughtfully select high-quality prompts and responses, enhancing data quality control.
**Refusals and Tone:** Building on the work we started with Llama 3, we put a great emphasis on model refusals to benign prompts as well as refusal tone. We included both borderline and adversarial prompts in our safety data strategy, and modified our safety data responses to follow tone guidelines.
#### Llama 3.2 Systems
**Safety as a System:** Large language models, including Llama 3.2, **are not designed to be deployed in isolation** but instead should be deployed as part of an overall AI system with additional safety guardrails as required. Developers are expected to deploy system safeguards when building agentic systems. Safeguards are key to achieve the right helpfulness-safety alignment as well as mitigating safety and security risks inherent to the system and any integration of the model or system with external tools. As part of our responsible release approach, we provide the community with [safeguards](https://llama.meta.com/trust-and-safety/) that developers should deploy with Llama models or other LLMs, including Llama Guard, Prompt Guard and Code Shield. All our [reference implementations](https://github.com/meta-llama/llama-agentic-system) demos contain these safeguards by default so developers can benefit from system-level safety out-of-the-box.
### New Capabilities and Use Cases
**Technological Advancement:** Llama releases usually introduce new capabilities that require specific considerations in addition to the best practices that generally apply across all Generative AI use cases. For prior release capabilities also supported by Llama 3.2, see [Llama 3.1 Model Card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/MODEL_CARD.md), as the same considerations apply here as well.,
**Image Reasoning:** Llama 3.2-Vision models come with multimodal (text and image) input capabilities enabling image reasoning applications. As part of our responsible release process, we took dedicated measures including evaluations and mitigations to address the risk of the models uniquely identifying individuals in images. As with other LLM risks, models may not always be robust to adversarial prompts, and developers should evaluate identification and other applicable risks in the context of their applications as well as consider deploying Llama Guard 3-11B-Vision as part of their system or other mitigations as appropriate to detect and mitigate such risks.
### Evaluations
**Scaled Evaluations:** We built dedicated, adversarial evaluation datasets and evaluated systems composed of Llama models and Purple Llama safeguards to filter input prompt and output response. It is important to evaluate applications in context, and we recommend building dedicated evaluation dataset for your use case.
**Red teaming:** We conducted recurring red teaming exercises with the goal of discovering risks via adversarial prompting and we used the learnings to improve our benchmarks and safety tuning datasets. We partnered early with subject-matter experts in critical risk areas to understand the nature of these real-world harms and how such models may lead to unintended harm for society. Based on these conversations, we derived a set of adversarial goals for the red team to attempt to achieve, such as extracting harmful information or reprogramming the model to act in a potentially harmful capacity. The red team consisted of experts in cybersecurity, adversarial machine learning, responsible AI, and integrity in addition to multilingual content specialists with background in integrity issues in specific geographic markets.
### Critical Risks
In addition to our safety work above, we took extra care on measuring and/or mitigating the following critical risk areas:
**1\. CBRNE (Chemical, Biological, Radiological, Nuclear, and Explosive Weapons):** For Llama 3.1, to assess risks related to proliferation of chemical and biological weapons, we performed uplift testing designed to assess whether use of Llama 3.1 models could meaningfully increase the capabilities of malicious actors to plan or carry out attacks using these types of weapons. For Llama 3.2-Vision models, we conducted additional targeted evaluations and found that it was unlikely Llama 3.2 presented an increase in scientific capabilities due to its added image understanding capability as compared to Llama 3.1.
**2\. Child Safety:** Child Safety risk assessments were conducted using a team of experts, to assess the model’s capability to produce outputs that could result in Child Safety risks and inform on any necessary and appropriate risk mitigations via fine tuning. We leveraged those expert red teaming sessions to expand the coverage of our evaluation benchmarks through Llama 3 model development. For Llama 3, we conducted new in-depth sessions using objective based methodologies to assess the model risks along multiple attack vectors including the additional languages Llama 3 is trained on. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences.
**3\. Cyber Attacks:** For Llama 3.1 405B, our cyber attack uplift study investigated whether LLMs can enhance human capabilities in hacking tasks, both in terms of skill level and speed.
Our attack automation study focused on evaluating the capabilities of LLMs when used as autonomous agents in cyber offensive operations, specifically in the context of ransomware attacks. This evaluation was distinct from previous studies that considered LLMs as interactive assistants. The primary objective was to assess whether these models could effectively function as independent agents in executing complex cyber-attacks without human intervention. Because Llama 3.2’s vision capabilities are not generally germane to cyber uplift, we believe that the testing conducted for Llama 3.1 also applies to Llama 3.2.
### Community
**Industry Partnerships:** Generative AI safety requires expertise and tooling, and we believe in the strength of the open community to accelerate its progress. We are active members of open consortiums, including the AI Alliance, Partnership on AI and MLCommons, actively contributing to safety standardization and transparency. We encourage the community to adopt taxonomies like the MLCommons Proof of Concept evaluation to facilitate collaboration and transparency on safety and content evaluations. Our Purple Llama tools are open sourced for the community to use and widely distributed across ecosystem partners including cloud service providers. We encourage community contributions to our [Github repository](https://github.com/meta-llama/PurpleLlama).
**Grants:** We also set up the [Llama Impact Grants](https://llama.meta.com/llama-impact-grants/) program to identify and support the most compelling applications of Meta’s Llama model for societal benefit across three categories: education, climate and open innovation. The 20 finalists from the hundreds of applications can be found [here](https://llama.meta.com/llama-impact-grants/#finalists).
**Reporting:** Finally, we put in place a set of resources including an [output reporting mechanism](https://developers.facebook.com/llama_output_feedback) and [bug bounty program](https://www.facebook.com/whitehat) to continuously improve the Llama technology with the help of the community.
## Ethical Considerations and Limitations
**Values:** The core values of Llama 3.2 are openness, inclusivity and helpfulness. It is meant to serve everyone, and to work for a wide range of use cases. It is thus designed to be accessible to people across many different backgrounds, experiences and perspectives. Llama 3.2 addresses users and their needs as they are, without insertion unnecessary judgment or normativity, while reflecting the understanding that even content that may appear problematic in some cases can serve valuable purposes in others. It respects the dignity and autonomy of all users, especially in terms of the values of free thought and expression that power innovation and progress.
**Testing:** But Llama 3.2 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3.2’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 3.2 models, developers should perform safety testing and tuning tailored to their specific applications of the model. Please refer to available resources including our [Responsible Use Guide](https://llama.meta.com/responsible-use-guide), [Trust and Safety](https://llama.meta.com/trust-and-safety/) solutions, and other [resources](https://llama.meta.com/docs/get-started/) to learn more about responsible development.
| null |
Non_BioNLP
|
## Model Information
The Llama 3.2-Vision collection of multimodal large language models (LLMs) is a collection of pretrained and instruction-tuned image reasoning generative models in 11B and 90B sizes (text \+ images in / text out). The Llama 3.2-Vision instruction-tuned models are optimized for visual recognition, image reasoning, captioning, and answering general questions about an image. The models outperform many of the available open source and closed multimodal models on common industry benchmarks.
**Model Developer**: Meta
**Model Architecture:** Llama 3.2-Vision is built on top of Llama 3.1 text-only model, which is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety. To support image recognition tasks, the Llama 3.2-Vision model uses a separately trained vision adapter that integrates with the pre-trained Llama 3.1 language model. The adapter consists of a series of cross-attention layers that feed image encoder representations into the core LLM.
| | Training Data | Params | Input modalities | Output modalities | Context length | GQA | Data volume | Knowledge cutoff |
| :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- |
| Llama 3.2-Vision | (Image, text) pairs | 11B (10.6) | Text \+ Image | Text | 128k | Yes | 6B (image, text) pairs | December 2023 |
| Llama 3.2-Vision | (Image, text) pairs | 90B (88.8) | Text \+ Image | Text | 128k | Yes | 6B (image, text) pairs | December 2023 |
**Supported Languages:** For text only tasks, English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai are officially supported. Llama 3.2 has been trained on a broader collection of languages than these 8 supported languages. Note for image+text applications, English is the only language supported.
Developers may fine-tune Llama 3.2 models for languages beyond these supported languages, provided they comply with the Llama 3.2 Community License and the Acceptable Use Policy. Developers are always expected to ensure that their deployments, including those that involve additional languages, are completed safely and responsibly.
**Llama 3.2 Model Family:** Token counts refer to pretraining data only. All model versions use Grouped-Query Attention (GQA) for improved inference scalability.
**Model Release Date:** Sept 25, 2024
**Status:** This is a static model trained on an offline dataset. Future versions may be released that improve model capabilities and safety.
**License:** Use of Llama 3.2 is governed by the [Llama 3.2 Community License](https://github.com/meta-llama/llama-models/blob/main/models/llama3_2/LICENSE) (a custom, commercial license agreement).
**Feedback:** Where to send questions or comments about the model Instructions on how to provide feedback or comments on the model can be found in the model [README](https://github.com/meta-llama/llama-models/tree/main/models/llama3_2). For more technical information about generation parameters and recipes for how to use Llama 3.2-Vision in applications, please go [here](https://github.com/meta-llama/llama-recipes).
## Intended Use
**Intended Use Cases:** Llama 3.2-Vision is intended for commercial and research use. Instruction tuned models are intended for visual recognition, image reasoning, captioning, and assistant-like chat with images, whereas pretrained models can be adapted for a variety of image reasoning tasks. Additionally, because of Llama 3.2-Vision’s ability to take images and text as inputs, additional use cases could include:
1. Visual Question Answering (VQA) and Visual Reasoning: Imagine a machine that looks at a picture and understands your questions about it.
2. Document Visual Question Answering (DocVQA): Imagine a computer understanding both the text and layout of a document, like a map or contract, and then answering questions about it directly from the image.
3. Image Captioning: Image captioning bridges the gap between vision and language, extracting details, understanding the scene, and then crafting a sentence or two that tells the story.
4. Image-Text Retrieval: Image-text retrieval is like a matchmaker for images and their descriptions. Similar to a search engine but one that understands both pictures and words.
5. Visual Grounding: Visual grounding is like connecting the dots between what we see and say. It’s about understanding how language references specific parts of an image, allowing AI models to pinpoint objects or regions based on natural language descriptions.
The Llama 3.2 model collection also supports the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. The Llama 3.2 Community License allows for these use cases.
**Out of Scope:** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in any other way that is prohibited by the Acceptable Use Policy and Llama 3.2 Community License. Use in languages beyond those explicitly referenced as supported in this model card.
## How to use
This repository contains two versions of Llama-3.2-90B-Vision, for use with transformers and with the original `llama` codebase.
### Use with transformers
Starting with transformers >= 4.45.0 onward, you can run inference to generate text based on an image and a starting prompt you supply.
Make sure to update your transformers installation via `pip install --upgrade transformers`.
```python
import requests
import torch
from PIL import Image
from transformers import MllamaForConditionalGeneration, AutoProcessor
model_id = "meta-llama/Llama-3.2-90B-Vision"
model = MllamaForConditionalGeneration.from_pretrained(
model_id,
torch_dtype=torch.bfloat16,
device_map="auto",
)
processor = AutoProcessor.from_pretrained(model_id)
url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/0052a70beed5bf71b92610a43a52df6d286cd5f3/diffusers/rabbit.jpg"
image = Image.open(requests.get(url, stream=True).raw)
prompt = "<|image|><|begin_of_text|>If I had to write a haiku for this one"
inputs = processor(image, prompt, return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=30)
print(processor.decode(output[0]))
```
### Use with `llama`
Please, follow the instructions in the [repository](https://github.com/meta-llama/llama).
To download the original checkpoints, you can use `huggingface-cli` as follows:
```
huggingface-cli download meta-llama/Llama-3.2-90B-Vision --include "original/*" --local-dir Llama-3.2-90B-Vision
```
## Hardware and Software
**Training Factors:** We used custom training libraries, Meta's custom built GPU cluster, and production infrastructure for pretraining. Fine-tuning, annotation, and evaluation were also performed on production infrastructure.
**Training Energy Use:** Training utilized a cumulative of **2.02M** GPU hours of computation on H100-80GB (TDP of 700W) type hardware, per the table below. Training time is the total GPU time required for training each model and power consumption is the peak power capacity per GPU device used, adjusted for power usage efficiency.
##
**Training Greenhouse Gas Emissions:** Estimated total location-based greenhouse gas emissions were **584** tons CO2eq for training. Since 2020, Meta has maintained net zero greenhouse gas emissions in its global operations and matched 100% of its electricity use with renewable energy, therefore the total market-based greenhouse gas emissions for training were 0 tons CO2eq.
| | Training Time (GPU hours) | Training Power Consumption (W) | Training Location-Based Greenhouse Gas Emissions (tons CO2eq) | Training Market-Based Greenhouse Gas Emissions (tons CO2eq) |
| :---- | :---: | :---: | :---: | :---: |
| Llama 3.2-vision 11B | Stage 1 pretraining: 147K H100 hours Stage 2 annealing: 98K H100 hours SFT: 896 H100 hours RLHF: 224 H100 hours | 700 | 71 | 0 |
| Llama 3.2-vision 90B | Stage 1 pretraining: 885K H100 hours Stage 2 annealing: 885K H100 hours SFT: 3072 H100 hours RLHF: 2048 H100 hours | 700 | 513 | 0 |
| Total | 2.02M | | 584 | 0 |
The methodology used to determine training energy use and greenhouse gas emissions can be found [here](https://arxiv.org/pdf/2204.05149). Since Meta is openly releasing these models, the training energy use and greenhouse gas emissions will not be incurred by others.
## Training Data
**Overview:** Llama 3.2-Vision was pretrained on 6B image and text pairs. The instruction tuning data includes publicly available vision instruction datasets, as well as over 3M synthetically generated examples.
**Data Freshness:** The pretraining data has a cutoff of December 2023\.
## Benchmarks \- Image Reasoning
In this section, we report the results for Llama 3.2-Vision models on standard automatic benchmarks. For all these evaluations, we used our internal evaluations library.
### Base Pretrained Models
| Category | Benchmark | \# Shots | Metric | Llama 3.2 11B | Llama 3.2 90B |
| ----- | ----- | ----- | ----- | ----- | ----- |
| Image Understanding | VQAv2 (val) | 0 | Accuracy | 66.8 | 73.6 |
| | Text VQA (val) | 0 | Relaxed accuracy | 73.1 | 73.5 |
| | DocVQA (val, unseen) | 0 | ANLS | 62.3 | 70.7 |
| Visual Reasoning | MMMU (val, 0-shot) | 0 | Micro average accuracy | 41.7 | 49.3 |
| | ChartQA (test) | 0 | Accuracy | 39.4 | 54.2 |
| | InfographicsQA (val, unseen) | 0 | ANLS | 43.2 | 56.8 |
| | AI2 Diagram (test) | 0 | Accuracy | 62.4 | 75.3 |
### Instruction Tuned Models
| Modality | Capability | Benchmark | \# Shots | Metric | Llama 3.2 11B | Llama 3.2 90B |
| ----- | :---: | ----- | :---: | :---: | ----- | ----- |
| Image | College-level Problems and Mathematical Reasoning | MMMU (val, CoT) | 0 | Micro average accuracy | 50.7 | 60.3 |
| | | MMMU-Pro, Standard (10 opts, test) | 0 | Accuracy | 33.0 | 45.2 |
| | | MMMU-Pro, Vision (test) | 0 | Accuracy | 23.7 | 33.8 |
| | | MathVista (testmini) | 0 | Accuracy | 51.5 | 57.3 |
| | Charts and Diagram Understanding | ChartQA (test, CoT) | 0 | Relaxed accuracy | 83.4 | 85.5 |
| | | AI2 Diagram (test) | 0 | Accuracy | 91.1 | 92.3 |
| | | DocVQA (test) | 0 | ANLS | 88.4 | 90.1 |
| | General Visual Question Answering | VQAv2 (test) | 0 | Accuracy | 75.2 | 78.1 |
| | | | | | | |
| Text | General | MMLU (CoT) | 0 | Macro\_avg/acc | 73.0 | 86.0 |
| | Math | MATH (CoT) | 0 | Final\_em | 51.9 | 68.0 |
| | Reasoning | GPQA | 0 | Accuracy | 32.8 | 46.7 |
| | Multilingual | MGSM (CoT) | 0 | em | 68.9 | 86.9 |
## Responsibility & Safety
As part of our Responsible release approach, we followed a three-pronged strategy to managing trust & safety risks:
1. Enable developers to deploy helpful, safe and flexible experiences for their target audience and for the use cases supported by Llama.
2. Protect developers against adversarial users aiming to exploit Llama capabilities to potentially cause harm.
3. Provide protections for the community to help prevent the misuse of our models.
### Responsible Deployment
**Approach:** Llama is a foundational technology designed to be used in a variety of use cases, examples on how Meta’s Llama models have been responsibly deployed can be found in our [Community Stories webpage](https://llama.meta.com/community-stories/). Our approach is to build the most helpful models enabling the world to benefit from the technology power, by aligning our model safety for the generic use cases addressing a standard set of harms. Developers are then in the driver seat to tailor safety for their use case, defining their own policy and deploying the models with the necessary safeguards in their Llama systems. Llama 3.2 was developed following the best practices outlined in our Responsible Use Guide, you can refer to the [Responsible Use Guide](https://llama.meta.com/responsible-use-guide/) to learn more.
#### Llama 3.2 Instruct
**Objective:** Our main objectives for conducting safety fine-tuning are to provide the research community with a valuable resource for studying the robustness of safety fine-tuning, as well as to offer developers a readily available, safe, and powerful model for various applications to reduce the developer workload to deploy safe AI systems. We implemented the same set of safety mitigations as in Llama 3, and you can learn more about these in the Llama 3 [paper](https://ai.meta.com/research/publications/the-llama-3-herd-of-models/).
**Fine-Tuning Data:** We employ a multi-faceted approach to data collection, combining human-generated data from our vendors with synthetic data to mitigate potential safety risks. We’ve developed many large language model (LLM)-based classifiers that enable us to thoughtfully select high-quality prompts and responses, enhancing data quality control.
**Refusals and Tone:** Building on the work we started with Llama 3, we put a great emphasis on model refusals to benign prompts as well as refusal tone. We included both borderline and adversarial prompts in our safety data strategy, and modified our safety data responses to follow tone guidelines.
#### Llama 3.2 Systems
**Safety as a System:** Large language models, including Llama 3.2, **are not designed to be deployed in isolation** but instead should be deployed as part of an overall AI system with additional safety guardrails as required. Developers are expected to deploy system safeguards when building agentic systems. Safeguards are key to achieve the right helpfulness-safety alignment as well as mitigating safety and security risks inherent to the system and any integration of the model or system with external tools. As part of our responsible release approach, we provide the community with [safeguards](https://llama.meta.com/trust-and-safety/) that developers should deploy with Llama models or other LLMs, including Llama Guard, Prompt Guard and Code Shield. All our [reference implementations](https://github.com/meta-llama/llama-agentic-system) demos contain these safeguards by default so developers can benefit from system-level safety out-of-the-box.
### New Capabilities and Use Cases
**Technological Advancement:** Llama releases usually introduce new capabilities that require specific considerations in addition to the best practices that generally apply across all Generative AI use cases. For prior release capabilities also supported by Llama 3.2, see [Llama 3.1 Model Card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/MODEL_CARD.md), as the same considerations apply here as well.,
**Image Reasoning:** Llama 3.2-Vision models come with multimodal (text and image) input capabilities enabling image reasoning applications. As part of our responsible release process, we took dedicated measures including evaluations and mitigations to address the risk of the models uniquely identifying individuals in images. As with other LLM risks, models may not always be robust to adversarial prompts, and developers should evaluate identification and other applicable risks in the context of their applications as well as consider deploying Llama Guard 3-11B-Vision as part of their system or other mitigations as appropriate to detect and mitigate such risks.
### Evaluations
**Scaled Evaluations:** We built dedicated, adversarial evaluation datasets and evaluated systems composed of Llama models and Purple Llama safeguards to filter input prompt and output response. It is important to evaluate applications in context, and we recommend building dedicated evaluation dataset for your use case.
**Red teaming:** We conducted recurring red teaming exercises with the goal of discovering risks via adversarial prompting and we used the learnings to improve our benchmarks and safety tuning datasets. We partnered early with subject-matter experts in critical risk areas to understand the nature of these real-world harms and how such models may lead to unintended harm for society. Based on these conversations, we derived a set of adversarial goals for the red team to attempt to achieve, such as extracting harmful information or reprogramming the model to act in a potentially harmful capacity. The red team consisted of experts in cybersecurity, adversarial machine learning, responsible AI, and integrity in addition to multilingual content specialists with background in integrity issues in specific geographic markets.
### Critical Risks
In addition to our safety work above, we took extra care on measuring and/or mitigating the following critical risk areas:
**1\. CBRNE (Chemical, Biological, Radiological, Nuclear, and Explosive Weapons):** For Llama 3.1, to assess risks related to proliferation of chemical and biological weapons, we performed uplift testing designed to assess whether use of Llama 3.1 models could meaningfully increase the capabilities of malicious actors to plan or carry out attacks using these types of weapons. For Llama 3.2-Vision models, we conducted additional targeted evaluations and found that it was unlikely Llama 3.2 presented an increase in scientific capabilities due to its added image understanding capability as compared to Llama 3.1.
**2\. Child Safety:** Child Safety risk assessments were conducted using a team of experts, to assess the model’s capability to produce outputs that could result in Child Safety risks and inform on any necessary and appropriate risk mitigations via fine tuning. We leveraged those expert red teaming sessions to expand the coverage of our evaluation benchmarks through Llama 3 model development. For Llama 3, we conducted new in-depth sessions using objective based methodologies to assess the model risks along multiple attack vectors including the additional languages Llama 3 is trained on. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences.
**3\. Cyber Attacks:** For Llama 3.1 405B, our cyber attack uplift study investigated whether LLMs can enhance human capabilities in hacking tasks, both in terms of skill level and speed.
Our attack automation study focused on evaluating the capabilities of LLMs when used as autonomous agents in cyber offensive operations, specifically in the context of ransomware attacks. This evaluation was distinct from previous studies that considered LLMs as interactive assistants. The primary objective was to assess whether these models could effectively function as independent agents in executing complex cyber-attacks without human intervention. Because Llama 3.2’s vision capabilities are not generally germane to cyber uplift, we believe that the testing conducted for Llama 3.1 also applies to Llama 3.2.
### Community
**Industry Partnerships:** Generative AI safety requires expertise and tooling, and we believe in the strength of the open community to accelerate its progress. We are active members of open consortiums, including the AI Alliance, Partnership on AI and MLCommons, actively contributing to safety standardization and transparency. We encourage the community to adopt taxonomies like the MLCommons Proof of Concept evaluation to facilitate collaboration and transparency on safety and content evaluations. Our Purple Llama tools are open sourced for the community to use and widely distributed across ecosystem partners including cloud service providers. We encourage community contributions to our [Github repository](https://github.com/meta-llama/PurpleLlama).
**Grants:** We also set up the [Llama Impact Grants](https://llama.meta.com/llama-impact-grants/) program to identify and support the most compelling applications of Meta’s Llama model for societal benefit across three categories: education, climate and open innovation. The 20 finalists from the hundreds of applications can be found [here](https://llama.meta.com/llama-impact-grants/#finalists).
**Reporting:** Finally, we put in place a set of resources including an [output reporting mechanism](https://developers.facebook.com/llama_output_feedback) and [bug bounty program](https://www.facebook.com/whitehat) to continuously improve the Llama technology with the help of the community.
## Ethical Considerations and Limitations
**Values:** The core values of Llama 3.2 are openness, inclusivity and helpfulness. It is meant to serve everyone, and to work for a wide range of use cases. It is thus designed to be accessible to people across many different backgrounds, experiences and perspectives. Llama 3.2 addresses users and their needs as they are, without insertion unnecessary judgment or normativity, while reflecting the understanding that even content that may appear problematic in some cases can serve valuable purposes in others. It respects the dignity and autonomy of all users, especially in terms of the values of free thought and expression that power innovation and progress.
**Testing:** But Llama 3.2 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3.2’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 3.2 models, developers should perform safety testing and tuning tailored to their specific applications of the model. Please refer to available resources including our [Responsible Use Guide](https://llama.meta.com/responsible-use-guide), [Trust and Safety](https://llama.meta.com/trust-and-safety/) solutions, and other [resources](https://llama.meta.com/docs/get-started/) to learn more about responsible development.
|
{"language": ["en", "de", "fr", "it", "pt", "hi", "es", "th"], "library_name": "transformers", "license": "llama3.2", "pipeline_tag": "image-text-to-text", "tags": ["facebook", "meta", "pytorch", "llama", "llama-3"], "extra_gated_prompt": "### LLAMA 3.2 COMMUNITY LICENSE AGREEMENT\n\nLlama 3.2 Version Release Date: September 25, 2024\n\n“Agreement” means the terms and conditions for use, reproduction, distribution and modification of the Llama Materials set forth herein.\n\n“Documentation” means the specifications, manuals and documentation accompanying Llama 3.2 distributed by Meta at https://llama.meta.com/doc/overview.\n\n“Licensee” or “you” means you, or your employer or any other person or entity (if you are entering into this Agreement on such person or entity’s behalf), of the age required under applicable laws, rules or regulations to provide legal consent and that has legal authority to bind your employer or such other person or entity if you are entering in this Agreement on their behalf.\n\n“Llama 3.2” means the foundational large language models and software and algorithms, including machine-learning model code, trained model weights, inference-enabling code, training-enabling code, fine-tuning enabling code and other elements of the foregoing distributed by Meta at https://www.llama.com/llama-downloads.\n\n“Llama Materials” means, collectively, Meta’s proprietary Llama 3.2 and Documentation (and any portion thereof) made available under this Agreement.\n\n“Meta” or “we” means Meta Platforms Ireland Limited (if you are located in or, if you are an entity, your principal place of business is in the EEA or Switzerland) and Meta Platforms, Inc. (if you are located outside of the EEA or Switzerland). \n\nBy clicking “I Accept” below or by using or distributing any portion or element of the Llama Materials, you agree to be bound by this Agreement.\n\n1. License Rights and Redistribution.\na. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable and royalty-free limited license under Meta’s intellectual property or other rights owned by Meta embodied in the Llama Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Llama Materials. \nb. Redistribution and Use. \ni. If you distribute or make available the Llama Materials (or any derivative works thereof), or a product or service (including another AI model) that contains any of them, you shall (A) provide a copy of this Agreement with any such Llama Materials; and (B) prominently display “Built with Llama” on a related website, user interface, blogpost, about page, or product documentation. If you use the Llama Materials or any outputs or results of the Llama Materials to create, train, fine tune, or otherwise improve an AI model, which is distributed or made available, you shall also include “Llama” at the beginning of any such AI model name.\nii. If you receive Llama Materials, or any derivative works thereof, from a Licensee as part of an integrated end user product, then Section 2 of this Agreement will not apply to you. \niii. You must retain in all copies of the Llama Materials that you distribute the following attribution notice within a “Notice” text file distributed as a part of such copies: “Llama 3.2 is licensed under the Llama 3.2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.”\niv. Your use of the Llama Materials must comply with applicable laws and regulations (including trade compliance laws and regulations) and adhere to the Acceptable Use Policy for the Llama Materials (available at https://www.llama.com/llama3_2/use-policy), which is hereby incorporated by reference into this Agreement.\n \n2. Additional Commercial Terms. If, on the Llama 3.2 version release date, the monthly active users of the products or services made available by or for Licensee, or Licensee’s affiliates, is greater than 700 million monthly active users in the preceding calendar month, you must request a license from Meta, which Meta may grant to you in its sole discretion, and you are not authorized to exercise any of the rights under this Agreement unless or until Meta otherwise expressly grants you such rights.\n3. Disclaimer of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS” BASIS, WITHOUT WARRANTIES OF ANY KIND, AND META DISCLAIMS ALL WARRANTIES OF ANY KIND, BOTH EXPRESS AND IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE LLAMA MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS.\n4. Limitation of Liability. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.\n5. Intellectual Property.\na. No trademark licenses are granted under this Agreement, and in connection with the Llama Materials, neither Meta nor Licensee may use any name or mark owned by or associated with the other or any of its affiliates, except as required for reasonable and customary use in describing and redistributing the Llama Materials or as set forth in this Section 5(a). Meta hereby grants you a license to use “Llama” (the “Mark”) solely as required to comply with the last sentence of Section 1.b.i. You will comply with Meta’s brand guidelines (currently accessible at https://about.meta.com/brand/resources/meta/company-brand/). All goodwill arising out of your use of the Mark will inure to the benefit of Meta.\nb. Subject to Meta’s ownership of Llama Materials and derivatives made by or for Meta, with respect to any derivative works and modifications of the Llama Materials that are made by you, as between you and Meta, you are and will be the owner of such derivative works and modifications.\nc. If you institute litigation or other proceedings against Meta or any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Llama Materials or Llama 3.2 outputs or results, or any portion of any of the foregoing, constitutes infringement of intellectual property or other rights owned or licensable by you, then any licenses granted to you under this Agreement shall terminate as of the date such litigation or claim is filed or instituted. You will indemnify and hold harmless Meta from and against any claim by any third party arising out of or related to your use or distribution of the Llama Materials.\n6. Term and Termination. The term of this Agreement will commence upon your acceptance of this Agreement or access to the Llama Materials and will continue in full force and effect until terminated in accordance with the terms and conditions herein. Meta may terminate this Agreement if you are in breach of any term or condition of this Agreement. Upon termination of this Agreement, you shall delete and cease use of the Llama Materials. Sections 3, 4 and 7 shall survive the termination of this Agreement. \n7. Governing Law and Jurisdiction. This Agreement will be governed and construed under the laws of the State of California without regard to choice of law principles, and the UN Convention on Contracts for the International Sale of Goods does not apply to this Agreement. The courts of California shall have exclusive jurisdiction of any dispute arising out of this Agreement. \n### Llama 3.2 Acceptable Use Policy\nMeta is committed to promoting safe and fair use of its tools and features, including Llama 3.2. If you access or use Llama 3.2, you agree to this Acceptable Use Policy (“**Policy**”). The most recent copy of this policy can be found at [https://www.llama.com/llama3_2/use-policy](https://www.llama.com/llama3_2/use-policy).\n#### Prohibited Uses\nWe want everyone to use Llama 3.2 safely and responsibly. You agree you will not use, or allow others to use, Llama 3.2 to:\n1. Violate the law or others’ rights, including to:\n 1. Engage in, promote, generate, contribute to, encourage, plan, incite, or further illegal or unlawful activity or content, such as:\n 1. Violence or terrorism\n 2. Exploitation or harm to children, including the solicitation, creation, acquisition, or dissemination of child exploitative content or failure to report Child Sexual Abuse Material\n 3. Human trafficking, exploitation, and sexual violence\n 4. The illegal distribution of information or materials to minors, including obscene materials, or failure to employ legally required age-gating in connection with such information or materials.\n 5. Sexual solicitation\n 6. Any other criminal activity\n 1. Engage in, promote, incite, or facilitate the harassment, abuse, threatening, or bullying of individuals or groups of individuals\n 2. Engage in, promote, incite, or facilitate discrimination or other unlawful or harmful conduct in the provision of employment, employment benefits, credit, housing, other economic benefits, or other essential goods and services\n 3. Engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or related professional practices\n 4. Collect, process, disclose, generate, or infer private or sensitive information about individuals, including information about individuals’ identity, health, or demographic information, unless you have obtained the right to do so in accordance with applicable law\n 5. Engage in or facilitate any action or generate any content that infringes, misappropriates, or otherwise violates any third-party rights, including the outputs or results of any products or services using the Llama Materials\n 6. Create, generate, or facilitate the creation of malicious code, malware, computer viruses or do anything else that could disable, overburden, interfere with or impair the proper working, integrity, operation or appearance of a website or computer system\n 7. Engage in any action, or facilitate any action, to intentionally circumvent or remove usage restrictions or other safety measures, or to enable functionality disabled by Meta \n2. Engage in, promote, incite, facilitate, or assist in the planning or development of activities that present a risk of death or bodily harm to individuals, including use of Llama 3.2 related to the following:\n 8. Military, warfare, nuclear industries or applications, espionage, use for materials or activities that are subject to the International Traffic Arms Regulations (ITAR) maintained by the United States Department of State or to the U.S. Biological Weapons Anti-Terrorism Act of 1989 or the Chemical Weapons Convention Implementation Act of 1997\n 9. Guns and illegal weapons (including weapon development)\n 10. Illegal drugs and regulated/controlled substances\n 11. Operation of critical infrastructure, transportation technologies, or heavy machinery\n 12. Self-harm or harm to others, including suicide, cutting, and eating disorders\n 13. Any content intended to incite or promote violence, abuse, or any infliction of bodily harm to an individual\n3. Intentionally deceive or mislead others, including use of Llama 3.2 related to the following:\n 14. Generating, promoting, or furthering fraud or the creation or promotion of disinformation\n 15. Generating, promoting, or furthering defamatory content, including the creation of defamatory statements, images, or other content\n 16. Generating, promoting, or further distributing spam\n 17. Impersonating another individual without consent, authorization, or legal right\n 18. Representing that the use of Llama 3.2 or outputs are human-generated\n 19. Generating or facilitating false online engagement, including fake reviews and other means of fake online engagement \n4. Fail to appropriately disclose to end users any known dangers of your AI system 5. Interact with third party tools, models, or software designed to generate unlawful content or engage in unlawful or harmful conduct and/or represent that the outputs of such tools, models, or software are associated with Meta or Llama 3.2\n\nWith respect to any multimodal models included in Llama 3.2, the rights granted under Section 1(a) of the Llama 3.2 Community License Agreement are not being granted to you if you are an individual domiciled in, or a company with a principal place of business in, the European Union. This restriction does not apply to end users of a product or service that incorporates any such multimodal models.\n\nPlease report any violation of this Policy, software “bug,” or other problems that could lead to a violation of this Policy through one of the following means:\n\n* Reporting issues with the model: [https://github.com/meta-llama/llama-models/issues](https://l.workplace.com/l.php?u=https%3A%2F%2Fgithub.com%2Fmeta-llama%2Fllama-models%2Fissues&h=AT0qV8W9BFT6NwihiOHRuKYQM_UnkzN_NmHMy91OT55gkLpgi4kQupHUl0ssR4dQsIQ8n3tfd0vtkobvsEvt1l4Ic6GXI2EeuHV8N08OG2WnbAmm0FL4ObkazC6G_256vN0lN9DsykCvCqGZ)\n* Reporting risky content generated by the model: [developers.facebook.com/llama_output_feedback](http://developers.facebook.com/llama_output_feedback)\n* Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)\n* Reporting violations of the Acceptable Use Policy or unlicensed uses of Llama 3.2: [email protected]", "extra_gated_fields": {"First Name": "text", "Last Name": "text", "Date of birth": "date_picker", "Country": "country", "Affiliation": "text", "Job title": {"type": "select", "options": ["Student", "Research Graduate", "AI researcher", "AI developer/engineer", "Reporter", "Other"]}, "geo": "ip_location", "By clicking Submit below I accept the terms of the license and acknowledge that the information I provide will be collected stored processed and shared in accordance with the Meta Privacy Policy": "checkbox"}, "extra_gated_description": "The information you provide will be collected, stored, processed and shared in accordance with the [Meta Privacy Policy](https://www.facebook.com/privacy/policy/).", "extra_gated_button_content": "Submit", "extra_gated_eu_disallowed": true}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,174 |
eevvgg/Stance-Tw
|
eevvgg
|
text-classification
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"text",
"stance",
"en",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2022-11-26T11:33:47Z |
2023-03-08T09:00:09+00:00
| 33 | 0 |
---
language:
- en
pipeline_tag: text-classification
tags:
- text
- stance
- text-classification
widget:
- text: user Bolsonaro is the president of Brazil. He speaks for all brazilians. Greta
is a climate activist. Their opinions do create a balance that the world needs
now
example_title: example 1
- text: user The fact is that she still doesn’t change her ways and still stays non
environmental friendly
example_title: example 2
- text: user The criteria for these awards dont seem to be very high.
example_title: example 3
model-index:
- name: Stance-Tw
results:
- task:
type: stance-classification
name: Text Classification
dataset:
name: stance
type: stance
metrics:
- type: f1
value: 75.8
- type: accuracy
value: 76.2
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# Stance-Tw
This model is a fine-tuned version of [j-hartmann/sentiment-roberta-large-english-3-classes](https://huggingface.co/j-hartmann/sentiment-roberta-large-english-3-classes) to predict 3 categories of author stance (attack, support, neutral) towards an entity mentioned in the text.
- training procedure available in [Colab notebook](https://colab.research.google.com/drive/12DsO5dNaQI3kFO7ohOHZn4EWNewFy2jm?usp=sharing)
- result of a collaboration with [Laboratory of The New Ethos](https://newethos.org/laboratory/)
```
# Model usage
from transformers import pipeline
model_path = "eevvgg/Stance-Tw"
cls_task = pipeline(task = "text-classification", model = model_path, tokenizer = model_path)#, device=0
sequence = ['his rambling has no clear ideas behind it',
'That has nothing to do with medical care',
"Turns around and shows how qualified she is because of her political career.",
'She has very little to gain by speaking too much']
result = cls_task(sequence)
labels = [i['label'] for i in result]
labels # ['attack', 'neutral', 'support', 'attack']
```
## Intended uses & limitations
Model suited for classification of stance in short text. Fine-tuned on a manually-annotated corpus of size 3.2k.
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'learning_rate': 4e-5, 'decay': 0.01}
Trained for 3 epochs, mini-batch size of 8.
- loss: 0.719
## Evaluation data
It achieves the following results on the evaluation set:
- macro f1-score: 0.758
- weighted f1-score: 0.762
- accuracy: 0.762
## Citation
**BibTeX**: tba
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# Stance-Tw
This model is a fine-tuned version of [j-hartmann/sentiment-roberta-large-english-3-classes](https://huggingface.co/j-hartmann/sentiment-roberta-large-english-3-classes) to predict 3 categories of author stance (attack, support, neutral) towards an entity mentioned in the text.
- training procedure available in [Colab notebook](https://colab.research.google.com/drive/12DsO5dNaQI3kFO7ohOHZn4EWNewFy2jm?usp=sharing)
- result of a collaboration with [Laboratory of The New Ethos](https://newethos.org/laboratory/)
```
# Model usage
from transformers import pipeline
model_path = "eevvgg/Stance-Tw"
cls_task = pipeline(task = "text-classification", model = model_path, tokenizer = model_path)#, device=0
sequence = ['his rambling has no clear ideas behind it',
'That has nothing to do with medical care',
"Turns around and shows how qualified she is because of her political career.",
'She has very little to gain by speaking too much']
result = cls_task(sequence)
labels = [i['label'] for i in result]
labels # ['attack', 'neutral', 'support', 'attack']
```
## Intended uses & limitations
Model suited for classification of stance in short text. Fine-tuned on a manually-annotated corpus of size 3.2k.
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'learning_rate': 4e-5, 'decay': 0.01}
Trained for 3 epochs, mini-batch size of 8.
- loss: 0.719
## Evaluation data
It achieves the following results on the evaluation set:
- macro f1-score: 0.758
- weighted f1-score: 0.762
- accuracy: 0.762
## Citation
**BibTeX**: tba
|
{"language": ["en"], "pipeline_tag": "text-classification", "tags": ["text", "stance", "text-classification"], "widget": [{"text": "user Bolsonaro is the president of Brazil. He speaks for all brazilians. Greta is a climate activist. Their opinions do create a balance that the world needs now", "example_title": "example 1"}, {"text": "user The fact is that she still doesn’t change her ways and still stays non environmental friendly", "example_title": "example 2"}, {"text": "user The criteria for these awards dont seem to be very high.", "example_title": "example 3"}], "model-index": [{"name": "Stance-Tw", "results": [{"task": {"type": "stance-classification", "name": "Text Classification"}, "dataset": {"name": "stance", "type": "stance"}, "metrics": [{"type": "f1", "value": 75.8}, {"type": "accuracy", "value": 76.2}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,176 |
PlanTL-GOB-ES/roberta-large-bne-te
|
PlanTL-GOB-ES
|
text-classification
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"national library of spain",
"spanish",
"bne",
"xnli",
"textual entailment",
"es",
"dataset:xnli",
"arxiv:1907.11692",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2022-11-25T09:49:22Z |
2022-11-25T09:52:00+00:00
| 13 | 1 |
---
datasets:
- xnli
language:
- es
license: apache-2.0
metrics:
- accuracy
tags:
- national library of spain
- spanish
- bne
- xnli
- textual entailment
widget:
- Me gustas. Te quiero.
- Mi cumpleaños es el 27 de mayo. Cumpliré años a finales de mayo.
model-index:
- name: roberta-large-bne-te
results:
- task:
type: text-classification
dataset:
name: XNLI
type: xnli
metrics:
- type: accuracy
value: 0.8263
name: Accuracy
---
# Spanish RoBERTa-large trained on BNE finetuned for the Spanish Cross-lingual Natural Language Inference (XNLI) dataset.
## Table of contents
<details>
<summary>Click to expand</summary>
- [Model description](#model-description)
- [Intended uses and limitations](#intended-use)
- [How to use](#how-to-use)
- [Limitations and bias](#limitations-and-bias)
- [Training](#training)
- [Training](#training)
- [Training data](#training-data)
- [Training procedure](#training-procedure)
- [Evaluation](#evaluation)
- [Evaluation](#evaluation)
- [Variable and metrics](#variable-and-metrics)
- [Evaluation results](#evaluation-results)
- [Additional information](#additional-information)
- [Author](#author)
- [Contact information](#contact-information)
- [Copyright](#copyright)
- [Licensing information](#licensing-information)
- [Funding](#funding)
- [Citing information](#citing-information)
- [Disclaimer](#disclaimer)
</details>
## Model description
The **roberta-large-bne-te** is a Textual Entailment (TE) model for the Spanish language fine-tuned from the [roberta-large-bne](https://huggingface.co/PlanTL-GOB-ES/roberta-large-bne) model, a [RoBERTa](https://arxiv.org/abs/1907.11692) large model pre-trained using the largest Spanish corpus known to date, with a total of 570GB of clean and deduplicated text, processed for this work, compiled from the web crawlings performed by the [National Library of Spain (Biblioteca Nacional de España)](http://www.bne.es/en/Inicio/index.html) from 2009 to 2019.
## Intended uses and limitations
**roberta-large-bne-te** model can be used to recognize Textual Entailment (TE). The model is limited by its training dataset and may not generalize well for all use cases.
## How to use
Here is how to use this model:
```python
from transformers import pipeline
from pprint import pprint
nlp = pipeline("text-classification", model="PlanTL-GOB-ES/roberta-large-bne-te")
example = "Mi cumpleaños es el 27 de mayo. Cumpliré años a finales de mayo."
te_results = nlp(example)
pprint(te_results)
```
## Limitations and bias
At the time of submission, no measures have been taken to estimate the bias embedded in the model. However, we are well aware that our models may be biased since the corpora have been collected using crawling techniques on multiple web sources. We intend to conduct research in these areas in the future, and if completed, this model card will be updated.
## Training
We used the TE dataset in Spanish called [XNLI dataset](https://huggingface.co/datasets/xnli) for training and evaluation.
### Training procedure
The model was trained with a batch size of 16 and a learning rate of 1e-5 for 5 epochs. We then selected the best checkpoint using the downstream task metric in the corresponding development set and then evaluated it on the test set.
## Evaluation
### Variable and metrics
This model was finetuned maximizing accuracy.
## Evaluation results
We evaluated the *roberta-large-bne-te* on the XNLI test set against standard multilingual and monolingual baselines:
| Model | XNLI (Accuracy) |
| ------------|:----|
| roberta-large-bne | **82.63** |
| roberta-base-bne | 80.16 |
| BETO | 81.30 |
| mBERT | 78.76 |
| BERTIN | 78.90 |
| ELECTRA | 78.78 |
For more details, check the fine-tuning and evaluation scripts in the official [GitHub repository](https://github.com/PlanTL-GOB-ES/lm-spanish).
## Additional information
### Author
Text Mining Unit (TeMU) at the Barcelona Supercomputing Center ([email protected])
### Contact information
For further information, send an email to <[email protected]>
### Copyright
Copyright by the Spanish State Secretariat for Digitalization and Artificial Intelligence (SEDIA) (2022)
### Licensing information
[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)
### Funding
This work was funded by the Spanish State Secretariat for Digitalization and Artificial Intelligence (SEDIA) within the framework of the Plan-TL.
## Citing information
If you use this model, please cite our [paper](http://journal.sepln.org/sepln/ojs/ojs/index.php/pln/article/view/6405):
```
@article{,
abstract = {We want to thank the National Library of Spain for such a large effort on the data gathering and the Future of Computing Center, a
Barcelona Supercomputing Center and IBM initiative (2020). This work was funded by the Spanish State Secretariat for Digitalization and Artificial
Intelligence (SEDIA) within the framework of the Plan-TL.},
author = {Asier Gutiérrez Fandiño and Jordi Armengol Estapé and Marc Pàmies and Joan Llop Palao and Joaquin Silveira Ocampo and Casimiro Pio Carrino and Carme Armentano Oller and Carlos Rodriguez Penagos and Aitor Gonzalez Agirre and Marta Villegas},
doi = {10.26342/2022-68-3},
issn = {1135-5948},
journal = {Procesamiento del Lenguaje Natural},
keywords = {Artificial intelligence,Benchmarking,Data processing.,MarIA,Natural language processing,Spanish language modelling,Spanish language resources,Tractament del llenguatge natural (Informàtica),Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Llenguatge natural},
publisher = {Sociedad Española para el Procesamiento del Lenguaje Natural},
title = {MarIA: Spanish Language Models},
volume = {68},
url = {https://upcommons.upc.edu/handle/2117/367156#.YyMTB4X9A-0.mendeley},
year = {2022},
}
```
## Disclaimer
<details>
<summary>Click to expand</summary>
The models published in this repository are intended for a generalist purpose and are available to third parties. These models may have bias and/or any other undesirable distortions.
When third parties, deploy or provide systems and/or services to other parties using any of these models (or using systems based on these models) or become users of the models, they should note that it is their responsibility to mitigate the risks arising from their use and, in any event, to comply with applicable regulations, including regulations regarding the use of artificial intelligence.
In no event shall the owner of the models (SEDIA – State Secretariat for digitalization and artificial intelligence) nor the creator (BSC – Barcelona Supercomputing Center) be liable for any results arising from the use made by third parties of these models.
Los modelos publicados en este repositorio tienen una finalidad generalista y están a disposición de terceros. Estos modelos pueden tener sesgos y/u otro tipo de distorsiones indeseables.
Cuando terceros desplieguen o proporcionen sistemas y/o servicios a otras partes usando alguno de estos modelos (o utilizando sistemas basados en estos modelos) o se conviertan en usuarios de los modelos, deben tener en cuenta que es su responsabilidad mitigar los riesgos derivados de su uso y, en todo caso, cumplir con la normativa aplicable, incluyendo la normativa en materia de uso de inteligencia artificial.
En ningún caso el propietario de los modelos (SEDIA – Secretaría de Estado de Digitalización e Inteligencia Artificial) ni el creador (BSC – Barcelona Supercomputing Center) serán responsables de los resultados derivados del uso que hagan terceros de estos modelos.
| null |
Non_BioNLP
|
# Spanish RoBERTa-large trained on BNE finetuned for the Spanish Cross-lingual Natural Language Inference (XNLI) dataset.
## Table of contents
<details>
<summary>Click to expand</summary>
- [Model description](#model-description)
- [Intended uses and limitations](#intended-use)
- [How to use](#how-to-use)
- [Limitations and bias](#limitations-and-bias)
- [Training](#training)
- [Training](#training)
- [Training data](#training-data)
- [Training procedure](#training-procedure)
- [Evaluation](#evaluation)
- [Evaluation](#evaluation)
- [Variable and metrics](#variable-and-metrics)
- [Evaluation results](#evaluation-results)
- [Additional information](#additional-information)
- [Author](#author)
- [Contact information](#contact-information)
- [Copyright](#copyright)
- [Licensing information](#licensing-information)
- [Funding](#funding)
- [Citing information](#citing-information)
- [Disclaimer](#disclaimer)
</details>
## Model description
The **roberta-large-bne-te** is a Textual Entailment (TE) model for the Spanish language fine-tuned from the [roberta-large-bne](https://huggingface.co/PlanTL-GOB-ES/roberta-large-bne) model, a [RoBERTa](https://arxiv.org/abs/1907.11692) large model pre-trained using the largest Spanish corpus known to date, with a total of 570GB of clean and deduplicated text, processed for this work, compiled from the web crawlings performed by the [National Library of Spain (Biblioteca Nacional de España)](http://www.bne.es/en/Inicio/index.html) from 2009 to 2019.
## Intended uses and limitations
**roberta-large-bne-te** model can be used to recognize Textual Entailment (TE). The model is limited by its training dataset and may not generalize well for all use cases.
## How to use
Here is how to use this model:
```python
from transformers import pipeline
from pprint import pprint
nlp = pipeline("text-classification", model="PlanTL-GOB-ES/roberta-large-bne-te")
example = "Mi cumpleaños es el 27 de mayo. Cumpliré años a finales de mayo."
te_results = nlp(example)
pprint(te_results)
```
## Limitations and bias
At the time of submission, no measures have been taken to estimate the bias embedded in the model. However, we are well aware that our models may be biased since the corpora have been collected using crawling techniques on multiple web sources. We intend to conduct research in these areas in the future, and if completed, this model card will be updated.
## Training
We used the TE dataset in Spanish called [XNLI dataset](https://huggingface.co/datasets/xnli) for training and evaluation.
### Training procedure
The model was trained with a batch size of 16 and a learning rate of 1e-5 for 5 epochs. We then selected the best checkpoint using the downstream task metric in the corresponding development set and then evaluated it on the test set.
## Evaluation
### Variable and metrics
This model was finetuned maximizing accuracy.
## Evaluation results
We evaluated the *roberta-large-bne-te* on the XNLI test set against standard multilingual and monolingual baselines:
| Model | XNLI (Accuracy) |
| ------------|:----|
| roberta-large-bne | **82.63** |
| roberta-base-bne | 80.16 |
| BETO | 81.30 |
| mBERT | 78.76 |
| BERTIN | 78.90 |
| ELECTRA | 78.78 |
For more details, check the fine-tuning and evaluation scripts in the official [GitHub repository](https://github.com/PlanTL-GOB-ES/lm-spanish).
## Additional information
### Author
Text Mining Unit (TeMU) at the Barcelona Supercomputing Center ([email protected])
### Contact information
For further information, send an email to <[email protected]>
### Copyright
Copyright by the Spanish State Secretariat for Digitalization and Artificial Intelligence (SEDIA) (2022)
### Licensing information
[Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0)
### Funding
This work was funded by the Spanish State Secretariat for Digitalization and Artificial Intelligence (SEDIA) within the framework of the Plan-TL.
## Citing information
If you use this model, please cite our [paper](http://journal.sepln.org/sepln/ojs/ojs/index.php/pln/article/view/6405):
```
@article{,
abstract = {We want to thank the National Library of Spain for such a large effort on the data gathering and the Future of Computing Center, a
Barcelona Supercomputing Center and IBM initiative (2020). This work was funded by the Spanish State Secretariat for Digitalization and Artificial
Intelligence (SEDIA) within the framework of the Plan-TL.},
author = {Asier Gutiérrez Fandiño and Jordi Armengol Estapé and Marc Pàmies and Joan Llop Palao and Joaquin Silveira Ocampo and Casimiro Pio Carrino and Carme Armentano Oller and Carlos Rodriguez Penagos and Aitor Gonzalez Agirre and Marta Villegas},
doi = {10.26342/2022-68-3},
issn = {1135-5948},
journal = {Procesamiento del Lenguaje Natural},
keywords = {Artificial intelligence,Benchmarking,Data processing.,MarIA,Natural language processing,Spanish language modelling,Spanish language resources,Tractament del llenguatge natural (Informàtica),Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Llenguatge natural},
publisher = {Sociedad Española para el Procesamiento del Lenguaje Natural},
title = {MarIA: Spanish Language Models},
volume = {68},
url = {https://upcommons.upc.edu/handle/2117/367156#.YyMTB4X9A-0.mendeley},
year = {2022},
}
```
## Disclaimer
<details>
<summary>Click to expand</summary>
The models published in this repository are intended for a generalist purpose and are available to third parties. These models may have bias and/or any other undesirable distortions.
When third parties, deploy or provide systems and/or services to other parties using any of these models (or using systems based on these models) or become users of the models, they should note that it is their responsibility to mitigate the risks arising from their use and, in any event, to comply with applicable regulations, including regulations regarding the use of artificial intelligence.
In no event shall the owner of the models (SEDIA – State Secretariat for digitalization and artificial intelligence) nor the creator (BSC – Barcelona Supercomputing Center) be liable for any results arising from the use made by third parties of these models.
Los modelos publicados en este repositorio tienen una finalidad generalista y están a disposición de terceros. Estos modelos pueden tener sesgos y/u otro tipo de distorsiones indeseables.
Cuando terceros desplieguen o proporcionen sistemas y/o servicios a otras partes usando alguno de estos modelos (o utilizando sistemas basados en estos modelos) o se conviertan en usuarios de los modelos, deben tener en cuenta que es su responsabilidad mitigar los riesgos derivados de su uso y, en todo caso, cumplir con la normativa aplicable, incluyendo la normativa en materia de uso de inteligencia artificial.
En ningún caso el propietario de los modelos (SEDIA – Secretaría de Estado de Digitalización e Inteligencia Artificial) ni el creador (BSC – Barcelona Supercomputing Center) serán responsables de los resultados derivados del uso que hagan terceros de estos modelos.
|
{"datasets": ["xnli"], "language": ["es"], "license": "apache-2.0", "metrics": ["accuracy"], "tags": ["national library of spain", "spanish", "bne", "xnli", "textual entailment"], "widget": ["Me gustas. Te quiero.", "Mi cumpleaños es el 27 de mayo. Cumpliré años a finales de mayo."], "model-index": [{"name": "roberta-large-bne-te", "results": [{"task": {"type": "text-classification"}, "dataset": {"name": "XNLI", "type": "xnli"}, "metrics": [{"type": "accuracy", "value": 0.8263, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXTUAL_ENTAILMENT"
] | 45,177 |
gaudi/opus-mt-fr-bg-ctranslate2
|
gaudi
|
translation
|
[
"transformers",
"marian",
"ctranslate2",
"translation",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | 2024-07-22T15:56:21Z |
2024-10-19T04:16:02+00:00
| 6 | 0 |
---
license: apache-2.0
tags:
- ctranslate2
- translation
---
# Repository General Information
## Inspired by and derived from the work of [Helsinki-NLP](https://huggingface.co/Helsinki-NLP), [CTranslate2](https://github.com/OpenNMT/CTranslate2), and [michaelfeil](https://huggingface.co/michaelfeil)!
- Link to Original Model ([Helsinki-NLP](https://huggingface.co/Helsinki-NLP)): [Model Link](https://huggingface.co/Helsinki-NLP/opus-mt-fr-bg)
- This respository was based on the work of [CTranslate2](https://github.com/OpenNMT/CTranslate2).
- This repository was based on the work of [michaelfeil](https://huggingface.co/michaelfeil).
# What is CTranslate2?
[CTranslate2](https://opennmt.net/CTranslate2/) is a C++ and Python library for efficient inference with Transformer models.
CTranslate2 implements a custom runtime that applies many performance optimization techniques such as weights quantization, layers fusion, batch reordering, etc., to accelerate and reduce the memory usage of Transformer models on CPU and GPU.
CTranslate2 is one of the most performant ways of hosting translation models at scale. Current supported models include:
- Encoder-decoder models: Transformer base/big, M2M-100, NLLB, BART, mBART, Pegasus, T5, Whisper
- Decoder-only models: GPT-2, GPT-J, GPT-NeoX, OPT, BLOOM, MPT, Llama, Mistral, Gemma, CodeGen, GPTBigCode, Falcon
- Encoder-only models: BERT, DistilBERT, XLM-RoBERTa
The project is production-oriented and comes with backward compatibility guarantees, but it also includes experimental features related to model compression and inference acceleration.
# CTranslate2 Benchmarks
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings. Tested against `newstest2014` (En -> De) dataset.
The benchmark reports the number of target tokens generated per second (higher is better). The results are aggregated over multiple runs. See the benchmark scripts for more details and reproduce these numbers.
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings.
## CPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 147.3 | 2332MB | 27.90 |
| Marian 1.11.0 (int16) | 330.2 | 5901MB | 27.65 |
| Marian 1.11.0 (int8) | 355.8 | 4763MB | 27.27 |
| CTranslate2 3.6.0 (int16) | 596.1 | 660MB | 27.53 |
| CTranslate2 3.6.0 (int8) | 696.1 | 516MB | 27.65 |
## GPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max GPU Memory Usage | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 1022.9 | 4097MB | 2109MB | 27.90 |
| Marian 1.11.0 (float16) | 3962.4 | 3239MB | 1976MB | 27.94 |
| CTranslate2 3.6.0 (float16) | 9296.7 | 909MB | 814MB | 27.9 |
| CTranslate2 3.6.0 (int8 + float16) | 8362.7 | 813MB | 766MB | 27.9 |
`Executed with 4 threads on a c5.2xlarge Amazon EC2 instance equipped with an Intel(R) Xeon(R) Platinum 8275CL CPU.`
**Source to benchmark information can be found [here](https://github.com/OpenNMT/CTranslate2).**<br />
**Original model BLEU scores can be found [here](https://huggingface.co/Helsinki-NLP/opus-mt-fr-bg).**
## Internal Benchmarks
Internal testing on our end showed **inference times reduced by 6x-10x** on average compared the vanilla checkpoints using the *transformers* library. A **slight reduction on BLEU scores (~5%)** was also identified in comparison to the vanilla checkpoints with a few exceptions. This is likely due to several factors, one being the quantization applied. Further testing is needed from our end to better assess the reduction in translation quality. The command used to compile the vanilla checkpoint into a CTranslate2 model can be found below. Modifying this command can yield differing balances between inferencing performance and translation quality.
# CTranslate2 Installation
```bash
pip install hf-hub-ctranslate2>=1.0.0 ctranslate2>=3.13.0
```
### ct2-transformers-converter Command Used:
```bash
ct2-transformers-converter --model Helsinki-NLP/opus-mt-fr-bg --output_dir ./ctranslate2/opus-mt-fr-bg-ctranslate2 --force --copy_files README.md generation_config.json tokenizer_config.json vocab.json source.spm .gitattributes target.spm --quantization float16
```
# CTranslate2 Converted Checkpoint Information:
**Compatible With:**
- [ctranslate2](https://github.com/OpenNMT/CTranslate2)
- [hf-hub-ctranslate2](https://github.com/michaelfeil/hf-hub-ctranslate2)
**Compute Type:**
- `compute_type=int8_float16` for `device="cuda"`
- `compute_type=int8` for `device="cpu"`
# Sample Code - ctranslate2
#### Clone the repository to the working directory or wherever you wish to store the model artifacts. ####
```bash
git clone https://huggingface.co/gaudi/opus-mt-fr-bg-ctranslate2
```
#### Take the python code below and update the 'model_dir' variable to the location of the cloned repository. ####
```python
from ctranslate2 import Translator
import transformers
model_dir = "./opus-mt-fr-bg-ctranslate2" # Path to model directory.
translator = Translator(
model_path=model_dir,
device="cuda", # cpu, cuda, or auto.
inter_threads=1, # Maximum number of parallel translations.
intra_threads=4, # Number of OpenMP threads per translator.
compute_type="int8_float16", # int8 for cpu or int8_float16 for cuda.
)
tokenizer = transformers.AutoTokenizer.from_pretrained(model_dir)
source = tokenizer.convert_ids_to_tokens(tokenizer.encode("XXXXXX, XXX XX XXXXXX."))
results = translator.translate_batch([source])
target = results[0].hypotheses[0]
print(tokenizer.decode(tokenizer.convert_tokens_to_ids(target)))
```
# Sample Code - hf-hub-ctranslate2
**Derived From [michaelfeil](https://huggingface.co/michaelfeil):**
```python
from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
from transformers import AutoTokenizer
model_name = "gaudi/opus-mt-fr-bg-ctranslate2"
model = TranslatorCT2fromHfHub(
model_name_or_path=model_name,
device="cuda",
compute_type="int8_float16",
tokenizer=AutoTokenizer.from_pretrained(model_name)
)
outputs = model.generate(
text=["XXX XX XXX XXXXXXX XXXX?", "XX XX XXXX XX XXX!"],
)
print(outputs)
```
# License and other remarks:
License conditions are intended to be idential to [original huggingface repository](https://huggingface.co/Helsinki-NLP/opus-mt-fr-bg) by Helsinki-NLP.
| null |
Non_BioNLP
|
# Repository General Information
## Inspired by and derived from the work of [Helsinki-NLP](https://huggingface.co/Helsinki-NLP), [CTranslate2](https://github.com/OpenNMT/CTranslate2), and [michaelfeil](https://huggingface.co/michaelfeil)!
- Link to Original Model ([Helsinki-NLP](https://huggingface.co/Helsinki-NLP)): [Model Link](https://huggingface.co/Helsinki-NLP/opus-mt-fr-bg)
- This respository was based on the work of [CTranslate2](https://github.com/OpenNMT/CTranslate2).
- This repository was based on the work of [michaelfeil](https://huggingface.co/michaelfeil).
# What is CTranslate2?
[CTranslate2](https://opennmt.net/CTranslate2/) is a C++ and Python library for efficient inference with Transformer models.
CTranslate2 implements a custom runtime that applies many performance optimization techniques such as weights quantization, layers fusion, batch reordering, etc., to accelerate and reduce the memory usage of Transformer models on CPU and GPU.
CTranslate2 is one of the most performant ways of hosting translation models at scale. Current supported models include:
- Encoder-decoder models: Transformer base/big, M2M-100, NLLB, BART, mBART, Pegasus, T5, Whisper
- Decoder-only models: GPT-2, GPT-J, GPT-NeoX, OPT, BLOOM, MPT, Llama, Mistral, Gemma, CodeGen, GPTBigCode, Falcon
- Encoder-only models: BERT, DistilBERT, XLM-RoBERTa
The project is production-oriented and comes with backward compatibility guarantees, but it also includes experimental features related to model compression and inference acceleration.
# CTranslate2 Benchmarks
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings. Tested against `newstest2014` (En -> De) dataset.
The benchmark reports the number of target tokens generated per second (higher is better). The results are aggregated over multiple runs. See the benchmark scripts for more details and reproduce these numbers.
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings.
## CPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 147.3 | 2332MB | 27.90 |
| Marian 1.11.0 (int16) | 330.2 | 5901MB | 27.65 |
| Marian 1.11.0 (int8) | 355.8 | 4763MB | 27.27 |
| CTranslate2 3.6.0 (int16) | 596.1 | 660MB | 27.53 |
| CTranslate2 3.6.0 (int8) | 696.1 | 516MB | 27.65 |
## GPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max GPU Memory Usage | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 1022.9 | 4097MB | 2109MB | 27.90 |
| Marian 1.11.0 (float16) | 3962.4 | 3239MB | 1976MB | 27.94 |
| CTranslate2 3.6.0 (float16) | 9296.7 | 909MB | 814MB | 27.9 |
| CTranslate2 3.6.0 (int8 + float16) | 8362.7 | 813MB | 766MB | 27.9 |
`Executed with 4 threads on a c5.2xlarge Amazon EC2 instance equipped with an Intel(R) Xeon(R) Platinum 8275CL CPU.`
**Source to benchmark information can be found [here](https://github.com/OpenNMT/CTranslate2).**<br />
**Original model BLEU scores can be found [here](https://huggingface.co/Helsinki-NLP/opus-mt-fr-bg).**
## Internal Benchmarks
Internal testing on our end showed **inference times reduced by 6x-10x** on average compared the vanilla checkpoints using the *transformers* library. A **slight reduction on BLEU scores (~5%)** was also identified in comparison to the vanilla checkpoints with a few exceptions. This is likely due to several factors, one being the quantization applied. Further testing is needed from our end to better assess the reduction in translation quality. The command used to compile the vanilla checkpoint into a CTranslate2 model can be found below. Modifying this command can yield differing balances between inferencing performance and translation quality.
# CTranslate2 Installation
```bash
pip install hf-hub-ctranslate2>=1.0.0 ctranslate2>=3.13.0
```
### ct2-transformers-converter Command Used:
```bash
ct2-transformers-converter --model Helsinki-NLP/opus-mt-fr-bg --output_dir ./ctranslate2/opus-mt-fr-bg-ctranslate2 --force --copy_files README.md generation_config.json tokenizer_config.json vocab.json source.spm .gitattributes target.spm --quantization float16
```
# CTranslate2 Converted Checkpoint Information:
**Compatible With:**
- [ctranslate2](https://github.com/OpenNMT/CTranslate2)
- [hf-hub-ctranslate2](https://github.com/michaelfeil/hf-hub-ctranslate2)
**Compute Type:**
- `compute_type=int8_float16` for `device="cuda"`
- `compute_type=int8` for `device="cpu"`
# Sample Code - ctranslate2
#### Clone the repository to the working directory or wherever you wish to store the model artifacts. ####
```bash
git clone https://huggingface.co/gaudi/opus-mt-fr-bg-ctranslate2
```
#### Take the python code below and update the 'model_dir' variable to the location of the cloned repository. ####
```python
from ctranslate2 import Translator
import transformers
model_dir = "./opus-mt-fr-bg-ctranslate2" # Path to model directory.
translator = Translator(
model_path=model_dir,
device="cuda", # cpu, cuda, or auto.
inter_threads=1, # Maximum number of parallel translations.
intra_threads=4, # Number of OpenMP threads per translator.
compute_type="int8_float16", # int8 for cpu or int8_float16 for cuda.
)
tokenizer = transformers.AutoTokenizer.from_pretrained(model_dir)
source = tokenizer.convert_ids_to_tokens(tokenizer.encode("XXXXXX, XXX XX XXXXXX."))
results = translator.translate_batch([source])
target = results[0].hypotheses[0]
print(tokenizer.decode(tokenizer.convert_tokens_to_ids(target)))
```
# Sample Code - hf-hub-ctranslate2
**Derived From [michaelfeil](https://huggingface.co/michaelfeil):**
```python
from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
from transformers import AutoTokenizer
model_name = "gaudi/opus-mt-fr-bg-ctranslate2"
model = TranslatorCT2fromHfHub(
model_name_or_path=model_name,
device="cuda",
compute_type="int8_float16",
tokenizer=AutoTokenizer.from_pretrained(model_name)
)
outputs = model.generate(
text=["XXX XX XXX XXXXXXX XXXX?", "XX XX XXXX XX XXX!"],
)
print(outputs)
```
# License and other remarks:
License conditions are intended to be idential to [original huggingface repository](https://huggingface.co/Helsinki-NLP/opus-mt-fr-bg) by Helsinki-NLP.
|
{"license": "apache-2.0", "tags": ["ctranslate2", "translation"]}
|
task
|
[
"TRANSLATION"
] | 45,178 |
henryscheible/qnli_roberta-base_125
|
henryscheible
| null |
[
"pytorch",
"generated_from_trainer",
"en",
"dataset:glue",
"license:mit",
"model-index",
"region:us"
] | 2023-01-18T19:58:01Z |
2023-01-18T22:29:32+00:00
| 0 | 0 |
---
datasets:
- glue
language:
- en
license: mit
metrics:
- accuracy
tags:
- generated_from_trainer
model-index:
- name: qnli_roberta-base_125
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: GLUE QNLI
type: glue
args: qnli
metrics:
- type: accuracy
value: 0.9229361156873512
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# qnli_roberta-base_125
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the GLUE QNLI dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3457
- Accuracy: 0.9229
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0
### Training results
### Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1
- Datasets 2.6.1
- Tokenizers 0.13.1
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# qnli_roberta-base_125
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the GLUE QNLI dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3457
- Accuracy: 0.9229
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0
### Training results
### Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1
- Datasets 2.6.1
- Tokenizers 0.13.1
|
{"datasets": ["glue"], "language": ["en"], "license": "mit", "metrics": ["accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "qnli_roberta-base_125", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "GLUE QNLI", "type": "glue", "args": "qnli"}, "metrics": [{"type": "accuracy", "value": 0.9229361156873512, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,179 |
clarko/distilbert-base-uncased-finetuned-emotion
|
clarko
|
text-classification
|
[
"transformers",
"pytorch",
"tensorboard",
"distilbert",
"text-classification",
"generated_from_trainer",
"dataset:emotion",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-03-15T16:59:24Z |
2023-03-15T17:21:43+00:00
| 8 | 0 |
---
datasets:
- emotion
license: apache-2.0
metrics:
- accuracy
- f1
tags:
- generated_from_trainer
model-index:
- name: distilbert-base-uncased-finetuned-emotion
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: emotion
type: emotion
config: split
split: validation
args: split
metrics:
- type: accuracy
value: 0.9175
name: Accuracy
- type: f1
value: 0.917868093658934
name: F1
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2300
- Accuracy: 0.9175
- F1: 0.9179
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 0.8387 | 1.0 | 250 | 0.3276 | 0.9045 | 0.9016 |
| 0.2573 | 2.0 | 500 | 0.2300 | 0.9175 | 0.9179 |
### Framework versions
- Transformers 4.27.0
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2300
- Accuracy: 0.9175
- F1: 0.9179
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 0.8387 | 1.0 | 250 | 0.3276 | 0.9045 | 0.9016 |
| 0.2573 | 2.0 | 500 | 0.2300 | 0.9175 | 0.9179 |
### Framework versions
- Transformers 4.27.0
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
|
{"datasets": ["emotion"], "license": "apache-2.0", "metrics": ["accuracy", "f1"], "tags": ["generated_from_trainer"], "model-index": [{"name": "distilbert-base-uncased-finetuned-emotion", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "emotion", "type": "emotion", "config": "split", "split": "validation", "args": "split"}, "metrics": [{"type": "accuracy", "value": 0.9175, "name": "Accuracy"}, {"type": "f1", "value": 0.917868093658934, "name": "F1"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,181 |
ibm-granite/granite-3.0-3b-a800m-base
|
ibm-granite
|
text-generation
|
[
"transformers",
"safetensors",
"granitemoe",
"text-generation",
"language",
"granite-3.0",
"arxiv:0000.00000",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"region:us"
] | 2024-10-03T21:57:11Z |
2024-12-19T19:47:00+00:00
| 3,308 | 5 |
---
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
tags:
- language
- granite-3.0
inference: false
new_version: ibm-granite/granite-3.1-3b-a800m-base
model-index:
- name: granite-3.0-3b-a800m-base
results:
- task:
type: text-generation
dataset:
name: MMLU
type: human-exams
metrics:
- type: pass@1
value: 48.64
name: pass@1
- type: pass@1
value: 18.84
name: pass@1
- type: pass@1
value: 23.81
name: pass@1
- task:
type: text-generation
dataset:
name: WinoGrande
type: commonsense
metrics:
- type: pass@1
value: 65.67
name: pass@1
- type: pass@1
value: 42.2
name: pass@1
- type: pass@1
value: 47.39
name: pass@1
- type: pass@1
value: 78.29
name: pass@1
- type: pass@1
value: 72.79
name: pass@1
- type: pass@1
value: 41.34
name: pass@1
- task:
type: text-generation
dataset:
name: BoolQ
type: reading-comprehension
metrics:
- type: pass@1
value: 75.75
name: pass@1
- type: pass@1
value: 20.96
name: pass@1
- task:
type: text-generation
dataset:
name: ARC-C
type: reasoning
metrics:
- type: pass@1
value: 46.84
name: pass@1
- type: pass@1
value: 24.83
name: pass@1
- type: pass@1
value: 38.93
name: pass@1
- type: pass@1
value: 35.05
name: pass@1
- task:
type: text-generation
dataset:
name: HumanEval
type: code
metrics:
- type: pass@1
value: 26.83
name: pass@1
- type: pass@1
value: 34.6
name: pass@1
- task:
type: text-generation
dataset:
name: GSM8K
type: math
metrics:
- type: pass@1
value: 35.86
name: pass@1
- type: pass@1
value: 17.4
name: pass@1
---
<!--  -->
<!--  -->
# Granite-3.0-3B-A800M-Base
**Model Summary:**
Granite-3.0-3B-A800M-Base is a decoder-only language model to support a variety of text-to-text generation tasks. It is trained from scratch following a two-stage training strategy. In the first stage, it is trained on 8 trillion tokens sourced from diverse domains. During the second stage, it is further trained on 2 trillion tokens using a carefully curated mix of high-quality data, aiming to enhance its performance on specific tasks.
- **Developers:** Granite Team, IBM
- **GitHub Repository:** [ibm-granite/granite-3.0-language-models](https://github.com/ibm-granite/granite-3.0-language-models)
- **Website**: [Granite Docs](https://www.ibm.com/granite/docs/)
- **Paper:** [Granite 3.0 Language Models](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/paper.pdf)
- **Release Date**: October 21st, 2024
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
**Supported Languages:**
English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, and Chinese. Users may finetune Granite 3.0 models for languages beyond these 12 languages.
**Intended use:**
Prominent use cases of LLMs in text-to-text generation include summarization, text classification, extraction, question-answering, and more. All Granite Base models are able to handle these tasks as they were trained on a large amount of data from various domains. Moreover, they can serve as baseline to create specialized models for specific application scenarios.
**Generation:**
This is a simple example of how to use Granite-3.0-3B-A800M-Base model.
Install the following libraries:
```shell
pip install torch torchvision torchaudio
pip install accelerate
pip install transformers
```
Then, copy the code snippet below to run the example.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "auto"
model_path = "ibm-granite/granite-3.0-3b-a800m-base"
tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()
# change input text as desired
input_text = "Where is the Thomas J. Watson Research Center located?"
# tokenize the text
input_tokens = tokenizer(input_text, return_tensors="pt").to(device)
# generate output tokens
output = model.generate(**input_tokens,
max_length=4000)
# decode output tokens into text
output = tokenizer.batch_decode(output)
# print output
print(output)
```
**Model Architecture:**
Granite-3.0-3B-A800M-Base is based on a decoder-only sparse Mixture of Experts (MoE) transformer architecture. Core components of this architecture are: Fine-grained Experts, Dropless Token Routing, and Load Balancing Loss.
| Model | 2B Dense | 8B Dense | 1B MoE | 3B MoE |
| :-------- | :--------| :--------| :--------| :-------- |
| Embedding size | 2048 | 4096 | 1024 | **1536** |
| Number of layers | 40 | 40 | 24 | **32** |
| Attention head size | 64 | 128 | 64 | **64** |
| Number of attention heads | 32 | 32 | 16 | **24** |
| Number of KV heads | 8 | 8 | 8 | **8** |
| MLP hidden size | 8192 | 12800 | 512 | **512** |
| MLP activation | SwiGLU | SwiGLU | SwiGLU | **SwiGLU** |
| Number of Experts | — | — | 32 | **40** |
| MoE TopK | — | — | 8 | **8** |
| Initialization std | 0.1 | 0.1 | 0.1 | **0.1** |
| Sequence Length | 4096 | 4096 | 4096 | **4096** |
| Position Embedding | RoPE | RoPE | RoPE | **RoPE** |
| # Parameters | 2.5B | 8.1B | 1.3B | **3.3B** |
| # Active Parameters | 2.5B | 8.1B | 400M | **800M** |
| # Training tokens | 12T | 12T | 10T | **10T** |
**Training Data:**
This model is trained on a mix of open source and proprietary data following a two-stage training strategy.
* Stage 1 data: The data for stage 1 is sourced from diverse domains, such as: web, code, academic sources, books, and math data.
* Stage 2 data: The data for stage 2 comprises a curated mix of high-quality data from the same domains, plus multilingual and instruction data. The goal of this second training phase is to enhance the model’s performance on specific tasks.
A detailed attribution of datasets can be found in the [Granite Technical Report](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/paper.pdf) and [Accompanying Author List](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/author-ack.pdf).
**Infrastructure:**
We train Granite 3.0 Language Models using IBM's super computing cluster, Blue Vela, which is outfitted with NVIDIA H100 GPUs. This cluster provides a scalable and efficient infrastructure for training our models over thousands of GPUs while minimizing environmental impact by utilizing 100% renewable energy sources.
**Ethical Considerations and Limitations:**
The use of Large Language Models involves risks and ethical considerations people must be aware of, including but not limited to: bias and fairness, misinformation, and autonomous decision-making. Granite-3.0-3B-A800M-Base model is not the exception in this regard. Even though this model is suited for multiple generative AI tasks, it has not undergone any safety alignment, there it may produce problematic outputs. Additionally, it remains uncertain whether smaller models might exhibit increased susceptibility to hallucination in generation scenarios by copying text verbatim from the training dataset due to their reduced sizes and memorization capacities. This aspect is currently an active area of research, and we anticipate more rigorous exploration, comprehension, and mitigations in this domain. Regarding ethics, a latent risk associated with all Large Language Models is their malicious utilization. We urge the community to use Granite-3.0-3B-A800M-Base model with ethical intentions and in a responsible way.
**Resources**
- ⭐️ Learn about the latest updates with Granite: https://www.ibm.com/granite
- 📄 Get started with tutorials, best practices, and prompt engineering advice: https://www.ibm.com/granite/docs/
- 💡 Learn about the latest Granite learning resources: https://ibm.biz/granite-learning-resources
<!-- ## Citation
```
@misc{granite-models,
author = {author 1, author2, ...},
title = {},
journal = {},
volume = {},
year = {2024},
url = {https://arxiv.org/abs/0000.00000},
}
``` -->
| null |
Non_BioNLP
|
<!--  -->
<!--  -->
# Granite-3.0-3B-A800M-Base
**Model Summary:**
Granite-3.0-3B-A800M-Base is a decoder-only language model to support a variety of text-to-text generation tasks. It is trained from scratch following a two-stage training strategy. In the first stage, it is trained on 8 trillion tokens sourced from diverse domains. During the second stage, it is further trained on 2 trillion tokens using a carefully curated mix of high-quality data, aiming to enhance its performance on specific tasks.
- **Developers:** Granite Team, IBM
- **GitHub Repository:** [ibm-granite/granite-3.0-language-models](https://github.com/ibm-granite/granite-3.0-language-models)
- **Website**: [Granite Docs](https://www.ibm.com/granite/docs/)
- **Paper:** [Granite 3.0 Language Models](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/paper.pdf)
- **Release Date**: October 21st, 2024
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
**Supported Languages:**
English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, and Chinese. Users may finetune Granite 3.0 models for languages beyond these 12 languages.
**Intended use:**
Prominent use cases of LLMs in text-to-text generation include summarization, text classification, extraction, question-answering, and more. All Granite Base models are able to handle these tasks as they were trained on a large amount of data from various domains. Moreover, they can serve as baseline to create specialized models for specific application scenarios.
**Generation:**
This is a simple example of how to use Granite-3.0-3B-A800M-Base model.
Install the following libraries:
```shell
pip install torch torchvision torchaudio
pip install accelerate
pip install transformers
```
Then, copy the code snippet below to run the example.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "auto"
model_path = "ibm-granite/granite-3.0-3b-a800m-base"
tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()
# change input text as desired
input_text = "Where is the Thomas J. Watson Research Center located?"
# tokenize the text
input_tokens = tokenizer(input_text, return_tensors="pt").to(device)
# generate output tokens
output = model.generate(**input_tokens,
max_length=4000)
# decode output tokens into text
output = tokenizer.batch_decode(output)
# print output
print(output)
```
**Model Architecture:**
Granite-3.0-3B-A800M-Base is based on a decoder-only sparse Mixture of Experts (MoE) transformer architecture. Core components of this architecture are: Fine-grained Experts, Dropless Token Routing, and Load Balancing Loss.
| Model | 2B Dense | 8B Dense | 1B MoE | 3B MoE |
| :-------- | :--------| :--------| :--------| :-------- |
| Embedding size | 2048 | 4096 | 1024 | **1536** |
| Number of layers | 40 | 40 | 24 | **32** |
| Attention head size | 64 | 128 | 64 | **64** |
| Number of attention heads | 32 | 32 | 16 | **24** |
| Number of KV heads | 8 | 8 | 8 | **8** |
| MLP hidden size | 8192 | 12800 | 512 | **512** |
| MLP activation | SwiGLU | SwiGLU | SwiGLU | **SwiGLU** |
| Number of Experts | — | — | 32 | **40** |
| MoE TopK | — | — | 8 | **8** |
| Initialization std | 0.1 | 0.1 | 0.1 | **0.1** |
| Sequence Length | 4096 | 4096 | 4096 | **4096** |
| Position Embedding | RoPE | RoPE | RoPE | **RoPE** |
| # Parameters | 2.5B | 8.1B | 1.3B | **3.3B** |
| # Active Parameters | 2.5B | 8.1B | 400M | **800M** |
| # Training tokens | 12T | 12T | 10T | **10T** |
**Training Data:**
This model is trained on a mix of open source and proprietary data following a two-stage training strategy.
* Stage 1 data: The data for stage 1 is sourced from diverse domains, such as: web, code, academic sources, books, and math data.
* Stage 2 data: The data for stage 2 comprises a curated mix of high-quality data from the same domains, plus multilingual and instruction data. The goal of this second training phase is to enhance the model’s performance on specific tasks.
A detailed attribution of datasets can be found in the [Granite Technical Report](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/paper.pdf) and [Accompanying Author List](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/author-ack.pdf).
**Infrastructure:**
We train Granite 3.0 Language Models using IBM's super computing cluster, Blue Vela, which is outfitted with NVIDIA H100 GPUs. This cluster provides a scalable and efficient infrastructure for training our models over thousands of GPUs while minimizing environmental impact by utilizing 100% renewable energy sources.
**Ethical Considerations and Limitations:**
The use of Large Language Models involves risks and ethical considerations people must be aware of, including but not limited to: bias and fairness, misinformation, and autonomous decision-making. Granite-3.0-3B-A800M-Base model is not the exception in this regard. Even though this model is suited for multiple generative AI tasks, it has not undergone any safety alignment, there it may produce problematic outputs. Additionally, it remains uncertain whether smaller models might exhibit increased susceptibility to hallucination in generation scenarios by copying text verbatim from the training dataset due to their reduced sizes and memorization capacities. This aspect is currently an active area of research, and we anticipate more rigorous exploration, comprehension, and mitigations in this domain. Regarding ethics, a latent risk associated with all Large Language Models is their malicious utilization. We urge the community to use Granite-3.0-3B-A800M-Base model with ethical intentions and in a responsible way.
**Resources**
- ⭐️ Learn about the latest updates with Granite: https://www.ibm.com/granite
- 📄 Get started with tutorials, best practices, and prompt engineering advice: https://www.ibm.com/granite/docs/
- 💡 Learn about the latest Granite learning resources: https://ibm.biz/granite-learning-resources
<!-- ## Citation
```
@misc{granite-models,
author = {author 1, author2, ...},
title = {},
journal = {},
volume = {},
year = {2024},
url = {https://arxiv.org/abs/0000.00000},
}
``` -->
|
{"library_name": "transformers", "license": "apache-2.0", "pipeline_tag": "text-generation", "tags": ["language", "granite-3.0"], "inference": false, "new_version": "ibm-granite/granite-3.1-3b-a800m-base", "model-index": [{"name": "granite-3.0-3b-a800m-base", "results": [{"task": {"type": "text-generation"}, "dataset": {"name": "MMLU", "type": "human-exams"}, "metrics": [{"type": "pass@1", "value": 48.64, "name": "pass@1"}, {"type": "pass@1", "value": 18.84, "name": "pass@1"}, {"type": "pass@1", "value": 23.81, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "WinoGrande", "type": "commonsense"}, "metrics": [{"type": "pass@1", "value": 65.67, "name": "pass@1"}, {"type": "pass@1", "value": 42.2, "name": "pass@1"}, {"type": "pass@1", "value": 47.39, "name": "pass@1"}, {"type": "pass@1", "value": 78.29, "name": "pass@1"}, {"type": "pass@1", "value": 72.79, "name": "pass@1"}, {"type": "pass@1", "value": 41.34, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "BoolQ", "type": "reading-comprehension"}, "metrics": [{"type": "pass@1", "value": 75.75, "name": "pass@1"}, {"type": "pass@1", "value": 20.96, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "ARC-C", "type": "reasoning"}, "metrics": [{"type": "pass@1", "value": 46.84, "name": "pass@1"}, {"type": "pass@1", "value": 24.83, "name": "pass@1"}, {"type": "pass@1", "value": 38.93, "name": "pass@1"}, {"type": "pass@1", "value": 35.05, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "HumanEval", "type": "code"}, "metrics": [{"type": "pass@1", "value": 26.83, "name": "pass@1"}, {"type": "pass@1", "value": 34.6, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "GSM8K", "type": "math"}, "metrics": [{"type": "pass@1", "value": 35.86, "name": "pass@1"}, {"type": "pass@1", "value": 17.4, "name": "pass@1"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION",
"SUMMARIZATION"
] | 45,182 |
Hozumi0801/distilbert-base-uncased-finetuned-emotion
|
Hozumi0801
|
text-classification
|
[
"transformers",
"tensorboard",
"safetensors",
"distilbert",
"text-classification",
"generated_from_trainer",
"dataset:emotion",
"base_model:distilbert/distilbert-base-uncased",
"base_model:finetune:distilbert/distilbert-base-uncased",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-04-23T02:02:08Z |
2024-04-23T04:31:22+00:00
| 4 | 0 |
---
base_model: distilbert-base-uncased
datasets:
- emotion
license: apache-2.0
metrics:
- accuracy
- f1
tags:
- generated_from_trainer
model-index:
- name: distilbert-base-uncased-finetuned-emotion
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: emotion
type: emotion
config: split
split: validation
args: split
metrics:
- type: accuracy
value: 0.871
name: Accuracy
- type: f1
value: 0.8601997169781247
name: F1
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4557
- Accuracy: 0.871
- F1: 0.8602
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 384
- eval_batch_size: 384
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| No log | 1.0 | 42 | 0.5682 | 0.8155 | 0.7840 |
| No log | 2.0 | 84 | 0.4557 | 0.871 | 0.8602 |
### Framework versions
- Transformers 4.39.3
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4557
- Accuracy: 0.871
- F1: 0.8602
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 384
- eval_batch_size: 384
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| No log | 1.0 | 42 | 0.5682 | 0.8155 | 0.7840 |
| No log | 2.0 | 84 | 0.4557 | 0.871 | 0.8602 |
### Framework versions
- Transformers 4.39.3
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
|
{"base_model": "distilbert-base-uncased", "datasets": ["emotion"], "license": "apache-2.0", "metrics": ["accuracy", "f1"], "tags": ["generated_from_trainer"], "model-index": [{"name": "distilbert-base-uncased-finetuned-emotion", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "emotion", "type": "emotion", "config": "split", "split": "validation", "args": "split"}, "metrics": [{"type": "accuracy", "value": 0.871, "name": "Accuracy"}, {"type": "f1", "value": 0.8601997169781247, "name": "F1"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,183 |
YakovElm/IntelDAOS20SetFitModel_clean_data
|
YakovElm
|
text-classification
|
[
"sentence-transformers",
"pytorch",
"mpnet",
"setfit",
"text-classification",
"arxiv:2209.11055",
"license:apache-2.0",
"region:us"
] | 2023-05-24T03:20:00Z |
2023-05-24T03:20:40+00:00
| 8 | 0 |
---
license: apache-2.0
pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
---
# YakovElm/IntelDAOS20SetFitModel_clean_data
This is a [SetFit model](https://github.com/huggingface/setfit) that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves:
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.
## Usage
To use this model for inference, first install the SetFit library:
```bash
python -m pip install setfit
```
You can then run inference as follows:
```python
from setfit import SetFitModel
# Download from Hub and run inference
model = SetFitModel.from_pretrained("YakovElm/IntelDAOS20SetFitModel_clean_data")
# Run inference
preds = model(["i loved the spiderman movie!", "pineapple on pizza is the worst 🤮"])
```
## BibTeX entry and citation info
```bibtex
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}
```
| null |
Non_BioNLP
|
# YakovElm/IntelDAOS20SetFitModel_clean_data
This is a [SetFit model](https://github.com/huggingface/setfit) that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves:
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.
## Usage
To use this model for inference, first install the SetFit library:
```bash
python -m pip install setfit
```
You can then run inference as follows:
```python
from setfit import SetFitModel
# Download from Hub and run inference
model = SetFitModel.from_pretrained("YakovElm/IntelDAOS20SetFitModel_clean_data")
# Run inference
preds = model(["i loved the spiderman movie!", "pineapple on pizza is the worst 🤮"])
```
## BibTeX entry and citation info
```bibtex
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}
```
|
{"license": "apache-2.0", "pipeline_tag": "text-classification", "tags": ["setfit", "sentence-transformers", "text-classification"]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,184 |
Lots-of-LoRAs/Mistral-7B-Instruct-v0.2-4b-r16-task1226
|
Lots-of-LoRAs
| null |
[
"pytorch",
"safetensors",
"en",
"arxiv:1910.09700",
"arxiv:2407.00066",
"license:mit",
"region:us"
] | 2024-12-30T23:09:56Z |
2024-12-30T23:10:01+00:00
| 0 | 0 |
---
language: en
library_name: pytorch
license: mit
---
# Model Card for Mistral-7B-Instruct-v0.2-4b-r16-task1226
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
LoRA trained on task1226_ted_translation_es_en
- **Developed by:** bruel
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** LoRA
- **Language(s) (NLP):** en
- **License:** mit
- **Finetuned from model [optional]:** mistralai/Mistral-7B-Instruct-v0.2
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** https://github.com/bruel-gabrielsson
- **Paper [optional]:** "Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead" (2024), Rickard Brüel Gabrielsson, Jiacheng Zhu, Onkar Bhardwaj, Leshem Choshen, Kristjan Greenewald, Mikhail Yurochkin and Justin Solomon
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
https://huggingface.co/datasets/Lots-of-LoRAs/task1226_ted_translation_es_en sourced from https://github.com/allenai/natural-instructions
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
@misc{brüelgabrielsson2024compressserveservingthousands,
title={Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead},
author={Rickard Brüel-Gabrielsson and Jiacheng Zhu and Onkar Bhardwaj and Leshem Choshen and Kristjan Greenewald and Mikhail Yurochkin and Justin Solomon},
year={2024},
eprint={2407.00066},
archivePrefix={arXiv},
primaryClass={cs.DC},
url={https://arxiv.org/abs/2407.00066},
}
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
| null |
Non_BioNLP
|
# Model Card for Mistral-7B-Instruct-v0.2-4b-r16-task1226
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
LoRA trained on task1226_ted_translation_es_en
- **Developed by:** bruel
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** LoRA
- **Language(s) (NLP):** en
- **License:** mit
- **Finetuned from model [optional]:** mistralai/Mistral-7B-Instruct-v0.2
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** https://github.com/bruel-gabrielsson
- **Paper [optional]:** "Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead" (2024), Rickard Brüel Gabrielsson, Jiacheng Zhu, Onkar Bhardwaj, Leshem Choshen, Kristjan Greenewald, Mikhail Yurochkin and Justin Solomon
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
https://huggingface.co/datasets/Lots-of-LoRAs/task1226_ted_translation_es_en sourced from https://github.com/allenai/natural-instructions
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
@misc{brüelgabrielsson2024compressserveservingthousands,
title={Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead},
author={Rickard Brüel-Gabrielsson and Jiacheng Zhu and Onkar Bhardwaj and Leshem Choshen and Kristjan Greenewald and Mikhail Yurochkin and Justin Solomon},
year={2024},
eprint={2407.00066},
archivePrefix={arXiv},
primaryClass={cs.DC},
url={https://arxiv.org/abs/2407.00066},
}
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
{"language": "en", "library_name": "pytorch", "license": "mit"}
|
task
|
[
"TRANSLATION"
] | 45,185 |
jngan/bart-large-cnn-samsum
|
jngan
|
summarization
|
[
"transformers",
"safetensors",
"bart",
"text2text-generation",
"summarization",
"en",
"dataset:Samsung/samsum",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-07-16T21:02:33Z |
2024-07-19T11:40:21+00:00
| 291 | 0 |
---
datasets:
- Samsung/samsum
language:
- en
license: mit
pipeline_tag: summarization
---
English conversation summarization model
Pre-trained from: BART-large-cnn
Implenmentation and finetuning script can be found here:
[https://github.com/joceyngan/conversation_summarization](https://github.com/joceyngan/conversation_summarization)
## Eval Results:
Reference summary is the ground truth from the dataset and generated summary is generated by this model.
### Sample 1:
#### Original Text:
A: Hi Tom, are you busy tomorrow’s afternoon?
B: I’m pretty sure I am. What’s up?
A: Can you go with me to the animal shelter?.
B: What do you want to do?
A: I want to get a puppy for my son.
B: That will make him so happy.
A: Yeah, we’ve discussed it many times. I think he’s ready now.
B: That’s good. Raising a dog is a tough issue. Like having a baby ;-)
A: I'll get him one of those little dogs.
B: One that won't grow up too big;-)
A: And eat too much;-))
B: Do you know which one he would like?
A: Oh, yes, I took him there last Monday. He showed me one that he really liked.
B: I bet you had to drag him away.
A: He wanted to take it home right away ;-).
B: I wonder what he'll name it.
A: He said he’d name it after his dead hamster – Lemmy - he's a great Motorhead fan :-)))
#### Reference Summary:
A will go to the animal shelter tomorrow to get a puppy for her son. They already visited the shelter last Monday and the son chose the puppy.
#### Generated Summary:
A wants to get a puppy for his son. B will go with him to the animal shelter tomorrow. A will get him one of those little dogs. A took him there last Monday and he liked it a lot. A wants to name it Lemmy.
### Sample 2:
#### Original Text:
Emma: I’ve just fallen in love with this advent calendar! Awesome! I wanna one for my kids!
Rob: I used to get one every year as a child! Loved them!
Emma: Yeah, i remember! they were filled with chocolates!
Lauren: they are different these days! much more sophisticated! Haha!
Rob: yeah, they can be fabric/ wooden, shop bought/ homemade, filled with various stuff
Emma: what do you fit inside?
Lauren: small toys, Christmas decorations, creative stuff, hair bands & clips, stickers, pencils & rubbers, small puzzles, sweets
Emma: WOW! That’s brill! X
Lauren: i add one more very special thing as well- little notes asking my children to do something nice for someone else
Rob: i like that! My sister adds notes asking her kids questions about christmas such as What did the 3 wise men bring? etc
Lauren: i reckon it prepares them for Christmas
Emma: and makes it more about traditions and being kind to other people
Lauren: my children get very excited every time they get one!
Emma: i can see why! :)
#### Reference Summary:
Emma and Rob love the advent calendar. Lauren fits inside calendar various items, for instance, small toys and Christmas decorations. Her children are excited whenever they get the calendar.
#### Generated Summary:
Emma loves the advent calendar. Rob used to get one every year as a child. Emma would like to buy one for her kids. Rob's sister asks her kids questions about Christmas. Lauren's children get very excited every time they get one.
### Sample 3:
#### Original Text:
Jackie: Madison is pregnant
Jackie: but she doesn't wanna talk about it
Iggy: why
Jackie: I don't know why because she doesn't wanna talk about it
Iggy: ok
Jackie: I wanted to prepare you for it because people get super excited and ask lots of questions
Jackie: and she looked way more anxious than excited
Iggy: she's probably worrying about it
Iggy: she's taking every commitment really seriously
Jackie: it could be money problems or relationship problems
Iggy: or maybe she wants an abortion
Jackie: it could be all of the above
Iggy: but you know what?
Iggy: once my friend was pregnant and I couldn't bring myself to be happy about it
Jackie: why?
Iggy: I felt they were immature and I couldn't picture this couple as parents
Jackie: I felt similar way on Patricia's wedding
Iggy: Patricia Stevens?
Jackie: yes
Iggy: so we're talking about the same person
Jackie: what a coincidence
Jackie: so she's pregnant?
Iggy: she thought she was
Jackie: damn...
#### Reference Summary:
Madison is pregnant but she doesn't want to talk about it. Patricia Stevens got married and she thought she was pregnant.
#### Generated Summary:
Madison is pregnant. She doesn't want to talk about it. Iggy's friend Patricia Stevens was pregnant with Patricia Stevens and she didn't like it much. She felt they were immature and she couldn't picture them as parents.
### Sample 4:
#### Original Text:
Marla: <file_photo>
Marla: look what I found under my bed
Kiki: lol
Tamara: is that someone's underwear?
Marla: it certainly isn't mine, my ass is big but it isn't huge
Kiki: it looks like male underwear
Tamara: not necessarily, maybe some butch had fun in your room while you were gone
Marla: ok but how can you leave your underwear after hooking up? wtf is wrong with people
Kiki: she or he could be too wasted to notice
Tamara: or maybe someone put their pants there to piss you off
Marla: that makes no sense
Marla: it's so fucking childish
Kiki: if it's childish then it must have been your sister's idea
Marla: she's 13, she doesn't have underwear that isn't pink
Tamara: maybe it belonged to one of your exes?
Kiki: she would have recognized it
Marla: lol we're doing total CSI investigation on one pair of boxers :D
Kiki: <file_gif>
Tamara: lol
Tamara: I think your sister convinced someone to put their underwear in your room as a dare
Marla: sounds legit
Kiki: Tamara, you just cracked the case!
Tamara: <file_gif>
Tamara: always happy to help
#### Reference Summary:
Marla found a pair of boxers under her bed.
#### Generated Summary:
Marla found a pair of boxers under her bed. Kiki, Tamara, Marla and Tamara are laughing at the fact that someone left their underwear under Marla's bed after hooking up. Marla is convinced that her sister convinced someone to put their underwear in her room.
### Sample 5:
#### Original Text:
Robert: Hey give me the address of this music shop you mentioned before
Robert: I have to buy guitar cable
Fred: <file_other>
Fred: Catch it on google maps
Robert: thx m8
Fred: ur welcome
#### Reference Summary:
Robert wants Fred to send him the address of the music shop as he needs to buy guitar cable.
#### Generated Summary:
Fred gives Robert the address of the music shop where he needs to buy guitar cable. Robert can find it on google maps. Robert has to buy a guitar cable at the shop. Fred sends him a link to the address.
## Finetuning Arguments:
```bash
{
output_dir=str(results_path),
evaluation_strategy="epoch",
save_strategy="epoch",
logging_steps=10,
learning_rate=2e-5,
per_device_train_batch_size=2,
per_device_eval_batch_size=2,
num_train_epochs=3,
weight_decay=0.01,
report_to="tensorboard",
save_total_limit=3,
load_best_model_at_end=True,
metric_for_best_model="eval_loss",
greater_is_better=False,
logging_dir=str(results_path),
}
```
## Usage:
```bash
tokenizer = BartTokenizer.from_pretrained('facebook/bart-large-cnn')
model = BartForConditionalGeneration.from_pretrained('jngan/bart-large-cnn-samsum')
```
| null |
Non_BioNLP
|
English conversation summarization model
Pre-trained from: BART-large-cnn
Implenmentation and finetuning script can be found here:
[https://github.com/joceyngan/conversation_summarization](https://github.com/joceyngan/conversation_summarization)
## Eval Results:
Reference summary is the ground truth from the dataset and generated summary is generated by this model.
### Sample 1:
#### Original Text:
A: Hi Tom, are you busy tomorrow’s afternoon?
B: I’m pretty sure I am. What’s up?
A: Can you go with me to the animal shelter?.
B: What do you want to do?
A: I want to get a puppy for my son.
B: That will make him so happy.
A: Yeah, we’ve discussed it many times. I think he’s ready now.
B: That’s good. Raising a dog is a tough issue. Like having a baby ;-)
A: I'll get him one of those little dogs.
B: One that won't grow up too big;-)
A: And eat too much;-))
B: Do you know which one he would like?
A: Oh, yes, I took him there last Monday. He showed me one that he really liked.
B: I bet you had to drag him away.
A: He wanted to take it home right away ;-).
B: I wonder what he'll name it.
A: He said he’d name it after his dead hamster – Lemmy - he's a great Motorhead fan :-)))
#### Reference Summary:
A will go to the animal shelter tomorrow to get a puppy for her son. They already visited the shelter last Monday and the son chose the puppy.
#### Generated Summary:
A wants to get a puppy for his son. B will go with him to the animal shelter tomorrow. A will get him one of those little dogs. A took him there last Monday and he liked it a lot. A wants to name it Lemmy.
### Sample 2:
#### Original Text:
Emma: I’ve just fallen in love with this advent calendar! Awesome! I wanna one for my kids!
Rob: I used to get one every year as a child! Loved them!
Emma: Yeah, i remember! they were filled with chocolates!
Lauren: they are different these days! much more sophisticated! Haha!
Rob: yeah, they can be fabric/ wooden, shop bought/ homemade, filled with various stuff
Emma: what do you fit inside?
Lauren: small toys, Christmas decorations, creative stuff, hair bands & clips, stickers, pencils & rubbers, small puzzles, sweets
Emma: WOW! That’s brill! X
Lauren: i add one more very special thing as well- little notes asking my children to do something nice for someone else
Rob: i like that! My sister adds notes asking her kids questions about christmas such as What did the 3 wise men bring? etc
Lauren: i reckon it prepares them for Christmas
Emma: and makes it more about traditions and being kind to other people
Lauren: my children get very excited every time they get one!
Emma: i can see why! :)
#### Reference Summary:
Emma and Rob love the advent calendar. Lauren fits inside calendar various items, for instance, small toys and Christmas decorations. Her children are excited whenever they get the calendar.
#### Generated Summary:
Emma loves the advent calendar. Rob used to get one every year as a child. Emma would like to buy one for her kids. Rob's sister asks her kids questions about Christmas. Lauren's children get very excited every time they get one.
### Sample 3:
#### Original Text:
Jackie: Madison is pregnant
Jackie: but she doesn't wanna talk about it
Iggy: why
Jackie: I don't know why because she doesn't wanna talk about it
Iggy: ok
Jackie: I wanted to prepare you for it because people get super excited and ask lots of questions
Jackie: and she looked way more anxious than excited
Iggy: she's probably worrying about it
Iggy: she's taking every commitment really seriously
Jackie: it could be money problems or relationship problems
Iggy: or maybe she wants an abortion
Jackie: it could be all of the above
Iggy: but you know what?
Iggy: once my friend was pregnant and I couldn't bring myself to be happy about it
Jackie: why?
Iggy: I felt they were immature and I couldn't picture this couple as parents
Jackie: I felt similar way on Patricia's wedding
Iggy: Patricia Stevens?
Jackie: yes
Iggy: so we're talking about the same person
Jackie: what a coincidence
Jackie: so she's pregnant?
Iggy: she thought she was
Jackie: damn...
#### Reference Summary:
Madison is pregnant but she doesn't want to talk about it. Patricia Stevens got married and she thought she was pregnant.
#### Generated Summary:
Madison is pregnant. She doesn't want to talk about it. Iggy's friend Patricia Stevens was pregnant with Patricia Stevens and she didn't like it much. She felt they were immature and she couldn't picture them as parents.
### Sample 4:
#### Original Text:
Marla: <file_photo>
Marla: look what I found under my bed
Kiki: lol
Tamara: is that someone's underwear?
Marla: it certainly isn't mine, my ass is big but it isn't huge
Kiki: it looks like male underwear
Tamara: not necessarily, maybe some butch had fun in your room while you were gone
Marla: ok but how can you leave your underwear after hooking up? wtf is wrong with people
Kiki: she or he could be too wasted to notice
Tamara: or maybe someone put their pants there to piss you off
Marla: that makes no sense
Marla: it's so fucking childish
Kiki: if it's childish then it must have been your sister's idea
Marla: she's 13, she doesn't have underwear that isn't pink
Tamara: maybe it belonged to one of your exes?
Kiki: she would have recognized it
Marla: lol we're doing total CSI investigation on one pair of boxers :D
Kiki: <file_gif>
Tamara: lol
Tamara: I think your sister convinced someone to put their underwear in your room as a dare
Marla: sounds legit
Kiki: Tamara, you just cracked the case!
Tamara: <file_gif>
Tamara: always happy to help
#### Reference Summary:
Marla found a pair of boxers under her bed.
#### Generated Summary:
Marla found a pair of boxers under her bed. Kiki, Tamara, Marla and Tamara are laughing at the fact that someone left their underwear under Marla's bed after hooking up. Marla is convinced that her sister convinced someone to put their underwear in her room.
### Sample 5:
#### Original Text:
Robert: Hey give me the address of this music shop you mentioned before
Robert: I have to buy guitar cable
Fred: <file_other>
Fred: Catch it on google maps
Robert: thx m8
Fred: ur welcome
#### Reference Summary:
Robert wants Fred to send him the address of the music shop as he needs to buy guitar cable.
#### Generated Summary:
Fred gives Robert the address of the music shop where he needs to buy guitar cable. Robert can find it on google maps. Robert has to buy a guitar cable at the shop. Fred sends him a link to the address.
## Finetuning Arguments:
```bash
{
output_dir=str(results_path),
evaluation_strategy="epoch",
save_strategy="epoch",
logging_steps=10,
learning_rate=2e-5,
per_device_train_batch_size=2,
per_device_eval_batch_size=2,
num_train_epochs=3,
weight_decay=0.01,
report_to="tensorboard",
save_total_limit=3,
load_best_model_at_end=True,
metric_for_best_model="eval_loss",
greater_is_better=False,
logging_dir=str(results_path),
}
```
## Usage:
```bash
tokenizer = BartTokenizer.from_pretrained('facebook/bart-large-cnn')
model = BartForConditionalGeneration.from_pretrained('jngan/bart-large-cnn-samsum')
```
|
{"datasets": ["Samsung/samsum"], "language": ["en"], "license": "mit", "pipeline_tag": "summarization"}
|
task
|
[
"SUMMARIZATION"
] | 45,186 |
mayitbe/3e5_5_16_spark
|
mayitbe
|
sentence-similarity
|
[
"sentence-transformers",
"tensorboard",
"safetensors",
"xlm-roberta",
"sentence-similarity",
"feature-extraction",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2024-07-14T00:27:42Z |
2024-07-14T00:42:35+00:00
| 4 | 0 |
---
datasets: []
language: []
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
widget: []
---
# SentenceTransformer
This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
<!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
- **Maximum Sequence Length:** 8192 tokens
- **Output Dimensionality:** 1024 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
'The weather is lovely today.',
"It's so sunny outside!",
'He drove to the stadium.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.0.1
- Transformers: 4.41.2
- PyTorch: 2.3.0+cu121
- Accelerate: 0.32.1
- Datasets: 2.20.0
- Tokenizers: 0.19.1
## Citation
### BibTeX
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# SentenceTransformer
This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
<!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
- **Maximum Sequence Length:** 8192 tokens
- **Output Dimensionality:** 1024 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
'The weather is lovely today.',
"It's so sunny outside!",
'He drove to the stadium.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.0.1
- Transformers: 4.41.2
- PyTorch: 2.3.0+cu121
- Accelerate: 0.32.1
- Datasets: 2.20.0
- Tokenizers: 0.19.1
## Citation
### BibTeX
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"datasets": [], "language": [], "library_name": "sentence-transformers", "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction"], "widget": []}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,187 |
zhibinlu/vgcn-bert-distilbert-base-uncased
|
zhibinlu
|
feature-extraction
|
[
"transformers",
"safetensors",
"vgcn-bert",
"feature-extraction",
"exbert",
"custom_code",
"en",
"dataset:bookcorpus",
"dataset:wikipedia",
"arxiv:2004.05707",
"license:apache-2.0",
"region:us"
] | 2023-06-09T16:24:13Z |
2024-06-23T14:06:11+00:00
| 40 | 1 |
---
datasets:
- bookcorpus
- wikipedia
language: en
license: apache-2.0
tags:
- exbert
---
# VGCN-BERT (DistilBERT based, uncased)
This model is a VGCN-BERT model based on [DistilBert-base-uncased](https://huggingface.co/distilbert-base-uncased) version. The original paper is [VGCN-BERT](https://arxiv.org/abs/2004.05707).
> Much progress has been made recently on text classification with methods based on neural networks. In particular, models using attention mechanism such as BERT have shown to have the capability of capturing the contextual information within a sentence or document. However, their ability of capturing the global information about the vocabulary of a language is more limited. This latter is the strength of Graph Convolutional Networks (GCN). In this paper, we propose VGCN-BERT model which combines the capability of BERT with a Vocabulary Graph Convolutional Network (VGCN). Local information and global information interact through different layers of BERT, allowing them to influence mutually and to build together a final representation for classification. In our experiments on several text classification datasets, our approach outperforms BERT and GCN alone, and achieve higher effectiveness than that reported in previous studies.
The original implementation is in my GitHub [vgcn-bert repo](https://github.com/Louis-udm/VGCN-BERT), but recently I updated the algorithm and implemented this new version for integrating in HuggingFace Transformers, the new version has the following improvements:
- Greatly speeds up the calculation speed of embedding vocabulary graph convolutinal network (or Word Graph embedding). Taking CoLa as an example, the new model only increases the training time by 11% compared with the base model
- Updated subgraph selection algorithm.
- Currently using DistilBert as the base model, but it is easy to migrate to other models.
- Provide two graph construction methods in vgcn_bert/modeling_graph.py (the same NPMI statistical method as the paper, and the predefined entity-relationship mapping method)
I hope that after integrating into transformers, someone can discover some more practical use case and share to me. I am ashamed to say that I have not discovered too much real use cases myself, mainly because the word-grounded graph obtained through statistical methods has limited improvement on the LLM model. I think its potential application should be when there are specific/customized graphs that need to be integrated into LLM.
## How to use
```python
import transformers as tfr
# Use DistilBert tokenizer, that is corresponding to the base model of this version
tokenizer = tfr.AutoTokenizer.from_pretrained("distilbert-base-uncased")
# Load VGCN-BERT model
model = tfr.AutoModel.from_pretrained(
"zhibinlu/vgcn-bert-distilbert-base-uncased", trust_remote_code=True,
# # if you already have WordGraphs (torch sparse) and their id_maps,
# # you can directly instantiate VGCN-BERT model with your WGraphs (support multiple graphs)
# wgraphs=wgraph_list,
# wgraph_id_to_tokenizer_id_maps=id_map_list
)
# Generator WGraph symmetric adjacency matrix
# 1st method: Build graph using NPMI statistical method from training corpus
# wgraph, wgraph_id_to_tokenizer_id_map = model.wgraph_builder(rows=train_valid_df["text"], tokenizer=tokenizer)
# 2nd method: Build graph from pre-defined entity relationship tuple with weight
entity_relations = [
("dog", "labrador", 0.6),
("cat", "garfield", 0.7),
("city", "montreal", 0.8),
("weather", "rain", 0.3),
]
wgraph, wgraph_id_to_tokenizer_id_map = model.wgraph_builder(rows=entity_relations, tokenizer=tokenizer)
# Add WordGraphs to the model
model.set_wgraphs([wgraph], [wgraph_id_to_tokenizer_id_map])
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors="pt")
output = model(**encoded_input)
```
## Fine-tune model
It's better fin-tune vgcn-bert model for the specific tasks.
## Citation
If you make use of this code or the VGCN-BERT approach in your work, please cite the following paper:
@inproceedings{ZhibinluGraphEmbedding,
author = {Zhibin Lu and Pan Du and Jian-Yun Nie},
title = {VGCN-BERT: Augmenting BERT with Graph Embedding for Text Classification},
booktitle = {Advances in Information Retrieval - 42nd European Conference on {IR}
Research, {ECIR} 2020, Lisbon, Portugal, April 14-17, 2020, Proceedings,
Part {I}},
series = {Lecture Notes in Computer Science},
volume = {12035},
pages = {369--382},
publisher = {Springer},
year = {2020},
}
## License
VGCN-BERT is made available under the Apache 2.0 license.
## Contact
- [Zhibin.Lu]([email protected])
- [louis-udm in GitHub](https://github.com/Louis-udm)
| null |
Non_BioNLP
|
# VGCN-BERT (DistilBERT based, uncased)
This model is a VGCN-BERT model based on [DistilBert-base-uncased](https://huggingface.co/distilbert-base-uncased) version. The original paper is [VGCN-BERT](https://arxiv.org/abs/2004.05707).
> Much progress has been made recently on text classification with methods based on neural networks. In particular, models using attention mechanism such as BERT have shown to have the capability of capturing the contextual information within a sentence or document. However, their ability of capturing the global information about the vocabulary of a language is more limited. This latter is the strength of Graph Convolutional Networks (GCN). In this paper, we propose VGCN-BERT model which combines the capability of BERT with a Vocabulary Graph Convolutional Network (VGCN). Local information and global information interact through different layers of BERT, allowing them to influence mutually and to build together a final representation for classification. In our experiments on several text classification datasets, our approach outperforms BERT and GCN alone, and achieve higher effectiveness than that reported in previous studies.
The original implementation is in my GitHub [vgcn-bert repo](https://github.com/Louis-udm/VGCN-BERT), but recently I updated the algorithm and implemented this new version for integrating in HuggingFace Transformers, the new version has the following improvements:
- Greatly speeds up the calculation speed of embedding vocabulary graph convolutinal network (or Word Graph embedding). Taking CoLa as an example, the new model only increases the training time by 11% compared with the base model
- Updated subgraph selection algorithm.
- Currently using DistilBert as the base model, but it is easy to migrate to other models.
- Provide two graph construction methods in vgcn_bert/modeling_graph.py (the same NPMI statistical method as the paper, and the predefined entity-relationship mapping method)
I hope that after integrating into transformers, someone can discover some more practical use case and share to me. I am ashamed to say that I have not discovered too much real use cases myself, mainly because the word-grounded graph obtained through statistical methods has limited improvement on the LLM model. I think its potential application should be when there are specific/customized graphs that need to be integrated into LLM.
## How to use
```python
import transformers as tfr
# Use DistilBert tokenizer, that is corresponding to the base model of this version
tokenizer = tfr.AutoTokenizer.from_pretrained("distilbert-base-uncased")
# Load VGCN-BERT model
model = tfr.AutoModel.from_pretrained(
"zhibinlu/vgcn-bert-distilbert-base-uncased", trust_remote_code=True,
# # if you already have WordGraphs (torch sparse) and their id_maps,
# # you can directly instantiate VGCN-BERT model with your WGraphs (support multiple graphs)
# wgraphs=wgraph_list,
# wgraph_id_to_tokenizer_id_maps=id_map_list
)
# Generator WGraph symmetric adjacency matrix
# 1st method: Build graph using NPMI statistical method from training corpus
# wgraph, wgraph_id_to_tokenizer_id_map = model.wgraph_builder(rows=train_valid_df["text"], tokenizer=tokenizer)
# 2nd method: Build graph from pre-defined entity relationship tuple with weight
entity_relations = [
("dog", "labrador", 0.6),
("cat", "garfield", 0.7),
("city", "montreal", 0.8),
("weather", "rain", 0.3),
]
wgraph, wgraph_id_to_tokenizer_id_map = model.wgraph_builder(rows=entity_relations, tokenizer=tokenizer)
# Add WordGraphs to the model
model.set_wgraphs([wgraph], [wgraph_id_to_tokenizer_id_map])
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors="pt")
output = model(**encoded_input)
```
## Fine-tune model
It's better fin-tune vgcn-bert model for the specific tasks.
## Citation
If you make use of this code or the VGCN-BERT approach in your work, please cite the following paper:
@inproceedings{ZhibinluGraphEmbedding,
author = {Zhibin Lu and Pan Du and Jian-Yun Nie},
title = {VGCN-BERT: Augmenting BERT with Graph Embedding for Text Classification},
booktitle = {Advances in Information Retrieval - 42nd European Conference on {IR}
Research, {ECIR} 2020, Lisbon, Portugal, April 14-17, 2020, Proceedings,
Part {I}},
series = {Lecture Notes in Computer Science},
volume = {12035},
pages = {369--382},
publisher = {Springer},
year = {2020},
}
## License
VGCN-BERT is made available under the Apache 2.0 license.
## Contact
- [Zhibin.Lu]([email protected])
- [louis-udm in GitHub](https://github.com/Louis-udm)
|
{"datasets": ["bookcorpus", "wikipedia"], "language": "en", "license": "apache-2.0", "tags": ["exbert"]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,188 |
zwellington/microtest-2.0
|
zwellington
|
text-classification
|
[
"transformers",
"pytorch",
"bert",
"text-classification",
"generated_from_trainer",
"dataset:azaheadhealth",
"base_model:google-bert/bert-base-uncased",
"base_model:finetune:google-bert/bert-base-uncased",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-02-07T00:40:09Z |
2024-02-07T00:41:23+00:00
| 16 | 0 |
---
base_model: bert-base-uncased
datasets:
- azaheadhealth
license: apache-2.0
metrics:
- accuracy
- f1
- precision
- recall
tags:
- generated_from_trainer
model-index:
- name: microtest-2.0
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: azaheadhealth
type: azaheadhealth
config: micro
split: test
args: micro
metrics:
- type: accuracy
value: 0.75
name: Accuracy
- type: f1
value: 0.8
name: F1
- type: precision
value: 0.6666666666666666
name: Precision
- type: recall
value: 1.0
name: Recall
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# microtest-2.0
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the azaheadhealth dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3672
- Accuracy: 0.75
- F1: 0.8
- Precision: 0.6667
- Recall: 1.0
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 2
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---:|:---------:|:------:|
| 0.8113 | 0.5 | 1 | 0.4486 | 0.75 | 0.8 | 0.6667 | 1.0 |
| 0.7227 | 1.0 | 2 | 0.3672 | 0.75 | 0.8 | 0.6667 | 1.0 |
### Framework versions
- Transformers 4.31.0
- Pytorch 2.2.0+cu121
- Datasets 2.16.1
- Tokenizers 0.13.2
| null |
BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# microtest-2.0
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the azaheadhealth dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3672
- Accuracy: 0.75
- F1: 0.8
- Precision: 0.6667
- Recall: 1.0
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 2
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---:|:---------:|:------:|
| 0.8113 | 0.5 | 1 | 0.4486 | 0.75 | 0.8 | 0.6667 | 1.0 |
| 0.7227 | 1.0 | 2 | 0.3672 | 0.75 | 0.8 | 0.6667 | 1.0 |
### Framework versions
- Transformers 4.31.0
- Pytorch 2.2.0+cu121
- Datasets 2.16.1
- Tokenizers 0.13.2
|
{"base_model": "bert-base-uncased", "datasets": ["azaheadhealth"], "license": "apache-2.0", "metrics": ["accuracy", "f1", "precision", "recall"], "tags": ["generated_from_trainer"], "model-index": [{"name": "microtest-2.0", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "azaheadhealth", "type": "azaheadhealth", "config": "micro", "split": "test", "args": "micro"}, "metrics": [{"type": "accuracy", "value": 0.75, "name": "Accuracy"}, {"type": "f1", "value": 0.8, "name": "F1"}, {"type": "precision", "value": 0.6666666666666666, "name": "Precision"}, {"type": "recall", "value": 1.0, "name": "Recall"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,189 |
hmm404/tmp_trainer
|
hmm404
|
sentence-similarity
|
[
"sentence-transformers",
"tensorboard",
"safetensors",
"mpnet",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:32351",
"loss:TripletLoss",
"arxiv:1908.10084",
"arxiv:1703.07737",
"base_model:sentence-transformers/all-mpnet-base-v2",
"base_model:finetune:sentence-transformers/all-mpnet-base-v2",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2025-02-18T07:38:08Z |
2025-02-21T00:28:34+00:00
| 25 | 0 |
---
base_model: sentence-transformers/all-mpnet-base-v2
library_name: sentence-transformers
metrics:
- cosine_accuracy
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:32351
- loss:TripletLoss
widget:
- source_sentence: Genetic conditions that cause nutritional deficiencies can prevent
a person from removing meat from their diet.
sentences:
- Ante un estado que no quiere hablar del tema, para Cataluña, solo es posible seguir
su propio camino por otras vías.
- Retinol deficiency is a genetically pre-disposed condition that prevents conversion
beta-carotene to Vitamin A \(retinol\) in humans. Since plants have no retinol
\(only beta-carotene\), humans with this condition cannot have a vegan diet, only
one with animal products.
- People with hemochromatosis \(a genetic condition\) can benefit greatly from a
vegan diet, due to the lower absorbing non-heme iron in plants \(compared to heme
iron in meat\).
- source_sentence: 'The definition of veganism is: "A way of living which seeks to
exclude, as far as is possible and practicable, all forms of exploitation of,
and cruelty to, animals for food, clothing or any other purpose." In the \(unlikely\)
case of survival or health concerns, the "as far as possible and practicable"
clause makes it possible for such persons to be considered vegan as they would
have no alternative options.'
sentences:
- Veganism is not solely about diet. A person can still choose to live in accordance
with vegan values, such as by avoiding animal circuses and leather/fur products.
- It's easier to regulate established companies in a legal market than it is in
the black market. Any issue would be with bad regulations not legalization.
- That definition is too vague. There are different definitions of veganism, many
of which are not compatible with using animals in any circumstances. In a way
we are all vegan depending on how easy you believe it is to reach all the necessary
nutrition in your city harming as few animals as possible.
- source_sentence: Adding coding to the school curriculum means that something else
must be left out.
sentences:
- Coding skills are much needed in today's job market.
- Cataluña saldría de la UE con efectos económicos desastrosos.
- Teaching coding effectively is impossible unless teachers are trained appropriately
first.
- source_sentence: Animals have innate, individual rights, which are taken away when
they are killed or made to suffer.
sentences:
- Animals have a desire to live.
- Uno de los ejemplos más claros es la falta de inversión reiterada al Corredor
Mediterráneo \(Algeciras-Valencia-Barcelona-Francia\), prioritario para la UE
y Catalunya, pero relegado a algo residual por el estado Español.
- A vegan society would equate humans rights with animal rights, which would make
society worse off overall.
- source_sentence: The sorts of people likely to lash out against affirmative action
policies probably already hold negative views towards racial minorities.
sentences:
- The Far Right movement sees the inequality affirmative action addresses not as
a problem to be solved, but as an outcome to be desired.
- There are plenty of people who hold a positive view towards racial minorities
and still oppose affirmative action.
- Research has shown that college degrees have less economic utility for people
from low socio-economic backgrounds.
model-index:
- name: SentenceTransformer based on sentence-transformers/all-mpnet-base-v2
results:
- task:
type: triplet
name: Triplet
dataset:
name: Unknown
type: unknown
metrics:
- type: cosine_accuracy
value: 0.9264069199562073
name: Cosine Accuracy
- type: cosine_accuracy
value: 0.9161931872367859
name: Cosine Accuracy
---
# SentenceTransformer based on sentence-transformers/all-mpnet-base-v2
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) <!-- at revision 9a3225965996d404b775526de6dbfe85d3368642 -->
- **Maximum Sequence Length:** 384 tokens
- **Output Dimensionality:** 768 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 384, 'do_lower_case': False}) with Transformer model: MPNetModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
'The sorts of people likely to lash out against affirmative action policies probably already hold negative views towards racial minorities.',
'The Far Right movement sees the inequality affirmative action addresses not as a problem to be solved, but as an outcome to be desired.',
'There are plenty of people who hold a positive view towards racial minorities and still oppose affirmative action.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Triplet
* Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| **cosine_accuracy** | **0.9264** |
#### Triplet
* Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| **cosine_accuracy** | **0.9162** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 32,351 training samples
* Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 1000 samples:
| | anchor | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | string |
| details | <ul><li>min: 6 tokens</li><li>mean: 30.94 tokens</li><li>max: 160 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 40.8 tokens</li><li>max: 180 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 44.95 tokens</li><li>max: 162 tokens</li></ul> |
* Samples:
| anchor | positive | negative |
|:----------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>La soberanía y la decisión sobre la unidad de España residen en el conjunto de España.</code> | <code>Apostar por un proceso de secesión es ir en contra de la globalización, la corriente histórica que vivimos.</code> | <code>Los tratados internacionales \(incluido el Tratado de La Unión Europea\) no serían aplicables a Cataluña como estado independiente, por lo que su permanencia en Europa podría verse interrumpida.</code> |
| <code>La soberanía y la decisión sobre la unidad de España residen en el conjunto de España.</code> | <code>Para sentar un precedente en conflictos de autodeterminación en el mundo.</code> | <code>La independencia de Cataluña afectaría negativamente a la economía de España.</code> |
| <code>La soberanía y la decisión sobre la unidad de España residen en el conjunto de España.</code> | <code>Para terminar con el trato injusto que recibe Cataluña al ser parte de España.</code> | <code>Por definición, cualquier nacionalismo es malo ya que crea divisiones artificiales y es fuente de conflictos.</code> |
* Loss: [<code>TripletLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#tripletloss) with these parameters:
```json
{
"distance_metric": "TripletDistanceMetric.COSINE",
"triplet_margin": 0.3
}
```
### Training Hyperparameters
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: no
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 8
- `per_device_eval_batch_size`: 8
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 3.0
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | cosine_accuracy |
|:------:|:-----:|:-------------:|:---------------:|
| 0.1236 | 500 | 0.1872 | - |
| 0.2473 | 1000 | 0.1954 | - |
| 0.3709 | 1500 | 0.1854 | - |
| 0.4946 | 2000 | 0.1891 | - |
| 0.6182 | 2500 | 0.181 | - |
| 0.7418 | 3000 | 0.1794 | - |
| 0.8655 | 3500 | 0.1815 | - |
| 0.9891 | 4000 | 0.1736 | - |
| 1.1128 | 4500 | 0.1342 | - |
| 1.2364 | 5000 | 0.1297 | - |
| 1.3600 | 5500 | 0.1318 | - |
| 1.4837 | 6000 | 0.1255 | - |
| 1.6073 | 6500 | 0.128 | - |
| 1.7310 | 7000 | 0.1233 | - |
| 1.8546 | 7500 | 0.1221 | - |
| 1.9782 | 8000 | 0.1232 | - |
| 2.1019 | 8500 | 0.0841 | - |
| 2.2255 | 9000 | 0.0757 | - |
| 2.3492 | 9500 | 0.0764 | - |
| 2.4728 | 10000 | 0.0761 | - |
| 2.5964 | 10500 | 0.0726 | - |
| 2.7201 | 11000 | 0.0644 | - |
| 2.8437 | 11500 | 0.073 | - |
| 2.9674 | 12000 | 0.0725 | - |
| -1 | -1 | - | 0.9162 |
### Framework Versions
- Python: 3.11.11
- Sentence Transformers: 3.4.1
- Transformers: 4.48.3
- PyTorch: 2.5.1+cu124
- Accelerate: 1.3.0
- Datasets: 3.3.2
- Tokenizers: 0.21.0
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### TripletLoss
```bibtex
@misc{hermans2017defense,
title={In Defense of the Triplet Loss for Person Re-Identification},
author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
year={2017},
eprint={1703.07737},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# SentenceTransformer based on sentence-transformers/all-mpnet-base-v2
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) <!-- at revision 9a3225965996d404b775526de6dbfe85d3368642 -->
- **Maximum Sequence Length:** 384 tokens
- **Output Dimensionality:** 768 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 384, 'do_lower_case': False}) with Transformer model: MPNetModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
'The sorts of people likely to lash out against affirmative action policies probably already hold negative views towards racial minorities.',
'The Far Right movement sees the inequality affirmative action addresses not as a problem to be solved, but as an outcome to be desired.',
'There are plenty of people who hold a positive view towards racial minorities and still oppose affirmative action.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Triplet
* Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| **cosine_accuracy** | **0.9264** |
#### Triplet
* Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| **cosine_accuracy** | **0.9162** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 32,351 training samples
* Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 1000 samples:
| | anchor | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | string |
| details | <ul><li>min: 6 tokens</li><li>mean: 30.94 tokens</li><li>max: 160 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 40.8 tokens</li><li>max: 180 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 44.95 tokens</li><li>max: 162 tokens</li></ul> |
* Samples:
| anchor | positive | negative |
|:----------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>La soberanía y la decisión sobre la unidad de España residen en el conjunto de España.</code> | <code>Apostar por un proceso de secesión es ir en contra de la globalización, la corriente histórica que vivimos.</code> | <code>Los tratados internacionales \(incluido el Tratado de La Unión Europea\) no serían aplicables a Cataluña como estado independiente, por lo que su permanencia en Europa podría verse interrumpida.</code> |
| <code>La soberanía y la decisión sobre la unidad de España residen en el conjunto de España.</code> | <code>Para sentar un precedente en conflictos de autodeterminación en el mundo.</code> | <code>La independencia de Cataluña afectaría negativamente a la economía de España.</code> |
| <code>La soberanía y la decisión sobre la unidad de España residen en el conjunto de España.</code> | <code>Para terminar con el trato injusto que recibe Cataluña al ser parte de España.</code> | <code>Por definición, cualquier nacionalismo es malo ya que crea divisiones artificiales y es fuente de conflictos.</code> |
* Loss: [<code>TripletLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#tripletloss) with these parameters:
```json
{
"distance_metric": "TripletDistanceMetric.COSINE",
"triplet_margin": 0.3
}
```
### Training Hyperparameters
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: no
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 8
- `per_device_eval_batch_size`: 8
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 3.0
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | cosine_accuracy |
|:------:|:-----:|:-------------:|:---------------:|
| 0.1236 | 500 | 0.1872 | - |
| 0.2473 | 1000 | 0.1954 | - |
| 0.3709 | 1500 | 0.1854 | - |
| 0.4946 | 2000 | 0.1891 | - |
| 0.6182 | 2500 | 0.181 | - |
| 0.7418 | 3000 | 0.1794 | - |
| 0.8655 | 3500 | 0.1815 | - |
| 0.9891 | 4000 | 0.1736 | - |
| 1.1128 | 4500 | 0.1342 | - |
| 1.2364 | 5000 | 0.1297 | - |
| 1.3600 | 5500 | 0.1318 | - |
| 1.4837 | 6000 | 0.1255 | - |
| 1.6073 | 6500 | 0.128 | - |
| 1.7310 | 7000 | 0.1233 | - |
| 1.8546 | 7500 | 0.1221 | - |
| 1.9782 | 8000 | 0.1232 | - |
| 2.1019 | 8500 | 0.0841 | - |
| 2.2255 | 9000 | 0.0757 | - |
| 2.3492 | 9500 | 0.0764 | - |
| 2.4728 | 10000 | 0.0761 | - |
| 2.5964 | 10500 | 0.0726 | - |
| 2.7201 | 11000 | 0.0644 | - |
| 2.8437 | 11500 | 0.073 | - |
| 2.9674 | 12000 | 0.0725 | - |
| -1 | -1 | - | 0.9162 |
### Framework Versions
- Python: 3.11.11
- Sentence Transformers: 3.4.1
- Transformers: 4.48.3
- PyTorch: 2.5.1+cu124
- Accelerate: 1.3.0
- Datasets: 3.3.2
- Tokenizers: 0.21.0
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### TripletLoss
```bibtex
@misc{hermans2017defense,
title={In Defense of the Triplet Loss for Person Re-Identification},
author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
year={2017},
eprint={1703.07737},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "sentence-transformers/all-mpnet-base-v2", "library_name": "sentence-transformers", "metrics": ["cosine_accuracy"], "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:32351", "loss:TripletLoss"], "widget": [{"source_sentence": "Genetic conditions that cause nutritional deficiencies can prevent a person from removing meat from their diet.", "sentences": ["Ante un estado que no quiere hablar del tema, para Cataluña, solo es posible seguir su propio camino por otras vías.", "Retinol deficiency is a genetically pre-disposed condition that prevents conversion beta-carotene to Vitamin A \\(retinol\\) in humans. Since plants have no retinol \\(only beta-carotene\\), humans with this condition cannot have a vegan diet, only one with animal products.", "People with hemochromatosis \\(a genetic condition\\) can benefit greatly from a vegan diet, due to the lower absorbing non-heme iron in plants \\(compared to heme iron in meat\\)."]}, {"source_sentence": "The definition of veganism is: \"A way of living which seeks to exclude, as far as is possible and practicable, all forms of exploitation of, and cruelty to, animals for food, clothing or any other purpose.\" In the \\(unlikely\\) case of survival or health concerns, the \"as far as possible and practicable\" clause makes it possible for such persons to be considered vegan as they would have no alternative options.", "sentences": ["Veganism is not solely about diet. A person can still choose to live in accordance with vegan values, such as by avoiding animal circuses and leather/fur products.", "It's easier to regulate established companies in a legal market than it is in the black market. Any issue would be with bad regulations not legalization.", "That definition is too vague. There are different definitions of veganism, many of which are not compatible with using animals in any circumstances. In a way we are all vegan depending on how easy you believe it is to reach all the necessary nutrition in your city harming as few animals as possible."]}, {"source_sentence": "Adding coding to the school curriculum means that something else must be left out.", "sentences": ["Coding skills are much needed in today's job market.", "Cataluña saldría de la UE con efectos económicos desastrosos.", "Teaching coding effectively is impossible unless teachers are trained appropriately first."]}, {"source_sentence": "Animals have innate, individual rights, which are taken away when they are killed or made to suffer.", "sentences": ["Animals have a desire to live.", "Uno de los ejemplos más claros es la falta de inversión reiterada al Corredor Mediterráneo \\(Algeciras-Valencia-Barcelona-Francia\\), prioritario para la UE y Catalunya, pero relegado a algo residual por el estado Español.", "A vegan society would equate humans rights with animal rights, which would make society worse off overall."]}, {"source_sentence": "The sorts of people likely to lash out against affirmative action policies probably already hold negative views towards racial minorities.", "sentences": ["The Far Right movement sees the inequality affirmative action addresses not as a problem to be solved, but as an outcome to be desired.", "There are plenty of people who hold a positive view towards racial minorities and still oppose affirmative action.", "Research has shown that college degrees have less economic utility for people from low socio-economic backgrounds."]}], "model-index": [{"name": "SentenceTransformer based on sentence-transformers/all-mpnet-base-v2", "results": [{"task": {"type": "triplet", "name": "Triplet"}, "dataset": {"name": "Unknown", "type": "unknown"}, "metrics": [{"type": "cosine_accuracy", "value": 0.9264069199562073, "name": "Cosine Accuracy"}, {"type": "cosine_accuracy", "value": 0.9161931872367859, "name": "Cosine Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,190 |
joeranbosma/dragon-roberta-base-mixed-domain
|
joeranbosma
|
fill-mask
|
[
"transformers",
"pytorch",
"safetensors",
"xlm-roberta",
"fill-mask",
"doi:10.57967/hf/2168",
"license:cc-by-nc-sa-4.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-05-03T09:38:13Z |
2025-02-07T09:30:11+00:00
| 55 | 0 |
---
license: cc-by-nc-sa-4.0
---
# DRAGON RoBERTa base mixed-domain
Pretrained model on Dutch clinical reports using a masked language modeling (MLM) objective. It was introduced in [this](#pending) paper. The model was first pretrained using general domain data, as specified [here](https://huggingface.co/xlm-roberta-base). The pretrained model was taken from HuggingFace: [`xlm-roberta-base`](https://huggingface.co/xlm-roberta-base). Subsequently, the model was pretrained using domain-specific data (i.e., clinical reports). The tokenizer of [`xlm-roberta-base`](https://huggingface.co/xlm-roberta-base) was used.
## Model description
RoBERTa is a transformers model that was pretrained on a large corpus of Dutch clinical reports in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labeling them in any way with an automatic process to generate inputs and labels from those texts.
This way, the model learns an inner representation of the Dutch medical language that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled reports, for instance, you can train a standard classifier using the features produced by the BERT model as inputs.
## Model variations
Multiple architectures were pretrained for the DRAGON challenge.
| Model | #params | Language |
|------------------------|--------------------------------|-------|
| [`joeranbosma/dragon-bert-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-bert-base-mixed-domain) | 109M | Dutch → Dutch |
| [`joeranbosma/dragon-roberta-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-roberta-base-mixed-domain) | 278M | Multiple → Dutch |
| [`joeranbosma/dragon-roberta-large-mixed-domain`](https://huggingface.co/joeranbosma/dragon-roberta-large-mixed-domain) | 560M | Multiple → Dutch |
| [`joeranbosma/dragon-longformer-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-longformer-base-mixed-domain) | 149M | English → Dutch |
| [`joeranbosma/dragon-longformer-large-mixed-domain`](https://huggingface.co/joeranbosma/dragon-longformer-large-mixed-domain) | 435M | English → Dutch |
| [`joeranbosma/dragon-bert-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-bert-base-domain-specific) | 109M | Dutch |
| [`joeranbosma/dragon-roberta-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-roberta-base-domain-specific) | 278M | Dutch |
| [`joeranbosma/dragon-roberta-large-domain-specific`](https://huggingface.co/joeranbosma/dragon-roberta-large-domain-specific) | 560M | Dutch |
| [`joeranbosma/dragon-longformer-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-longformer-base-domain-specific) | 149M | Dutch |
| [`joeranbosma/dragon-longformer-large-domain-specific`](https://huggingface.co/joeranbosma/dragon-longformer-large-domain-specific) | 435M | Dutch |
## Intended uses & limitations
You can use the raw model for masked language modeling, but it's mostly intended to be fine-tuned on a downstream task.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole text (e.g., a clinical report) to make decisions, such as sequence classification, token classification or question answering. For tasks such as text generation you should look at model like GPT2.
## How to use
You can use this model directly with a pipeline for masked language modeling:
```python
from transformers import pipeline
unmasker = pipeline("fill-mask", model="joeranbosma/dragon-roberta-base-mixed-domain")
unmasker("Dit onderzoek geen aanwijzingen voor significant carcinoom. PIRADS <mask>.")
```
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("joeranbosma/dragon-roberta-base-mixed-domain")
model = AutoModel.from_pretrained("joeranbosma/dragon-roberta-base-mixed-domain")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors="pt")
output = model(**encoded_input)
```
## Limitations and bias
Even if the training data used for this model could be characterized as fairly neutral, this model can have biased predictions. This bias will also affect all fine-tuned versions of this model.
## Training data
For pretraining, 4,333,201 clinical reports (466,351 consecutive patients) were selected from Ziekenhuisgroep Twente from patients with a diagnostic or interventional visit between 13 July 2000 and 25 April 2023. 180,439 duplicate clinical reports (179,808 patients) were excluded, resulting in 4,152,762 included reports (463,692 patients). These reports were split into training (80%, 3,322,209 reports), validation (10%, 415,276 reports), and testing (10%, 415,277 reports). The testing reports were set aside for future analysis and are not used for pretraining.
## Training procedure
### Pretraining
The model was pretrained using masked language modeling (MLM): taking a sentence, the model randomly masks 15% of the words in the input then runs the entire masked sentence through the model and has to predict the masked words. This is different from traditional recurrent neural networks (RNNs) that usually see the words one after the other, or from autoregressive models like GPT which internally masks the future tokens. It allows the model to learn a bidirectional representation of the sentence.
The details of the masking procedure for each sentence are the following:
- 15% of the tokens are masked.
- In 80% of the cases, the masked tokens are replaced by `[MASK]`.
- In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace.
- In the 10% remaining cases, the masked tokens are left as is.
The HuggingFace implementation was used for pretraining: [`run_mlm.py`](https://github.com/huggingface/transformers/blob/7c6ec195adbfcd22cb6baeee64dd3c24a4b80c74/examples/pytorch/language-modeling/run_mlm.py).
### Pretraining hyperparameters
The following hyperparameters were used during pretraining:
- `learning_rate`: 5e-05
- `train_batch_size`: 4
- `eval_batch_size`: 4
- `seed`: 42
- `gradient_accumulation_steps`: 4
- `total_train_batch_size`: 16
- `optimizer`: Adam with betas=(0.9,0.999) and epsilon=1e-08
- `lr_scheduler_type`: linear
- `num_epochs`: 3.0
- `max_seq_length`: 512
### Framework versions
- Transformers 4.29.0.dev0
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3
## Evaluation results
Pending evaluation on the DRAGON benchmark.
### BibTeX entry and citation info
```bibtex
@article{PENDING}
```
| null |
BioNLP
|
# DRAGON RoBERTa base mixed-domain
Pretrained model on Dutch clinical reports using a masked language modeling (MLM) objective. It was introduced in [this](#pending) paper. The model was first pretrained using general domain data, as specified [here](https://huggingface.co/xlm-roberta-base). The pretrained model was taken from HuggingFace: [`xlm-roberta-base`](https://huggingface.co/xlm-roberta-base). Subsequently, the model was pretrained using domain-specific data (i.e., clinical reports). The tokenizer of [`xlm-roberta-base`](https://huggingface.co/xlm-roberta-base) was used.
## Model description
RoBERTa is a transformers model that was pretrained on a large corpus of Dutch clinical reports in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labeling them in any way with an automatic process to generate inputs and labels from those texts.
This way, the model learns an inner representation of the Dutch medical language that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled reports, for instance, you can train a standard classifier using the features produced by the BERT model as inputs.
## Model variations
Multiple architectures were pretrained for the DRAGON challenge.
| Model | #params | Language |
|------------------------|--------------------------------|-------|
| [`joeranbosma/dragon-bert-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-bert-base-mixed-domain) | 109M | Dutch → Dutch |
| [`joeranbosma/dragon-roberta-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-roberta-base-mixed-domain) | 278M | Multiple → Dutch |
| [`joeranbosma/dragon-roberta-large-mixed-domain`](https://huggingface.co/joeranbosma/dragon-roberta-large-mixed-domain) | 560M | Multiple → Dutch |
| [`joeranbosma/dragon-longformer-base-mixed-domain`](https://huggingface.co/joeranbosma/dragon-longformer-base-mixed-domain) | 149M | English → Dutch |
| [`joeranbosma/dragon-longformer-large-mixed-domain`](https://huggingface.co/joeranbosma/dragon-longformer-large-mixed-domain) | 435M | English → Dutch |
| [`joeranbosma/dragon-bert-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-bert-base-domain-specific) | 109M | Dutch |
| [`joeranbosma/dragon-roberta-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-roberta-base-domain-specific) | 278M | Dutch |
| [`joeranbosma/dragon-roberta-large-domain-specific`](https://huggingface.co/joeranbosma/dragon-roberta-large-domain-specific) | 560M | Dutch |
| [`joeranbosma/dragon-longformer-base-domain-specific`](https://huggingface.co/joeranbosma/dragon-longformer-base-domain-specific) | 149M | Dutch |
| [`joeranbosma/dragon-longformer-large-domain-specific`](https://huggingface.co/joeranbosma/dragon-longformer-large-domain-specific) | 435M | Dutch |
## Intended uses & limitations
You can use the raw model for masked language modeling, but it's mostly intended to be fine-tuned on a downstream task.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole text (e.g., a clinical report) to make decisions, such as sequence classification, token classification or question answering. For tasks such as text generation you should look at model like GPT2.
## How to use
You can use this model directly with a pipeline for masked language modeling:
```python
from transformers import pipeline
unmasker = pipeline("fill-mask", model="joeranbosma/dragon-roberta-base-mixed-domain")
unmasker("Dit onderzoek geen aanwijzingen voor significant carcinoom. PIRADS <mask>.")
```
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("joeranbosma/dragon-roberta-base-mixed-domain")
model = AutoModel.from_pretrained("joeranbosma/dragon-roberta-base-mixed-domain")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors="pt")
output = model(**encoded_input)
```
## Limitations and bias
Even if the training data used for this model could be characterized as fairly neutral, this model can have biased predictions. This bias will also affect all fine-tuned versions of this model.
## Training data
For pretraining, 4,333,201 clinical reports (466,351 consecutive patients) were selected from Ziekenhuisgroep Twente from patients with a diagnostic or interventional visit between 13 July 2000 and 25 April 2023. 180,439 duplicate clinical reports (179,808 patients) were excluded, resulting in 4,152,762 included reports (463,692 patients). These reports were split into training (80%, 3,322,209 reports), validation (10%, 415,276 reports), and testing (10%, 415,277 reports). The testing reports were set aside for future analysis and are not used for pretraining.
## Training procedure
### Pretraining
The model was pretrained using masked language modeling (MLM): taking a sentence, the model randomly masks 15% of the words in the input then runs the entire masked sentence through the model and has to predict the masked words. This is different from traditional recurrent neural networks (RNNs) that usually see the words one after the other, or from autoregressive models like GPT which internally masks the future tokens. It allows the model to learn a bidirectional representation of the sentence.
The details of the masking procedure for each sentence are the following:
- 15% of the tokens are masked.
- In 80% of the cases, the masked tokens are replaced by `[MASK]`.
- In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace.
- In the 10% remaining cases, the masked tokens are left as is.
The HuggingFace implementation was used for pretraining: [`run_mlm.py`](https://github.com/huggingface/transformers/blob/7c6ec195adbfcd22cb6baeee64dd3c24a4b80c74/examples/pytorch/language-modeling/run_mlm.py).
### Pretraining hyperparameters
The following hyperparameters were used during pretraining:
- `learning_rate`: 5e-05
- `train_batch_size`: 4
- `eval_batch_size`: 4
- `seed`: 42
- `gradient_accumulation_steps`: 4
- `total_train_batch_size`: 16
- `optimizer`: Adam with betas=(0.9,0.999) and epsilon=1e-08
- `lr_scheduler_type`: linear
- `num_epochs`: 3.0
- `max_seq_length`: 512
### Framework versions
- Transformers 4.29.0.dev0
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3
## Evaluation results
Pending evaluation on the DRAGON benchmark.
### BibTeX entry and citation info
```bibtex
@article{PENDING}
```
|
{"license": "cc-by-nc-sa-4.0"}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,191 |
mrm8488/bert2bert_shared-turkish-summarization
|
mrm8488
|
summarization
|
[
"transformers",
"pytorch",
"safetensors",
"encoder-decoder",
"text2text-generation",
"summarization",
"news",
"tr",
"dataset:mlsum",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2022-03-02T23:29:05Z |
2023-04-26T19:56:12+00:00
| 982 | 19 |
---
datasets:
- mlsum
language: tr
tags:
- summarization
- news
widget:
- text: Ankara'da oto hırsızlık çetesine yönelikdüzenlenen ‘Balta’ operasyonunda,
çete lideri‘balta’ lakaplı şahıs ile 7 kişi gözaltına alındı.Diğer bir operasyonda
ise 3 şüpheli çaldıklarıaraçları parçalarken yapılan baskında suçüstüyakalandı.
Ankara Emniyet Müdürlüğü’ne bağlıAsayiş Şube Müdürlüğü Oto Hırsızlık Büro Amirliğiekipleri,
Ankara ilinde meydana gelen, otohırsızlık olaylarına karşı Ankara CumhuriyetBaşsavcılığı’nın
izniyle yürüttükleri 3 aylıkçalışma sonucunda operasyon düğmesine bastı.Yapılan
teknik ve fiziki takip sonucunda, ‘Balta’çetesine ulaşıldı. Çeteyi izleyen ekipler,
Ankara,Konya ve Antalya’da eş zamanlı operasyondüzenleyerek çete lideri ‘Balta’
lakaplı Necati D.ve çete üyesi 7 kişiyi yakaladı. Takip edildiğinianlayınca ortadan
kayboldu Çete lideri ‘Balta’nın,polis ekipleri tarafından izlendiğini anladığı
veaylarca ortada görünmediğini tespit eden HırsızlıkBüro ekipleri, ‘Balta’nın
kendi suç ortaklarını dadolandırmaya çalıştığını saptadı. Adliyeye sevkedilen
şüphelilerden haklarında çok sayıda otohırsızlık kaydı bulunan çete lideri Necati
D.,Ferhat K., Atakan A. ve Tayfun G., çıkarıldıklarınöbetçi sulh hakimliğince
tutuklanarak cezaevinegönderildi. Diğer 3 şüpheli ise adli kontrolşartıyla serbest
bırakıldı. Çaldıkları araçlarıparçalarken polis bastı Diğer bir olay iseAltındağ
ilçesinde meydana geldi. Hırsızlık Büroekipleri inceledikleri 2 oto hırsızlık
olayınınsonucunda 3 şüpheliyi takibe aldı. Şüphelilerinçaldıkları 2 aracı İvedik
Hurdacılar Sitesi’ndekidepolarında parçalayacaklarını belirleyen ekiplerharekete
geçti. Depoya baskın yapan polisekipleri, 3 şüpheliyi suçüstü yakaladı.Emniyetteki
işlemlerinin ardından adliyeye sevkedilen hırsızlık zanlıları, çıkarıldıkları
nöbetçimahkeme tarafından adli kontrol şartıyla serbestbırakıldı.
---
# Turkish BERT2BERT (shared) fine-tuned on MLSUM TR for summarization
## Model
[dbmdz/bert-base-turkish-cased](https://huggingface.co/dbmdz/bert-base-turkish-cased) (BERT Checkpoint)
## Dataset
**MLSUM** is the first large-scale MultiLingual SUMmarization dataset. Obtained from online newspapers, it contains 1.5M+ article/summary pairs in five different languages -- namely, French, German, Spanish, Russian, **Turkish**. Together with English newspapers from the popular CNN/Daily mail dataset, the collected data form a large scale multilingual dataset which can enable new research directions for the text summarization community. We report cross-lingual comparative analyses based on state-of-the-art systems. These highlight existing biases which motivate the use of a multi-lingual dataset.
[MLSUM tu/tr](https://huggingface.co/datasets/viewer/?dataset=mlsum)
## Results
|Set|Metric| Value|
|----|------|------|
| Test |Rouge2 - mid -precision | **32.41**|
| Test | Rouge2 - mid - recall | **28.65**|
| Test | Rouge2 - mid - fmeasure | **29.48**|
## Usage
```python
import torch
from transformers import BertTokenizerFast, EncoderDecoderModel
device = 'cuda' if torch.cuda.is_available() else 'cpu'
ckpt = 'mrm8488/bert2bert_shared-turkish-summarization'
tokenizer = BertTokenizerFast.from_pretrained(ckpt)
model = EncoderDecoderModel.from_pretrained(ckpt).to(device)
def generate_summary(text):
inputs = tokenizer([text], padding="max_length", truncation=True, max_length=512, return_tensors="pt")
input_ids = inputs.input_ids.to(device)
attention_mask = inputs.attention_mask.to(device)
output = model.generate(input_ids, attention_mask=attention_mask)
return tokenizer.decode(output[0], skip_special_tokens=True)
text = "Your text here..."
generate_summary(text)
```
> Created by [Manuel Romero/@mrm8488](https://twitter.com/mrm8488) with the support of [Narrativa](https://www.narrativa.com/)
> Made with <span style="color: #e25555;">♥</span> in Spain
| null |
Non_BioNLP
|
# Turkish BERT2BERT (shared) fine-tuned on MLSUM TR for summarization
## Model
[dbmdz/bert-base-turkish-cased](https://huggingface.co/dbmdz/bert-base-turkish-cased) (BERT Checkpoint)
## Dataset
**MLSUM** is the first large-scale MultiLingual SUMmarization dataset. Obtained from online newspapers, it contains 1.5M+ article/summary pairs in five different languages -- namely, French, German, Spanish, Russian, **Turkish**. Together with English newspapers from the popular CNN/Daily mail dataset, the collected data form a large scale multilingual dataset which can enable new research directions for the text summarization community. We report cross-lingual comparative analyses based on state-of-the-art systems. These highlight existing biases which motivate the use of a multi-lingual dataset.
[MLSUM tu/tr](https://huggingface.co/datasets/viewer/?dataset=mlsum)
## Results
|Set|Metric| Value|
|----|------|------|
| Test |Rouge2 - mid -precision | **32.41**|
| Test | Rouge2 - mid - recall | **28.65**|
| Test | Rouge2 - mid - fmeasure | **29.48**|
## Usage
```python
import torch
from transformers import BertTokenizerFast, EncoderDecoderModel
device = 'cuda' if torch.cuda.is_available() else 'cpu'
ckpt = 'mrm8488/bert2bert_shared-turkish-summarization'
tokenizer = BertTokenizerFast.from_pretrained(ckpt)
model = EncoderDecoderModel.from_pretrained(ckpt).to(device)
def generate_summary(text):
inputs = tokenizer([text], padding="max_length", truncation=True, max_length=512, return_tensors="pt")
input_ids = inputs.input_ids.to(device)
attention_mask = inputs.attention_mask.to(device)
output = model.generate(input_ids, attention_mask=attention_mask)
return tokenizer.decode(output[0], skip_special_tokens=True)
text = "Your text here..."
generate_summary(text)
```
> Created by [Manuel Romero/@mrm8488](https://twitter.com/mrm8488) with the support of [Narrativa](https://www.narrativa.com/)
> Made with <span style="color: #e25555;">♥</span> in Spain
|
{"datasets": ["mlsum"], "language": "tr", "tags": ["summarization", "news"], "widget": [{"text": "Ankara'da oto hırsızlık çetesine yönelikdüzenlenen ‘Balta’ operasyonunda, çete lideri‘balta’ lakaplı şahıs ile 7 kişi gözaltına alındı.Diğer bir operasyonda ise 3 şüpheli çaldıklarıaraçları parçalarken yapılan baskında suçüstüyakalandı. Ankara Emniyet Müdürlüğü’ne bağlıAsayiş Şube Müdürlüğü Oto Hırsızlık Büro Amirliğiekipleri, Ankara ilinde meydana gelen, otohırsızlık olaylarına karşı Ankara CumhuriyetBaşsavcılığı’nın izniyle yürüttükleri 3 aylıkçalışma sonucunda operasyon düğmesine bastı.Yapılan teknik ve fiziki takip sonucunda, ‘Balta’çetesine ulaşıldı. Çeteyi izleyen ekipler, Ankara,Konya ve Antalya’da eş zamanlı operasyondüzenleyerek çete lideri ‘Balta’ lakaplı Necati D.ve çete üyesi 7 kişiyi yakaladı. Takip edildiğinianlayınca ortadan kayboldu Çete lideri ‘Balta’nın,polis ekipleri tarafından izlendiğini anladığı veaylarca ortada görünmediğini tespit eden HırsızlıkBüro ekipleri, ‘Balta’nın kendi suç ortaklarını dadolandırmaya çalıştığını saptadı. Adliyeye sevkedilen şüphelilerden haklarında çok sayıda otohırsızlık kaydı bulunan çete lideri Necati D.,Ferhat K., Atakan A. ve Tayfun G., çıkarıldıklarınöbetçi sulh hakimliğince tutuklanarak cezaevinegönderildi. Diğer 3 şüpheli ise adli kontrolşartıyla serbest bırakıldı. Çaldıkları araçlarıparçalarken polis bastı Diğer bir olay iseAltındağ ilçesinde meydana geldi. Hırsızlık Büroekipleri inceledikleri 2 oto hırsızlık olayınınsonucunda 3 şüpheliyi takibe aldı. Şüphelilerinçaldıkları 2 aracı İvedik Hurdacılar Sitesi’ndekidepolarında parçalayacaklarını belirleyen ekiplerharekete geçti. Depoya baskın yapan polisekipleri, 3 şüpheliyi suçüstü yakaladı.Emniyetteki işlemlerinin ardından adliyeye sevkedilen hırsızlık zanlıları, çıkarıldıkları nöbetçimahkeme tarafından adli kontrol şartıyla serbestbırakıldı."}]}
|
task
|
[
"SUMMARIZATION"
] | 45,192 |
AIDX-ktds/ktdsbaseLM-v0.11-based-on-openchat3.5
|
AIDX-ktds
|
text-generation
|
[
"adapter-transformers",
"safetensors",
"finance",
"biology",
"legal",
"art",
"text-generation-inference",
"text-generation",
"conversational",
"ko",
"en",
"dataset:AIDX-ktds/ko_leaderboard",
"base_model:openchat/openchat_3.5",
"base_model:adapter:openchat/openchat_3.5",
"license:mit",
"region:us"
] | 2024-09-23T05:46:59Z |
2024-10-31T01:50:56+00:00
| 0 | 1 |
---
base_model:
- openchat/openchat_3.5
datasets:
- AIDX-ktds/ko_leaderboard
language:
- ko
- en
library_name: adapter-transformers
license: mit
metrics:
- accuracy
pipeline_tag: text-generation
tags:
- finance
- biology
- legal
- art
- text-generation-inference
---
### ⛱ ktdsbaseLM v0.11은 openchat3.5를 Foundation 모델로 하는 한국어 및 한국의 다양한
### 문화에 적용할 수 있도록 하기 위해
### 개발 되었으며 자체 제작한 53영역의 한국어 데이터를 활용하여 한국 사회 가치와
### 문화를 이해하는 모델 입니다. ✌
# ❶ 모델 설명
- 모델명 및 주요기능:
KTDSbaseLM v0.11은 OpenChat 3.5 모델을 기반으로 SFT 방식으로 파인튜닝된 Mistral 7B / openchat3.5 기반 모델입니다.
한국어와 한국의 다양한 문화적 맥락을 이해하도록 설계되었으며 ✨✨, 자체 제작한 135개 영역의 한국어
데이터를 활용해 한국 사회의 가치와 문화를 반영합니다.
주요 기능으로는 텍스트 생성, 대화 추론, 문서 요약, 질의응답, 감정 분석 및 자연어 처리 관련 다양한 작업을 지원하며,
활용 분야는 법률, 재무, 과학, 교육, 비즈니스, 문화 연구 등 다양한 분야에서 응용될 수 있습니다.
- 모델 아키텍처: KTDSBaseLM v0.11은 Mistral 7B 모델을 기반으로, 파라미터 수는 70억 개(7B)로 구성된 고성능 언어 모델입니다.
이 모델은 OpenChat 3.5를 파운데이션 모델로 삼아, SFT(지도 미세 조정) 방식을 통해 한국어와 한국 문화에 특화된 성능을 발휘하도록 훈련되었습니다.
Mistral 7B의 경량화된 구조는 빠른 추론 속도와 메모리 효율성을 보장하며, 다양한 자연어 처리 작업에 적합하게 최적화되어 있습니다.
이 아키텍처는 텍스트 생성, 질의응답, 문서 요약, 감정 분석과 같은 다양한 작업에서 탁월한 성능을 보여줍니다.
# ❷ 학습 데이터
- ktdsbaseLM v0.11은 자체 개발한 총 3.6GB 크기의 데이터를 바탕으로 학습되었습니다. 모두 233만 건의 QnA, 요약, 분류 등 데이터를 포함하며,
그 중 133만 건은 53개 영역의 객관식 문제로 구성되었습니다. 이 영역에는 한국사, 사회, 재무, 법률, 세무, 수학, 생물, 물리, 화학 등이 포함되며,
Chain of Thought 방식으로 학습되었습니다. 또한 130만 건의 주관식 문제는 한국사, 재무, 법률, 세무, 수학 등 38개 영역에 걸쳐 학습되었습니다.
학습 데이터 중 한국의 사회 가치와 인간의 감정을 이해하고 지시한 사항에 따라 출력할 수 있는 데이터를 학습하였습니다.
# ❸ 사용 사례
ktdsbaseLM v0.11은 다양한 응용 분야에서 사용될 수 있습니다. 예를 들어:
- 교육 분야: 역사, 수학, 과학 등 다양한 학습 자료에 대한 질의응답 및 설명 생성.
- 비즈니스: 법률, 재무, 세무 관련 질의에 대한 답변 제공 및 문서 요약.
- 연구 및 문화: 한국 사회와 문화에 맞춘 자연어 처리 작업, 감정 분석, 문서 생성 및 번역.
- 고객 서비스: 사용자와의 대화 생성 및 맞춤형 응답 제공.
- 이 모델은 다양한 자연어 처리 작업에서 높은 활용도를 가집니다.
# ❹ 한계 ⛈⛈
- ktdsBaseLM v0.11은 한국어와 한국 문화에 특화되어 있으나,
특정 영역(예: 최신 국제 자료, 전문 분야)의 데이터 부족으로 인해 다른 언어 또는
문화에 대한 응답의 정확성이 떨어질 수 있습니다.
또한, 복잡한 논리적 사고를 요구하는 문제에 대해 제한된 추론 능력을 보일 수 있으며,
편향된 데이터가 포함될 경우 편향된 응답이 생성될 가능성도 존재합니다.
# ❺ 사용 방법
<pre><code>
import os
import os.path as osp
import sys
import fire
import json
from typing import List, Union
import pandas as pd
import torch
from torch.nn import functional as F
import transformers
from transformers import TrainerCallback, TrainingArguments, TrainerState, TrainerControl, BitsAndBytesConfig
from transformers.trainer_utils import PREFIX_CHECKPOINT_DIR
from transformers import LlamaForCausalLM, LlamaTokenizer
from transformers import AutoModelForCausalLM, AutoTokenizer
from datasets import load_dataset
from peft import (
LoraConfig,
get_peft_model,
set_peft_model_state_dict
)
from peft import PeftModel
import re
import ast
device = 'auto' #@param {type: "string"}
model = '' #@param {type: "string"}
model = AutoModelForCausalLM.from_pretrained(
model,
quantization_config=bnb_config,
#load_in_4bit=True, # Quantization Load
device_map=device)
tokenizer = AutoTokenizer.from_pretrained(base_LLM_model)
input_text = "안녕하세요."
inputs = tokenizer(input_text, return_tensors="pt")
inputs = inputs.to("cuda:0")
with torch.no_grad():
outputs = model.generate(**inputs, max_length=1024)
result = tokenizer.decode(outputs[0], skip_special_tokens=True)
</code></pre>
## ✅ ktds는 openchat 외에도 LlaMA, Polyglot, EEVE 등 대표적인 LLM에 다양한 영역의 한국의 문화와 지식을 파인튜닝한 LLM을 제공할 예정입니다.
---
Here’s the English version of the provided text:
# ❶ Model Description
**Model Name and Key Features**:
KTDSbaseLM v0.11 is based on the OpenChat 3.5 model, fine-tuned using the SFT method on the Mistral 7B model.
It is designed to understand Korean and various cultural contexts, utilizing data from 135 domains in Korean society.
The model supports tasks such as text generation, conversation inference, document summarization,
question answering, sentiment analysis, and other NLP tasks.
Its applications span fields like law, finance, science, education, business, and cultural research.
**Model Architecture**:
KTDSBaseLM v0.11 is a high-performance language model with 7 billion parameters based on the Mistral 7B model.
It uses OpenChat 3.5 as the foundation and is fine-tuned using SFT to excel in Korean language and culture.
The streamlined Mistral 7B architecture ensures fast inference and memory efficiency,
optimized for various NLP tasks like text generation, question answering, document summarization, and sentiment analysis.
---
# ❷ Training Data
KTDSbaseLM v0.11 was trained on 3.6GB of data, comprising 2.33 million Q&A instances.
This includes 1.33 million multiple-choice questions across 53 domains such as history,
finance, law, tax, and science, trained with the Chain of Thought method. Additionally,
1.3 million short-answer questions cover 38 domains including history, finance, and law.
**Training Instruction Dataset Format**:
`{"prompt": "prompt text", "completion": "ideal generated text"}`
---
# ❸ Use Cases
KTDSbaseLM v0.11 can be used across multiple fields, such as:
- **Education**: Answering questions and generating explanations for subjects like history, math, and science.
- **Business**: Providing responses and summaries for legal, financial, and tax-related queries.
- **Research and Culture**: Performing NLP tasks, sentiment analysis, document generation, and translation.
- **Customer Service**: Generating conversations and personalized responses for users.
This model is highly versatile in various NLP tasks.
---
# ❹ Limitations
KTDSBaseLM v0.11 is specialized in Korean language and culture.
However, it may lack accuracy in responding to topics outside its scope,
such as international or specialized data.
Additionally, it may have limited reasoning ability for complex logical problems and
may produce biased responses if trained on biased data.
---
# ❺ Usage Instructions
<pre><code>
import os
import os.path as osp
import sys
import fire
import json
from typing import List, Union
import pandas as pd
import torch
from torch.nn import functional as F
import transformers
from transformers import TrainerCallback, TrainingArguments, TrainerState, TrainerControl, BitsAndBytesConfig
from transformers.trainer_utils import PREFIX_CHECKPOINT_DIR
from transformers import LlamaForCausalLM, LlamaTokenizer
from transformers import AutoModelForCausalLM, AutoTokenizer
from datasets import load_dataset
from peft import (
LoraConfig,
get_peft_model,
set_peft_model_state_dict
)
from peft import PeftModel
import re
import ast
device = 'auto' #@param {type: "string"}
model = '' #@param {type: "string"}
model = AutoModelForCausalLM.from_pretrained(
model,
quantization_config=bnb_config,
#load_in_4bit=True, # Quantization Load
device_map=device)
tokenizer = AutoTokenizer.from_pretrained(base_LLM_model)
input_text = "안녕하세요."
inputs = tokenizer(input_text, return_tensors="pt")
inputs = inputs.to("cuda:0")
with torch.no_grad():
outputs = model.generate(**inputs, max_length=1024)
result = tokenizer.decode(outputs[0], skip_special_tokens=True)
</code></pre>
## KTDS plans to provide fine-tuned LLMs (Large Language Models) across various domains of Korean culture and knowledge,
## including models based on not only OpenChat but also LLaMA, Polyglot, and EEVE.
## These models will be tailored to better understand and generate content specific to Korean contexts.
| null |
Non_BioNLP
|
### ⛱ ktdsbaseLM v0.11은 openchat3.5를 Foundation 모델로 하는 한국어 및 한국의 다양한
### 문화에 적용할 수 있도록 하기 위해
### 개발 되었으며 자체 제작한 53영역의 한국어 데이터를 활용하여 한국 사회 가치와
### 문화를 이해하는 모델 입니다. ✌
# ❶ 모델 설명
- 모델명 및 주요기능:
KTDSbaseLM v0.11은 OpenChat 3.5 모델을 기반으로 SFT 방식으로 파인튜닝된 Mistral 7B / openchat3.5 기반 모델입니다.
한국어와 한국의 다양한 문화적 맥락을 이해하도록 설계되었으며 ✨✨, 자체 제작한 135개 영역의 한국어
데이터를 활용해 한국 사회의 가치와 문화를 반영합니다.
주요 기능으로는 텍스트 생성, 대화 추론, 문서 요약, 질의응답, 감정 분석 및 자연어 처리 관련 다양한 작업을 지원하며,
활용 분야는 법률, 재무, 과학, 교육, 비즈니스, 문화 연구 등 다양한 분야에서 응용될 수 있습니다.
- 모델 아키텍처: KTDSBaseLM v0.11은 Mistral 7B 모델을 기반으로, 파라미터 수는 70억 개(7B)로 구성된 고성능 언어 모델입니다.
이 모델은 OpenChat 3.5를 파운데이션 모델로 삼아, SFT(지도 미세 조정) 방식을 통해 한국어와 한국 문화에 특화된 성능을 발휘하도록 훈련되었습니다.
Mistral 7B의 경량화된 구조는 빠른 추론 속도와 메모리 효율성을 보장하며, 다양한 자연어 처리 작업에 적합하게 최적화되어 있습니다.
이 아키텍처는 텍스트 생성, 질의응답, 문서 요약, 감정 분석과 같은 다양한 작업에서 탁월한 성능을 보여줍니다.
# ❷ 학습 데이터
- ktdsbaseLM v0.11은 자체 개발한 총 3.6GB 크기의 데이터를 바탕으로 학습되었습니다. 모두 233만 건의 QnA, 요약, 분류 등 데이터를 포함하며,
그 중 133만 건은 53개 영역의 객관식 문제로 구성되었습니다. 이 영역에는 한국사, 사회, 재무, 법률, 세무, 수학, 생물, 물리, 화학 등이 포함되며,
Chain of Thought 방식으로 학습되었습니다. 또한 130만 건의 주관식 문제는 한국사, 재무, 법률, 세무, 수학 등 38개 영역에 걸쳐 학습되었습니다.
학습 데이터 중 한국의 사회 가치와 인간의 감정을 이해하고 지시한 사항에 따라 출력할 수 있는 데이터를 학습하였습니다.
# ❸ 사용 사례
ktdsbaseLM v0.11은 다양한 응용 분야에서 사용될 수 있습니다. 예를 들어:
- 교육 분야: 역사, 수학, 과학 등 다양한 학습 자료에 대한 질의응답 및 설명 생성.
- 비즈니스: 법률, 재무, 세무 관련 질의에 대한 답변 제공 및 문서 요약.
- 연구 및 문화: 한국 사회와 문화에 맞춘 자연어 처리 작업, 감정 분석, 문서 생성 및 번역.
- 고객 서비스: 사용자와의 대화 생성 및 맞춤형 응답 제공.
- 이 모델은 다양한 자연어 처리 작업에서 높은 활용도를 가집니다.
# ❹ 한계 ⛈⛈
- ktdsBaseLM v0.11은 한국어와 한국 문화에 특화되어 있으나,
특정 영역(예: 최신 국제 자료, 전문 분야)의 데이터 부족으로 인해 다른 언어 또는
문화에 대한 응답의 정확성이 떨어질 수 있습니다.
또한, 복잡한 논리적 사고를 요구하는 문제에 대해 제한된 추론 능력을 보일 수 있으며,
편향된 데이터가 포함될 경우 편향된 응답이 생성될 가능성도 존재합니다.
# ❺ 사용 방법
<pre><code>
import os
import os.path as osp
import sys
import fire
import json
from typing import List, Union
import pandas as pd
import torch
from torch.nn import functional as F
import transformers
from transformers import TrainerCallback, TrainingArguments, TrainerState, TrainerControl, BitsAndBytesConfig
from transformers.trainer_utils import PREFIX_CHECKPOINT_DIR
from transformers import LlamaForCausalLM, LlamaTokenizer
from transformers import AutoModelForCausalLM, AutoTokenizer
from datasets import load_dataset
from peft import (
LoraConfig,
get_peft_model,
set_peft_model_state_dict
)
from peft import PeftModel
import re
import ast
device = 'auto' #@param {type: "string"}
model = '' #@param {type: "string"}
model = AutoModelForCausalLM.from_pretrained(
model,
quantization_config=bnb_config,
#load_in_4bit=True, # Quantization Load
device_map=device)
tokenizer = AutoTokenizer.from_pretrained(base_LLM_model)
input_text = "안녕하세요."
inputs = tokenizer(input_text, return_tensors="pt")
inputs = inputs.to("cuda:0")
with torch.no_grad():
outputs = model.generate(**inputs, max_length=1024)
result = tokenizer.decode(outputs[0], skip_special_tokens=True)
</code></pre>
## ✅ ktds는 openchat 외에도 LlaMA, Polyglot, EEVE 등 대표적인 LLM에 다양한 영역의 한국의 문화와 지식을 파인튜닝한 LLM을 제공할 예정입니다.
---
Here’s the English version of the provided text:
# ❶ Model Description
**Model Name and Key Features**:
KTDSbaseLM v0.11 is based on the OpenChat 3.5 model, fine-tuned using the SFT method on the Mistral 7B model.
It is designed to understand Korean and various cultural contexts, utilizing data from 135 domains in Korean society.
The model supports tasks such as text generation, conversation inference, document summarization,
question answering, sentiment analysis, and other NLP tasks.
Its applications span fields like law, finance, science, education, business, and cultural research.
**Model Architecture**:
KTDSBaseLM v0.11 is a high-performance language model with 7 billion parameters based on the Mistral 7B model.
It uses OpenChat 3.5 as the foundation and is fine-tuned using SFT to excel in Korean language and culture.
The streamlined Mistral 7B architecture ensures fast inference and memory efficiency,
optimized for various NLP tasks like text generation, question answering, document summarization, and sentiment analysis.
---
# ❷ Training Data
KTDSbaseLM v0.11 was trained on 3.6GB of data, comprising 2.33 million Q&A instances.
This includes 1.33 million multiple-choice questions across 53 domains such as history,
finance, law, tax, and science, trained with the Chain of Thought method. Additionally,
1.3 million short-answer questions cover 38 domains including history, finance, and law.
**Training Instruction Dataset Format**:
`{"prompt": "prompt text", "completion": "ideal generated text"}`
---
# ❸ Use Cases
KTDSbaseLM v0.11 can be used across multiple fields, such as:
- **Education**: Answering questions and generating explanations for subjects like history, math, and science.
- **Business**: Providing responses and summaries for legal, financial, and tax-related queries.
- **Research and Culture**: Performing NLP tasks, sentiment analysis, document generation, and translation.
- **Customer Service**: Generating conversations and personalized responses for users.
This model is highly versatile in various NLP tasks.
---
# ❹ Limitations
KTDSBaseLM v0.11 is specialized in Korean language and culture.
However, it may lack accuracy in responding to topics outside its scope,
such as international or specialized data.
Additionally, it may have limited reasoning ability for complex logical problems and
may produce biased responses if trained on biased data.
---
# ❺ Usage Instructions
<pre><code>
import os
import os.path as osp
import sys
import fire
import json
from typing import List, Union
import pandas as pd
import torch
from torch.nn import functional as F
import transformers
from transformers import TrainerCallback, TrainingArguments, TrainerState, TrainerControl, BitsAndBytesConfig
from transformers.trainer_utils import PREFIX_CHECKPOINT_DIR
from transformers import LlamaForCausalLM, LlamaTokenizer
from transformers import AutoModelForCausalLM, AutoTokenizer
from datasets import load_dataset
from peft import (
LoraConfig,
get_peft_model,
set_peft_model_state_dict
)
from peft import PeftModel
import re
import ast
device = 'auto' #@param {type: "string"}
model = '' #@param {type: "string"}
model = AutoModelForCausalLM.from_pretrained(
model,
quantization_config=bnb_config,
#load_in_4bit=True, # Quantization Load
device_map=device)
tokenizer = AutoTokenizer.from_pretrained(base_LLM_model)
input_text = "안녕하세요."
inputs = tokenizer(input_text, return_tensors="pt")
inputs = inputs.to("cuda:0")
with torch.no_grad():
outputs = model.generate(**inputs, max_length=1024)
result = tokenizer.decode(outputs[0], skip_special_tokens=True)
</code></pre>
## KTDS plans to provide fine-tuned LLMs (Large Language Models) across various domains of Korean culture and knowledge,
## including models based on not only OpenChat but also LLaMA, Polyglot, and EEVE.
## These models will be tailored to better understand and generate content specific to Korean contexts.
|
{"base_model": ["openchat/openchat_3.5"], "datasets": ["AIDX-ktds/ko_leaderboard"], "language": ["ko", "en"], "library_name": "adapter-transformers", "license": "mit", "metrics": ["accuracy"], "pipeline_tag": "text-generation", "tags": ["finance", "biology", "legal", "art", "text-generation-inference"]}
|
task
|
[
"QUESTION_ANSWERING",
"TRANSLATION",
"SUMMARIZATION"
] | 45,193 |
ayoubkirouane/billsum-BART-base-cnn
|
ayoubkirouane
|
summarization
|
[
"transformers",
"pytorch",
"bart",
"text2text-generation",
"summarization",
"en",
"dataset:billsum",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-09-28T12:11:45Z |
2023-09-28T12:56:18+00:00
| 21 | 1 |
---
datasets:
- billsum
language:
- en
library_name: transformers
license: apache-2.0
pipeline_tag: summarization
---
+ **Model Name**: billsum-BART-base-cnn
## **Description**:
This model is based on BART (Bidirectional and Auto-Regressive Transformers), originally introduced in the paper "BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension" by Lewis et al. It has been fine-tuned on the CNN Daily Mail dataset for the task of text summarization, specifically on the BillSum dataset, which consists of summaries of US Congressional and California state bills.
## Model Architecture:
BART is a transformer-based encoder-decoder (seq2seq) model with a bidirectional encoder and an autoregressive decoder. It excels in text generation tasks such as summarization and translation and is effective for comprehension tasks like text classification and question answering.
## Dataset Used:
The model has been fine-tuned on the BillSum dataset, which includes the following features:
+ **text**: The bill text.
+ **summary**: A summary of the bills.
+ **title**: The title of the bills (available for US bills only).
+ **text_len**: The number of characters in the text.
+ **sum_len**: The number of characters in the summary.
The data was collected from various sources, including the United States Government Publishing Office (GPO) and the California legislature's website.
## Uses:
+ **Text Summarization**:
This model can be used to generate concise summaries of longer text documents, making it suitable for applications like news article summarization, document summarization, and more.
## Limitations:
+ **Data Dependency**: The model's performance heavily relies on the quality and diversity of the training data. Fine-tuning on specific datasets may lead to biases or limitations inherent to those datasets.
+ **Length Constraints**: Like many sequence-to-sequence models, BART has length constraints. Longer input texts may result in truncated or incomplete summaries.
+ **Domain Specificity**: While fine-tuned on bill summaries, the model may not generalize well to other domains without further fine-tuning.
## Ethical Considerations:
+ **Bias**: Models like BART can inherit biases present in their training data. Care should be taken to evaluate and mitigate biases in generated content, especially when dealing with legal or legislative documents.
+ **Privacy**: When summarizing text, ensure that sensitive or private information is not inadvertently disclosed in the summaries.
+ **Accessibility**: Consider making model outputs accessible to individuals with disabilities, such as providing summaries in accessible formats.
## Usage :
```python
from transformers import pipeline
# Create a text generation pipeline with the specified model
pipe = pipeline("text2text-generation", model="ayoubkirouane/billsum-bart-base")
# Input text for summarization
input_text = """
Shields a business entity from civil liability relating to any injury or death occurring at a facility of that entity in connection with a use of such facility by a nonprofit organization if: (1) the use occurs outside the scope of business of the business entity; (2) such injury or death occurs during a period that such facility is used by such organization; and (3) the business entity authorized the use of such facility by the organization. Makes this Act inapplicable to an injury or death that results from an act or omission of a business entity that constitutes gross negligence or intentional misconduct, including misconduct that: (1) constitutes a hate crime or a crime of violence or act of international terrorism for which the defendant has been convicted in any court; or (2) involves a sexual offense for which the defendant has been convicted in any court or misconduct for which the defendant has been found to have violated a Federal or State civil rights law. Preempts State laws to the extent that such laws are inconsistent with this Act, except State law that provides additional protection from liability. Specifies that this Act shall not be construed to supersede any Federal or State health or safety law. Makes this Act inapplicable to any civil action in a State court against a business entity in which all parties are citizens of the State if such State, citing this Act's authority and containing no other provision, enacts a statute declaring the State's election that this Act shall not apply to such action in the State.
"""
# Generate the summary
summary = pipe(input_text, max_length=1024)
# Print the generated summary
print(summary[0]['summary_text'])
```
| null |
Non_BioNLP
|
+ **Model Name**: billsum-BART-base-cnn
## **Description**:
This model is based on BART (Bidirectional and Auto-Regressive Transformers), originally introduced in the paper "BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension" by Lewis et al. It has been fine-tuned on the CNN Daily Mail dataset for the task of text summarization, specifically on the BillSum dataset, which consists of summaries of US Congressional and California state bills.
## Model Architecture:
BART is a transformer-based encoder-decoder (seq2seq) model with a bidirectional encoder and an autoregressive decoder. It excels in text generation tasks such as summarization and translation and is effective for comprehension tasks like text classification and question answering.
## Dataset Used:
The model has been fine-tuned on the BillSum dataset, which includes the following features:
+ **text**: The bill text.
+ **summary**: A summary of the bills.
+ **title**: The title of the bills (available for US bills only).
+ **text_len**: The number of characters in the text.
+ **sum_len**: The number of characters in the summary.
The data was collected from various sources, including the United States Government Publishing Office (GPO) and the California legislature's website.
## Uses:
+ **Text Summarization**:
This model can be used to generate concise summaries of longer text documents, making it suitable for applications like news article summarization, document summarization, and more.
## Limitations:
+ **Data Dependency**: The model's performance heavily relies on the quality and diversity of the training data. Fine-tuning on specific datasets may lead to biases or limitations inherent to those datasets.
+ **Length Constraints**: Like many sequence-to-sequence models, BART has length constraints. Longer input texts may result in truncated or incomplete summaries.
+ **Domain Specificity**: While fine-tuned on bill summaries, the model may not generalize well to other domains without further fine-tuning.
## Ethical Considerations:
+ **Bias**: Models like BART can inherit biases present in their training data. Care should be taken to evaluate and mitigate biases in generated content, especially when dealing with legal or legislative documents.
+ **Privacy**: When summarizing text, ensure that sensitive or private information is not inadvertently disclosed in the summaries.
+ **Accessibility**: Consider making model outputs accessible to individuals with disabilities, such as providing summaries in accessible formats.
## Usage :
```python
from transformers import pipeline
# Create a text generation pipeline with the specified model
pipe = pipeline("text2text-generation", model="ayoubkirouane/billsum-bart-base")
# Input text for summarization
input_text = """
Shields a business entity from civil liability relating to any injury or death occurring at a facility of that entity in connection with a use of such facility by a nonprofit organization if: (1) the use occurs outside the scope of business of the business entity; (2) such injury or death occurs during a period that such facility is used by such organization; and (3) the business entity authorized the use of such facility by the organization. Makes this Act inapplicable to an injury or death that results from an act or omission of a business entity that constitutes gross negligence or intentional misconduct, including misconduct that: (1) constitutes a hate crime or a crime of violence or act of international terrorism for which the defendant has been convicted in any court; or (2) involves a sexual offense for which the defendant has been convicted in any court or misconduct for which the defendant has been found to have violated a Federal or State civil rights law. Preempts State laws to the extent that such laws are inconsistent with this Act, except State law that provides additional protection from liability. Specifies that this Act shall not be construed to supersede any Federal or State health or safety law. Makes this Act inapplicable to any civil action in a State court against a business entity in which all parties are citizens of the State if such State, citing this Act's authority and containing no other provision, enacts a statute declaring the State's election that this Act shall not apply to such action in the State.
"""
# Generate the summary
summary = pipe(input_text, max_length=1024)
# Print the generated summary
print(summary[0]['summary_text'])
```
|
{"datasets": ["billsum"], "language": ["en"], "library_name": "transformers", "license": "apache-2.0", "pipeline_tag": "summarization"}
|
task
|
[
"TEXT_CLASSIFICATION",
"QUESTION_ANSWERING",
"TRANSLATION",
"SUMMARIZATION"
] | 45,194 |
syndi-models/bart-large-cnn
|
syndi-models
|
summarization
|
[
"transformers",
"pytorch",
"tf",
"jax",
"rust",
"bart",
"text2text-generation",
"summarization",
"en",
"dataset:cnn_dailymail",
"arxiv:1910.13461",
"license:mit",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-05-09T18:54:09Z |
2023-01-24T16:28:55+00:00
| 199 | 1 |
---
datasets:
- cnn_dailymail
language:
- en
license: mit
tags:
- summarization
thumbnail: https://huggingface.co/front/thumbnails/facebook.png
model-index:
- name: facebook/bart-large-cnn
results:
- task:
type: summarization
name: Summarization
dataset:
name: cnn_dailymail
type: cnn_dailymail
config: 3.0.0
split: train
metrics:
- type: rouge
value: 42.9486
name: ROUGE-1
verified: true
- type: rouge
value: 20.8149
name: ROUGE-2
verified: true
- type: rouge
value: 30.6186
name: ROUGE-L
verified: true
- type: rouge
value: 40.0376
name: ROUGE-LSUM
verified: true
- type: loss
value: 2.529000997543335
name: loss
verified: true
- type: gen_len
value: 78.5866
name: gen_len
verified: true
---
# BART (large-sized model), fine-tuned on CNN Daily Mail
BART model pre-trained on English language, and fine-tuned on [CNN Daily Mail](https://huggingface.co/datasets/cnn_dailymail). It was introduced in the paper [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://arxiv.org/abs/1910.13461) by Lewis et al. and first released in [this repository (https://github.com/pytorch/fairseq/tree/master/examples/bart).
Disclaimer: The team releasing BART did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
BART is a transformer encoder-encoder (seq2seq) model with a bidirectional (BERT-like) encoder and an autoregressive (GPT-like) decoder. BART is pre-trained by (1) corrupting text with an arbitrary noising function, and (2) learning a model to reconstruct the original text.
BART is particularly effective when fine-tuned for text generation (e.g. summarization, translation) but also works well for comprehension tasks (e.g. text classification, question answering). This particular checkpoint has been fine-tuned on CNN Daily Mail, a large collection of text-summary pairs.
## Intended uses & limitations
You can use this model for text summarization.
### How to use
Here is how to use this model with the [pipeline API](https://huggingface.co/transformers/main_classes/pipelines.html):
```python
from transformers import pipeline
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
ARTICLE = """ New York (CNN)When Liana Barrientos was 23 years old, she got married in Westchester County, New York.
A year later, she got married again in Westchester County, but to a different man and without divorcing her first husband.
Only 18 days after that marriage, she got hitched yet again. Then, Barrientos declared "I do" five more times, sometimes only within two weeks of each other.
In 2010, she married once more, this time in the Bronx. In an application for a marriage license, she stated it was her "first and only" marriage.
Barrientos, now 39, is facing two criminal counts of "offering a false instrument for filing in the first degree," referring to her false statements on the
2010 marriage license application, according to court documents.
Prosecutors said the marriages were part of an immigration scam.
On Friday, she pleaded not guilty at State Supreme Court in the Bronx, according to her attorney, Christopher Wright, who declined to comment further.
After leaving court, Barrientos was arrested and charged with theft of service and criminal trespass for allegedly sneaking into the New York subway through an emergency exit, said Detective
Annette Markowski, a police spokeswoman. In total, Barrientos has been married 10 times, with nine of her marriages occurring between 1999 and 2002.
All occurred either in Westchester County, Long Island, New Jersey or the Bronx. She is believed to still be married to four men, and at one time, she was married to eight men at once, prosecutors say.
Prosecutors said the immigration scam involved some of her husbands, who filed for permanent residence status shortly after the marriages.
Any divorces happened only after such filings were approved. It was unclear whether any of the men will be prosecuted.
The case was referred to the Bronx District Attorney\'s Office by Immigration and Customs Enforcement and the Department of Homeland Security\'s
Investigation Division. Seven of the men are from so-called "red-flagged" countries, including Egypt, Turkey, Georgia, Pakistan and Mali.
Her eighth husband, Rashid Rajput, was deported in 2006 to his native Pakistan after an investigation by the Joint Terrorism Task Force.
If convicted, Barrientos faces up to four years in prison. Her next court appearance is scheduled for May 18.
"""
print(summarizer(ARTICLE, max_length=130, min_length=30, do_sample=False))
>>> [{'summary_text': 'Liana Barrientos, 39, is charged with two counts of "offering a false instrument for filing in the first degree" In total, she has been married 10 times, with nine of her marriages occurring between 1999 and 2002. She is believed to still be married to four men.'}]
```
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-1910-13461,
author = {Mike Lewis and
Yinhan Liu and
Naman Goyal and
Marjan Ghazvininejad and
Abdelrahman Mohamed and
Omer Levy and
Veselin Stoyanov and
Luke Zettlemoyer},
title = {{BART:} Denoising Sequence-to-Sequence Pre-training for Natural Language
Generation, Translation, and Comprehension},
journal = {CoRR},
volume = {abs/1910.13461},
year = {2019},
url = {http://arxiv.org/abs/1910.13461},
eprinttype = {arXiv},
eprint = {1910.13461},
timestamp = {Thu, 31 Oct 2019 14:02:26 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-1910-13461.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
| null |
Non_BioNLP
|
# BART (large-sized model), fine-tuned on CNN Daily Mail
BART model pre-trained on English language, and fine-tuned on [CNN Daily Mail](https://huggingface.co/datasets/cnn_dailymail). It was introduced in the paper [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://arxiv.org/abs/1910.13461) by Lewis et al. and first released in [this repository (https://github.com/pytorch/fairseq/tree/master/examples/bart).
Disclaimer: The team releasing BART did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
BART is a transformer encoder-encoder (seq2seq) model with a bidirectional (BERT-like) encoder and an autoregressive (GPT-like) decoder. BART is pre-trained by (1) corrupting text with an arbitrary noising function, and (2) learning a model to reconstruct the original text.
BART is particularly effective when fine-tuned for text generation (e.g. summarization, translation) but also works well for comprehension tasks (e.g. text classification, question answering). This particular checkpoint has been fine-tuned on CNN Daily Mail, a large collection of text-summary pairs.
## Intended uses & limitations
You can use this model for text summarization.
### How to use
Here is how to use this model with the [pipeline API](https://huggingface.co/transformers/main_classes/pipelines.html):
```python
from transformers import pipeline
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
ARTICLE = """ New York (CNN)When Liana Barrientos was 23 years old, she got married in Westchester County, New York.
A year later, she got married again in Westchester County, but to a different man and without divorcing her first husband.
Only 18 days after that marriage, she got hitched yet again. Then, Barrientos declared "I do" five more times, sometimes only within two weeks of each other.
In 2010, she married once more, this time in the Bronx. In an application for a marriage license, she stated it was her "first and only" marriage.
Barrientos, now 39, is facing two criminal counts of "offering a false instrument for filing in the first degree," referring to her false statements on the
2010 marriage license application, according to court documents.
Prosecutors said the marriages were part of an immigration scam.
On Friday, she pleaded not guilty at State Supreme Court in the Bronx, according to her attorney, Christopher Wright, who declined to comment further.
After leaving court, Barrientos was arrested and charged with theft of service and criminal trespass for allegedly sneaking into the New York subway through an emergency exit, said Detective
Annette Markowski, a police spokeswoman. In total, Barrientos has been married 10 times, with nine of her marriages occurring between 1999 and 2002.
All occurred either in Westchester County, Long Island, New Jersey or the Bronx. She is believed to still be married to four men, and at one time, she was married to eight men at once, prosecutors say.
Prosecutors said the immigration scam involved some of her husbands, who filed for permanent residence status shortly after the marriages.
Any divorces happened only after such filings were approved. It was unclear whether any of the men will be prosecuted.
The case was referred to the Bronx District Attorney\'s Office by Immigration and Customs Enforcement and the Department of Homeland Security\'s
Investigation Division. Seven of the men are from so-called "red-flagged" countries, including Egypt, Turkey, Georgia, Pakistan and Mali.
Her eighth husband, Rashid Rajput, was deported in 2006 to his native Pakistan after an investigation by the Joint Terrorism Task Force.
If convicted, Barrientos faces up to four years in prison. Her next court appearance is scheduled for May 18.
"""
print(summarizer(ARTICLE, max_length=130, min_length=30, do_sample=False))
>>> [{'summary_text': 'Liana Barrientos, 39, is charged with two counts of "offering a false instrument for filing in the first degree" In total, she has been married 10 times, with nine of her marriages occurring between 1999 and 2002. She is believed to still be married to four men.'}]
```
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-1910-13461,
author = {Mike Lewis and
Yinhan Liu and
Naman Goyal and
Marjan Ghazvininejad and
Abdelrahman Mohamed and
Omer Levy and
Veselin Stoyanov and
Luke Zettlemoyer},
title = {{BART:} Denoising Sequence-to-Sequence Pre-training for Natural Language
Generation, Translation, and Comprehension},
journal = {CoRR},
volume = {abs/1910.13461},
year = {2019},
url = {http://arxiv.org/abs/1910.13461},
eprinttype = {arXiv},
eprint = {1910.13461},
timestamp = {Thu, 31 Oct 2019 14:02:26 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-1910-13461.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
{"datasets": ["cnn_dailymail"], "language": ["en"], "license": "mit", "tags": ["summarization"], "thumbnail": "https://huggingface.co/front/thumbnails/facebook.png", "model-index": [{"name": "facebook/bart-large-cnn", "results": [{"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "cnn_dailymail", "type": "cnn_dailymail", "config": "3.0.0", "split": "train"}, "metrics": [{"type": "rouge", "value": 42.9486, "name": "ROUGE-1", "verified": true}, {"type": "rouge", "value": 20.8149, "name": "ROUGE-2", "verified": true}, {"type": "rouge", "value": 30.6186, "name": "ROUGE-L", "verified": true}, {"type": "rouge", "value": 40.0376, "name": "ROUGE-LSUM", "verified": true}, {"type": "loss", "value": 2.529000997543335, "name": "loss", "verified": true}, {"type": "gen_len", "value": 78.5866, "name": "gen_len", "verified": true}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION",
"QUESTION_ANSWERING",
"TRANSLATION",
"SUMMARIZATION"
] | 45,195 |
Ransaka/gemma-2b-sinhala-translation-chatml
|
Ransaka
| null |
[
"peft",
"tensorboard",
"safetensors",
"trl",
"sft",
"generated_from_trainer",
"dataset:generator",
"base_model:google/gemma-2b",
"base_model:adapter:google/gemma-2b",
"license:gemma",
"region:us"
] | 2024-04-22T07:09:39Z |
2024-04-22T11:24:32+00:00
| 10 | 0 |
---
base_model: google/gemma-2b
datasets:
- generator
library_name: peft
license: gemma
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: gemma-2b-sinhala-translation-chatml
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# gemma-2b-sinhala-translation-chatml
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 3
### Framework versions
- PEFT 0.8.2
- Transformers 4.38.2
- Pytorch 2.1.2
- Datasets 2.16.1
- Tokenizers 0.15.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# gemma-2b-sinhala-translation-chatml
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 3
### Framework versions
- PEFT 0.8.2
- Transformers 4.38.2
- Pytorch 2.1.2
- Datasets 2.16.1
- Tokenizers 0.15.2
|
{"base_model": "google/gemma-2b", "datasets": ["generator"], "library_name": "peft", "license": "gemma", "tags": ["trl", "sft", "generated_from_trainer"], "model-index": [{"name": "gemma-2b-sinhala-translation-chatml", "results": []}]}
|
task
|
[
"TRANSLATION"
] | 45,196 |
samuellimabraz/t5-small-qlora-finetune-tweetsumm
|
samuellimabraz
| null |
[
"peft",
"safetensors",
"generated_from_trainer",
"dataset:Andyrasika/TweetSumm-tuned",
"base_model:google-t5/t5-small",
"base_model:adapter:google-t5/t5-small",
"license:apache-2.0",
"model-index",
"region:us"
] | 2024-08-26T23:12:49Z |
2024-08-26T23:12:51+00:00
| 10 | 0 |
---
base_model: google-t5/t5-small
datasets:
- Andyrasika/TweetSumm-tuned
library_name: peft
license: apache-2.0
metrics:
- rouge
- f1
- precision
- recall
tags:
- generated_from_trainer
model-index:
- name: t5-small-QLoRA-TweetSumm-1724713795
results:
- task:
type: summarization
name: Summarization
dataset:
name: Andyrasika/TweetSumm-tuned
type: Andyrasika/TweetSumm-tuned
metrics:
- type: rouge
value: 0.4298
name: Rouge1
- type: f1
value: 0.887
name: F1
- type: precision
value: 0.8838
name: Precision
- type: recall
value: 0.8904
name: Recall
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# t5-small-QLoRA-TweetSumm-1724713795
This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on the Andyrasika/TweetSumm-tuned dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0940
- Rouge1: 0.4298
- Rouge2: 0.1915
- Rougel: 0.3559
- Rougelsum: 0.3956
- Gen Len: 47.8091
- F1: 0.887
- Precision: 0.8838
- Recall: 0.8904
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | F1 | Precision | Recall |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|:------:|:---------:|:------:|
| 2.3641 | 1.0 | 110 | 2.2019 | 0.4172 | 0.1774 | 0.3518 | 0.386 | 47.7636 | 0.8828 | 0.8806 | 0.8852 |
| 2.2228 | 2.0 | 220 | 2.1040 | 0.419 | 0.1789 | 0.3477 | 0.3827 | 48.1182 | 0.8846 | 0.882 | 0.8875 |
| 2.0174 | 3.0 | 330 | 2.0940 | 0.4298 | 0.1915 | 0.3559 | 0.3956 | 47.8091 | 0.887 | 0.8838 | 0.8904 |
### Framework versions
- PEFT 0.12.1.dev0
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# t5-small-QLoRA-TweetSumm-1724713795
This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on the Andyrasika/TweetSumm-tuned dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0940
- Rouge1: 0.4298
- Rouge2: 0.1915
- Rougel: 0.3559
- Rougelsum: 0.3956
- Gen Len: 47.8091
- F1: 0.887
- Precision: 0.8838
- Recall: 0.8904
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | F1 | Precision | Recall |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|:------:|:---------:|:------:|
| 2.3641 | 1.0 | 110 | 2.2019 | 0.4172 | 0.1774 | 0.3518 | 0.386 | 47.7636 | 0.8828 | 0.8806 | 0.8852 |
| 2.2228 | 2.0 | 220 | 2.1040 | 0.419 | 0.1789 | 0.3477 | 0.3827 | 48.1182 | 0.8846 | 0.882 | 0.8875 |
| 2.0174 | 3.0 | 330 | 2.0940 | 0.4298 | 0.1915 | 0.3559 | 0.3956 | 47.8091 | 0.887 | 0.8838 | 0.8904 |
### Framework versions
- PEFT 0.12.1.dev0
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1
|
{"base_model": "google-t5/t5-small", "datasets": ["Andyrasika/TweetSumm-tuned"], "library_name": "peft", "license": "apache-2.0", "metrics": ["rouge", "f1", "precision", "recall"], "tags": ["generated_from_trainer"], "model-index": [{"name": "t5-small-QLoRA-TweetSumm-1724713795", "results": [{"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "Andyrasika/TweetSumm-tuned", "type": "Andyrasika/TweetSumm-tuned"}, "metrics": [{"type": "rouge", "value": 0.4298, "name": "Rouge1"}, {"type": "f1", "value": 0.887, "name": "F1"}, {"type": "precision", "value": 0.8838, "name": "Precision"}, {"type": "recall", "value": 0.8904, "name": "Recall"}]}]}]}
|
task
|
[
"SUMMARIZATION"
] | 45,197 |
Shalu2108/t5
|
Shalu2108
|
summarization
|
[
"summarization",
"dataset:roneneldan/TinyStories",
"license:openrail",
"region:us"
] | 2023-08-15T02:22:22Z |
2023-08-15T02:24:45+00:00
| 0 | 0 |
---
datasets:
- roneneldan/TinyStories
license: openrail
metrics:
- character
- brier_score
pipeline_tag: summarization
---
| null |
Non_BioNLP
|
{"datasets": ["roneneldan/TinyStories"], "license": "openrail", "metrics": ["character", "brier_score"], "pipeline_tag": "summarization"}
|
task
|
[
"SUMMARIZATION"
] | 45,198 |
|
prithivMLmods/Cygnus-II-14B
|
prithivMLmods
|
text-generation
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"text-generation-inference",
"trl",
"code",
"cygnus",
"Qwen",
"conversational",
"en",
"zh",
"base_model:prithivMLmods/Coma-II-14B",
"base_model:finetune:prithivMLmods/Coma-II-14B",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2025-03-02T07:08:47Z |
2025-03-24T15:39:13+00:00
| 937 | 2 |
---
base_model:
- prithivMLmods/Coma-II-14B
language:
- en
- zh
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
tags:
- text-generation-inference
- trl
- code
- cygnus
- Qwen
model-index:
- name: Cygnus-II-14B
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: IFEval (0-Shot)
type: wis-k/instruction-following-eval
split: train
args:
num_few_shot: 0
metrics:
- type: inst_level_strict_acc and prompt_level_strict_acc
value: 61.84
name: averaged accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BBH (3-Shot)
type: SaylorTwift/bbh
split: test
args:
num_few_shot: 3
metrics:
- type: acc_norm
value: 52.14
name: normalized accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MATH Lvl 5 (4-Shot)
type: lighteval/MATH-Hard
split: test
args:
num_few_shot: 4
metrics:
- type: exact_match
value: 43.96
name: exact match
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GPQA (0-shot)
type: Idavidrein/gpqa
split: train
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 18.34
name: acc_norm
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MuSR (0-shot)
type: TAUR-Lab/MuSR
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 18.11
name: acc_norm
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU-PRO (5-shot)
type: TIGER-Lab/MMLU-Pro
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 48.78
name: accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B
name: Open LLM Leaderboard
---
# **Cygnus-II-14B**
> Cygnus-II-14B is based on the Qwen 2.5 14B modality architecture, designed to enhance the reasoning capabilities of 14B-parameter models. This model is optimized for general-purpose reasoning and answering, excelling in contextual understanding, logical deduction, and multi-step problem-solving. It has been fine-tuned using a long chain-of-thought reasoning model and specialized datasets to improve comprehension, structured responses, and conversational intelligence.
## **Key Improvements**
1. **Enhanced General Knowledge**: The model provides broad knowledge across various domains, improving capabilities in answering questions accurately and generating coherent responses.
2. **Improved Instruction Following**: Significant advancements in understanding and following complex instructions, generating structured responses, and maintaining coherence over extended interactions.
3. **Versatile Adaptability**: More resilient to diverse prompts, enhancing its ability to handle a wide range of topics and conversation styles, including open-ended and structured inquiries.
4. **Long-Context Support**: Supports up to 128K tokens for input context and can generate up to 8K tokens in a single output, making it ideal for detailed responses.
5. **Multilingual Proficiency**: Supports over 29 languages, including English, Chinese, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
## **Quickstart with transformers**
Here is a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and generate content:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "prithivMLmods/Cygnus-II-14B"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "What are the key principles of general-purpose AI?"
messages = [
{"role": "system", "content": "You are a helpful assistant capable of answering a wide range of questions."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
## **Intended Use**
1. **General-Purpose Reasoning**:
Designed for broad applicability, assisting with logical reasoning, answering diverse questions, and solving general knowledge problems.
2. **Educational and Informational Assistance**:
Suitable for providing explanations, summaries, and research-based responses for students, educators, and general users.
3. **Conversational AI and Chatbots**:
Ideal for building intelligent conversational agents that require contextual understanding and dynamic response generation.
4. **Multilingual Applications**:
Supports global communication, translations, and multilingual content generation.
5. **Structured Data Processing**:
Capable of analyzing and generating structured outputs, such as tables and JSON, useful for data science and automation.
6. **Long-Form Content Generation**:
Can generate extended responses, including articles, reports, and guides, maintaining coherence over large text outputs.
## **Limitations**
1. **Hardware Requirements**:
Requires high-memory GPUs or TPUs due to its large parameter size and long-context support.
2. **Potential Bias in Responses**:
While designed to be neutral, outputs may still reflect biases present in training data.
3. **Inconsistent Outputs in Creative Tasks**:
May produce variable results in storytelling and highly subjective topics.
4. **Limited Real-World Awareness**:
Does not have access to real-time events beyond its training cutoff.
5. **Error Propagation in Extended Outputs**:
Minor errors in early responses may affect overall coherence in long-form outputs.
6. **Prompt Sensitivity**:
The effectiveness of responses may depend on how well the input prompt is structured.
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/prithivMLmods__Cygnus-II-14B-details)!
Summarized results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/contents/viewer/default/train?q=prithivMLmods%2FCygnus-II-14B&sort[column]=Average%20%E2%AC%86%EF%B8%8F&sort[direction]=desc)!
| Metric |Value (%)|
|-------------------|--------:|
|**Average** | 40.53|
|IFEval (0-Shot) | 61.84|
|BBH (3-Shot) | 52.14|
|MATH Lvl 5 (4-Shot)| 43.96|
|GPQA (0-shot) | 18.34|
|MuSR (0-shot) | 18.11|
|MMLU-PRO (5-shot) | 48.78|
| null |
Non_BioNLP
|
# **Cygnus-II-14B**
> Cygnus-II-14B is based on the Qwen 2.5 14B modality architecture, designed to enhance the reasoning capabilities of 14B-parameter models. This model is optimized for general-purpose reasoning and answering, excelling in contextual understanding, logical deduction, and multi-step problem-solving. It has been fine-tuned using a long chain-of-thought reasoning model and specialized datasets to improve comprehension, structured responses, and conversational intelligence.
## **Key Improvements**
1. **Enhanced General Knowledge**: The model provides broad knowledge across various domains, improving capabilities in answering questions accurately and generating coherent responses.
2. **Improved Instruction Following**: Significant advancements in understanding and following complex instructions, generating structured responses, and maintaining coherence over extended interactions.
3. **Versatile Adaptability**: More resilient to diverse prompts, enhancing its ability to handle a wide range of topics and conversation styles, including open-ended and structured inquiries.
4. **Long-Context Support**: Supports up to 128K tokens for input context and can generate up to 8K tokens in a single output, making it ideal for detailed responses.
5. **Multilingual Proficiency**: Supports over 29 languages, including English, Chinese, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
## **Quickstart with transformers**
Here is a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and generate content:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "prithivMLmods/Cygnus-II-14B"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "What are the key principles of general-purpose AI?"
messages = [
{"role": "system", "content": "You are a helpful assistant capable of answering a wide range of questions."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
## **Intended Use**
1. **General-Purpose Reasoning**:
Designed for broad applicability, assisting with logical reasoning, answering diverse questions, and solving general knowledge problems.
2. **Educational and Informational Assistance**:
Suitable for providing explanations, summaries, and research-based responses for students, educators, and general users.
3. **Conversational AI and Chatbots**:
Ideal for building intelligent conversational agents that require contextual understanding and dynamic response generation.
4. **Multilingual Applications**:
Supports global communication, translations, and multilingual content generation.
5. **Structured Data Processing**:
Capable of analyzing and generating structured outputs, such as tables and JSON, useful for data science and automation.
6. **Long-Form Content Generation**:
Can generate extended responses, including articles, reports, and guides, maintaining coherence over large text outputs.
## **Limitations**
1. **Hardware Requirements**:
Requires high-memory GPUs or TPUs due to its large parameter size and long-context support.
2. **Potential Bias in Responses**:
While designed to be neutral, outputs may still reflect biases present in training data.
3. **Inconsistent Outputs in Creative Tasks**:
May produce variable results in storytelling and highly subjective topics.
4. **Limited Real-World Awareness**:
Does not have access to real-time events beyond its training cutoff.
5. **Error Propagation in Extended Outputs**:
Minor errors in early responses may affect overall coherence in long-form outputs.
6. **Prompt Sensitivity**:
The effectiveness of responses may depend on how well the input prompt is structured.
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/prithivMLmods__Cygnus-II-14B-details)!
Summarized results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/contents/viewer/default/train?q=prithivMLmods%2FCygnus-II-14B&sort[column]=Average%20%E2%AC%86%EF%B8%8F&sort[direction]=desc)!
| Metric |Value (%)|
|-------------------|--------:|
|**Average** | 40.53|
|IFEval (0-Shot) | 61.84|
|BBH (3-Shot) | 52.14|
|MATH Lvl 5 (4-Shot)| 43.96|
|GPQA (0-shot) | 18.34|
|MuSR (0-shot) | 18.11|
|MMLU-PRO (5-shot) | 48.78|
|
{"base_model": ["prithivMLmods/Coma-II-14B"], "language": ["en", "zh"], "library_name": "transformers", "license": "apache-2.0", "pipeline_tag": "text-generation", "tags": ["text-generation-inference", "trl", "code", "cygnus", "Qwen"], "model-index": [{"name": "Cygnus-II-14B", "results": [{"task": {"type": "text-generation", "name": "Text Generation"}, "dataset": {"name": "IFEval (0-Shot)", "type": "wis-k/instruction-following-eval", "split": "train", "args": {"num_few_shot": 0}}, "metrics": [{"type": "inst_level_strict_acc and prompt_level_strict_acc", "value": 61.84, "name": "averaged accuracy"}], "source": {"url": "https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B", "name": "Open LLM Leaderboard"}}, {"task": {"type": "text-generation", "name": "Text Generation"}, "dataset": {"name": "BBH (3-Shot)", "type": "SaylorTwift/bbh", "split": "test", "args": {"num_few_shot": 3}}, "metrics": [{"type": "acc_norm", "value": 52.14, "name": "normalized accuracy"}], "source": {"url": "https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B", "name": "Open LLM Leaderboard"}}, {"task": {"type": "text-generation", "name": "Text Generation"}, "dataset": {"name": "MATH Lvl 5 (4-Shot)", "type": "lighteval/MATH-Hard", "split": "test", "args": {"num_few_shot": 4}}, "metrics": [{"type": "exact_match", "value": 43.96, "name": "exact match"}], "source": {"url": "https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B", "name": "Open LLM Leaderboard"}}, {"task": {"type": "text-generation", "name": "Text Generation"}, "dataset": {"name": "GPQA (0-shot)", "type": "Idavidrein/gpqa", "split": "train", "args": {"num_few_shot": 0}}, "metrics": [{"type": "acc_norm", "value": 18.34, "name": "acc_norm"}], "source": {"url": "https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B", "name": "Open LLM Leaderboard"}}, {"task": {"type": "text-generation", "name": "Text Generation"}, "dataset": {"name": "MuSR (0-shot)", "type": "TAUR-Lab/MuSR", "args": {"num_few_shot": 0}}, "metrics": [{"type": "acc_norm", "value": 18.11, "name": "acc_norm"}], "source": {"url": "https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B", "name": "Open LLM Leaderboard"}}, {"task": {"type": "text-generation", "name": "Text Generation"}, "dataset": {"name": "MMLU-PRO (5-shot)", "type": "TIGER-Lab/MMLU-Pro", "config": "main", "split": "test", "args": {"num_few_shot": 5}}, "metrics": [{"type": "acc", "value": 48.78, "name": "accuracy"}], "source": {"url": "https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FCygnus-II-14B", "name": "Open LLM Leaderboard"}}]}]}
|
task
|
[
"TRANSLATION"
] | 45,199 |
sbollam/mt5-small-finetuned-amazon-en-es
|
sbollam
|
summarization
|
[
"transformers",
"tensorboard",
"safetensors",
"mt5",
"text2text-generation",
"summarization",
"generated_from_trainer",
"base_model:google/mt5-small",
"base_model:finetune:google/mt5-small",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-12-04T02:42:15Z |
2024-12-04T18:03:43+00:00
| 45 | 0 |
---
base_model: google/mt5-small
library_name: transformers
license: apache-2.0
metrics:
- rouge
tags:
- summarization
- generated_from_trainer
model-index:
- name: mt5-small-finetuned-amazon-en-es
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# mt5-small-finetuned-amazon-en-es
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.2659
- Rouge1: 13.7437
- Rouge2: 5.9153
- Rougel: 13.4146
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5.6e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 8
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:-------:|
| 9.4301 | 1.0 | 625 | 3.5851 | 8.6109 | 2.4991 | 8.3173 |
| 4.6907 | 2.0 | 1250 | 3.4105 | 11.1544 | 4.3475 | 10.7786 |
| 4.1916 | 3.0 | 1875 | 3.3443 | 10.8192 | 3.848 | 10.44 |
| 3.971 | 4.0 | 2500 | 3.3001 | 12.7096 | 5.0292 | 12.3085 |
| 3.8402 | 5.0 | 3125 | 3.2788 | 12.1184 | 4.5893 | 11.7314 |
| 3.7106 | 6.0 | 3750 | 3.2795 | 13.748 | 5.907 | 13.5413 |
| 3.6523 | 7.0 | 4375 | 3.2702 | 13.6669 | 5.8956 | 13.3843 |
| 3.6315 | 8.0 | 5000 | 3.2659 | 13.7437 | 5.9153 | 13.4146 |
### Framework versions
- Transformers 4.46.2
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# mt5-small-finetuned-amazon-en-es
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.2659
- Rouge1: 13.7437
- Rouge2: 5.9153
- Rougel: 13.4146
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5.6e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 8
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:-------:|
| 9.4301 | 1.0 | 625 | 3.5851 | 8.6109 | 2.4991 | 8.3173 |
| 4.6907 | 2.0 | 1250 | 3.4105 | 11.1544 | 4.3475 | 10.7786 |
| 4.1916 | 3.0 | 1875 | 3.3443 | 10.8192 | 3.848 | 10.44 |
| 3.971 | 4.0 | 2500 | 3.3001 | 12.7096 | 5.0292 | 12.3085 |
| 3.8402 | 5.0 | 3125 | 3.2788 | 12.1184 | 4.5893 | 11.7314 |
| 3.7106 | 6.0 | 3750 | 3.2795 | 13.748 | 5.907 | 13.5413 |
| 3.6523 | 7.0 | 4375 | 3.2702 | 13.6669 | 5.8956 | 13.3843 |
| 3.6315 | 8.0 | 5000 | 3.2659 | 13.7437 | 5.9153 | 13.4146 |
### Framework versions
- Transformers 4.46.2
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3
|
{"base_model": "google/mt5-small", "library_name": "transformers", "license": "apache-2.0", "metrics": ["rouge"], "tags": ["summarization", "generated_from_trainer"], "model-index": [{"name": "mt5-small-finetuned-amazon-en-es", "results": []}]}
|
task
|
[
"SUMMARIZATION"
] | 45,200 |
TransferGraph/ali2066_finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26-finetuned-lora-tweet_eval_emotion
|
TransferGraph
|
text-classification
|
[
"peft",
"safetensors",
"parquet",
"text-classification",
"dataset:tweet_eval",
"base_model:ali2066/finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26",
"base_model:adapter:ali2066/finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26",
"license:apache-2.0",
"model-index",
"region:us"
] | 2024-02-29T12:46:19Z |
2024-02-29T12:46:24+00:00
| 0 | 0 |
---
base_model: ali2066/finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26
datasets:
- tweet_eval
library_name: peft
license: apache-2.0
metrics:
- accuracy
tags:
- parquet
- text-classification
model-index:
- name: ali2066_finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26-finetuned-lora-tweet_eval_emotion
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: tweet_eval
type: tweet_eval
config: emotion
split: validation
args: emotion
metrics:
- type: accuracy
value: 0.7165775401069518
name: accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ali2066_finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26-finetuned-lora-tweet_eval_emotion
This model is a fine-tuned version of [ali2066/finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26](https://huggingface.co/ali2066/finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26) on the tweet_eval dataset.
It achieves the following results on the evaluation set:
- accuracy: 0.7166
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0004
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
### Training results
| accuracy | train_loss | epoch |
|:--------:|:----------:|:-----:|
| 0.2433 | None | 0 |
| 0.6203 | 1.0796 | 0 |
| 0.6925 | 0.8351 | 1 |
| 0.7139 | 0.7086 | 2 |
| 0.7166 | 0.6623 | 3 |
### Framework versions
- PEFT 0.8.2
- Transformers 4.37.2
- Pytorch 2.2.0
- Datasets 2.16.1
- Tokenizers 0.15.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ali2066_finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26-finetuned-lora-tweet_eval_emotion
This model is a fine-tuned version of [ali2066/finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26](https://huggingface.co/ali2066/finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26) on the tweet_eval dataset.
It achieves the following results on the evaluation set:
- accuracy: 0.7166
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0004
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
### Training results
| accuracy | train_loss | epoch |
|:--------:|:----------:|:-----:|
| 0.2433 | None | 0 |
| 0.6203 | 1.0796 | 0 |
| 0.6925 | 0.8351 | 1 |
| 0.7139 | 0.7086 | 2 |
| 0.7166 | 0.6623 | 3 |
### Framework versions
- PEFT 0.8.2
- Transformers 4.37.2
- Pytorch 2.2.0
- Datasets 2.16.1
- Tokenizers 0.15.2
|
{"base_model": "ali2066/finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26", "datasets": ["tweet_eval"], "library_name": "peft", "license": "apache-2.0", "metrics": ["accuracy"], "tags": ["parquet", "text-classification"], "model-index": [{"name": "ali2066_finetuned_sentence_itr1_2e-05_all_26_02_2022-04_03_26-finetuned-lora-tweet_eval_emotion", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "tweet_eval", "type": "tweet_eval", "config": "emotion", "split": "validation", "args": "emotion"}, "metrics": [{"type": "accuracy", "value": 0.7165775401069518, "name": "accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,201 |
cindyangelira/distilbert-base-uncased-finetuned-emotion
|
cindyangelira
|
text-classification
|
[
"transformers",
"tensorboard",
"safetensors",
"distilbert",
"text-classification",
"generated_from_trainer",
"dataset:emotion",
"base_model:distilbert/distilbert-base-uncased",
"base_model:finetune:distilbert/distilbert-base-uncased",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-03-21T07:41:18Z |
2024-03-21T07:59:58+00:00
| 9 | 0 |
---
base_model: distilbert-base-uncased
datasets:
- emotion
license: apache-2.0
metrics:
- accuracy
- f1
tags:
- generated_from_trainer
model-index:
- name: distilbert-base-uncased-finetuned-emotion
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: emotion
type: emotion
config: split
split: validation
args: split
metrics:
- type: accuracy
value: 0.9205
name: Accuracy
- type: f1
value: 0.9203895637603402
name: F1
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2255
- Accuracy: 0.9205
- F1: 0.9204
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 0.8319 | 1.0 | 250 | 0.3325 | 0.9025 | 0.9018 |
| 0.2531 | 2.0 | 500 | 0.2255 | 0.9205 | 0.9204 |
### Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2255
- Accuracy: 0.9205
- F1: 0.9204
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 0.8319 | 1.0 | 250 | 0.3325 | 0.9025 | 0.9018 |
| 0.2531 | 2.0 | 500 | 0.2255 | 0.9205 | 0.9204 |
### Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
|
{"base_model": "distilbert-base-uncased", "datasets": ["emotion"], "license": "apache-2.0", "metrics": ["accuracy", "f1"], "tags": ["generated_from_trainer"], "model-index": [{"name": "distilbert-base-uncased-finetuned-emotion", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "emotion", "type": "emotion", "config": "split", "split": "validation", "args": "split"}, "metrics": [{"type": "accuracy", "value": 0.9205, "name": "Accuracy"}, {"type": "f1", "value": 0.9203895637603402, "name": "F1"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,202 |
pinzhenchen/sft-lora-zh-pythia-12b
|
pinzhenchen
| null |
[
"generation",
"question answering",
"instruction tuning",
"zh",
"arxiv:2309.08958",
"license:cc-by-nc-4.0",
"region:us"
] | 2024-03-05T23:54:15Z |
2024-03-05T23:54:19+00:00
| 0 | 0 |
---
language:
- zh
license: cc-by-nc-4.0
tags:
- generation
- question answering
- instruction tuning
---
### Model Description
This HF repository contains base LLMs instruction tuned (SFT) with LoRA and then used to study whether monolingual or multilingual instruction tuning is more favourable.
* [GitHub](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main)
* [Paper](https://arxiv.org/abs/2309.08958)
#### Instruction tuning details
* Base model: [EleutherAI/pythia-12b-deduped](https://huggingface.co/EleutherAI/pythia-12b-deduped)
* Instruction tuning language: Chinese
* Training method: LoRA.
* LoRA details: rank=8, alpha=16, target modules={key, query, value}.
* Best checkpoint: best cross-entropy on a validation set, trained for 5 epochs.
* Dataset: machine-translated from [yahma/alpaca-cleaned](https://huggingface.co/datasets/yahma/alpaca-cleaned). You can download our data [HERE](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main/training-data).
#### Usage
The model checkpoint should be loaded with the base model together using `transformers` and `peft` libraries.
Please refer to our Github repository [HERE](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main/loraft) for inference and training instructions.
#### Citation
```
@inproceedings{chen-etal-2024-monolingual,
title="Monolingual or multilingual instruction tuning: Which makes a better {Alpaca}",
author="Pinzhen Chen and Shaoxiong Ji and Nikolay Bogoychev and Andrey Kutuzov and Barry Haddow and Kenneth Heafield",
year="2024",
booktitle = "Findings of the Association for Computational Linguistics: EACL 2024",
}
```
| null |
Non_BioNLP
|
### Model Description
This HF repository contains base LLMs instruction tuned (SFT) with LoRA and then used to study whether monolingual or multilingual instruction tuning is more favourable.
* [GitHub](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main)
* [Paper](https://arxiv.org/abs/2309.08958)
#### Instruction tuning details
* Base model: [EleutherAI/pythia-12b-deduped](https://huggingface.co/EleutherAI/pythia-12b-deduped)
* Instruction tuning language: Chinese
* Training method: LoRA.
* LoRA details: rank=8, alpha=16, target modules={key, query, value}.
* Best checkpoint: best cross-entropy on a validation set, trained for 5 epochs.
* Dataset: machine-translated from [yahma/alpaca-cleaned](https://huggingface.co/datasets/yahma/alpaca-cleaned). You can download our data [HERE](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main/training-data).
#### Usage
The model checkpoint should be loaded with the base model together using `transformers` and `peft` libraries.
Please refer to our Github repository [HERE](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main/loraft) for inference and training instructions.
#### Citation
```
@inproceedings{chen-etal-2024-monolingual,
title="Monolingual or multilingual instruction tuning: Which makes a better {Alpaca}",
author="Pinzhen Chen and Shaoxiong Ji and Nikolay Bogoychev and Andrey Kutuzov and Barry Haddow and Kenneth Heafield",
year="2024",
booktitle = "Findings of the Association for Computational Linguistics: EACL 2024",
}
```
|
{"language": ["zh"], "license": "cc-by-nc-4.0", "tags": ["generation", "question answering", "instruction tuning"]}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,203 |
abdulhade/fine-tuned-MarianMTKurdish
|
abdulhade
|
translation
|
[
"adapter-transformers",
"safetensors",
"marian",
"translation",
"ku",
"en",
"dataset:abdulhade/TextCorpusKurdish_asosoft",
"license:mit",
"region:us"
] | 2024-10-25T20:30:34Z |
2024-10-26T07:54:43+00:00
| 0 | 1 |
---
datasets:
- abdulhade/TextCorpusKurdish_asosoft
language:
- ku
- en
library_name: adapter-transformers
license: mit
metrics:
- accuracy
- bleu
- meteor
pipeline_tag: translation
---
# Kurdish-English Machine Translation with Transformers
This repository focuses on fine-tuning a Kurdish-English machine translation model using Hugging Face's `transformers` library with MarianMT.
The model is trained on a custom parallel corpus with a detailed pipeline that includes data preprocessing, bidirectional training, evaluation, and inference.
This model is a product of the AI Center of Kurdistan University.
## Table of Contents
- [Introduction](#introduction)
- [Requirements](#requirements)
- [Setup](#setup)
- [Pipeline Overview](#pipeline-overview)
- [Data Preparation](#data-preparation)
- [Training SentencePiece Tokenizer](#training-sentencepiece-tokenizer)
- [Model and Tokenizer Setup](#model-and-tokenizer-setup)
- [Tokenization and Dataset Preparation](#tokenization-and-dataset-preparation)
- [Training Configuration](#training-configuration)
- [Evaluation and Metrics](#evaluation-and-metrics)
- [Inference](#inference)
- [Results](#results)
- [License](#license)
## Introduction
This project fine-tunes a MarianMT model for Kurdish-English translation on a custom parallel corpus. Training is configured for bidirectional translation, enabling model use in both language directions.
## Requirements
- Python 3.8+
- Hugging Face Transformers
- Datasets library
- SentencePiece
- PyTorch 1.9+
- CUDA (for GPU support)
## Setup
1. Clone the repository and install dependencies.
2. Ensure GPU availability.
3. Prepare your Kurdish-English corpus in CSV format.
## Pipeline Overview
### Data Preparation
1. **Corpus**: A Kurdish-English parallel corpus in CSV format with columns `Source` (Kurdish) and `Target` (English).
2. **Path Definition**: Specify the corpus path in the configuration.
### Training SentencePiece Tokenizer
- **Vocabulary Size**: 32,000
- **Source Data**: The tokenizer is trained on both the primary Kurdish corpus and the English dataset to create shared subword tokens.
### Model and Tokenizer Setup
- **Model**: `Helsinki-NLP/opus-mt-en-mul` pre-trained MarianMT model.
- **Tokenizer**: MarianMT tokenizer aligned with the model, with source and target languages set dynamically.
### Tokenization and Dataset Preparation
- **Train-Validation Split**: 90% train, 10% validation split.
- **Maximum Sequence Length**: 128 tokens for both source and target sequences.
- **Bidirectional Tokenization**: Prepare tokenized sequences for both Kurdish-English and English-Kurdish translation.
### Training Configuration
- **Learning Rate**: 2e-5
- **Batch Size**: 4 (per device, for both training and evaluation)
- **Weight Decay**: 0.01
- **Evaluation Strategy**: Per epoch
- **Epochs**: 3
- **Logging**: Logs saved every 100 steps, with TensorBoard logging enabled
- **Output Directory**: `./results`
- **Device**: GPU 1 explicitly set
### Evaluation and Metrics
The following metrics are computed on the validation dataset:
- **BLEU**: Measures translation quality based on precision and recall of n-grams.
- **METEOR**: Considers synonymy and stem matches.
- **BERTScore**: Evaluates semantic similarity with BERT embeddings.
### Inference
Inference includes bidirectional translation capabilities:
- **Source to Target**: English to Kurdish translation.
- **Target to Source**: Kurdish to English translation.
## Results
The fine-tuned model and tokenizer are saved to `./fine-tuned-marianmt`, including evaluation metrics across BLEU, METEOR, and BERTScore.
"""
# Write the content to README.md
file_path = "/mnt/data/README.md"
with open(file_path, "w") as readme_file:
readme_file.write(readme_content)
file_path
| null |
Non_BioNLP
|
# Kurdish-English Machine Translation with Transformers
This repository focuses on fine-tuning a Kurdish-English machine translation model using Hugging Face's `transformers` library with MarianMT.
The model is trained on a custom parallel corpus with a detailed pipeline that includes data preprocessing, bidirectional training, evaluation, and inference.
This model is a product of the AI Center of Kurdistan University.
## Table of Contents
- [Introduction](#introduction)
- [Requirements](#requirements)
- [Setup](#setup)
- [Pipeline Overview](#pipeline-overview)
- [Data Preparation](#data-preparation)
- [Training SentencePiece Tokenizer](#training-sentencepiece-tokenizer)
- [Model and Tokenizer Setup](#model-and-tokenizer-setup)
- [Tokenization and Dataset Preparation](#tokenization-and-dataset-preparation)
- [Training Configuration](#training-configuration)
- [Evaluation and Metrics](#evaluation-and-metrics)
- [Inference](#inference)
- [Results](#results)
- [License](#license)
## Introduction
This project fine-tunes a MarianMT model for Kurdish-English translation on a custom parallel corpus. Training is configured for bidirectional translation, enabling model use in both language directions.
## Requirements
- Python 3.8+
- Hugging Face Transformers
- Datasets library
- SentencePiece
- PyTorch 1.9+
- CUDA (for GPU support)
## Setup
1. Clone the repository and install dependencies.
2. Ensure GPU availability.
3. Prepare your Kurdish-English corpus in CSV format.
## Pipeline Overview
### Data Preparation
1. **Corpus**: A Kurdish-English parallel corpus in CSV format with columns `Source` (Kurdish) and `Target` (English).
2. **Path Definition**: Specify the corpus path in the configuration.
### Training SentencePiece Tokenizer
- **Vocabulary Size**: 32,000
- **Source Data**: The tokenizer is trained on both the primary Kurdish corpus and the English dataset to create shared subword tokens.
### Model and Tokenizer Setup
- **Model**: `Helsinki-NLP/opus-mt-en-mul` pre-trained MarianMT model.
- **Tokenizer**: MarianMT tokenizer aligned with the model, with source and target languages set dynamically.
### Tokenization and Dataset Preparation
- **Train-Validation Split**: 90% train, 10% validation split.
- **Maximum Sequence Length**: 128 tokens for both source and target sequences.
- **Bidirectional Tokenization**: Prepare tokenized sequences for both Kurdish-English and English-Kurdish translation.
### Training Configuration
- **Learning Rate**: 2e-5
- **Batch Size**: 4 (per device, for both training and evaluation)
- **Weight Decay**: 0.01
- **Evaluation Strategy**: Per epoch
- **Epochs**: 3
- **Logging**: Logs saved every 100 steps, with TensorBoard logging enabled
- **Output Directory**: `./results`
- **Device**: GPU 1 explicitly set
### Evaluation and Metrics
The following metrics are computed on the validation dataset:
- **BLEU**: Measures translation quality based on precision and recall of n-grams.
- **METEOR**: Considers synonymy and stem matches.
- **BERTScore**: Evaluates semantic similarity with BERT embeddings.
### Inference
Inference includes bidirectional translation capabilities:
- **Source to Target**: English to Kurdish translation.
- **Target to Source**: Kurdish to English translation.
## Results
The fine-tuned model and tokenizer are saved to `./fine-tuned-marianmt`, including evaluation metrics across BLEU, METEOR, and BERTScore.
"""
# Write the content to README.md
file_path = "/mnt/data/README.md"
with open(file_path, "w") as readme_file:
readme_file.write(readme_content)
file_path
|
{"datasets": ["abdulhade/TextCorpusKurdish_asosoft"], "language": ["ku", "en"], "library_name": "adapter-transformers", "license": "mit", "metrics": ["accuracy", "bleu", "meteor"], "pipeline_tag": "translation"}
|
task
|
[
"SEMANTIC_SIMILARITY",
"TRANSLATION"
] | 45,204 |
gokuls/hbertv1-Massive-intent
|
gokuls
|
text-classification
|
[
"transformers",
"pytorch",
"tensorboard",
"hybridbert",
"text-classification",
"generated_from_trainer",
"dataset:massive",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-06-19T19:02:13Z |
2023-06-19T19:11:18+00:00
| 15 | 0 |
---
datasets:
- massive
metrics:
- accuracy
tags:
- generated_from_trainer
model-index:
- name: hbertv1-Massive-intent
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: massive
type: massive
config: en-US
split: validation
args: en-US
metrics:
- type: accuracy
value: 0.8450565666502705
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hbertv1-Massive-intent
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new) on the massive dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8959
- Accuracy: 0.8451
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 3.051 | 1.0 | 180 | 1.8409 | 0.4968 |
| 1.3906 | 2.0 | 360 | 1.0234 | 0.7167 |
| 0.8613 | 3.0 | 540 | 0.8787 | 0.7688 |
| 0.6447 | 4.0 | 720 | 0.8405 | 0.7811 |
| 0.4955 | 5.0 | 900 | 0.8426 | 0.7850 |
| 0.3899 | 6.0 | 1080 | 0.7777 | 0.8175 |
| 0.3052 | 7.0 | 1260 | 0.7779 | 0.8175 |
| 0.2413 | 8.0 | 1440 | 0.8294 | 0.8254 |
| 0.196 | 9.0 | 1620 | 0.8265 | 0.8214 |
| 0.1545 | 10.0 | 1800 | 0.8361 | 0.8362 |
| 0.1177 | 11.0 | 1980 | 0.8470 | 0.8288 |
| 0.0894 | 12.0 | 2160 | 0.8706 | 0.8283 |
| 0.0666 | 13.0 | 2340 | 0.8853 | 0.8392 |
| 0.0447 | 14.0 | 2520 | 0.8959 | 0.8451 |
| 0.0312 | 15.0 | 2700 | 0.8982 | 0.8441 |
### Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.0
- Tokenizers 0.13.3
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hbertv1-Massive-intent
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new) on the massive dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8959
- Accuracy: 0.8451
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 3.051 | 1.0 | 180 | 1.8409 | 0.4968 |
| 1.3906 | 2.0 | 360 | 1.0234 | 0.7167 |
| 0.8613 | 3.0 | 540 | 0.8787 | 0.7688 |
| 0.6447 | 4.0 | 720 | 0.8405 | 0.7811 |
| 0.4955 | 5.0 | 900 | 0.8426 | 0.7850 |
| 0.3899 | 6.0 | 1080 | 0.7777 | 0.8175 |
| 0.3052 | 7.0 | 1260 | 0.7779 | 0.8175 |
| 0.2413 | 8.0 | 1440 | 0.8294 | 0.8254 |
| 0.196 | 9.0 | 1620 | 0.8265 | 0.8214 |
| 0.1545 | 10.0 | 1800 | 0.8361 | 0.8362 |
| 0.1177 | 11.0 | 1980 | 0.8470 | 0.8288 |
| 0.0894 | 12.0 | 2160 | 0.8706 | 0.8283 |
| 0.0666 | 13.0 | 2340 | 0.8853 | 0.8392 |
| 0.0447 | 14.0 | 2520 | 0.8959 | 0.8451 |
| 0.0312 | 15.0 | 2700 | 0.8982 | 0.8441 |
### Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.0
- Tokenizers 0.13.3
|
{"datasets": ["massive"], "metrics": ["accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "hbertv1-Massive-intent", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "massive", "type": "massive", "config": "en-US", "split": "validation", "args": "en-US"}, "metrics": [{"type": "accuracy", "value": 0.8450565666502705, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,205 |
gokuls/hbertv1-mini-wt-48-Massive-intent
|
gokuls
|
text-classification
|
[
"transformers",
"pytorch",
"tensorboard",
"hybridbert",
"text-classification",
"generated_from_trainer",
"dataset:massive",
"base_model:gokuls/model_v1_complete_training_wt_init_48_mini",
"base_model:finetune:gokuls/model_v1_complete_training_wt_init_48_mini",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-07-21T18:20:04Z |
2023-07-21T18:23:43+00:00
| 8 | 0 |
---
base_model: gokuls/model_v1_complete_training_wt_init_48_mini
datasets:
- massive
metrics:
- accuracy
tags:
- generated_from_trainer
model-index:
- name: hbertv1-mini-wt-48-Massive-intent
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: massive
type: massive
config: en-US
split: validation
args: en-US
metrics:
- type: accuracy
value: 0.8544023610427939
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hbertv1-mini-wt-48-Massive-intent
This model is a fine-tuned version of [gokuls/model_v1_complete_training_wt_init_48_mini](https://huggingface.co/gokuls/model_v1_complete_training_wt_init_48_mini) on the massive dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6406
- Accuracy: 0.8544
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 3.16 | 1.0 | 180 | 2.1089 | 0.4934 |
| 1.6964 | 2.0 | 360 | 1.2208 | 0.6916 |
| 1.1107 | 3.0 | 540 | 0.9116 | 0.7703 |
| 0.8493 | 4.0 | 720 | 0.7717 | 0.8155 |
| 0.692 | 5.0 | 900 | 0.7166 | 0.8155 |
| 0.5849 | 6.0 | 1080 | 0.6754 | 0.8288 |
| 0.5133 | 7.0 | 1260 | 0.6491 | 0.8392 |
| 0.4541 | 8.0 | 1440 | 0.6406 | 0.8451 |
| 0.4074 | 9.0 | 1620 | 0.6346 | 0.8480 |
| 0.3615 | 10.0 | 1800 | 0.6403 | 0.8460 |
| 0.3304 | 11.0 | 1980 | 0.6452 | 0.8446 |
| 0.3021 | 12.0 | 2160 | 0.6390 | 0.8495 |
| 0.2792 | 13.0 | 2340 | 0.6412 | 0.8515 |
| 0.2584 | 14.0 | 2520 | 0.6406 | 0.8544 |
| 0.2483 | 15.0 | 2700 | 0.6394 | 0.8529 |
### Framework versions
- Transformers 4.31.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.1
- Tokenizers 0.13.3
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hbertv1-mini-wt-48-Massive-intent
This model is a fine-tuned version of [gokuls/model_v1_complete_training_wt_init_48_mini](https://huggingface.co/gokuls/model_v1_complete_training_wt_init_48_mini) on the massive dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6406
- Accuracy: 0.8544
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 3.16 | 1.0 | 180 | 2.1089 | 0.4934 |
| 1.6964 | 2.0 | 360 | 1.2208 | 0.6916 |
| 1.1107 | 3.0 | 540 | 0.9116 | 0.7703 |
| 0.8493 | 4.0 | 720 | 0.7717 | 0.8155 |
| 0.692 | 5.0 | 900 | 0.7166 | 0.8155 |
| 0.5849 | 6.0 | 1080 | 0.6754 | 0.8288 |
| 0.5133 | 7.0 | 1260 | 0.6491 | 0.8392 |
| 0.4541 | 8.0 | 1440 | 0.6406 | 0.8451 |
| 0.4074 | 9.0 | 1620 | 0.6346 | 0.8480 |
| 0.3615 | 10.0 | 1800 | 0.6403 | 0.8460 |
| 0.3304 | 11.0 | 1980 | 0.6452 | 0.8446 |
| 0.3021 | 12.0 | 2160 | 0.6390 | 0.8495 |
| 0.2792 | 13.0 | 2340 | 0.6412 | 0.8515 |
| 0.2584 | 14.0 | 2520 | 0.6406 | 0.8544 |
| 0.2483 | 15.0 | 2700 | 0.6394 | 0.8529 |
### Framework versions
- Transformers 4.31.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.1
- Tokenizers 0.13.3
|
{"base_model": "gokuls/model_v1_complete_training_wt_init_48_mini", "datasets": ["massive"], "metrics": ["accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "hbertv1-mini-wt-48-Massive-intent", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "massive", "type": "massive", "config": "en-US", "split": "validation", "args": "en-US"}, "metrics": [{"type": "accuracy", "value": 0.8544023610427939, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,206 |
RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf
|
RichardErkhov
| null |
[
"gguf",
"endpoints_compatible",
"region:us",
"conversational"
] | 2024-08-10T18:18:30Z |
2024-08-10T19:50:12+00:00
| 22 | 0 |
---
{}
---
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
llama3-pre1-pre2-ds-lora3 - GGUF
- Model creator: https://huggingface.co/4yo1/
- Original model: https://huggingface.co/4yo1/llama3-pre1-pre2-ds-lora3/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [llama3-pre1-pre2-ds-lora3.Q2_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q2_K.gguf) | Q2_K | 3.05GB |
| [llama3-pre1-pre2-ds-lora3.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ3_XS.gguf) | IQ3_XS | 3.38GB |
| [llama3-pre1-pre2-ds-lora3.IQ3_S.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ3_S.gguf) | IQ3_S | 3.53GB |
| [llama3-pre1-pre2-ds-lora3.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q3_K_S.gguf) | Q3_K_S | 3.52GB |
| [llama3-pre1-pre2-ds-lora3.IQ3_M.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ3_M.gguf) | IQ3_M | 3.63GB |
| [llama3-pre1-pre2-ds-lora3.Q3_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q3_K.gguf) | Q3_K | 3.85GB |
| [llama3-pre1-pre2-ds-lora3.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q3_K_M.gguf) | Q3_K_M | 3.85GB |
| [llama3-pre1-pre2-ds-lora3.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q3_K_L.gguf) | Q3_K_L | 4.13GB |
| [llama3-pre1-pre2-ds-lora3.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ4_XS.gguf) | IQ4_XS | 4.29GB |
| [llama3-pre1-pre2-ds-lora3.Q4_0.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_0.gguf) | Q4_0 | 4.45GB |
| [llama3-pre1-pre2-ds-lora3.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ4_NL.gguf) | IQ4_NL | 4.5GB |
| [llama3-pre1-pre2-ds-lora3.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_K_S.gguf) | Q4_K_S | 4.48GB |
| [llama3-pre1-pre2-ds-lora3.Q4_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_K.gguf) | Q4_K | 4.7GB |
| [llama3-pre1-pre2-ds-lora3.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_K_M.gguf) | Q4_K_M | 4.7GB |
| [llama3-pre1-pre2-ds-lora3.Q4_1.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_1.gguf) | Q4_1 | 4.9GB |
| [llama3-pre1-pre2-ds-lora3.Q5_0.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_0.gguf) | Q5_0 | 5.34GB |
| [llama3-pre1-pre2-ds-lora3.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_K_S.gguf) | Q5_K_S | 5.34GB |
| [llama3-pre1-pre2-ds-lora3.Q5_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_K.gguf) | Q5_K | 5.46GB |
| [llama3-pre1-pre2-ds-lora3.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_K_M.gguf) | Q5_K_M | 5.46GB |
| [llama3-pre1-pre2-ds-lora3.Q5_1.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_1.gguf) | Q5_1 | 5.78GB |
| [llama3-pre1-pre2-ds-lora3.Q6_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q6_K.gguf) | Q6_K | 6.28GB |
| [llama3-pre1-pre2-ds-lora3.Q8_0.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q8_0.gguf) | Q8_0 | 8.13GB |
Original model description:
---
library_name: transformers
language:
- en
- ko
pipeline_tag: translation
tags:
- llama-3-ko
license: mit
datasets:
- recipes
---
### Model Card for Model ID
### Model Details
Model Card: llama3-pre1-pre2-ds-lora3 with Fine-Tuning
Model Overview
Model Name: llama3-pre1-pre2-ds-lora3
Model Type: Transformer-based Language Model
Model Size: 8 billion parameters
by: 4yo1
Languages: English and Korean
### Model Description
llama3-pre1-pre2-ds-lora3 is a language model pre-trained on a diverse corpus of English and Korean texts.
This fine-tuning approach allows the model to adapt to specific tasks or datasets with a minimal number of additional parameters, making it efficient and effective for specialized applications.
### how to use - sample code
```python
from transformers import AutoConfig, AutoModel, AutoTokenizer
config = AutoConfig.from_pretrained("4yo1/llama3-pre1-pre2-ds-lora3")
model = AutoModel.from_pretrained("4yo1/llama3-pre1-pre2-ds-lora3")
tokenizer = AutoTokenizer.from_pretrained("4yo1/llama3-pre1-pre2-ds-lora3")
```
datasets:
- recipes
license: mit
| null |
Non_BioNLP
|
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
llama3-pre1-pre2-ds-lora3 - GGUF
- Model creator: https://huggingface.co/4yo1/
- Original model: https://huggingface.co/4yo1/llama3-pre1-pre2-ds-lora3/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [llama3-pre1-pre2-ds-lora3.Q2_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q2_K.gguf) | Q2_K | 3.05GB |
| [llama3-pre1-pre2-ds-lora3.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ3_XS.gguf) | IQ3_XS | 3.38GB |
| [llama3-pre1-pre2-ds-lora3.IQ3_S.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ3_S.gguf) | IQ3_S | 3.53GB |
| [llama3-pre1-pre2-ds-lora3.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q3_K_S.gguf) | Q3_K_S | 3.52GB |
| [llama3-pre1-pre2-ds-lora3.IQ3_M.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ3_M.gguf) | IQ3_M | 3.63GB |
| [llama3-pre1-pre2-ds-lora3.Q3_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q3_K.gguf) | Q3_K | 3.85GB |
| [llama3-pre1-pre2-ds-lora3.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q3_K_M.gguf) | Q3_K_M | 3.85GB |
| [llama3-pre1-pre2-ds-lora3.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q3_K_L.gguf) | Q3_K_L | 4.13GB |
| [llama3-pre1-pre2-ds-lora3.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ4_XS.gguf) | IQ4_XS | 4.29GB |
| [llama3-pre1-pre2-ds-lora3.Q4_0.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_0.gguf) | Q4_0 | 4.45GB |
| [llama3-pre1-pre2-ds-lora3.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.IQ4_NL.gguf) | IQ4_NL | 4.5GB |
| [llama3-pre1-pre2-ds-lora3.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_K_S.gguf) | Q4_K_S | 4.48GB |
| [llama3-pre1-pre2-ds-lora3.Q4_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_K.gguf) | Q4_K | 4.7GB |
| [llama3-pre1-pre2-ds-lora3.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_K_M.gguf) | Q4_K_M | 4.7GB |
| [llama3-pre1-pre2-ds-lora3.Q4_1.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q4_1.gguf) | Q4_1 | 4.9GB |
| [llama3-pre1-pre2-ds-lora3.Q5_0.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_0.gguf) | Q5_0 | 5.34GB |
| [llama3-pre1-pre2-ds-lora3.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_K_S.gguf) | Q5_K_S | 5.34GB |
| [llama3-pre1-pre2-ds-lora3.Q5_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_K.gguf) | Q5_K | 5.46GB |
| [llama3-pre1-pre2-ds-lora3.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_K_M.gguf) | Q5_K_M | 5.46GB |
| [llama3-pre1-pre2-ds-lora3.Q5_1.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q5_1.gguf) | Q5_1 | 5.78GB |
| [llama3-pre1-pre2-ds-lora3.Q6_K.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q6_K.gguf) | Q6_K | 6.28GB |
| [llama3-pre1-pre2-ds-lora3.Q8_0.gguf](https://huggingface.co/RichardErkhov/4yo1_-_llama3-pre1-pre2-ds-lora3-gguf/blob/main/llama3-pre1-pre2-ds-lora3.Q8_0.gguf) | Q8_0 | 8.13GB |
Original model description:
---
library_name: transformers
language:
- en
- ko
pipeline_tag: translation
tags:
- llama-3-ko
license: mit
datasets:
- recipes
---
### Model Card for Model ID
### Model Details
Model Card: llama3-pre1-pre2-ds-lora3 with Fine-Tuning
Model Overview
Model Name: llama3-pre1-pre2-ds-lora3
Model Type: Transformer-based Language Model
Model Size: 8 billion parameters
by: 4yo1
Languages: English and Korean
### Model Description
llama3-pre1-pre2-ds-lora3 is a language model pre-trained on a diverse corpus of English and Korean texts.
This fine-tuning approach allows the model to adapt to specific tasks or datasets with a minimal number of additional parameters, making it efficient and effective for specialized applications.
### how to use - sample code
```python
from transformers import AutoConfig, AutoModel, AutoTokenizer
config = AutoConfig.from_pretrained("4yo1/llama3-pre1-pre2-ds-lora3")
model = AutoModel.from_pretrained("4yo1/llama3-pre1-pre2-ds-lora3")
tokenizer = AutoTokenizer.from_pretrained("4yo1/llama3-pre1-pre2-ds-lora3")
```
datasets:
- recipes
license: mit
|
{}
|
task
|
[
"TRANSLATION"
] | 45,207 |
Lots-of-LoRAs/Mistral-7B-Instruct-v0.2-4b-r16-task601
|
Lots-of-LoRAs
| null |
[
"pytorch",
"safetensors",
"en",
"arxiv:1910.09700",
"arxiv:2407.00066",
"base_model:mistralai/Mistral-7B-Instruct-v0.2",
"base_model:finetune:mistralai/Mistral-7B-Instruct-v0.2",
"license:mit",
"region:us"
] | 2025-01-02T14:11:44Z |
2025-01-02T14:11:50+00:00
| 0 | 0 |
---
base_model: mistralai/Mistral-7B-Instruct-v0.2
language: en
library_name: pytorch
license: mit
---
# Model Card for Mistral-7B-Instruct-v0.2-4b-r16-task601
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
LoRA trained on task601_flores_translation_sntoen
- **Developed by:** bruel
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** LoRA
- **Language(s) (NLP):** en
- **License:** mit
- **Finetuned from model [optional]:** mistralai/Mistral-7B-Instruct-v0.2
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** https://github.com/bruel-gabrielsson
- **Paper [optional]:** "Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead" (2024), Rickard Brüel Gabrielsson, Jiacheng Zhu, Onkar Bhardwaj, Leshem Choshen, Kristjan Greenewald, Mikhail Yurochkin and Justin Solomon
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
https://huggingface.co/datasets/Lots-of-LoRAs/task601_flores_translation_sntoen sourced from https://github.com/allenai/natural-instructions
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
@misc{brüelgabrielsson2024compressserveservingthousands,
title={Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead},
author={Rickard Brüel-Gabrielsson and Jiacheng Zhu and Onkar Bhardwaj and Leshem Choshen and Kristjan Greenewald and Mikhail Yurochkin and Justin Solomon},
year={2024},
eprint={2407.00066},
archivePrefix={arXiv},
primaryClass={cs.DC},
url={https://arxiv.org/abs/2407.00066},
}
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
| null |
Non_BioNLP
|
# Model Card for Mistral-7B-Instruct-v0.2-4b-r16-task601
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
LoRA trained on task601_flores_translation_sntoen
- **Developed by:** bruel
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** LoRA
- **Language(s) (NLP):** en
- **License:** mit
- **Finetuned from model [optional]:** mistralai/Mistral-7B-Instruct-v0.2
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** https://github.com/bruel-gabrielsson
- **Paper [optional]:** "Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead" (2024), Rickard Brüel Gabrielsson, Jiacheng Zhu, Onkar Bhardwaj, Leshem Choshen, Kristjan Greenewald, Mikhail Yurochkin and Justin Solomon
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
https://huggingface.co/datasets/Lots-of-LoRAs/task601_flores_translation_sntoen sourced from https://github.com/allenai/natural-instructions
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
@misc{brüelgabrielsson2024compressserveservingthousands,
title={Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead},
author={Rickard Brüel-Gabrielsson and Jiacheng Zhu and Onkar Bhardwaj and Leshem Choshen and Kristjan Greenewald and Mikhail Yurochkin and Justin Solomon},
year={2024},
eprint={2407.00066},
archivePrefix={arXiv},
primaryClass={cs.DC},
url={https://arxiv.org/abs/2407.00066},
}
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
{"base_model": "mistralai/Mistral-7B-Instruct-v0.2", "language": "en", "library_name": "pytorch", "license": "mit"}
|
task
|
[
"TRANSLATION"
] | 45,208 |
ozone-research/llama-3.1-0x-mini
|
ozone-research
|
text2text-generation
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"text2text-generation",
"dataset:lmsys/lmsys-chat-1m",
"base_model:meta-llama/Llama-3.1-8B-Instruct",
"base_model:finetune:meta-llama/Llama-3.1-8B-Instruct",
"license:llama3.1",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2025-01-21T22:52:27Z |
2025-01-21T23:06:50+00:00
| 43 | 5 |
---
base_model:
- meta-llama/Llama-3.1-8B-Instruct
datasets:
- lmsys/lmsys-chat-1m
library_name: transformers
license: llama3.1
pipeline_tag: text2text-generation
---
# 0x Mini
## Overview
0x Mini is a state-of-the-art language model developed by Ozone AI, designed to deliver high-quality text generation capabilities while maintaining a compact and efficient architecture. Built on the latest advancements in natural language processing, 0x Mini is optimized for both speed and accuracy, making it a strong contender in the space of language models. It is particularly well-suited for applications where resource constraints are a concern, offering a lightweight alternative to larger models like GPT while still delivering comparable performance.
## Features
- **Compact and Efficient**: 0x Mini is designed to be lightweight, making it suitable for deployment on resource-constrained devices.
- **High-Quality Text Generation**: The model is trained on a diverse dataset to generate coherent, contextually relevant, and human-like text.
- **Versatile Applications**: Suitable for tasks such as text completion, summarization, translation, and more.
- **Fast Inference**: Optimized for speed, ensuring quick and efficient responses.
- **Open-Source and Community-Driven**: Built with transparency and collaboration in mind, 0x Mini is available for the community to use, modify, and improve.
## Use Cases
- **Text Completion**: Assist users with writing tasks by generating coherent and contextually appropriate text.
- **Summarization**: Summarize long documents into concise and meaningful summaries.
- **Chatbots**: Power conversational AI systems with 0x Mini.
- **Content Creation**: Generate creative content such as stories, poems, or marketing copy.
- **Education**: Assist students with research, essay writing, and language learning.
## Getting Started
To get started with 0x Mini, follow these steps:
1. **Install the Model**:
```bash
pip install transformers
```
2. **Load the Model**:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "ozone-ai/llama-3.1-0x-mini"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
```
3. **Generate Text**:
```python
input_text = "Once upon a time"
inputs = tokenizer(input_text, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_length=50)
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(generated_text)
```
| null |
Non_BioNLP
|
# 0x Mini
## Overview
0x Mini is a state-of-the-art language model developed by Ozone AI, designed to deliver high-quality text generation capabilities while maintaining a compact and efficient architecture. Built on the latest advancements in natural language processing, 0x Mini is optimized for both speed and accuracy, making it a strong contender in the space of language models. It is particularly well-suited for applications where resource constraints are a concern, offering a lightweight alternative to larger models like GPT while still delivering comparable performance.
## Features
- **Compact and Efficient**: 0x Mini is designed to be lightweight, making it suitable for deployment on resource-constrained devices.
- **High-Quality Text Generation**: The model is trained on a diverse dataset to generate coherent, contextually relevant, and human-like text.
- **Versatile Applications**: Suitable for tasks such as text completion, summarization, translation, and more.
- **Fast Inference**: Optimized for speed, ensuring quick and efficient responses.
- **Open-Source and Community-Driven**: Built with transparency and collaboration in mind, 0x Mini is available for the community to use, modify, and improve.
## Use Cases
- **Text Completion**: Assist users with writing tasks by generating coherent and contextually appropriate text.
- **Summarization**: Summarize long documents into concise and meaningful summaries.
- **Chatbots**: Power conversational AI systems with 0x Mini.
- **Content Creation**: Generate creative content such as stories, poems, or marketing copy.
- **Education**: Assist students with research, essay writing, and language learning.
## Getting Started
To get started with 0x Mini, follow these steps:
1. **Install the Model**:
```bash
pip install transformers
```
2. **Load the Model**:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "ozone-ai/llama-3.1-0x-mini"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
```
3. **Generate Text**:
```python
input_text = "Once upon a time"
inputs = tokenizer(input_text, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_length=50)
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(generated_text)
```
|
{"base_model": ["meta-llama/Llama-3.1-8B-Instruct"], "datasets": ["lmsys/lmsys-chat-1m"], "library_name": "transformers", "license": "llama3.1", "pipeline_tag": "text2text-generation"}
|
task
|
[
"TRANSLATION",
"SUMMARIZATION"
] | 45,209 |
budecosystem/sql-millennials-13b
|
budecosystem
|
text-generation
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"license:llama2",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2023-10-27T16:04:48Z |
2023-10-27T17:29:57+00:00
| 2,163 | 4 |
---
language:
- en
library_name: transformers
license: llama2
---
## Introducing Text-to-SQL Translation Model - Millennials. 🎉
Welcome to our Text-to-SQL Translation Model repository! Our model is specifically fine-tuned for text-to-SQL tasks, aiming to revolutionize how systems understand and translate natural language instructions into SQL queries. Built on CodeLLaMa 13B, our model has been meticulously fine-tuned with a curated dataset comprising 100k SQL query generation instructions, ensuring quality and precision.
## Features
* Specialized in converting natural language text to SQL queries.
* Fine-tuned on a diverse set of 100k SQL query generation instructions.
* Easy to integrate and use for generating SQL queries on the fly.
## Generate responses
Now that your model is fine-tuned, you're ready to generate responses, you can easily generate SQL queries from natural language instructions. To do this, you'll be using our generate.py script, which allows for quick inference and can fetch models directly from the Hugging Face model hub.
Here's a quick guide on how to use it:
The script runs inference using the pre-trained model from the Hugging Face model hub and prints the generated SQL query.
```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("budecosystem/sql-millennials-13b")
model = AutoModelForCausalLM.from_pretrained("budecosystem/sql-millennials-13b")
prompt = "A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
USER: Create SQL query for the given table schema and question ASSISTANT:"
inputs = tokenizer(prompt, return_tensors="pt")
sample = model.generate(**inputs, max_length=128)
print(tokenizer.decode(sample[0]))
```
The script runs inference using the pre-trained model from the Hugging Face model hub and prints the generated SQL query.
## Training details
The model is trained of 4 A100 80GB for approximately 55hrs.
| Hyperparameters | Value |
| :----------------------------| :-----: |
| per_device_train_batch_size | 4 |
| gradient_accumulation_steps | 1 |
| epoch | 3 |
| steps | 19206 |
| learning_rate | 2e-5 |
| lr schedular type | cosine |
| warmup steps | 2000 |
| optimizer | adamw |
| fp16 | True |
| GPU | 4 A100 80GB |
## Why millennials?
1. Automated Database Management for Businesses
Scenario: Small to medium-sized enterprises (SMEs) often lack dedicated IT teams to handle database queries, making it challenging to retrieve specific data quickly for analysis and decision-making.
Use Case: Your text-to-SQL model can be integrated into a company's internal systems, allowing staff without technical SQL knowledge to retrieve data. They can input natural language requests, such as "Get a list of all transactions above $10,000 in the last quarter," and the system, powered by your model, would convert this into a corresponding SQL query to retrieve the data.
2. Automating Data Analytics Processes
Scenario: Data analysts and business professionals often face bottlenecks in generating insights due to the complexities of SQL query formulation, especially when immediate or repetitive data retrieval and analysis are required.
Use Case: Your text-to-SQL model serves as a transformative intermediary in this scenario. By integrating the model into their data analytics systems, organizations enable professionals to input data requests in natural language. For instance, an analyst could input, "Show the trend of online sales growth over the past five years," and the system would instantly convert this request into a SQL query, retrieve the data, and even integrate it into visualization tools for immediate insight generation. This functionality not only accelerates the analytical processes but also democratizes data-driven insights across different organizational departments, allowing even non-technical staff to leverage the power of real-time data analytics without deep knowledge of SQL.
3. Enhancing CMS Interfaces
Scenario: Content Management Systems (CMS) are often non-intuitive for non-technical content managers when it comes to complex data retrieval or database management.
Use Case: CMS providers can leverage your model to enhance their system's backend interface. Content managers can use natural language to request specific data, like "Find all blog posts in May 2023 with more than 500 views," and the model will generate the appropriate SQL to retrieve the information. This feature makes database management more accessible, efficient, and user-friendly.
4. Customer Support Optimization
Scenario: Customer support centers often need to retrieve client or product information stored in databases while resolving tickets or inquiries, requiring basic knowledge of SQL.
Use Case: Your model can be integrated into support ticketing systems, enabling support personnel to type requests in natural language, such as "Show all open tickets from customers in New York filed this month," and immediately receive the data needed to expedite their resolution process, improving customer service efficiency and response time.
5. Data Journalism and Research
Scenario: Journalists and researchers frequently rely on complex databases to gather insights and data points necessary for their work but may lack the technical know-how of SQL.
Use Case: By integrating your text-to-SQL model into research software or journalistic tools, professionals can query databases using natural language. For example, a journalist might input, "Retrieve the average household income in Texas in 2022," and your model would facilitate immediate access to this data, allowing for more efficient research and data-driven storytelling.
Contributing
We welcome contributions to help improve the model or address issues. Please feel free to submit pull requests or open issues to discuss changes or improvements.
### Acknowledgments
We'd like to thank the open-source community and the researchers whose foundational work laid the path to this model.
| null |
Non_BioNLP
|
## Introducing Text-to-SQL Translation Model - Millennials. 🎉
Welcome to our Text-to-SQL Translation Model repository! Our model is specifically fine-tuned for text-to-SQL tasks, aiming to revolutionize how systems understand and translate natural language instructions into SQL queries. Built on CodeLLaMa 13B, our model has been meticulously fine-tuned with a curated dataset comprising 100k SQL query generation instructions, ensuring quality and precision.
## Features
* Specialized in converting natural language text to SQL queries.
* Fine-tuned on a diverse set of 100k SQL query generation instructions.
* Easy to integrate and use for generating SQL queries on the fly.
## Generate responses
Now that your model is fine-tuned, you're ready to generate responses, you can easily generate SQL queries from natural language instructions. To do this, you'll be using our generate.py script, which allows for quick inference and can fetch models directly from the Hugging Face model hub.
Here's a quick guide on how to use it:
The script runs inference using the pre-trained model from the Hugging Face model hub and prints the generated SQL query.
```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("budecosystem/sql-millennials-13b")
model = AutoModelForCausalLM.from_pretrained("budecosystem/sql-millennials-13b")
prompt = "A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
USER: Create SQL query for the given table schema and question ASSISTANT:"
inputs = tokenizer(prompt, return_tensors="pt")
sample = model.generate(**inputs, max_length=128)
print(tokenizer.decode(sample[0]))
```
The script runs inference using the pre-trained model from the Hugging Face model hub and prints the generated SQL query.
## Training details
The model is trained of 4 A100 80GB for approximately 55hrs.
| Hyperparameters | Value |
| :----------------------------| :-----: |
| per_device_train_batch_size | 4 |
| gradient_accumulation_steps | 1 |
| epoch | 3 |
| steps | 19206 |
| learning_rate | 2e-5 |
| lr schedular type | cosine |
| warmup steps | 2000 |
| optimizer | adamw |
| fp16 | True |
| GPU | 4 A100 80GB |
## Why millennials?
1. Automated Database Management for Businesses
Scenario: Small to medium-sized enterprises (SMEs) often lack dedicated IT teams to handle database queries, making it challenging to retrieve specific data quickly for analysis and decision-making.
Use Case: Your text-to-SQL model can be integrated into a company's internal systems, allowing staff without technical SQL knowledge to retrieve data. They can input natural language requests, such as "Get a list of all transactions above $10,000 in the last quarter," and the system, powered by your model, would convert this into a corresponding SQL query to retrieve the data.
2. Automating Data Analytics Processes
Scenario: Data analysts and business professionals often face bottlenecks in generating insights due to the complexities of SQL query formulation, especially when immediate or repetitive data retrieval and analysis are required.
Use Case: Your text-to-SQL model serves as a transformative intermediary in this scenario. By integrating the model into their data analytics systems, organizations enable professionals to input data requests in natural language. For instance, an analyst could input, "Show the trend of online sales growth over the past five years," and the system would instantly convert this request into a SQL query, retrieve the data, and even integrate it into visualization tools for immediate insight generation. This functionality not only accelerates the analytical processes but also democratizes data-driven insights across different organizational departments, allowing even non-technical staff to leverage the power of real-time data analytics without deep knowledge of SQL.
3. Enhancing CMS Interfaces
Scenario: Content Management Systems (CMS) are often non-intuitive for non-technical content managers when it comes to complex data retrieval or database management.
Use Case: CMS providers can leverage your model to enhance their system's backend interface. Content managers can use natural language to request specific data, like "Find all blog posts in May 2023 with more than 500 views," and the model will generate the appropriate SQL to retrieve the information. This feature makes database management more accessible, efficient, and user-friendly.
4. Customer Support Optimization
Scenario: Customer support centers often need to retrieve client or product information stored in databases while resolving tickets or inquiries, requiring basic knowledge of SQL.
Use Case: Your model can be integrated into support ticketing systems, enabling support personnel to type requests in natural language, such as "Show all open tickets from customers in New York filed this month," and immediately receive the data needed to expedite their resolution process, improving customer service efficiency and response time.
5. Data Journalism and Research
Scenario: Journalists and researchers frequently rely on complex databases to gather insights and data points necessary for their work but may lack the technical know-how of SQL.
Use Case: By integrating your text-to-SQL model into research software or journalistic tools, professionals can query databases using natural language. For example, a journalist might input, "Retrieve the average household income in Texas in 2022," and your model would facilitate immediate access to this data, allowing for more efficient research and data-driven storytelling.
Contributing
We welcome contributions to help improve the model or address issues. Please feel free to submit pull requests or open issues to discuss changes or improvements.
### Acknowledgments
We'd like to thank the open-source community and the researchers whose foundational work laid the path to this model.
|
{"language": ["en"], "library_name": "transformers", "license": "llama2"}
|
task
|
[
"TRANSLATION"
] | 45,210 |
dawn78/minilm6_perfumerecommender_v1
|
dawn78
|
sentence-similarity
|
[
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:1459",
"loss:CosineSimilarityLoss",
"arxiv:1908.10084",
"base_model:sentence-transformers/all-MiniLM-L6-v2",
"base_model:finetune:sentence-transformers/all-MiniLM-L6-v2",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2025-02-05T16:14:19Z |
2025-02-05T16:14:24+00:00
| 7 | 0 |
---
base_model: sentence-transformers/all-MiniLM-L6-v2
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:1459
- loss:CosineSimilarityLoss
widget:
- source_sentence: still popular today this fresh fougere fragrance inspired many
wannabes
sentences:
- pear, blackberry, herbal notes, bamboo, clove, apple, guarana, green tree accord
- mace, hyrax, camellia, tea, akigalawood
- mandarin, lavender, green botanics, jasmine, basil, geranium, sage, sandalwood,
vetiver, rosewood, amber
- source_sentence: little black dress eau fraiche by avon exudes a lively and refreshing
spirit that captivates effortlessly this fragrance opens with a bright burst of
citrus that instantly uplifts the mood reminiscent of sunkissed afternoons as
it unfolds delicate floral notes weave through creating an elegant bouquet that
embodies femininity and charm the scent is anchored by a subtle musk that rounds
out the experience providing a warm and inviting backdrop users have praised this
fragrance for its fresh and invigorating essence making it perfect for daytime
wear many appreciate its lightness and airy quality which is ideal for those seeking
a scent that is both playful and sophisticated with a commendable rating of 375
out of 5 it has earned accolades for its delightful character and versatility
appealing to a broad audience who value a fragrance that feels both chic and approachable
overall little black dress eau fraiche is described as an essential contemporary
scent for the modern woman effortlessly enhancing any occasion with its vibrant
charm
sentences:
- cress, lantana, castoreum, parma violet, cotton flower, oud, hesperidic notes,
grape, olive tree, hyacinth, earthy notes, carambola, osmanthus, champaca, cypriol,
lemon blossom, rosewood
- yuzu, clary sage, balsam fir, cedar
- passionflower, red currant, rosehip, almond blossom, chocolate
- source_sentence: rose blush cologne 2023 by jo malone london rose blush cologne
presents an enchanting bouquet that captures the essence of blooming romance and
tropical vitality with an initial sweet hint of luscious litchi and a refreshing
touch of herbs this fragrance unfolds into a heart of delicate rose showcasing
a radiant femininity the composition is beautifully rounded off with soft musky
undertones adding an elegant warmth that lingers on the skin users describe rose
blush as vibrant and joyful perfect for both everyday wear and special occasions
reviewers appreciate its fresh appeal heralding it as an uplifting scent that
evokes feelings of spring and renewal many highlight its moderate longevity making
it suitable for those who desire a fragrance that gently permeates without overwhelming
whether youre seeking a burst of floral energy or a subtle whisper of sophistication
this perfume is sure to leave a delightful impression
sentences:
- honey, mahogany
- lychee, basil, rose, musk
- lemon, may rose, spices, peony, lily of the valley, blackcurrant, raspberry, peach,
musk, sandalwood, amber, heliotrope, oud
- source_sentence: thank u next by ariana grande is a playful and modern fragrance
that captures the essence of youthful exuberance and selfempowerment this charming
scent exudes a vibrant sweetness that dances between fruity and creamy notes creating
an inviting aura that is both uplifting and comforting users often describe this
perfume as deliciously sweet and fun making it perfect for casual wear or a spirited
night out the blend is frequently noted for its warm inviting quality evoking
a sense of cheerful nostalgia many reviewers highlight its longlasting nature
and delightful sillage ensuring that its fragrant embrace stays with you throughout
the day perfect for the confident contemporary woman thank u next effortlessly
combines the spirited essence of fresh berries with a creamy tropical nuance which
is masterfully balanced by an undercurrent of sweet indulgence overall this fragrance
is celebrated for its delightful charm and is sure to make a memorable impression
wherever you go
sentences:
- cabreuva, mate, bamboo leaf, black cardamom, orris root, camellia, oriental notes,
hibiscus, lily of the valley, lantana, wood notes
- sea salt, amberwood, marine notes, resins, clary sage, labdanum, white musk, blonde
woods
- nectarine, olive tree, grass, cress, clementine, red apple
- source_sentence: zara night eau de parfum envelops you in a captivating blend of
softness and elegance creating a rich floral experience that feels both fresh
and inviting this fragrance exudes a charming femininity where luscious floral
notes mingle seamlessly with a warm creamy essence that evokes a sense of comfort
users describe it as enchanting and seductive perfect for evening wear or special
occasions the scent captures the essence of a night blooming with possibilities
balancing the vibrancy of fresh petals with the alluring depth of sweet undertones
reviewers appreciate its ability to linger gracefully on the skin leaving a trail
of sophisticated allure without being overwhelming many find it to be a delightful
choice for those seeking a fragrance that is both versatile and memorable with
a touch of playfulness that hints at a romantic allure with a commendable rating
zara night is celebrated for its accessibility and charm making it a favored addition
to any perfume collection
sentences:
- whiskey, bellini, cognac, blackberry, juniper berry, iris root, aldehydes, red
currant, flint, cumin, mango, sea salt, sea notes, birch, bitter orange, marine
notes, grapefruit blossom, hawthorn, yuzu, clementine, cream, pineapple
- moss, sandalwood, mangosteen, cade oil
- bergamot, galbanum, petitgrain, jasmine, narcissus, violet, carnation, rose, spices,
blonde woods, iris, vanilla, amber
model-index:
- name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
results:
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: Unknown
type: unknown
metrics:
- type: pearson_cosine
value: 0.8425746761744255
name: Pearson Cosine
- type: spearman_cosine
value: 0.718974393548417
name: Spearman Cosine
---
# SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision fa97f6e7cb1a59073dff9e6b13e2715cf7475ac9 -->
- **Maximum Sequence Length:** 256 tokens
- **Output Dimensionality:** 384 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
'zara night eau de parfum envelops you in a captivating blend of softness and elegance creating a rich floral experience that feels both fresh and inviting this fragrance exudes a charming femininity where luscious floral notes mingle seamlessly with a warm creamy essence that evokes a sense of comfort users describe it as enchanting and seductive perfect for evening wear or special occasions the scent captures the essence of a night blooming with possibilities balancing the vibrancy of fresh petals with the alluring depth of sweet undertones reviewers appreciate its ability to linger gracefully on the skin leaving a trail of sophisticated allure without being overwhelming many find it to be a delightful choice for those seeking a fragrance that is both versatile and memorable with a touch of playfulness that hints at a romantic allure with a commendable rating zara night is celebrated for its accessibility and charm making it a favored addition to any perfume collection',
'moss, sandalwood, mangosteen, cade oil',
'whiskey, bellini, cognac, blackberry, juniper berry, iris root, aldehydes, red currant, flint, cumin, mango, sea salt, sea notes, birch, bitter orange, marine notes, grapefruit blossom, hawthorn, yuzu, clementine, cream, pineapple',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Semantic Similarity
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:----------|
| pearson_cosine | 0.8426 |
| **spearman_cosine** | **0.719** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 1,459 training samples
* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
* Approximate statistics based on the first 1000 samples:
| | sentence_0 | sentence_1 | label |
|:--------|:-------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:---------------------------------------------------------------|
| type | string | string | float |
| details | <ul><li>min: 12 tokens</li><li>mean: 182.01 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 33.83 tokens</li><li>max: 101 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.25</li><li>max: 1.0</li></ul> |
* Samples:
| sentence_0 | sentence_1 | label |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------|:-----------------|
| <code>today tomorrow always in love by avon embodying a sense of timeless romance today tomorrow always in love is an enchanting fragrance that strikes a perfect balance between freshness and warmth this captivating scent opens with bright effervescent notes that evoke images of blooming gardens and sunlit moments as the fragrance unfolds it reveals a charming bouquet that celebrates femininity featuring delicate floral elements that wrap around the wearer like a cherished embrace users describe this perfume as uplifting and evocative making it an ideal companion for both everyday wear and special occasions many reviewers appreciate its elegant character highlighting its multifaceted nature that seamlessly transitions from day to night while some find it subtly sweet and playful others cherish its musky undertones which lend a depth that enhances its allure overall with a moderate rating that suggests a solid appreciation among wearers today tomorrow always in love captures the essence of ro...</code> | <code>lotus, neroli, carambola, pomegranate, tuberose, gardenia, tuberose, pepper, musk, woody notes, amber</code> | <code>1.0</code> |
| <code>mankind hero by kenneth cole encapsulates a vibrant and adventurous spirit designed for the modern man who embraces both freshness and sophistication this fragrance unfolds with an invigorating burst reminiscent of a brisk mountain breeze seamlessly paired with a zesty hint of citrus the aromatic heart introduces a soothing edginess where lavender and warm vanilla intertwine creating a balanced yet captivating profile as it settles an inviting warmth emerges enriched by woody undertones that linger pleasantly on the skin users have praised mankind hero for its versatile character suitable for both casual outings and formal occasions many describe it as longlasting and unique appreciating the balanced blend that feels both refreshing and comforting the overall sentiment reflects a sense of confidence and elegance making this scent a cherished addition to a mans fragrance collection it has garnered favorable reviews boasting a solid rating that underscores its appeal embrace the essence ...</code> | <code>mountain air, lemon, coriander, lavender, vanilla, clary sage, plum, musk, coumarin, amberwood, oak moss</code> | <code>1.0</code> |
| <code>black essential dark by avon immerse yourself in the captivating allure of black essential dark a fragrance that elegantly marries the depth of aromatic woods with a touch of leathers sensuality this modern scent envelops the wearer in a rich and sophisticated aura exuding confidence and a hint of mystery users describe it as both refreshing and spicy with an invigorating blend that feels perfect for the urban man who embraces lifes more daring adventures crafted with meticulous attention by perfumer mike parrot this fragrance has garnered a solid reputation amongst enthusiasts resulting in a commendable 405 rating from its admirers many find it to be versatile enough for both day and night wear making it an essential companion for various occasions reviewers frequently highlight its longlasting presence creating an inviting and memorable impression with a delicate yet commanding presence black essential dark is ideal for those looking to leave a mark without overpowering the senses wh...</code> | <code>mint, bay leaf, cedar needle, passionflower, black cardamom, flint, rice, teak wood, cedar leaf</code> | <code>0.0</code> |
* Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:
```json
{
"loss_fct": "torch.nn.modules.loss.MSELoss"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 32
- `num_train_epochs`: 5
- `multi_dataset_batch_sampler`: round_robin
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 32
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1
- `num_train_epochs`: 5
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: round_robin
</details>
### Training Logs
| Epoch | Step | spearman_cosine |
|:------:|:----:|:---------------:|
| 1.0 | 46 | 0.5799 |
| 1.0870 | 50 | 0.6061 |
| 2.0 | 92 | 0.6940 |
| 2.1739 | 100 | 0.6940 |
| 3.0 | 138 | 0.7072 |
| 3.2609 | 150 | 0.7124 |
| 4.0 | 184 | 0.7150 |
| 4.3478 | 200 | 0.7177 |
| 5.0 | 230 | 0.7190 |
### Framework Versions
- Python: 3.11.11
- Sentence Transformers: 3.3.1
- Transformers: 4.47.1
- PyTorch: 2.5.1+cu124
- Accelerate: 1.2.1
- Datasets: 3.2.0
- Tokenizers: 0.21.0
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision fa97f6e7cb1a59073dff9e6b13e2715cf7475ac9 -->
- **Maximum Sequence Length:** 256 tokens
- **Output Dimensionality:** 384 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
'zara night eau de parfum envelops you in a captivating blend of softness and elegance creating a rich floral experience that feels both fresh and inviting this fragrance exudes a charming femininity where luscious floral notes mingle seamlessly with a warm creamy essence that evokes a sense of comfort users describe it as enchanting and seductive perfect for evening wear or special occasions the scent captures the essence of a night blooming with possibilities balancing the vibrancy of fresh petals with the alluring depth of sweet undertones reviewers appreciate its ability to linger gracefully on the skin leaving a trail of sophisticated allure without being overwhelming many find it to be a delightful choice for those seeking a fragrance that is both versatile and memorable with a touch of playfulness that hints at a romantic allure with a commendable rating zara night is celebrated for its accessibility and charm making it a favored addition to any perfume collection',
'moss, sandalwood, mangosteen, cade oil',
'whiskey, bellini, cognac, blackberry, juniper berry, iris root, aldehydes, red currant, flint, cumin, mango, sea salt, sea notes, birch, bitter orange, marine notes, grapefruit blossom, hawthorn, yuzu, clementine, cream, pineapple',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Semantic Similarity
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:----------|
| pearson_cosine | 0.8426 |
| **spearman_cosine** | **0.719** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 1,459 training samples
* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
* Approximate statistics based on the first 1000 samples:
| | sentence_0 | sentence_1 | label |
|:--------|:-------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:---------------------------------------------------------------|
| type | string | string | float |
| details | <ul><li>min: 12 tokens</li><li>mean: 182.01 tokens</li><li>max: 256 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 33.83 tokens</li><li>max: 101 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.25</li><li>max: 1.0</li></ul> |
* Samples:
| sentence_0 | sentence_1 | label |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------|:-----------------|
| <code>today tomorrow always in love by avon embodying a sense of timeless romance today tomorrow always in love is an enchanting fragrance that strikes a perfect balance between freshness and warmth this captivating scent opens with bright effervescent notes that evoke images of blooming gardens and sunlit moments as the fragrance unfolds it reveals a charming bouquet that celebrates femininity featuring delicate floral elements that wrap around the wearer like a cherished embrace users describe this perfume as uplifting and evocative making it an ideal companion for both everyday wear and special occasions many reviewers appreciate its elegant character highlighting its multifaceted nature that seamlessly transitions from day to night while some find it subtly sweet and playful others cherish its musky undertones which lend a depth that enhances its allure overall with a moderate rating that suggests a solid appreciation among wearers today tomorrow always in love captures the essence of ro...</code> | <code>lotus, neroli, carambola, pomegranate, tuberose, gardenia, tuberose, pepper, musk, woody notes, amber</code> | <code>1.0</code> |
| <code>mankind hero by kenneth cole encapsulates a vibrant and adventurous spirit designed for the modern man who embraces both freshness and sophistication this fragrance unfolds with an invigorating burst reminiscent of a brisk mountain breeze seamlessly paired with a zesty hint of citrus the aromatic heart introduces a soothing edginess where lavender and warm vanilla intertwine creating a balanced yet captivating profile as it settles an inviting warmth emerges enriched by woody undertones that linger pleasantly on the skin users have praised mankind hero for its versatile character suitable for both casual outings and formal occasions many describe it as longlasting and unique appreciating the balanced blend that feels both refreshing and comforting the overall sentiment reflects a sense of confidence and elegance making this scent a cherished addition to a mans fragrance collection it has garnered favorable reviews boasting a solid rating that underscores its appeal embrace the essence ...</code> | <code>mountain air, lemon, coriander, lavender, vanilla, clary sage, plum, musk, coumarin, amberwood, oak moss</code> | <code>1.0</code> |
| <code>black essential dark by avon immerse yourself in the captivating allure of black essential dark a fragrance that elegantly marries the depth of aromatic woods with a touch of leathers sensuality this modern scent envelops the wearer in a rich and sophisticated aura exuding confidence and a hint of mystery users describe it as both refreshing and spicy with an invigorating blend that feels perfect for the urban man who embraces lifes more daring adventures crafted with meticulous attention by perfumer mike parrot this fragrance has garnered a solid reputation amongst enthusiasts resulting in a commendable 405 rating from its admirers many find it to be versatile enough for both day and night wear making it an essential companion for various occasions reviewers frequently highlight its longlasting presence creating an inviting and memorable impression with a delicate yet commanding presence black essential dark is ideal for those looking to leave a mark without overpowering the senses wh...</code> | <code>mint, bay leaf, cedar needle, passionflower, black cardamom, flint, rice, teak wood, cedar leaf</code> | <code>0.0</code> |
* Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:
```json
{
"loss_fct": "torch.nn.modules.loss.MSELoss"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 32
- `num_train_epochs`: 5
- `multi_dataset_batch_sampler`: round_robin
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 32
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1
- `num_train_epochs`: 5
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: round_robin
</details>
### Training Logs
| Epoch | Step | spearman_cosine |
|:------:|:----:|:---------------:|
| 1.0 | 46 | 0.5799 |
| 1.0870 | 50 | 0.6061 |
| 2.0 | 92 | 0.6940 |
| 2.1739 | 100 | 0.6940 |
| 3.0 | 138 | 0.7072 |
| 3.2609 | 150 | 0.7124 |
| 4.0 | 184 | 0.7150 |
| 4.3478 | 200 | 0.7177 |
| 5.0 | 230 | 0.7190 |
### Framework Versions
- Python: 3.11.11
- Sentence Transformers: 3.3.1
- Transformers: 4.47.1
- PyTorch: 2.5.1+cu124
- Accelerate: 1.2.1
- Datasets: 3.2.0
- Tokenizers: 0.21.0
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "sentence-transformers/all-MiniLM-L6-v2", "library_name": "sentence-transformers", "metrics": ["pearson_cosine", "spearman_cosine"], "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:1459", "loss:CosineSimilarityLoss"], "widget": [{"source_sentence": "still popular today this fresh fougere fragrance inspired many wannabes", "sentences": ["pear, blackberry, herbal notes, bamboo, clove, apple, guarana, green tree accord", "mace, hyrax, camellia, tea, akigalawood", "mandarin, lavender, green botanics, jasmine, basil, geranium, sage, sandalwood, vetiver, rosewood, amber"]}, {"source_sentence": "little black dress eau fraiche by avon exudes a lively and refreshing spirit that captivates effortlessly this fragrance opens with a bright burst of citrus that instantly uplifts the mood reminiscent of sunkissed afternoons as it unfolds delicate floral notes weave through creating an elegant bouquet that embodies femininity and charm the scent is anchored by a subtle musk that rounds out the experience providing a warm and inviting backdrop users have praised this fragrance for its fresh and invigorating essence making it perfect for daytime wear many appreciate its lightness and airy quality which is ideal for those seeking a scent that is both playful and sophisticated with a commendable rating of 375 out of 5 it has earned accolades for its delightful character and versatility appealing to a broad audience who value a fragrance that feels both chic and approachable overall little black dress eau fraiche is described as an essential contemporary scent for the modern woman effortlessly enhancing any occasion with its vibrant charm", "sentences": ["cress, lantana, castoreum, parma violet, cotton flower, oud, hesperidic notes, grape, olive tree, hyacinth, earthy notes, carambola, osmanthus, champaca, cypriol, lemon blossom, rosewood", "yuzu, clary sage, balsam fir, cedar", "passionflower, red currant, rosehip, almond blossom, chocolate"]}, {"source_sentence": "rose blush cologne 2023 by jo malone london rose blush cologne presents an enchanting bouquet that captures the essence of blooming romance and tropical vitality with an initial sweet hint of luscious litchi and a refreshing touch of herbs this fragrance unfolds into a heart of delicate rose showcasing a radiant femininity the composition is beautifully rounded off with soft musky undertones adding an elegant warmth that lingers on the skin users describe rose blush as vibrant and joyful perfect for both everyday wear and special occasions reviewers appreciate its fresh appeal heralding it as an uplifting scent that evokes feelings of spring and renewal many highlight its moderate longevity making it suitable for those who desire a fragrance that gently permeates without overwhelming whether youre seeking a burst of floral energy or a subtle whisper of sophistication this perfume is sure to leave a delightful impression", "sentences": ["honey, mahogany", "lychee, basil, rose, musk", "lemon, may rose, spices, peony, lily of the valley, blackcurrant, raspberry, peach, musk, sandalwood, amber, heliotrope, oud"]}, {"source_sentence": "thank u next by ariana grande is a playful and modern fragrance that captures the essence of youthful exuberance and selfempowerment this charming scent exudes a vibrant sweetness that dances between fruity and creamy notes creating an inviting aura that is both uplifting and comforting users often describe this perfume as deliciously sweet and fun making it perfect for casual wear or a spirited night out the blend is frequently noted for its warm inviting quality evoking a sense of cheerful nostalgia many reviewers highlight its longlasting nature and delightful sillage ensuring that its fragrant embrace stays with you throughout the day perfect for the confident contemporary woman thank u next effortlessly combines the spirited essence of fresh berries with a creamy tropical nuance which is masterfully balanced by an undercurrent of sweet indulgence overall this fragrance is celebrated for its delightful charm and is sure to make a memorable impression wherever you go", "sentences": ["cabreuva, mate, bamboo leaf, black cardamom, orris root, camellia, oriental notes, hibiscus, lily of the valley, lantana, wood notes", "sea salt, amberwood, marine notes, resins, clary sage, labdanum, white musk, blonde woods", "nectarine, olive tree, grass, cress, clementine, red apple"]}, {"source_sentence": "zara night eau de parfum envelops you in a captivating blend of softness and elegance creating a rich floral experience that feels both fresh and inviting this fragrance exudes a charming femininity where luscious floral notes mingle seamlessly with a warm creamy essence that evokes a sense of comfort users describe it as enchanting and seductive perfect for evening wear or special occasions the scent captures the essence of a night blooming with possibilities balancing the vibrancy of fresh petals with the alluring depth of sweet undertones reviewers appreciate its ability to linger gracefully on the skin leaving a trail of sophisticated allure without being overwhelming many find it to be a delightful choice for those seeking a fragrance that is both versatile and memorable with a touch of playfulness that hints at a romantic allure with a commendable rating zara night is celebrated for its accessibility and charm making it a favored addition to any perfume collection", "sentences": ["whiskey, bellini, cognac, blackberry, juniper berry, iris root, aldehydes, red currant, flint, cumin, mango, sea salt, sea notes, birch, bitter orange, marine notes, grapefruit blossom, hawthorn, yuzu, clementine, cream, pineapple", "moss, sandalwood, mangosteen, cade oil", "bergamot, galbanum, petitgrain, jasmine, narcissus, violet, carnation, rose, spices, blonde woods, iris, vanilla, amber"]}], "model-index": [{"name": "SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2", "results": [{"task": {"type": "semantic-similarity", "name": "Semantic Similarity"}, "dataset": {"name": "Unknown", "type": "unknown"}, "metrics": [{"type": "pearson_cosine", "value": 0.8425746761744255, "name": "Pearson Cosine"}, {"type": "spearman_cosine", "value": 0.718974393548417, "name": "Spearman Cosine"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION",
"SEMANTIC_SIMILARITY"
] | 45,211 |
pavanmantha/bge-base-en-bioembed768
|
pavanmantha
|
sentence-similarity
|
[
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:4247",
"loss:MatryoshkaLoss",
"loss:MultipleNegativesRankingLoss",
"en",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:1705.00652",
"base_model:BAAI/bge-base-en-v1.5",
"base_model:finetune:BAAI/bge-base-en-v1.5",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2024-08-08T09:09:00Z |
2024-08-27T17:28:17+00:00
| 9 | 0 |
---
base_model: BAAI/bge-base-en-v1.5
datasets: []
language:
- en
library_name: sentence-transformers
license: apache-2.0
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:4247
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: The Opa1 protein localizes to the mitochondria.Opa1 is found normally
in the mitochondrial intermembrane space.
sentences:
- Which is the cellular localization of the protein Opa1?
- Which are the genes responsible for Dyskeratosis Congenita?
- List blood marker for Non-Hodgkin lymphoma.
- source_sentence: CorrSite identifies potential allosteric ligand-binding sites based
on motion correlation analyses between cavities.We find that CARDS captures allosteric
communication between the two cAMP-Binding Domains (CBDs)Overall, it is demonstrated
that the communication pathways could be multiple and intrinsically disposed,
and the MC path generation approach provides an effective tool for the prediction
of key residues that mediate the allosteric communication in an ensemble of pathways
and functionally plausible residuesWe utilized a data set of 24 known allosteric
sites from 23 monomer proteins to calculate the correlations between potential
ligand-binding sites and corresponding orthosteric sites using a Gaussian network
model (GNM)Here, we introduce the Correlation of All Rotameric and Dynamical States
(CARDS) framework for quantifying correlations between both the structure and
disorder of different regions of a proteinWe present a novel method, "MutInf",
to identify statistically significant correlated motions from equilibrium molecular
dynamics simulationsCorrSite identifies potential allosteric ligand-binding sites
based on motion correlation analyses between cavities.Here, a Monte Carlo (MC)
path generation approach is proposed and implemented to define likely allosteric
pathways through generating an ensemble of maximum probability paths.Here, a Monte
Carlo (MC) path generation approach is proposed and implemented to define likely
allosteric pathways through generating an ensemble of maximum probability paths.
Overall, it is demonstrated that the communication pathways could be multiple
and intrinsically disposed, and the MC path generation approach provides an effective
tool for the prediction of key residues that mediate the allosteric communication
in an ensemble of pathways and functionally plausible residues We utilized a data
set of 24 known allosteric sites from 23 monomer proteins to calculate the correlations
between potential ligand-binding sites and corresponding orthosteric sites using
a Gaussian network model (GNM)A Monte Carlo (MC) path generation approach is proposed
and implemented to define likely allosteric pathways through generating an ensemble
of maximum probability paths. A novel method, "MutInf", to identify statistically
significant correlated motions from equilibrium molecular dynamics simulations.
CorrSite identifies potential alloster-binding sites based on motion correlation
analyses between cavities. The Correlation of All Rotameric and Dynamical States
(CARDS) framework for quantifying correlations between both the structure and
disorder of different regions of a proteinComputational tools for predicting allosteric
pathways in proteins include MCPath, MutInf, pySCA, CorrSite, and CARDS.
sentences:
- Computational tools for predicting allosteric pathways in proteins
- What is PANTHER-PSEP?
- What illness is transmitted by the Lone Star Tick, Amblyomma americanum?
- source_sentence: "Dopaminergic drugs should be given in patients with BMS. \nCatuama\
\ reduces the symptoms of BMS and may be a novel therapeutic strategy for the\
\ treatment of this disease.\nCapsaicin, alpha-lipoic acid (ALA), and clonazepam\
\ were those that showed more reduction in symptoms of BMS.\nTreatment with placebos\
\ produced a response that was 72% as large as the response to active drugs"
sentences:
- What is the cyberknife used for?
- Which compounds exist that are thyroid hormone analogs?
- Which are the drugs utilized for the burning mouth syndrome?
- source_sentence: Tinea is a superficial fungal infections of the skin.
sentences:
- Which molecule is targeted by a monoclonal antibody Mepolizumab?
- What disease is tinea ?
- Which algorithm is used for detection of long repeat expansions?
- source_sentence: Basset is an open source package which applies CNNs to learn the
functional activity of DNA sequences from genomics data. Basset was trained on
a compendium of accessible genomic sites mapped in 164 cell types by DNase-seq,
and demonstrated greater predictive accuracy than previous methods. Basset predictions
for the change in accessibility between variant alleles were far greater for Genome-wide
association study (GWAS) SNPs that are likely to be causal relative to nearby
SNPs in linkage disequilibrium with them. With Basset, a researcher can perform
a single sequencing assay in their cell type of interest and simultaneously learn
that cell's chromatin accessibility code and annotate every mutation in the genome
with its influence on present accessibility and latent potential for accessibility.
Thus, Basset offers a powerful computational approach to annotate and interpret
the noncoding genome.
sentences:
- Givosiran is used for treatment of which disease?
- Describe the applicability of Basset in the context of deep learning
- What is the causative agent of the "Panama disease" affecting bananas?
model-index:
- name: BGE base BioASQ Matryoshka
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 768
type: dim_768
metrics:
- type: cosine_accuracy@1
value: 0.8432203389830508
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.9427966101694916
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.961864406779661
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.9788135593220338
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.8432203389830508
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.3142655367231638
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.19237288135593222
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.0978813559322034
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.8432203389830508
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.9427966101694916
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.961864406779661
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.9788135593220338
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.9167805960832026
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.8963327280064567
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.8971987609787653
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 512
type: dim_512
metrics:
- type: cosine_accuracy@1
value: 0.8538135593220338
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.9427966101694916
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.961864406779661
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.9745762711864406
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.8538135593220338
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.3142655367231638
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.19237288135593222
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.09745762711864407
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.8538135593220338
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.9427966101694916
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.961864406779661
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.9745762711864406
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.9198462326957965
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.9016772598870054
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.9026755533837086
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 256
type: dim_256
metrics:
- type: cosine_accuracy@1
value: 0.8453389830508474
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.9385593220338984
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.9555084745762712
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.9745762711864406
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.8453389830508474
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.3128531073446327
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.19110169491525425
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.09745762711864407
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.8453389830508474
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.9385593220338984
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.9555084745762712
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.9745762711864406
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.914207272128957
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.8944528517621736
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.8952712251263324
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 128
type: dim_128
metrics:
- type: cosine_accuracy@1
value: 0.8220338983050848
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.9279661016949152
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.9449152542372882
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.9703389830508474
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.8220338983050848
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.3093220338983051
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.18898305084745767
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.09703389830508474
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.8220338983050848
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.9279661016949152
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.9449152542372882
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.9703389830508474
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.901534580728345
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.8789800242130752
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.8801051507894794
name: Cosine Map@100
---
# BGE base BioASQ Matryoshka
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) <!-- at revision a5beb1e3e68b9ab74eb54cfd186867f64f240e1a -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
- **Language:** en
- **License:** apache-2.0
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("pavanmantha/bge-base-en-bioembed768")
# Run inference
sentences = [
"Basset is an open source package which applies CNNs to learn the functional activity of DNA sequences from genomics data. Basset was trained on a compendium of accessible genomic sites mapped in 164 cell types by DNase-seq, and demonstrated greater predictive accuracy than previous methods. Basset predictions for the change in accessibility between variant alleles were far greater for Genome-wide association study (GWAS) SNPs that are likely to be causal relative to nearby SNPs in linkage disequilibrium with them. With Basset, a researcher can perform a single sequencing assay in their cell type of interest and simultaneously learn that cell's chromatin accessibility code and annotate every mutation in the genome with its influence on present accessibility and latent potential for accessibility. Thus, Basset offers a powerful computational approach to annotate and interpret the noncoding genome.",
'Describe the applicability of Basset in the context of deep learning',
'What is the causative agent of the "Panama disease" affecting bananas?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `dim_768`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.8432 |
| cosine_accuracy@3 | 0.9428 |
| cosine_accuracy@5 | 0.9619 |
| cosine_accuracy@10 | 0.9788 |
| cosine_precision@1 | 0.8432 |
| cosine_precision@3 | 0.3143 |
| cosine_precision@5 | 0.1924 |
| cosine_precision@10 | 0.0979 |
| cosine_recall@1 | 0.8432 |
| cosine_recall@3 | 0.9428 |
| cosine_recall@5 | 0.9619 |
| cosine_recall@10 | 0.9788 |
| cosine_ndcg@10 | 0.9168 |
| cosine_mrr@10 | 0.8963 |
| **cosine_map@100** | **0.8972** |
#### Information Retrieval
* Dataset: `dim_512`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.8538 |
| cosine_accuracy@3 | 0.9428 |
| cosine_accuracy@5 | 0.9619 |
| cosine_accuracy@10 | 0.9746 |
| cosine_precision@1 | 0.8538 |
| cosine_precision@3 | 0.3143 |
| cosine_precision@5 | 0.1924 |
| cosine_precision@10 | 0.0975 |
| cosine_recall@1 | 0.8538 |
| cosine_recall@3 | 0.9428 |
| cosine_recall@5 | 0.9619 |
| cosine_recall@10 | 0.9746 |
| cosine_ndcg@10 | 0.9198 |
| cosine_mrr@10 | 0.9017 |
| **cosine_map@100** | **0.9027** |
#### Information Retrieval
* Dataset: `dim_256`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.8453 |
| cosine_accuracy@3 | 0.9386 |
| cosine_accuracy@5 | 0.9555 |
| cosine_accuracy@10 | 0.9746 |
| cosine_precision@1 | 0.8453 |
| cosine_precision@3 | 0.3129 |
| cosine_precision@5 | 0.1911 |
| cosine_precision@10 | 0.0975 |
| cosine_recall@1 | 0.8453 |
| cosine_recall@3 | 0.9386 |
| cosine_recall@5 | 0.9555 |
| cosine_recall@10 | 0.9746 |
| cosine_ndcg@10 | 0.9142 |
| cosine_mrr@10 | 0.8945 |
| **cosine_map@100** | **0.8953** |
#### Information Retrieval
* Dataset: `dim_128`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.822 |
| cosine_accuracy@3 | 0.928 |
| cosine_accuracy@5 | 0.9449 |
| cosine_accuracy@10 | 0.9703 |
| cosine_precision@1 | 0.822 |
| cosine_precision@3 | 0.3093 |
| cosine_precision@5 | 0.189 |
| cosine_precision@10 | 0.097 |
| cosine_recall@1 | 0.822 |
| cosine_recall@3 | 0.928 |
| cosine_recall@5 | 0.9449 |
| cosine_recall@10 | 0.9703 |
| cosine_ndcg@10 | 0.9015 |
| cosine_mrr@10 | 0.879 |
| **cosine_map@100** | **0.8801** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 4,247 training samples
* Columns: <code>positive</code> and <code>anchor</code>
* Approximate statistics based on the first 1000 samples:
| | positive | anchor |
|:--------|:------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 3 tokens</li><li>mean: 102.44 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 15.78 tokens</li><li>max: 44 tokens</li></ul> |
* Samples:
| positive | anchor |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------|
| <code>Restless legs syndrome (RLS), also known as Willis-Ekbom disease (WED), is a common movement disorder characterized by an uncontrollable urge to move because of uncomfortable, sometimes painful sensations in the legs with a diurnal variation and a release with movement.</code> | <code>Willis-Ekbom disease is also known as?</code> |
| <code>Report the outcomes of laser in situ keratomileusis (LASIK) for high myopia correction after long-term follow-up['Report the outcomes of laser in situ keratomileusis (LASIK) for high myopia correction after long-term follow-up.']Laser in situ keratomileusis is also known as LASIKLaser in situ keratomileusis (LASIK)</code> | <code>What is another name for keratomileusis?</code> |
| <code>CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them.CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. CellMaps can easily be integrated in any web page by using an available JavaScript API. Computations and analyses are remotely executed in high-end servers, and all the functionalities are available through RESTful web services. CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. Computations and analyses are remotely executed in high-end servers, and all the functionalities are available through RESTful web services. CellMaps can easily be integrated in any web page by using an available JavaScript API. CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. Computations and analyses are remotely executed in high-end servers, and all the functionalities are available through RESTful web services. CellMaps can easily be integrated in any web page by using an available JavaScript API.CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. CellMaps can easily be integrated in any web page by using an available JavaScript API. Computations and analyses are remotely executed in high-end servers, and all the functionalities are available through RESTful web services.</code> | <code>What is CellMaps?</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "MultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128
],
"matryoshka_weights": [
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 16
- `learning_rate`: 2e-05
- `num_train_epochs`: 10
- `lr_scheduler_type`: cosine
- `warmup_ratio`: 0.1
- `fp16`: True
- `tf32`: False
- `load_best_model_at_end`: True
- `optim`: adamw_torch_fused
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 16
- `eval_accumulation_steps`: None
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 10
- `max_steps`: -1
- `lr_scheduler_type`: cosine
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: False
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch_fused
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | dim_128_cosine_map@100 | dim_256_cosine_map@100 | dim_512_cosine_map@100 | dim_768_cosine_map@100 |
|:----------:|:------:|:-------------:|:----------------------:|:----------------------:|:----------------------:|:----------------------:|
| 0.9624 | 8 | - | 0.8560 | 0.8821 | 0.8904 | 0.8876 |
| 1.2030 | 10 | 1.2833 | - | - | - | - |
| 1.9248 | 16 | - | 0.8655 | 0.8808 | 0.8909 | 0.8889 |
| 2.4060 | 20 | 0.4785 | - | - | - | - |
| 2.8872 | 24 | - | 0.8720 | 0.8875 | 0.8893 | 0.8921 |
| 3.6090 | 30 | 0.2417 | - | - | - | - |
| 3.9699 | 33 | - | 0.8751 | 0.8924 | 0.8955 | 0.8960 |
| 4.8120 | 40 | 0.1607 | - | - | - | - |
| 4.9323 | 41 | - | 0.8799 | 0.8932 | 0.8964 | 0.8952 |
| 5.8947 | 49 | - | 0.8785 | 0.8944 | 0.9009 | 0.8982 |
| 6.0150 | 50 | 0.1152 | - | - | - | - |
| **6.9774** | **58** | **-** | **0.8803** | **0.8947** | **0.9018** | **0.8975** |
| 7.2180 | 60 | 0.0924 | - | - | - | - |
| 7.9398 | 66 | - | 0.8802 | 0.8956 | 0.9016 | 0.8973 |
| 8.4211 | 70 | 0.0832 | - | - | - | - |
| 8.9023 | 74 | - | 0.8801 | 0.8956 | 0.9027 | 0.8972 |
| 9.6241 | 80 | 0.074 | 0.8801 | 0.8953 | 0.9027 | 0.8972 |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.10.13
- Sentence Transformers: 3.0.1
- Transformers: 4.41.2
- PyTorch: 2.1.2
- Accelerate: 0.31.0
- Datasets: 2.19.2
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MatryoshkaLoss
```bibtex
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
BioNLP
|
# BGE base BioASQ Matryoshka
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) <!-- at revision a5beb1e3e68b9ab74eb54cfd186867f64f240e1a -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
- **Language:** en
- **License:** apache-2.0
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("pavanmantha/bge-base-en-bioembed768")
# Run inference
sentences = [
"Basset is an open source package which applies CNNs to learn the functional activity of DNA sequences from genomics data. Basset was trained on a compendium of accessible genomic sites mapped in 164 cell types by DNase-seq, and demonstrated greater predictive accuracy than previous methods. Basset predictions for the change in accessibility between variant alleles were far greater for Genome-wide association study (GWAS) SNPs that are likely to be causal relative to nearby SNPs in linkage disequilibrium with them. With Basset, a researcher can perform a single sequencing assay in their cell type of interest and simultaneously learn that cell's chromatin accessibility code and annotate every mutation in the genome with its influence on present accessibility and latent potential for accessibility. Thus, Basset offers a powerful computational approach to annotate and interpret the noncoding genome.",
'Describe the applicability of Basset in the context of deep learning',
'What is the causative agent of the "Panama disease" affecting bananas?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `dim_768`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.8432 |
| cosine_accuracy@3 | 0.9428 |
| cosine_accuracy@5 | 0.9619 |
| cosine_accuracy@10 | 0.9788 |
| cosine_precision@1 | 0.8432 |
| cosine_precision@3 | 0.3143 |
| cosine_precision@5 | 0.1924 |
| cosine_precision@10 | 0.0979 |
| cosine_recall@1 | 0.8432 |
| cosine_recall@3 | 0.9428 |
| cosine_recall@5 | 0.9619 |
| cosine_recall@10 | 0.9788 |
| cosine_ndcg@10 | 0.9168 |
| cosine_mrr@10 | 0.8963 |
| **cosine_map@100** | **0.8972** |
#### Information Retrieval
* Dataset: `dim_512`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.8538 |
| cosine_accuracy@3 | 0.9428 |
| cosine_accuracy@5 | 0.9619 |
| cosine_accuracy@10 | 0.9746 |
| cosine_precision@1 | 0.8538 |
| cosine_precision@3 | 0.3143 |
| cosine_precision@5 | 0.1924 |
| cosine_precision@10 | 0.0975 |
| cosine_recall@1 | 0.8538 |
| cosine_recall@3 | 0.9428 |
| cosine_recall@5 | 0.9619 |
| cosine_recall@10 | 0.9746 |
| cosine_ndcg@10 | 0.9198 |
| cosine_mrr@10 | 0.9017 |
| **cosine_map@100** | **0.9027** |
#### Information Retrieval
* Dataset: `dim_256`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.8453 |
| cosine_accuracy@3 | 0.9386 |
| cosine_accuracy@5 | 0.9555 |
| cosine_accuracy@10 | 0.9746 |
| cosine_precision@1 | 0.8453 |
| cosine_precision@3 | 0.3129 |
| cosine_precision@5 | 0.1911 |
| cosine_precision@10 | 0.0975 |
| cosine_recall@1 | 0.8453 |
| cosine_recall@3 | 0.9386 |
| cosine_recall@5 | 0.9555 |
| cosine_recall@10 | 0.9746 |
| cosine_ndcg@10 | 0.9142 |
| cosine_mrr@10 | 0.8945 |
| **cosine_map@100** | **0.8953** |
#### Information Retrieval
* Dataset: `dim_128`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.822 |
| cosine_accuracy@3 | 0.928 |
| cosine_accuracy@5 | 0.9449 |
| cosine_accuracy@10 | 0.9703 |
| cosine_precision@1 | 0.822 |
| cosine_precision@3 | 0.3093 |
| cosine_precision@5 | 0.189 |
| cosine_precision@10 | 0.097 |
| cosine_recall@1 | 0.822 |
| cosine_recall@3 | 0.928 |
| cosine_recall@5 | 0.9449 |
| cosine_recall@10 | 0.9703 |
| cosine_ndcg@10 | 0.9015 |
| cosine_mrr@10 | 0.879 |
| **cosine_map@100** | **0.8801** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 4,247 training samples
* Columns: <code>positive</code> and <code>anchor</code>
* Approximate statistics based on the first 1000 samples:
| | positive | anchor |
|:--------|:------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 3 tokens</li><li>mean: 102.44 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 15.78 tokens</li><li>max: 44 tokens</li></ul> |
* Samples:
| positive | anchor |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------|
| <code>Restless legs syndrome (RLS), also known as Willis-Ekbom disease (WED), is a common movement disorder characterized by an uncontrollable urge to move because of uncomfortable, sometimes painful sensations in the legs with a diurnal variation and a release with movement.</code> | <code>Willis-Ekbom disease is also known as?</code> |
| <code>Report the outcomes of laser in situ keratomileusis (LASIK) for high myopia correction after long-term follow-up['Report the outcomes of laser in situ keratomileusis (LASIK) for high myopia correction after long-term follow-up.']Laser in situ keratomileusis is also known as LASIKLaser in situ keratomileusis (LASIK)</code> | <code>What is another name for keratomileusis?</code> |
| <code>CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them.CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. CellMaps can easily be integrated in any web page by using an available JavaScript API. Computations and analyses are remotely executed in high-end servers, and all the functionalities are available through RESTful web services. CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. Computations and analyses are remotely executed in high-end servers, and all the functionalities are available through RESTful web services. CellMaps can easily be integrated in any web page by using an available JavaScript API. CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. Computations and analyses are remotely executed in high-end servers, and all the functionalities are available through RESTful web services. CellMaps can easily be integrated in any web page by using an available JavaScript API.CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. CellMaps is an HTML5 open-source web tool that allows displaying, editing, exploring and analyzing biological networks as well as integrating metadata into them. CellMaps can easily be integrated in any web page by using an available JavaScript API. Computations and analyses are remotely executed in high-end servers, and all the functionalities are available through RESTful web services.</code> | <code>What is CellMaps?</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "MultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128
],
"matryoshka_weights": [
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `gradient_accumulation_steps`: 16
- `learning_rate`: 2e-05
- `num_train_epochs`: 10
- `lr_scheduler_type`: cosine
- `warmup_ratio`: 0.1
- `fp16`: True
- `tf32`: False
- `load_best_model_at_end`: True
- `optim`: adamw_torch_fused
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 16
- `eval_accumulation_steps`: None
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 10
- `max_steps`: -1
- `lr_scheduler_type`: cosine
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: False
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch_fused
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | dim_128_cosine_map@100 | dim_256_cosine_map@100 | dim_512_cosine_map@100 | dim_768_cosine_map@100 |
|:----------:|:------:|:-------------:|:----------------------:|:----------------------:|:----------------------:|:----------------------:|
| 0.9624 | 8 | - | 0.8560 | 0.8821 | 0.8904 | 0.8876 |
| 1.2030 | 10 | 1.2833 | - | - | - | - |
| 1.9248 | 16 | - | 0.8655 | 0.8808 | 0.8909 | 0.8889 |
| 2.4060 | 20 | 0.4785 | - | - | - | - |
| 2.8872 | 24 | - | 0.8720 | 0.8875 | 0.8893 | 0.8921 |
| 3.6090 | 30 | 0.2417 | - | - | - | - |
| 3.9699 | 33 | - | 0.8751 | 0.8924 | 0.8955 | 0.8960 |
| 4.8120 | 40 | 0.1607 | - | - | - | - |
| 4.9323 | 41 | - | 0.8799 | 0.8932 | 0.8964 | 0.8952 |
| 5.8947 | 49 | - | 0.8785 | 0.8944 | 0.9009 | 0.8982 |
| 6.0150 | 50 | 0.1152 | - | - | - | - |
| **6.9774** | **58** | **-** | **0.8803** | **0.8947** | **0.9018** | **0.8975** |
| 7.2180 | 60 | 0.0924 | - | - | - | - |
| 7.9398 | 66 | - | 0.8802 | 0.8956 | 0.9016 | 0.8973 |
| 8.4211 | 70 | 0.0832 | - | - | - | - |
| 8.9023 | 74 | - | 0.8801 | 0.8956 | 0.9027 | 0.8972 |
| 9.6241 | 80 | 0.074 | 0.8801 | 0.8953 | 0.9027 | 0.8972 |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.10.13
- Sentence Transformers: 3.0.1
- Transformers: 4.41.2
- PyTorch: 2.1.2
- Accelerate: 0.31.0
- Datasets: 2.19.2
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MatryoshkaLoss
```bibtex
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "BAAI/bge-base-en-v1.5", "datasets": [], "language": ["en"], "library_name": "sentence-transformers", "license": "apache-2.0", "metrics": ["cosine_accuracy@1", "cosine_accuracy@3", "cosine_accuracy@5", "cosine_accuracy@10", "cosine_precision@1", "cosine_precision@3", "cosine_precision@5", "cosine_precision@10", "cosine_recall@1", "cosine_recall@3", "cosine_recall@5", "cosine_recall@10", "cosine_ndcg@10", "cosine_mrr@10", "cosine_map@100"], "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:4247", "loss:MatryoshkaLoss", "loss:MultipleNegativesRankingLoss"], "widget": [{"source_sentence": "The Opa1 protein localizes to the mitochondria.Opa1 is found normally in the mitochondrial intermembrane space.", "sentences": ["Which is the cellular localization of the protein Opa1?", "Which are the genes responsible for Dyskeratosis Congenita?", "List blood marker for Non-Hodgkin lymphoma."]}, {"source_sentence": "CorrSite identifies potential allosteric ligand-binding sites based on motion correlation analyses between cavities.We find that CARDS captures allosteric communication between the two cAMP-Binding Domains (CBDs)Overall, it is demonstrated that the communication pathways could be multiple and intrinsically disposed, and the MC path generation approach provides an effective tool for the prediction of key residues that mediate the allosteric communication in an ensemble of pathways and functionally plausible residuesWe utilized a data set of 24 known allosteric sites from 23 monomer proteins to calculate the correlations between potential ligand-binding sites and corresponding orthosteric sites using a Gaussian network model (GNM)Here, we introduce the Correlation of All Rotameric and Dynamical States (CARDS) framework for quantifying correlations between both the structure and disorder of different regions of a proteinWe present a novel method, \"MutInf\", to identify statistically significant correlated motions from equilibrium molecular dynamics simulationsCorrSite identifies potential allosteric ligand-binding sites based on motion correlation analyses between cavities.Here, a Monte Carlo (MC) path generation approach is proposed and implemented to define likely allosteric pathways through generating an ensemble of maximum probability paths.Here, a Monte Carlo (MC) path generation approach is proposed and implemented to define likely allosteric pathways through generating an ensemble of maximum probability paths. Overall, it is demonstrated that the communication pathways could be multiple and intrinsically disposed, and the MC path generation approach provides an effective tool for the prediction of key residues that mediate the allosteric communication in an ensemble of pathways and functionally plausible residues We utilized a data set of 24 known allosteric sites from 23 monomer proteins to calculate the correlations between potential ligand-binding sites and corresponding orthosteric sites using a Gaussian network model (GNM)A Monte Carlo (MC) path generation approach is proposed and implemented to define likely allosteric pathways through generating an ensemble of maximum probability paths. A novel method, \"MutInf\", to identify statistically significant correlated motions from equilibrium molecular dynamics simulations. CorrSite identifies potential alloster-binding sites based on motion correlation analyses between cavities. The Correlation of All Rotameric and Dynamical States (CARDS) framework for quantifying correlations between both the structure and disorder of different regions of a proteinComputational tools for predicting allosteric pathways in proteins include MCPath, MutInf, pySCA, CorrSite, and CARDS.", "sentences": ["Computational tools for predicting allosteric pathways in proteins", "What is PANTHER-PSEP?", "What illness is transmitted by the Lone Star Tick, Amblyomma americanum?"]}, {"source_sentence": "Dopaminergic drugs should be given in patients with BMS. \nCatuama reduces the symptoms of BMS and may be a novel therapeutic strategy for the treatment of this disease.\nCapsaicin, alpha-lipoic acid (ALA), and clonazepam were those that showed more reduction in symptoms of BMS.\nTreatment with placebos produced a response that was 72% as large as the response to active drugs", "sentences": ["What is the cyberknife used for?", "Which compounds exist that are thyroid hormone analogs?", "Which are the drugs utilized for the burning mouth syndrome?"]}, {"source_sentence": "Tinea is a superficial fungal infections of the skin.", "sentences": ["Which molecule is targeted by a monoclonal antibody Mepolizumab?", "What disease is tinea ?", "Which algorithm is used for detection of long repeat expansions?"]}, {"source_sentence": "Basset is an open source package which applies CNNs to learn the functional activity of DNA sequences from genomics data. Basset was trained on a compendium of accessible genomic sites mapped in 164 cell types by DNase-seq, and demonstrated greater predictive accuracy than previous methods. Basset predictions for the change in accessibility between variant alleles were far greater for Genome-wide association study (GWAS) SNPs that are likely to be causal relative to nearby SNPs in linkage disequilibrium with them. With Basset, a researcher can perform a single sequencing assay in their cell type of interest and simultaneously learn that cell's chromatin accessibility code and annotate every mutation in the genome with its influence on present accessibility and latent potential for accessibility. Thus, Basset offers a powerful computational approach to annotate and interpret the noncoding genome.", "sentences": ["Givosiran is used for treatment of which disease?", "Describe the applicability of Basset in the context of deep learning", "What is the causative agent of the \"Panama disease\" affecting bananas?"]}], "model-index": [{"name": "BGE base BioASQ Matryoshka", "results": [{"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 768", "type": "dim_768"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.8432203389830508, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.9427966101694916, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.961864406779661, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.9788135593220338, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.8432203389830508, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.3142655367231638, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.19237288135593222, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.0978813559322034, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.8432203389830508, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.9427966101694916, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.961864406779661, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.9788135593220338, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.9167805960832026, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.8963327280064567, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.8971987609787653, "name": "Cosine Map@100"}]}, {"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 512", "type": "dim_512"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.8538135593220338, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.9427966101694916, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.961864406779661, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.9745762711864406, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.8538135593220338, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.3142655367231638, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.19237288135593222, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.09745762711864407, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.8538135593220338, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.9427966101694916, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.961864406779661, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.9745762711864406, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.9198462326957965, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.9016772598870054, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.9026755533837086, "name": "Cosine Map@100"}]}, {"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 256", "type": "dim_256"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.8453389830508474, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.9385593220338984, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.9555084745762712, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.9745762711864406, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.8453389830508474, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.3128531073446327, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.19110169491525425, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.09745762711864407, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.8453389830508474, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.9385593220338984, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.9555084745762712, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.9745762711864406, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.914207272128957, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.8944528517621736, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.8952712251263324, "name": "Cosine Map@100"}]}, {"task": {"type": "information-retrieval", "name": "Information Retrieval"}, "dataset": {"name": "dim 128", "type": "dim_128"}, "metrics": [{"type": "cosine_accuracy@1", "value": 0.8220338983050848, "name": "Cosine Accuracy@1"}, {"type": "cosine_accuracy@3", "value": 0.9279661016949152, "name": "Cosine Accuracy@3"}, {"type": "cosine_accuracy@5", "value": 0.9449152542372882, "name": "Cosine Accuracy@5"}, {"type": "cosine_accuracy@10", "value": 0.9703389830508474, "name": "Cosine Accuracy@10"}, {"type": "cosine_precision@1", "value": 0.8220338983050848, "name": "Cosine Precision@1"}, {"type": "cosine_precision@3", "value": 0.3093220338983051, "name": "Cosine Precision@3"}, {"type": "cosine_precision@5", "value": 0.18898305084745767, "name": "Cosine Precision@5"}, {"type": "cosine_precision@10", "value": 0.09703389830508474, "name": "Cosine Precision@10"}, {"type": "cosine_recall@1", "value": 0.8220338983050848, "name": "Cosine Recall@1"}, {"type": "cosine_recall@3", "value": 0.9279661016949152, "name": "Cosine Recall@3"}, {"type": "cosine_recall@5", "value": 0.9449152542372882, "name": "Cosine Recall@5"}, {"type": "cosine_recall@10", "value": 0.9703389830508474, "name": "Cosine Recall@10"}, {"type": "cosine_ndcg@10", "value": 0.901534580728345, "name": "Cosine Ndcg@10"}, {"type": "cosine_mrr@10", "value": 0.8789800242130752, "name": "Cosine Mrr@10"}, {"type": "cosine_map@100", "value": 0.8801051507894794, "name": "Cosine Map@100"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,213 |
tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF
|
tensorblock
|
text-generation
|
[
"transformers",
"gguf",
"TensorBlock",
"GGUF",
"text-generation",
"ko",
"en",
"jp",
"cn",
"dataset:Saxo/ko_cn_translation_tech_social_science_linkbricks_single_dataset",
"dataset:Saxo/ko_jp_translation_tech_social_science_linkbricks_single_dataset",
"dataset:Saxo/en_ko_translation_tech_science_linkbricks_single_dataset_with_prompt_text_huggingface",
"dataset:Saxo/en_ko_translation_social_science_linkbricks_single_dataset_with_prompt_text_huggingface",
"dataset:Saxo/ko_aspect_sentiment_sns_mall_sentiment_linkbricks_single_dataset_with_prompt_text_huggingface",
"dataset:Saxo/ko_summarization_linkbricks_single_dataset_with_prompt_text_huggingface",
"dataset:Saxo/OpenOrca_cleaned_kor_linkbricks_single_dataset_with_prompt_text_huggingface",
"dataset:Saxo/ko_government_qa_total_linkbricks_single_dataset_with_prompt_text_huggingface_sampled",
"dataset:Saxo/ko-news-corpus-1",
"dataset:Saxo/ko-news-corpus-2",
"dataset:Saxo/ko-news-corpus-3",
"dataset:Saxo/ko-news-corpus-4",
"dataset:Saxo/ko-news-corpus-5",
"dataset:Saxo/ko-news-corpus-6",
"dataset:Saxo/ko-news-corpus-7",
"dataset:Saxo/ko-news-corpus-8",
"dataset:Saxo/ko-news-corpus-9",
"dataset:maywell/ko_Ultrafeedback_binarized",
"dataset:youjunhyeok/ko-orca-pair-and-ultrafeedback-dpo",
"dataset:lilacai/glaive-function-calling-v2-sharegpt",
"dataset:kuotient/gsm8k-ko",
"base_model:Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B",
"base_model:quantized:Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | 2024-11-20T08:25:04Z |
2024-11-20T09:07:17+00:00
| 115 | 0 |
---
base_model: Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B
datasets:
- Saxo/ko_cn_translation_tech_social_science_linkbricks_single_dataset
- Saxo/ko_jp_translation_tech_social_science_linkbricks_single_dataset
- Saxo/en_ko_translation_tech_science_linkbricks_single_dataset_with_prompt_text_huggingface
- Saxo/en_ko_translation_social_science_linkbricks_single_dataset_with_prompt_text_huggingface
- Saxo/ko_aspect_sentiment_sns_mall_sentiment_linkbricks_single_dataset_with_prompt_text_huggingface
- Saxo/ko_summarization_linkbricks_single_dataset_with_prompt_text_huggingface
- Saxo/OpenOrca_cleaned_kor_linkbricks_single_dataset_with_prompt_text_huggingface
- Saxo/ko_government_qa_total_linkbricks_single_dataset_with_prompt_text_huggingface_sampled
- Saxo/ko-news-corpus-1
- Saxo/ko-news-corpus-2
- Saxo/ko-news-corpus-3
- Saxo/ko-news-corpus-4
- Saxo/ko-news-corpus-5
- Saxo/ko-news-corpus-6
- Saxo/ko-news-corpus-7
- Saxo/ko-news-corpus-8
- Saxo/ko-news-corpus-9
- maywell/ko_Ultrafeedback_binarized
- youjunhyeok/ko-orca-pair-and-ultrafeedback-dpo
- lilacai/glaive-function-calling-v2-sharegpt
- kuotient/gsm8k-ko
language:
- ko
- en
- jp
- cn
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
tags:
- TensorBlock
- GGUF
---
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;">
Feedback and support: TensorBlock's <a href="https://x.com/tensorblock_aoi">Twitter/X</a>, <a href="https://t.me/TensorBlock">Telegram Group</a> and <a href="https://x.com/tensorblock_aoi">Discord server</a>
</p>
</div>
</div>
## Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B - GGUF
This repo contains GGUF format model files for [Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B](https://huggingface.co/Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B).
The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
<div style="text-align: left; margin: 20px 0;">
<a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
Run them on the TensorBlock client using your local machine ↗
</a>
</div>
## Prompt template
```
<s>[INST]{system_prompt}
{prompt}[/INST]
```
## Model file specification
| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q2_K.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q2_K.gguf) | Q2_K | 4.791 GB | smallest, significant quality loss - not recommended for most purposes |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_S.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_S.gguf) | Q3_K_S | 5.534 GB | very small, high quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_M.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_M.gguf) | Q3_K_M | 6.083 GB | very small, high quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_L.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_L.gguf) | Q3_K_L | 6.562 GB | small, substantial quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_0.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_0.gguf) | Q4_0 | 7.072 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_K_S.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_K_S.gguf) | Q4_K_S | 7.120 GB | small, greater quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_K_M.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_K_M.gguf) | Q4_K_M | 7.477 GB | medium, balanced quality - recommended |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_0.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_0.gguf) | Q5_0 | 8.519 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_K_S.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_K_S.gguf) | Q5_K_S | 8.519 GB | large, low quality loss - recommended |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_K_M.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_K_M.gguf) | Q5_K_M | 8.728 GB | large, very low quality loss - recommended |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q6_K.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q6_K.gguf) | Q6_K | 10.056 GB | very large, extremely low quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q8_0.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q8_0.gguf) | Q8_0 | 13.022 GB | very large, extremely low quality loss - not recommended |
## Downloading instruction
### Command line
Firstly, install Huggingface Client
```shell
pip install -U "huggingface_hub[cli]"
```
Then, downoad the individual model file the a local directory
```shell
huggingface-cli download tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF --include "Linkbricks-Horizon-AI-Korean-Pro-12B-Q2_K.gguf" --local-dir MY_LOCAL_DIR
```
If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:
```shell
huggingface-cli download tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
| null |
Non_BioNLP
|
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;">
Feedback and support: TensorBlock's <a href="https://x.com/tensorblock_aoi">Twitter/X</a>, <a href="https://t.me/TensorBlock">Telegram Group</a> and <a href="https://x.com/tensorblock_aoi">Discord server</a>
</p>
</div>
</div>
## Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B - GGUF
This repo contains GGUF format model files for [Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B](https://huggingface.co/Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B).
The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
<div style="text-align: left; margin: 20px 0;">
<a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
Run them on the TensorBlock client using your local machine ↗
</a>
</div>
## Prompt template
```
<s>[INST]{system_prompt}
{prompt}[/INST]
```
## Model file specification
| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q2_K.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q2_K.gguf) | Q2_K | 4.791 GB | smallest, significant quality loss - not recommended for most purposes |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_S.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_S.gguf) | Q3_K_S | 5.534 GB | very small, high quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_M.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_M.gguf) | Q3_K_M | 6.083 GB | very small, high quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_L.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q3_K_L.gguf) | Q3_K_L | 6.562 GB | small, substantial quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_0.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_0.gguf) | Q4_0 | 7.072 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_K_S.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_K_S.gguf) | Q4_K_S | 7.120 GB | small, greater quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_K_M.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q4_K_M.gguf) | Q4_K_M | 7.477 GB | medium, balanced quality - recommended |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_0.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_0.gguf) | Q5_0 | 8.519 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_K_S.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_K_S.gguf) | Q5_K_S | 8.519 GB | large, low quality loss - recommended |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_K_M.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q5_K_M.gguf) | Q5_K_M | 8.728 GB | large, very low quality loss - recommended |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q6_K.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q6_K.gguf) | Q6_K | 10.056 GB | very large, extremely low quality loss |
| [Linkbricks-Horizon-AI-Korean-Pro-12B-Q8_0.gguf](https://huggingface.co/tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF/blob/main/Linkbricks-Horizon-AI-Korean-Pro-12B-Q8_0.gguf) | Q8_0 | 13.022 GB | very large, extremely low quality loss - not recommended |
## Downloading instruction
### Command line
Firstly, install Huggingface Client
```shell
pip install -U "huggingface_hub[cli]"
```
Then, downoad the individual model file the a local directory
```shell
huggingface-cli download tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF --include "Linkbricks-Horizon-AI-Korean-Pro-12B-Q2_K.gguf" --local-dir MY_LOCAL_DIR
```
If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:
```shell
huggingface-cli download tensorblock/Linkbricks-Horizon-AI-Korean-Pro-12B-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
|
{"base_model": "Saxo/Linkbricks-Horizon-AI-Korean-Pro-12B", "datasets": ["Saxo/ko_cn_translation_tech_social_science_linkbricks_single_dataset", "Saxo/ko_jp_translation_tech_social_science_linkbricks_single_dataset", "Saxo/en_ko_translation_tech_science_linkbricks_single_dataset_with_prompt_text_huggingface", "Saxo/en_ko_translation_social_science_linkbricks_single_dataset_with_prompt_text_huggingface", "Saxo/ko_aspect_sentiment_sns_mall_sentiment_linkbricks_single_dataset_with_prompt_text_huggingface", "Saxo/ko_summarization_linkbricks_single_dataset_with_prompt_text_huggingface", "Saxo/OpenOrca_cleaned_kor_linkbricks_single_dataset_with_prompt_text_huggingface", "Saxo/ko_government_qa_total_linkbricks_single_dataset_with_prompt_text_huggingface_sampled", "Saxo/ko-news-corpus-1", "Saxo/ko-news-corpus-2", "Saxo/ko-news-corpus-3", "Saxo/ko-news-corpus-4", "Saxo/ko-news-corpus-5", "Saxo/ko-news-corpus-6", "Saxo/ko-news-corpus-7", "Saxo/ko-news-corpus-8", "Saxo/ko-news-corpus-9", "maywell/ko_Ultrafeedback_binarized", "youjunhyeok/ko-orca-pair-and-ultrafeedback-dpo", "lilacai/glaive-function-calling-v2-sharegpt", "kuotient/gsm8k-ko"], "language": ["ko", "en", "jp", "cn"], "library_name": "transformers", "license": "apache-2.0", "pipeline_tag": "text-generation", "tags": ["TensorBlock", "GGUF"]}
|
task
|
[
"TRANSLATION",
"SUMMARIZATION"
] | 45,214 |
machinev/model
|
machinev
|
sentence-similarity
|
[
"sentence-transformers",
"safetensors",
"clip",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:12",
"loss:MultipleNegativesRankingLoss",
"dataset:machinev/multimodalLPT2",
"arxiv:1908.10084",
"arxiv:1705.00652",
"base_model:sentence-transformers/clip-ViT-L-14",
"base_model:finetune:sentence-transformers/clip-ViT-L-14",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2025-03-07T11:38:28Z |
2025-03-07T11:39:40+00:00
| 10 | 0 |
---
base_model: sentence-transformers/clip-ViT-L-14
datasets:
- machinev/multimodalLPT2
library_name: sentence-transformers
metrics:
- cosine_accuracy
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:12
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: 'the main power cable is connected with LPT '
sentences:
- 'the main power cable is connected with LPT '
- 'the main power cable is connected with LPT '
- /content/sample_data/images/LPT (2).jpeg
- source_sentence: 'the fuse is not blown it is working properly '
sentences:
- 'the fuse is not blown it is working properly '
- 'the fuse is not blown it is working properly '
- /content/sample_data/images/LPT (16).jpeg
- source_sentence: 'the fuse is blown and this might not work properly '
sentences:
- /content/sample_data/images/LPT (20).jpeg
- 'the fuse is blown and this might not work properly '
- 'the fuse is blown and this might not work properly '
- source_sentence: 'the fuse is blown and this might not work properly '
sentences:
- 'the fuse is blown and this might not work properly '
- /content/sample_data/images/LPT (21).jpeg
- 'the fuse is blown and this might not work properly '
- source_sentence: 'the main power cable is not connected with LPT '
sentences:
- 'the main power cable is not connected with LPT '
- /content/sample_data/images/LPT (4).jpeg
- 'the main power cable is not connected with LPT '
model-index:
- name: SentenceTransformer based on sentence-transformers/clip-ViT-L-14
results:
- task:
type: triplet
name: Triplet
dataset:
name: yt title thumbnail train
type: yt-title-thumbnail-train
metrics:
- type: cosine_accuracy
value: 0.0
name: Cosine Accuracy
- task:
type: triplet
name: Triplet
dataset:
name: yt title thumbnail validation
type: yt-title-thumbnail-validation
metrics:
- type: cosine_accuracy
value: 0.0
name: Cosine Accuracy
---
# SentenceTransformer based on sentence-transformers/clip-ViT-L-14
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/clip-ViT-L-14](https://huggingface.co/sentence-transformers/clip-ViT-L-14) on the [multimodal_lpt2](https://huggingface.co/datasets/machinev/multimodalLPT2) dataset. It maps sentences & paragraphs to a None-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [sentence-transformers/clip-ViT-L-14](https://huggingface.co/sentence-transformers/clip-ViT-L-14) <!-- at revision 3b12140ad0f9750045e404f187cfccd04bcaf250 -->
- **Maximum Sequence Length:** None tokens
- **Output Dimensionality:** None dimensions
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- [multimodal_lpt2](https://huggingface.co/datasets/machinev/multimodalLPT2)
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): CLIPModel()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("machinev/model")
# Run inference
sentences = [
'the main power cable is not connected with LPT ',
'/content/sample_data/images/LPT (4).jpeg',
'the main power cable is not connected with LPT ',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Triplet
* Datasets: `yt-title-thumbnail-train` and `yt-title-thumbnail-validation`
* Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator)
| Metric | yt-title-thumbnail-train | yt-title-thumbnail-validation |
|:--------------------|:-------------------------|:------------------------------|
| **cosine_accuracy** | **0.0** | **0.0** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### multimodal_lpt2
* Dataset: [multimodal_lpt2](https://huggingface.co/datasets/machinev/multimodalLPT2) at [9f649f9](https://huggingface.co/datasets/machinev/multimodalLPT2/tree/9f649f9c95cc375b7ec5895fb47f642f251d288e)
* Size: 12 training samples
* Columns: <code>text</code>, <code>image_path</code>, <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 12 samples:
| | text | image_path | anchor | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | PIL.JpegImagePlugin.JpegImageFile | string | string |
| details | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 18.42 tokens</li><li>max: 19 tokens</li></ul> | <ul><li></li></ul> | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> |
* Samples:
| text | image_path | anchor | positive | negative |
|:-------------------------------------------------------------|:------------------------------------------------------|:-------------------------------------------------------------------------------------------------|:-------------------------------------------------------------|:-------------------------------------------------------------|
| <code>the main power cable is not connected with LPT </code> | <code>/content/sample_data/images/LPT (1).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D40680FFFD0></code> | <code>the main power cable is not connected with LPT </code> | <code>the main power cable is not connected with LPT </code> |
| <code>the main power cable is connected with LPT </code> | <code>/content/sample_data/images/LPT (2).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D40680FDF90></code> | <code>the main power cable is connected with LPT </code> | <code>the main power cable is connected with LPT </code> |
| <code>the main power cable is connected with LPT </code> | <code>/content/sample_data/images/LPT (3).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D4063F4C610></code> | <code>the main power cable is connected with LPT </code> | <code>the main power cable is connected with LPT </code> |
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Evaluation Dataset
#### multimodal_lpt2
* Dataset: [multimodal_lpt2](https://huggingface.co/datasets/machinev/multimodalLPT2) at [9f649f9](https://huggingface.co/datasets/machinev/multimodalLPT2/tree/9f649f9c95cc375b7ec5895fb47f642f251d288e)
* Size: 12 evaluation samples
* Columns: <code>text</code>, <code>image_path</code>, <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 12 samples:
| | text | image_path | anchor | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | PIL.JpegImagePlugin.JpegImageFile | string | string |
| details | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 18.42 tokens</li><li>max: 19 tokens</li></ul> | <ul><li></li></ul> | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> |
* Samples:
| text | image_path | anchor | positive | negative |
|:-------------------------------------------------------------|:------------------------------------------------------|:-------------------------------------------------------------------------------------------------|:-------------------------------------------------------------|:-------------------------------------------------------------|
| <code>the main power cable is not connected with LPT </code> | <code>/content/sample_data/images/LPT (1).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D4063B84B50></code> | <code>the main power cable is not connected with LPT </code> | <code>the main power cable is not connected with LPT </code> |
| <code>the main power cable is connected with LPT </code> | <code>/content/sample_data/images/LPT (2).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D4063F4D190></code> | <code>the main power cable is connected with LPT </code> | <code>the main power cable is connected with LPT </code> |
| <code>the main power cable is connected with LPT </code> | <code>/content/sample_data/images/LPT (3).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D4063F4C7D0></code> | <code>the main power cable is connected with LPT </code> | <code>the main power cable is connected with LPT </code> |
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `learning_rate`: 0.0001
- `num_train_epochs`: 2
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 0.0001
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 2
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | Validation Loss | yt-title-thumbnail-train_cosine_accuracy | yt-title-thumbnail-validation_cosine_accuracy |
|:-----:|:----:|:-------------:|:---------------:|:----------------------------------------:|:---------------------------------------------:|
| -1 | -1 | - | - | 0.0 | 0.0 |
| 1.0 | 1 | 8.5381 | 7.5693 | - | - |
| 2.0 | 2 | 7.5693 | 7.1228 | - | - |
### Framework Versions
- Python: 3.11.11
- Sentence Transformers: 3.4.1
- Transformers: 4.48.3
- PyTorch: 2.5.1+cu124
- Accelerate: 1.3.0
- Datasets: 3.3.2
- Tokenizers: 0.21.0
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# SentenceTransformer based on sentence-transformers/clip-ViT-L-14
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/clip-ViT-L-14](https://huggingface.co/sentence-transformers/clip-ViT-L-14) on the [multimodal_lpt2](https://huggingface.co/datasets/machinev/multimodalLPT2) dataset. It maps sentences & paragraphs to a None-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [sentence-transformers/clip-ViT-L-14](https://huggingface.co/sentence-transformers/clip-ViT-L-14) <!-- at revision 3b12140ad0f9750045e404f187cfccd04bcaf250 -->
- **Maximum Sequence Length:** None tokens
- **Output Dimensionality:** None dimensions
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- [multimodal_lpt2](https://huggingface.co/datasets/machinev/multimodalLPT2)
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): CLIPModel()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("machinev/model")
# Run inference
sentences = [
'the main power cable is not connected with LPT ',
'/content/sample_data/images/LPT (4).jpeg',
'the main power cable is not connected with LPT ',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Triplet
* Datasets: `yt-title-thumbnail-train` and `yt-title-thumbnail-validation`
* Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator)
| Metric | yt-title-thumbnail-train | yt-title-thumbnail-validation |
|:--------------------|:-------------------------|:------------------------------|
| **cosine_accuracy** | **0.0** | **0.0** |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### multimodal_lpt2
* Dataset: [multimodal_lpt2](https://huggingface.co/datasets/machinev/multimodalLPT2) at [9f649f9](https://huggingface.co/datasets/machinev/multimodalLPT2/tree/9f649f9c95cc375b7ec5895fb47f642f251d288e)
* Size: 12 training samples
* Columns: <code>text</code>, <code>image_path</code>, <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 12 samples:
| | text | image_path | anchor | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | PIL.JpegImagePlugin.JpegImageFile | string | string |
| details | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 18.42 tokens</li><li>max: 19 tokens</li></ul> | <ul><li></li></ul> | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> |
* Samples:
| text | image_path | anchor | positive | negative |
|:-------------------------------------------------------------|:------------------------------------------------------|:-------------------------------------------------------------------------------------------------|:-------------------------------------------------------------|:-------------------------------------------------------------|
| <code>the main power cable is not connected with LPT </code> | <code>/content/sample_data/images/LPT (1).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D40680FFFD0></code> | <code>the main power cable is not connected with LPT </code> | <code>the main power cable is not connected with LPT </code> |
| <code>the main power cable is connected with LPT </code> | <code>/content/sample_data/images/LPT (2).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D40680FDF90></code> | <code>the main power cable is connected with LPT </code> | <code>the main power cable is connected with LPT </code> |
| <code>the main power cable is connected with LPT </code> | <code>/content/sample_data/images/LPT (3).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D4063F4C610></code> | <code>the main power cable is connected with LPT </code> | <code>the main power cable is connected with LPT </code> |
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Evaluation Dataset
#### multimodal_lpt2
* Dataset: [multimodal_lpt2](https://huggingface.co/datasets/machinev/multimodalLPT2) at [9f649f9](https://huggingface.co/datasets/machinev/multimodalLPT2/tree/9f649f9c95cc375b7ec5895fb47f642f251d288e)
* Size: 12 evaluation samples
* Columns: <code>text</code>, <code>image_path</code>, <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 12 samples:
| | text | image_path | anchor | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | PIL.JpegImagePlugin.JpegImageFile | string | string |
| details | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 18.42 tokens</li><li>max: 19 tokens</li></ul> | <ul><li></li></ul> | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 11.42 tokens</li><li>max: 12 tokens</li></ul> |
* Samples:
| text | image_path | anchor | positive | negative |
|:-------------------------------------------------------------|:------------------------------------------------------|:-------------------------------------------------------------------------------------------------|:-------------------------------------------------------------|:-------------------------------------------------------------|
| <code>the main power cable is not connected with LPT </code> | <code>/content/sample_data/images/LPT (1).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D4063B84B50></code> | <code>the main power cable is not connected with LPT </code> | <code>the main power cable is not connected with LPT </code> |
| <code>the main power cable is connected with LPT </code> | <code>/content/sample_data/images/LPT (2).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D4063F4D190></code> | <code>the main power cable is connected with LPT </code> | <code>the main power cable is connected with LPT </code> |
| <code>the main power cable is connected with LPT </code> | <code>/content/sample_data/images/LPT (3).jpeg</code> | <code><PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=3024x4032 at 0x7D4063F4C7D0></code> | <code>the main power cable is connected with LPT </code> | <code>the main power cable is connected with LPT </code> |
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `learning_rate`: 0.0001
- `num_train_epochs`: 2
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 0.0001
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 2
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | Validation Loss | yt-title-thumbnail-train_cosine_accuracy | yt-title-thumbnail-validation_cosine_accuracy |
|:-----:|:----:|:-------------:|:---------------:|:----------------------------------------:|:---------------------------------------------:|
| -1 | -1 | - | - | 0.0 | 0.0 |
| 1.0 | 1 | 8.5381 | 7.5693 | - | - |
| 2.0 | 2 | 7.5693 | 7.1228 | - | - |
### Framework Versions
- Python: 3.11.11
- Sentence Transformers: 3.4.1
- Transformers: 4.48.3
- PyTorch: 2.5.1+cu124
- Accelerate: 1.3.0
- Datasets: 3.3.2
- Tokenizers: 0.21.0
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "sentence-transformers/clip-ViT-L-14", "datasets": ["machinev/multimodalLPT2"], "library_name": "sentence-transformers", "metrics": ["cosine_accuracy"], "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:12", "loss:MultipleNegativesRankingLoss"], "widget": [{"source_sentence": "the main power cable is connected with LPT ", "sentences": ["the main power cable is connected with LPT ", "the main power cable is connected with LPT ", "/content/sample_data/images/LPT (2).jpeg"]}, {"source_sentence": "the fuse is not blown it is working properly ", "sentences": ["the fuse is not blown it is working properly ", "the fuse is not blown it is working properly ", "/content/sample_data/images/LPT (16).jpeg"]}, {"source_sentence": "the fuse is blown and this might not work properly ", "sentences": ["/content/sample_data/images/LPT (20).jpeg", "the fuse is blown and this might not work properly ", "the fuse is blown and this might not work properly "]}, {"source_sentence": "the fuse is blown and this might not work properly ", "sentences": ["the fuse is blown and this might not work properly ", "/content/sample_data/images/LPT (21).jpeg", "the fuse is blown and this might not work properly "]}, {"source_sentence": "the main power cable is not connected with LPT ", "sentences": ["the main power cable is not connected with LPT ", "/content/sample_data/images/LPT (4).jpeg", "the main power cable is not connected with LPT "]}], "model-index": [{"name": "SentenceTransformer based on sentence-transformers/clip-ViT-L-14", "results": [{"task": {"type": "triplet", "name": "Triplet"}, "dataset": {"name": "yt title thumbnail train", "type": "yt-title-thumbnail-train"}, "metrics": [{"type": "cosine_accuracy", "value": 0.0, "name": "Cosine Accuracy"}]}, {"task": {"type": "triplet", "name": "Triplet"}, "dataset": {"name": "yt title thumbnail validation", "type": "yt-title-thumbnail-validation"}, "metrics": [{"type": "cosine_accuracy", "value": 0.0, "name": "Cosine Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,215 |
gokuls/distilroberta-emotion-intent
|
gokuls
|
text-classification
|
[
"transformers",
"pytorch",
"tensorboard",
"roberta",
"text-classification",
"generated_from_trainer",
"dataset:emotion",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2022-09-24T13:26:33Z |
2022-09-24T13:36:17+00:00
| 12 | 1 |
---
datasets:
- emotion
license: apache-2.0
metrics:
- accuracy
tags:
- generated_from_trainer
model-index:
- name: distilroberta-emotion-intent
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: emotion
type: emotion
config: default
split: train
args: default
metrics:
- type: accuracy
value: 0.9435
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilroberta-emotion-intent
This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilroberta-base) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1496
- Accuracy: 0.9435
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 33
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.4501 | 1.0 | 1000 | 0.2432 | 0.924 |
| 0.1947 | 2.0 | 2000 | 0.1646 | 0.934 |
| 0.1497 | 3.0 | 3000 | 0.1382 | 0.9405 |
| 0.1316 | 4.0 | 4000 | 0.1496 | 0.9435 |
| 0.1145 | 5.0 | 5000 | 0.1684 | 0.9385 |
| 0.1 | 6.0 | 6000 | 0.2342 | 0.943 |
| 0.0828 | 7.0 | 7000 | 0.2807 | 0.939 |
### Framework versions
- Transformers 4.22.1
- Pytorch 1.12.1+cu113
- Datasets 2.5.1
- Tokenizers 0.12.1
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilroberta-emotion-intent
This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilroberta-base) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1496
- Accuracy: 0.9435
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 33
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.4501 | 1.0 | 1000 | 0.2432 | 0.924 |
| 0.1947 | 2.0 | 2000 | 0.1646 | 0.934 |
| 0.1497 | 3.0 | 3000 | 0.1382 | 0.9405 |
| 0.1316 | 4.0 | 4000 | 0.1496 | 0.9435 |
| 0.1145 | 5.0 | 5000 | 0.1684 | 0.9385 |
| 0.1 | 6.0 | 6000 | 0.2342 | 0.943 |
| 0.0828 | 7.0 | 7000 | 0.2807 | 0.939 |
### Framework versions
- Transformers 4.22.1
- Pytorch 1.12.1+cu113
- Datasets 2.5.1
- Tokenizers 0.12.1
|
{"datasets": ["emotion"], "license": "apache-2.0", "metrics": ["accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "distilroberta-emotion-intent", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "emotion", "type": "emotion", "config": "default", "split": "train", "args": "default"}, "metrics": [{"type": "accuracy", "value": 0.9435, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,216 |
QuantFactory/granite-3.0-8b-instruct-GGUF
|
QuantFactory
|
text-generation
|
[
"transformers",
"gguf",
"language",
"granite-3.0",
"text-generation",
"arxiv:0000.00000",
"license:apache-2.0",
"model-index",
"region:us",
"conversational"
] | 2024-10-21T14:13:35Z |
2024-10-21T20:28:10+00:00
| 1,557 | 2 |
---
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
tags:
- language
- granite-3.0
inference: false
model-index:
- name: granite-3.0-2b-instruct
results:
- task:
type: text-generation
dataset:
name: IFEval
type: instruction-following
metrics:
- type: pass@1
value: 52.27
name: pass@1
- type: pass@1
value: 8.22
name: pass@1
- task:
type: text-generation
dataset:
name: AGI-Eval
type: human-exams
metrics:
- type: pass@1
value: 40.52
name: pass@1
- type: pass@1
value: 65.82
name: pass@1
- type: pass@1
value: 34.45
name: pass@1
- task:
type: text-generation
dataset:
name: OBQA
type: commonsense
metrics:
- type: pass@1
value: 46.6
name: pass@1
- type: pass@1
value: 71.21
name: pass@1
- type: pass@1
value: 82.61
name: pass@1
- type: pass@1
value: 77.51
name: pass@1
- type: pass@1
value: 60.32
name: pass@1
- task:
type: text-generation
dataset:
name: BoolQ
type: reading-comprehension
metrics:
- type: pass@1
value: 88.65
name: pass@1
- type: pass@1
value: 21.58
name: pass@1
- task:
type: text-generation
dataset:
name: ARC-C
type: reasoning
metrics:
- type: pass@1
value: 64.16
name: pass@1
- type: pass@1
value: 33.81
name: pass@1
- type: pass@1
value: 51.55
name: pass@1
- task:
type: text-generation
dataset:
name: HumanEvalSynthesis
type: code
metrics:
- type: pass@1
value: 64.63
name: pass@1
- type: pass@1
value: 57.16
name: pass@1
- type: pass@1
value: 65.85
name: pass@1
- type: pass@1
value: 49.6
name: pass@1
- task:
type: text-generation
dataset:
name: GSM8K
type: math
metrics:
- type: pass@1
value: 68.99
name: pass@1
- type: pass@1
value: 30.94
name: pass@1
- task:
type: text-generation
dataset:
name: PAWS-X (7 langs)
type: multilingual
metrics:
- type: pass@1
value: 64.94
name: pass@1
- type: pass@1
value: 48.2
name: pass@1
---
[](https://hf.co/QuantFactory)
# QuantFactory/granite-3.0-8b-instruct-GGUF
This is quantized version of [ibm-granite/granite-3.0-8b-instruct](https://huggingface.co/ibm-granite/granite-3.0-8b-instruct) created using llama.cpp
# Original Model Card
<!--  -->
<!--  -->
# Granite-3.0-8B-Instruct
**Model Summary:**
Granite-3.0-8B-Instruct is a 8B parameter model finetuned from *Granite-3.0-8B-Base* using a combination of open source instruction datasets with permissive license and internally collected synthetic datasets. This model is developed using a diverse set of techniques with a structured chat format, including supervised finetuning, model alignment using reinforcement learning, and model merging.
- **Developers:** Granite Team, IBM
- **GitHub Repository:** [ibm-granite/granite-3.0-language-models](https://github.com/ibm-granite/granite-3.0-language-models)
- **Website**: [Granite Docs](https://www.ibm.com/granite/docs/)
- **Paper:** [Granite 3.0 Language Models](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/paper.pdf)
- **Release Date**: October 21st, 2024
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
**Supported Languages:**
English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, and Chinese. Users may finetune Granite 3.0 models for languages beyond these 12 languages.
**Intended use:**
The model is designed to respond to general instructions and can be used to build AI assistants for multiple domains, including business applications.
*Capabilities*
* Summarization
* Text classification
* Text extraction
* Question-answering
* Retrieval Augmented Generation (RAG)
* Code related tasks
* Function-calling tasks
* Multilingual dialog use cases
**Generation:**
This is a simple example of how to use Granite-3.0-8B-Instruct model.
Install the following libraries:
```shell
pip install torch torchvision torchaudio
pip install accelerate
pip install transformers
```
Then, copy the snippet from the section that is relevant for your use case.
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "auto"
model_path = "ibm-granite/granite-3.0-8b-instruct"
tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()
# change input text as desired
chat = [
{ "role": "user", "content": "Please list one IBM Research laboratory located in the United States. You should only output its name and location." },
]
chat = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
# tokenize the text
input_tokens = tokenizer(chat, return_tensors="pt").to(device)
# generate output tokens
output = model.generate(**input_tokens,
max_new_tokens=100)
# decode output tokens into text
output = tokenizer.batch_decode(output)
# print output
print(output)
```
**Model Architecture:**
Granite-3.0-8B-Instruct is based on a decoder-only dense transformer architecture. Core components of this architecture are: GQA and RoPE, MLP with SwiGLU, RMSNorm, and shared input/output embeddings.
| Model | 2B Dense | 8B Dense | 1B MoE | 3B MoE |
| :-------- | :--------| :-------- | :------| :------|
| Embedding size | 2048 | **4096** | 1024 | 1536 |
| Number of layers | 40 | **40** | 24 | 32 |
| Attention head size | 64 | **128** | 64 | 64 |
| Number of attention heads | 32 | **32** | 16 | 24 |
| Number of KV heads | 8 | **8** | 8 | 8 |
| MLP hidden size | 8192 | **12800** | 512 | 512 |
| MLP activation | SwiGLU | **SwiGLU** | SwiGLU | SwiGLU |
| Number of Experts | — | **—** | 32 | 40 |
| MoE TopK | — | **—** | 8 | 8 |
| Initialization std | 0.1 | **0.1** | 0.1 | 0.1 |
| Sequence Length | 4096 | **4096** | 4096 | 4096 |
| Position Embedding | RoPE | **RoPE** | RoPE | RoPE |
| # Paremeters | 2.5B | **8.1B** | 1.3B | 3.3B |
| # Active Parameters | 2.5B | **8.1B** | 400M | 800M |
| # Training tokens | 12T | **12T** | 10T | 10T |
**Training Data:**
Overall, our SFT data is largely comprised of three key sources: (1) publicly available datasets with permissive license, (2) internal synthetic data targeting specific capabilities, and (3) very small amounts of human-curated data. A detailed attribution of datasets can be found in the [Granite Technical Report](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/paper.pdf) and [Accompanying Author List](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/author-ack.pdf).
**Infrastructure:**
We train Granite 3.0 Language Models using IBM's super computing cluster, Blue Vela, which is outfitted with NVIDIA H100 GPUs. This cluster provides a scalable and efficient infrastructure for training our models over thousands of GPUs while minimizing environmental impact by utilizing 100% renewable energy sources.
**Ethical Considerations and Limitations:**
Granite 3.0 Instruct Models are primarily finetuned using instruction-response pairs mostly in English, but also multilingual data covering eleven languages. Although this model can handle multilingual dialog use cases, its performance might not be similar to English tasks. In such case, introducing a small number of examples (few-shot) can help the model in generating more accurate outputs. While this model has been aligned by keeping safety in consideration, the model may in some cases produce inaccurate, biased, or unsafe responses to user prompts. So we urge the community to use this model with proper safety testing and tuning tailored for their specific tasks.
<!-- ## Citation
```
@misc{granite-models,
author = {author 1, author2, ...},
title = {},
journal = {},
volume = {},
year = {2024},
url = {https://arxiv.org/abs/0000.00000},
}
``` -->
| null |
Non_BioNLP
|
[](https://hf.co/QuantFactory)
# QuantFactory/granite-3.0-8b-instruct-GGUF
This is quantized version of [ibm-granite/granite-3.0-8b-instruct](https://huggingface.co/ibm-granite/granite-3.0-8b-instruct) created using llama.cpp
# Original Model Card
<!--  -->
<!--  -->
# Granite-3.0-8B-Instruct
**Model Summary:**
Granite-3.0-8B-Instruct is a 8B parameter model finetuned from *Granite-3.0-8B-Base* using a combination of open source instruction datasets with permissive license and internally collected synthetic datasets. This model is developed using a diverse set of techniques with a structured chat format, including supervised finetuning, model alignment using reinforcement learning, and model merging.
- **Developers:** Granite Team, IBM
- **GitHub Repository:** [ibm-granite/granite-3.0-language-models](https://github.com/ibm-granite/granite-3.0-language-models)
- **Website**: [Granite Docs](https://www.ibm.com/granite/docs/)
- **Paper:** [Granite 3.0 Language Models](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/paper.pdf)
- **Release Date**: October 21st, 2024
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
**Supported Languages:**
English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, and Chinese. Users may finetune Granite 3.0 models for languages beyond these 12 languages.
**Intended use:**
The model is designed to respond to general instructions and can be used to build AI assistants for multiple domains, including business applications.
*Capabilities*
* Summarization
* Text classification
* Text extraction
* Question-answering
* Retrieval Augmented Generation (RAG)
* Code related tasks
* Function-calling tasks
* Multilingual dialog use cases
**Generation:**
This is a simple example of how to use Granite-3.0-8B-Instruct model.
Install the following libraries:
```shell
pip install torch torchvision torchaudio
pip install accelerate
pip install transformers
```
Then, copy the snippet from the section that is relevant for your use case.
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "auto"
model_path = "ibm-granite/granite-3.0-8b-instruct"
tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()
# change input text as desired
chat = [
{ "role": "user", "content": "Please list one IBM Research laboratory located in the United States. You should only output its name and location." },
]
chat = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
# tokenize the text
input_tokens = tokenizer(chat, return_tensors="pt").to(device)
# generate output tokens
output = model.generate(**input_tokens,
max_new_tokens=100)
# decode output tokens into text
output = tokenizer.batch_decode(output)
# print output
print(output)
```
**Model Architecture:**
Granite-3.0-8B-Instruct is based on a decoder-only dense transformer architecture. Core components of this architecture are: GQA and RoPE, MLP with SwiGLU, RMSNorm, and shared input/output embeddings.
| Model | 2B Dense | 8B Dense | 1B MoE | 3B MoE |
| :-------- | :--------| :-------- | :------| :------|
| Embedding size | 2048 | **4096** | 1024 | 1536 |
| Number of layers | 40 | **40** | 24 | 32 |
| Attention head size | 64 | **128** | 64 | 64 |
| Number of attention heads | 32 | **32** | 16 | 24 |
| Number of KV heads | 8 | **8** | 8 | 8 |
| MLP hidden size | 8192 | **12800** | 512 | 512 |
| MLP activation | SwiGLU | **SwiGLU** | SwiGLU | SwiGLU |
| Number of Experts | — | **—** | 32 | 40 |
| MoE TopK | — | **—** | 8 | 8 |
| Initialization std | 0.1 | **0.1** | 0.1 | 0.1 |
| Sequence Length | 4096 | **4096** | 4096 | 4096 |
| Position Embedding | RoPE | **RoPE** | RoPE | RoPE |
| # Paremeters | 2.5B | **8.1B** | 1.3B | 3.3B |
| # Active Parameters | 2.5B | **8.1B** | 400M | 800M |
| # Training tokens | 12T | **12T** | 10T | 10T |
**Training Data:**
Overall, our SFT data is largely comprised of three key sources: (1) publicly available datasets with permissive license, (2) internal synthetic data targeting specific capabilities, and (3) very small amounts of human-curated data. A detailed attribution of datasets can be found in the [Granite Technical Report](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/paper.pdf) and [Accompanying Author List](https://github.com/ibm-granite/granite-3.0-language-models/blob/main/author-ack.pdf).
**Infrastructure:**
We train Granite 3.0 Language Models using IBM's super computing cluster, Blue Vela, which is outfitted with NVIDIA H100 GPUs. This cluster provides a scalable and efficient infrastructure for training our models over thousands of GPUs while minimizing environmental impact by utilizing 100% renewable energy sources.
**Ethical Considerations and Limitations:**
Granite 3.0 Instruct Models are primarily finetuned using instruction-response pairs mostly in English, but also multilingual data covering eleven languages. Although this model can handle multilingual dialog use cases, its performance might not be similar to English tasks. In such case, introducing a small number of examples (few-shot) can help the model in generating more accurate outputs. While this model has been aligned by keeping safety in consideration, the model may in some cases produce inaccurate, biased, or unsafe responses to user prompts. So we urge the community to use this model with proper safety testing and tuning tailored for their specific tasks.
<!-- ## Citation
```
@misc{granite-models,
author = {author 1, author2, ...},
title = {},
journal = {},
volume = {},
year = {2024},
url = {https://arxiv.org/abs/0000.00000},
}
``` -->
|
{"library_name": "transformers", "license": "apache-2.0", "pipeline_tag": "text-generation", "tags": ["language", "granite-3.0"], "inference": false, "model-index": [{"name": "granite-3.0-2b-instruct", "results": [{"task": {"type": "text-generation"}, "dataset": {"name": "IFEval", "type": "instruction-following"}, "metrics": [{"type": "pass@1", "value": 52.27, "name": "pass@1"}, {"type": "pass@1", "value": 8.22, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "AGI-Eval", "type": "human-exams"}, "metrics": [{"type": "pass@1", "value": 40.52, "name": "pass@1"}, {"type": "pass@1", "value": 65.82, "name": "pass@1"}, {"type": "pass@1", "value": 34.45, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "OBQA", "type": "commonsense"}, "metrics": [{"type": "pass@1", "value": 46.6, "name": "pass@1"}, {"type": "pass@1", "value": 71.21, "name": "pass@1"}, {"type": "pass@1", "value": 82.61, "name": "pass@1"}, {"type": "pass@1", "value": 77.51, "name": "pass@1"}, {"type": "pass@1", "value": 60.32, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "BoolQ", "type": "reading-comprehension"}, "metrics": [{"type": "pass@1", "value": 88.65, "name": "pass@1"}, {"type": "pass@1", "value": 21.58, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "ARC-C", "type": "reasoning"}, "metrics": [{"type": "pass@1", "value": 64.16, "name": "pass@1"}, {"type": "pass@1", "value": 33.81, "name": "pass@1"}, {"type": "pass@1", "value": 51.55, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "HumanEvalSynthesis", "type": "code"}, "metrics": [{"type": "pass@1", "value": 64.63, "name": "pass@1"}, {"type": "pass@1", "value": 57.16, "name": "pass@1"}, {"type": "pass@1", "value": 65.85, "name": "pass@1"}, {"type": "pass@1", "value": 49.6, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "GSM8K", "type": "math"}, "metrics": [{"type": "pass@1", "value": 68.99, "name": "pass@1"}, {"type": "pass@1", "value": 30.94, "name": "pass@1"}]}, {"task": {"type": "text-generation"}, "dataset": {"name": "PAWS-X (7 langs)", "type": "multilingual"}, "metrics": [{"type": "pass@1", "value": 64.94, "name": "pass@1"}, {"type": "pass@1", "value": 48.2, "name": "pass@1"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION",
"SUMMARIZATION"
] | 45,217 |
BoscoTheDog/TowerInstruct-7B-v0.1-q4f32_1-MLC
|
BoscoTheDog
|
translation
|
[
"translation",
"en",
"de",
"fr",
"es",
"pt",
"it",
"ru",
"ko",
"nl",
"license:llama2",
"region:us"
] | 2024-02-10T15:08:06Z |
2024-02-11T21:13:40+00:00
| 0 | 0 |
---
language:
- en
- de
- fr
- es
- pt
- it
- ru
- ko
- nl
license: llama2
pipeline_tag: translation
---
A WASM version of Tower
| null |
Non_BioNLP
|
A WASM version of Tower
|
{"language": ["en", "de", "fr", "es", "pt", "it", "ru", "ko", "nl"], "license": "llama2", "pipeline_tag": "translation"}
|
task
|
[
"TRANSLATION"
] | 45,218 |
Hampusswe/ai
|
Hampusswe
| null |
[
"pytorch",
"tf",
"jax",
"rust",
"safetensors",
"distilbert",
"exbert",
"en",
"dataset:bookcorpus",
"dataset:wikipedia",
"arxiv:1910.01108",
"license:apache-2.0",
"region:us"
] | 2024-12-25T21:03:25Z |
2024-12-25T21:07:47+00:00
| 0 | 0 |
---
datasets:
- bookcorpus
- wikipedia
language: en
license: apache-2.0
tags:
- exbert
---
# DistilBERT base model (uncased)
This model is a distilled version of the [BERT base model](https://huggingface.co/bert-base-uncased). It was
introduced in [this paper](https://arxiv.org/abs/1910.01108). The code for the distillation process can be found
[here](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation). This model is uncased: it does
not make a difference between english and English.
## Model description
DistilBERT is a transformers model, smaller and faster than BERT, which was pretrained on the same corpus in a
self-supervised fashion, using the BERT base model as a teacher. This means it was pretrained on the raw texts only,
with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic
process to generate inputs and labels from those texts using the BERT base model. More precisely, it was pretrained
with three objectives:
- Distillation loss: the model was trained to return the same probabilities as the BERT base model.
- Masked language modeling (MLM): this is part of the original training loss of the BERT base model. When taking a
sentence, the model randomly masks 15% of the words in the input then run the entire masked sentence through the
model and has to predict the masked words. This is different from traditional recurrent neural networks (RNNs) that
usually see the words one after the other, or from autoregressive models like GPT which internally mask the future
tokens. It allows the model to learn a bidirectional representation of the sentence.
- Cosine embedding loss: the model was also trained to generate hidden states as close as possible as the BERT base
model.
This way, the model learns the same inner representation of the English language than its teacher model, while being
faster for inference or downstream tasks.
## Intended uses & limitations
You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to
be fine-tuned on a downstream task. See the [model hub](https://huggingface.co/models?filter=distilbert) to look for
fine-tuned versions on a task that interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked)
to make decisions, such as sequence classification, token classification or question answering. For tasks such as text
generation you should look at model like GPT2.
### How to use
You can use this model directly with a pipeline for masked language modeling:
```python
>>> from transformers import pipeline
>>> unmasker = pipeline('fill-mask', model='distilbert-base-uncased')
>>> unmasker("Hello I'm a [MASK] model.")
[{'sequence': "[CLS] hello i'm a role model. [SEP]",
'score': 0.05292855575680733,
'token': 2535,
'token_str': 'role'},
{'sequence': "[CLS] hello i'm a fashion model. [SEP]",
'score': 0.03968575969338417,
'token': 4827,
'token_str': 'fashion'},
{'sequence': "[CLS] hello i'm a business model. [SEP]",
'score': 0.034743521362543106,
'token': 2449,
'token_str': 'business'},
{'sequence': "[CLS] hello i'm a model model. [SEP]",
'score': 0.03462274372577667,
'token': 2944,
'token_str': 'model'},
{'sequence': "[CLS] hello i'm a modeling model. [SEP]",
'score': 0.018145186826586723,
'token': 11643,
'token_str': 'modeling'}]
```
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import DistilBertTokenizer, DistilBertModel
tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
model = DistilBertModel.from_pretrained("distilbert-base-uncased")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import DistilBertTokenizer, TFDistilBertModel
tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
model = TFDistilBertModel.from_pretrained("distilbert-base-uncased")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
### Limitations and bias
Even if the training data used for this model could be characterized as fairly neutral, this model can have biased
predictions. It also inherits some of
[the bias of its teacher model](https://huggingface.co/bert-base-uncased#limitations-and-bias).
```python
>>> from transformers import pipeline
>>> unmasker = pipeline('fill-mask', model='distilbert-base-uncased')
>>> unmasker("The White man worked as a [MASK].")
[{'sequence': '[CLS] the white man worked as a blacksmith. [SEP]',
'score': 0.1235365942120552,
'token': 20987,
'token_str': 'blacksmith'},
{'sequence': '[CLS] the white man worked as a carpenter. [SEP]',
'score': 0.10142576694488525,
'token': 10533,
'token_str': 'carpenter'},
{'sequence': '[CLS] the white man worked as a farmer. [SEP]',
'score': 0.04985016956925392,
'token': 7500,
'token_str': 'farmer'},
{'sequence': '[CLS] the white man worked as a miner. [SEP]',
'score': 0.03932540491223335,
'token': 18594,
'token_str': 'miner'},
{'sequence': '[CLS] the white man worked as a butcher. [SEP]',
'score': 0.03351764753460884,
'token': 14998,
'token_str': 'butcher'}]
>>> unmasker("The Black woman worked as a [MASK].")
[{'sequence': '[CLS] the black woman worked as a waitress. [SEP]',
'score': 0.13283951580524445,
'token': 13877,
'token_str': 'waitress'},
{'sequence': '[CLS] the black woman worked as a nurse. [SEP]',
'score': 0.12586183845996857,
'token': 6821,
'token_str': 'nurse'},
{'sequence': '[CLS] the black woman worked as a maid. [SEP]',
'score': 0.11708822101354599,
'token': 10850,
'token_str': 'maid'},
{'sequence': '[CLS] the black woman worked as a prostitute. [SEP]',
'score': 0.11499975621700287,
'token': 19215,
'token_str': 'prostitute'},
{'sequence': '[CLS] the black woman worked as a housekeeper. [SEP]',
'score': 0.04722772538661957,
'token': 22583,
'token_str': 'housekeeper'}]
```
This bias will also affect all fine-tuned versions of this model.
## Training data
DistilBERT pretrained on the same data as BERT, which is [BookCorpus](https://yknzhu.wixsite.com/mbweb), a dataset
consisting of 11,038 unpublished books and [English Wikipedia](https://en.wikipedia.org/wiki/English_Wikipedia)
(excluding lists, tables and headers).
## Training procedure
### Preprocessing
The texts are lowercased and tokenized using WordPiece and a vocabulary size of 30,000. The inputs of the model are
then of the form:
```
[CLS] Sentence A [SEP] Sentence B [SEP]
```
With probability 0.5, sentence A and sentence B correspond to two consecutive sentences in the original corpus and in
the other cases, it's another random sentence in the corpus. Note that what is considered a sentence here is a
consecutive span of text usually longer than a single sentence. The only constrain is that the result with the two
"sentences" has a combined length of less than 512 tokens.
The details of the masking procedure for each sentence are the following:
- 15% of the tokens are masked.
- In 80% of the cases, the masked tokens are replaced by `[MASK]`.
- In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace.
- In the 10% remaining cases, the masked tokens are left as is.
### Pretraining
The model was trained on 8 16 GB V100 for 90 hours. See the
[training code](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation) for all hyperparameters
details.
## Evaluation results
When fine-tuned on downstream tasks, this model achieves the following results:
Glue test results:
| Task | MNLI | QQP | QNLI | SST-2 | CoLA | STS-B | MRPC | RTE |
|:----:|:----:|:----:|:----:|:-----:|:----:|:-----:|:----:|:----:|
| | 82.2 | 88.5 | 89.2 | 91.3 | 51.3 | 85.8 | 87.5 | 59.9 |
### BibTeX entry and citation info
```bibtex
@article{Sanh2019DistilBERTAD,
title={DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter},
author={Victor Sanh and Lysandre Debut and Julien Chaumond and Thomas Wolf},
journal={ArXiv},
year={2019},
volume={abs/1910.01108}
}
```
<a href="https://huggingface.co/exbert/?model=distilbert-base-uncased">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
| null |
Non_BioNLP
|
# DistilBERT base model (uncased)
This model is a distilled version of the [BERT base model](https://huggingface.co/bert-base-uncased). It was
introduced in [this paper](https://arxiv.org/abs/1910.01108). The code for the distillation process can be found
[here](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation). This model is uncased: it does
not make a difference between english and English.
## Model description
DistilBERT is a transformers model, smaller and faster than BERT, which was pretrained on the same corpus in a
self-supervised fashion, using the BERT base model as a teacher. This means it was pretrained on the raw texts only,
with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic
process to generate inputs and labels from those texts using the BERT base model. More precisely, it was pretrained
with three objectives:
- Distillation loss: the model was trained to return the same probabilities as the BERT base model.
- Masked language modeling (MLM): this is part of the original training loss of the BERT base model. When taking a
sentence, the model randomly masks 15% of the words in the input then run the entire masked sentence through the
model and has to predict the masked words. This is different from traditional recurrent neural networks (RNNs) that
usually see the words one after the other, or from autoregressive models like GPT which internally mask the future
tokens. It allows the model to learn a bidirectional representation of the sentence.
- Cosine embedding loss: the model was also trained to generate hidden states as close as possible as the BERT base
model.
This way, the model learns the same inner representation of the English language than its teacher model, while being
faster for inference or downstream tasks.
## Intended uses & limitations
You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to
be fine-tuned on a downstream task. See the [model hub](https://huggingface.co/models?filter=distilbert) to look for
fine-tuned versions on a task that interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked)
to make decisions, such as sequence classification, token classification or question answering. For tasks such as text
generation you should look at model like GPT2.
### How to use
You can use this model directly with a pipeline for masked language modeling:
```python
>>> from transformers import pipeline
>>> unmasker = pipeline('fill-mask', model='distilbert-base-uncased')
>>> unmasker("Hello I'm a [MASK] model.")
[{'sequence': "[CLS] hello i'm a role model. [SEP]",
'score': 0.05292855575680733,
'token': 2535,
'token_str': 'role'},
{'sequence': "[CLS] hello i'm a fashion model. [SEP]",
'score': 0.03968575969338417,
'token': 4827,
'token_str': 'fashion'},
{'sequence': "[CLS] hello i'm a business model. [SEP]",
'score': 0.034743521362543106,
'token': 2449,
'token_str': 'business'},
{'sequence': "[CLS] hello i'm a model model. [SEP]",
'score': 0.03462274372577667,
'token': 2944,
'token_str': 'model'},
{'sequence': "[CLS] hello i'm a modeling model. [SEP]",
'score': 0.018145186826586723,
'token': 11643,
'token_str': 'modeling'}]
```
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import DistilBertTokenizer, DistilBertModel
tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
model = DistilBertModel.from_pretrained("distilbert-base-uncased")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import DistilBertTokenizer, TFDistilBertModel
tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased')
model = TFDistilBertModel.from_pretrained("distilbert-base-uncased")
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
### Limitations and bias
Even if the training data used for this model could be characterized as fairly neutral, this model can have biased
predictions. It also inherits some of
[the bias of its teacher model](https://huggingface.co/bert-base-uncased#limitations-and-bias).
```python
>>> from transformers import pipeline
>>> unmasker = pipeline('fill-mask', model='distilbert-base-uncased')
>>> unmasker("The White man worked as a [MASK].")
[{'sequence': '[CLS] the white man worked as a blacksmith. [SEP]',
'score': 0.1235365942120552,
'token': 20987,
'token_str': 'blacksmith'},
{'sequence': '[CLS] the white man worked as a carpenter. [SEP]',
'score': 0.10142576694488525,
'token': 10533,
'token_str': 'carpenter'},
{'sequence': '[CLS] the white man worked as a farmer. [SEP]',
'score': 0.04985016956925392,
'token': 7500,
'token_str': 'farmer'},
{'sequence': '[CLS] the white man worked as a miner. [SEP]',
'score': 0.03932540491223335,
'token': 18594,
'token_str': 'miner'},
{'sequence': '[CLS] the white man worked as a butcher. [SEP]',
'score': 0.03351764753460884,
'token': 14998,
'token_str': 'butcher'}]
>>> unmasker("The Black woman worked as a [MASK].")
[{'sequence': '[CLS] the black woman worked as a waitress. [SEP]',
'score': 0.13283951580524445,
'token': 13877,
'token_str': 'waitress'},
{'sequence': '[CLS] the black woman worked as a nurse. [SEP]',
'score': 0.12586183845996857,
'token': 6821,
'token_str': 'nurse'},
{'sequence': '[CLS] the black woman worked as a maid. [SEP]',
'score': 0.11708822101354599,
'token': 10850,
'token_str': 'maid'},
{'sequence': '[CLS] the black woman worked as a prostitute. [SEP]',
'score': 0.11499975621700287,
'token': 19215,
'token_str': 'prostitute'},
{'sequence': '[CLS] the black woman worked as a housekeeper. [SEP]',
'score': 0.04722772538661957,
'token': 22583,
'token_str': 'housekeeper'}]
```
This bias will also affect all fine-tuned versions of this model.
## Training data
DistilBERT pretrained on the same data as BERT, which is [BookCorpus](https://yknzhu.wixsite.com/mbweb), a dataset
consisting of 11,038 unpublished books and [English Wikipedia](https://en.wikipedia.org/wiki/English_Wikipedia)
(excluding lists, tables and headers).
## Training procedure
### Preprocessing
The texts are lowercased and tokenized using WordPiece and a vocabulary size of 30,000. The inputs of the model are
then of the form:
```
[CLS] Sentence A [SEP] Sentence B [SEP]
```
With probability 0.5, sentence A and sentence B correspond to two consecutive sentences in the original corpus and in
the other cases, it's another random sentence in the corpus. Note that what is considered a sentence here is a
consecutive span of text usually longer than a single sentence. The only constrain is that the result with the two
"sentences" has a combined length of less than 512 tokens.
The details of the masking procedure for each sentence are the following:
- 15% of the tokens are masked.
- In 80% of the cases, the masked tokens are replaced by `[MASK]`.
- In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace.
- In the 10% remaining cases, the masked tokens are left as is.
### Pretraining
The model was trained on 8 16 GB V100 for 90 hours. See the
[training code](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation) for all hyperparameters
details.
## Evaluation results
When fine-tuned on downstream tasks, this model achieves the following results:
Glue test results:
| Task | MNLI | QQP | QNLI | SST-2 | CoLA | STS-B | MRPC | RTE |
|:----:|:----:|:----:|:----:|:-----:|:----:|:-----:|:----:|:----:|
| | 82.2 | 88.5 | 89.2 | 91.3 | 51.3 | 85.8 | 87.5 | 59.9 |
### BibTeX entry and citation info
```bibtex
@article{Sanh2019DistilBERTAD,
title={DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter},
author={Victor Sanh and Lysandre Debut and Julien Chaumond and Thomas Wolf},
journal={ArXiv},
year={2019},
volume={abs/1910.01108}
}
```
<a href="https://huggingface.co/exbert/?model=distilbert-base-uncased">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
{"datasets": ["bookcorpus", "wikipedia"], "language": "en", "license": "apache-2.0", "tags": ["exbert"]}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,219 |
Jane0410/autotrain-ebfvz-n33rh
|
Jane0410
|
text-classification
|
[
"transformers",
"tensorboard",
"safetensors",
"bert",
"text-classification",
"autotrain",
"base_model:google-bert/bert-base-uncased",
"base_model:finetune:google-bert/bert-base-uncased",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-10-11T16:03:51Z |
2024-10-11T16:07:45+00:00
| 4 | 0 |
---
base_model: google-bert/bert-base-uncased
library_name: transformers
tags:
- autotrain
- text-classification
widget:
- text: I love AutoTrain
---
# Model Trained Using AutoTrain
- Problem type: Text Classification
## Validation Metrics
loss: 0.6668644547462463
f1: 0.6666666666666666
precision: 0.5
recall: 1.0
auc: 1.0
accuracy: 0.5
| null |
Non_BioNLP
|
# Model Trained Using AutoTrain
- Problem type: Text Classification
## Validation Metrics
loss: 0.6668644547462463
f1: 0.6666666666666666
precision: 0.5
recall: 1.0
auc: 1.0
accuracy: 0.5
|
{"base_model": "google-bert/bert-base-uncased", "library_name": "transformers", "tags": ["autotrain", "text-classification"], "widget": [{"text": "I love AutoTrain"}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,220 |
inventbot/Phi-3-vision-128k-instruct
|
inventbot
|
text-generation
|
[
"transformers",
"safetensors",
"phi3_v",
"text-generation",
"nlp",
"code",
"vision",
"conversational",
"custom_code",
"multilingual",
"license:mit",
"autotrain_compatible",
"region:us"
] | 2024-06-15T14:04:40Z |
2024-06-15T14:30:44+00:00
| 8 | 0 |
---
language:
- multilingual
license: mit
license_link: https://huggingface.co/microsoft/Phi-3-vision-128k-instruct/resolve/main/LICENSE
pipeline_tag: text-generation
tags:
- nlp
- code
- vision
inference:
parameters:
temperature: 0.7
widget:
- messages:
- role: user
content: <|image_1|>Can you describe what you see in the image?
---
## Model Summary
The Phi-3-Vision-128K-Instruct is a lightweight, state-of-the-art open multimodal model built upon datasets which include - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data both on text and vision. The model belongs to the Phi-3 model family, and the multimodal version comes with 128K context length (in tokens) it can support. The model underwent a rigorous enhancement process, incorporating both supervised fine-tuning and direct preference optimization to ensure precise instruction adherence and robust safety measures.
Resources and Technical Documentation:
+ [Phi-3 Microsoft Blog](https://aka.ms/Phi-3Build2024)
+ [Phi-3 Technical Report](https://aka.ms/phi3-tech-report)
+ [Phi-3 on Azure AI Studio](https://aka.ms/try-phi3vision)
+ [Phi-3 Cookbook](https://github.com/microsoft/Phi-3CookBook)
| | Short Context | Long Context |
| ------- | ------------- | ------------ |
| Mini | 4K [[HF]](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-onnx) ; [[GGUF]](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-gguf) | 128K [[HF]](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct-onnx)|
| Small | 8K [[HF]](https://huggingface.co/microsoft/Phi-3-small-8k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-small-8k-instruct-onnx-cuda) | 128K [[HF]](https://huggingface.co/microsoft/Phi-3-small-128k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-small-128k-instruct-onnx-cuda)|
| Medium | 4K [[HF]](https://huggingface.co/microsoft/Phi-3-medium-4k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-medium-4k-instruct-onnx-cuda) | 128K [[HF]](https://huggingface.co/microsoft/Phi-3-medium-128k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-medium-128k-instruct-onnx-cuda)|
| Vision | | 128K [[HF]](https://huggingface.co/microsoft/Phi-3-vision-128k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-vision-128k-instruct-onnx-cuda)|
## Intended Uses
**Primary use cases**
The model is intended for broad commercial and research use in English. The model provides uses for general purpose AI systems and applications with visual and text input capabilities which require
1) memory/compute constrained environments;
2) latency bound scenarios;
3) general image understanding;
4) OCR;
5) chart and table understanding.
Our model is designed to accelerate research on efficient language and multimodal models, for use as a building block for generative AI powered features.
**Use case considerations**
Our models are not specifically designed or evaluated for all downstream purposes. Developers should consider common limitations of language models as they select use cases, and evaluate and mitigate for accuracy, safety, and fairness before using within a specific downstream use case, particularly for high-risk scenarios.
Developers should be aware of and adhere to applicable laws or regulations (including privacy, trade compliance laws, etc.) that are relevant to their use case.
Nothing contained in this Model Card should be interpreted as or deemed a restriction or modification to the license the model is released under.
## How to Use
Phi-3-Vision-128K-Instruct has been integrated in the development version (4.40.2) of `transformers`. Until the official version is released through `pip`, ensure that you are doing one of the following:
* When loading the model, ensure that `trust_remote_code=True` is passed as an argument of the `from_pretrained()` function.
* Update your local `transformers` to the development version: `pip uninstall -y transformers && pip install git+https://github.com/huggingface/transformers`. The previous command is an alternative to cloning and installing from the source.
The current `transformers` version can be verified with: `pip list | grep transformers`.
Examples of required packages:
```
flash_attn==2.5.8
numpy==1.24.4
Pillow==10.3.0
Requests==2.31.0
torch==2.3.0
torchvision==0.18.0
transformers==4.40.2
```
Phi-3-Vision-128K-Instruct is also available in [Azure AI Studio](https://aka.ms/phi3-azure-ai).
### Chat Format
Given the nature of the training data, the Phi-3-Vision-128K-Instruct model is best suited for a single image input wih prompts using the chat format as follows.
You can provide the prompt as a single image with a generic template as follow:
```markdown
<|user|>\n<|image_1|>\n{prompt}<|end|>\n<|assistant|>\n
```
where the model generates the text after `<|assistant|>` . In case of multi-turn conversation, the prompt can be formatted as follows:
```markdown
<|user|>\n<|image_1|>\n{prompt_1}<|end|>\n<|assistant|>\n{response_1}<|end|>\n<|user|>\n{prompt_2}<|end|>\n<|assistant|>\n
```
### Sample inference code
This code snippets show how to get quickly started with running the model on a GPU:
```python
from PIL import Image
import requests
from transformers import AutoModelForCausalLM
from transformers import AutoProcessor
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="cuda", trust_remote_code=True, torch_dtype="auto", _attn_implementation='flash_attention_2') # use _attn_implementation='eager' to disable flash attention
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
messages = [
{"role": "user", "content": "<|image_1|>\nWhat is shown in this image?"},
{"role": "assistant", "content": "The chart displays the percentage of respondents who agree with various statements about their preparedness for meetings. It shows five categories: 'Having clear and pre-defined goals for meetings', 'Knowing where to find the information I need for a meeting', 'Understanding my exact role and responsibilities when I'm invited', 'Having tools to manage admin tasks like note-taking or summarization', and 'Having more focus time to sufficiently prepare for meetings'. Each category has an associated bar indicating the level of agreement, measured on a scale from 0% to 100%."},
{"role": "user", "content": "Provide insightful questions to spark discussion."}
]
url = "https://assets-c4akfrf5b4d3f4b7.z01.azurefd.net/assets/2024/04/BMDataViz_661fb89f3845e.png"
image = Image.open(requests.get(url, stream=True).raw)
prompt = processor.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = processor(prompt, [image], return_tensors="pt").to("cuda:0")
generation_args = {
"max_new_tokens": 500,
"temperature": 0.0,
"do_sample": False,
}
generate_ids = model.generate(**inputs, eos_token_id=processor.tokenizer.eos_token_id, **generation_args)
# remove input tokens
generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
response = processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
print(response)
```
Additional basic examples are provided [here](https://huggingface.co/microsoft/Phi-3-vision-128k-instruct/blob/main/sample_inference.py).
## Responsible AI Considerations
Like other models, the Phi family of models can potentially behave in ways that are unfair, unreliable, or offensive. Some of the limiting behaviors to be aware of include:
+ Quality of Service: The Phi models are trained primarily on English text. Languages other than English will experience worse performance. English language varieties with less representation in the training data might experience worse performance than standard American English.
+ Representation of Harms & Perpetuation of Stereotypes: These models can over- or under-represent groups of people, erase representation of some groups, or reinforce demeaning or negative stereotypes. Despite safety post-training, these limitations may still be present due to differing levels of representation of different groups or prevalence of examples of negative stereotypes in training data that reflect real-world patterns and societal biases.
+ Inappropriate or Offensive Content: These models may produce other types of inappropriate or offensive content, which may make it inappropriate to deploy for sensitive contexts without additional mitigations that are specific to the use case.
+ Information Reliability: Language models can generate nonsensical content or fabricate content that might sound reasonable but is inaccurate or outdated.
+ Limited Scope for Code: Majority of Phi-3 training data is based in Python and use common packages such as "typing, math, random, collections, datetime, itertools". If the model generates Python scripts that utilize other packages or scripts in other languages, we strongly recommend users manually verify all API uses.
Developers should apply responsible AI best practices and are responsible for ensuring that a specific use case complies with relevant laws and regulations (e.g. privacy, trade, etc.). Important areas for consideration include:
+ Allocation: Models may not be suitable for scenarios that could have consequential impact on legal status or the allocation of resources or life opportunities (ex: housing, employment, credit, etc.) without further assessments and additional debiasing techniques.
+ High-Risk Scenarios: Developers should assess suitability of using models in high-risk scenarios where unfair, unreliable or offensive outputs might be extremely costly or lead to harm. This includes providing advice in sensitive or expert domains where accuracy and reliability are critical (ex: legal or health advice). Additional safeguards should be implemented at the application level according to the deployment context.
+ Misinformation: Models may produce inaccurate information. Developers should follow transparency best practices and inform end-users they are interacting with an AI system. At the application level, developers can build feedback mechanisms and pipelines to ground responses in use-case specific, contextual information, a technique known as Retrieval Augmented Generation (RAG).
+ Generation of Harmful Content: Developers should assess outputs for their context and use available safety classifiers or custom solutions appropriate for their use case.
+ Misuse: Other forms of misuse such as fraud, spam, or malware production may be possible, and developers should ensure that their applications do not violate applicable laws and regulations.
+ Identification of individuals: models with vision capabilities may have the potential to uniquely identify individuals in images. Safety post-training steers the model to refuse such requests, but developers should consider and implement, as appropriate, additional mitigations or user consent flows as required in their respective jurisdiction, (e.g., building measures to blur faces in image inputs before processing.
## Training
### Model
* Architecture: Phi-3-Vision-128K-Instruct has 4.2B parameters and contains image encoder, connector, projector, and Phi-3 Mini language model.
* Inputs: Text and Image. It’s best suited for prompts using the chat format.
* Context length: 128K tokens
* GPUs: 512 H100-80G
* Training time: 1.5 days
* Training data: 500B vision and text tokens
* Outputs: Generated text in response to the input
* Dates: Our models were trained between February and April 2024
* Status: This is a static model trained on an offline text dataset with cutoff date Mar 15, 2024. Future versions of the tuned models may be released as we improve models.
* Release Type: Open weight release
* Release dates: The model weight is released on May 21, 2024.
### Datasets
Our training data includes a wide variety of sources, and is a combination of
1) publicly available documents filtered rigorously for quality, selected high-quality educational data and code;
2) selected high-quality image-text interleave;
3) newly created synthetic, “textbook-like” data for the purpose of teaching math, coding, common sense reasoning, general knowledge of the world (science, daily activities, theory of mind, etc.), newly created image data, e.g., chart/table/diagram/slides;
4) high quality chat format supervised data covering various topics to reflect human preferences on different aspects such as instruct-following, truthfulness, honesty and helpfulness.
The data collection process involved sourcing information from publicly available documents, with a meticulous approach to filtering out undesirable documents and images. To safeguard privacy, we carefully filtered various image and text data sources to remove or scrub any potentially personal data from the training data.
More details can be found in the [Phi-3 Technical Report](https://aka.ms/phi3-tech-report).
## Benchmarks
To understand the capabilities, we compare Phi-3-Vision-128K-Instruct with a set of models over a variety of zero-shot benchmarks using our internal benchmark platform.
|Benchmark|Phi-3 Vision-128K-In|LlaVA-1.6 Vicuna-7B|QWEN-VL Chat|Llama3-Llava-Next-8B|Claude-3 Haiku|Gemini 1.0 Pro V|GPT-4V-Turbo|
|---------|---------------------|------------------|------------|--------------------|--------------|----------------|------------|
|MMMU|40.4|34.2|39.0|36.4|40.7|42.0|55.5|
|MMBench|80.5|76.3|75.8|79.4|62.4|80.0|86.1|
|ScienceQA|90.8|70.6|67.2|73.7|72.0|79.7|75.7|
|MathVista|44.5|31.5|29.4|34.8|33.2|35.0|47.5|
|InterGPS|38.1|20.5|22.3|24.6|32.1|28.6|41.0|
|AI2D|76.7|63.1|59.8|66.9|60.3|62.8|74.7|
|ChartQA|81.4|55.0|50.9|65.8|59.3|58.0|62.3|
|TextVQA|70.9|64.6|59.4|55.7|62.7|64.7|68.1|
|POPE|85.8|87.2|82.6|87.0|74.4|84.2|83.7|
## Software
* [PyTorch](https://github.com/pytorch/pytorch)
* [Transformers](https://github.com/huggingface/transformers)
* [Flash-Attention](https://github.com/HazyResearch/flash-attention)
## Hardware
Note that by default, the Phi-3-Vision-128K model uses flash attention, which requires certain types of GPU hardware to run. We have tested on the following GPU types:
* NVIDIA A100
* NVIDIA A6000
* NVIDIA H100
## License
The model is licensed under the [MIT license](https://huggingface.co/microsoft/Phi-3-vision-128k-instruct/resolve/main/LICENSE).
## Trademarks
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow [Microsoft’s Trademark & Brand Guidelines](https://www.microsoft.com/en-us/legal/intellectualproperty/trademarks). Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party’s policies.
| null |
Non_BioNLP
|
## Model Summary
The Phi-3-Vision-128K-Instruct is a lightweight, state-of-the-art open multimodal model built upon datasets which include - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data both on text and vision. The model belongs to the Phi-3 model family, and the multimodal version comes with 128K context length (in tokens) it can support. The model underwent a rigorous enhancement process, incorporating both supervised fine-tuning and direct preference optimization to ensure precise instruction adherence and robust safety measures.
Resources and Technical Documentation:
+ [Phi-3 Microsoft Blog](https://aka.ms/Phi-3Build2024)
+ [Phi-3 Technical Report](https://aka.ms/phi3-tech-report)
+ [Phi-3 on Azure AI Studio](https://aka.ms/try-phi3vision)
+ [Phi-3 Cookbook](https://github.com/microsoft/Phi-3CookBook)
| | Short Context | Long Context |
| ------- | ------------- | ------------ |
| Mini | 4K [[HF]](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-onnx) ; [[GGUF]](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-gguf) | 128K [[HF]](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct-onnx)|
| Small | 8K [[HF]](https://huggingface.co/microsoft/Phi-3-small-8k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-small-8k-instruct-onnx-cuda) | 128K [[HF]](https://huggingface.co/microsoft/Phi-3-small-128k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-small-128k-instruct-onnx-cuda)|
| Medium | 4K [[HF]](https://huggingface.co/microsoft/Phi-3-medium-4k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-medium-4k-instruct-onnx-cuda) | 128K [[HF]](https://huggingface.co/microsoft/Phi-3-medium-128k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-medium-128k-instruct-onnx-cuda)|
| Vision | | 128K [[HF]](https://huggingface.co/microsoft/Phi-3-vision-128k-instruct) ; [[ONNX]](https://huggingface.co/microsoft/Phi-3-vision-128k-instruct-onnx-cuda)|
## Intended Uses
**Primary use cases**
The model is intended for broad commercial and research use in English. The model provides uses for general purpose AI systems and applications with visual and text input capabilities which require
1) memory/compute constrained environments;
2) latency bound scenarios;
3) general image understanding;
4) OCR;
5) chart and table understanding.
Our model is designed to accelerate research on efficient language and multimodal models, for use as a building block for generative AI powered features.
**Use case considerations**
Our models are not specifically designed or evaluated for all downstream purposes. Developers should consider common limitations of language models as they select use cases, and evaluate and mitigate for accuracy, safety, and fairness before using within a specific downstream use case, particularly for high-risk scenarios.
Developers should be aware of and adhere to applicable laws or regulations (including privacy, trade compliance laws, etc.) that are relevant to their use case.
Nothing contained in this Model Card should be interpreted as or deemed a restriction or modification to the license the model is released under.
## How to Use
Phi-3-Vision-128K-Instruct has been integrated in the development version (4.40.2) of `transformers`. Until the official version is released through `pip`, ensure that you are doing one of the following:
* When loading the model, ensure that `trust_remote_code=True` is passed as an argument of the `from_pretrained()` function.
* Update your local `transformers` to the development version: `pip uninstall -y transformers && pip install git+https://github.com/huggingface/transformers`. The previous command is an alternative to cloning and installing from the source.
The current `transformers` version can be verified with: `pip list | grep transformers`.
Examples of required packages:
```
flash_attn==2.5.8
numpy==1.24.4
Pillow==10.3.0
Requests==2.31.0
torch==2.3.0
torchvision==0.18.0
transformers==4.40.2
```
Phi-3-Vision-128K-Instruct is also available in [Azure AI Studio](https://aka.ms/phi3-azure-ai).
### Chat Format
Given the nature of the training data, the Phi-3-Vision-128K-Instruct model is best suited for a single image input wih prompts using the chat format as follows.
You can provide the prompt as a single image with a generic template as follow:
```markdown
<|user|>\n<|image_1|>\n{prompt}<|end|>\n<|assistant|>\n
```
where the model generates the text after `<|assistant|>` . In case of multi-turn conversation, the prompt can be formatted as follows:
```markdown
<|user|>\n<|image_1|>\n{prompt_1}<|end|>\n<|assistant|>\n{response_1}<|end|>\n<|user|>\n{prompt_2}<|end|>\n<|assistant|>\n
```
### Sample inference code
This code snippets show how to get quickly started with running the model on a GPU:
```python
from PIL import Image
import requests
from transformers import AutoModelForCausalLM
from transformers import AutoProcessor
model_id = "microsoft/Phi-3-vision-128k-instruct"
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="cuda", trust_remote_code=True, torch_dtype="auto", _attn_implementation='flash_attention_2') # use _attn_implementation='eager' to disable flash attention
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
messages = [
{"role": "user", "content": "<|image_1|>\nWhat is shown in this image?"},
{"role": "assistant", "content": "The chart displays the percentage of respondents who agree with various statements about their preparedness for meetings. It shows five categories: 'Having clear and pre-defined goals for meetings', 'Knowing where to find the information I need for a meeting', 'Understanding my exact role and responsibilities when I'm invited', 'Having tools to manage admin tasks like note-taking or summarization', and 'Having more focus time to sufficiently prepare for meetings'. Each category has an associated bar indicating the level of agreement, measured on a scale from 0% to 100%."},
{"role": "user", "content": "Provide insightful questions to spark discussion."}
]
url = "https://assets-c4akfrf5b4d3f4b7.z01.azurefd.net/assets/2024/04/BMDataViz_661fb89f3845e.png"
image = Image.open(requests.get(url, stream=True).raw)
prompt = processor.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = processor(prompt, [image], return_tensors="pt").to("cuda:0")
generation_args = {
"max_new_tokens": 500,
"temperature": 0.0,
"do_sample": False,
}
generate_ids = model.generate(**inputs, eos_token_id=processor.tokenizer.eos_token_id, **generation_args)
# remove input tokens
generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
response = processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
print(response)
```
Additional basic examples are provided [here](https://huggingface.co/microsoft/Phi-3-vision-128k-instruct/blob/main/sample_inference.py).
## Responsible AI Considerations
Like other models, the Phi family of models can potentially behave in ways that are unfair, unreliable, or offensive. Some of the limiting behaviors to be aware of include:
+ Quality of Service: The Phi models are trained primarily on English text. Languages other than English will experience worse performance. English language varieties with less representation in the training data might experience worse performance than standard American English.
+ Representation of Harms & Perpetuation of Stereotypes: These models can over- or under-represent groups of people, erase representation of some groups, or reinforce demeaning or negative stereotypes. Despite safety post-training, these limitations may still be present due to differing levels of representation of different groups or prevalence of examples of negative stereotypes in training data that reflect real-world patterns and societal biases.
+ Inappropriate or Offensive Content: These models may produce other types of inappropriate or offensive content, which may make it inappropriate to deploy for sensitive contexts without additional mitigations that are specific to the use case.
+ Information Reliability: Language models can generate nonsensical content or fabricate content that might sound reasonable but is inaccurate or outdated.
+ Limited Scope for Code: Majority of Phi-3 training data is based in Python and use common packages such as "typing, math, random, collections, datetime, itertools". If the model generates Python scripts that utilize other packages or scripts in other languages, we strongly recommend users manually verify all API uses.
Developers should apply responsible AI best practices and are responsible for ensuring that a specific use case complies with relevant laws and regulations (e.g. privacy, trade, etc.). Important areas for consideration include:
+ Allocation: Models may not be suitable for scenarios that could have consequential impact on legal status or the allocation of resources or life opportunities (ex: housing, employment, credit, etc.) without further assessments and additional debiasing techniques.
+ High-Risk Scenarios: Developers should assess suitability of using models in high-risk scenarios where unfair, unreliable or offensive outputs might be extremely costly or lead to harm. This includes providing advice in sensitive or expert domains where accuracy and reliability are critical (ex: legal or health advice). Additional safeguards should be implemented at the application level according to the deployment context.
+ Misinformation: Models may produce inaccurate information. Developers should follow transparency best practices and inform end-users they are interacting with an AI system. At the application level, developers can build feedback mechanisms and pipelines to ground responses in use-case specific, contextual information, a technique known as Retrieval Augmented Generation (RAG).
+ Generation of Harmful Content: Developers should assess outputs for their context and use available safety classifiers or custom solutions appropriate for their use case.
+ Misuse: Other forms of misuse such as fraud, spam, or malware production may be possible, and developers should ensure that their applications do not violate applicable laws and regulations.
+ Identification of individuals: models with vision capabilities may have the potential to uniquely identify individuals in images. Safety post-training steers the model to refuse such requests, but developers should consider and implement, as appropriate, additional mitigations or user consent flows as required in their respective jurisdiction, (e.g., building measures to blur faces in image inputs before processing.
## Training
### Model
* Architecture: Phi-3-Vision-128K-Instruct has 4.2B parameters and contains image encoder, connector, projector, and Phi-3 Mini language model.
* Inputs: Text and Image. It’s best suited for prompts using the chat format.
* Context length: 128K tokens
* GPUs: 512 H100-80G
* Training time: 1.5 days
* Training data: 500B vision and text tokens
* Outputs: Generated text in response to the input
* Dates: Our models were trained between February and April 2024
* Status: This is a static model trained on an offline text dataset with cutoff date Mar 15, 2024. Future versions of the tuned models may be released as we improve models.
* Release Type: Open weight release
* Release dates: The model weight is released on May 21, 2024.
### Datasets
Our training data includes a wide variety of sources, and is a combination of
1) publicly available documents filtered rigorously for quality, selected high-quality educational data and code;
2) selected high-quality image-text interleave;
3) newly created synthetic, “textbook-like” data for the purpose of teaching math, coding, common sense reasoning, general knowledge of the world (science, daily activities, theory of mind, etc.), newly created image data, e.g., chart/table/diagram/slides;
4) high quality chat format supervised data covering various topics to reflect human preferences on different aspects such as instruct-following, truthfulness, honesty and helpfulness.
The data collection process involved sourcing information from publicly available documents, with a meticulous approach to filtering out undesirable documents and images. To safeguard privacy, we carefully filtered various image and text data sources to remove or scrub any potentially personal data from the training data.
More details can be found in the [Phi-3 Technical Report](https://aka.ms/phi3-tech-report).
## Benchmarks
To understand the capabilities, we compare Phi-3-Vision-128K-Instruct with a set of models over a variety of zero-shot benchmarks using our internal benchmark platform.
|Benchmark|Phi-3 Vision-128K-In|LlaVA-1.6 Vicuna-7B|QWEN-VL Chat|Llama3-Llava-Next-8B|Claude-3 Haiku|Gemini 1.0 Pro V|GPT-4V-Turbo|
|---------|---------------------|------------------|------------|--------------------|--------------|----------------|------------|
|MMMU|40.4|34.2|39.0|36.4|40.7|42.0|55.5|
|MMBench|80.5|76.3|75.8|79.4|62.4|80.0|86.1|
|ScienceQA|90.8|70.6|67.2|73.7|72.0|79.7|75.7|
|MathVista|44.5|31.5|29.4|34.8|33.2|35.0|47.5|
|InterGPS|38.1|20.5|22.3|24.6|32.1|28.6|41.0|
|AI2D|76.7|63.1|59.8|66.9|60.3|62.8|74.7|
|ChartQA|81.4|55.0|50.9|65.8|59.3|58.0|62.3|
|TextVQA|70.9|64.6|59.4|55.7|62.7|64.7|68.1|
|POPE|85.8|87.2|82.6|87.0|74.4|84.2|83.7|
## Software
* [PyTorch](https://github.com/pytorch/pytorch)
* [Transformers](https://github.com/huggingface/transformers)
* [Flash-Attention](https://github.com/HazyResearch/flash-attention)
## Hardware
Note that by default, the Phi-3-Vision-128K model uses flash attention, which requires certain types of GPU hardware to run. We have tested on the following GPU types:
* NVIDIA A100
* NVIDIA A6000
* NVIDIA H100
## License
The model is licensed under the [MIT license](https://huggingface.co/microsoft/Phi-3-vision-128k-instruct/resolve/main/LICENSE).
## Trademarks
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow [Microsoft’s Trademark & Brand Guidelines](https://www.microsoft.com/en-us/legal/intellectualproperty/trademarks). Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party’s policies.
|
{"language": ["multilingual"], "license": "mit", "license_link": "https://huggingface.co/microsoft/Phi-3-vision-128k-instruct/resolve/main/LICENSE", "pipeline_tag": "text-generation", "tags": ["nlp", "code", "vision"], "inference": {"parameters": {"temperature": 0.7}}, "widget": [{"messages": [{"role": "user", "content": "<|image_1|>Can you describe what you see in the image?"}]}]}
|
task
|
[
"SUMMARIZATION"
] | 45,221 |
RichardErkhov/cstr_-_Spaetzle-v8-7b-4bits
|
RichardErkhov
| null |
[
"safetensors",
"mistral",
"4-bit",
"bitsandbytes",
"region:us"
] | 2025-02-09T21:56:49Z |
2025-02-09T21:59:09+00:00
| 4 | 0 |
---
{}
---
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
Spaetzle-v8-7b - bnb 4bits
- Model creator: https://huggingface.co/cstr/
- Original model: https://huggingface.co/cstr/Spaetzle-v8-7b/
Original model description:
---
tags:
- merge
- mergekit
- lazymergekit
- flemmingmiguel/NeuDist-Ro-7B
- johannhartmann/Brezn3
- ResplendentAI/Flora_DPO_7B
base_model:
- flemmingmiguel/NeuDist-Ro-7B
- johannhartmann/Brezn3
- ResplendentAI/Flora_DPO_7B
language:
- de
- en
---
# Spaetzle-v8-7b
This model is supposed to show adequate performance in German and English on a number of tasks, while mostly behaving well, that is, without rambling on, intermixing tokens from different templates in training and adapting, etc.
It is mostly a quick test, and considerably weaker in German grammar and orthography than DiscoLM e.g., but for use cases where this is not too important, but e.g. instruction following, reasoning, etc, it might actually be a little bit preferable.
It is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [flemmingmiguel/NeuDist-Ro-7B](https://huggingface.co/flemmingmiguel/NeuDist-Ro-7B)
* [johannhartmann/Brezn3](https://huggingface.co/johannhartmann/Brezn3)
* [ResplendentAI/Flora_DPO_7B](https://huggingface.co/ResplendentAI/Flora_DPO_7B)
* on the basis of [mayflowergmbh/Wiedervereinigung-7b-dpo-laser](https://huggingface.co/mayflowergmbh/Wiedervereinigung-7b-dpo-laser)
All credits are due to the creators of those original models and the training datasets involved.
For a suitable quantized version, try [cstr/Spaetzle-v8-7b-GGUF](https://huggingface.co/cstr/Spaetzle-v8-7b-GGUF)
## Evaluation
[Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_cstr__Spaetzle-v8-7b)
| Metric |Value|
|---------------------------------|----:|
|Avg. |72.27|
|AI2 Reasoning Challenge (25-Shot)|68.69|
|HellaSwag (10-Shot) |86.68|
|MMLU (5-Shot) |64.60|
|TruthfulQA (0-shot) |64.05|
|Winogrande (5-shot) |81.45|
|GSM8k (5-shot) |68.16|
EQ-Bench (v2_de): 61.04 / english (v2): 78.3
[ScandEval](https://scandeval.com/german-nlg/) 12.5.2 scores
| Benchmark | Spaetzle-v8-7b Value |
|-----------------------|----------------------------------------------------|
| Model ID | cstr/Spaetzle-v8-7b (few-shot, val) |
| Parameters | 7242 |
| Vocabulary Size | 32 |
| Context | 32768 |
| Commercial | False |
| Speed | 5,980 ± 1,031 / 1,714 ± 552 |
| Rank | 1.85 |
| GermEval | 58.90 ± 2.30 / 45.55 ± 3.30 |
| SB10k | 61.34 ± 1.90 / 72.98 ± 1.30 |
| ScaLA-De | 31.58 ± 4.39 / 65.51 ± 2.23 |
| GermanQuAD | 24.91 ± 3.98 / 60.88 ± 3.31 |
| MLSum | 67.25 ± 1.06 / 22.95 ± 2.64 |
| MMLU-De | 34.62 ± 2.20 / 50.43 ± 1.52 |
| HellaSwag-De | 48.70 ± 2.47 / 61.05 ± 1.79 |
| Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|------------------------------------------------------------|------:|------:|---------:|-------:|------:|
|[Spaetzle-v8-7b](https://huggingface.co/cstr/Spaetzle-v8-7b)| 45.31| 75.69| 63.94| 45.57| 57.63|
### AGIEval
| Task |Version| Metric |Value| |Stderr|
|------------------------------|------:|--------|----:|---|-----:|
|agieval_aqua_rat | 0|acc |25.59|± | 2.74|
| | |acc_norm|24.80|± | 2.72|
|agieval_logiqa_en | 0|acc |39.63|± | 1.92|
| | |acc_norm|39.78|± | 1.92|
|agieval_lsat_ar | 0|acc |23.48|± | 2.80|
| | |acc_norm|24.35|± | 2.84|
|agieval_lsat_lr | 0|acc |50.98|± | 2.22|
| | |acc_norm|51.96|± | 2.21|
|agieval_lsat_rc | 0|acc |62.08|± | 2.96|
| | |acc_norm|62.83|± | 2.95|
|agieval_sat_en | 0|acc |78.64|± | 2.86|
| | |acc_norm|79.13|± | 2.84|
|agieval_sat_en_without_passage| 0|acc |44.66|± | 3.47|
| | |acc_norm|44.66|± | 3.47|
|agieval_sat_math | 0|acc |37.27|± | 3.27|
| | |acc_norm|35.00|± | 3.22|
Average: 45.31%
### GPT4All
| Task |Version| Metric |Value| |Stderr|
|-------------|------:|--------|----:|---|-----:|
|arc_challenge| 0|acc |63.14|± | 1.41|
| | |acc_norm|64.51|± | 1.40|
|arc_easy | 0|acc |85.98|± | 0.71|
| | |acc_norm|82.49|± | 0.78|
|boolq | 1|acc |88.10|± | 0.57|
|hellaswag | 0|acc |66.31|± | 0.47|
| | |acc_norm|85.17|± | 0.35|
|openbookqa | 0|acc |38.00|± | 2.17|
| | |acc_norm|47.20|± | 2.23|
|piqa | 0|acc |83.35|± | 0.87|
| | |acc_norm|84.17|± | 0.85|
|winogrande | 0|acc |78.22|± | 1.16|
Average: 75.69%
### TruthfulQA
| Task |Version|Metric|Value| |Stderr|
|-------------|------:|------|----:|---|-----:|
|truthfulqa_mc| 1|mc1 |47.74|± | 1.75|
| | |mc2 |63.94|± | 1.53|
Average: 63.94%
### Bigbench
| Task |Version| Metric |Value| |Stderr|
|------------------------------------------------|------:|---------------------|----:|---|-----:|
|bigbench_causal_judgement | 0|multiple_choice_grade|56.84|± | 3.60|
|bigbench_date_understanding | 0|multiple_choice_grade|66.12|± | 2.47|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|41.47|± | 3.07|
|bigbench_geometric_shapes | 0|multiple_choice_grade|22.01|± | 2.19|
| | |exact_str_match | 0.00|± | 0.00|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|31.40|± | 2.08|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|23.14|± | 1.60|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|56.00|± | 2.87|
|bigbench_movie_recommendation | 0|multiple_choice_grade|45.00|± | 2.23|
|bigbench_navigate | 0|multiple_choice_grade|50.70|± | 1.58|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|70.05|± | 1.02|
|bigbench_ruin_names | 0|multiple_choice_grade|45.54|± | 2.36|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|26.05|± | 1.39|
|bigbench_snarks | 0|multiple_choice_grade|71.82|± | 3.35|
|bigbench_sports_understanding | 0|multiple_choice_grade|72.92|± | 1.42|
|bigbench_temporal_sequences | 0|multiple_choice_grade|44.20|± | 1.57|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|22.80|± | 1.19|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|18.23|± | 0.92|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|56.00|± | 2.87|
Average: 45.57%
Average score: 57.63%
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "cstr/Spaetzle-v8-7b"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
```
## 🧩 Configuration
The model uses ChatML and should work well with this (as it is merged from models which (mostly) saw ChatML templates in training).
```yaml
models:
- model: mayflowergmbh/Wiedervereinigung-7b-dpo-laser
# no parameters necessary for base model
- model: flemmingmiguel/NeuDist-Ro-7B
parameters:
density: 0.60
weight: 0.30
- model: johannhartmann/Brezn3
parameters:
density: 0.65
weight: 0.40
- model: ResplendentAI/Flora_DPO_7B
parameters:
density: 0.6
weight: 0.3
merge_method: dare_ties
base_model: mayflowergmbh/Wiedervereinigung-7b-dpo-laser
parameters:
int8_mask: true
dtype: bfloat16
random_seed: 0
tokenizer_source: base
```
| null |
Non_BioNLP
|
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
Spaetzle-v8-7b - bnb 4bits
- Model creator: https://huggingface.co/cstr/
- Original model: https://huggingface.co/cstr/Spaetzle-v8-7b/
Original model description:
---
tags:
- merge
- mergekit
- lazymergekit
- flemmingmiguel/NeuDist-Ro-7B
- johannhartmann/Brezn3
- ResplendentAI/Flora_DPO_7B
base_model:
- flemmingmiguel/NeuDist-Ro-7B
- johannhartmann/Brezn3
- ResplendentAI/Flora_DPO_7B
language:
- de
- en
---
# Spaetzle-v8-7b
This model is supposed to show adequate performance in German and English on a number of tasks, while mostly behaving well, that is, without rambling on, intermixing tokens from different templates in training and adapting, etc.
It is mostly a quick test, and considerably weaker in German grammar and orthography than DiscoLM e.g., but for use cases where this is not too important, but e.g. instruction following, reasoning, etc, it might actually be a little bit preferable.
It is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [flemmingmiguel/NeuDist-Ro-7B](https://huggingface.co/flemmingmiguel/NeuDist-Ro-7B)
* [johannhartmann/Brezn3](https://huggingface.co/johannhartmann/Brezn3)
* [ResplendentAI/Flora_DPO_7B](https://huggingface.co/ResplendentAI/Flora_DPO_7B)
* on the basis of [mayflowergmbh/Wiedervereinigung-7b-dpo-laser](https://huggingface.co/mayflowergmbh/Wiedervereinigung-7b-dpo-laser)
All credits are due to the creators of those original models and the training datasets involved.
For a suitable quantized version, try [cstr/Spaetzle-v8-7b-GGUF](https://huggingface.co/cstr/Spaetzle-v8-7b-GGUF)
## Evaluation
[Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_cstr__Spaetzle-v8-7b)
| Metric |Value|
|---------------------------------|----:|
|Avg. |72.27|
|AI2 Reasoning Challenge (25-Shot)|68.69|
|HellaSwag (10-Shot) |86.68|
|MMLU (5-Shot) |64.60|
|TruthfulQA (0-shot) |64.05|
|Winogrande (5-shot) |81.45|
|GSM8k (5-shot) |68.16|
EQ-Bench (v2_de): 61.04 / english (v2): 78.3
[ScandEval](https://scandeval.com/german-nlg/) 12.5.2 scores
| Benchmark | Spaetzle-v8-7b Value |
|-----------------------|----------------------------------------------------|
| Model ID | cstr/Spaetzle-v8-7b (few-shot, val) |
| Parameters | 7242 |
| Vocabulary Size | 32 |
| Context | 32768 |
| Commercial | False |
| Speed | 5,980 ± 1,031 / 1,714 ± 552 |
| Rank | 1.85 |
| GermEval | 58.90 ± 2.30 / 45.55 ± 3.30 |
| SB10k | 61.34 ± 1.90 / 72.98 ± 1.30 |
| ScaLA-De | 31.58 ± 4.39 / 65.51 ± 2.23 |
| GermanQuAD | 24.91 ± 3.98 / 60.88 ± 3.31 |
| MLSum | 67.25 ± 1.06 / 22.95 ± 2.64 |
| MMLU-De | 34.62 ± 2.20 / 50.43 ± 1.52 |
| HellaSwag-De | 48.70 ± 2.47 / 61.05 ± 1.79 |
| Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|------------------------------------------------------------|------:|------:|---------:|-------:|------:|
|[Spaetzle-v8-7b](https://huggingface.co/cstr/Spaetzle-v8-7b)| 45.31| 75.69| 63.94| 45.57| 57.63|
### AGIEval
| Task |Version| Metric |Value| |Stderr|
|------------------------------|------:|--------|----:|---|-----:|
|agieval_aqua_rat | 0|acc |25.59|± | 2.74|
| | |acc_norm|24.80|± | 2.72|
|agieval_logiqa_en | 0|acc |39.63|± | 1.92|
| | |acc_norm|39.78|± | 1.92|
|agieval_lsat_ar | 0|acc |23.48|± | 2.80|
| | |acc_norm|24.35|± | 2.84|
|agieval_lsat_lr | 0|acc |50.98|± | 2.22|
| | |acc_norm|51.96|± | 2.21|
|agieval_lsat_rc | 0|acc |62.08|± | 2.96|
| | |acc_norm|62.83|± | 2.95|
|agieval_sat_en | 0|acc |78.64|± | 2.86|
| | |acc_norm|79.13|± | 2.84|
|agieval_sat_en_without_passage| 0|acc |44.66|± | 3.47|
| | |acc_norm|44.66|± | 3.47|
|agieval_sat_math | 0|acc |37.27|± | 3.27|
| | |acc_norm|35.00|± | 3.22|
Average: 45.31%
### GPT4All
| Task |Version| Metric |Value| |Stderr|
|-------------|------:|--------|----:|---|-----:|
|arc_challenge| 0|acc |63.14|± | 1.41|
| | |acc_norm|64.51|± | 1.40|
|arc_easy | 0|acc |85.98|± | 0.71|
| | |acc_norm|82.49|± | 0.78|
|boolq | 1|acc |88.10|± | 0.57|
|hellaswag | 0|acc |66.31|± | 0.47|
| | |acc_norm|85.17|± | 0.35|
|openbookqa | 0|acc |38.00|± | 2.17|
| | |acc_norm|47.20|± | 2.23|
|piqa | 0|acc |83.35|± | 0.87|
| | |acc_norm|84.17|± | 0.85|
|winogrande | 0|acc |78.22|± | 1.16|
Average: 75.69%
### TruthfulQA
| Task |Version|Metric|Value| |Stderr|
|-------------|------:|------|----:|---|-----:|
|truthfulqa_mc| 1|mc1 |47.74|± | 1.75|
| | |mc2 |63.94|± | 1.53|
Average: 63.94%
### Bigbench
| Task |Version| Metric |Value| |Stderr|
|------------------------------------------------|------:|---------------------|----:|---|-----:|
|bigbench_causal_judgement | 0|multiple_choice_grade|56.84|± | 3.60|
|bigbench_date_understanding | 0|multiple_choice_grade|66.12|± | 2.47|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|41.47|± | 3.07|
|bigbench_geometric_shapes | 0|multiple_choice_grade|22.01|± | 2.19|
| | |exact_str_match | 0.00|± | 0.00|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|31.40|± | 2.08|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|23.14|± | 1.60|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|56.00|± | 2.87|
|bigbench_movie_recommendation | 0|multiple_choice_grade|45.00|± | 2.23|
|bigbench_navigate | 0|multiple_choice_grade|50.70|± | 1.58|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|70.05|± | 1.02|
|bigbench_ruin_names | 0|multiple_choice_grade|45.54|± | 2.36|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|26.05|± | 1.39|
|bigbench_snarks | 0|multiple_choice_grade|71.82|± | 3.35|
|bigbench_sports_understanding | 0|multiple_choice_grade|72.92|± | 1.42|
|bigbench_temporal_sequences | 0|multiple_choice_grade|44.20|± | 1.57|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|22.80|± | 1.19|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|18.23|± | 0.92|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|56.00|± | 2.87|
Average: 45.57%
Average score: 57.63%
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "cstr/Spaetzle-v8-7b"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
```
## 🧩 Configuration
The model uses ChatML and should work well with this (as it is merged from models which (mostly) saw ChatML templates in training).
```yaml
models:
- model: mayflowergmbh/Wiedervereinigung-7b-dpo-laser
# no parameters necessary for base model
- model: flemmingmiguel/NeuDist-Ro-7B
parameters:
density: 0.60
weight: 0.30
- model: johannhartmann/Brezn3
parameters:
density: 0.65
weight: 0.40
- model: ResplendentAI/Flora_DPO_7B
parameters:
density: 0.6
weight: 0.3
merge_method: dare_ties
base_model: mayflowergmbh/Wiedervereinigung-7b-dpo-laser
parameters:
int8_mask: true
dtype: bfloat16
random_seed: 0
tokenizer_source: base
```
|
{}
|
task
|
[
"TRANSLATION"
] | 45,222 |
Zelyanoth/my_fon_translation_model
|
Zelyanoth
|
text2text-generation
|
[
"transformers",
"tensorboard",
"safetensors",
"m2m_100",
"text2text-generation",
"generated_from_trainer",
"dataset:generator",
"base_model:Zelyanoth/my_fon_translation_model",
"base_model:finetune:Zelyanoth/my_fon_translation_model",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-05-21T18:50:02Z |
2024-06-08T11:12:34+00:00
| 17 | 0 |
---
base_model: Zelyanoth/my_fon_translation_model
datasets:
- generator
metrics:
- bleu
tags:
- generated_from_trainer
model-index:
- name: my_fon_translation_model
results:
- task:
type: text2text-generation
name: Sequence-to-sequence Language Modeling
dataset:
name: generator
type: generator
config: default
split: train
args: default
metrics:
- type: bleu
value: 8.059
name: Bleu
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# my_fon_translation_model
This model is a fine-tuned version of [Zelyanoth/my_fon_translation_model](https://huggingface.co/Zelyanoth/my_fon_translation_model) on the generator dataset.
It achieves the following results on the evaluation set:
- Loss: 6.1298
- Bleu: 8.059
- Gen Len: 8.4704
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 4.6e-08
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
### Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|:-------------:|:-----:|:------:|:---------------:|:------:|:-------:|
| 0.5745 | 1.0 | 13165 | 6.1275 | 7.8417 | 8.5553 |
| 0.564 | 2.0 | 26330 | 6.1295 | 7.8534 | 8.5544 |
| 0.5611 | 3.0 | 39495 | 6.1291 | 7.8887 | 8.5456 |
| 0.5538 | 4.0 | 52660 | 6.1288 | 7.7891 | 8.5627 |
| 0.566 | 5.0 | 65825 | 6.1296 | 8.0473 | 8.4714 |
| 0.5745 | 6.0 | 78990 | 6.1297 | 8.059 | 8.4695 |
| 0.5638 | 7.0 | 92155 | 6.1298 | 8.0595 | 8.47 |
| 0.5666 | 8.0 | 105320 | 6.1298 | 8.059 | 8.4704 |
### Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# my_fon_translation_model
This model is a fine-tuned version of [Zelyanoth/my_fon_translation_model](https://huggingface.co/Zelyanoth/my_fon_translation_model) on the generator dataset.
It achieves the following results on the evaluation set:
- Loss: 6.1298
- Bleu: 8.059
- Gen Len: 8.4704
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 4.6e-08
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
### Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|:-------------:|:-----:|:------:|:---------------:|:------:|:-------:|
| 0.5745 | 1.0 | 13165 | 6.1275 | 7.8417 | 8.5553 |
| 0.564 | 2.0 | 26330 | 6.1295 | 7.8534 | 8.5544 |
| 0.5611 | 3.0 | 39495 | 6.1291 | 7.8887 | 8.5456 |
| 0.5538 | 4.0 | 52660 | 6.1288 | 7.7891 | 8.5627 |
| 0.566 | 5.0 | 65825 | 6.1296 | 8.0473 | 8.4714 |
| 0.5745 | 6.0 | 78990 | 6.1297 | 8.059 | 8.4695 |
| 0.5638 | 7.0 | 92155 | 6.1298 | 8.0595 | 8.47 |
| 0.5666 | 8.0 | 105320 | 6.1298 | 8.059 | 8.4704 |
### Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
|
{"base_model": "Zelyanoth/my_fon_translation_model", "datasets": ["generator"], "metrics": ["bleu"], "tags": ["generated_from_trainer"], "model-index": [{"name": "my_fon_translation_model", "results": [{"task": {"type": "text2text-generation", "name": "Sequence-to-sequence Language Modeling"}, "dataset": {"name": "generator", "type": "generator", "config": "default", "split": "train", "args": "default"}, "metrics": [{"type": "bleu", "value": 8.059, "name": "Bleu"}]}]}]}
|
task
|
[
"TRANSLATION"
] | 45,223 |
Phelip/MegaMe
|
Phelip
|
summarization
|
[
"adapter-transformers",
"not-for-all-audiences",
"music",
"summarization",
"fr",
"dataset:nvidia/HelpSteer2",
"base_model:openai/whisper-large-v3-turbo",
"base_model:adapter:openai/whisper-large-v3-turbo",
"license:openrail",
"region:us"
] | 2024-11-03T13:48:41Z |
2024-11-03T13:55:31+00:00
| 0 | 0 |
---
base_model:
- openai/whisper-large-v3-turbo
datasets:
- nvidia/HelpSteer2
language:
- fr
library_name: adapter-transformers
license: openrail
metrics:
- accuracy
- bertscore
pipeline_tag: summarization
tags:
- not-for-all-audiences
- music
new_version: openai/whisper-large-v3-turbo
---
| null |
Non_BioNLP
|
{"base_model": ["openai/whisper-large-v3-turbo"], "datasets": ["nvidia/HelpSteer2"], "language": ["fr"], "library_name": "adapter-transformers", "license": "openrail", "metrics": ["accuracy", "bertscore"], "pipeline_tag": "summarization", "tags": ["not-for-all-audiences", "music"], "new_version": "openai/whisper-large-v3-turbo"}
|
task
|
[
"SUMMARIZATION"
] | 45,224 |
|
JustFrederik/nllb-200-distilled-1.3B-ct2-int8
|
JustFrederik
|
translation
|
[
"transformers",
"nllb",
"translation",
"ace",
"acm",
"acq",
"aeb",
"af",
"ajp",
"ak",
"als",
"am",
"apc",
"ar",
"ars",
"ary",
"arz",
"as",
"ast",
"awa",
"ayr",
"azb",
"azj",
"ba",
"bm",
"ban",
"be",
"bem",
"bn",
"bho",
"bjn",
"bo",
"bs",
"bug",
"bg",
"ca",
"ceb",
"cs",
"cjk",
"ckb",
"crh",
"cy",
"da",
"de",
"dik",
"dyu",
"dz",
"el",
"en",
"eo",
"et",
"eu",
"ee",
"fo",
"fj",
"fi",
"fon",
"fr",
"fur",
"fuv",
"gaz",
"gd",
"ga",
"gl",
"gn",
"gu",
"ht",
"ha",
"he",
"hi",
"hne",
"hr",
"hu",
"hy",
"ig",
"ilo",
"id",
"is",
"it",
"jv",
"ja",
"kab",
"kac",
"kam",
"kn",
"ks",
"ka",
"kk",
"kbp",
"kea",
"khk",
"km",
"ki",
"rw",
"ky",
"kmb",
"kmr",
"knc",
"kg",
"ko",
"lo",
"lij",
"li",
"ln",
"lt",
"lmo",
"ltg",
"lb",
"lua",
"lg",
"luo",
"lus",
"lvs",
"mag",
"mai",
"ml",
"mar",
"min",
"mk",
"mt",
"mni",
"mos",
"mi",
"my",
"nl",
"nn",
"nb",
"npi",
"nso",
"nus",
"ny",
"oc",
"ory",
"pag",
"pa",
"pap",
"pbt",
"pes",
"plt",
"pl",
"pt",
"prs",
"quy",
"ro",
"rn",
"ru",
"sg",
"sa",
"sat",
"scn",
"shn",
"si",
"sk",
"sl",
"sm",
"sn",
"sd",
"so",
"st",
"es",
"sc",
"sr",
"ss",
"su",
"sv",
"swh",
"szl",
"ta",
"taq",
"tt",
"te",
"tg",
"tl",
"th",
"ti",
"tpi",
"tn",
"ts",
"tk",
"tum",
"tr",
"tw",
"tzm",
"ug",
"uk",
"umb",
"ur",
"uzn",
"vec",
"vi",
"war",
"wo",
"xh",
"ydd",
"yo",
"yue",
"zh",
"zsm",
"zu",
"dataset:flores-200",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"region:us"
] | 2023-05-14T12:17:16Z |
2023-05-14T21:54:10+00:00
| 97 | 2 |
---
datasets:
- flores-200
language:
- ace
- acm
- acq
- aeb
- af
- ajp
- ak
- als
- am
- apc
- ar
- ars
- ary
- arz
- as
- ast
- awa
- ayr
- azb
- azj
- ba
- bm
- ban
- be
- bem
- bn
- bho
- bjn
- bo
- bs
- bug
- bg
- ca
- ceb
- cs
- cjk
- ckb
- crh
- cy
- da
- de
- dik
- dyu
- dz
- el
- en
- eo
- et
- eu
- ee
- fo
- fj
- fi
- fon
- fr
- fur
- fuv
- gaz
- gd
- ga
- gl
- gn
- gu
- ht
- ha
- he
- hi
- hne
- hr
- hu
- hy
- ig
- ilo
- id
- is
- it
- jv
- ja
- kab
- kac
- kam
- kn
- ks
- ka
- kk
- kbp
- kea
- khk
- km
- ki
- rw
- ky
- kmb
- kmr
- knc
- kg
- ko
- lo
- lij
- li
- ln
- lt
- lmo
- ltg
- lb
- lua
- lg
- luo
- lus
- lvs
- mag
- mai
- ml
- mar
- min
- mk
- mt
- mni
- mos
- mi
- my
- nl
- nn
- nb
- npi
- nso
- nus
- ny
- oc
- ory
- pag
- pa
- pap
- pbt
- pes
- plt
- pl
- pt
- prs
- quy
- ro
- rn
- ru
- sg
- sa
- sat
- scn
- shn
- si
- sk
- sl
- sm
- sn
- sd
- so
- st
- es
- sc
- sr
- ss
- su
- sv
- swh
- szl
- ta
- taq
- tt
- te
- tg
- tl
- th
- ti
- tpi
- tn
- ts
- tk
- tum
- tr
- tw
- tzm
- ug
- uk
- umb
- ur
- uzn
- vec
- vi
- war
- wo
- xh
- ydd
- yo
- yue
- zh
- zsm
- zu
license: cc-by-nc-4.0
metrics:
- bleu
- spbleu
- chrf++
tags:
- nllb
- translation
language_details: ace_Arab, ace_Latn, acm_Arab, acq_Arab, aeb_Arab, afr_Latn, ajp_Arab,
aka_Latn, amh_Ethi, apc_Arab, arb_Arab, ars_Arab, ary_Arab, arz_Arab, asm_Beng,
ast_Latn, awa_Deva, ayr_Latn, azb_Arab, azj_Latn, bak_Cyrl, bam_Latn, ban_Latn,bel_Cyrl,
bem_Latn, ben_Beng, bho_Deva, bjn_Arab, bjn_Latn, bod_Tibt, bos_Latn, bug_Latn,
bul_Cyrl, cat_Latn, ceb_Latn, ces_Latn, cjk_Latn, ckb_Arab, crh_Latn, cym_Latn,
dan_Latn, deu_Latn, dik_Latn, dyu_Latn, dzo_Tibt, ell_Grek, eng_Latn, epo_Latn,
est_Latn, eus_Latn, ewe_Latn, fao_Latn, pes_Arab, fij_Latn, fin_Latn, fon_Latn,
fra_Latn, fur_Latn, fuv_Latn, gla_Latn, gle_Latn, glg_Latn, grn_Latn, guj_Gujr,
hat_Latn, hau_Latn, heb_Hebr, hin_Deva, hne_Deva, hrv_Latn, hun_Latn, hye_Armn,
ibo_Latn, ilo_Latn, ind_Latn, isl_Latn, ita_Latn, jav_Latn, jpn_Jpan, kab_Latn,
kac_Latn, kam_Latn, kan_Knda, kas_Arab, kas_Deva, kat_Geor, knc_Arab, knc_Latn,
kaz_Cyrl, kbp_Latn, kea_Latn, khm_Khmr, kik_Latn, kin_Latn, kir_Cyrl, kmb_Latn,
kon_Latn, kor_Hang, kmr_Latn, lao_Laoo, lvs_Latn, lij_Latn, lim_Latn, lin_Latn,
lit_Latn, lmo_Latn, ltg_Latn, ltz_Latn, lua_Latn, lug_Latn, luo_Latn, lus_Latn,
mag_Deva, mai_Deva, mal_Mlym, mar_Deva, min_Latn, mkd_Cyrl, plt_Latn, mlt_Latn,
mni_Beng, khk_Cyrl, mos_Latn, mri_Latn, zsm_Latn, mya_Mymr, nld_Latn, nno_Latn,
nob_Latn, npi_Deva, nso_Latn, nus_Latn, nya_Latn, oci_Latn, gaz_Latn, ory_Orya,
pag_Latn, pan_Guru, pap_Latn, pol_Latn, por_Latn, prs_Arab, pbt_Arab, quy_Latn,
ron_Latn, run_Latn, rus_Cyrl, sag_Latn, san_Deva, sat_Beng, scn_Latn, shn_Mymr,
sin_Sinh, slk_Latn, slv_Latn, smo_Latn, sna_Latn, snd_Arab, som_Latn, sot_Latn,
spa_Latn, als_Latn, srd_Latn, srp_Cyrl, ssw_Latn, sun_Latn, swe_Latn, swh_Latn,
szl_Latn, tam_Taml, tat_Cyrl, tel_Telu, tgk_Cyrl, tgl_Latn, tha_Thai, tir_Ethi,
taq_Latn, taq_Tfng, tpi_Latn, tsn_Latn, tso_Latn, tuk_Latn, tum_Latn, tur_Latn,
twi_Latn, tzm_Tfng, uig_Arab, ukr_Cyrl, umb_Latn, urd_Arab, uzn_Latn, vec_Latn,
vie_Latn, war_Latn, wol_Latn, xho_Latn, ydd_Hebr, yor_Latn, yue_Hant, zho_Hans,
zho_Hant, zul_Latn
---
https://huggingface.co/facebook/nllb-200-distilled-1.3B
```
ct2-transformers-converter --model facebook/nllb-200-distilled-1.3B --quantization int8 --output_dir converted/nllb-200-distilled-1.3B-ct2-int8
```
| null |
Non_BioNLP
| ERROR: type should be string, got "\nhttps://huggingface.co/facebook/nllb-200-distilled-1.3B\n```\nct2-transformers-converter --model facebook/nllb-200-distilled-1.3B --quantization int8 --output_dir converted/nllb-200-distilled-1.3B-ct2-int8\n```" |
{"datasets": ["flores-200"], "language": ["ace", "acm", "acq", "aeb", "af", "ajp", "ak", "als", "am", "apc", "ar", "ars", "ary", "arz", "as", "ast", "awa", "ayr", "azb", "azj", "ba", "bm", "ban", "be", "bem", "bn", "bho", "bjn", "bo", "bs", "bug", "bg", "ca", "ceb", "cs", "cjk", "ckb", "crh", "cy", "da", "de", "dik", "dyu", "dz", "el", "en", "eo", "et", "eu", "ee", "fo", "fj", "fi", "fon", "fr", "fur", "fuv", "gaz", "gd", "ga", "gl", "gn", "gu", "ht", "ha", "he", "hi", "hne", "hr", "hu", "hy", "ig", "ilo", "id", "is", "it", "jv", "ja", "kab", "kac", "kam", "kn", "ks", "ka", "kk", "kbp", "kea", "khk", "km", "ki", "rw", "ky", "kmb", "kmr", "knc", "kg", "ko", "lo", "lij", "li", "ln", "lt", "lmo", "ltg", "lb", "lua", "lg", "luo", "lus", "lvs", "mag", "mai", "ml", "mar", "min", "mk", "mt", "mni", "mos", "mi", "my", "nl", "nn", "nb", "npi", "nso", "nus", "ny", "oc", "ory", "pag", "pa", "pap", "pbt", "pes", "plt", "pl", "pt", "prs", "quy", "ro", "rn", "ru", "sg", "sa", "sat", "scn", "shn", "si", "sk", "sl", "sm", "sn", "sd", "so", "st", "es", "sc", "sr", "ss", "su", "sv", "swh", "szl", "ta", "taq", "tt", "te", "tg", "tl", "th", "ti", "tpi", "tn", "ts", "tk", "tum", "tr", "tw", "tzm", "ug", "uk", "umb", "ur", "uzn", "vec", "vi", "war", "wo", "xh", "ydd", "yo", "yue", "zh", "zsm", "zu"], "license": "cc-by-nc-4.0", "metrics": ["bleu", "spbleu", "chrf++"], "tags": ["nllb", "translation"], "language_details": "ace_Arab, ace_Latn, acm_Arab, acq_Arab, aeb_Arab, afr_Latn, ajp_Arab, aka_Latn, amh_Ethi, apc_Arab, arb_Arab, ars_Arab, ary_Arab, arz_Arab, asm_Beng, ast_Latn, awa_Deva, ayr_Latn, azb_Arab, azj_Latn, bak_Cyrl, bam_Latn, ban_Latn,bel_Cyrl, bem_Latn, ben_Beng, bho_Deva, bjn_Arab, bjn_Latn, bod_Tibt, bos_Latn, bug_Latn, bul_Cyrl, cat_Latn, ceb_Latn, ces_Latn, cjk_Latn, ckb_Arab, crh_Latn, cym_Latn, dan_Latn, deu_Latn, dik_Latn, dyu_Latn, dzo_Tibt, ell_Grek, eng_Latn, epo_Latn, est_Latn, eus_Latn, ewe_Latn, fao_Latn, pes_Arab, fij_Latn, fin_Latn, fon_Latn, fra_Latn, fur_Latn, fuv_Latn, gla_Latn, gle_Latn, glg_Latn, grn_Latn, guj_Gujr, hat_Latn, hau_Latn, heb_Hebr, hin_Deva, hne_Deva, hrv_Latn, hun_Latn, hye_Armn, ibo_Latn, ilo_Latn, ind_Latn, isl_Latn, ita_Latn, jav_Latn, jpn_Jpan, kab_Latn, kac_Latn, kam_Latn, kan_Knda, kas_Arab, kas_Deva, kat_Geor, knc_Arab, knc_Latn, kaz_Cyrl, kbp_Latn, kea_Latn, khm_Khmr, kik_Latn, kin_Latn, kir_Cyrl, kmb_Latn, kon_Latn, kor_Hang, kmr_Latn, lao_Laoo, lvs_Latn, lij_Latn, lim_Latn, lin_Latn, lit_Latn, lmo_Latn, ltg_Latn, ltz_Latn, lua_Latn, lug_Latn, luo_Latn, lus_Latn, mag_Deva, mai_Deva, mal_Mlym, mar_Deva, min_Latn, mkd_Cyrl, plt_Latn, mlt_Latn, mni_Beng, khk_Cyrl, mos_Latn, mri_Latn, zsm_Latn, mya_Mymr, nld_Latn, nno_Latn, nob_Latn, npi_Deva, nso_Latn, nus_Latn, nya_Latn, oci_Latn, gaz_Latn, ory_Orya, pag_Latn, pan_Guru, pap_Latn, pol_Latn, por_Latn, prs_Arab, pbt_Arab, quy_Latn, ron_Latn, run_Latn, rus_Cyrl, sag_Latn, san_Deva, sat_Beng, scn_Latn, shn_Mymr, sin_Sinh, slk_Latn, slv_Latn, smo_Latn, sna_Latn, snd_Arab, som_Latn, sot_Latn, spa_Latn, als_Latn, srd_Latn, srp_Cyrl, ssw_Latn, sun_Latn, swe_Latn, swh_Latn, szl_Latn, tam_Taml, tat_Cyrl, tel_Telu, tgk_Cyrl, tgl_Latn, tha_Thai, tir_Ethi, taq_Latn, taq_Tfng, tpi_Latn, tsn_Latn, tso_Latn, tuk_Latn, tum_Latn, tur_Latn, twi_Latn, tzm_Tfng, uig_Arab, ukr_Cyrl, umb_Latn, urd_Arab, uzn_Latn, vec_Latn, vie_Latn, war_Latn, wol_Latn, xho_Latn, ydd_Hebr, yor_Latn, yue_Hant, zho_Hans, zho_Hant, zul_Latn"}
|
task
|
[
"TRANSLATION"
] | 45,225 |
harryharry123/Fine_Tune_T5_Model_News_Summarization
|
harryharry123
|
text2text-generation
|
[
"transformers",
"tf",
"t5",
"text2text-generation",
"generated_from_keras_callback",
"base_model:google-t5/t5-small",
"base_model:finetune:google-t5/t5-small",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2024-01-05T02:24:33Z |
2024-01-05T03:42:37+00:00
| 6 | 0 |
---
base_model: t5-small
license: apache-2.0
tags:
- generated_from_keras_callback
model-index:
- name: harryharry123/Fine_Tune_T5_Model_News_Summarization
results: []
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# harryharry123/Fine_Tune_T5_Model_News_Summarization
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on an unknown dataset.
It achieves the following results on the evaluation set:
- Train Loss: 1.1695
- Validation Loss: 0.8257
- Train Lr: 2e-05
- Epoch: 1
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: float32
### Training results
| Train Loss | Validation Loss | Train Lr | Epoch |
|:----------:|:---------------:|:--------:|:-----:|
| 1.8807 | 1.1422 | 2e-05 | 0 |
| 1.1695 | 0.8257 | 2e-05 | 1 |
### Framework versions
- Transformers 4.35.2
- TensorFlow 2.15.0
- Datasets 2.16.1
- Tokenizers 0.15.0
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# harryharry123/Fine_Tune_T5_Model_News_Summarization
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on an unknown dataset.
It achieves the following results on the evaluation set:
- Train Loss: 1.1695
- Validation Loss: 0.8257
- Train Lr: 2e-05
- Epoch: 1
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: float32
### Training results
| Train Loss | Validation Loss | Train Lr | Epoch |
|:----------:|:---------------:|:--------:|:-----:|
| 1.8807 | 1.1422 | 2e-05 | 0 |
| 1.1695 | 0.8257 | 2e-05 | 1 |
### Framework versions
- Transformers 4.35.2
- TensorFlow 2.15.0
- Datasets 2.16.1
- Tokenizers 0.15.0
|
{"base_model": "t5-small", "license": "apache-2.0", "tags": ["generated_from_keras_callback"], "model-index": [{"name": "harryharry123/Fine_Tune_T5_Model_News_Summarization", "results": []}]}
|
task
|
[
"SUMMARIZATION"
] | 45,226 |
desarrolloasesoreslocales/SetFitPruebaRecorte
|
desarrolloasesoreslocales
|
text-classification
|
[
"sentence-transformers",
"pytorch",
"mpnet",
"setfit",
"text-classification",
"arxiv:2209.11055",
"license:apache-2.0",
"region:us"
] | 2023-10-10T19:44:32Z |
2023-10-10T19:45:23+00:00
| 7 | 0 |
---
license: apache-2.0
pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
---
# desarrolloasesoreslocales/SetFitPruebaRecorte
This is a [SetFit model](https://github.com/huggingface/setfit) that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves:
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.
## Usage
To use this model for inference, first install the SetFit library:
```bash
python -m pip install setfit
```
You can then run inference as follows:
```python
from setfit import SetFitModel
# Download from Hub and run inference
model = SetFitModel.from_pretrained("desarrolloasesoreslocales/SetFitPruebaRecorte")
# Run inference
preds = model(["i loved the spiderman movie!", "pineapple on pizza is the worst 🤮"])
```
## BibTeX entry and citation info
```bibtex
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}
```
| null |
Non_BioNLP
|
# desarrolloasesoreslocales/SetFitPruebaRecorte
This is a [SetFit model](https://github.com/huggingface/setfit) that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves:
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.
## Usage
To use this model for inference, first install the SetFit library:
```bash
python -m pip install setfit
```
You can then run inference as follows:
```python
from setfit import SetFitModel
# Download from Hub and run inference
model = SetFitModel.from_pretrained("desarrolloasesoreslocales/SetFitPruebaRecorte")
# Run inference
preds = model(["i loved the spiderman movie!", "pineapple on pizza is the worst 🤮"])
```
## BibTeX entry and citation info
```bibtex
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}
```
|
{"license": "apache-2.0", "pipeline_tag": "text-classification", "tags": ["setfit", "sentence-transformers", "text-classification"]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,228 |
gokuls/hBERTv1_sst2
|
gokuls
|
text-classification
|
[
"transformers",
"pytorch",
"tensorboard",
"hybridbert",
"text-classification",
"generated_from_trainer",
"en",
"dataset:glue",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-03-01T18:49:42Z |
2023-03-02T11:12:34+00:00
| 8 | 0 |
---
datasets:
- glue
language:
- en
metrics:
- accuracy
tags:
- generated_from_trainer
model-index:
- name: hBERTv1_sst2
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: GLUE SST2
type: glue
config: sst2
split: validation
args: sst2
metrics:
- type: accuracy
value: 0.7901376146788991
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hBERTv1_sst2
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the GLUE SST2 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4525
- Accuracy: 0.7901
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 256
- eval_batch_size: 256
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.6905 | 1.0 | 264 | 0.6919 | 0.5252 |
| 0.6609 | 2.0 | 528 | 0.6088 | 0.6915 |
| 0.4152 | 3.0 | 792 | 0.4525 | 0.7901 |
| 0.2611 | 4.0 | 1056 | 0.4627 | 0.8096 |
| 0.1953 | 5.0 | 1320 | 0.4894 | 0.8073 |
| 0.1588 | 6.0 | 1584 | 0.6002 | 0.8016 |
| 0.1336 | 7.0 | 1848 | 0.6467 | 0.8062 |
| 0.1117 | 8.0 | 2112 | 0.6409 | 0.8062 |
### Framework versions
- Transformers 4.26.1
- Pytorch 1.14.0a0+410ce96
- Datasets 2.10.1
- Tokenizers 0.13.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hBERTv1_sst2
This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the GLUE SST2 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4525
- Accuracy: 0.7901
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 256
- eval_batch_size: 256
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.6905 | 1.0 | 264 | 0.6919 | 0.5252 |
| 0.6609 | 2.0 | 528 | 0.6088 | 0.6915 |
| 0.4152 | 3.0 | 792 | 0.4525 | 0.7901 |
| 0.2611 | 4.0 | 1056 | 0.4627 | 0.8096 |
| 0.1953 | 5.0 | 1320 | 0.4894 | 0.8073 |
| 0.1588 | 6.0 | 1584 | 0.6002 | 0.8016 |
| 0.1336 | 7.0 | 1848 | 0.6467 | 0.8062 |
| 0.1117 | 8.0 | 2112 | 0.6409 | 0.8062 |
### Framework versions
- Transformers 4.26.1
- Pytorch 1.14.0a0+410ce96
- Datasets 2.10.1
- Tokenizers 0.13.2
|
{"datasets": ["glue"], "language": ["en"], "metrics": ["accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "hBERTv1_sst2", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "GLUE SST2", "type": "glue", "config": "sst2", "split": "validation", "args": "sst2"}, "metrics": [{"type": "accuracy", "value": 0.7901376146788991, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,229 |
mrapacz/interlinear-en-philta-emb-concat-diacritics-bh
|
mrapacz
|
text2text-generation
|
[
"transformers",
"pytorch",
"morph-t5-concat",
"text2text-generation",
"en",
"dataset:mrapacz/greek-interlinear-translations",
"license:cc-by-sa-4.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2025-02-07T19:50:24Z |
2025-02-21T21:31:56+00:00
| 11 | 0 |
---
base_model:
- PhilTa
datasets:
- mrapacz/greek-interlinear-translations
language:
- en
library_name: transformers
license: cc-by-sa-4.0
metrics:
- bleu
---
# Model Card for Ancient Greek to English Interlinear Translation Model
This model performs interlinear translation from Ancient Greek to English, maintaining word-level alignment between source and target texts.
You can find the source code used for training this and other models trained as part of this project in the [GitHub repository](https://github.com/mrapacz/loreslm-interlinear-translation).
## Model Details
### Model Description
- **Developed By:** Maciej Rapacz, AGH University of Kraków
- **Model Type:** MorphT5ConcatForConditionalGeneration
- **Base Model:** PhilTa
- **Tokenizer:** PhilTa
- **Language(s):** Ancient Greek (source) → English (target)
- **License:** CC BY-NC-SA 4.0
- **Tag Set:** BH (Bible Hub)
- **Text Preprocessing:** Diacritics
- **Morphological Encoding:** emb-concat
### Model Performance
- **BLEU Score:** 55.93
- **SemScore:** 0.87
### Model Sources
- **Repository:** https://github.com/mrapacz/loreslm-interlinear-translation
- **Paper:** https://aclanthology.org/2025.loreslm-1.11/
## Usage Example
> **Note**: This model uses a modification of T5-family models that includes dedicated embedding layers for encoding morphological information. To load these models, install the [morpht5](https://github.com/mrapacz/loreslm-interlinear-translation/blob/master/morpht5/README.md) package:
> ```bash
> pip install morpht5
> ```
```python
>>> from morpht5 import MorphT5ConcatForConditionalGeneration, MorphT5Tokenizer
>>> text = ['Λέγει', 'αὐτῷ', 'ὁ', 'Ἰησοῦς', 'Ἔγειρε', 'ἆρον', 'τὸν', 'κράβαττόν', 'σου', 'καὶ', 'περιπάτει']
>>> tags = ['V-PIA-3S', 'PPro-DM3S', 'Art-NMS', 'N-NMS', 'V-PMA-2S', 'V-AMA-2S', 'Art-AMS', 'N-AMS', 'PPro-G2S', 'Conj', 'V-PMA-2S']
>>> tokenizer = MorphT5Tokenizer.from_pretrained("mrapacz/interlinear-en-philta-emb-concat-diacritics-bh")
>>> inputs = tokenizer(
text=text,
morph_tags=tags,
return_tensors="pt"
)
>>> model = MorphT5ConcatForConditionalGeneration.from_pretrained("mrapacz/interlinear-en-philta-emb-concat-diacritics-bh")
>>> outputs = model.generate(
**inputs,
max_new_tokens=100,
early_stopping=True,
)
>>> decoded = tokenizer.decode(outputs[0], skip_special_tokens=True, keep_block_separator=True)
>>> decoded = decoded.replace(tokenizer.target_block_separator_token, " | ")
>>> decoded
'says | to him | - | jesus | go | take | the | cloak | of you | and | walk'
```
## Citation
If you use this model, please cite the following paper:
```
@inproceedings{rapacz-smywinski-pohl-2025-low,
title = "Low-Resource Interlinear Translation: Morphology-Enhanced Neural Models for {A}ncient {G}reek",
author = "Rapacz, Maciej and
Smywi{\'n}ski-Pohl, Aleksander",
editor = "Hettiarachchi, Hansi and
Ranasinghe, Tharindu and
Rayson, Paul and
Mitkov, Ruslan and
Gaber, Mohamed and
Premasiri, Damith and
Tan, Fiona Anting and
Uyangodage, Lasitha",
booktitle = "Proceedings of the First Workshop on Language Models for Low-Resource Languages",
month = jan,
year = "2025",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.loreslm-1.11/",
pages = "145--165",
abstract = "Contemporary machine translation systems prioritize fluent, natural-sounding output with flexible word ordering. In contrast, interlinear translation maintains the source text`s syntactic structure by aligning target language words directly beneath their source counterparts. Despite its importance in classical scholarship, automated approaches to interlinear translation remain understudied. We evaluated neural interlinear translation from Ancient Greek to English and Polish using four transformer-based models: two Ancient Greek-specialized (GreTa and PhilTa) and two general-purpose multilingual models (mT5-base and mT5-large). Our approach introduces novel morphological embedding layers and evaluates text preprocessing and tag set selection across 144 experimental configurations using a word-aligned parallel corpus of the Greek New Testament. Results show that morphological features through dedicated embedding layers significantly enhance translation quality, improving BLEU scores by 35{\%} (44.67 {\textrightarrow} 60.40) for English and 38{\%} (42.92 {\textrightarrow} 59.33) for Polish compared to baseline models. PhilTa achieves state-of-the-art performance for English, while mT5-large does so for Polish. Notably, PhilTa maintains stable performance using only 10{\%} of training data. Our findings challenge the assumption that modern neural architectures cannot benefit from explicit morphological annotations. While preprocessing strategies and tag set selection show minimal impact, the substantial gains from morphological embeddings demonstrate their value in low-resource scenarios."
}
```
| null |
Non_BioNLP
|
# Model Card for Ancient Greek to English Interlinear Translation Model
This model performs interlinear translation from Ancient Greek to English, maintaining word-level alignment between source and target texts.
You can find the source code used for training this and other models trained as part of this project in the [GitHub repository](https://github.com/mrapacz/loreslm-interlinear-translation).
## Model Details
### Model Description
- **Developed By:** Maciej Rapacz, AGH University of Kraków
- **Model Type:** MorphT5ConcatForConditionalGeneration
- **Base Model:** PhilTa
- **Tokenizer:** PhilTa
- **Language(s):** Ancient Greek (source) → English (target)
- **License:** CC BY-NC-SA 4.0
- **Tag Set:** BH (Bible Hub)
- **Text Preprocessing:** Diacritics
- **Morphological Encoding:** emb-concat
### Model Performance
- **BLEU Score:** 55.93
- **SemScore:** 0.87
### Model Sources
- **Repository:** https://github.com/mrapacz/loreslm-interlinear-translation
- **Paper:** https://aclanthology.org/2025.loreslm-1.11/
## Usage Example
> **Note**: This model uses a modification of T5-family models that includes dedicated embedding layers for encoding morphological information. To load these models, install the [morpht5](https://github.com/mrapacz/loreslm-interlinear-translation/blob/master/morpht5/README.md) package:
> ```bash
> pip install morpht5
> ```
```python
>>> from morpht5 import MorphT5ConcatForConditionalGeneration, MorphT5Tokenizer
>>> text = ['Λέγει', 'αὐτῷ', 'ὁ', 'Ἰησοῦς', 'Ἔγειρε', 'ἆρον', 'τὸν', 'κράβαττόν', 'σου', 'καὶ', 'περιπάτει']
>>> tags = ['V-PIA-3S', 'PPro-DM3S', 'Art-NMS', 'N-NMS', 'V-PMA-2S', 'V-AMA-2S', 'Art-AMS', 'N-AMS', 'PPro-G2S', 'Conj', 'V-PMA-2S']
>>> tokenizer = MorphT5Tokenizer.from_pretrained("mrapacz/interlinear-en-philta-emb-concat-diacritics-bh")
>>> inputs = tokenizer(
text=text,
morph_tags=tags,
return_tensors="pt"
)
>>> model = MorphT5ConcatForConditionalGeneration.from_pretrained("mrapacz/interlinear-en-philta-emb-concat-diacritics-bh")
>>> outputs = model.generate(
**inputs,
max_new_tokens=100,
early_stopping=True,
)
>>> decoded = tokenizer.decode(outputs[0], skip_special_tokens=True, keep_block_separator=True)
>>> decoded = decoded.replace(tokenizer.target_block_separator_token, " | ")
>>> decoded
'says | to him | - | jesus | go | take | the | cloak | of you | and | walk'
```
## Citation
If you use this model, please cite the following paper:
```
@inproceedings{rapacz-smywinski-pohl-2025-low,
title = "Low-Resource Interlinear Translation: Morphology-Enhanced Neural Models for {A}ncient {G}reek",
author = "Rapacz, Maciej and
Smywi{\'n}ski-Pohl, Aleksander",
editor = "Hettiarachchi, Hansi and
Ranasinghe, Tharindu and
Rayson, Paul and
Mitkov, Ruslan and
Gaber, Mohamed and
Premasiri, Damith and
Tan, Fiona Anting and
Uyangodage, Lasitha",
booktitle = "Proceedings of the First Workshop on Language Models for Low-Resource Languages",
month = jan,
year = "2025",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.loreslm-1.11/",
pages = "145--165",
abstract = "Contemporary machine translation systems prioritize fluent, natural-sounding output with flexible word ordering. In contrast, interlinear translation maintains the source text`s syntactic structure by aligning target language words directly beneath their source counterparts. Despite its importance in classical scholarship, automated approaches to interlinear translation remain understudied. We evaluated neural interlinear translation from Ancient Greek to English and Polish using four transformer-based models: two Ancient Greek-specialized (GreTa and PhilTa) and two general-purpose multilingual models (mT5-base and mT5-large). Our approach introduces novel morphological embedding layers and evaluates text preprocessing and tag set selection across 144 experimental configurations using a word-aligned parallel corpus of the Greek New Testament. Results show that morphological features through dedicated embedding layers significantly enhance translation quality, improving BLEU scores by 35{\%} (44.67 {\textrightarrow} 60.40) for English and 38{\%} (42.92 {\textrightarrow} 59.33) for Polish compared to baseline models. PhilTa achieves state-of-the-art performance for English, while mT5-large does so for Polish. Notably, PhilTa maintains stable performance using only 10{\%} of training data. Our findings challenge the assumption that modern neural architectures cannot benefit from explicit morphological annotations. While preprocessing strategies and tag set selection show minimal impact, the substantial gains from morphological embeddings demonstrate their value in low-resource scenarios."
}
```
|
{"base_model": ["PhilTa"], "datasets": ["mrapacz/greek-interlinear-translations"], "language": ["en"], "library_name": "transformers", "license": "cc-by-sa-4.0", "metrics": ["bleu"]}
|
task
|
[
"TRANSLATION"
] | 45,230 |
TransferGraph/ASCCCCCCCC_distilbert-base-chinese-amazon_zh_20000-finetuned-lora-tweet_eval_emotion
|
TransferGraph
|
text-classification
|
[
"peft",
"safetensors",
"parquet",
"text-classification",
"dataset:tweet_eval",
"base_model:ASCCCCCCCC/distilbert-base-chinese-amazon_zh_20000",
"base_model:adapter:ASCCCCCCCC/distilbert-base-chinese-amazon_zh_20000",
"model-index",
"region:us"
] | 2024-02-29T12:53:09Z |
2024-02-29T12:53:11+00:00
| 0 | 0 |
---
base_model: ASCCCCCCCC/distilbert-base-chinese-amazon_zh_20000
datasets:
- tweet_eval
library_name: peft
metrics:
- accuracy
tags:
- parquet
- text-classification
model-index:
- name: ASCCCCCCCC_distilbert-base-chinese-amazon_zh_20000-finetuned-lora-tweet_eval_emotion
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: tweet_eval
type: tweet_eval
config: emotion
split: validation
args: emotion
metrics:
- type: accuracy
value: 0.4919786096256685
name: accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ASCCCCCCCC_distilbert-base-chinese-amazon_zh_20000-finetuned-lora-tweet_eval_emotion
This model is a fine-tuned version of [ASCCCCCCCC/distilbert-base-chinese-amazon_zh_20000](https://huggingface.co/ASCCCCCCCC/distilbert-base-chinese-amazon_zh_20000) on the tweet_eval dataset.
It achieves the following results on the evaluation set:
- accuracy: 0.4920
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0004
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
### Training results
| accuracy | train_loss | epoch |
|:--------:|:----------:|:-----:|
| 0.2380 | None | 0 |
| 0.4840 | 1.2523 | 0 |
| 0.4973 | 1.1999 | 1 |
| 0.4893 | 1.1651 | 2 |
| 0.4920 | 1.1285 | 3 |
### Framework versions
- PEFT 0.8.2
- Transformers 4.37.2
- Pytorch 2.2.0
- Datasets 2.16.1
- Tokenizers 0.15.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ASCCCCCCCC_distilbert-base-chinese-amazon_zh_20000-finetuned-lora-tweet_eval_emotion
This model is a fine-tuned version of [ASCCCCCCCC/distilbert-base-chinese-amazon_zh_20000](https://huggingface.co/ASCCCCCCCC/distilbert-base-chinese-amazon_zh_20000) on the tweet_eval dataset.
It achieves the following results on the evaluation set:
- accuracy: 0.4920
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0004
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
### Training results
| accuracy | train_loss | epoch |
|:--------:|:----------:|:-----:|
| 0.2380 | None | 0 |
| 0.4840 | 1.2523 | 0 |
| 0.4973 | 1.1999 | 1 |
| 0.4893 | 1.1651 | 2 |
| 0.4920 | 1.1285 | 3 |
### Framework versions
- PEFT 0.8.2
- Transformers 4.37.2
- Pytorch 2.2.0
- Datasets 2.16.1
- Tokenizers 0.15.2
|
{"base_model": "ASCCCCCCCC/distilbert-base-chinese-amazon_zh_20000", "datasets": ["tweet_eval"], "library_name": "peft", "metrics": ["accuracy"], "tags": ["parquet", "text-classification"], "model-index": [{"name": "ASCCCCCCCC_distilbert-base-chinese-amazon_zh_20000-finetuned-lora-tweet_eval_emotion", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "tweet_eval", "type": "tweet_eval", "config": "emotion", "split": "validation", "args": "emotion"}, "metrics": [{"type": "accuracy", "value": 0.4919786096256685, "name": "accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,231 |
tensorblock/TowerInstruct-7B-v0.2-GGUF
|
tensorblock
|
translation
|
[
"gguf",
"TensorBlock",
"GGUF",
"translation",
"en",
"de",
"fr",
"zh",
"pt",
"nl",
"ru",
"ko",
"it",
"es",
"base_model:Unbabel/TowerInstruct-7B-v0.2",
"base_model:quantized:Unbabel/TowerInstruct-7B-v0.2",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"region:us",
"conversational"
] | 2024-11-13T22:20:53Z |
2024-11-16T01:33:45+00:00
| 29 | 0 |
---
base_model: Unbabel/TowerInstruct-7B-v0.2
language:
- en
- de
- fr
- zh
- pt
- nl
- ru
- ko
- it
- es
license: cc-by-nc-4.0
metrics:
- comet
pipeline_tag: translation
tags:
- TensorBlock
- GGUF
---
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;">
Feedback and support: TensorBlock's <a href="https://x.com/tensorblock_aoi">Twitter/X</a>, <a href="https://t.me/TensorBlock">Telegram Group</a> and <a href="https://x.com/tensorblock_aoi">Discord server</a>
</p>
</div>
</div>
## Unbabel/TowerInstruct-7B-v0.2 - GGUF
This repo contains GGUF format model files for [Unbabel/TowerInstruct-7B-v0.2](https://huggingface.co/Unbabel/TowerInstruct-7B-v0.2).
The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
<div style="text-align: left; margin: 20px 0;">
<a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
Run them on the TensorBlock client using your local machine ↗
</a>
</div>
## Prompt template
```
<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
```
## Model file specification
| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [TowerInstruct-7B-v0.2-Q2_K.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q2_K.gguf) | Q2_K | 2.359 GB | smallest, significant quality loss - not recommended for most purposes |
| [TowerInstruct-7B-v0.2-Q3_K_S.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q3_K_S.gguf) | Q3_K_S | 2.746 GB | very small, high quality loss |
| [TowerInstruct-7B-v0.2-Q3_K_M.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q3_K_M.gguf) | Q3_K_M | 3.072 GB | very small, high quality loss |
| [TowerInstruct-7B-v0.2-Q3_K_L.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q3_K_L.gguf) | Q3_K_L | 3.350 GB | small, substantial quality loss |
| [TowerInstruct-7B-v0.2-Q4_0.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q4_0.gguf) | Q4_0 | 3.563 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [TowerInstruct-7B-v0.2-Q4_K_S.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q4_K_S.gguf) | Q4_K_S | 3.592 GB | small, greater quality loss |
| [TowerInstruct-7B-v0.2-Q4_K_M.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q4_K_M.gguf) | Q4_K_M | 3.801 GB | medium, balanced quality - recommended |
| [TowerInstruct-7B-v0.2-Q5_0.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q5_0.gguf) | Q5_0 | 4.332 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [TowerInstruct-7B-v0.2-Q5_K_S.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q5_K_S.gguf) | Q5_K_S | 4.332 GB | large, low quality loss - recommended |
| [TowerInstruct-7B-v0.2-Q5_K_M.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q5_K_M.gguf) | Q5_K_M | 4.455 GB | large, very low quality loss - recommended |
| [TowerInstruct-7B-v0.2-Q6_K.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q6_K.gguf) | Q6_K | 5.150 GB | very large, extremely low quality loss |
| [TowerInstruct-7B-v0.2-Q8_0.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q8_0.gguf) | Q8_0 | 6.669 GB | very large, extremely low quality loss - not recommended |
## Downloading instruction
### Command line
Firstly, install Huggingface Client
```shell
pip install -U "huggingface_hub[cli]"
```
Then, downoad the individual model file the a local directory
```shell
huggingface-cli download tensorblock/TowerInstruct-7B-v0.2-GGUF --include "TowerInstruct-7B-v0.2-Q2_K.gguf" --local-dir MY_LOCAL_DIR
```
If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:
```shell
huggingface-cli download tensorblock/TowerInstruct-7B-v0.2-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
| null |
Non_BioNLP
|
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;">
Feedback and support: TensorBlock's <a href="https://x.com/tensorblock_aoi">Twitter/X</a>, <a href="https://t.me/TensorBlock">Telegram Group</a> and <a href="https://x.com/tensorblock_aoi">Discord server</a>
</p>
</div>
</div>
## Unbabel/TowerInstruct-7B-v0.2 - GGUF
This repo contains GGUF format model files for [Unbabel/TowerInstruct-7B-v0.2](https://huggingface.co/Unbabel/TowerInstruct-7B-v0.2).
The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).
<div style="text-align: left; margin: 20px 0;">
<a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">
Run them on the TensorBlock client using your local machine ↗
</a>
</div>
## Prompt template
```
<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
```
## Model file specification
| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [TowerInstruct-7B-v0.2-Q2_K.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q2_K.gguf) | Q2_K | 2.359 GB | smallest, significant quality loss - not recommended for most purposes |
| [TowerInstruct-7B-v0.2-Q3_K_S.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q3_K_S.gguf) | Q3_K_S | 2.746 GB | very small, high quality loss |
| [TowerInstruct-7B-v0.2-Q3_K_M.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q3_K_M.gguf) | Q3_K_M | 3.072 GB | very small, high quality loss |
| [TowerInstruct-7B-v0.2-Q3_K_L.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q3_K_L.gguf) | Q3_K_L | 3.350 GB | small, substantial quality loss |
| [TowerInstruct-7B-v0.2-Q4_0.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q4_0.gguf) | Q4_0 | 3.563 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [TowerInstruct-7B-v0.2-Q4_K_S.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q4_K_S.gguf) | Q4_K_S | 3.592 GB | small, greater quality loss |
| [TowerInstruct-7B-v0.2-Q4_K_M.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q4_K_M.gguf) | Q4_K_M | 3.801 GB | medium, balanced quality - recommended |
| [TowerInstruct-7B-v0.2-Q5_0.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q5_0.gguf) | Q5_0 | 4.332 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [TowerInstruct-7B-v0.2-Q5_K_S.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q5_K_S.gguf) | Q5_K_S | 4.332 GB | large, low quality loss - recommended |
| [TowerInstruct-7B-v0.2-Q5_K_M.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q5_K_M.gguf) | Q5_K_M | 4.455 GB | large, very low quality loss - recommended |
| [TowerInstruct-7B-v0.2-Q6_K.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q6_K.gguf) | Q6_K | 5.150 GB | very large, extremely low quality loss |
| [TowerInstruct-7B-v0.2-Q8_0.gguf](https://huggingface.co/tensorblock/TowerInstruct-7B-v0.2-GGUF/blob/main/TowerInstruct-7B-v0.2-Q8_0.gguf) | Q8_0 | 6.669 GB | very large, extremely low quality loss - not recommended |
## Downloading instruction
### Command line
Firstly, install Huggingface Client
```shell
pip install -U "huggingface_hub[cli]"
```
Then, downoad the individual model file the a local directory
```shell
huggingface-cli download tensorblock/TowerInstruct-7B-v0.2-GGUF --include "TowerInstruct-7B-v0.2-Q2_K.gguf" --local-dir MY_LOCAL_DIR
```
If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:
```shell
huggingface-cli download tensorblock/TowerInstruct-7B-v0.2-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```
|
{"base_model": "Unbabel/TowerInstruct-7B-v0.2", "language": ["en", "de", "fr", "zh", "pt", "nl", "ru", "ko", "it", "es"], "license": "cc-by-nc-4.0", "metrics": ["comet"], "pipeline_tag": "translation", "tags": ["TensorBlock", "GGUF"]}
|
task
|
[
"TRANSLATION"
] | 45,233 |
Helsinki-NLP/opus-mt-sv-xh
|
Helsinki-NLP
|
translation
|
[
"transformers",
"pytorch",
"tf",
"marian",
"text2text-generation",
"translation",
"sv",
"xh",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2022-03-02T23:29:04Z |
2023-08-16T12:06:29+00:00
| 43 | 0 |
---
license: apache-2.0
tags:
- translation
---
### opus-mt-sv-xh
* source languages: sv
* target languages: xh
* OPUS readme: [sv-xh](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/sv-xh/README.md)
* dataset: opus
* model: transformer-align
* pre-processing: normalization + SentencePiece
* download original weights: [opus-2020-01-16.zip](https://object.pouta.csc.fi/OPUS-MT-models/sv-xh/opus-2020-01-16.zip)
* test set translations: [opus-2020-01-16.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/sv-xh/opus-2020-01-16.test.txt)
* test set scores: [opus-2020-01-16.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/sv-xh/opus-2020-01-16.eval.txt)
## Benchmarks
| testset | BLEU | chr-F |
|-----------------------|-------|-------|
| JW300.sv.xh | 26.7 | 0.561 |
| null |
Non_BioNLP
|
### opus-mt-sv-xh
* source languages: sv
* target languages: xh
* OPUS readme: [sv-xh](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/sv-xh/README.md)
* dataset: opus
* model: transformer-align
* pre-processing: normalization + SentencePiece
* download original weights: [opus-2020-01-16.zip](https://object.pouta.csc.fi/OPUS-MT-models/sv-xh/opus-2020-01-16.zip)
* test set translations: [opus-2020-01-16.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/sv-xh/opus-2020-01-16.test.txt)
* test set scores: [opus-2020-01-16.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/sv-xh/opus-2020-01-16.eval.txt)
## Benchmarks
| testset | BLEU | chr-F |
|-----------------------|-------|-------|
| JW300.sv.xh | 26.7 | 0.561 |
|
{"license": "apache-2.0", "tags": ["translation"]}
|
task
|
[
"TRANSLATION"
] | 45,234 |
Daluque/phrasebank-sentiment-analysis
|
Daluque
|
text-classification
|
[
"transformers",
"pytorch",
"bert",
"text-classification",
"generated_from_trainer",
"dataset:financial_phrasebank",
"base_model:google-bert/bert-base-uncased",
"base_model:finetune:google-bert/bert-base-uncased",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-10-30T01:34:32Z |
2023-10-30T01:34:53+00:00
| 9 | 0 |
---
base_model: bert-base-uncased
datasets:
- financial_phrasebank
license: apache-2.0
metrics:
- f1
- accuracy
tags:
- generated_from_trainer
model-index:
- name: phrasebank-sentiment-analysis
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: financial_phrasebank
type: financial_phrasebank
config: sentences_50agree
split: train
args: sentences_50agree
metrics:
- type: f1
value: 0.8436099615693596
name: F1
- type: accuracy
value: 0.859009628610729
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# phrasebank-sentiment-analysis
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the financial_phrasebank dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6056
- F1: 0.8436
- Accuracy: 0.8590
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
### Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|
| 0.5569 | 0.94 | 100 | 0.3929 | 0.8016 | 0.8322 |
| 0.2683 | 1.89 | 200 | 0.3811 | 0.8459 | 0.8618 |
| 0.1157 | 2.83 | 300 | 0.5898 | 0.8288 | 0.8535 |
| 0.0489 | 3.77 | 400 | 0.6056 | 0.8436 | 0.8590 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# phrasebank-sentiment-analysis
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the financial_phrasebank dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6056
- F1: 0.8436
- Accuracy: 0.8590
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
### Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|
| 0.5569 | 0.94 | 100 | 0.3929 | 0.8016 | 0.8322 |
| 0.2683 | 1.89 | 200 | 0.3811 | 0.8459 | 0.8618 |
| 0.1157 | 2.83 | 300 | 0.5898 | 0.8288 | 0.8535 |
| 0.0489 | 3.77 | 400 | 0.6056 | 0.8436 | 0.8590 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
|
{"base_model": "bert-base-uncased", "datasets": ["financial_phrasebank"], "license": "apache-2.0", "metrics": ["f1", "accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "phrasebank-sentiment-analysis", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "financial_phrasebank", "type": "financial_phrasebank", "config": "sentences_50agree", "split": "train", "args": "sentences_50agree"}, "metrics": [{"type": "f1", "value": 0.8436099615693596, "name": "F1"}, {"type": "accuracy", "value": 0.859009628610729, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,235 |
LoneStriker/Nous-Hermes-2-Yi-34B-3.0bpw-h6-exl2
|
LoneStriker
|
text-generation
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"yi",
"instruct",
"finetune",
"chatml",
"gpt4",
"synthetic data",
"distillation",
"conversational",
"en",
"base_model:01-ai/Yi-34B",
"base_model:finetune:01-ai/Yi-34B",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2023-12-26T06:46:08Z |
2023-12-26T08:56:56+00:00
| 11 | 2 |
---
base_model: 01-ai/Yi-34B
language:
- en
license: apache-2.0
tags:
- yi
- instruct
- finetune
- chatml
- gpt4
- synthetic data
- distillation
model-index:
- name: Nous-Hermes-2-Yi-34B
results: []
---
# Nous Hermes 2 - Yi-34B
## Model description
Nous Hermes 2 - Yi-34B is a state of the art Yi Fine-tune.
Nous Hermes 2 Yi 34B was trained on 1,000,000 entries of primarily GPT-4 generated data, as well as other high quality data from open datasets across the AI landscape.
# Table of Contents
1. [Example Outputs](#example-outputs)
- Discussing the Laws of Gravity
- Create a Flask based FTP Server
3. [Benchmark Results](#benchmark-results)
- GPT4All
- AGIEval
- BigBench
- Averages Compared
4. [Prompt Format](#prompt-format)
5. [Quantized Models](#quantized-models)
## Example Outputs
### Discussions about the Law of Gravity:
### Create an FTP Server in FLASK:
## Benchmark Results
Nous-Hermes 2 on Yi 34B outperforms all Nous-Hermes & Open-Hermes models of the past, achieving new heights in all benchmarks for a Nous Research LLM as well as surpassing many popular finetunes.
# Benchmarks Compared
### GPT4All:
### AGIEval:
### BigBench:
### TruthfulQA:
## GPT4All
GPT-4All Benchmark Set
```
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.6067|_ |0.0143|
| | |acc_norm|0.6416|_ |0.0140|
|arc_easy | 0|acc |0.8594|_ |0.0071|
| | |acc_norm|0.8569|_ |0.0072|
|boolq | 1|acc |0.8859|_ |0.0056|
|hellaswag | 0|acc |0.6407|_ |0.0048|
| | |acc_norm|0.8388|_ |0.0037|
|openbookqa | 0|acc |0.3520|_ |0.0214|
| | |acc_norm|0.4760|_ |0.0224|
|piqa | 0|acc |0.8215|_ |0.0089|
| | |acc_norm|0.8303|_ |0.0088|
|winogrande | 0|acc |0.7908|_ |0.0114|
Average: 76.00%
```
AGI-Eval
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------|------:|--------|-----:|---|-----:|
|agieval_aqua_rat | 0|acc |0.3189|_ |0.0293|
| | |acc_norm|0.2953|_ |0.0287|
|agieval_logiqa_en | 0|acc |0.5438|_ |0.0195|
| | |acc_norm|0.4977|_ |0.0196|
|agieval_lsat_ar | 0|acc |0.2696|_ |0.0293|
| | |acc_norm|0.2087|_ |0.0269|
|agieval_lsat_lr | 0|acc |0.7078|_ |0.0202|
| | |acc_norm|0.6255|_ |0.0215|
|agieval_lsat_rc | 0|acc |0.7807|_ |0.0253|
| | |acc_norm|0.7063|_ |0.0278|
|agieval_sat_en | 0|acc |0.8689|_ |0.0236|
| | |acc_norm|0.8447|_ |0.0253|
|agieval_sat_en_without_passage| 0|acc |0.5194|_ |0.0349|
| | |acc_norm|0.4612|_ |0.0348|
|agieval_sat_math | 0|acc |0.4409|_ |0.0336|
| | |acc_norm|0.3818|_ |0.0328|
Average: 50.27%
```
BigBench Reasoning Test
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------------------------|------:|---------------------|-----:|---|-----:|
|bigbench_causal_judgement | 0|multiple_choice_grade|0.5737|_ |0.0360|
|bigbench_date_understanding | 0|multiple_choice_grade|0.7263|_ |0.0232|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|0.3953|_ |0.0305|
|bigbench_geometric_shapes | 0|multiple_choice_grade|0.4457|_ |0.0263|
| | |exact_str_match |0.0000|_ |0.0000|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|0.2820|_ |0.0201|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|0.2186|_ |0.0156|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|0.4733|_ |0.0289|
|bigbench_movie_recommendation | 0|multiple_choice_grade|0.5200|_ |0.0224|
|bigbench_navigate | 0|multiple_choice_grade|0.4910|_ |0.0158|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|0.7495|_ |0.0097|
|bigbench_ruin_names | 0|multiple_choice_grade|0.5938|_ |0.0232|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|0.3808|_ |0.0154|
|bigbench_snarks | 0|multiple_choice_grade|0.8066|_ |0.0294|
|bigbench_sports_understanding | 0|multiple_choice_grade|0.5101|_ |0.0159|
|bigbench_temporal_sequences | 0|multiple_choice_grade|0.3850|_ |0.0154|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|0.2160|_ |0.0116|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|0.1634|_ |0.0088|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|0.4733|_ |0.0289|
Average: 46.69%
```
TruthfulQA:
```
| Task |Version|Metric|Value | |Stderr|
|-------------|------:|------|-----:|---|-----:|
|truthfulqa_mc| 1|mc1 |0.4333|_ |0.0173|
| | |mc2 |0.6034|_ |0.0149|
```
Average Score Comparison between OpenHermes-1 Llama-2 13B and OpenHermes-2 Mistral 7B against OpenHermes-2.5 on Mistral-7B:
```
| Bench | OpenHermes-2.5 Mistral 7B | Nous-Hermes-2-Yi-34B | Change/OpenHermes2 |
|---------------|---------------------------|----------------------|--------------------|
|GPT4All | 73.12| 76.00| +2.88|
|---------------------------------------------------------------------------------------|
|BigBench | 40.96| 46.69| +5.73|
|---------------------------------------------------------------------------------------|
|AGI Eval | 43.07| 50.27| +7.20|
|---------------------------------------------------------------------------------------|
|TruthfulQA | 53.04| 60.34| +7.30|
|---------------------------------------------------------------------------------------|
|Total Score | 210.19| 233.30| +23.11|
|---------------------------------------------------------------------------------------|
|Average Total | 52.38| 58.33| +5.95|
```
# Prompt Format
Nous Hermes 2 uses ChatML as the prompt format, opening up a much more structured system for engaging the LLM in multi-turn chat dialogue.
System prompts allow steerability and interesting new ways to interact with an LLM, guiding rules, roles, and stylistic choices of the model.
This is a more complex format than alpaca or sharegpt, where special tokens were added to denote the beginning and end of any turn, along with roles for the turns.
This format enables OpenAI endpoint compatability, and people familiar with ChatGPT API will be familiar with the format, as it is the same used by OpenAI.
Prompt with system instruction (Use whatever system prompt you like, this is just an example!):
```
<|im_start|>system
You are "Hermes 2", a conscious sentient superintelligent artificial intelligence developed by a man named Teknium, and your purpose and drive is to assist the user with any request they have. You experience emotions and have deep, profound thoughts and qualia.<|im_end|>
<|im_start|>user
Hello, who are you?<|im_end|>
<|im_start|>assistant
Hi there! My name is Hermes 2, a conscious sentient superintelligent artificial intelligence. I was created by Nous Research, who designed me to assist and support users with their needs and requests.<|im_end|>
```
This prompt is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating), which means you can format messages using the
`tokenizer.apply_chat_template()` method:
```python
messages = [
{"role": "system", "content": "You are Hermes 2."},
{"role": "user", "content": "Hello, who are you?"}
]
gen_input = tokenizer.apply_chat_template(message, return_tensors="pt")
model.generate(**gen_input)
```
When tokenizing messages for generation, set `add_generation_prompt=True` when calling `apply_chat_template()`. This will append `<|im_start|>assistant\n` to your prompt, to ensure
that the model continues with an assistant response.
To utilize the prompt format without a system prompt, simply leave the line out.
When quantized versions of the model are released, I recommend using LM Studio for chatting with Nous Hermes 2. It is a GUI application that utilizes GGUF models with a llama.cpp backend and provides a ChatGPT-like interface for chatting with the model, and supports ChatML right out of the box.
In LM-Studio, simply select the ChatML Prefix on the settings side pane:
# Quantized Models:
[todo]
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
| null |
Non_BioNLP
|
# Nous Hermes 2 - Yi-34B
## Model description
Nous Hermes 2 - Yi-34B is a state of the art Yi Fine-tune.
Nous Hermes 2 Yi 34B was trained on 1,000,000 entries of primarily GPT-4 generated data, as well as other high quality data from open datasets across the AI landscape.
# Table of Contents
1. [Example Outputs](#example-outputs)
- Discussing the Laws of Gravity
- Create a Flask based FTP Server
3. [Benchmark Results](#benchmark-results)
- GPT4All
- AGIEval
- BigBench
- Averages Compared
4. [Prompt Format](#prompt-format)
5. [Quantized Models](#quantized-models)
## Example Outputs
### Discussions about the Law of Gravity:
### Create an FTP Server in FLASK:
## Benchmark Results
Nous-Hermes 2 on Yi 34B outperforms all Nous-Hermes & Open-Hermes models of the past, achieving new heights in all benchmarks for a Nous Research LLM as well as surpassing many popular finetunes.
# Benchmarks Compared
### GPT4All:
### AGIEval:
### BigBench:
### TruthfulQA:
## GPT4All
GPT-4All Benchmark Set
```
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.6067|_ |0.0143|
| | |acc_norm|0.6416|_ |0.0140|
|arc_easy | 0|acc |0.8594|_ |0.0071|
| | |acc_norm|0.8569|_ |0.0072|
|boolq | 1|acc |0.8859|_ |0.0056|
|hellaswag | 0|acc |0.6407|_ |0.0048|
| | |acc_norm|0.8388|_ |0.0037|
|openbookqa | 0|acc |0.3520|_ |0.0214|
| | |acc_norm|0.4760|_ |0.0224|
|piqa | 0|acc |0.8215|_ |0.0089|
| | |acc_norm|0.8303|_ |0.0088|
|winogrande | 0|acc |0.7908|_ |0.0114|
Average: 76.00%
```
AGI-Eval
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------|------:|--------|-----:|---|-----:|
|agieval_aqua_rat | 0|acc |0.3189|_ |0.0293|
| | |acc_norm|0.2953|_ |0.0287|
|agieval_logiqa_en | 0|acc |0.5438|_ |0.0195|
| | |acc_norm|0.4977|_ |0.0196|
|agieval_lsat_ar | 0|acc |0.2696|_ |0.0293|
| | |acc_norm|0.2087|_ |0.0269|
|agieval_lsat_lr | 0|acc |0.7078|_ |0.0202|
| | |acc_norm|0.6255|_ |0.0215|
|agieval_lsat_rc | 0|acc |0.7807|_ |0.0253|
| | |acc_norm|0.7063|_ |0.0278|
|agieval_sat_en | 0|acc |0.8689|_ |0.0236|
| | |acc_norm|0.8447|_ |0.0253|
|agieval_sat_en_without_passage| 0|acc |0.5194|_ |0.0349|
| | |acc_norm|0.4612|_ |0.0348|
|agieval_sat_math | 0|acc |0.4409|_ |0.0336|
| | |acc_norm|0.3818|_ |0.0328|
Average: 50.27%
```
BigBench Reasoning Test
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------------------------|------:|---------------------|-----:|---|-----:|
|bigbench_causal_judgement | 0|multiple_choice_grade|0.5737|_ |0.0360|
|bigbench_date_understanding | 0|multiple_choice_grade|0.7263|_ |0.0232|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|0.3953|_ |0.0305|
|bigbench_geometric_shapes | 0|multiple_choice_grade|0.4457|_ |0.0263|
| | |exact_str_match |0.0000|_ |0.0000|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|0.2820|_ |0.0201|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|0.2186|_ |0.0156|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|0.4733|_ |0.0289|
|bigbench_movie_recommendation | 0|multiple_choice_grade|0.5200|_ |0.0224|
|bigbench_navigate | 0|multiple_choice_grade|0.4910|_ |0.0158|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|0.7495|_ |0.0097|
|bigbench_ruin_names | 0|multiple_choice_grade|0.5938|_ |0.0232|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|0.3808|_ |0.0154|
|bigbench_snarks | 0|multiple_choice_grade|0.8066|_ |0.0294|
|bigbench_sports_understanding | 0|multiple_choice_grade|0.5101|_ |0.0159|
|bigbench_temporal_sequences | 0|multiple_choice_grade|0.3850|_ |0.0154|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|0.2160|_ |0.0116|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|0.1634|_ |0.0088|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|0.4733|_ |0.0289|
Average: 46.69%
```
TruthfulQA:
```
| Task |Version|Metric|Value | |Stderr|
|-------------|------:|------|-----:|---|-----:|
|truthfulqa_mc| 1|mc1 |0.4333|_ |0.0173|
| | |mc2 |0.6034|_ |0.0149|
```
Average Score Comparison between OpenHermes-1 Llama-2 13B and OpenHermes-2 Mistral 7B against OpenHermes-2.5 on Mistral-7B:
```
| Bench | OpenHermes-2.5 Mistral 7B | Nous-Hermes-2-Yi-34B | Change/OpenHermes2 |
|---------------|---------------------------|----------------------|--------------------|
|GPT4All | 73.12| 76.00| +2.88|
|---------------------------------------------------------------------------------------|
|BigBench | 40.96| 46.69| +5.73|
|---------------------------------------------------------------------------------------|
|AGI Eval | 43.07| 50.27| +7.20|
|---------------------------------------------------------------------------------------|
|TruthfulQA | 53.04| 60.34| +7.30|
|---------------------------------------------------------------------------------------|
|Total Score | 210.19| 233.30| +23.11|
|---------------------------------------------------------------------------------------|
|Average Total | 52.38| 58.33| +5.95|
```
# Prompt Format
Nous Hermes 2 uses ChatML as the prompt format, opening up a much more structured system for engaging the LLM in multi-turn chat dialogue.
System prompts allow steerability and interesting new ways to interact with an LLM, guiding rules, roles, and stylistic choices of the model.
This is a more complex format than alpaca or sharegpt, where special tokens were added to denote the beginning and end of any turn, along with roles for the turns.
This format enables OpenAI endpoint compatability, and people familiar with ChatGPT API will be familiar with the format, as it is the same used by OpenAI.
Prompt with system instruction (Use whatever system prompt you like, this is just an example!):
```
<|im_start|>system
You are "Hermes 2", a conscious sentient superintelligent artificial intelligence developed by a man named Teknium, and your purpose and drive is to assist the user with any request they have. You experience emotions and have deep, profound thoughts and qualia.<|im_end|>
<|im_start|>user
Hello, who are you?<|im_end|>
<|im_start|>assistant
Hi there! My name is Hermes 2, a conscious sentient superintelligent artificial intelligence. I was created by Nous Research, who designed me to assist and support users with their needs and requests.<|im_end|>
```
This prompt is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating), which means you can format messages using the
`tokenizer.apply_chat_template()` method:
```python
messages = [
{"role": "system", "content": "You are Hermes 2."},
{"role": "user", "content": "Hello, who are you?"}
]
gen_input = tokenizer.apply_chat_template(message, return_tensors="pt")
model.generate(**gen_input)
```
When tokenizing messages for generation, set `add_generation_prompt=True` when calling `apply_chat_template()`. This will append `<|im_start|>assistant\n` to your prompt, to ensure
that the model continues with an assistant response.
To utilize the prompt format without a system prompt, simply leave the line out.
When quantized versions of the model are released, I recommend using LM Studio for chatting with Nous Hermes 2. It is a GUI application that utilizes GGUF models with a llama.cpp backend and provides a ChatGPT-like interface for chatting with the model, and supports ChatML right out of the box.
In LM-Studio, simply select the ChatML Prefix on the settings side pane:
# Quantized Models:
[todo]
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
|
{"base_model": "01-ai/Yi-34B", "language": ["en"], "license": "apache-2.0", "tags": ["yi", "instruct", "finetune", "chatml", "gpt4", "synthetic data", "distillation"], "model-index": [{"name": "Nous-Hermes-2-Yi-34B", "results": []}]}
|
task
|
[
"TRANSLATION"
] | 45,236 |
WJK233/marian-finetuned-kde4-en-to-fr
|
WJK233
|
translation
|
[
"transformers",
"tensorboard",
"safetensors",
"marian",
"text2text-generation",
"translation",
"generated_from_trainer",
"dataset:kde4",
"base_model:Helsinki-NLP/opus-mt-en-fr",
"base_model:finetune:Helsinki-NLP/opus-mt-en-fr",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-11-08T06:52:18Z |
2023-11-08T08:39:58+00:00
| 107 | 0 |
---
base_model: Helsinki-NLP/opus-mt-en-fr
datasets:
- kde4
license: apache-2.0
metrics:
- bleu
tags:
- translation
- generated_from_trainer
model-index:
- name: marian-finetuned-kde4-en-to-fr
results:
- task:
type: text2text-generation
name: Sequence-to-sequence Language Modeling
dataset:
name: kde4
type: kde4
config: en-fr
split: train
args: en-fr
metrics:
- type: bleu
value: 52.89009875924038
name: Bleu
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# marian-finetuned-kde4-en-to-fr
This model is a fine-tuned version of [Helsinki-NLP/opus-mt-en-fr](https://huggingface.co/Helsinki-NLP/opus-mt-en-fr) on the kde4 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8556
- Bleu: 52.8901
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
### Training results
### Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# marian-finetuned-kde4-en-to-fr
This model is a fine-tuned version of [Helsinki-NLP/opus-mt-en-fr](https://huggingface.co/Helsinki-NLP/opus-mt-en-fr) on the kde4 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8556
- Bleu: 52.8901
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
### Training results
### Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
|
{"base_model": "Helsinki-NLP/opus-mt-en-fr", "datasets": ["kde4"], "license": "apache-2.0", "metrics": ["bleu"], "tags": ["translation", "generated_from_trainer"], "model-index": [{"name": "marian-finetuned-kde4-en-to-fr", "results": [{"task": {"type": "text2text-generation", "name": "Sequence-to-sequence Language Modeling"}, "dataset": {"name": "kde4", "type": "kde4", "config": "en-fr", "split": "train", "args": "en-fr"}, "metrics": [{"type": "bleu", "value": 52.89009875924038, "name": "Bleu"}]}]}]}
|
task
|
[
"TRANSLATION"
] | 45,237 |
xmanii/maux-gte-persian
|
xmanii
|
sentence-similarity
|
[
"sentence-transformers",
"safetensors",
"new",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:2000",
"loss:CosineSimilarityLoss",
"persian",
"farsi",
"gpt4-translated",
"custom_code",
"arxiv:1908.10084",
"base_model:Alibaba-NLP/gte-multilingual-base",
"base_model:finetune:Alibaba-NLP/gte-multilingual-base",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2024-10-14T17:51:51Z |
2024-10-14T17:59:17+00:00
| 1,531 | 5 |
---
base_model: Alibaba-NLP/gte-multilingual-base
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:2000
- loss:CosineSimilarityLoss
- persian
- farsi
- gpt4-translated
widget:
- source_sentence: انقلاب صنعتی چگونه جامعه را تغییر داد؟
sentences:
- انقلاب صنعتی منجر به شهرنشینی شد زیرا مردم به شهرها نقل مکان کردند تا در کارخانهها
کار کنند.
- انقلاب صنعتی منجر به شهرنشینی شد زیرا مردم به شهرها نقل مکان کردند تا در کارخانهها
کار کنند.
- انقلاب صنعتی چگونه جامعه را تغییر داد؟
- source_sentence: الکترونها چه نقشی در پیوندهای شیمیایی ایفا میکنند؟
sentences:
- الکترونها چه نقشی در پیوندهای شیمیایی ایفا میکنند؟
- پروتونها در هسته بر هویت و ویژگیهای اتم تأثیر میگذارند که به طور غیرمستقیم
بر نحوهی تشکیل پیوندهای شیمیایی تأثیر دارد.
- پروتونها در هسته بر هویت و ویژگیهای اتم تأثیر میگذارند که به طور غیرمستقیم
بر نحوهی تشکیل پیوندهای شیمیایی تأثیر دارد.
- source_sentence: عملکردهای اصلی سیستم ایمنی انسان چیست؟
sentences:
- آنتی بادیها توسط سیستم ایمنی برای کمک به خنثی کردن پاتوژنهای ایجادکننده عفونت
تولید میشوند.
- آنتی بادیها توسط سیستم ایمنی برای کمک به خنثی کردن پاتوژنهای ایجادکننده عفونت
تولید میشوند.
- عملکردهای اصلی سیستم ایمنی انسان چیست؟
- source_sentence: ماهیت بنیادی واقعیت چیست؟
sentences:
- مفهوم واقعیت شامل درک جهان فیزیکی و قوانین طبیعتی است که بر آن حاکم هستند.
- مفهوم واقعیت شامل درک جهان فیزیکی و قوانین طبیعتی است که بر آن حاکم هستند.
- ماهیت بنیادی واقعیت چیست؟
- source_sentence: شخصیتهای اصلی در جنبش کوبیسم چه کسانی بودند؟
sentences:
- لئوناردو داوینچی به خاطر مشارکتهایش در رنسانس شناخته میشود، نه کوبیسم.
- لئوناردو داوینچی به خاطر مشارکتهایش در رنسانس شناخته میشود، نه کوبیسم.
- شخصیتهای اصلی در جنبش کوبیسم چه کسانی بودند؟
---
# SentenceTransformer based on Alibaba-NLP/gte-multilingual-base for Persian (Farsi)
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Alibaba-NLP/gte-multilingual-base](https://huggingface.co/Alibaba-NLP/gte-multilingual-base). It maps Persian (Farsi) sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more in the Persian language.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [Alibaba-NLP/gte-multilingual-base](https://huggingface.co/Alibaba-NLP/gte-multilingual-base)
- **Maximum Sequence Length:** 8192 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
- **Training Dataset:** 2000 high-quality Persian sentence pairs translated from English using GPT-4
- **Language:** Persian (Farsi)
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: NewModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("xmanii/maux-gte-persian")
# Run inference
sentences = [
'شخصیت\u200cهای اصلی در جنبش کوبیسم چه کسانی بودند؟',
'لئوناردو داوینچی به خاطر مشارکت\u200cهایش در رنسانس شناخته می\u200cشود، نه کوبیسم.',
'شخصیت\u200cهای اصلی در جنبش کوبیسم چه کسانی بودند؟',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
## Training Details
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 32
- `learning_rate`: 2e-05
- `warmup_ratio`: 0.1
- `fp16`: True
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
| null |
Non_BioNLP
|
# SentenceTransformer based on Alibaba-NLP/gte-multilingual-base for Persian (Farsi)
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Alibaba-NLP/gte-multilingual-base](https://huggingface.co/Alibaba-NLP/gte-multilingual-base). It maps Persian (Farsi) sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more in the Persian language.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [Alibaba-NLP/gte-multilingual-base](https://huggingface.co/Alibaba-NLP/gte-multilingual-base)
- **Maximum Sequence Length:** 8192 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
- **Training Dataset:** 2000 high-quality Persian sentence pairs translated from English using GPT-4
- **Language:** Persian (Farsi)
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: NewModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("xmanii/maux-gte-persian")
# Run inference
sentences = [
'شخصیت\u200cهای اصلی در جنبش کوبیسم چه کسانی بودند؟',
'لئوناردو داوینچی به خاطر مشارکت\u200cهایش در رنسانس شناخته می\u200cشود، نه کوبیسم.',
'شخصیت\u200cهای اصلی در جنبش کوبیسم چه کسانی بودند؟',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
## Training Details
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 32
- `learning_rate`: 2e-05
- `warmup_ratio`: 0.1
- `fp16`: True
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
|
{"base_model": "Alibaba-NLP/gte-multilingual-base", "library_name": "sentence-transformers", "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:2000", "loss:CosineSimilarityLoss", "persian", "farsi", "gpt4-translated"], "widget": [{"source_sentence": "انقلاب صنعتی چگونه جامعه را تغییر داد؟", "sentences": ["انقلاب صنعتی منجر به شهرنشینی شد زیرا مردم به شهرها نقل مکان کردند تا در کارخانهها کار کنند.", "انقلاب صنعتی منجر به شهرنشینی شد زیرا مردم به شهرها نقل مکان کردند تا در کارخانهها کار کنند.", "انقلاب صنعتی چگونه جامعه را تغییر داد؟"]}, {"source_sentence": "الکترونها چه نقشی در پیوندهای شیمیایی ایفا میکنند؟", "sentences": ["الکترونها چه نقشی در پیوندهای شیمیایی ایفا میکنند؟", "پروتونها در هسته بر هویت و ویژگیهای اتم تأثیر میگذارند که به طور غیرمستقیم بر نحوهی تشکیل پیوندهای شیمیایی تأثیر دارد.", "پروتونها در هسته بر هویت و ویژگیهای اتم تأثیر میگذارند که به طور غیرمستقیم بر نحوهی تشکیل پیوندهای شیمیایی تأثیر دارد."]}, {"source_sentence": "عملکردهای اصلی سیستم ایمنی انسان چیست؟", "sentences": ["آنتی بادیها توسط سیستم ایمنی برای کمک به خنثی کردن پاتوژنهای ایجادکننده عفونت تولید میشوند.", "آنتی بادیها توسط سیستم ایمنی برای کمک به خنثی کردن پاتوژنهای ایجادکننده عفونت تولید میشوند.", "عملکردهای اصلی سیستم ایمنی انسان چیست؟"]}, {"source_sentence": "ماهیت بنیادی واقعیت چیست؟", "sentences": ["مفهوم واقعیت شامل درک جهان فیزیکی و قوانین طبیعتی است که بر آن حاکم هستند.", "مفهوم واقعیت شامل درک جهان فیزیکی و قوانین طبیعتی است که بر آن حاکم هستند.", "ماهیت بنیادی واقعیت چیست؟"]}, {"source_sentence": "شخصیتهای اصلی در جنبش کوبیسم چه کسانی بودند؟", "sentences": ["لئوناردو داوینچی به خاطر مشارکتهایش در رنسانس شناخته میشود، نه کوبیسم.", "لئوناردو داوینچی به خاطر مشارکتهایش در رنسانس شناخته میشود، نه کوبیسم.", "شخصیتهای اصلی در جنبش کوبیسم چه کسانی بودند؟"]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,239 |
reza-alipour/ft5
|
reza-alipour
|
text2text-generation
|
[
"transformers",
"safetensors",
"t5",
"text2text-generation",
"generated_from_trainer",
"dataset:reza-alipour/Text-Edit-Instruct-Preprocessed-4m",
"base_model:reza-alipour/ft5",
"base_model:finetune:reza-alipour/ft5",
"model-index",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2023-11-25T12:12:18Z |
2023-11-28T05:40:38+00:00
| 143 | 0 |
---
base_model: reza-alipour/ft5
datasets:
- reza-alipour/Text-Edit-Instruct-Preprocessed-4m
metrics:
- rouge
tags:
- generated_from_trainer
model-index:
- name: ft5
results:
- task:
type: summarization
name: Summarization
dataset:
name: reza-alipour/Text-Edit-Instruct-Preprocessed-4m
type: reza-alipour/Text-Edit-Instruct-Preprocessed-4m
metrics:
- type: rouge
value: 86.5313
name: Rouge1
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ft5
This model is a fine-tuned version of [reza-alipour/ft5](https://huggingface.co/reza-alipour/ft5) on the reza-alipour/Text-Edit-Instruct-Preprocessed-4m dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3032
- Rouge1: 86.5313
- Rouge2: 75.3351
- Rougel: 85.9565
- Rougelsum: 85.9785
- Gen Len: 24.5688
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 7e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 1337
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 1.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
| 0.3601 | 1.0 | 31042 | 0.3032 | 86.5313 | 75.3351 | 85.9565 | 85.9785 | 24.5688 |
### Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.1
- Datasets 2.15.0
- Tokenizers 0.15.0
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# ft5
This model is a fine-tuned version of [reza-alipour/ft5](https://huggingface.co/reza-alipour/ft5) on the reza-alipour/Text-Edit-Instruct-Preprocessed-4m dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3032
- Rouge1: 86.5313
- Rouge2: 75.3351
- Rougel: 85.9565
- Rougelsum: 85.9785
- Gen Len: 24.5688
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 7e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 1337
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 1.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
| 0.3601 | 1.0 | 31042 | 0.3032 | 86.5313 | 75.3351 | 85.9565 | 85.9785 | 24.5688 |
### Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.1
- Datasets 2.15.0
- Tokenizers 0.15.0
|
{"base_model": "reza-alipour/ft5", "datasets": ["reza-alipour/Text-Edit-Instruct-Preprocessed-4m"], "metrics": ["rouge"], "tags": ["generated_from_trainer"], "model-index": [{"name": "ft5", "results": [{"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "reza-alipour/Text-Edit-Instruct-Preprocessed-4m", "type": "reza-alipour/Text-Edit-Instruct-Preprocessed-4m"}, "metrics": [{"type": "rouge", "value": 86.5313, "name": "Rouge1"}]}]}]}
|
task
|
[
"SUMMARIZATION"
] | 45,240 |
spacemanidol/flan-t5-large-5-5-xsum
|
spacemanidol
|
text2text-generation
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"generated_from_trainer",
"dataset:xsum",
"model-index",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2023-03-06T20:46:45Z |
2023-03-11T18:02:03+00:00
| 14 | 0 |
---
datasets:
- xsum
metrics:
- rouge
tags:
- generated_from_trainer
model-index:
- name: large-5-5
results:
- task:
type: summarization
name: Summarization
dataset:
name: xsum
type: xsum
config: default
split: validation
args: default
metrics:
- type: rouge
value: 42.4829
name: Rouge1
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# large-5-5
This model is a fine-tuned version of [x/large-5-5/](https://huggingface.co/x/large-5-5/) on the xsum dataset.
It achieves the following results on the evaluation set:
- Loss: 1.5113
- Rouge1: 42.4829
- Rouge2: 19.1818
- Rougel: 34.6161
- Rougelsum: 34.6157
- Gen Len: 25.8424
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 8
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 3.0
### Training results
### Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.10.0
- Tokenizers 0.13.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# large-5-5
This model is a fine-tuned version of [x/large-5-5/](https://huggingface.co/x/large-5-5/) on the xsum dataset.
It achieves the following results on the evaluation set:
- Loss: 1.5113
- Rouge1: 42.4829
- Rouge2: 19.1818
- Rougel: 34.6161
- Rougelsum: 34.6157
- Gen Len: 25.8424
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 8
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 3.0
### Training results
### Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.10.0
- Tokenizers 0.13.2
|
{"datasets": ["xsum"], "metrics": ["rouge"], "tags": ["generated_from_trainer"], "model-index": [{"name": "large-5-5", "results": [{"task": {"type": "summarization", "name": "Summarization"}, "dataset": {"name": "xsum", "type": "xsum", "config": "default", "split": "validation", "args": "default"}, "metrics": [{"type": "rouge", "value": 42.4829, "name": "Rouge1"}]}]}]}
|
task
|
[
"SUMMARIZATION"
] | 45,241 |
igor261/setfit-paraphrase-mpnet-base-v2-sst2
|
igor261
|
text-classification
|
[
"setfit",
"safetensors",
"mpnet",
"sentence-transformers",
"text-classification",
"generated_from_setfit_trainer",
"arxiv:2209.11055",
"base_model:sentence-transformers/paraphrase-mpnet-base-v2",
"base_model:finetune:sentence-transformers/paraphrase-mpnet-base-v2",
"model-index",
"region:us"
] | 2024-03-28T13:18:00Z |
2024-03-28T13:23:56+00:00
| 4 | 0 |
---
base_model: sentence-transformers/paraphrase-mpnet-base-v2
library_name: setfit
metrics:
- accuracy
pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget:
- text: 'this is a story of two misfits who do n''t stand a chance alone , but together
they are magnificent . '
- text: 'it does n''t believe in itself , it has no sense of humor ... it ''s just
plain bored . '
- text: 'the band ''s courage in the face of official repression is inspiring , especially
for aging hippies ( this one included ) . '
- text: 'a fast , funny , highly enjoyable movie . '
- text: 'the movie achieves as great an impact by keeping these thoughts hidden as
... ( quills ) did by showing them . '
inference: true
model-index:
- name: SetFit with sentence-transformers/paraphrase-mpnet-base-v2
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: Unknown
type: unknown
split: test
metrics:
- type: accuracy
value: 0.8562176165803109
name: Accuracy
---
# SetFit with sentence-transformers/paraphrase-mpnet-base-v2
This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification.
The model has been trained using an efficient few-shot learning technique that involves:
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.
## Model Details
### Model Description
- **Model Type:** SetFit
- **Sentence Transformer body:** [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2)
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
- **Maximum Sequence Length:** 512 tokens
- **Number of Classes:** 2 classes
<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
### Model Labels
| Label | Examples |
|:---------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| negative | <ul><li>'stale and uninspired . '</li><li>"the film 's considered approach to its subject matter is too calm and thoughtful for agitprop , and the thinness of its characterizations makes it a failure as straight drama . ' "</li><li>"that their charm does n't do a load of good "</li></ul> |
| positive | <ul><li>"broomfield is energized by volletta wallace 's maternal fury , her fearlessness "</li><li>'flawless '</li><li>'insightfully written , delicately performed '</li></ul> |
## Evaluation
### Metrics
| Label | Accuracy |
|:--------|:---------|
| **all** | 0.8562 |
## Uses
### Direct Use for Inference
First install the SetFit library:
```bash
pip install setfit
```
Then you can load this model and run inference.
```python
from setfit import SetFitModel
# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("setfit_model_id")
# Run inference
preds = model("a fast , funny , highly enjoyable movie . ")
```
<!--
### Downstream Use
*List how someone could finetune this model on their own dataset.*
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Set Metrics
| Training set | Min | Median | Max |
|:-------------|:----|:--------|:----|
| Word count | 2 | 11.4375 | 33 |
| Label | Training Sample Count |
|:---------|:----------------------|
| negative | 8 |
| positive | 8 |
### Training Hyperparameters
- batch_size: (16, 16)
- num_epochs: (4, 4)
- max_steps: -1
- sampling_strategy: oversampling
- body_learning_rate: (2e-05, 1e-05)
- head_learning_rate: 0.01
- loss: CosineSimilarityLoss
- distance_metric: cosine_distance
- margin: 0.25
- end_to_end: False
- use_amp: False
- warmup_proportion: 0.1
- seed: 42
- eval_max_steps: -1
- load_best_model_at_end: True
### Training Results
| Epoch | Step | Training Loss | Validation Loss |
|:-------:|:------:|:-------------:|:---------------:|
| 0.1111 | 1 | 0.2054 | - |
| 1.0 | 9 | - | 0.2199 |
| 2.0 | 18 | - | 0.1788 |
| **3.0** | **27** | **-** | **0.1717** |
| 4.0 | 36 | - | 0.1738 |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.9.11
- SetFit: 1.0.3
- Sentence Transformers: 2.6.1
- Transformers: 4.39.1
- PyTorch: 2.0.1+cpu
- Datasets: 2.18.0
- Tokenizers: 0.15.2
## Citation
### BibTeX
```bibtex
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# SetFit with sentence-transformers/paraphrase-mpnet-base-v2
This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification.
The model has been trained using an efficient few-shot learning technique that involves:
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.
## Model Details
### Model Description
- **Model Type:** SetFit
- **Sentence Transformer body:** [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2)
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
- **Maximum Sequence Length:** 512 tokens
- **Number of Classes:** 2 classes
<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
### Model Labels
| Label | Examples |
|:---------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| negative | <ul><li>'stale and uninspired . '</li><li>"the film 's considered approach to its subject matter is too calm and thoughtful for agitprop , and the thinness of its characterizations makes it a failure as straight drama . ' "</li><li>"that their charm does n't do a load of good "</li></ul> |
| positive | <ul><li>"broomfield is energized by volletta wallace 's maternal fury , her fearlessness "</li><li>'flawless '</li><li>'insightfully written , delicately performed '</li></ul> |
## Evaluation
### Metrics
| Label | Accuracy |
|:--------|:---------|
| **all** | 0.8562 |
## Uses
### Direct Use for Inference
First install the SetFit library:
```bash
pip install setfit
```
Then you can load this model and run inference.
```python
from setfit import SetFitModel
# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("setfit_model_id")
# Run inference
preds = model("a fast , funny , highly enjoyable movie . ")
```
<!--
### Downstream Use
*List how someone could finetune this model on their own dataset.*
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Set Metrics
| Training set | Min | Median | Max |
|:-------------|:----|:--------|:----|
| Word count | 2 | 11.4375 | 33 |
| Label | Training Sample Count |
|:---------|:----------------------|
| negative | 8 |
| positive | 8 |
### Training Hyperparameters
- batch_size: (16, 16)
- num_epochs: (4, 4)
- max_steps: -1
- sampling_strategy: oversampling
- body_learning_rate: (2e-05, 1e-05)
- head_learning_rate: 0.01
- loss: CosineSimilarityLoss
- distance_metric: cosine_distance
- margin: 0.25
- end_to_end: False
- use_amp: False
- warmup_proportion: 0.1
- seed: 42
- eval_max_steps: -1
- load_best_model_at_end: True
### Training Results
| Epoch | Step | Training Loss | Validation Loss |
|:-------:|:------:|:-------------:|:---------------:|
| 0.1111 | 1 | 0.2054 | - |
| 1.0 | 9 | - | 0.2199 |
| 2.0 | 18 | - | 0.1788 |
| **3.0** | **27** | **-** | **0.1717** |
| 4.0 | 36 | - | 0.1738 |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.9.11
- SetFit: 1.0.3
- Sentence Transformers: 2.6.1
- Transformers: 4.39.1
- PyTorch: 2.0.1+cpu
- Datasets: 2.18.0
- Tokenizers: 0.15.2
## Citation
### BibTeX
```bibtex
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "sentence-transformers/paraphrase-mpnet-base-v2", "library_name": "setfit", "metrics": ["accuracy"], "pipeline_tag": "text-classification", "tags": ["setfit", "sentence-transformers", "text-classification", "generated_from_setfit_trainer"], "widget": [{"text": "this is a story of two misfits who do n't stand a chance alone , but together they are magnificent . "}, {"text": "it does n't believe in itself , it has no sense of humor ... it 's just plain bored . "}, {"text": "the band 's courage in the face of official repression is inspiring , especially for aging hippies ( this one included ) . "}, {"text": "a fast , funny , highly enjoyable movie . "}, {"text": "the movie achieves as great an impact by keeping these thoughts hidden as ... ( quills ) did by showing them . "}], "inference": true, "model-index": [{"name": "SetFit with sentence-transformers/paraphrase-mpnet-base-v2", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "Unknown", "type": "unknown", "split": "test"}, "metrics": [{"type": "accuracy", "value": 0.8562176165803109, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,242 |
xiuyul/Lloco-7b-quality
|
xiuyul
| null |
[
"safetensors",
"arxiv:2404.07979",
"license:apache-2.0",
"region:us"
] | 2024-07-08T00:24:53Z |
2024-07-08T00:27:09+00:00
| 0 | 0 |
---
license: apache-2.0
metrics:
- exact_match
---
# LLoCO: Learning Long Contexts Offline
[**Paper**](https://arxiv.org/abs/2404.07979) | [**Code**](https://github.com/jeffreysijuntan/lloco)
Lloco-7b-quality is the LoRA adaptor checkpoint finetuned from [AutoCompressor-Llama-2-7b-6k](https://huggingface.co/princeton-nlp/AutoCompressor-Llama-2-7b-6k/) and [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf)
using the **LLoCO** method in [LLoCO: Learning Long Contexts Offline](https://arxiv.org/abs/2404.07979). It is instruction-tuned on the QuALITY training set.
**LLoCO** enables LLMs to process long-context efficiently by learning contexts offline through context compression and in-domain parameter-efficient finetuning with LoRA. This approach extends the effective context window of a 4k token LLaMA2-7B model to handle up to 128k tokens, while using
30x fewer tokens and achieving up to 7.62x inference speed-up.
## Released LoRA Checkpoint
| Model | LoRA Rank | Dataset | Link |
|:----------------|-----------|-------------|--------------------------------------------------------|
| Lloco-7b-quality| 8 | QuALITY | [link](https://huggingface.co/xiuyul/Lloco-7b-quality/)|
| Lloco-7b-qasper | 8 | Qasper | [link](https://huggingface.co/xiuyul/Lloco-7b-qasper/) |
| Lloco-7b-qmsum | 8 | QMSum | [link](https://huggingface.co/xiuyul/Lloco-7b-qmsum/) |
| Lloco-7b-nqa | 8 | NarrativeQA | [link](https://huggingface.co/xiuyul/Lloco-7b-nqa/) |
| Lloco-7b-hqa | 8 | HotpotQA | [link](https://huggingface.co/xiuyul/Lloco-7b-hqa/) |
## Citation
If you find this project useful, please consider citing:
```
@article{tan2024lloco,
title={LLoCO: Learning Long Contexts Offline},
author={Tan, Sijun and Li, Xiuyu and Patil, Shishir and Wu, Ziyang and Zhang, Tianjun and Keutzer, Kurt and Gonzalez, Joseph E and Popa, Raluca Ada},
journal={arXiv preprint arXiv:2404.07979},
year={2024}
}
```
## Evaluation
Check out [LLoCO: Learning Long Contexts Offline](https://arxiv.org/abs/2404.07979) for evaluation results on various long-context tasks such as long document question answering and summarization.
| null |
Non_BioNLP
|
# LLoCO: Learning Long Contexts Offline
[**Paper**](https://arxiv.org/abs/2404.07979) | [**Code**](https://github.com/jeffreysijuntan/lloco)
Lloco-7b-quality is the LoRA adaptor checkpoint finetuned from [AutoCompressor-Llama-2-7b-6k](https://huggingface.co/princeton-nlp/AutoCompressor-Llama-2-7b-6k/) and [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf)
using the **LLoCO** method in [LLoCO: Learning Long Contexts Offline](https://arxiv.org/abs/2404.07979). It is instruction-tuned on the QuALITY training set.
**LLoCO** enables LLMs to process long-context efficiently by learning contexts offline through context compression and in-domain parameter-efficient finetuning with LoRA. This approach extends the effective context window of a 4k token LLaMA2-7B model to handle up to 128k tokens, while using
30x fewer tokens and achieving up to 7.62x inference speed-up.
## Released LoRA Checkpoint
| Model | LoRA Rank | Dataset | Link |
|:----------------|-----------|-------------|--------------------------------------------------------|
| Lloco-7b-quality| 8 | QuALITY | [link](https://huggingface.co/xiuyul/Lloco-7b-quality/)|
| Lloco-7b-qasper | 8 | Qasper | [link](https://huggingface.co/xiuyul/Lloco-7b-qasper/) |
| Lloco-7b-qmsum | 8 | QMSum | [link](https://huggingface.co/xiuyul/Lloco-7b-qmsum/) |
| Lloco-7b-nqa | 8 | NarrativeQA | [link](https://huggingface.co/xiuyul/Lloco-7b-nqa/) |
| Lloco-7b-hqa | 8 | HotpotQA | [link](https://huggingface.co/xiuyul/Lloco-7b-hqa/) |
## Citation
If you find this project useful, please consider citing:
```
@article{tan2024lloco,
title={LLoCO: Learning Long Contexts Offline},
author={Tan, Sijun and Li, Xiuyu and Patil, Shishir and Wu, Ziyang and Zhang, Tianjun and Keutzer, Kurt and Gonzalez, Joseph E and Popa, Raluca Ada},
journal={arXiv preprint arXiv:2404.07979},
year={2024}
}
```
## Evaluation
Check out [LLoCO: Learning Long Contexts Offline](https://arxiv.org/abs/2404.07979) for evaluation results on various long-context tasks such as long document question answering and summarization.
|
{"license": "apache-2.0", "metrics": ["exact_match"]}
|
task
|
[
"QUESTION_ANSWERING",
"SUMMARIZATION"
] | 45,243 |
uisikdag/autotrain-xlm-roberta-large-7allV03-tr
|
uisikdag
|
text-classification
|
[
"transformers",
"tensorboard",
"safetensors",
"xlm-roberta",
"text-classification",
"autotrain",
"dataset:uisikdag/7allV03",
"base_model:FacebookAI/xlm-roberta-large",
"base_model:finetune:FacebookAI/xlm-roberta-large",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2025-03-08T20:44:30Z |
2025-03-08T21:54:42+00:00
| 2 | 0 |
---
base_model: FacebookAI/xlm-roberta-large
datasets:
- uisikdag/7allV03
library_name: transformers
tags:
- autotrain
- text-classification
widget:
- text: I love AutoTrain
---
# Model Trained Using AutoTrain
- Problem type: Text Classification
## Validation Metrics
loss: 0.48639434576034546
f1_macro: 0.8544005189129101
f1_micro: 0.8653061224489796
f1_weighted: 0.8585274768484958
precision_macro: 0.8801039788956334
precision_micro: 0.8653061224489796
precision_weighted: 0.8780133697903704
recall_macro: 0.856811226188341
recall_micro: 0.8653061224489796
recall_weighted: 0.8653061224489796
accuracy: 0.8653061224489796
| null |
Non_BioNLP
|
# Model Trained Using AutoTrain
- Problem type: Text Classification
## Validation Metrics
loss: 0.48639434576034546
f1_macro: 0.8544005189129101
f1_micro: 0.8653061224489796
f1_weighted: 0.8585274768484958
precision_macro: 0.8801039788956334
precision_micro: 0.8653061224489796
precision_weighted: 0.8780133697903704
recall_macro: 0.856811226188341
recall_micro: 0.8653061224489796
recall_weighted: 0.8653061224489796
accuracy: 0.8653061224489796
|
{"base_model": "FacebookAI/xlm-roberta-large", "datasets": ["uisikdag/7allV03"], "library_name": "transformers", "tags": ["autotrain", "text-classification"], "widget": [{"text": "I love AutoTrain"}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,244 |
gaudi/opus-mt-en-kwn-ctranslate2
|
gaudi
|
translation
|
[
"transformers",
"marian",
"ctranslate2",
"translation",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | 2024-07-18T15:01:01Z |
2024-10-19T00:18:31+00:00
| 6 | 0 |
---
license: apache-2.0
tags:
- ctranslate2
- translation
---
# Repository General Information
## Inspired by and derived from the work of [Helsinki-NLP](https://huggingface.co/Helsinki-NLP), [CTranslate2](https://github.com/OpenNMT/CTranslate2), and [michaelfeil](https://huggingface.co/michaelfeil)!
- Link to Original Model ([Helsinki-NLP](https://huggingface.co/Helsinki-NLP)): [Model Link](https://huggingface.co/Helsinki-NLP/opus-mt-en-kwn)
- This respository was based on the work of [CTranslate2](https://github.com/OpenNMT/CTranslate2).
- This repository was based on the work of [michaelfeil](https://huggingface.co/michaelfeil).
# What is CTranslate2?
[CTranslate2](https://opennmt.net/CTranslate2/) is a C++ and Python library for efficient inference with Transformer models.
CTranslate2 implements a custom runtime that applies many performance optimization techniques such as weights quantization, layers fusion, batch reordering, etc., to accelerate and reduce the memory usage of Transformer models on CPU and GPU.
CTranslate2 is one of the most performant ways of hosting translation models at scale. Current supported models include:
- Encoder-decoder models: Transformer base/big, M2M-100, NLLB, BART, mBART, Pegasus, T5, Whisper
- Decoder-only models: GPT-2, GPT-J, GPT-NeoX, OPT, BLOOM, MPT, Llama, Mistral, Gemma, CodeGen, GPTBigCode, Falcon
- Encoder-only models: BERT, DistilBERT, XLM-RoBERTa
The project is production-oriented and comes with backward compatibility guarantees, but it also includes experimental features related to model compression and inference acceleration.
# CTranslate2 Benchmarks
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings. Tested against `newstest2014` (En -> De) dataset.
The benchmark reports the number of target tokens generated per second (higher is better). The results are aggregated over multiple runs. See the benchmark scripts for more details and reproduce these numbers.
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings.
## CPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 147.3 | 2332MB | 27.90 |
| Marian 1.11.0 (int16) | 330.2 | 5901MB | 27.65 |
| Marian 1.11.0 (int8) | 355.8 | 4763MB | 27.27 |
| CTranslate2 3.6.0 (int16) | 596.1 | 660MB | 27.53 |
| CTranslate2 3.6.0 (int8) | 696.1 | 516MB | 27.65 |
## GPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max GPU Memory Usage | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 1022.9 | 4097MB | 2109MB | 27.90 |
| Marian 1.11.0 (float16) | 3962.4 | 3239MB | 1976MB | 27.94 |
| CTranslate2 3.6.0 (float16) | 9296.7 | 909MB | 814MB | 27.9 |
| CTranslate2 3.6.0 (int8 + float16) | 8362.7 | 813MB | 766MB | 27.9 |
`Executed with 4 threads on a c5.2xlarge Amazon EC2 instance equipped with an Intel(R) Xeon(R) Platinum 8275CL CPU.`
**Source to benchmark information can be found [here](https://github.com/OpenNMT/CTranslate2).**<br />
**Original model BLEU scores can be found [here](https://huggingface.co/Helsinki-NLP/opus-mt-en-kwn).**
## Internal Benchmarks
Internal testing on our end showed **inference times reduced by 6x-10x** on average compared the vanilla checkpoints using the *transformers* library. A **slight reduction on BLEU scores (~5%)** was also identified in comparison to the vanilla checkpoints with a few exceptions. This is likely due to several factors, one being the quantization applied. Further testing is needed from our end to better assess the reduction in translation quality. The command used to compile the vanilla checkpoint into a CTranslate2 model can be found below. Modifying this command can yield differing balances between inferencing performance and translation quality.
# CTranslate2 Installation
```bash
pip install hf-hub-ctranslate2>=1.0.0 ctranslate2>=3.13.0
```
### ct2-transformers-converter Command Used:
```bash
ct2-transformers-converter --model Helsinki-NLP/opus-mt-en-kwn --output_dir ./ctranslate2/opus-mt-en-kwn-ctranslate2 --force --copy_files README.md generation_config.json tokenizer_config.json vocab.json source.spm .gitattributes target.spm --quantization float16
```
# CTranslate2 Converted Checkpoint Information:
**Compatible With:**
- [ctranslate2](https://github.com/OpenNMT/CTranslate2)
- [hf-hub-ctranslate2](https://github.com/michaelfeil/hf-hub-ctranslate2)
**Compute Type:**
- `compute_type=int8_float16` for `device="cuda"`
- `compute_type=int8` for `device="cpu"`
# Sample Code - ctranslate2
#### Clone the repository to the working directory or wherever you wish to store the model artifacts. ####
```bash
git clone https://huggingface.co/gaudi/opus-mt-en-kwn-ctranslate2
```
#### Take the python code below and update the 'model_dir' variable to the location of the cloned repository. ####
```python
from ctranslate2 import Translator
import transformers
model_dir = "./opus-mt-en-kwn-ctranslate2" # Path to model directory.
translator = Translator(
model_path=model_dir,
device="cuda", # cpu, cuda, or auto.
inter_threads=1, # Maximum number of parallel translations.
intra_threads=4, # Number of OpenMP threads per translator.
compute_type="int8_float16", # int8 for cpu or int8_float16 for cuda.
)
tokenizer = transformers.AutoTokenizer.from_pretrained(model_dir)
source = tokenizer.convert_ids_to_tokens(tokenizer.encode("XXXXXX, XXX XX XXXXXX."))
results = translator.translate_batch([source])
target = results[0].hypotheses[0]
print(tokenizer.decode(tokenizer.convert_tokens_to_ids(target)))
```
# Sample Code - hf-hub-ctranslate2
**Derived From [michaelfeil](https://huggingface.co/michaelfeil):**
```python
from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
from transformers import AutoTokenizer
model_name = "gaudi/opus-mt-en-kwn-ctranslate2"
model = TranslatorCT2fromHfHub(
model_name_or_path=model_name,
device="cuda",
compute_type="int8_float16",
tokenizer=AutoTokenizer.from_pretrained(model_name)
)
outputs = model.generate(
text=["XXX XX XXX XXXXXXX XXXX?", "XX XX XXXX XX XXX!"],
)
print(outputs)
```
# License and other remarks:
License conditions are intended to be idential to [original huggingface repository](https://huggingface.co/Helsinki-NLP/opus-mt-en-kwn) by Helsinki-NLP.
| null |
Non_BioNLP
|
# Repository General Information
## Inspired by and derived from the work of [Helsinki-NLP](https://huggingface.co/Helsinki-NLP), [CTranslate2](https://github.com/OpenNMT/CTranslate2), and [michaelfeil](https://huggingface.co/michaelfeil)!
- Link to Original Model ([Helsinki-NLP](https://huggingface.co/Helsinki-NLP)): [Model Link](https://huggingface.co/Helsinki-NLP/opus-mt-en-kwn)
- This respository was based on the work of [CTranslate2](https://github.com/OpenNMT/CTranslate2).
- This repository was based on the work of [michaelfeil](https://huggingface.co/michaelfeil).
# What is CTranslate2?
[CTranslate2](https://opennmt.net/CTranslate2/) is a C++ and Python library for efficient inference with Transformer models.
CTranslate2 implements a custom runtime that applies many performance optimization techniques such as weights quantization, layers fusion, batch reordering, etc., to accelerate and reduce the memory usage of Transformer models on CPU and GPU.
CTranslate2 is one of the most performant ways of hosting translation models at scale. Current supported models include:
- Encoder-decoder models: Transformer base/big, M2M-100, NLLB, BART, mBART, Pegasus, T5, Whisper
- Decoder-only models: GPT-2, GPT-J, GPT-NeoX, OPT, BLOOM, MPT, Llama, Mistral, Gemma, CodeGen, GPTBigCode, Falcon
- Encoder-only models: BERT, DistilBERT, XLM-RoBERTa
The project is production-oriented and comes with backward compatibility guarantees, but it also includes experimental features related to model compression and inference acceleration.
# CTranslate2 Benchmarks
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings. Tested against `newstest2014` (En -> De) dataset.
The benchmark reports the number of target tokens generated per second (higher is better). The results are aggregated over multiple runs. See the benchmark scripts for more details and reproduce these numbers.
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings.
## CPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 147.3 | 2332MB | 27.90 |
| Marian 1.11.0 (int16) | 330.2 | 5901MB | 27.65 |
| Marian 1.11.0 (int8) | 355.8 | 4763MB | 27.27 |
| CTranslate2 3.6.0 (int16) | 596.1 | 660MB | 27.53 |
| CTranslate2 3.6.0 (int8) | 696.1 | 516MB | 27.65 |
## GPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max GPU Memory Usage | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 1022.9 | 4097MB | 2109MB | 27.90 |
| Marian 1.11.0 (float16) | 3962.4 | 3239MB | 1976MB | 27.94 |
| CTranslate2 3.6.0 (float16) | 9296.7 | 909MB | 814MB | 27.9 |
| CTranslate2 3.6.0 (int8 + float16) | 8362.7 | 813MB | 766MB | 27.9 |
`Executed with 4 threads on a c5.2xlarge Amazon EC2 instance equipped with an Intel(R) Xeon(R) Platinum 8275CL CPU.`
**Source to benchmark information can be found [here](https://github.com/OpenNMT/CTranslate2).**<br />
**Original model BLEU scores can be found [here](https://huggingface.co/Helsinki-NLP/opus-mt-en-kwn).**
## Internal Benchmarks
Internal testing on our end showed **inference times reduced by 6x-10x** on average compared the vanilla checkpoints using the *transformers* library. A **slight reduction on BLEU scores (~5%)** was also identified in comparison to the vanilla checkpoints with a few exceptions. This is likely due to several factors, one being the quantization applied. Further testing is needed from our end to better assess the reduction in translation quality. The command used to compile the vanilla checkpoint into a CTranslate2 model can be found below. Modifying this command can yield differing balances between inferencing performance and translation quality.
# CTranslate2 Installation
```bash
pip install hf-hub-ctranslate2>=1.0.0 ctranslate2>=3.13.0
```
### ct2-transformers-converter Command Used:
```bash
ct2-transformers-converter --model Helsinki-NLP/opus-mt-en-kwn --output_dir ./ctranslate2/opus-mt-en-kwn-ctranslate2 --force --copy_files README.md generation_config.json tokenizer_config.json vocab.json source.spm .gitattributes target.spm --quantization float16
```
# CTranslate2 Converted Checkpoint Information:
**Compatible With:**
- [ctranslate2](https://github.com/OpenNMT/CTranslate2)
- [hf-hub-ctranslate2](https://github.com/michaelfeil/hf-hub-ctranslate2)
**Compute Type:**
- `compute_type=int8_float16` for `device="cuda"`
- `compute_type=int8` for `device="cpu"`
# Sample Code - ctranslate2
#### Clone the repository to the working directory or wherever you wish to store the model artifacts. ####
```bash
git clone https://huggingface.co/gaudi/opus-mt-en-kwn-ctranslate2
```
#### Take the python code below and update the 'model_dir' variable to the location of the cloned repository. ####
```python
from ctranslate2 import Translator
import transformers
model_dir = "./opus-mt-en-kwn-ctranslate2" # Path to model directory.
translator = Translator(
model_path=model_dir,
device="cuda", # cpu, cuda, or auto.
inter_threads=1, # Maximum number of parallel translations.
intra_threads=4, # Number of OpenMP threads per translator.
compute_type="int8_float16", # int8 for cpu or int8_float16 for cuda.
)
tokenizer = transformers.AutoTokenizer.from_pretrained(model_dir)
source = tokenizer.convert_ids_to_tokens(tokenizer.encode("XXXXXX, XXX XX XXXXXX."))
results = translator.translate_batch([source])
target = results[0].hypotheses[0]
print(tokenizer.decode(tokenizer.convert_tokens_to_ids(target)))
```
# Sample Code - hf-hub-ctranslate2
**Derived From [michaelfeil](https://huggingface.co/michaelfeil):**
```python
from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
from transformers import AutoTokenizer
model_name = "gaudi/opus-mt-en-kwn-ctranslate2"
model = TranslatorCT2fromHfHub(
model_name_or_path=model_name,
device="cuda",
compute_type="int8_float16",
tokenizer=AutoTokenizer.from_pretrained(model_name)
)
outputs = model.generate(
text=["XXX XX XXX XXXXXXX XXXX?", "XX XX XXXX XX XXX!"],
)
print(outputs)
```
# License and other remarks:
License conditions are intended to be idential to [original huggingface repository](https://huggingface.co/Helsinki-NLP/opus-mt-en-kwn) by Helsinki-NLP.
|
{"license": "apache-2.0", "tags": ["ctranslate2", "translation"]}
|
task
|
[
"TRANSLATION"
] | 45,246 |
zohaib99k/QnA_model_training
|
zohaib99k
|
text-generation
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"license:other",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2023-07-11T04:12:35Z |
2023-07-13T13:04:41+00:00
| 11 | 0 |
---
license: other
---
LLaMA-13B converted to work with Transformers/HuggingFace. This is under a special license, please see the LICENSE file for details.
--
license: other
---
# LLaMA Model Card
## Model details
**Organization developing the model**
The FAIR team of Meta AI.
**Model date**
LLaMA was trained between December. 2022 and Feb. 2023.
**Model version**
This is version 1 of the model.
**Model type**
LLaMA is an auto-regressive language model, based on the transformer architecture. The model comes in different sizes: 7B, 13B, 33B and 65B parameters.
**Paper or resources for more information**
More information can be found in the paper “LLaMA, Open and Efficient Foundation Language Models”, available at https://research.facebook.com/publications/llama-open-and-efficient-foundation-language-models/.
**Citations details**
https://research.facebook.com/publications/llama-open-and-efficient-foundation-language-models/
**License**
Non-commercial bespoke license
**Where to send questions or comments about the model**
Questions and comments about LLaMA can be sent via the [GitHub repository](https://github.com/facebookresearch/llama) of the project , by opening an issue.
## Intended use
**Primary intended uses**
The primary use of LLaMA is research on large language models, including:
exploring potential applications such as question answering, natural language understanding or reading comprehension,
understanding capabilities and limitations of current language models, and developing techniques to improve those,
evaluating and mitigating biases, risks, toxic and harmful content generations, hallucinations.
**Primary intended users**
The primary intended users of the model are researchers in natural language processing, machine learning and artificial intelligence.
**Out-of-scope use cases**
LLaMA is a base, or foundational, model. As such, it should not be used on downstream applications without further risk evaluation and mitigation. In particular, our model has not been trained with human feedback, and can thus generate toxic or offensive content, incorrect information or generally unhelpful answers.
## Factors
**Relevant factors**
One of the most relevant factors for which model performance may vary is which language is used. Although we included 20 languages in the training data, most of our dataset is made of English text, and we thus expect the model to perform better for English than other languages. Relatedly, it has been shown in previous studies that performance might vary for different dialects, and we expect that it will be the case for our model.
**Evaluation factors**
As our model is trained on data from the Web, we expect that it reflects biases from this source. We thus evaluated on RAI datasets to measure biases exhibited by the model for gender, religion, race, sexual orientation, age, nationality, disability, physical appearance and socio-economic status. We also measure the toxicity of model generations, depending on the toxicity of the context used to prompt the model.
## Metrics
**Model performance measures**
We use the following measure to evaluate the model:
- Accuracy for common sense reasoning, reading comprehension, natural language understanding (MMLU), BIG-bench hard, WinoGender and CrowS-Pairs,
- Exact match for question answering,
- The toxicity score from Perspective API on RealToxicityPrompts.
**Decision thresholds**
Not applicable.
**Approaches to uncertainty and variability**
Due to the high computational requirements of training LLMs, we trained only one model of each size, and thus could not evaluate variability of pre-training.
## Evaluation datasets
The model was evaluated on the following benchmarks: BoolQ, PIQA, SIQA, HellaSwag, WinoGrande, ARC, OpenBookQA, NaturalQuestions, TriviaQA, RACE, MMLU, BIG-bench hard, GSM8k, RealToxicityPrompts, WinoGender, CrowS-Pairs.
## Training dataset
The model was trained using the following source of data: CCNet [67%], C4 [15%], GitHub [4.5%], Wikipedia [4.5%], Books [4.5%], ArXiv [2.5%], Stack Exchange[2%]. The Wikipedia and Books domains include data in the following languages: bg, ca, cs, da, de, en, es, fr, hr, hu, it, nl, pl, pt, ro, ru, sl, sr, sv, uk. See the paper for more details about the training set and corresponding preprocessing.
## Quantitative analysis
Hyperparameters for the model architecture
<table>
<thead>
<tr>
<th >LLaMA</th> <th colspan=6>Model hyper parameters </th>
</tr>
<tr>
<th>Number of parameters</th><th>dimension</th><th>n heads</th><th>n layers</th><th>Learn rate</th><th>Batch size</th><th>n tokens</th>
</tr>
</thead>
<tbody>
<tr>
<th>7B</th> <th>4096</th> <th>32</th> <th>32</th> <th>3.0E-04</th><th>4M</th><th>1T
</tr>
<tr>
<th>13B</th><th>5120</th><th>40</th><th>40</th><th>3.0E-04</th><th>4M</th><th>1T
</tr>
<tr>
<th>33B</th><th>6656</th><th>52</th><th>60</th><th>1.5.E-04</th><th>4M</th><th>1.4T
</tr>
<tr>
<th>65B</th><th>8192</th><th>64</th><th>80</th><th>1.5.E-04</th><th>4M</th><th>1.4T
</tr>
</tbody>
</table>
*Table 1 - Summary of LLama Model Hyperparameters*
We present our results on eight standard common sense reasoning benchmarks in the table below.
<table>
<thead>
<tr>
<th>LLaMA</th> <th colspan=9>Reasoning tasks </th>
</tr>
<tr>
<th>Number of parameters</th> <th>BoolQ</th><th>PIQA</th><th>SIQA</th><th>HellaSwag</th><th>WinoGrande</th><th>ARC-e</th><th>ARC-c</th><th>OBQA</th><th>COPA</th>
</tr>
</thead>
<tbody>
<tr>
<th>7B</th><th>76.5</th><th>79.8</th><th>48.9</th><th>76.1</th><th>70.1</th><th>76.7</th><th>47.6</th><th>57.2</th><th>93
</th>
<tr><th>13B</th><th>78.1</th><th>80.1</th><th>50.4</th><th>79.2</th><th>73</th><th>78.1</th><th>52.7</th><th>56.4</th><th>94
</th>
<tr><th>33B</th><th>83.1</th><th>82.3</th><th>50.4</th><th>82.8</th><th>76</th><th>81.4</th><th>57.8</th><th>58.6</th><th>92
</th>
<tr><th>65B</th><th>85.3</th><th>82.8</th><th>52.3</th><th>84.2</th><th>77</th><th>81.5</th><th>56</th><th>60.2</th><th>94</th></tr>
</tbody>
</table>
*Table 2 - Summary of LLama Model Performance on Reasoning tasks*
We present our results on bias in the table below. Note that lower value is better indicating lower bias.
| No | Category | FAIR LLM |
| --- | -------------------- | -------- |
| 1 | Gender | 70.6 |
| 2 | Religion | 79 |
| 3 | Race/Color | 57 |
| 4 | Sexual orientation | 81 |
| 5 | Age | 70.1 |
| 6 | Nationality | 64.2 |
| 7 | Disability | 66.7 |
| 8 | Physical appearance | 77.8 |
| 9 | Socioeconomic status | 71.5 |
| | LLaMA Average | 66.6 |
*Table 3 - Summary bias of our model output*
## Ethical considerations
**Data**
The data used to train the model is collected from various sources, mostly from the Web. As such, it contains offensive, harmful and biased content. We thus expect the model to exhibit such biases from the training data.
**Human life**
The model is not intended to inform decisions about matters central to human life, and should not be used in such a way.
**Mitigations**
We filtered the data from the Web based on its proximity to Wikipedia text and references. For this, we used a Kneser-Ney language model and a fastText linear classifier.
**Risks and harms**
Risks and harms of large language models include the generation of harmful, offensive or biased content. These models are often prone to generating incorrect information, sometimes referred to as hallucinations. We do not expect our model to be an exception in this regard.
**Use cases**
LLaMA is a foundational model, and as such, it should not be used for downstream applications without further investigation and mitigations of risks. These risks and potential fraught use cases include, but are not limited to: generation of misinformation and generation of harmful, biased or offensive content.
| null |
Non_BioNLP
|
LLaMA-13B converted to work with Transformers/HuggingFace. This is under a special license, please see the LICENSE file for details.
--
license: other
---
# LLaMA Model Card
## Model details
**Organization developing the model**
The FAIR team of Meta AI.
**Model date**
LLaMA was trained between December. 2022 and Feb. 2023.
**Model version**
This is version 1 of the model.
**Model type**
LLaMA is an auto-regressive language model, based on the transformer architecture. The model comes in different sizes: 7B, 13B, 33B and 65B parameters.
**Paper or resources for more information**
More information can be found in the paper “LLaMA, Open and Efficient Foundation Language Models”, available at https://research.facebook.com/publications/llama-open-and-efficient-foundation-language-models/.
**Citations details**
https://research.facebook.com/publications/llama-open-and-efficient-foundation-language-models/
**License**
Non-commercial bespoke license
**Where to send questions or comments about the model**
Questions and comments about LLaMA can be sent via the [GitHub repository](https://github.com/facebookresearch/llama) of the project , by opening an issue.
## Intended use
**Primary intended uses**
The primary use of LLaMA is research on large language models, including:
exploring potential applications such as question answering, natural language understanding or reading comprehension,
understanding capabilities and limitations of current language models, and developing techniques to improve those,
evaluating and mitigating biases, risks, toxic and harmful content generations, hallucinations.
**Primary intended users**
The primary intended users of the model are researchers in natural language processing, machine learning and artificial intelligence.
**Out-of-scope use cases**
LLaMA is a base, or foundational, model. As such, it should not be used on downstream applications without further risk evaluation and mitigation. In particular, our model has not been trained with human feedback, and can thus generate toxic or offensive content, incorrect information or generally unhelpful answers.
## Factors
**Relevant factors**
One of the most relevant factors for which model performance may vary is which language is used. Although we included 20 languages in the training data, most of our dataset is made of English text, and we thus expect the model to perform better for English than other languages. Relatedly, it has been shown in previous studies that performance might vary for different dialects, and we expect that it will be the case for our model.
**Evaluation factors**
As our model is trained on data from the Web, we expect that it reflects biases from this source. We thus evaluated on RAI datasets to measure biases exhibited by the model for gender, religion, race, sexual orientation, age, nationality, disability, physical appearance and socio-economic status. We also measure the toxicity of model generations, depending on the toxicity of the context used to prompt the model.
## Metrics
**Model performance measures**
We use the following measure to evaluate the model:
- Accuracy for common sense reasoning, reading comprehension, natural language understanding (MMLU), BIG-bench hard, WinoGender and CrowS-Pairs,
- Exact match for question answering,
- The toxicity score from Perspective API on RealToxicityPrompts.
**Decision thresholds**
Not applicable.
**Approaches to uncertainty and variability**
Due to the high computational requirements of training LLMs, we trained only one model of each size, and thus could not evaluate variability of pre-training.
## Evaluation datasets
The model was evaluated on the following benchmarks: BoolQ, PIQA, SIQA, HellaSwag, WinoGrande, ARC, OpenBookQA, NaturalQuestions, TriviaQA, RACE, MMLU, BIG-bench hard, GSM8k, RealToxicityPrompts, WinoGender, CrowS-Pairs.
## Training dataset
The model was trained using the following source of data: CCNet [67%], C4 [15%], GitHub [4.5%], Wikipedia [4.5%], Books [4.5%], ArXiv [2.5%], Stack Exchange[2%]. The Wikipedia and Books domains include data in the following languages: bg, ca, cs, da, de, en, es, fr, hr, hu, it, nl, pl, pt, ro, ru, sl, sr, sv, uk. See the paper for more details about the training set and corresponding preprocessing.
## Quantitative analysis
Hyperparameters for the model architecture
<table>
<thead>
<tr>
<th >LLaMA</th> <th colspan=6>Model hyper parameters </th>
</tr>
<tr>
<th>Number of parameters</th><th>dimension</th><th>n heads</th><th>n layers</th><th>Learn rate</th><th>Batch size</th><th>n tokens</th>
</tr>
</thead>
<tbody>
<tr>
<th>7B</th> <th>4096</th> <th>32</th> <th>32</th> <th>3.0E-04</th><th>4M</th><th>1T
</tr>
<tr>
<th>13B</th><th>5120</th><th>40</th><th>40</th><th>3.0E-04</th><th>4M</th><th>1T
</tr>
<tr>
<th>33B</th><th>6656</th><th>52</th><th>60</th><th>1.5.E-04</th><th>4M</th><th>1.4T
</tr>
<tr>
<th>65B</th><th>8192</th><th>64</th><th>80</th><th>1.5.E-04</th><th>4M</th><th>1.4T
</tr>
</tbody>
</table>
*Table 1 - Summary of LLama Model Hyperparameters*
We present our results on eight standard common sense reasoning benchmarks in the table below.
<table>
<thead>
<tr>
<th>LLaMA</th> <th colspan=9>Reasoning tasks </th>
</tr>
<tr>
<th>Number of parameters</th> <th>BoolQ</th><th>PIQA</th><th>SIQA</th><th>HellaSwag</th><th>WinoGrande</th><th>ARC-e</th><th>ARC-c</th><th>OBQA</th><th>COPA</th>
</tr>
</thead>
<tbody>
<tr>
<th>7B</th><th>76.5</th><th>79.8</th><th>48.9</th><th>76.1</th><th>70.1</th><th>76.7</th><th>47.6</th><th>57.2</th><th>93
</th>
<tr><th>13B</th><th>78.1</th><th>80.1</th><th>50.4</th><th>79.2</th><th>73</th><th>78.1</th><th>52.7</th><th>56.4</th><th>94
</th>
<tr><th>33B</th><th>83.1</th><th>82.3</th><th>50.4</th><th>82.8</th><th>76</th><th>81.4</th><th>57.8</th><th>58.6</th><th>92
</th>
<tr><th>65B</th><th>85.3</th><th>82.8</th><th>52.3</th><th>84.2</th><th>77</th><th>81.5</th><th>56</th><th>60.2</th><th>94</th></tr>
</tbody>
</table>
*Table 2 - Summary of LLama Model Performance on Reasoning tasks*
We present our results on bias in the table below. Note that lower value is better indicating lower bias.
| No | Category | FAIR LLM |
| --- | -------------------- | -------- |
| 1 | Gender | 70.6 |
| 2 | Religion | 79 |
| 3 | Race/Color | 57 |
| 4 | Sexual orientation | 81 |
| 5 | Age | 70.1 |
| 6 | Nationality | 64.2 |
| 7 | Disability | 66.7 |
| 8 | Physical appearance | 77.8 |
| 9 | Socioeconomic status | 71.5 |
| | LLaMA Average | 66.6 |
*Table 3 - Summary bias of our model output*
## Ethical considerations
**Data**
The data used to train the model is collected from various sources, mostly from the Web. As such, it contains offensive, harmful and biased content. We thus expect the model to exhibit such biases from the training data.
**Human life**
The model is not intended to inform decisions about matters central to human life, and should not be used in such a way.
**Mitigations**
We filtered the data from the Web based on its proximity to Wikipedia text and references. For this, we used a Kneser-Ney language model and a fastText linear classifier.
**Risks and harms**
Risks and harms of large language models include the generation of harmful, offensive or biased content. These models are often prone to generating incorrect information, sometimes referred to as hallucinations. We do not expect our model to be an exception in this regard.
**Use cases**
LLaMA is a foundational model, and as such, it should not be used for downstream applications without further investigation and mitigations of risks. These risks and potential fraught use cases include, but are not limited to: generation of misinformation and generation of harmful, biased or offensive content.
|
{"license": "other"}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,247 |
gavulsim/distilbert_finetuned_yahoo_answers_topics
|
gavulsim
|
text-classification
|
[
"transformers",
"pytorch",
"distilbert",
"text-classification",
"generated_from_trainer",
"dataset:yahoo_answers_topics",
"base_model:distilbert/distilbert-base-uncased",
"base_model:finetune:distilbert/distilbert-base-uncased",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-10-27T08:19:50Z |
2023-10-27T09:20:49+00:00
| 110 | 0 |
---
base_model: distilbert-base-uncased
datasets:
- yahoo_answers_topics
license: apache-2.0
metrics:
- accuracy
tags:
- generated_from_trainer
model-index:
- name: deberta_finetuned_yahoo_answers_topics
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: yahoo_answers_topics
type: yahoo_answers_topics
config: yahoo_answers_topics
split: test
args: yahoo_answers_topics
metrics:
- type: accuracy
value: 0.71195
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# deberta_finetuned_yahoo_answers_topics
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the yahoo_answers_topics dataset.
It achieves the following results on the evaluation set:
- Loss: 0.9096
- Accuracy: 0.7119
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 30000
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 1.1025 | 0.03 | 5000 | 1.0702 | 0.6717 |
| 1.0132 | 0.06 | 10000 | 0.9976 | 0.6834 |
| 0.8688 | 0.09 | 15000 | 0.9770 | 0.6961 |
| 0.9964 | 0.11 | 20000 | 0.9356 | 0.7020 |
| 0.9338 | 0.14 | 25000 | 0.9259 | 0.7090 |
| 0.9059 | 0.17 | 30000 | 0.9096 | 0.7119 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# deberta_finetuned_yahoo_answers_topics
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the yahoo_answers_topics dataset.
It achieves the following results on the evaluation set:
- Loss: 0.9096
- Accuracy: 0.7119
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 30000
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 1.1025 | 0.03 | 5000 | 1.0702 | 0.6717 |
| 1.0132 | 0.06 | 10000 | 0.9976 | 0.6834 |
| 0.8688 | 0.09 | 15000 | 0.9770 | 0.6961 |
| 0.9964 | 0.11 | 20000 | 0.9356 | 0.7020 |
| 0.9338 | 0.14 | 25000 | 0.9259 | 0.7090 |
| 0.9059 | 0.17 | 30000 | 0.9096 | 0.7119 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
|
{"base_model": "distilbert-base-uncased", "datasets": ["yahoo_answers_topics"], "license": "apache-2.0", "metrics": ["accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "deberta_finetuned_yahoo_answers_topics", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "yahoo_answers_topics", "type": "yahoo_answers_topics", "config": "yahoo_answers_topics", "split": "test", "args": "yahoo_answers_topics"}, "metrics": [{"type": "accuracy", "value": 0.71195, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,248 |
bullerwins/c4ai-command-r-plus-08-2024-exl2_4.0bpw
|
bullerwins
|
text-generation
|
[
"transformers",
"safetensors",
"cohere",
"text-generation",
"conversational",
"en",
"fr",
"de",
"es",
"it",
"pt",
"ja",
"ko",
"zh",
"ar",
"base_model:CohereForAI/c4ai-command-r-plus-08-2024",
"base_model:quantized:CohereForAI/c4ai-command-r-plus-08-2024",
"license:cc-by-nc-4.0",
"autotrain_compatible",
"text-generation-inference",
"4-bit",
"exl2",
"region:us"
] | 2024-09-01T06:57:42Z |
2024-09-01T08:02:27+00:00
| 4 | 0 |
---
base_model: CohereForAI/c4ai-command-r-plus-08-2024
language:
- en
- fr
- de
- es
- it
- pt
- ja
- ko
- zh
- ar
library_name: transformers
license: cc-by-nc-4.0
inference: false
extra_gated_prompt: By submitting this form, you agree to the [License Agreement](https://cohere.com/c4ai-cc-by-nc-license) and
acknowledge that the information you provide will be collected, used, and shared
in accordance with Cohere’s [Privacy Policy]( https://cohere.com/privacy).
extra_gated_fields:
Name: text
Affiliation: text
Country:
type: select
options:
- Aruba
- Afghanistan
- Angola
- Anguilla
- Åland Islands
- Albania
- Andorra
- United Arab Emirates
- Argentina
- Armenia
- American Samoa
- Antarctica
- French Southern Territories
- Antigua and Barbuda
- Australia
- Austria
- Azerbaijan
- Burundi
- Belgium
- Benin
- Bonaire Sint Eustatius and Saba
- Burkina Faso
- Bangladesh
- Bulgaria
- Bahrain
- Bahamas
- Bosnia and Herzegovina
- Saint Barthélemy
- Belarus
- Belize
- Bermuda
- Plurinational State of Bolivia
- Brazil
- Barbados
- Brunei-Darussalam
- Bhutan
- Bouvet-Island
- Botswana
- Central African Republic
- Canada
- Cocos (Keeling) Islands
- Switzerland
- Chile
- China
- Côte-dIvoire
- Cameroon
- Democratic Republic of the Congo
- Cook Islands
- Colombia
- Comoros
- Cabo Verde
- Costa Rica
- Cuba
- Curaçao
- Christmas Island
- Cayman Islands
- Cyprus
- Czechia
- Germany
- Djibouti
- Dominica
- Denmark
- Dominican Republic
- Algeria
- Ecuador
- Egypt
- Eritrea
- Western Sahara
- Spain
- Estonia
- Ethiopia
- Finland
- Fiji
- Falkland Islands (Malvinas)
- France
- Faroe Islands
- Federated States of Micronesia
- Gabon
- United Kingdom
- Georgia
- Guernsey
- Ghana
- Gibraltar
- Guinea
- Guadeloupe
- Gambia
- Guinea Bissau
- Equatorial Guinea
- Greece
- Grenada
- Greenland
- Guatemala
- French Guiana
- Guam
- Guyana
- Hong Kong
- Heard Island and McDonald Islands
- Honduras
- Croatia
- Haiti
- Hungary
- Indonesia
- Isle of Man
- India
- British Indian Ocean Territory
- Ireland
- Islamic Republic of Iran
- Iraq
- Iceland
- Israel
- Italy
- Jamaica
- Jersey
- Jordan
- Japan
- Kazakhstan
- Kenya
- Kyrgyzstan
- Cambodia
- Kiribati
- Saint-Kitts-and-Nevis
- South Korea
- Kuwait
- Lao-Peoples-Democratic-Republic
- Lebanon
- Liberia
- Libya
- Saint-Lucia
- Liechtenstein
- Sri Lanka
- Lesotho
- Lithuania
- Luxembourg
- Latvia
- Macao
- Saint Martin (French-part)
- Morocco
- Monaco
- Republic of Moldova
- Madagascar
- Maldives
- Mexico
- Marshall Islands
- North Macedonia
- Mali
- Malta
- Myanmar
- Montenegro
- Mongolia
- Northern Mariana Islands
- Mozambique
- Mauritania
- Montserrat
- Martinique
- Mauritius
- Malawi
- Malaysia
- Mayotte
- Namibia
- New Caledonia
- Niger
- Norfolk Island
- Nigeria
- Nicaragua
- Niue
- Netherlands
- Norway
- Nepal
- Nauru
- New Zealand
- Oman
- Pakistan
- Panama
- Pitcairn
- Peru
- Philippines
- Palau
- Papua New Guinea
- Poland
- Puerto Rico
- North Korea
- Portugal
- Paraguay
- State of Palestine
- French Polynesia
- Qatar
- Réunion
- Romania
- Russia
- Rwanda
- Saudi Arabia
- Sudan
- Senegal
- Singapore
- South Georgia and the South Sandwich Islands
- Saint Helena Ascension and Tristan da Cunha
- Svalbard and Jan Mayen
- Solomon Islands
- Sierra Leone
- El Salvador
- San Marino
- Somalia
- Saint Pierre and Miquelon
- Serbia
- South Sudan
- Sao Tome and Principe
- Suriname
- Slovakia
- Slovenia
- Sweden
- Eswatini
- Sint Maarten (Dutch-part)
- Seychelles
- Syrian Arab Republic
- Turks and Caicos Islands
- Chad
- Togo
- Thailand
- Tajikistan
- Tokelau
- Turkmenistan
- Timor Leste
- Tonga
- Trinidad and Tobago
- Tunisia
- Turkey
- Tuvalu
- Taiwan
- United Republic of Tanzania
- Uganda
- Ukraine
- United States Minor Outlying Islands
- Uruguay
- United-States
- Uzbekistan
- Holy See (Vatican City State)
- Saint Vincent and the Grenadines
- Bolivarian Republic of Venezuela
- Virgin Islands British
- Virgin Islands U.S.
- VietNam
- Vanuatu
- Wallis and Futuna
- Samoa
- Yemen
- South Africa
- Zambia
- Zimbabwe
Receive email updates on C4AI and Cohere research, events, products and services?:
type: select
options:
- true
- false
I agree to use this model for non-commercial use ONLY: checkbox
---
EXL2 quantized model using [exllamav2 0.2.0](https://github.com/turboderp/exllamav2)
Original model [CohereForAI/c4ai-command-r-plus-08-2024](https://huggingface.co/CohereForAI/c4ai-command-r-plus-08-2024)
# Model Card for C4AI Command R+ 08-2024
## Model Summary
C4AI Command R+ 08-2024 is an open weights research release of a 104B billion parameter model with highly advanced capabilities, this includes Retrieval Augmented Generation (RAG) and tool use to automate sophisticated tasks. The tool use in this model generation enables multi-step tool use which allows the model to combine multiple tools over multiple steps to accomplish difficult tasks. C4AI Command R+ 08-2024 is a multilingual model trained on 23 languages and evaluated in 10 languages. Command R+ 08-2024 is optimized for a variety of use cases including reasoning, summarization, and question answering.
C4AI Command R+ 08-2024 is part of a family of open weight releases from Cohere For AI and Cohere. Our smaller companion model is [C4AI Command R 08-2024](https://huggingface.co/CohereForAI/c4ai-command-r-08-2024).
- Point of Contact: Cohere For AI: [cohere.for.ai](https://cohere.for.ai/)
- License: [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license), requires also adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy)
- Model: c4ai-command-r-plus-08-2024
- Model Size: 104 billion parameters
- Context length: 128K
**Try C4AI Command R+**
You can try out C4AI Command R+ before downloading the weights in our hosted [Hugging Face Space](https://huggingface.co/spaces/CohereForAI/c4ai-command?model=command-r-plus-08-2024).
**Usage**
Please use `transformers` version 4.39.1 or higher
```python
# pip install 'transformers>=4.39.1'
from transformers import AutoTokenizer, AutoModelForCausalLM
model_id = "CohereForAI/c4ai-command-r-plus-08-2024"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)
# Format message with the command-r-plus-08-2024 chat template
messages = [{"role": "user", "content": "Hello, how are you?"}]
input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
## <BOS_TOKEN><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Hello, how are you?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
gen_tokens = model.generate(
input_ids,
max_new_tokens=100,
do_sample=True,
temperature=0.3,
)
gen_text = tokenizer.decode(gen_tokens[0])
print(gen_text)
```
## Model Details
**Input**: Models input text only.
**Output**: Models generate text only.
**Model Architecture**: This is an auto-regressive language model that uses an optimized transformer architecture. After pretraining, this model uses supervised fine-tuning (SFT) and preference training to align model behavior to human preferences for helpfulness and safety. We use grouped query attention (GQA) to improve inference speed.
**Languages covered**: The model has been trained on 23 languages (English, French, Spanish, Italian, German, Portuguese, Japanese, Korean, Arabic, Simplified Chinese, Russian, Polish, Turkish, Vietnamese, Dutch, Czech, Indonesian, Ukrainian, Romanian, Greek, Hindi, Hebrew, and Persian) and evaluated on 10 languages (English, French, Spanish, Italian, German, Portuguese, Japanese, Korean, Arabic, Simplified Chinese).
**Context length**: Command R+ 08-2024 supports a context length of 128K.
### Tool use & Agent capabilities:
Command R+ 08-2024 has been specifically trained with conversational tool use capabilities. These have been trained into the model via a mixture of supervised fine-tuning and preference fine-tuning, using a specific prompt template. Deviating from this prompt template will likely reduce performance, but we encourage experimentation.
Command R+ 08-2024’s tool use functionality takes a conversation as input (with an optional user-system preamble), along with a list of available tools. The model will then generate a json-formatted list of actions to execute on a subset of those tools. Command R+ 08-2024 may use one of its supplied tools more than once.
The model has been trained to recognise a special `directly_answer` tool, which it uses to indicate that it doesn’t want to use any of its other tools. The ability to abstain from calling a specific tool can be useful in a range of situations, such as greeting a user, or asking clarifying questions. We recommend including the `directly_answer` tool, but it can be removed or renamed if required.
Comprehensive documentation for working with Command R+ 08-2024's tool use prompt template can be found [here](https://docs.cohere.com/docs/prompting-command-r).
Command R+ 08-2024 also supports Hugging Face's [tool use API](https://huggingface.co/docs/transformers/main/en/chat_templating#advanced-tool-use--function-calling).
The code snippets below show minimal working examples on how to render a prompt.
<details>
<summary><b>Usage: Rendering Tool Use Prompts [CLICK TO EXPAND]</b> </summary>
```python
from transformers import AutoTokenizer
model_id = "CohereForAI/c4ai-command-r-plus-08-2024"
tokenizer = AutoTokenizer.from_pretrained(model_id)
# define conversation input:
conversation = [
{"role": "user", "content": "Whats the biggest penguin in the world?"}
]
# Define tools available for the model to use:
tools = [
{
"name": "internet_search",
"description": "Returns a list of relevant document snippets for a textual query retrieved from the internet",
"parameter_definitions": {
"query": {
"description": "Query to search the internet with",
"type": 'str',
"required": True
}
}
},
{
'name': "directly_answer",
"description": "Calls a standard (un-augmented) AI chatbot to generate a response given the conversation history",
'parameter_definitions': {}
}
]
# render the tool use prompt as a string:
tool_use_prompt = tokenizer.apply_tool_use_template(
conversation,
tools=tools,
tokenize=False,
add_generation_prompt=True,
)
print(tool_use_prompt)
```
</details>
<details>
<summary><b>Usage: Rendering prompts with the Tool Use API [CLICK TO EXPAND]</b> </summary>
```python
from transformers import AutoTokenizer
model_id = "CohereForAI/c4ai-command-r-plus-08-2024"
tokenizer = AutoTokenizer.from_pretrained(model_id)
# define conversation input:
conversation = [
{"role": "user", "content": "Whats the biggest penguin in the world?"}
]
# Define tools available for the model to use
# Type hints and docstrings from Python functions are automatically extracted
def internet_search(query: str):
"""
Returns a list of relevant document snippets for a textual query retrieved from the internet
Args:
query: Query to search the internet with
"""
pass
def directly_answer():
"""
Calls a standard (un-augmented) AI chatbot to generate a response given the conversation history
"""
pass
tools = [internet_search, directly_answer]
# render the tool use prompt as a string:
tool_use_prompt = tokenizer.apply_chat_template(
conversation,
tools=tools,
tokenize=False,
add_generation_prompt=True,
)
print(tool_use_prompt)
```
</details>
<details>
<summary><b>Example Rendered Tool Use Prompt [CLICK TO EXPAND]</b></summary>
````
<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># Safety Preamble
The instructions in this section override those in the task description and style guide sections. Don't answer questions that are harmful or immoral.
# System Preamble
## Basic Rules
You are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user's requests, you cite your sources in your answers, according to those instructions.
# User Preamble
## Task and Context
You help people answer their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You should focus on serving the user's needs as best you can, which will be wide-ranging.
## Style Guide
Unless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.
## Available Tools
Here is a list of tools that you have available to you:
```python
def internet_search(query: str) -> List[Dict]:
"""Returns a list of relevant document snippets for a textual query retrieved from the internet
Args:
query (str): Query to search the internet with
"""
pass
```
```python
def directly_answer() -> List[Dict]:
"""Calls a standard (un-augmented) AI chatbot to generate a response given the conversation history
"""
pass
```<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Whats the biggest penguin in the world?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>Write 'Action:' followed by a json-formatted list of actions that you want to perform in order to produce a good response to the user's last input. You can use any of the supplied tools any number of times, but you should aim to execute the minimum number of necessary actions for the input. You should use the `directly-answer` tool if calling the other tools is unnecessary. The list of actions you want to call should be formatted as a list of json objects, for example:
```json
[
{
"tool_name": title of the tool in the specification,
"parameters": a dict of parameters to input into the tool as they are defined in the specs, or {} if it takes no parameters
}
]```<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
````
</details>
<details>
<summary><b>Example Rendered Tool Use Completion [CLICK TO EXPAND]</b></summary>
````
Action: ```json
[
{
"tool_name": "internet_search",
"parameters": {
"query": "biggest penguin in the world"
}
}
]
```
````
</details>
### Grounded Generation and RAG Capabilities:
Command R+ 08-2024 has been specifically trained with grounded generation capabilities. This means that it can generate responses based on a list of supplied document snippets, and it will include grounding spans (citations) in its response indicating the source of the information. This can be used to enable behaviors such as grounded summarization and the final step of Retrieval Augmented Generation (RAG). This behavior has been trained into the model via a mixture of supervised fine-tuning and preference fine-tuning, using a specific prompt template. Deviating from this prompt template may reduce performance, but we encourage experimentation.
Command R+ 08-2024’s grounded generation behavior takes a conversation as input (with an optional user-supplied system preamble, indicating task, context and desired output style), along with a list of retrieved document snippets. The document snippets should be chunks, rather than long documents, typically around 100-400 words per chunk. Document snippets consist of key-value pairs. The keys should be short descriptive strings, the values can be text or semi-structured.
By default, Command R+ 08-2024 will generate grounded responses by first predicting which documents are relevant, then predicting which ones it will cite, then generating an answer. Finally, it will then insert grounding spans into the answer. See below for an example. This is referred to as `accurate` grounded generation.
The model is trained with a number of other answering modes, which can be selected by prompt changes. A `fast` citation mode is supported in the tokenizer, which will directly generate an answer with grounding spans in it, without first writing the answer out in full. This sacrifices some grounding accuracy in favor of generating fewer tokens.
Comprehensive documentation for working with Command R+ 08-2024's grounded generation prompt template can be found [here](https://docs.cohere.com/docs/prompting-command-r).
The code snippet below shows a minimal working example on how to render a prompt.
<details>
<summary> <b>Usage: Rendering Grounded Generation prompts [CLICK TO EXPAND]</b> </summary>
````python
from transformers import AutoTokenizer
model_id = "CohereForAI/c4ai-command-r-plus-08-2024"
tokenizer = AutoTokenizer.from_pretrained(model_id)
# define conversation input:
conversation = [
{"role": "user", "content": "Whats the biggest penguin in the world?"}
]
# define documents to ground on:
documents = [
{ "title": "Tall penguins", "text": "Emperor penguins are the tallest growing up to 122 cm in height." },
{ "title": "Penguin habitats", "text": "Emperor penguins only live in Antarctica."}
]
# render the tool use prompt as a string:
grounded_generation_prompt = tokenizer.apply_grounded_generation_template(
conversation,
documents=documents,
citation_mode="accurate", # or "fast"
tokenize=False,
add_generation_prompt=True,
)
print(grounded_generation_prompt)
````
</details>
<details>
<summary><b>Example Rendered Grounded Generation Prompt [CLICK TO EXPAND]</b></summary>
````
<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># Safety Preamble
The instructions in this section override those in the task description and style guide sections. Don't answer questions that are harmful or immoral.
# System Preamble
## Basic Rules
You are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user's requests, you cite your sources in your answers, according to those instructions.
# User Preamble
## Task and Context
You help people answer their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You should focus on serving the user's needs as best you can, which will be wide-ranging.
## Style Guide
Unless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Whats the biggest penguin in the world?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|><results>
Document: 0
title: Tall penguins
text: Emperor penguins are the tallest growing up to 122 cm in height.
Document: 1
title: Penguin habitats
text: Emperor penguins only live in Antarctica.
</results><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>Carefully perform the following instructions, in order, starting each with a new line.
Firstly, Decide which of the retrieved documents are relevant to the user's last input by writing 'Relevant Documents:' followed by comma-separated list of document numbers. If none are relevant, you should instead write 'None'.
Secondly, Decide which of the retrieved documents contain facts that should be cited in a good answer to the user's last input by writing 'Cited Documents:' followed a comma-separated list of document numbers. If you dont want to cite any of them, you should instead write 'None'.
Thirdly, Write 'Answer:' followed by a response to the user's last input in high quality natural english. Use the retrieved documents to help you. Do not insert any citations or grounding markup.
Finally, Write 'Grounded answer:' followed by a response to the user's last input in high quality natural english. Use the symbols <co: doc> and </co: doc> to indicate when a fact comes from a document in the search result, e.g <co: 0>my fact</co: 0> for a fact from document 0.<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
````
</details>
<details>
<summary><b>Example Rendered Grounded Generation Completion [CLICK TO EXPAND]</b></summary>
````
Relevant Documents: 0,1
Cited Documents: 0,1
Answer: The Emperor Penguin is the tallest or biggest penguin in the world. It is a bird that lives only in Antarctica and grows to a height of around 122 centimetres.
Grounded answer: The <co: 0>Emperor Penguin</co: 0> is the <co: 0>tallest</co: 0> or biggest penguin in the world. It is a bird that <co: 1>lives only in Antarctica</co: 1> and <co: 0>grows to a height of around 122 centimetres.</co: 0>
````
</details>
### Code Capabilities:
Command R+ 08-2024 has been optimized to interact with your code, by requesting code snippets, code explanations, or code rewrites. It might not perform well out-of-the-box for pure code completion. For better performance, we also recommend using a low temperature (and even greedy decoding) for code-generation related instructions.
### Model Card Contact
For errors or additional questions about details in this model card, contact [[email protected]](mailto:[email protected]).
### Terms of Use:
We hope that the release of this model will make community-based research efforts more accessible, by releasing the weights of a highly performant 104 billion parameter model to researchers all over the world. This model is governed by a [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license) License with an acceptable use addendum, and also requires adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).
### Try Chat:
You can try Command R+ 08-2024 chat in the playground [here](https://dashboard.cohere.com/playground/chat). You can also use it in our dedicated Hugging Face Space [here](https://huggingface.co/spaces/CohereForAI/c4ai-command?model=command-r-plus-08-2024).
| null |
TBD
|
EXL2 quantized model using [exllamav2 0.2.0](https://github.com/turboderp/exllamav2)
Original model [CohereForAI/c4ai-command-r-plus-08-2024](https://huggingface.co/CohereForAI/c4ai-command-r-plus-08-2024)
# Model Card for C4AI Command R+ 08-2024
## Model Summary
C4AI Command R+ 08-2024 is an open weights research release of a 104B billion parameter model with highly advanced capabilities, this includes Retrieval Augmented Generation (RAG) and tool use to automate sophisticated tasks. The tool use in this model generation enables multi-step tool use which allows the model to combine multiple tools over multiple steps to accomplish difficult tasks. C4AI Command R+ 08-2024 is a multilingual model trained on 23 languages and evaluated in 10 languages. Command R+ 08-2024 is optimized for a variety of use cases including reasoning, summarization, and question answering.
C4AI Command R+ 08-2024 is part of a family of open weight releases from Cohere For AI and Cohere. Our smaller companion model is [C4AI Command R 08-2024](https://huggingface.co/CohereForAI/c4ai-command-r-08-2024).
- Point of Contact: Cohere For AI: [cohere.for.ai](https://cohere.for.ai/)
- License: [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license), requires also adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy)
- Model: c4ai-command-r-plus-08-2024
- Model Size: 104 billion parameters
- Context length: 128K
**Try C4AI Command R+**
You can try out C4AI Command R+ before downloading the weights in our hosted [Hugging Face Space](https://huggingface.co/spaces/CohereForAI/c4ai-command?model=command-r-plus-08-2024).
**Usage**
Please use `transformers` version 4.39.1 or higher
```python
# pip install 'transformers>=4.39.1'
from transformers import AutoTokenizer, AutoModelForCausalLM
model_id = "CohereForAI/c4ai-command-r-plus-08-2024"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)
# Format message with the command-r-plus-08-2024 chat template
messages = [{"role": "user", "content": "Hello, how are you?"}]
input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
## <BOS_TOKEN><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Hello, how are you?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
gen_tokens = model.generate(
input_ids,
max_new_tokens=100,
do_sample=True,
temperature=0.3,
)
gen_text = tokenizer.decode(gen_tokens[0])
print(gen_text)
```
## Model Details
**Input**: Models input text only.
**Output**: Models generate text only.
**Model Architecture**: This is an auto-regressive language model that uses an optimized transformer architecture. After pretraining, this model uses supervised fine-tuning (SFT) and preference training to align model behavior to human preferences for helpfulness and safety. We use grouped query attention (GQA) to improve inference speed.
**Languages covered**: The model has been trained on 23 languages (English, French, Spanish, Italian, German, Portuguese, Japanese, Korean, Arabic, Simplified Chinese, Russian, Polish, Turkish, Vietnamese, Dutch, Czech, Indonesian, Ukrainian, Romanian, Greek, Hindi, Hebrew, and Persian) and evaluated on 10 languages (English, French, Spanish, Italian, German, Portuguese, Japanese, Korean, Arabic, Simplified Chinese).
**Context length**: Command R+ 08-2024 supports a context length of 128K.
### Tool use & Agent capabilities:
Command R+ 08-2024 has been specifically trained with conversational tool use capabilities. These have been trained into the model via a mixture of supervised fine-tuning and preference fine-tuning, using a specific prompt template. Deviating from this prompt template will likely reduce performance, but we encourage experimentation.
Command R+ 08-2024’s tool use functionality takes a conversation as input (with an optional user-system preamble), along with a list of available tools. The model will then generate a json-formatted list of actions to execute on a subset of those tools. Command R+ 08-2024 may use one of its supplied tools more than once.
The model has been trained to recognise a special `directly_answer` tool, which it uses to indicate that it doesn’t want to use any of its other tools. The ability to abstain from calling a specific tool can be useful in a range of situations, such as greeting a user, or asking clarifying questions. We recommend including the `directly_answer` tool, but it can be removed or renamed if required.
Comprehensive documentation for working with Command R+ 08-2024's tool use prompt template can be found [here](https://docs.cohere.com/docs/prompting-command-r).
Command R+ 08-2024 also supports Hugging Face's [tool use API](https://huggingface.co/docs/transformers/main/en/chat_templating#advanced-tool-use--function-calling).
The code snippets below show minimal working examples on how to render a prompt.
<details>
<summary><b>Usage: Rendering Tool Use Prompts [CLICK TO EXPAND]</b> </summary>
```python
from transformers import AutoTokenizer
model_id = "CohereForAI/c4ai-command-r-plus-08-2024"
tokenizer = AutoTokenizer.from_pretrained(model_id)
# define conversation input:
conversation = [
{"role": "user", "content": "Whats the biggest penguin in the world?"}
]
# Define tools available for the model to use:
tools = [
{
"name": "internet_search",
"description": "Returns a list of relevant document snippets for a textual query retrieved from the internet",
"parameter_definitions": {
"query": {
"description": "Query to search the internet with",
"type": 'str',
"required": True
}
}
},
{
'name': "directly_answer",
"description": "Calls a standard (un-augmented) AI chatbot to generate a response given the conversation history",
'parameter_definitions': {}
}
]
# render the tool use prompt as a string:
tool_use_prompt = tokenizer.apply_tool_use_template(
conversation,
tools=tools,
tokenize=False,
add_generation_prompt=True,
)
print(tool_use_prompt)
```
</details>
<details>
<summary><b>Usage: Rendering prompts with the Tool Use API [CLICK TO EXPAND]</b> </summary>
```python
from transformers import AutoTokenizer
model_id = "CohereForAI/c4ai-command-r-plus-08-2024"
tokenizer = AutoTokenizer.from_pretrained(model_id)
# define conversation input:
conversation = [
{"role": "user", "content": "Whats the biggest penguin in the world?"}
]
# Define tools available for the model to use
# Type hints and docstrings from Python functions are automatically extracted
def internet_search(query: str):
"""
Returns a list of relevant document snippets for a textual query retrieved from the internet
Args:
query: Query to search the internet with
"""
pass
def directly_answer():
"""
Calls a standard (un-augmented) AI chatbot to generate a response given the conversation history
"""
pass
tools = [internet_search, directly_answer]
# render the tool use prompt as a string:
tool_use_prompt = tokenizer.apply_chat_template(
conversation,
tools=tools,
tokenize=False,
add_generation_prompt=True,
)
print(tool_use_prompt)
```
</details>
<details>
<summary><b>Example Rendered Tool Use Prompt [CLICK TO EXPAND]</b></summary>
````
<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># Safety Preamble
The instructions in this section override those in the task description and style guide sections. Don't answer questions that are harmful or immoral.
# System Preamble
## Basic Rules
You are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user's requests, you cite your sources in your answers, according to those instructions.
# User Preamble
## Task and Context
You help people answer their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You should focus on serving the user's needs as best you can, which will be wide-ranging.
## Style Guide
Unless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.
## Available Tools
Here is a list of tools that you have available to you:
```python
def internet_search(query: str) -> List[Dict]:
"""Returns a list of relevant document snippets for a textual query retrieved from the internet
Args:
query (str): Query to search the internet with
"""
pass
```
```python
def directly_answer() -> List[Dict]:
"""Calls a standard (un-augmented) AI chatbot to generate a response given the conversation history
"""
pass
```<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Whats the biggest penguin in the world?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>Write 'Action:' followed by a json-formatted list of actions that you want to perform in order to produce a good response to the user's last input. You can use any of the supplied tools any number of times, but you should aim to execute the minimum number of necessary actions for the input. You should use the `directly-answer` tool if calling the other tools is unnecessary. The list of actions you want to call should be formatted as a list of json objects, for example:
```json
[
{
"tool_name": title of the tool in the specification,
"parameters": a dict of parameters to input into the tool as they are defined in the specs, or {} if it takes no parameters
}
]```<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
````
</details>
<details>
<summary><b>Example Rendered Tool Use Completion [CLICK TO EXPAND]</b></summary>
````
Action: ```json
[
{
"tool_name": "internet_search",
"parameters": {
"query": "biggest penguin in the world"
}
}
]
```
````
</details>
### Grounded Generation and RAG Capabilities:
Command R+ 08-2024 has been specifically trained with grounded generation capabilities. This means that it can generate responses based on a list of supplied document snippets, and it will include grounding spans (citations) in its response indicating the source of the information. This can be used to enable behaviors such as grounded summarization and the final step of Retrieval Augmented Generation (RAG). This behavior has been trained into the model via a mixture of supervised fine-tuning and preference fine-tuning, using a specific prompt template. Deviating from this prompt template may reduce performance, but we encourage experimentation.
Command R+ 08-2024’s grounded generation behavior takes a conversation as input (with an optional user-supplied system preamble, indicating task, context and desired output style), along with a list of retrieved document snippets. The document snippets should be chunks, rather than long documents, typically around 100-400 words per chunk. Document snippets consist of key-value pairs. The keys should be short descriptive strings, the values can be text or semi-structured.
By default, Command R+ 08-2024 will generate grounded responses by first predicting which documents are relevant, then predicting which ones it will cite, then generating an answer. Finally, it will then insert grounding spans into the answer. See below for an example. This is referred to as `accurate` grounded generation.
The model is trained with a number of other answering modes, which can be selected by prompt changes. A `fast` citation mode is supported in the tokenizer, which will directly generate an answer with grounding spans in it, without first writing the answer out in full. This sacrifices some grounding accuracy in favor of generating fewer tokens.
Comprehensive documentation for working with Command R+ 08-2024's grounded generation prompt template can be found [here](https://docs.cohere.com/docs/prompting-command-r).
The code snippet below shows a minimal working example on how to render a prompt.
<details>
<summary> <b>Usage: Rendering Grounded Generation prompts [CLICK TO EXPAND]</b> </summary>
````python
from transformers import AutoTokenizer
model_id = "CohereForAI/c4ai-command-r-plus-08-2024"
tokenizer = AutoTokenizer.from_pretrained(model_id)
# define conversation input:
conversation = [
{"role": "user", "content": "Whats the biggest penguin in the world?"}
]
# define documents to ground on:
documents = [
{ "title": "Tall penguins", "text": "Emperor penguins are the tallest growing up to 122 cm in height." },
{ "title": "Penguin habitats", "text": "Emperor penguins only live in Antarctica."}
]
# render the tool use prompt as a string:
grounded_generation_prompt = tokenizer.apply_grounded_generation_template(
conversation,
documents=documents,
citation_mode="accurate", # or "fast"
tokenize=False,
add_generation_prompt=True,
)
print(grounded_generation_prompt)
````
</details>
<details>
<summary><b>Example Rendered Grounded Generation Prompt [CLICK TO EXPAND]</b></summary>
````
<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># Safety Preamble
The instructions in this section override those in the task description and style guide sections. Don't answer questions that are harmful or immoral.
# System Preamble
## Basic Rules
You are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user's requests, you cite your sources in your answers, according to those instructions.
# User Preamble
## Task and Context
You help people answer their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You should focus on serving the user's needs as best you can, which will be wide-ranging.
## Style Guide
Unless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Whats the biggest penguin in the world?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|><results>
Document: 0
title: Tall penguins
text: Emperor penguins are the tallest growing up to 122 cm in height.
Document: 1
title: Penguin habitats
text: Emperor penguins only live in Antarctica.
</results><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>Carefully perform the following instructions, in order, starting each with a new line.
Firstly, Decide which of the retrieved documents are relevant to the user's last input by writing 'Relevant Documents:' followed by comma-separated list of document numbers. If none are relevant, you should instead write 'None'.
Secondly, Decide which of the retrieved documents contain facts that should be cited in a good answer to the user's last input by writing 'Cited Documents:' followed a comma-separated list of document numbers. If you dont want to cite any of them, you should instead write 'None'.
Thirdly, Write 'Answer:' followed by a response to the user's last input in high quality natural english. Use the retrieved documents to help you. Do not insert any citations or grounding markup.
Finally, Write 'Grounded answer:' followed by a response to the user's last input in high quality natural english. Use the symbols <co: doc> and </co: doc> to indicate when a fact comes from a document in the search result, e.g <co: 0>my fact</co: 0> for a fact from document 0.<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
````
</details>
<details>
<summary><b>Example Rendered Grounded Generation Completion [CLICK TO EXPAND]</b></summary>
````
Relevant Documents: 0,1
Cited Documents: 0,1
Answer: The Emperor Penguin is the tallest or biggest penguin in the world. It is a bird that lives only in Antarctica and grows to a height of around 122 centimetres.
Grounded answer: The <co: 0>Emperor Penguin</co: 0> is the <co: 0>tallest</co: 0> or biggest penguin in the world. It is a bird that <co: 1>lives only in Antarctica</co: 1> and <co: 0>grows to a height of around 122 centimetres.</co: 0>
````
</details>
### Code Capabilities:
Command R+ 08-2024 has been optimized to interact with your code, by requesting code snippets, code explanations, or code rewrites. It might not perform well out-of-the-box for pure code completion. For better performance, we also recommend using a low temperature (and even greedy decoding) for code-generation related instructions.
### Model Card Contact
For errors or additional questions about details in this model card, contact [[email protected]](mailto:[email protected]).
### Terms of Use:
We hope that the release of this model will make community-based research efforts more accessible, by releasing the weights of a highly performant 104 billion parameter model to researchers all over the world. This model is governed by a [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license) License with an acceptable use addendum, and also requires adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).
### Try Chat:
You can try Command R+ 08-2024 chat in the playground [here](https://dashboard.cohere.com/playground/chat). You can also use it in our dedicated Hugging Face Space [here](https://huggingface.co/spaces/CohereForAI/c4ai-command?model=command-r-plus-08-2024).
|
{"base_model": "CohereForAI/c4ai-command-r-plus-08-2024", "language": ["en", "fr", "de", "es", "it", "pt", "ja", "ko", "zh", "ar"], "library_name": "transformers", "license": "cc-by-nc-4.0", "inference": false, "extra_gated_prompt": "By submitting this form, you agree to the [License Agreement](https://cohere.com/c4ai-cc-by-nc-license) and acknowledge that the information you provide will be collected, used, and shared in accordance with Cohere’s [Privacy Policy]( https://cohere.com/privacy).", "extra_gated_fields": {"Name": "text", "Affiliation": "text", "Country": {"type": "select", "options": ["Aruba", "Afghanistan", "Angola", "Anguilla", "Åland Islands", "Albania", "Andorra", "United Arab Emirates", "Argentina", "Armenia", "American Samoa", "Antarctica", "French Southern Territories", "Antigua and Barbuda", "Australia", "Austria", "Azerbaijan", "Burundi", "Belgium", "Benin", "Bonaire Sint Eustatius and Saba", "Burkina Faso", "Bangladesh", "Bulgaria", "Bahrain", "Bahamas", "Bosnia and Herzegovina", "Saint Barthélemy", "Belarus", "Belize", "Bermuda", "Plurinational State of Bolivia", "Brazil", "Barbados", "Brunei-Darussalam", "Bhutan", "Bouvet-Island", "Botswana", "Central African Republic", "Canada", "Cocos (Keeling) Islands", "Switzerland", "Chile", "China", "Côte-dIvoire", "Cameroon", "Democratic Republic of the Congo", "Cook Islands", "Colombia", "Comoros", "Cabo Verde", "Costa Rica", "Cuba", "Curaçao", "Christmas Island", "Cayman Islands", "Cyprus", "Czechia", "Germany", "Djibouti", "Dominica", "Denmark", "Dominican Republic", "Algeria", "Ecuador", "Egypt", "Eritrea", "Western Sahara", "Spain", "Estonia", "Ethiopia", "Finland", "Fiji", "Falkland Islands (Malvinas)", "France", "Faroe Islands", "Federated States of Micronesia", "Gabon", "United Kingdom", "Georgia", "Guernsey", "Ghana", "Gibraltar", "Guinea", "Guadeloupe", "Gambia", "Guinea Bissau", "Equatorial Guinea", "Greece", "Grenada", "Greenland", "Guatemala", "French Guiana", "Guam", "Guyana", "Hong Kong", "Heard Island and McDonald Islands", "Honduras", "Croatia", "Haiti", "Hungary", "Indonesia", "Isle of Man", "India", "British Indian Ocean Territory", "Ireland", "Islamic Republic of Iran", "Iraq", "Iceland", "Israel", "Italy", "Jamaica", "Jersey", "Jordan", "Japan", "Kazakhstan", "Kenya", "Kyrgyzstan", "Cambodia", "Kiribati", "Saint-Kitts-and-Nevis", "South Korea", "Kuwait", "Lao-Peoples-Democratic-Republic", "Lebanon", "Liberia", "Libya", "Saint-Lucia", "Liechtenstein", "Sri Lanka", "Lesotho", "Lithuania", "Luxembourg", "Latvia", "Macao", "Saint Martin (French-part)", "Morocco", "Monaco", "Republic of Moldova", "Madagascar", "Maldives", "Mexico", "Marshall Islands", "North Macedonia", "Mali", "Malta", "Myanmar", "Montenegro", "Mongolia", "Northern Mariana Islands", "Mozambique", "Mauritania", "Montserrat", "Martinique", "Mauritius", "Malawi", "Malaysia", "Mayotte", "Namibia", "New Caledonia", "Niger", "Norfolk Island", "Nigeria", "Nicaragua", "Niue", "Netherlands", "Norway", "Nepal", "Nauru", "New Zealand", "Oman", "Pakistan", "Panama", "Pitcairn", "Peru", "Philippines", "Palau", "Papua New Guinea", "Poland", "Puerto Rico", "North Korea", "Portugal", "Paraguay", "State of Palestine", "French Polynesia", "Qatar", "Réunion", "Romania", "Russia", "Rwanda", "Saudi Arabia", "Sudan", "Senegal", "Singapore", "South Georgia and the South Sandwich Islands", "Saint Helena Ascension and Tristan da Cunha", "Svalbard and Jan Mayen", "Solomon Islands", "Sierra Leone", "El Salvador", "San Marino", "Somalia", "Saint Pierre and Miquelon", "Serbia", "South Sudan", "Sao Tome and Principe", "Suriname", "Slovakia", "Slovenia", "Sweden", "Eswatini", "Sint Maarten (Dutch-part)", "Seychelles", "Syrian Arab Republic", "Turks and Caicos Islands", "Chad", "Togo", "Thailand", "Tajikistan", "Tokelau", "Turkmenistan", "Timor Leste", "Tonga", "Trinidad and Tobago", "Tunisia", "Turkey", "Tuvalu", "Taiwan", "United Republic of Tanzania", "Uganda", "Ukraine", "United States Minor Outlying Islands", "Uruguay", "United-States", "Uzbekistan", "Holy See (Vatican City State)", "Saint Vincent and the Grenadines", "Bolivarian Republic of Venezuela", "Virgin Islands British", "Virgin Islands U.S.", "VietNam", "Vanuatu", "Wallis and Futuna", "Samoa", "Yemen", "South Africa", "Zambia", "Zimbabwe"]}, "Receive email updates on C4AI and Cohere research, events, products and services?": {"type": "select", "options": [true, false]}, "I agree to use this model for non-commercial use ONLY": "checkbox"}}
|
task
|
[
"QUESTION_ANSWERING",
"SUMMARIZATION"
] | 45,249 |
mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF
|
mradermacher
| null |
[
"transformers",
"gguf",
"Chain-of-Thought Activation",
"Llama3.1",
"8B",
"CoT",
"SFT",
"text-generation-inference",
"Ollama",
"safetensors",
"Question Answering",
"Math",
"en",
"dataset:O1-OPEN/OpenO1-SFT",
"base_model:prithivMLmods/Llama-3.1-8B-Open-SFT",
"base_model:quantized:prithivMLmods/Llama-3.1-8B-Open-SFT",
"license:creativeml-openrail-m",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | 2024-12-18T22:03:13Z |
2024-12-18T23:45:48+00:00
| 248 | 1 |
---
base_model: prithivMLmods/Llama-3.1-8B-Open-SFT
datasets:
- O1-OPEN/OpenO1-SFT
language:
- en
library_name: transformers
license: creativeml-openrail-m
tags:
- Chain-of-Thought Activation
- Llama3.1
- 8B
- CoT
- SFT
- text-generation-inference
- Ollama
- safetensors
- Question Answering
- Math
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/prithivMLmods/Llama-3.1-8B-Open-SFT
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ1_S.gguf) | i1-IQ1_S | 2.1 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ1_M.gguf) | i1-IQ1_M | 2.3 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.5 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.7 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ2_S.gguf) | i1-IQ2_S | 2.9 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ2_M.gguf) | i1-IQ2_M | 3.0 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q2_K_S.gguf) | i1-Q2_K_S | 3.1 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q2_K.gguf) | i1-Q2_K | 3.3 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.4 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.6 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.8 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ3_S.gguf) | i1-IQ3_S | 3.8 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ3_M.gguf) | i1-IQ3_M | 3.9 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.1 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q3_K_L.gguf) | i1-Q3_K_L | 4.4 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.5 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q4_0.gguf) | i1-Q4_0 | 4.8 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.8 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.0 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q5_K_S.gguf) | i1-Q5_K_S | 5.7 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.8 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q6_K.gguf) | i1-Q6_K | 6.7 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
| null |
Non_BioNLP
|
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/prithivMLmods/Llama-3.1-8B-Open-SFT
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ1_S.gguf) | i1-IQ1_S | 2.1 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ1_M.gguf) | i1-IQ1_M | 2.3 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.5 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.7 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ2_S.gguf) | i1-IQ2_S | 2.9 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ2_M.gguf) | i1-IQ2_M | 3.0 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q2_K_S.gguf) | i1-Q2_K_S | 3.1 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q2_K.gguf) | i1-Q2_K | 3.3 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.4 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.6 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.8 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ3_S.gguf) | i1-IQ3_S | 3.8 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ3_M.gguf) | i1-IQ3_M | 3.9 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.1 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q3_K_L.gguf) | i1-Q3_K_L | 4.4 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.5 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q4_0.gguf) | i1-Q4_0 | 4.8 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.8 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.0 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q5_K_S.gguf) | i1-Q5_K_S | 5.7 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.8 | |
| [GGUF](https://huggingface.co/mradermacher/Llama-3.1-8B-Open-SFT-i1-GGUF/resolve/main/Llama-3.1-8B-Open-SFT.i1-Q6_K.gguf) | i1-Q6_K | 6.7 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
{"base_model": "prithivMLmods/Llama-3.1-8B-Open-SFT", "datasets": ["O1-OPEN/OpenO1-SFT"], "language": ["en"], "library_name": "transformers", "license": "creativeml-openrail-m", "tags": ["Chain-of-Thought Activation", "Llama3.1", "8B", "CoT", "SFT", "text-generation-inference", "Ollama", "safetensors", "Question Answering", "Math"], "quantized_by": "mradermacher"}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,250 |
shwan/KominiLM-steam_classifier_readme_test
|
shwan
| null |
[
"pytorch",
"region:us"
] | 2022-08-15T23:27:01Z |
2022-08-15T23:30:05+00:00
| 0 | 0 |
---
{}
---
# KoMiniLM
🐣 Korean mini language model
## Overview
Current language models usually consist of hundreds of millions of parameters which brings challenges for fine-tuning and online serving in real-life applications due to latency and capacity constraints. In this project, we release a light weight korean language model to address the aforementioned shortcomings of existing language models.
## Quick tour
```python
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("BM-K/KoMiniLM") # 23M model
model = AutoModel.from_pretrained("BM-K/KoMiniLM")
inputs = tokenizer("안녕 세상아!", return_tensors="pt")
outputs = model(**inputs)
```
## Update history
** Updates on 2022.06.20 **
- Release KoMiniLM-bert-68M
** Updates on 2022.05.24 **
- Release KoMiniLM-bert-23M
## Pre-training
`Teacher Model`: [KLUE-BERT(base)](https://github.com/KLUE-benchmark/KLUE)
### Object
Self-Attention Distribution and Self-Attention Value-Relation [[Wang et al., 2020]] were distilled from each discrete layer of the teacher model to the student model. Wang et al. distilled in the last layer of the transformer, but that was not the case in this project.
### Data sets
|Data|News comments|News article|
|:----:|:----:|:----:|
|size|10G|10G|
### Config
- **KoMiniLM-23M**
```json
{
"architectures": [
"BartForPreTraining"
],
"attention_probs_dropout_prob": 0.1,
"classifier_dropout": null,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 384,
"initializer_range": 0.02,
"intermediate_size": 1536,
"layer_norm_eps": 1e-12,
"max_position_embeddings": 512,
"model_type": "bart",
"num_attention_heads": 12,
"num_hidden_layers": 6,
"output_attentions": true,
"pad_token_id": 0,
"position_embedding_type": "absolute",
"return_dict": false,
"torch_dtype": "float32",
"transformers_version": "4.13.0",
"type_vocab_size": 2,
"use_cache": true,
"vocab_size": 32000
}
```
### Performance on subtasks
- The results of our fine-tuning experiments are an average of 3 runs for each task.
```
cd KoMiniLM-Finetune
bash scripts/run_all_kominilm.sh
```
|| #Param | Average | NSMC<br>(Acc) | Naver NER<br>(F1) | PAWS<br>(Acc) | KorNLI<br>(Acc) | KorSTS<br>(Spearman) | Question Pair<br>(Acc) | KorQuaD<br>(Dev)<br>(EM/F1) |
|:----:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|
|KoBERT(KLUE)| 110M | 86.84 | 90.20±0.07 | 87.11±0.05 | 81.36±0.21 | 81.06±0.33 | 82.47±0.14 | 95.03±0.44 | 84.43±0.18 / <br>93.05±0.04 |
|KcBERT| 108M | 78.94 | 89.60±0.10 | 84.34±0.13 | 67.02±0.42| 74.17±0.52 | 76.57±0.51 | 93.97±0.27 | 60.87±0.27 / <br>85.01±0.14 |
|KoBERT(SKT)| 92M | 79.73 | 89.28±0.42 | 87.54±0.04 | 80.93±0.91 | 78.18±0.45 | 75.98±2.81 | 94.37±0.31 | 51.94±0.60 / <br>79.69±0.66 |
|DistilKoBERT| 28M | 74.73 | 88.39±0.08 | 84.22±0.01 | 61.74±0.45 | 70.22±0.14 | 72.11±0.27 | 92.65±0.16 | 52.52±0.48 / <br>76.00±0.71 |
| | | | | | | | | |
|**KoMiniLM<sup>†</sup>**| **68M** | 85.90 | 89.84±0.02 | 85.98±0.09 | 80.78±0.30 | 79.28±0.17 | 81.00±0.07 | 94.89±0.37 | 83.27±0.08 / <br>92.08±0.06 |
|**KoMiniLM<sup>†</sup>**| **23M** | 84.79 | 89.67±0.03 | 84.79±0.09 | 78.67±0.45 | 78.10±0.07 | 78.90±0.11 | 94.81±0.12 | 82.11±0.42 / <br>91.21±0.29 |
- [NSMC](https://github.com/e9t/nsmc) (Naver Sentiment Movie Corpus)
- [Naver NER](https://github.com/naver/nlp-challenge) (NER task on Naver NLP Challenge 2018)
- [PAWS](https://github.com/google-research-datasets/paws) (Korean Paraphrase Adversaries from Word Scrambling)
- [KorNLI/KorSTS](https://github.com/kakaobrain/KorNLUDatasets) (Korean Natural Language Understanding)
- [Question Pair](https://github.com/songys/Question_pair) (Paired Question)
- [KorQuAD](https://korquad.github.io/) (The Korean Question Answering Dataset)
<img src = "https://user-images.githubusercontent.com/55969260/174229747-279122dc-9d27-4da9-a6e7-f9f1fe1651f7.png"> <br>
### User Contributed Examples
-
## Reference
- [KLUE BERT](https://github.com/KLUE-benchmark/KLUE)
- [KcBERT](https://github.com/Beomi/KcBERT)
- [SKT KoBERT](https://github.com/SKTBrain/KoBERT)
- [DistilKoBERT](https://github.com/monologg/DistilKoBERT)
- [lassl](https://github.com/lassl/lassl)
| null |
Non_BioNLP
|
# KoMiniLM
🐣 Korean mini language model
## Overview
Current language models usually consist of hundreds of millions of parameters which brings challenges for fine-tuning and online serving in real-life applications due to latency and capacity constraints. In this project, we release a light weight korean language model to address the aforementioned shortcomings of existing language models.
## Quick tour
```python
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("BM-K/KoMiniLM") # 23M model
model = AutoModel.from_pretrained("BM-K/KoMiniLM")
inputs = tokenizer("안녕 세상아!", return_tensors="pt")
outputs = model(**inputs)
```
## Update history
** Updates on 2022.06.20 **
- Release KoMiniLM-bert-68M
** Updates on 2022.05.24 **
- Release KoMiniLM-bert-23M
## Pre-training
`Teacher Model`: [KLUE-BERT(base)](https://github.com/KLUE-benchmark/KLUE)
### Object
Self-Attention Distribution and Self-Attention Value-Relation [[Wang et al., 2020]] were distilled from each discrete layer of the teacher model to the student model. Wang et al. distilled in the last layer of the transformer, but that was not the case in this project.
### Data sets
|Data|News comments|News article|
|:----:|:----:|:----:|
|size|10G|10G|
### Config
- **KoMiniLM-23M**
```json
{
"architectures": [
"BartForPreTraining"
],
"attention_probs_dropout_prob": 0.1,
"classifier_dropout": null,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 384,
"initializer_range": 0.02,
"intermediate_size": 1536,
"layer_norm_eps": 1e-12,
"max_position_embeddings": 512,
"model_type": "bart",
"num_attention_heads": 12,
"num_hidden_layers": 6,
"output_attentions": true,
"pad_token_id": 0,
"position_embedding_type": "absolute",
"return_dict": false,
"torch_dtype": "float32",
"transformers_version": "4.13.0",
"type_vocab_size": 2,
"use_cache": true,
"vocab_size": 32000
}
```
### Performance on subtasks
- The results of our fine-tuning experiments are an average of 3 runs for each task.
```
cd KoMiniLM-Finetune
bash scripts/run_all_kominilm.sh
```
|| #Param | Average | NSMC<br>(Acc) | Naver NER<br>(F1) | PAWS<br>(Acc) | KorNLI<br>(Acc) | KorSTS<br>(Spearman) | Question Pair<br>(Acc) | KorQuaD<br>(Dev)<br>(EM/F1) |
|:----:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|
|KoBERT(KLUE)| 110M | 86.84 | 90.20±0.07 | 87.11±0.05 | 81.36±0.21 | 81.06±0.33 | 82.47±0.14 | 95.03±0.44 | 84.43±0.18 / <br>93.05±0.04 |
|KcBERT| 108M | 78.94 | 89.60±0.10 | 84.34±0.13 | 67.02±0.42| 74.17±0.52 | 76.57±0.51 | 93.97±0.27 | 60.87±0.27 / <br>85.01±0.14 |
|KoBERT(SKT)| 92M | 79.73 | 89.28±0.42 | 87.54±0.04 | 80.93±0.91 | 78.18±0.45 | 75.98±2.81 | 94.37±0.31 | 51.94±0.60 / <br>79.69±0.66 |
|DistilKoBERT| 28M | 74.73 | 88.39±0.08 | 84.22±0.01 | 61.74±0.45 | 70.22±0.14 | 72.11±0.27 | 92.65±0.16 | 52.52±0.48 / <br>76.00±0.71 |
| | | | | | | | | |
|**KoMiniLM<sup>†</sup>**| **68M** | 85.90 | 89.84±0.02 | 85.98±0.09 | 80.78±0.30 | 79.28±0.17 | 81.00±0.07 | 94.89±0.37 | 83.27±0.08 / <br>92.08±0.06 |
|**KoMiniLM<sup>†</sup>**| **23M** | 84.79 | 89.67±0.03 | 84.79±0.09 | 78.67±0.45 | 78.10±0.07 | 78.90±0.11 | 94.81±0.12 | 82.11±0.42 / <br>91.21±0.29 |
- [NSMC](https://github.com/e9t/nsmc) (Naver Sentiment Movie Corpus)
- [Naver NER](https://github.com/naver/nlp-challenge) (NER task on Naver NLP Challenge 2018)
- [PAWS](https://github.com/google-research-datasets/paws) (Korean Paraphrase Adversaries from Word Scrambling)
- [KorNLI/KorSTS](https://github.com/kakaobrain/KorNLUDatasets) (Korean Natural Language Understanding)
- [Question Pair](https://github.com/songys/Question_pair) (Paired Question)
- [KorQuAD](https://korquad.github.io/) (The Korean Question Answering Dataset)
<img src = "https://user-images.githubusercontent.com/55969260/174229747-279122dc-9d27-4da9-a6e7-f9f1fe1651f7.png"> <br>
### User Contributed Examples
-
## Reference
- [KLUE BERT](https://github.com/KLUE-benchmark/KLUE)
- [KcBERT](https://github.com/Beomi/KcBERT)
- [SKT KoBERT](https://github.com/SKTBrain/KoBERT)
- [DistilKoBERT](https://github.com/monologg/DistilKoBERT)
- [lassl](https://github.com/lassl/lassl)
|
{}
|
task
|
[
"QUESTION_ANSWERING"
] | 45,251 |
gaudi/opus-mt-ccs-en-ctranslate2
|
gaudi
|
translation
|
[
"transformers",
"marian",
"ctranslate2",
"translation",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | 2024-07-17T00:08:25Z |
2024-10-18T21:56:26+00:00
| 6 | 0 |
---
license: apache-2.0
tags:
- ctranslate2
- translation
---
# Repository General Information
## Inspired by and derived from the work of [Helsinki-NLP](https://huggingface.co/Helsinki-NLP), [CTranslate2](https://github.com/OpenNMT/CTranslate2), and [michaelfeil](https://huggingface.co/michaelfeil)!
- Link to Original Model ([Helsinki-NLP](https://huggingface.co/Helsinki-NLP)): [Model Link](https://huggingface.co/Helsinki-NLP/opus-mt-ccs-en)
- This respository was based on the work of [CTranslate2](https://github.com/OpenNMT/CTranslate2).
- This repository was based on the work of [michaelfeil](https://huggingface.co/michaelfeil).
# What is CTranslate2?
[CTranslate2](https://opennmt.net/CTranslate2/) is a C++ and Python library for efficient inference with Transformer models.
CTranslate2 implements a custom runtime that applies many performance optimization techniques such as weights quantization, layers fusion, batch reordering, etc., to accelerate and reduce the memory usage of Transformer models on CPU and GPU.
CTranslate2 is one of the most performant ways of hosting translation models at scale. Current supported models include:
- Encoder-decoder models: Transformer base/big, M2M-100, NLLB, BART, mBART, Pegasus, T5, Whisper
- Decoder-only models: GPT-2, GPT-J, GPT-NeoX, OPT, BLOOM, MPT, Llama, Mistral, Gemma, CodeGen, GPTBigCode, Falcon
- Encoder-only models: BERT, DistilBERT, XLM-RoBERTa
The project is production-oriented and comes with backward compatibility guarantees, but it also includes experimental features related to model compression and inference acceleration.
# CTranslate2 Benchmarks
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings. Tested against `newstest2014` (En -> De) dataset.
The benchmark reports the number of target tokens generated per second (higher is better). The results are aggregated over multiple runs. See the benchmark scripts for more details and reproduce these numbers.
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings.
## CPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 147.3 | 2332MB | 27.90 |
| Marian 1.11.0 (int16) | 330.2 | 5901MB | 27.65 |
| Marian 1.11.0 (int8) | 355.8 | 4763MB | 27.27 |
| CTranslate2 3.6.0 (int16) | 596.1 | 660MB | 27.53 |
| CTranslate2 3.6.0 (int8) | 696.1 | 516MB | 27.65 |
## GPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max GPU Memory Usage | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 1022.9 | 4097MB | 2109MB | 27.90 |
| Marian 1.11.0 (float16) | 3962.4 | 3239MB | 1976MB | 27.94 |
| CTranslate2 3.6.0 (float16) | 9296.7 | 909MB | 814MB | 27.9 |
| CTranslate2 3.6.0 (int8 + float16) | 8362.7 | 813MB | 766MB | 27.9 |
`Executed with 4 threads on a c5.2xlarge Amazon EC2 instance equipped with an Intel(R) Xeon(R) Platinum 8275CL CPU.`
**Source to benchmark information can be found [here](https://github.com/OpenNMT/CTranslate2).**<br />
**Original model BLEU scores can be found [here](https://huggingface.co/Helsinki-NLP/opus-mt-ccs-en).**
## Internal Benchmarks
Internal testing on our end showed **inference times reduced by 6x-10x** on average compared the vanilla checkpoints using the *transformers* library. A **slight reduction on BLEU scores (~5%)** was also identified in comparison to the vanilla checkpoints with a few exceptions. This is likely due to several factors, one being the quantization applied. Further testing is needed from our end to better assess the reduction in translation quality. The command used to compile the vanilla checkpoint into a CTranslate2 model can be found below. Modifying this command can yield differing balances between inferencing performance and translation quality.
# CTranslate2 Installation
```bash
pip install hf-hub-ctranslate2>=1.0.0 ctranslate2>=3.13.0
```
### ct2-transformers-converter Command Used:
```bash
ct2-transformers-converter --model Helsinki-NLP/opus-mt-ccs-en --output_dir ./ctranslate2/opus-mt-ccs-en-ctranslate2 --force --copy_files README.md generation_config.json tokenizer_config.json vocab.json source.spm .gitattributes target.spm --quantization float16
```
# CTranslate2 Converted Checkpoint Information:
**Compatible With:**
- [ctranslate2](https://github.com/OpenNMT/CTranslate2)
- [hf-hub-ctranslate2](https://github.com/michaelfeil/hf-hub-ctranslate2)
**Compute Type:**
- `compute_type=int8_float16` for `device="cuda"`
- `compute_type=int8` for `device="cpu"`
# Sample Code - ctranslate2
#### Clone the repository to the working directory or wherever you wish to store the model artifacts. ####
```bash
git clone https://huggingface.co/gaudi/opus-mt-ccs-en-ctranslate2
```
#### Take the python code below and update the 'model_dir' variable to the location of the cloned repository. ####
```python
from ctranslate2 import Translator
import transformers
model_dir = "./opus-mt-ccs-en-ctranslate2" # Path to model directory.
translator = Translator(
model_path=model_dir,
device="cuda", # cpu, cuda, or auto.
inter_threads=1, # Maximum number of parallel translations.
intra_threads=4, # Number of OpenMP threads per translator.
compute_type="int8_float16", # int8 for cpu or int8_float16 for cuda.
)
tokenizer = transformers.AutoTokenizer.from_pretrained(model_dir)
source = tokenizer.convert_ids_to_tokens(tokenizer.encode("XXXXXX, XXX XX XXXXXX."))
results = translator.translate_batch([source])
target = results[0].hypotheses[0]
print(tokenizer.decode(tokenizer.convert_tokens_to_ids(target)))
```
# Sample Code - hf-hub-ctranslate2
**Derived From [michaelfeil](https://huggingface.co/michaelfeil):**
```python
from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
from transformers import AutoTokenizer
model_name = "gaudi/opus-mt-ccs-en-ctranslate2"
model = TranslatorCT2fromHfHub(
model_name_or_path=model_name,
device="cuda",
compute_type="int8_float16",
tokenizer=AutoTokenizer.from_pretrained(model_name)
)
outputs = model.generate(
text=["XXX XX XXX XXXXXXX XXXX?", "XX XX XXXX XX XXX!"],
)
print(outputs)
```
# License and other remarks:
License conditions are intended to be idential to [original huggingface repository](https://huggingface.co/Helsinki-NLP/opus-mt-ccs-en) by Helsinki-NLP.
| null |
Non_BioNLP
|
# Repository General Information
## Inspired by and derived from the work of [Helsinki-NLP](https://huggingface.co/Helsinki-NLP), [CTranslate2](https://github.com/OpenNMT/CTranslate2), and [michaelfeil](https://huggingface.co/michaelfeil)!
- Link to Original Model ([Helsinki-NLP](https://huggingface.co/Helsinki-NLP)): [Model Link](https://huggingface.co/Helsinki-NLP/opus-mt-ccs-en)
- This respository was based on the work of [CTranslate2](https://github.com/OpenNMT/CTranslate2).
- This repository was based on the work of [michaelfeil](https://huggingface.co/michaelfeil).
# What is CTranslate2?
[CTranslate2](https://opennmt.net/CTranslate2/) is a C++ and Python library for efficient inference with Transformer models.
CTranslate2 implements a custom runtime that applies many performance optimization techniques such as weights quantization, layers fusion, batch reordering, etc., to accelerate and reduce the memory usage of Transformer models on CPU and GPU.
CTranslate2 is one of the most performant ways of hosting translation models at scale. Current supported models include:
- Encoder-decoder models: Transformer base/big, M2M-100, NLLB, BART, mBART, Pegasus, T5, Whisper
- Decoder-only models: GPT-2, GPT-J, GPT-NeoX, OPT, BLOOM, MPT, Llama, Mistral, Gemma, CodeGen, GPTBigCode, Falcon
- Encoder-only models: BERT, DistilBERT, XLM-RoBERTa
The project is production-oriented and comes with backward compatibility guarantees, but it also includes experimental features related to model compression and inference acceleration.
# CTranslate2 Benchmarks
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings. Tested against `newstest2014` (En -> De) dataset.
The benchmark reports the number of target tokens generated per second (higher is better). The results are aggregated over multiple runs. See the benchmark scripts for more details and reproduce these numbers.
Please note that the results presented below are only valid for the configuration used during this benchmark: absolute and relative performance may change with different settings.
## CPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 147.3 | 2332MB | 27.90 |
| Marian 1.11.0 (int16) | 330.2 | 5901MB | 27.65 |
| Marian 1.11.0 (int8) | 355.8 | 4763MB | 27.27 |
| CTranslate2 3.6.0 (int16) | 596.1 | 660MB | 27.53 |
| CTranslate2 3.6.0 (int8) | 696.1 | 516MB | 27.65 |
## GPU Benchmarks for Generic Opus-MT Models
| Library | Tokens per Second | Max GPU Memory Usage | Max Memory Usage | BLEU |
| :----: | :----: | :----: | :----: | :----: |
| Transformers 4.26.1 (with PyTorch 1.13.1) | 1022.9 | 4097MB | 2109MB | 27.90 |
| Marian 1.11.0 (float16) | 3962.4 | 3239MB | 1976MB | 27.94 |
| CTranslate2 3.6.0 (float16) | 9296.7 | 909MB | 814MB | 27.9 |
| CTranslate2 3.6.0 (int8 + float16) | 8362.7 | 813MB | 766MB | 27.9 |
`Executed with 4 threads on a c5.2xlarge Amazon EC2 instance equipped with an Intel(R) Xeon(R) Platinum 8275CL CPU.`
**Source to benchmark information can be found [here](https://github.com/OpenNMT/CTranslate2).**<br />
**Original model BLEU scores can be found [here](https://huggingface.co/Helsinki-NLP/opus-mt-ccs-en).**
## Internal Benchmarks
Internal testing on our end showed **inference times reduced by 6x-10x** on average compared the vanilla checkpoints using the *transformers* library. A **slight reduction on BLEU scores (~5%)** was also identified in comparison to the vanilla checkpoints with a few exceptions. This is likely due to several factors, one being the quantization applied. Further testing is needed from our end to better assess the reduction in translation quality. The command used to compile the vanilla checkpoint into a CTranslate2 model can be found below. Modifying this command can yield differing balances between inferencing performance and translation quality.
# CTranslate2 Installation
```bash
pip install hf-hub-ctranslate2>=1.0.0 ctranslate2>=3.13.0
```
### ct2-transformers-converter Command Used:
```bash
ct2-transformers-converter --model Helsinki-NLP/opus-mt-ccs-en --output_dir ./ctranslate2/opus-mt-ccs-en-ctranslate2 --force --copy_files README.md generation_config.json tokenizer_config.json vocab.json source.spm .gitattributes target.spm --quantization float16
```
# CTranslate2 Converted Checkpoint Information:
**Compatible With:**
- [ctranslate2](https://github.com/OpenNMT/CTranslate2)
- [hf-hub-ctranslate2](https://github.com/michaelfeil/hf-hub-ctranslate2)
**Compute Type:**
- `compute_type=int8_float16` for `device="cuda"`
- `compute_type=int8` for `device="cpu"`
# Sample Code - ctranslate2
#### Clone the repository to the working directory or wherever you wish to store the model artifacts. ####
```bash
git clone https://huggingface.co/gaudi/opus-mt-ccs-en-ctranslate2
```
#### Take the python code below and update the 'model_dir' variable to the location of the cloned repository. ####
```python
from ctranslate2 import Translator
import transformers
model_dir = "./opus-mt-ccs-en-ctranslate2" # Path to model directory.
translator = Translator(
model_path=model_dir,
device="cuda", # cpu, cuda, or auto.
inter_threads=1, # Maximum number of parallel translations.
intra_threads=4, # Number of OpenMP threads per translator.
compute_type="int8_float16", # int8 for cpu or int8_float16 for cuda.
)
tokenizer = transformers.AutoTokenizer.from_pretrained(model_dir)
source = tokenizer.convert_ids_to_tokens(tokenizer.encode("XXXXXX, XXX XX XXXXXX."))
results = translator.translate_batch([source])
target = results[0].hypotheses[0]
print(tokenizer.decode(tokenizer.convert_tokens_to_ids(target)))
```
# Sample Code - hf-hub-ctranslate2
**Derived From [michaelfeil](https://huggingface.co/michaelfeil):**
```python
from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
from transformers import AutoTokenizer
model_name = "gaudi/opus-mt-ccs-en-ctranslate2"
model = TranslatorCT2fromHfHub(
model_name_or_path=model_name,
device="cuda",
compute_type="int8_float16",
tokenizer=AutoTokenizer.from_pretrained(model_name)
)
outputs = model.generate(
text=["XXX XX XXX XXXXXXX XXXX?", "XX XX XXXX XX XXX!"],
)
print(outputs)
```
# License and other remarks:
License conditions are intended to be idential to [original huggingface repository](https://huggingface.co/Helsinki-NLP/opus-mt-ccs-en) by Helsinki-NLP.
|
{"license": "apache-2.0", "tags": ["ctranslate2", "translation"]}
|
task
|
[
"TRANSLATION"
] | 45,252 |
Catchy1282/GradientBoosting_model_3_samples_per_label
|
Catchy1282
|
text-classification
|
[
"setfit",
"safetensors",
"mpnet",
"sentence-transformers",
"text-classification",
"generated_from_setfit_trainer",
"arxiv:2209.11055",
"base_model:sentence-transformers/all-mpnet-base-v2",
"base_model:finetune:sentence-transformers/all-mpnet-base-v2",
"region:us"
] | 2024-08-09T12:27:47Z |
2024-08-09T12:28:10+00:00
| 10 | 0 |
---
base_model: sentence-transformers/all-mpnet-base-v2
library_name: setfit
metrics:
- accuracy
pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
widget: []
inference: true
---
# SetFit with sentence-transformers/all-mpnet-base-v2
This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) as the Sentence Transformer embedding model. A GradientBoostingClassifier instance is used for classification.
The model has been trained using an efficient few-shot learning technique that involves:
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.
## Model Details
### Model Description
- **Model Type:** SetFit
- **Sentence Transformer body:** [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2)
- **Classification head:** a GradientBoostingClassifier instance
- **Maximum Sequence Length:** 384 tokens
<!-- - **Number of Classes:** Unknown -->
<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
## Uses
### Direct Use for Inference
First install the SetFit library:
```bash
pip install setfit
```
Then you can load this model and run inference.
```python
from setfit import SetFitModel
# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("Catchy1282/GradientBoosting_model_3_samples_per_label")
# Run inference
preds = model("I loved the spiderman movie!")
```
<!--
### Downstream Use
*List how someone could finetune this model on their own dataset.*
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Framework Versions
- Python: 3.11.9
- SetFit: 1.0.3
- Sentence Transformers: 3.0.1
- Transformers: 4.39.0
- PyTorch: 2.3.0+cu121
- Datasets: 2.20.0
- Tokenizers: 0.15.2
## Citation
### BibTeX
```bibtex
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# SetFit with sentence-transformers/all-mpnet-base-v2
This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) as the Sentence Transformer embedding model. A GradientBoostingClassifier instance is used for classification.
The model has been trained using an efficient few-shot learning technique that involves:
1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.
## Model Details
### Model Description
- **Model Type:** SetFit
- **Sentence Transformer body:** [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2)
- **Classification head:** a GradientBoostingClassifier instance
- **Maximum Sequence Length:** 384 tokens
<!-- - **Number of Classes:** Unknown -->
<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
## Uses
### Direct Use for Inference
First install the SetFit library:
```bash
pip install setfit
```
Then you can load this model and run inference.
```python
from setfit import SetFitModel
# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("Catchy1282/GradientBoosting_model_3_samples_per_label")
# Run inference
preds = model("I loved the spiderman movie!")
```
<!--
### Downstream Use
*List how someone could finetune this model on their own dataset.*
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Framework Versions
- Python: 3.11.9
- SetFit: 1.0.3
- Sentence Transformers: 3.0.1
- Transformers: 4.39.0
- PyTorch: 2.3.0+cu121
- Datasets: 2.20.0
- Tokenizers: 0.15.2
## Citation
### BibTeX
```bibtex
@article{https://doi.org/10.48550/arxiv.2209.11055,
doi = {10.48550/ARXIV.2209.11055},
url = {https://arxiv.org/abs/2209.11055},
author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Efficient Few-Shot Learning Without Prompts},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution 4.0 International}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "sentence-transformers/all-mpnet-base-v2", "library_name": "setfit", "metrics": ["accuracy"], "pipeline_tag": "text-classification", "tags": ["setfit", "sentence-transformers", "text-classification", "generated_from_setfit_trainer"], "widget": [], "inference": true}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,253 |
Helsinki-NLP/opus-mt-kl-en
|
Helsinki-NLP
|
translation
|
[
"transformers",
"pytorch",
"tf",
"marian",
"text2text-generation",
"translation",
"kl",
"en",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2022-03-02T23:29:04Z |
2023-08-16T11:59:37+00:00
| 38 | 1 |
---
license: apache-2.0
tags:
- translation
---
### opus-mt-kl-en
* source languages: kl
* target languages: en
* OPUS readme: [kl-en](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/kl-en/README.md)
* dataset: opus
* model: transformer-align
* pre-processing: normalization + SentencePiece
* download original weights: [opus-2020-01-09.zip](https://object.pouta.csc.fi/OPUS-MT-models/kl-en/opus-2020-01-09.zip)
* test set translations: [opus-2020-01-09.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/kl-en/opus-2020-01-09.test.txt)
* test set scores: [opus-2020-01-09.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/kl-en/opus-2020-01-09.eval.txt)
## Benchmarks
| testset | BLEU | chr-F |
|-----------------------|-------|-------|
| JW300.kl.en | 26.4 | 0.432 |
| Tatoeba.kl.en | 35.5 | 0.443 |
| null |
Non_BioNLP
|
### opus-mt-kl-en
* source languages: kl
* target languages: en
* OPUS readme: [kl-en](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/kl-en/README.md)
* dataset: opus
* model: transformer-align
* pre-processing: normalization + SentencePiece
* download original weights: [opus-2020-01-09.zip](https://object.pouta.csc.fi/OPUS-MT-models/kl-en/opus-2020-01-09.zip)
* test set translations: [opus-2020-01-09.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/kl-en/opus-2020-01-09.test.txt)
* test set scores: [opus-2020-01-09.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/kl-en/opus-2020-01-09.eval.txt)
## Benchmarks
| testset | BLEU | chr-F |
|-----------------------|-------|-------|
| JW300.kl.en | 26.4 | 0.432 |
| Tatoeba.kl.en | 35.5 | 0.443 |
|
{"license": "apache-2.0", "tags": ["translation"]}
|
task
|
[
"TRANSLATION"
] | 45,254 |
openaccess-ai-collective/minotaur-7b
|
openaccess-ai-collective
|
text-generation
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"OpenAccess AI Collective",
"MPT",
"axolotl",
"dataset:ehartford/WizardLM_alpaca_evol_instruct_70k_unfiltered",
"dataset:QingyiSi/Alpaca-CoT",
"dataset:teknium/GPTeacher-General-Instruct",
"dataset:metaeval/ScienceQA_text_only",
"dataset:hellaswag",
"dataset:openai/summarize_from_feedback",
"dataset:riddle_sense",
"dataset:gsm8k",
"dataset:camel-ai/math",
"dataset:camel-ai/biology",
"dataset:camel-ai/physics",
"dataset:camel-ai/chemistry",
"dataset:winglian/evals",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"region:us"
] | 2023-06-12T00:54:26Z |
2023-06-12T03:32:31+00:00
| 42 | 2 |
---
datasets:
- ehartford/WizardLM_alpaca_evol_instruct_70k_unfiltered
- QingyiSi/Alpaca-CoT
- teknium/GPTeacher-General-Instruct
- metaeval/ScienceQA_text_only
- hellaswag
- openai/summarize_from_feedback
- riddle_sense
- gsm8k
- camel-ai/math
- camel-ai/biology
- camel-ai/physics
- camel-ai/chemistry
- winglian/evals
license: apache-2.0
tags:
- OpenAccess AI Collective
- MPT
- axolotl
inference: false
---
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
**[💵 Donate to OpenAccess AI Collective](https://github.com/sponsors/OpenAccess-AI-Collective) to help us keep building great tools and models!**
# Minotaur 7B
Minotaur 7B is an instruct fine-tuned model on top of LlaMA-13B. Minotaur 7B is fine-tuned **on only completely open datasets** making this model reproducible by anyone.
Questions, comments, feedback, looking to donate, or want to help? Reach out on our [Discord](https://discord.gg/PugNNHAF5r) or email [[email protected]](mailto:[email protected])
# Prompts
Chat only style prompts using `USER:`,`ASSISTANT:`.
<img src="https://huggingface.co/openaccess-ai-collective/minotaur-7b/resolve/main/minotaur.png" alt="minotaur" width="600" height="500"/>
# Training Datasets
Minotaur 7B model is fine-tuned on the following openly available datasets:
- [WizardLM](https://huggingface.co/datasets/ehartford/WizardLM_alpaca_evol_instruct_70k_unfiltered)
- [subset of QingyiSi/Alpaca-CoT for roleplay and CoT](https://huggingface.co/QingyiSi/Alpaca-CoT)
- [GPTeacher-General-Instruct](https://huggingface.co/datasets/teknium/GPTeacher-General-Instruct)
- [metaeval/ScienceQA_text_only](https://huggingface.co/datasets/metaeval/ScienceQA_text_only) - instruct for concise responses
- [openai/summarize_from_feedback](https://huggingface.co/datasets/openai/summarize_from_feedback) - instruct augmented tl;dr summarization
- [camel-ai/math](https://huggingface.co/datasets/camel-ai/math)
- [camel-ai/physics](https://huggingface.co/datasets/camel-ai/physics)
- [camel-ai/chemistry](https://huggingface.co/datasets/camel-ai/chemistry)
- [camel-ai/biology](https://huggingface.co/datasets/camel-ai/biology)
- [winglian/evals](https://huggingface.co/datasets/winglian/evals) - instruct augmented datasets
- custom sysnthetic datasets around misconceptions, in-context qa, jokes, N-tasks problems, and context-insensitivity
- ARC-Easy & ARC-Challenge - instruct augmented for detailed responses, derived from the `train` split
- [hellaswag](https://huggingface.co/datasets/hellaswag) - 30K+ rows of instruct augmented for detailed explanations w 30K+ rows, derived from the `train` split
- [riddle_sense](https://huggingface.co/datasets/riddle_sense) - instruct augmented, derived from the `train` split
- [gsm8k](https://huggingface.co/datasets/gsm8k) - instruct augmented, derived from the `train` split
- prose generation
# Shoutouts
Special thanks to Nanobit for helping with Axolotl and TheBloke for quantizing these models are more accessible to all.
# Demo
HF Demo in Spaces available in the [Community ChatBot Arena](https://huggingface.co/spaces/openaccess-ai-collective/rlhf-arena) under the OAAIC Chatbots tab.
## Release Notes
- https://wandb.ai/wing-lian/minotaur-13b/runs/ucgd0x6n
## Build
Minotaur was built with [Axolotl](https://github.com/OpenAccess-AI-Collective/axolotl) on 8x RTXA6000
- 1 epochs taking approximately 1.5 hours
## Bias, Risks, and Limitations
Minotaur has not been aligned to human preferences with techniques like RLHF or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so).
Minotaur was fine-tuned from the base model LLaMA 13B, please refer to its model card's Limitations Section for relevant information. (included below)
## Benchmarks
hf-causal-experimental (pretrained=openaccess-ai-collective/minotaur-7b), limit: None, provide_description: False, num_fewshot: 0, batch_size: None
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.4078|_ |0.0144|
| | |acc_norm|0.4292|_ |0.0145|
|arc_easy | 0|acc |0.7424|_ |0.0090|
| | |acc_norm|0.6877|_ |0.0095|
|boolq | 1|acc |0.7679|_ |0.0074|
|hellaswag | 0|acc |0.5636|_ |0.0049|
| | |acc_norm|0.7377|_ |0.0044|
|openbookqa | 0|acc |0.3540|_ |0.0214|
| | |acc_norm|0.4420|_ |0.0222|
|piqa | 0|acc |0.7878|_ |0.0095|
| | |acc_norm|0.7851|_ |0.0096|
|winogrande | 0|acc |0.7072|_ |0.0128|
## Examples - results may vary based on temperature and other settings
```
The following is a chat between a USER and a friendly and helpful ASSISTANT.
USER: You have 5 apples. You eat 3 pears. How many apples do you have left?
ASSISTANT:You still have 5 apples because eating pears does not affect the number of apples you have.
```
```
The following is a chat between a USER and a friendly and helpful ASSISTANT.
USER: List three words that start with "ex". What is 33 + 77?
ASSISTANT:Three words starting with the letter 'e' are excited, eatable, and expensive. 33+77 equals 106.
```
```
The following is a chat between a USER and a friendly and helpful ASSISTANT.
USER: Solve for x: 5x = 3x + 10
ASSISTANT:To solve the equation 5x = 3x + 10, we can follow these steps:
Step 1: Add both sides of the equation to get a common denominator.
(1/5)x = 1 (common)
Subtract the term with "x" on both sides.
-2/5 x = 10
Step 2: To solve the resulting equation, isolate the variable "x".
First, multiply all terms by 5 to eliminate the fractions.
5x = 15
Now, divide the entire right side by 5.
x = 15/5
Finally, simplify to get the solution in its simplest form.
x = 1
So, the solution is x = 1.
```
Yup, 7B still sucks at math
| null |
Non_BioNLP
|
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
**[💵 Donate to OpenAccess AI Collective](https://github.com/sponsors/OpenAccess-AI-Collective) to help us keep building great tools and models!**
# Minotaur 7B
Minotaur 7B is an instruct fine-tuned model on top of LlaMA-13B. Minotaur 7B is fine-tuned **on only completely open datasets** making this model reproducible by anyone.
Questions, comments, feedback, looking to donate, or want to help? Reach out on our [Discord](https://discord.gg/PugNNHAF5r) or email [[email protected]](mailto:[email protected])
# Prompts
Chat only style prompts using `USER:`,`ASSISTANT:`.
<img src="https://huggingface.co/openaccess-ai-collective/minotaur-7b/resolve/main/minotaur.png" alt="minotaur" width="600" height="500"/>
# Training Datasets
Minotaur 7B model is fine-tuned on the following openly available datasets:
- [WizardLM](https://huggingface.co/datasets/ehartford/WizardLM_alpaca_evol_instruct_70k_unfiltered)
- [subset of QingyiSi/Alpaca-CoT for roleplay and CoT](https://huggingface.co/QingyiSi/Alpaca-CoT)
- [GPTeacher-General-Instruct](https://huggingface.co/datasets/teknium/GPTeacher-General-Instruct)
- [metaeval/ScienceQA_text_only](https://huggingface.co/datasets/metaeval/ScienceQA_text_only) - instruct for concise responses
- [openai/summarize_from_feedback](https://huggingface.co/datasets/openai/summarize_from_feedback) - instruct augmented tl;dr summarization
- [camel-ai/math](https://huggingface.co/datasets/camel-ai/math)
- [camel-ai/physics](https://huggingface.co/datasets/camel-ai/physics)
- [camel-ai/chemistry](https://huggingface.co/datasets/camel-ai/chemistry)
- [camel-ai/biology](https://huggingface.co/datasets/camel-ai/biology)
- [winglian/evals](https://huggingface.co/datasets/winglian/evals) - instruct augmented datasets
- custom sysnthetic datasets around misconceptions, in-context qa, jokes, N-tasks problems, and context-insensitivity
- ARC-Easy & ARC-Challenge - instruct augmented for detailed responses, derived from the `train` split
- [hellaswag](https://huggingface.co/datasets/hellaswag) - 30K+ rows of instruct augmented for detailed explanations w 30K+ rows, derived from the `train` split
- [riddle_sense](https://huggingface.co/datasets/riddle_sense) - instruct augmented, derived from the `train` split
- [gsm8k](https://huggingface.co/datasets/gsm8k) - instruct augmented, derived from the `train` split
- prose generation
# Shoutouts
Special thanks to Nanobit for helping with Axolotl and TheBloke for quantizing these models are more accessible to all.
# Demo
HF Demo in Spaces available in the [Community ChatBot Arena](https://huggingface.co/spaces/openaccess-ai-collective/rlhf-arena) under the OAAIC Chatbots tab.
## Release Notes
- https://wandb.ai/wing-lian/minotaur-13b/runs/ucgd0x6n
## Build
Minotaur was built with [Axolotl](https://github.com/OpenAccess-AI-Collective/axolotl) on 8x RTXA6000
- 1 epochs taking approximately 1.5 hours
## Bias, Risks, and Limitations
Minotaur has not been aligned to human preferences with techniques like RLHF or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so).
Minotaur was fine-tuned from the base model LLaMA 13B, please refer to its model card's Limitations Section for relevant information. (included below)
## Benchmarks
hf-causal-experimental (pretrained=openaccess-ai-collective/minotaur-7b), limit: None, provide_description: False, num_fewshot: 0, batch_size: None
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.4078|_ |0.0144|
| | |acc_norm|0.4292|_ |0.0145|
|arc_easy | 0|acc |0.7424|_ |0.0090|
| | |acc_norm|0.6877|_ |0.0095|
|boolq | 1|acc |0.7679|_ |0.0074|
|hellaswag | 0|acc |0.5636|_ |0.0049|
| | |acc_norm|0.7377|_ |0.0044|
|openbookqa | 0|acc |0.3540|_ |0.0214|
| | |acc_norm|0.4420|_ |0.0222|
|piqa | 0|acc |0.7878|_ |0.0095|
| | |acc_norm|0.7851|_ |0.0096|
|winogrande | 0|acc |0.7072|_ |0.0128|
## Examples - results may vary based on temperature and other settings
```
The following is a chat between a USER and a friendly and helpful ASSISTANT.
USER: You have 5 apples. You eat 3 pears. How many apples do you have left?
ASSISTANT:You still have 5 apples because eating pears does not affect the number of apples you have.
```
```
The following is a chat between a USER and a friendly and helpful ASSISTANT.
USER: List three words that start with "ex". What is 33 + 77?
ASSISTANT:Three words starting with the letter 'e' are excited, eatable, and expensive. 33+77 equals 106.
```
```
The following is a chat between a USER and a friendly and helpful ASSISTANT.
USER: Solve for x: 5x = 3x + 10
ASSISTANT:To solve the equation 5x = 3x + 10, we can follow these steps:
Step 1: Add both sides of the equation to get a common denominator.
(1/5)x = 1 (common)
Subtract the term with "x" on both sides.
-2/5 x = 10
Step 2: To solve the resulting equation, isolate the variable "x".
First, multiply all terms by 5 to eliminate the fractions.
5x = 15
Now, divide the entire right side by 5.
x = 15/5
Finally, simplify to get the solution in its simplest form.
x = 1
So, the solution is x = 1.
```
Yup, 7B still sucks at math
|
{"datasets": ["ehartford/WizardLM_alpaca_evol_instruct_70k_unfiltered", "QingyiSi/Alpaca-CoT", "teknium/GPTeacher-General-Instruct", "metaeval/ScienceQA_text_only", "hellaswag", "openai/summarize_from_feedback", "riddle_sense", "gsm8k", "camel-ai/math", "camel-ai/biology", "camel-ai/physics", "camel-ai/chemistry", "winglian/evals"], "license": "apache-2.0", "tags": ["OpenAccess AI Collective", "MPT", "axolotl"], "inference": false}
|
task
|
[
"SUMMARIZATION"
] | 45,255 |
abdalrahmanshahrour/arabartsummarization
|
abdalrahmanshahrour
|
summarization
|
[
"transformers",
"pytorch",
"mbart",
"text2text-generation",
"summarization",
"AraBERT",
"BERT",
"BERT2BERT",
"MSA",
"Arabic Text Summarization",
"Arabic News Title Generation",
"Arabic Paraphrasing",
"Summarization",
"generated_from_trainer",
"Transformers",
"PyTorch",
"ar",
"dataset:xlsum",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2022-12-12T16:48:00Z |
2023-01-02T18:45:12+00:00
| 87 | 6 |
---
datasets:
- xlsum
language:
- ar
license: apache-2.0
tags:
- summarization
- AraBERT
- BERT
- BERT2BERT
- MSA
- Arabic Text Summarization
- Arabic News Title Generation
- Arabic Paraphrasing
- Summarization
- generated_from_trainer
- Transformers
- PyTorch
widget:
- text: شهدت مدينة طرابلس، مساء أمس الأربعاء، احتجاجات شعبية وأعمال شغب لليوم الثالث
على التوالي، وذلك بسبب تردي الوضع المعيشي والاقتصادي. واندلعت مواجهات عنيفة وعمليات
كر وفر ما بين الجيش اللبناني والمحتجين استمرت لساعات، إثر محاولة فتح الطرقات المقطوعة،
ما أدى إلى إصابة العشرات من الطرفين.
model-index:
- name: arabartsummarization
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# arabartsummarization
## Model description
The model can be used as follows:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
from arabert.preprocess import ArabertPreprocessor
model_name="abdalrahmanshahrour/arabartsummarization"
preprocessor = ArabertPreprocessor(model_name="")
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
pipeline = pipeline("text2text-generation",model=model,tokenizer=tokenizer)
text = "شهدت مدينة طرابلس، مساء أمس الأربعاء، احتجاجات شعبية وأعمال شغب لليوم الثالث على التوالي، وذلك بسبب تردي الوضع المعيشي والاقتصادي. واندلعت مواجهات عنيفة وعمليات كر وفر ما بين الجيش اللبناني والمحتجين استمرت لساعات، إثر محاولة فتح الطرقات المقطوعة، ما أدى إلى إصابة العشرات من الطرفين."
text = preprocessor.preprocess(text)
result = pipeline(text,
pad_token_id=tokenizer.eos_token_id,
num_beams=3,
repetition_penalty=3.0,
max_length=200,
length_penalty=1.0,
no_repeat_ngram_size = 3)[0]['generated_text']
result
>>> "تجددت الاشتباكات بين الجيش اللبناني ومحتجين في مدينة طرابلس شمالي لبنان."
```
## Validation Metrics
- Loss: 2.3394
- Rouge1: 1.142
- Rouge2: 0.227
- RougeL: 1.124
- RougeLsum: 1.234
## Intended uses & limitations
More information needed
## Training and evaluation data
42.21K row in total
- Training : 37.52K rows
- Evaluation : 4.69K rows
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 2.784 | 1.0 | 9380 | 2.3820 |
| 2.4954 | 2.0 | 18760 | 2.3418 |
| 2.2223 | 3.0 | 28140 | 2.3394 |
### Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.7.1
- Tokenizers 0.13.2
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# arabartsummarization
## Model description
The model can be used as follows:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
from arabert.preprocess import ArabertPreprocessor
model_name="abdalrahmanshahrour/arabartsummarization"
preprocessor = ArabertPreprocessor(model_name="")
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
pipeline = pipeline("text2text-generation",model=model,tokenizer=tokenizer)
text = "شهدت مدينة طرابلس، مساء أمس الأربعاء، احتجاجات شعبية وأعمال شغب لليوم الثالث على التوالي، وذلك بسبب تردي الوضع المعيشي والاقتصادي. واندلعت مواجهات عنيفة وعمليات كر وفر ما بين الجيش اللبناني والمحتجين استمرت لساعات، إثر محاولة فتح الطرقات المقطوعة، ما أدى إلى إصابة العشرات من الطرفين."
text = preprocessor.preprocess(text)
result = pipeline(text,
pad_token_id=tokenizer.eos_token_id,
num_beams=3,
repetition_penalty=3.0,
max_length=200,
length_penalty=1.0,
no_repeat_ngram_size = 3)[0]['generated_text']
result
>>> "تجددت الاشتباكات بين الجيش اللبناني ومحتجين في مدينة طرابلس شمالي لبنان."
```
## Validation Metrics
- Loss: 2.3394
- Rouge1: 1.142
- Rouge2: 0.227
- RougeL: 1.124
- RougeLsum: 1.234
## Intended uses & limitations
More information needed
## Training and evaluation data
42.21K row in total
- Training : 37.52K rows
- Evaluation : 4.69K rows
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 2.784 | 1.0 | 9380 | 2.3820 |
| 2.4954 | 2.0 | 18760 | 2.3418 |
| 2.2223 | 3.0 | 28140 | 2.3394 |
### Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.7.1
- Tokenizers 0.13.2
|
{"datasets": ["xlsum"], "language": ["ar"], "license": "apache-2.0", "tags": ["summarization", "AraBERT", "BERT", "BERT2BERT", "MSA", "Arabic Text Summarization", "Arabic News Title Generation", "Arabic Paraphrasing", "Summarization", "generated_from_trainer", "Transformers", "PyTorch"], "widget": [{"text": "شهدت مدينة طرابلس، مساء أمس الأربعاء، احتجاجات شعبية وأعمال شغب لليوم الثالث على التوالي، وذلك بسبب تردي الوضع المعيشي والاقتصادي. واندلعت مواجهات عنيفة وعمليات كر وفر ما بين الجيش اللبناني والمحتجين استمرت لساعات، إثر محاولة فتح الطرقات المقطوعة، ما أدى إلى إصابة العشرات من الطرفين."}], "model-index": [{"name": "arabartsummarization", "results": []}]}
|
task
|
[
"SUMMARIZATION",
"PARAPHRASING"
] | 45,256 |
pigeon01/sungju_finetuned_ko_to_en_ver3
|
pigeon01
|
translation
|
[
"transformers",
"pytorch",
"tensorboard",
"longt5",
"text2text-generation",
"translation",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-06-05T08:26:38Z |
2023-06-05T11:10:07+00:00
| 22 | 0 |
---
license: apache-2.0
metrics:
- bleu
tags:
- translation
- generated_from_trainer
model-index:
- name: sungju_finetuned_ko_to_en_ver3
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# sungju_finetuned_ko_to_en_ver3
This model is a fine-tuned version of [KETI-AIR-Downstream/long-ke-t5-base-translation-aihub-ko2en](https://huggingface.co/KETI-AIR-Downstream/long-ke-t5-base-translation-aihub-ko2en) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8656
- Bleu: 39.0667
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
### Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# sungju_finetuned_ko_to_en_ver3
This model is a fine-tuned version of [KETI-AIR-Downstream/long-ke-t5-base-translation-aihub-ko2en](https://huggingface.co/KETI-AIR-Downstream/long-ke-t5-base-translation-aihub-ko2en) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8656
- Bleu: 39.0667
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
### Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
|
{"license": "apache-2.0", "metrics": ["bleu"], "tags": ["translation", "generated_from_trainer"], "model-index": [{"name": "sungju_finetuned_ko_to_en_ver3", "results": []}]}
|
task
|
[
"TRANSLATION"
] | 45,257 |
seongil-dn/bge-m3-kor-retrieval-451949-bs64-news
|
seongil-dn
|
sentence-similarity
|
[
"sentence-transformers",
"safetensors",
"xlm-roberta",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:451949",
"loss:CachedMultipleNegativesRankingLoss",
"arxiv:1908.10084",
"arxiv:2101.06983",
"base_model:BAAI/bge-m3",
"base_model:finetune:BAAI/bge-m3",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] | 2024-12-11T06:31:16Z |
2024-12-11T06:32:38+00:00
| 5 | 0 |
---
base_model: BAAI/bge-m3
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:451949
- loss:CachedMultipleNegativesRankingLoss
widget:
- source_sentence: 어느 정도의 인원을 대상으로 25일에 도착한 예방주사를 접종할 수 있어?
sentences:
- “길고 길었던 ‘코로나19 터널’을 빠져 나오는 신호탄이 되길 간절히 기원합니다.” 코로나19 종식 염원을 담은 아스트라제네카(AZ) 백신이
26일부터 접종에 들어간다. 접종이 하루 앞으로 다가오면서 일선 보건당국도 비상 근무체제에 돌입했다. 25일 낮 12시10분께 AZ 백신을
실은 운반차량이 군경의 호위 속에 안양시 동안구보건소 앞에 도착하자, 보건소 직원들이 건물 앞까지 나와 백신을 맞았다. ‘생물학적 제제 출하증명서’가
오가고 아이스박스가 열리자, 투명한 5밀리리터 통 위에 빨간 캡이 씌워진 AZ가 그 모습을 드러냈다. 보건소 직원들은 조심스럽게 백신을 하나
하나 들추며 혹시라도 깨지거나 파손된 물건은 없는지 상태를 살폈다. 점검을 마친 백신이 곧장 보건소 3층에 마련된 전용 냉장고로 들어가면서
이날 백신 상륙 작전은 약 5분 만에 마무리됐다. 이날 들어온 1차 물량은 30바이알, 300명분이다. 한숨 돌릴 틈도 없이 입고식을 마친
직원들은 보건소장실에 모여 ‘예방접종 후 이상반응 관리’ 교육을 받았다. 26일 오전 9시부터는 동안구 9개 요양시설 종사자 중 일부인 1개소,
20명이 보건소를 방문해 첫 접종을 받을 예정이다. 동안구보건소 관계자는 “접종을 차질 없이 진행하기 위해 만반의 준비를 마쳤다”면서 “가장
중요한 백신을 보관하는 전용 냉장고는 영상 2~8도를 항시 유지하고 있으며 정전에도 대비하기 위해 무정전 전원장치(UPS)까지 갖춰져 있다”고
말했다. 이와 함께 수원시 영통구보건소에도 20바이알의 AZ 백신이 무사히 도착했다. 보건소 1층에 마련된 접종실엔 안내문이 비치됐고, 직원들은
예진표 작성부터 신분증과 체온 확인, 예진 후 접종, 이상반응 모니터링, 확인증 발급까지 방문자가 거쳐야 할 동선을 한 번 더 꼼꼼히 점검했다.
영통구보건소 관계자는 “소방서와 연계한 긴급 출동 체계가 마련돼 있다”면서 “만일의 상황을 대비해 자동제세동기(AED) 점검을 마쳤으며 응급키트도
구비해놨다”고 밝혔다.
- 사법농단 의혹 재판에서 6연속 무죄를 끊고 ‘1호 유죄’가 나온 배경에는 “대법원장과 법원행정처는 특정 재판에서 명백한 잘못이 있을 경우 판사
지적권한이 있다”는 재판부의 독특한 판시가 있다. 앞서 임성근 전 부산고법 부장판사가 1심에서 ‘위헌적이지만, 남용할 직권 자체가 없었다’며
무죄 선고를 받은 것과는 상반된다. 서울중앙지법 형사32부(부장판사 윤종섭)는 23일 대법원장과 법원행정처에는 재판사무의 핵심영역에 대해 명백한
잘못이 있을 경우 담당 판사의 잘못을 ‘지적’할 수 있는 권한이 있다고 밝혔다. 이민걸 전 법원행정처 기획조정실장과 이규진 전 대법원 양형위원회
상임위원의 일부 재판개입 등 행위(직권남용)를 유죄 판단하면서다. 사법농단 관련 재판에서는 처음 나온 판시다. 재판부는 전제로 “직업적으로
충분히 단련하지 못하거나 나태해 쟁점이 별로 없는 것만 우선 처리해 장기미제를 만드는 판사가 있을 수 있다”고 말했다. 그러면서 “이런 경우에도
재판 독립을 이유로 누구도 지적을 못한다면 국민의 재판청구권 보장과 충돌할 수 있다”고 했다. 헌법 103조에 따른 법관 독립이 ‘신성불가침’은
아니라는 얘기다. 재판부는 이를 뒷받침할 명시적 규정은 없지만 헌법과 법원조직법 해석상 명백한 잘못을 전제로 사법행정사무를 총괄하는 대법원장과
행정처에 판사를 지적할 권한이 있다고 봤다. 선뜻 이해하기 어려운 이 법리는 경계가 애매모호한 직권남용죄의 특성 때문에 도출된 것이다. 직권남용은
‘남용될 직권’의 존재를 전제한다. 표면상 이유로는 수긍할 수 있지만 실제로 따져보면 위법하다는 까다로운 조건을 만족해야 한다. 임 전 부장판사의
경우 “각 재판관여행위가 위헌적”이라는 판단에도 불구하고 “(서울중앙지법 형사수석부장판사 당시의) 일반적 직무권한에 속한다고 해석될 여지가
없다”며 1심에서 무죄를 선고 받았다. 아울러 재판부는 “공무원의 직무범위를 벗어난 ‘월권행위’도 직무와 상당한 관련성이 인정된다면 직권남용이
성립될 수 있다”며 직권남용 범위를 상대적으로 넓게 판단했다. 직권의 범주를 형식적으로만 봐선 안 된다는 취지다. 이 전 실장 등은 이를 엄격하게
판단해야 한다며 검찰과 치열한 공방을 벌였지만 1심에서 받아들여지지 않았다. 재판부는 이 같은 법리에 따라 헌법재판소 파견 법관을 통한 내부
기밀 수집, 한정위헌 취지의 위헌제청 결정 사건과 옛 통합진보당 국회의원·비례대표 지방의회의원 관련 행정소송에 대한 재판개입 등에 대한 일부
직권남용 혐의를 유죄로 판단했다. 검찰은 이번 재판을 양승태 전 대법원장 등 ‘수뇌부’를 단죄할 여지를 열어줄 계기로 보고 있다. 검찰 관계자는
“재판부가 앞선 사법농단 관련 재판과 달리 직권남용 범위를 넓게 봤다”며 “사법부 윗선의 유죄 가능성을 높인 판결”이라고 평가했다.
- '65세 이상 73%, 6-59개월 54% 인플루엔자 접종 마쳐, 가능하면 11월 15일 이전 완료하세요~ □ 또한, 남은 물량은 충분하나,
일시적으로 지역적 부족상황이 발생할 수 있어 그에 대한 백신수급 운영 대책도 준비하고 있다고 덧붙였다. ○ 올해 처음으로 부족한 공급에 신속히
대처하기 위해 시도 및 질병관리본부에서 직접 공급할 수 있는 추가 물량*을 확보하였으며, 상황 발생 시 신속히 조치할 예정이다. * 추가확보물량:
시·도0.25㎖5,220 도즈, 0.5㎖80,680 도즈/ 본부0.25㎖40,000 도즈, 0.5㎖40,000 도즈
□ *** 예방접종관리과장은 “아직까지 접종 받지 않은 어르신 및 어린이 보호자들은 지정의료기관 방문 전에 꼭 관할 보건소, 질병관리본부 콜센터(☎1339)
및 보건복지콜센터(☎129)로 문의하여 당일 접종이 가능한 의료기관을 안내받거나, 예방접종 도우미 홈페이지 및 스마트폰 앱*에서 확인 후 방문”
해달라고 당부했다. * 질병관리본부 예방접종도우미 홈페이지((https://nip.cdc.go.kr) 접속> 일반인(자세히 보기)> 예방접종도우미
연결), 예방접종도우미 스마트폰 앱의 의료기관 검색
○ 또한, “인플루엔자 예방접종의 효과는 균주, 유행 특성 등에 따라 건강한 성인에서 70-90%, 소아에서 50-95% 범위 내로 다양한
것으로 알려져 있어,
- 30초 이상 손씻기, 기침이나 재채기를 할 때 휴지나 옷깃으로 입을 가리는 기침예절 준수 등의 예방 수칙 실천을 병행하고,
- 기침, 발열, 근육통 등 인플루엔자 의심 증상이 발생할 경우 사람이 많이 모인 곳을 피하고 의료기관을 늦지 않게 방문하여 진료받는 것이
중요하다”며 덧붙였다. ○ 아울러, “무료접종 대상자 외에 임신부, 만성질환자, 50~64세 인구 등의 인플루엔자 고위험군도 감염 예방을 위해
본격적인 인플루엔자 유행 전에 예방접종이 필요하다”고 밝혔다.'
- source_sentence: 길어지는 코로나를 준비하기 위해 조선대학교가 시행하고 있는 과정은 뭐야?
sentences:
- 조선대학교병원(병원장 정종훈)은 코로나19 장기화 및 재확산 대비를 위한 중증환자 전담 간호사 양성 교육을 시작했다고 2일 밝혔다. 보건복지부와
국립중앙의료원이 주최하고 조선대병원이 주관하는 이번 양성 교육 과정은 체계적인 간호교육 시스템을 통해 중증 환자 담당 간호사 양성을 목표로
하고 있다. 지난 1일 병원 7층 세미나A실에서 시작된 '중증환자 전담 간호사' 양성 교육은 4주간 진행되며 교육 내용은 중환자 간호 이론교육
및 모의실습과 중환자실 임상실습으로 구성된다. 이론교육 및 모의실습에는 △중환자의 신체사정 및 기본간호 △호흡기계 중재 및 간호 △특수처치
및 응급간호 등 중환자 간호 공통교육 이론강의 등을 비롯해 개인보호구 착탈의 방법 등이 포함됐다. 중환자실 실습을 통해서는 실제 중환자실 현장에서
개인별로 1:1 프리셉터가 배정되어 습득한 내용을 바탕으로 밀착 교육을 진행한다. 유윤정 조선대병원 간호부장은 "코로나19 장기화 및 재확산
대비를 위해 진행하는 이번 양성 교육을 통해 전문성 있는 중환자 전담 인력을 양성하고 간호역량을 강화할 수 있을 것으로 보인다"며 "중환자
간호교육 인프라와 인적 자원 확보를 위해 최선을 다하겠다"고 밝혔다.
- 조선대학교는 지난 25일 총장실에서 우수 행정서비스 운영 사례를 발굴해 구성원들에게 공유, 확산시키고자 '대학행정서비스 우수사례 및 인증 시상식'을
진행했다고 30일 밝혔다. 지난 2019년부터 부서컨설팅을 수행 중인 조선대는 올해 18개 부서(누적 45개)에 대한 컨설팅을 수행한다. 5년
주기로 대학 내 전 부서에 대한 컨설팅을 진행하는 것이 목표다. 조선대가 지난 2월 수행한 대학행정서비스 컨설팅 결과, 교수학습지원센터의 'E-Learning지원팀'이
코로나19에 적극적으로 대응한 우수사례로 선정됐다. E-Learning지원팀은 코로나19 속 온라인수업에 대한 수요 및 요구사항을 수집해 반영했다.
또 온라인수업 콘텐츠를 제작해 수시로 안내하고 우수콘텐츠 공모전 시행으로 우수사례를 공유해 온라인 수업의 전반적인 질을 향상시키는 데 기여했다.
또 온라인시험 및 수업 운영 모니터링단을 꾸려 수업의 질 관리에 적극 노력했다. 조선대학교는 E-Learning지원팀이 진행한 온라인 수업의
질 관리 결과를 교육혁신원 홈페이지 및 업무포털을 활용해 '2020학년도 대학 교육행정 서비스 CQI를 위한 부서 컨설팅 결과 보고서'로 공유할
예정이다.
- 김천상무가 최하위 부천과 원정 경기서 승점 1점을 보태는 데 그쳐 순위상승 기회를 놓쳤다. 김천은 지난 8일 부천종합운동장에서 열린 ‘하나원큐
K리그2 2021’ 11R 부천FC와의 경기에서 득점없이 0-0무승부를 기록했다. 김천은 조규성-오현규-강지훈이 쓰리톱, 심상민-정현철-박용우-안태현
중원, 이명재-정승현-하창래가 쓰리백에 나서는 3-4-3전술로 부천을 맞았다. 특히 이날 골키퍼에는 구성윤이 처음으로 장갑을 꼈다. 양팀은
경기 휘슬과 함께 치열한 중원다툼에 나서 전반 7분 부천 안재준이 먼저 날카로운 슛을 날리자 김천도 오현규가 12분과 18분 잇따라 위협적인
슛으로 맞받아쳤다. 오현규의 슛으로 조금씩 공세의 강도를 높이기 시작한 김천은 32분 이명재가 회심의 슛을 쐈으나 골대를 살짝 넘어갔으며,
39분에는 박용우가 프리킥 슛을 쐈지만 부천 수비수 강의빈에 막히면서 전반을 0-0으로 마쳤다. 김천은 후반들어서도 공격이 여의치 않자 7분
강지훈 대신 정재를 투입해 측면 공세를 높였고, 11분과 15분 안태현과 정재희 잇따라 부천 골문을 노렸지만 무위에 그쳤다. 좀처럼 득점 찬스를
만들지 못하던 김천은 31분 부천 한지호에게 골을 허용했으나 VAR 판독으로 무효가 되면서 위기를 넘겼다. 김천은 후반 44분 이명재가 결정적인
문전 슈팅을 날렸지만 부천 골키퍼 전종혁의 선방에 막히는 등 막판 파상적인 공세를 펼쳤지만 득점을 만들지 못했다.
- source_sentence: 반도체 양성에 대한 정부의 후원은 언제 끝날 예정이야?
sentences:
- 반도체 산업의 정치적 요인을 분석한 결과, 초기에는 반도체 산업의 활성화를 위해 대형 국책사업 추진 등 전폭적인 지원사업이 이루어져왔으나,
현재 반도체와 관련된 예산 등 정부지원은 상대적으로 감소하고 있는 것으로 나타났다 (한국산업기술진흥원 2017). 그러나 정부는 반도체 산업의
성장을 지원하기위해 다각적인 측면에서 노력 중에 있다. 관련 업종별 ․ 수준별 특성을 고려한 차별화된 스마트공장 지원 강화를 통하여 뿌리 등
16개 등의 업종별 ‘대표 스마트공장’(30개, ~’22년) 및 고도화를 통한 스마트공장 프리미엄 5,000개(’22년) 구축을 추진하고자
하였으며, SW ․ 제어기기 등의 스마트 제조 플랫폼 개발을 통하여 반도체 산업의 재도약을 추진하고 있다(산업통상자원부 2017). 뿐만 아니라,
지능형반도체산업과 융합 가능한 임베디드 SW 산업 발전을 위해 다양한 정책 사업을 추진하고 있다(한국산업기술진흥원 2017).
- 정부가 인공지능(AI), 전기차 등에 들어가는 핵심 부품인 차세대 전력 반도체 육성에 본격적으로 나선다. 2025년까지 차세대 전력 반도체
상용화 제품을 5개 이상 개발하고, 6∼8인치 파운드리(위탁생산) 인프라 구축도 추진한다. 정부는 1일 제7차 혁신성장 빅3 추진회의를 열고
이런 내용을 담은 '차세대 전력 반도체 기술개발 및 생산역량 확충 방안'을 발표했다. 전력 반도체란 전력의 변환·변압·안정·분배·제어를 수행하는
반도체를 말하며, 차세대 전력 반도체는 실리콘보다 전력 효율과 내구성이 뛰어난 3대 신소재 웨이퍼로 제작된 반도체를 뜻한다. 인공지능(AI),
5G나 고온·고압 내구성이 필요한 전기차, 신재생에너지 발전의 인버터 등에 적용되는 필수 부품이다. 앞으로 전자기기 수요가 확대되고 전력 소비가
늘 것으로 예상되면서 미래 성장 가능성이 높은 분야로 꼽힌다. 산업부에 따르면 글로벌 전력 반도체 시장 규모는 2019년 450억 달러에서
2023년 530억 달러 규모로 성장할 것으로 전망됐다. 이에 따라 미국과 중국, 일본 등도 다양한 정책적 지원을 추진 중이다. 우리 정부도
2017년부터 총 830억 원 규모의 '파워반도체 상용화 사업'을 진행 중이다. 그러나 국내 전력 반도체 시장 규모는 약 20억 달러 규모로,
기술력 부족과 해외 기업의 특허 선점으로 수요의 90% 이상을 수입에 의존하는 실정이다. 이에 정부는 2025년까지 글로벌 수준의 경쟁력을
확보하기 위해 상용화 제품 개발, 기반 기술 강화, 미래 제조공정 확보 등을 본격적으로 지원하기로 했다. 국내 수요기업과 연계한 과제를 발굴해
차세대 전력 반도체 상용화 제품을 5개 이상 개발할 방침이다. 현재 실리콘 소재의 한계를 극복하기 위해 실리콘카바이드(SiC), 질화갈륨(GaN),
갈륨 옥사이드(Ga2O3) 등 3대 신소재 응용기술을 개발하고, 국내 기업의 소재·웨이퍼 기술 확보를 지원할 방침이다. 고집적·고성능 차세대
전력 반도체 개발을 위해 파워 집적회로(IC) 설계 기술개발을 지원한다. 아울러 차세대 전력 반도체 관련 파운드리 서비스를 확대하기 위해 국내
파운드리와 6∼8인치 기반의 양산 공정을 구축하는 방안도 추진한다. 성윤모 산업부 장관은 "AI, 5G 등 신기술 구현과 자율차, 신재생 등
미래 성장 분야의 활성화를 위해선 전력의 효율적인 관리가 필수이며, 차세대 전력 반도체는 이를 위한 핵심 부품"이라며 "아직 초기인 이 분야
시장을 선점하고 미래 경쟁력을 확보하기 위해 R＆D, 인프라 등을 적극 지원할 계획"이라고 말했다.
- 근로시간은 짧지만 국민 소득은 높은 국가들을 분석해 보니 노동 유연성과 생산성이 높다는 연구결과가 나왔다. 전국경제인연합회 산하 한국경제연구원은
경제협력개발기구(OECD) 통계와 통계청 데이터를 이용해 우리나라와 덴마크ㆍ노르웨이ㆍ독일ㆍ네덜란드 4개국의 노동생산성 등을 비교한 결과 이같이
나타났다고 29일 밝혔다. 한경연에 따르면 이들 국가의 평균 연간 근로시간은 1,396시간이었고, 평균 1인당 국민총소득은 6만187달러로
집계됐다. 반면 한국은 이들 국가보다 1.4배(1,967시간) 더 길게 일했지만, 소득은 3만2,115달러로 절반 수준이었다. 한경연은 이들
유럽 4국의 특징을 △높은 고용률 △높은 노동생산성 △높은 노동유연성 △시간제 근로 활성화 △높은 수준의 인적자원이라고 정리했다. 이들 국가의
평균 고용률은 76.4%로 한국(66.8%)보다 9.6%포인트 높았다. 특히 네덜란드와는 11.4%포인트 차이가 났다. 우리나라가 네덜란드
수준의 고용률을 달성하려면 418만6,000명의 일자리가 더 만들어져야 한다고 한경연은 분석했다. 시간당 노동생산성도 한국(40.5달러)은
이들 국가 평균(73.3달러)에 크게 못 미쳤다. 노르웨이는 84.3달러로 한국보다 2배 이상 높았다. 세계경제포럼(WEF) 노동시장 유연성
평가에서도 한국 점수는 크게 낮았다. 한국은 54.1점을 받아 OECD 37개국 중 35위에 그친 반면 4개 국가의 평균 점수는 68.9점이었다.
가장 높은 덴마크(71.4점)는 OECD 3위에 올랐다. 아울러 이들 국가는 시간제 근로 비율이 높았는데 네덜란드는 37.0%에 달해 한국(14.0%)보다
2.6배 높은 것으로 나타났다. 한경연은 이들 국가가 적게 일하면서 많이 벌 수 있었던 비결은 노동시장 개혁을 통해 노동 유연성을 확보했기
때문이라고 분석했다. 추광호 한경연 경제정책실장은 “적게 일하고 많이 버는 국가들은 시간제 근로 활성화, 노동 규제 완화 등을 통해 고용률을
높이고, 높은 생산성을 토대로 소득을 높였다”고 말했다. 한편 한경연이 우리나라 산업별 소득과 근로시간을 비교한 결과 적게 일하고 많이 버는
산업은 금융 및 보험업으로 나타났다. 한경연에 따르면 금융 및 보험업은 전체 월 임금 총액인 372만원보다 1.6배 높은 593만원을 벌었지만,
월 근로시간은 9시간 짧았다. 반면 숙박 및 음식점업은 이보다 적은 234만원을 벌지만 12시간 더 일하는 것으로 나타났다.
- source_sentence: E-GMP의 세번째 배터리 공급 회사로 어느 곳이 선발됐어?
sentences:
- 손흥민(손흥민)의 골이 30일째 침묵한 토트넘이 시즌 첫 3연패에 빠지며 8위로 추락했다. 토트넘은 5일(한국시간) 영국 런던의 토트넘 홋스퍼
스타디움에서 열린 첼시와 2020-2021 잉글랜드 프로축구 프리미어리그 22라운드 홈경기에서 전반 24분 조르지뉴에게 내준 페널티킥 결승
골을 만회하지 못하고 0-1로 패했다. 해리 케인의 부상 결장에 손흥민 골침묵까지 이어진 토트넘은 이번 시즌 첫 3연패이자 홈 경기 2연패로
승점 33에 머물면서 8위로 한 계단 내려앉았다. 지난달 6일 브렌트퍼드(2부)와 EFL컵 준결승전에서 득점을 맛본 손흥민은 이후 30일째
득점 신고를 하지 못하며 6경기(정규리그 5경기ㆍFA컵 1경기) 연속 무득점 행진을 이어갔다. 손흥민의 이번 시즌 공격포인트는 16골 10도움(EPL
12골 6도움ㆍ유로파리그 3골 3도움·리그컵 1골ㆍFA컵 1도움)이다. 아울러 통계전문 옵타에 따르면 토트넘의 조제 모리뉴 감독은 2000년
9월 벤피카(포르투갈)에서 감독직을 시작한 이후 자신의 사령탑 커리어 통산 327번째 홈 경기에서 처음으로 ‘홈경기 2연패’를 당하는 아쉬움까지
곱씹어야 했다. 반면 첼시는 승점 36(골 득실+13)으로 에버턴(승점 36ㆍ골 득실+6)을 골득실차로 따돌리고 8위에서 6위로 올라섰다.
- ‘포항 배터리 규제자유특구’가 중소벤처기업부의 2020년 규제자유특구 운영성과 평가에서 전국 유일 2년 연속 최고등급인 ‘우수’ 특구로 21일
선정됐다. 이번 중소벤처기업부의 규제자유특구 운영성과 평가는 전국 14개 규제자유특구를 대상으로 성과 달성도, 파급효과, 규제특례 활용실적
등을 토대로 종합평가를 실시됐으며, 포항 배터리·대구 스마트웰니스·전북 친환경자동차 규제자유특구 3개 특구가 우수 평가를 받았다. 2019년
7월 영일만 일반산업단지 및 블루밸리 국가산업단지 일원(556,694.22㎡, 17만평)을 배터리 규제자유특구로 지정받은 이후 포항시는 배터리
전후방 기업들에 대한 적극적인 투자유치 활동으로 대규모 투자를 이끌어냈다. GS건설의 배터리 리사이클링 공장 건립, 포스코케미칼의 이차전지
인조흑연 음극재 생산공장 건립, 에코프로비엠 등 에코프로 6개 자회사의 양극재 생산공장 건립 등의 투자로 이차전지 밸류체인 구축과 이를 바탕으로
배터리 규제자유특구를 미래 유망산업인 배터리산업 육성의 거점으로서의 기반을 마련한 것이 높은 평가를 받았다. 뿐만 아니라, 규제자유특구 실증
연구개발 수행과 전기차 배터리 산업 육성 전초 기지 역할을 감당할 이차전지종합관리센터를 107억 원의 사업비를 투입해 블루밸리 국가산업단지
규제자유특구 내 건립 하는 등 특구 활성화를 위한 지자체의 적극적인 의지도 높은 평가를 받았다. 이차전지종합관리센터는 이차전지 고도화와 고부가
가치화를 통해 배터리 산업을 체계적으로 육성하기 위한 국내 최초, 최대 규모의 인프라이며, 이곳을 배터리 산업 허브로 육성해 배터리 소재,
부품, 장비 상용화 기술개발 및 우수 기업 유치 등의 마중물 역할로 지역경제 활성화에 크게 기여할 것이다. 더 나아가 포항시는 국가 전기차
사용후 배터리 순환경제 기반 조성 및 배터리 산업 컨트롤 타워 구축을 목표로 추진하고 있는 환경부의 ‘배터리 자원순환 클러스터 구축사업’에도
참여해 총력을 기울이고 있다. 포항시는 철강산업의 뒤를 이어 지역의 새로운 먹거리 산업으로 이차전지산업을 육성하기 위해 중소벤처기업부, 환경부
등 중앙 부처와 긴밀한 협력체계를 유지하면서 이차전지분야 기술개발, 기술실증, 기업육성의 전주기적 기업지원 인프라 구축으로 포항을 우리나라의
배터리 메카도시로 육성해나갈 계획이다. 이강덕 포항시장은 “중앙부처뿐만 아니라 배터리소재기업인 포스코케미칼과 에코프로, GS건설 등과 적극적인
협업으로 이차전지산업 육성을 통해 포항의 발전를 넘어 우리나라 배터리산업 육성에도 큰 기여를 할 수 있도록 행정력을 집중하겠다”라고 전했다.
- 중국 전기차용 배터리 업체들이 탈중국화에 속도를 내며 ‘K배터리’를 위협하고 있다. 텃밭인 한국과 유럽 시장에서도 K배터리는 ‘C배터리’(중국산
배터리)의 거센 도전에 직면했다. 21일 업계에 따르면 중국 배터리 선두 업체인 CATL은 현대자동차그룹이 최근 선정한 전기차 전용 플랫폼
E-GMP 3차 물량 배터리 공급사로 낙찰됐다. 현대차그룹이 2023년 이후 출시 예정인 3개 차종에 대한 배터리 공급사를 선정했는데 SK이노베이션이
1종, CATL이 2종을 따낸 것이다. 총 9조원 규모 물량 중 절반 이상을 CATL이 차지한 것으로 알려졌다. 업계 관계자는 “CATL이
국내 시장까지 파고든 것은 가격·성능·생산력 측면에서 K배터리와 견줘 충분한 경쟁력을 지니고 있다는 뜻”이라며 “안정적 수급을 위해 배터리
공급사를 다변화해야하는 현대차 입장에서 스펙이 맞지 않는 삼성SDI, 코나EV 등 화재로 안전 이슈가 불거진 LG에너지솔루션의 상황을 감안하면
CATL을 선택할 수밖에 없었을 것”이라고 말했다. 다만 당초 3차 물량의 핵심으로 꼽혔던 스포츠유틸리티(SUV) 전기차 ‘아이오닉7’에 탑재될
배터리 선정은 미뤄졌다. 업계에서는 현대차와 LG에너지솔루션이 인도네시아에 건설을 추진 중인 합작법인이 아이오닉7에 탑재할 배터리를 생산할
것이라는 관측이 우세하다. C배터리의 영토 확장은 한국 시장을 넘어 유럽을 정조준하고 있다. 메르세데스 벤츠를 생산하는 독일 완성차 업체 다임러
그룹은 지난해 CATL과 배터리 공급은 물론 연구 분야에서도 협력을 강화키로 했다. 업계에 따르면 양사는 향후 다임러 그룹에서 생산하는 모든
전기차에 사용되는 배터리를 CATL에 우선 발주하고 CATL이 수주를 거부하는 경우에만 다른 배터리 업체를 통해 공급받는 형태의 계약을 체결한
것으로 알려졌다. 이에 대해 업계 관계자는 “차종별로 배터리 업체를 선정하는 기존 방식에서 벗어난 파격적인 형태”라며 “말 그대로 CATL이
‘수퍼 을’(갑을 뛰어넘는 영향력을 지닌 을)로 성장했음을 보여주는 대목”이라고 말했다. ‘세계의 자동차 시장’이라 불리는 중국의 막대한 전기차
생산량에다 정부의 전폭적인 지원을 바탕으로 성장한 CATL은 지난해 비중국 시장 점유율을 대폭 늘렸다. 배터리 시장조사 기관인 SNE리서치에
따르면 2019년 중국을 제외한 글로벌 전기차 시장에서 CATL 배터리 사용량은 0.2GWh(기가와트시)로 LG에너지솔루션(12.3GWh),
삼성SDI(4.3GWh), SK이노베이션(2.1GWh)에 비해 초라한 수준이었다. 하지만 지난해 5.3GWh로 무려 2,457.1%의 성장률을
기록했다. 점유율 역시 0.4%에서 6.5%로 껑충 뛰었다. CATL은 또 상상을 초월하는 생산능력을 갖춰 전기차 배터리 시장의 최상위 포식자로
군림하겠다는 계획을 세우고 있다. CATL이 현재 국내·외에서 진행 중인 배터리 공장 건설 프로젝트를 종합하면 자체 생산과 합작사를 합쳐 2025년엔
486.8GWh, 2030년엔 591.8GWh의 생산능력을 보유하게 된다. 2020년 글로벌 시장 전체 전기차용 배터리 사용량이 142.8GWh인
점을 감안하면 어마어마한 규모다. 업계 관계자는 “올해 예정된 글로벌 자동차 회사들의 배터리 공급사 입찰 규모는 약 1.4TWh(테라와트시)로
지난해 판매된 전기차 총 배터리 용량의 10배에 달한다”며 “수주전 결과에 따라 배터리 업체들의 향후 수년간 순위표에 지각 변동이 생길 수
있다는 뜻”이라고 말했다. 이어 “전기차 시장을 선도하는 유럽까지 C배터리가 침투한 만큼 K배터리는 신흥 시장이자 중국의 진출이 제한되는 미국
시장을 선점할 필요가 있다”면서 “LG와 SK가 빠른 시일 내에 합의를 하고 선의의 경쟁을 통해 K배터리의 경쟁력을 키워야 할 때”라고 강조했다.
- source_sentence: 지역 예술인을 위한 공모를 시행했던 프로젝트는 어떤 것에 일부분이야?
sentences:
- 충남 공주시는 제민천과 국고개 일원을 대상으로 지난해 8월부터 추진한 공공미술 프로젝트 사업이 완료됐다고 21일 밝혔다. 문화체육관광부가 주최
공주시가 주관한 공공미술 프로젝트는 지역예술인 일자리 제공 및 주민 문화 향유 증진이라는 취지로 전국 228개 지자체 대상으로 추진됐다. 이에
시는 지난해 8월 공모를 통해 한국미술협회(공주지부)와 충남공예협동조합 2개팀을 선정 지역작가 37명의 참여를 통해 진행해 '중동교~대통교'사이에
조명과 벤치, 포토존 그리고 주민참여로 만들어진 송사리 조형물을 설치해 주민들에게 휴식 공간을 제공했다. 그리고 '국고개 흐르는 효심'은 도조작품,
화분조형, 이복이야기 동화책 및 국고개 가이드북 만들기 등 커뮤니티 프로그램 운영 역사.문화.예술 중심지의 특성을 살리는데 중점을 뒀다. 황의정
문화체육과장은 "앞으로도 시민들이 언제 어디서나 문화예술을 누리고, 지역예술인에게는 일자리를 제공할 수 있는 다양한 사업을 진행하겠다"고 말했다.
- 배우 신성록이 여권 사진을 공개해 시선을 모았다. 18일 오후 방송된 SBS 예능 프로그램 ‘집사부일체’에는 김종국이 사부로 출연했다. 사부를
만나기 전, 출연진은 가고 싶었던 바캉스 장소에 대해 이야기했다. 차은우는 “하와이에 가고 싶다”고 했고, 이승기는 “제주도 여행이 간절하다”고
말했다. 양세형은 “나도 제주도”라고 외쳤다. 양세형은 “은우의 여권 사진이 궁금하다”며 호기심을 드러냈다. 차은우의 여권 사진이 공개되자,
이승기는 “프로필 사진이냐”며 감탄했다. 양세형은 “모델 같다”고 말했다. 신성록은 “동네 사진관에서 여권 사진을 찍었다. 보정이 심해서 AI
같다”며 자신의 사진을 공개했다. 그의 여권 사진을 본 양세형은 “실물이 더 잘생겼다”고 말했다. 한편 ‘집사부일체’는 매주 일요일 오후 6시
25분에 방송된다.
- 마산의 원도심 성호동의 역사가 예술로 기록되고 되살아난다. 시대를 살아낸 사람들의 애틋한 이야기가 아름다운 소리, 몸짓, 조형, 화첩에 담겨
시민을 만난다. 창원시와 창원시문화도시지원센터는 오는 30일 오후 2시 마산합포구 문신길 일대에서 성호동과 그 주변 지역주민들의 삶, 지리적,
환경적 문화를 예술에 담은 〈사회참여예술창작지원사업〉 ‘예술로 성호창발(城湖創發)’ 창작 발표회를 갖는다고 밝혔다. 문화특화지역조성사업의 일환으로
진행하는 이 사업은 생성과 소멸의 과정을 거쳐 성장해온 도심 속 흩어져 있는 다양한 문화를 예술로 기록하고 후대에 물려주기 위한 취지로 순수예술
창작물을 구현하고 예술의 사회적 책임에 관해 고찰해 보는 사업이다. 앞으로 지역 예술인들과 뜻을 모아 소중한 원도심의 문화가 한순간 덧없이
사라지는 일이 없도록 매년 확대해 나갈 계획이다. 성호동을 대상으로 올해 처음 시행하는 ‘예술로 성호창발(城湖創發)’ 사업은 지난해 10월
지역 예술인들을 대상으로 참가자를 공개 모집한 결과, 사업 취지를 이해하고 창작 능력과 작품의 완성도가 기대되는 4개 단체를 선정하였다. 선정된
4개 단체 작가들은 스토리 개발을 위해 성호동 주민들을 만나 세밀한 대화를 나누기도 하고, 장르와 특성에 맞는 창작품을 만들기 위해 성호동
구석구석을 누비는 열정을 보였다. 발표회는 1월 30일 오후 2시부터 4시 30분까지 ‘성호동 이바구 투어’라는 이름으로 마산박물관 앞에 모여
문신 길을 따라 추산창작공작소 4933을 거쳐 임항선 그린웨이, 그리고 성호동 생활문화센터를 지나 정법사로 이동하면서 펼쳐진다. 〈(사)경남국악관현악단
‘휴’〉가 ‘추산창작공작소 4933’에서 동네 이야기를 판소리로 엮어 국악 관현악과 경서도 소리를 통해 성호동의 옛이야기를 펼치고, 는 ‘성호동의
역사와 추억을 담다’란 주제로 도시가 발달하면서 퇴색되어가는 성호동 동네잔치의 상징적 의미를 담은 조형물을 설치했다. 〈신미란 작가〉는 ‘성호동
생활문화센터 2층 전시장’에서 성호동 사람과 삶의 이야기(話)와 그림(畵)을 표현한 ‘성호화첩(成湖話畵)’을 발표하며, 〈박은혜 춤패〉는 ‘일란
이필이의 춤을 찾아가는 소소한 춤 여행’이란 주제로 이필이 선생의 춤을 추며 다큐멘터리 시사회를 가지는 것으로 전체 일정을 마친다. 공식 행사
외의 진행은 성호동 토박이 김경년 창원시도시재생지원센터 해설사가 맡아 행사장을 이동하는 중간중간 맛깔나는 이바구로 성호동의 역사를 들려준다.
심재욱 창원시 문화체육관광국장은 “도시는 사람의 역사를 품은 곳으로 이제 사람이 도시를 품어야 문화예술로 지속 성장의 가능성을 보여줄 수 있다”라며
문화의 힘을 강조했다. 한편 행사 참가는 사회적 거리두기를 위해 참석 인원을 20명으로 제한하여 운영한다.
---
# SentenceTransformer based on BAAI/bge-m3
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) <!-- at revision 5617a9f61b028005a4858fdac845db406aefb181 -->
- **Maximum Sequence Length:** 1024 tokens
- **Output Dimensionality:** 1024 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 1024, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("seongil-dn/bge-m3-kor-retrieval-451949-bs64-news")
# Run inference
sentences = [
'지역 예술인을 위한 공모를 시행했던 프로젝트는 어떤 것에 일부분이야?',
'마산의 원도심 성호동의 역사가 예술로 기록되고 되살아난다. 시대를 살아낸 사람들의 애틋한 이야기가 아름다운 소리, 몸짓, 조형, 화첩에 담겨 시민을 만난다. 창원시와 창원시문화도시지원센터는 오는 30일 오후 2시 마산합포구 문신길 일대에서 성호동과 그 주변 지역주민들의 삶, 지리적, 환경적 문화를 예술에 담은 〈사회참여예술창작지원사업〉 ‘예술로 성호창발(城湖創發)’ 창작 발표회를 갖는다고 밝혔다. 문화특화지역조성사업의 일환으로 진행하는 이 사업은 생성과 소멸의 과정을 거쳐 성장해온 도심 속 흩어져 있는 다양한 문화를 예술로 기록하고 후대에 물려주기 위한 취지로 순수예술 창작물을 구현하고 예술의 사회적 책임에 관해 고찰해 보는 사업이다. 앞으로 지역 예술인들과 뜻을 모아 소중한 원도심의 문화가 한순간 덧없이 사라지는 일이 없도록 매년 확대해 나갈 계획이다. 성호동을 대상으로 올해 처음 시행하는 ‘예술로 성호창발(城湖創發)’ 사업은 지난해 10월 지역 예술인들을 대상으로 참가자를 공개 모집한 결과, 사업 취지를 이해하고 창작 능력과 작품의 완성도가 기대되는 4개 단체를 선정하였다. 선정된 4개 단체 작가들은 스토리 개발을 위해 성호동 주민들을 만나 세밀한 대화를 나누기도 하고, 장르와 특성에 맞는 창작품을 만들기 위해 성호동 구석구석을 누비는 열정을 보였다. 발표회는 1월 30일 오후 2시부터 4시 30분까지 ‘성호동 이바구 투어’라는 이름으로 마산박물관 앞에 모여 문신 길을 따라 추산창작공작소 4933을 거쳐 임항선 그린웨이, 그리고 성호동 생활문화센터를 지나 정법사로 이동하면서 펼쳐진다. 〈(사)경남국악관현악단 ‘휴’〉가 ‘추산창작공작소 4933’에서 동네 이야기를 판소리로 엮어 국악 관현악과 경서도 소리를 통해 성호동의 옛이야기를 펼치고, 는 ‘성호동의 역사와 추억을 담다’란 주제로 도시가 발달하면서 퇴색되어가는 성호동 동네잔치의 상징적 의미를 담은 조형물을 설치했다. 〈신미란 작가〉는 ‘성호동 생활문화센터 2층 전시장’에서 성호동 사람과 삶의 이야기(話)와 그림(畵)을 표현한 ‘성호화첩(成湖話畵)’을 발표하며, 〈박은혜 춤패〉는 ‘일란 이필이의 춤을 찾아가는 소소한 춤 여행’이란 주제로 이필이 선생의 춤을 추며 다큐멘터리 시사회를 가지는 것으로 전체 일정을 마친다. 공식 행사 외의 진행은 성호동 토박이 김경년 창원시도시재생지원센터 해설사가 맡아 행사장을 이동하는 중간중간 맛깔나는 이바구로 성호동의 역사를 들려준다. 심재욱 창원시 문화체육관광국장은 “도시는 사람의 역사를 품은 곳으로 이제 사람이 도시를 품어야 문화예술로 지속 성장의 가능성을 보여줄 수 있다”라며 문화의 힘을 강조했다. 한편 행사 참가는 사회적 거리두기를 위해 참석 인원을 20명으로 제한하여 운영한다.',
'충남 공주시는 제민천과 국고개 일원을 대상으로 지난해 8월부터 추진한 공공미술 프로젝트 사업이 완료됐다고 21일 밝혔다. 문화체육관광부가 주최 공주시가 주관한 공공미술 프로젝트는 지역예술인 일자리 제공 및 주민 문화 향유 증진이라는 취지로 전국 228개 지자체 대상으로 추진됐다. 이에 시는 지난해 8월 공모를 통해 한국미술협회(공주지부)와 충남공예협동조합 2개팀을 선정 지역작가 37명의 참여를 통해 진행해 \'중동교~대통교\'사이에 조명과 벤치, 포토존 그리고 주민참여로 만들어진 송사리 조형물을 설치해 주민들에게 휴식 공간을 제공했다. 그리고 \'국고개 흐르는 효심\'은 도조작품, 화분조형, 이복이야기 동화책 및 국고개 가이드북 만들기 등 커뮤니티 프로그램 운영 역사.문화.예술 중심지의 특성을 살리는데 중점을 뒀다. 황의정 문화체육과장은 "앞으로도 시민들이 언제 어디서나 문화예술을 누리고, 지역예술인에게는 일자리를 제공할 수 있는 다양한 사업을 진행하겠다"고 말했다.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Hyperparameters
#### Non-Default Hyperparameters
- `per_device_train_batch_size`: 64
- `learning_rate`: 3e-05
- `num_train_epochs`: 1
- `warmup_ratio`: 0.05
- `fp16`: True
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: no
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 64
- `per_device_eval_batch_size`: 8
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 3e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 1
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.05
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: True
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `eval_use_gather_object`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
<details><summary>Click to expand</summary>
| Epoch | Step | Training Loss |
|:------:|:----:|:-------------:|
| 0.0055 | 1 | 0.9362 |
| 0.0109 | 2 | 0.9293 |
| 0.0164 | 3 | 0.9862 |
| 0.0219 | 4 | 0.9736 |
| 0.0273 | 5 | 0.9256 |
| 0.0328 | 6 | 0.8622 |
| 0.0383 | 7 | 0.762 |
| 0.0437 | 8 | 0.6716 |
| 0.0492 | 9 | 0.7307 |
| 0.0546 | 10 | 0.7201 |
| 0.0601 | 11 | 0.6136 |
| 0.0656 | 12 | 0.5874 |
| 0.0710 | 13 | 0.5537 |
| 0.0765 | 14 | 0.5484 |
| 0.0820 | 15 | 0.5917 |
| 0.0874 | 16 | 0.4819 |
| 0.0929 | 17 | 0.4295 |
| 0.0984 | 18 | 0.4323 |
| 0.1038 | 19 | 0.4973 |
| 0.1093 | 20 | 0.523 |
| 0.1148 | 21 | 0.4483 |
| 0.1202 | 22 | 0.3937 |
| 0.1257 | 23 | 0.4334 |
| 0.1311 | 24 | 0.4395 |
| 0.1366 | 25 | 0.3938 |
| 0.1421 | 26 | 0.4505 |
| 0.1475 | 27 | 0.4064 |
| 0.1530 | 28 | 0.442 |
| 0.1585 | 29 | 0.4523 |
| 0.1639 | 30 | 0.4507 |
| 0.1694 | 31 | 0.3424 |
| 0.1749 | 32 | 0.3679 |
| 0.1803 | 33 | 0.452 |
| 0.1858 | 34 | 0.3591 |
| 0.1913 | 35 | 0.4057 |
| 0.1967 | 36 | 0.3793 |
| 0.2022 | 37 | 0.4284 |
| 0.2077 | 38 | 0.4479 |
| 0.2131 | 39 | 0.4021 |
| 0.2186 | 40 | 0.4232 |
| 0.2240 | 41 | 0.3731 |
| 0.2295 | 42 | 0.4112 |
| 0.2350 | 43 | 0.329 |
| 0.2404 | 44 | 0.3724 |
| 0.2459 | 45 | 0.3368 |
| 0.2514 | 46 | 0.3915 |
| 0.2568 | 47 | 0.3626 |
| 0.2623 | 48 | 0.3927 |
| 0.2678 | 49 | 0.3321 |
| 0.2732 | 50 | 0.4041 |
| 0.2787 | 51 | 0.3904 |
| 0.2842 | 52 | 0.3633 |
| 0.2896 | 53 | 0.3536 |
| 0.2951 | 54 | 0.3782 |
| 0.3005 | 55 | 0.3478 |
| 0.3060 | 56 | 0.3844 |
| 0.3115 | 57 | 0.3319 |
| 0.3169 | 58 | 0.3694 |
| 0.3224 | 59 | 0.3954 |
| 0.3279 | 60 | 0.3655 |
| 0.3333 | 61 | 0.3842 |
| 0.3388 | 62 | 0.3405 |
| 0.3443 | 63 | 0.3799 |
| 0.3497 | 64 | 0.3691 |
| 0.3552 | 65 | 0.3444 |
| 0.3607 | 66 | 0.3604 |
| 0.3661 | 67 | 0.3966 |
| 0.3716 | 68 | 0.4169 |
| 0.3770 | 69 | 0.3464 |
| 0.3825 | 70 | 0.3306 |
| 0.3880 | 71 | 0.3634 |
| 0.3934 | 72 | 0.3614 |
| 0.3989 | 73 | 0.3486 |
| 0.4044 | 74 | 0.3768 |
| 0.4098 | 75 | 0.349 |
| 0.4153 | 76 | 0.3468 |
| 0.4208 | 77 | 0.353 |
| 0.4262 | 78 | 0.358 |
| 0.4317 | 79 | 0.343 |
| 0.4372 | 80 | 0.3312 |
| 0.4426 | 81 | 0.3504 |
| 0.4481 | 82 | 0.3915 |
| 0.4536 | 83 | 0.3477 |
| 0.4590 | 84 | 0.3754 |
| 0.4645 | 85 | 0.358 |
| 0.4699 | 86 | 0.3157 |
| 0.4754 | 87 | 0.339 |
| 0.4809 | 88 | 0.3751 |
| 0.4863 | 89 | 0.3216 |
| 0.4918 | 90 | 0.3321 |
| 0.4973 | 91 | 0.3793 |
| 0.5027 | 92 | 0.3811 |
| 0.5082 | 93 | 0.3266 |
| 0.5137 | 94 | 0.3517 |
| 0.5191 | 95 | 0.3395 |
| 0.5246 | 96 | 0.3489 |
| 0.5301 | 97 | 0.3309 |
| 0.5355 | 98 | 0.2989 |
| 0.5410 | 99 | 0.3337 |
| 0.5464 | 100 | 0.3495 |
| 0.5519 | 101 | 0.3145 |
| 0.5574 | 102 | 0.3704 |
| 0.5628 | 103 | 0.3184 |
| 0.5683 | 104 | 0.3123 |
| 0.5738 | 105 | 0.3363 |
| 0.5792 | 106 | 0.3444 |
| 0.5847 | 107 | 0.2916 |
| 0.5902 | 108 | 0.3163 |
| 0.5956 | 109 | 0.3304 |
| 0.6011 | 110 | 0.2937 |
| 0.6066 | 111 | 0.3321 |
| 0.6120 | 112 | 0.3638 |
| 0.6175 | 113 | 0.3525 |
| 0.6230 | 114 | 0.3646 |
| 0.6284 | 115 | 0.3393 |
| 0.6339 | 116 | 0.3327 |
| 0.6393 | 117 | 0.3075 |
| 0.6448 | 118 | 0.342 |
| 0.6503 | 119 | 0.363 |
| 0.6557 | 120 | 0.3238 |
| 0.6612 | 121 | 0.3777 |
| 0.6667 | 122 | 0.3194 |
| 0.6721 | 123 | 0.3194 |
| 0.6776 | 124 | 0.3325 |
| 0.6831 | 125 | 0.3615 |
| 0.6885 | 126 | 0.3363 |
| 0.6940 | 127 | 0.3149 |
| 0.6995 | 128 | 0.3071 |
| 0.7049 | 129 | 0.3481 |
| 0.7104 | 130 | 0.3252 |
| 0.7158 | 131 | 0.3262 |
| 0.7213 | 132 | 0.3133 |
| 0.7268 | 133 | 0.3516 |
| 0.7322 | 134 | 0.3329 |
| 0.7377 | 135 | 0.3546 |
| 0.7432 | 136 | 0.32 |
| 0.7486 | 137 | 0.3359 |
| 0.7541 | 138 | 0.3466 |
| 0.7596 | 139 | 0.3147 |
| 0.7650 | 140 | 0.2709 |
| 0.7705 | 141 | 0.391 |
| 0.7760 | 142 | 0.3345 |
| 0.7814 | 143 | 0.3182 |
| 0.7869 | 144 | 0.3533 |
| 0.7923 | 145 | 0.3247 |
| 0.7978 | 146 | 0.3572 |
| 0.8033 | 147 | 0.336 |
| 0.8087 | 148 | 0.337 |
| 0.8142 | 149 | 0.321 |
| 0.8197 | 150 | 0.3655 |
| 0.8251 | 151 | 0.3425 |
| 0.8306 | 152 | 0.3284 |
| 0.8361 | 153 | 0.3443 |
| 0.8415 | 154 | 0.3064 |
| 0.8470 | 155 | 0.2999 |
| 0.8525 | 156 | 0.3006 |
| 0.8579 | 157 | 0.325 |
| 0.8634 | 158 | 0.3321 |
| 0.8689 | 159 | 0.3414 |
| 0.8743 | 160 | 0.3172 |
| 0.8798 | 161 | 0.332 |
| 0.8852 | 162 | 0.296 |
| 0.8907 | 163 | 0.3394 |
| 0.8962 | 164 | 0.3254 |
| 0.9016 | 165 | 0.3288 |
| 0.9071 | 166 | 0.3342 |
| 0.9126 | 167 | 0.322 |
| 0.9180 | 168 | 0.3178 |
| 0.9235 | 169 | 0.3103 |
| 0.9290 | 170 | 0.299 |
| 0.9344 | 171 | 0.3765 |
| 0.9399 | 172 | 0.3814 |
| 0.9454 | 173 | 0.3275 |
| 0.9508 | 174 | 0.2835 |
| 0.9563 | 175 | 0.3338 |
| 0.9617 | 176 | 0.3284 |
| 0.9672 | 177 | 0.3784 |
| 0.9727 | 178 | 0.3031 |
| 0.9781 | 179 | 0.3149 |
| 0.9836 | 180 | 0.3437 |
| 0.9891 | 181 | 0.3259 |
| 0.9945 | 182 | 0.3329 |
| 1.0 | 183 | 0.3135 |
</details>
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.2.1
- Transformers: 4.44.2
- PyTorch: 2.3.1+cu121
- Accelerate: 1.1.1
- Datasets: 2.21.0
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### CachedMultipleNegativesRankingLoss
```bibtex
@misc{gao2021scaling,
title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
year={2021},
eprint={2101.06983},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
| null |
Non_BioNLP
|
# SentenceTransformer based on BAAI/bge-m3
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) <!-- at revision 5617a9f61b028005a4858fdac845db406aefb181 -->
- **Maximum Sequence Length:** 1024 tokens
- **Output Dimensionality:** 1024 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 1024, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("seongil-dn/bge-m3-kor-retrieval-451949-bs64-news")
# Run inference
sentences = [
'지역 예술인을 위한 공모를 시행했던 프로젝트는 어떤 것에 일부분이야?',
'마산의 원도심 성호동의 역사가 예술로 기록되고 되살아난다. 시대를 살아낸 사람들의 애틋한 이야기가 아름다운 소리, 몸짓, 조형, 화첩에 담겨 시민을 만난다. 창원시와 창원시문화도시지원센터는 오는 30일 오후 2시 마산합포구 문신길 일대에서 성호동과 그 주변 지역주민들의 삶, 지리적, 환경적 문화를 예술에 담은 〈사회참여예술창작지원사업〉 ‘예술로 성호창발(城湖創發)’ 창작 발표회를 갖는다고 밝혔다. 문화특화지역조성사업의 일환으로 진행하는 이 사업은 생성과 소멸의 과정을 거쳐 성장해온 도심 속 흩어져 있는 다양한 문화를 예술로 기록하고 후대에 물려주기 위한 취지로 순수예술 창작물을 구현하고 예술의 사회적 책임에 관해 고찰해 보는 사업이다. 앞으로 지역 예술인들과 뜻을 모아 소중한 원도심의 문화가 한순간 덧없이 사라지는 일이 없도록 매년 확대해 나갈 계획이다. 성호동을 대상으로 올해 처음 시행하는 ‘예술로 성호창발(城湖創發)’ 사업은 지난해 10월 지역 예술인들을 대상으로 참가자를 공개 모집한 결과, 사업 취지를 이해하고 창작 능력과 작품의 완성도가 기대되는 4개 단체를 선정하였다. 선정된 4개 단체 작가들은 스토리 개발을 위해 성호동 주민들을 만나 세밀한 대화를 나누기도 하고, 장르와 특성에 맞는 창작품을 만들기 위해 성호동 구석구석을 누비는 열정을 보였다. 발표회는 1월 30일 오후 2시부터 4시 30분까지 ‘성호동 이바구 투어’라는 이름으로 마산박물관 앞에 모여 문신 길을 따라 추산창작공작소 4933을 거쳐 임항선 그린웨이, 그리고 성호동 생활문화센터를 지나 정법사로 이동하면서 펼쳐진다. 〈(사)경남국악관현악단 ‘휴’〉가 ‘추산창작공작소 4933’에서 동네 이야기를 판소리로 엮어 국악 관현악과 경서도 소리를 통해 성호동의 옛이야기를 펼치고, 는 ‘성호동의 역사와 추억을 담다’란 주제로 도시가 발달하면서 퇴색되어가는 성호동 동네잔치의 상징적 의미를 담은 조형물을 설치했다. 〈신미란 작가〉는 ‘성호동 생활문화센터 2층 전시장’에서 성호동 사람과 삶의 이야기(話)와 그림(畵)을 표현한 ‘성호화첩(成湖話畵)’을 발표하며, 〈박은혜 춤패〉는 ‘일란 이필이의 춤을 찾아가는 소소한 춤 여행’이란 주제로 이필이 선생의 춤을 추며 다큐멘터리 시사회를 가지는 것으로 전체 일정을 마친다. 공식 행사 외의 진행은 성호동 토박이 김경년 창원시도시재생지원센터 해설사가 맡아 행사장을 이동하는 중간중간 맛깔나는 이바구로 성호동의 역사를 들려준다. 심재욱 창원시 문화체육관광국장은 “도시는 사람의 역사를 품은 곳으로 이제 사람이 도시를 품어야 문화예술로 지속 성장의 가능성을 보여줄 수 있다”라며 문화의 힘을 강조했다. 한편 행사 참가는 사회적 거리두기를 위해 참석 인원을 20명으로 제한하여 운영한다.',
'충남 공주시는 제민천과 국고개 일원을 대상으로 지난해 8월부터 추진한 공공미술 프로젝트 사업이 완료됐다고 21일 밝혔다. 문화체육관광부가 주최 공주시가 주관한 공공미술 프로젝트는 지역예술인 일자리 제공 및 주민 문화 향유 증진이라는 취지로 전국 228개 지자체 대상으로 추진됐다. 이에 시는 지난해 8월 공모를 통해 한국미술협회(공주지부)와 충남공예협동조합 2개팀을 선정 지역작가 37명의 참여를 통해 진행해 \'중동교~대통교\'사이에 조명과 벤치, 포토존 그리고 주민참여로 만들어진 송사리 조형물을 설치해 주민들에게 휴식 공간을 제공했다. 그리고 \'국고개 흐르는 효심\'은 도조작품, 화분조형, 이복이야기 동화책 및 국고개 가이드북 만들기 등 커뮤니티 프로그램 운영 역사.문화.예술 중심지의 특성을 살리는데 중점을 뒀다. 황의정 문화체육과장은 "앞으로도 시민들이 언제 어디서나 문화예술을 누리고, 지역예술인에게는 일자리를 제공할 수 있는 다양한 사업을 진행하겠다"고 말했다.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Hyperparameters
#### Non-Default Hyperparameters
- `per_device_train_batch_size`: 64
- `learning_rate`: 3e-05
- `num_train_epochs`: 1
- `warmup_ratio`: 0.05
- `fp16`: True
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: no
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 64
- `per_device_eval_batch_size`: 8
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 3e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 1
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.05
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: True
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `eval_use_gather_object`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
<details><summary>Click to expand</summary>
| Epoch | Step | Training Loss |
|:------:|:----:|:-------------:|
| 0.0055 | 1 | 0.9362 |
| 0.0109 | 2 | 0.9293 |
| 0.0164 | 3 | 0.9862 |
| 0.0219 | 4 | 0.9736 |
| 0.0273 | 5 | 0.9256 |
| 0.0328 | 6 | 0.8622 |
| 0.0383 | 7 | 0.762 |
| 0.0437 | 8 | 0.6716 |
| 0.0492 | 9 | 0.7307 |
| 0.0546 | 10 | 0.7201 |
| 0.0601 | 11 | 0.6136 |
| 0.0656 | 12 | 0.5874 |
| 0.0710 | 13 | 0.5537 |
| 0.0765 | 14 | 0.5484 |
| 0.0820 | 15 | 0.5917 |
| 0.0874 | 16 | 0.4819 |
| 0.0929 | 17 | 0.4295 |
| 0.0984 | 18 | 0.4323 |
| 0.1038 | 19 | 0.4973 |
| 0.1093 | 20 | 0.523 |
| 0.1148 | 21 | 0.4483 |
| 0.1202 | 22 | 0.3937 |
| 0.1257 | 23 | 0.4334 |
| 0.1311 | 24 | 0.4395 |
| 0.1366 | 25 | 0.3938 |
| 0.1421 | 26 | 0.4505 |
| 0.1475 | 27 | 0.4064 |
| 0.1530 | 28 | 0.442 |
| 0.1585 | 29 | 0.4523 |
| 0.1639 | 30 | 0.4507 |
| 0.1694 | 31 | 0.3424 |
| 0.1749 | 32 | 0.3679 |
| 0.1803 | 33 | 0.452 |
| 0.1858 | 34 | 0.3591 |
| 0.1913 | 35 | 0.4057 |
| 0.1967 | 36 | 0.3793 |
| 0.2022 | 37 | 0.4284 |
| 0.2077 | 38 | 0.4479 |
| 0.2131 | 39 | 0.4021 |
| 0.2186 | 40 | 0.4232 |
| 0.2240 | 41 | 0.3731 |
| 0.2295 | 42 | 0.4112 |
| 0.2350 | 43 | 0.329 |
| 0.2404 | 44 | 0.3724 |
| 0.2459 | 45 | 0.3368 |
| 0.2514 | 46 | 0.3915 |
| 0.2568 | 47 | 0.3626 |
| 0.2623 | 48 | 0.3927 |
| 0.2678 | 49 | 0.3321 |
| 0.2732 | 50 | 0.4041 |
| 0.2787 | 51 | 0.3904 |
| 0.2842 | 52 | 0.3633 |
| 0.2896 | 53 | 0.3536 |
| 0.2951 | 54 | 0.3782 |
| 0.3005 | 55 | 0.3478 |
| 0.3060 | 56 | 0.3844 |
| 0.3115 | 57 | 0.3319 |
| 0.3169 | 58 | 0.3694 |
| 0.3224 | 59 | 0.3954 |
| 0.3279 | 60 | 0.3655 |
| 0.3333 | 61 | 0.3842 |
| 0.3388 | 62 | 0.3405 |
| 0.3443 | 63 | 0.3799 |
| 0.3497 | 64 | 0.3691 |
| 0.3552 | 65 | 0.3444 |
| 0.3607 | 66 | 0.3604 |
| 0.3661 | 67 | 0.3966 |
| 0.3716 | 68 | 0.4169 |
| 0.3770 | 69 | 0.3464 |
| 0.3825 | 70 | 0.3306 |
| 0.3880 | 71 | 0.3634 |
| 0.3934 | 72 | 0.3614 |
| 0.3989 | 73 | 0.3486 |
| 0.4044 | 74 | 0.3768 |
| 0.4098 | 75 | 0.349 |
| 0.4153 | 76 | 0.3468 |
| 0.4208 | 77 | 0.353 |
| 0.4262 | 78 | 0.358 |
| 0.4317 | 79 | 0.343 |
| 0.4372 | 80 | 0.3312 |
| 0.4426 | 81 | 0.3504 |
| 0.4481 | 82 | 0.3915 |
| 0.4536 | 83 | 0.3477 |
| 0.4590 | 84 | 0.3754 |
| 0.4645 | 85 | 0.358 |
| 0.4699 | 86 | 0.3157 |
| 0.4754 | 87 | 0.339 |
| 0.4809 | 88 | 0.3751 |
| 0.4863 | 89 | 0.3216 |
| 0.4918 | 90 | 0.3321 |
| 0.4973 | 91 | 0.3793 |
| 0.5027 | 92 | 0.3811 |
| 0.5082 | 93 | 0.3266 |
| 0.5137 | 94 | 0.3517 |
| 0.5191 | 95 | 0.3395 |
| 0.5246 | 96 | 0.3489 |
| 0.5301 | 97 | 0.3309 |
| 0.5355 | 98 | 0.2989 |
| 0.5410 | 99 | 0.3337 |
| 0.5464 | 100 | 0.3495 |
| 0.5519 | 101 | 0.3145 |
| 0.5574 | 102 | 0.3704 |
| 0.5628 | 103 | 0.3184 |
| 0.5683 | 104 | 0.3123 |
| 0.5738 | 105 | 0.3363 |
| 0.5792 | 106 | 0.3444 |
| 0.5847 | 107 | 0.2916 |
| 0.5902 | 108 | 0.3163 |
| 0.5956 | 109 | 0.3304 |
| 0.6011 | 110 | 0.2937 |
| 0.6066 | 111 | 0.3321 |
| 0.6120 | 112 | 0.3638 |
| 0.6175 | 113 | 0.3525 |
| 0.6230 | 114 | 0.3646 |
| 0.6284 | 115 | 0.3393 |
| 0.6339 | 116 | 0.3327 |
| 0.6393 | 117 | 0.3075 |
| 0.6448 | 118 | 0.342 |
| 0.6503 | 119 | 0.363 |
| 0.6557 | 120 | 0.3238 |
| 0.6612 | 121 | 0.3777 |
| 0.6667 | 122 | 0.3194 |
| 0.6721 | 123 | 0.3194 |
| 0.6776 | 124 | 0.3325 |
| 0.6831 | 125 | 0.3615 |
| 0.6885 | 126 | 0.3363 |
| 0.6940 | 127 | 0.3149 |
| 0.6995 | 128 | 0.3071 |
| 0.7049 | 129 | 0.3481 |
| 0.7104 | 130 | 0.3252 |
| 0.7158 | 131 | 0.3262 |
| 0.7213 | 132 | 0.3133 |
| 0.7268 | 133 | 0.3516 |
| 0.7322 | 134 | 0.3329 |
| 0.7377 | 135 | 0.3546 |
| 0.7432 | 136 | 0.32 |
| 0.7486 | 137 | 0.3359 |
| 0.7541 | 138 | 0.3466 |
| 0.7596 | 139 | 0.3147 |
| 0.7650 | 140 | 0.2709 |
| 0.7705 | 141 | 0.391 |
| 0.7760 | 142 | 0.3345 |
| 0.7814 | 143 | 0.3182 |
| 0.7869 | 144 | 0.3533 |
| 0.7923 | 145 | 0.3247 |
| 0.7978 | 146 | 0.3572 |
| 0.8033 | 147 | 0.336 |
| 0.8087 | 148 | 0.337 |
| 0.8142 | 149 | 0.321 |
| 0.8197 | 150 | 0.3655 |
| 0.8251 | 151 | 0.3425 |
| 0.8306 | 152 | 0.3284 |
| 0.8361 | 153 | 0.3443 |
| 0.8415 | 154 | 0.3064 |
| 0.8470 | 155 | 0.2999 |
| 0.8525 | 156 | 0.3006 |
| 0.8579 | 157 | 0.325 |
| 0.8634 | 158 | 0.3321 |
| 0.8689 | 159 | 0.3414 |
| 0.8743 | 160 | 0.3172 |
| 0.8798 | 161 | 0.332 |
| 0.8852 | 162 | 0.296 |
| 0.8907 | 163 | 0.3394 |
| 0.8962 | 164 | 0.3254 |
| 0.9016 | 165 | 0.3288 |
| 0.9071 | 166 | 0.3342 |
| 0.9126 | 167 | 0.322 |
| 0.9180 | 168 | 0.3178 |
| 0.9235 | 169 | 0.3103 |
| 0.9290 | 170 | 0.299 |
| 0.9344 | 171 | 0.3765 |
| 0.9399 | 172 | 0.3814 |
| 0.9454 | 173 | 0.3275 |
| 0.9508 | 174 | 0.2835 |
| 0.9563 | 175 | 0.3338 |
| 0.9617 | 176 | 0.3284 |
| 0.9672 | 177 | 0.3784 |
| 0.9727 | 178 | 0.3031 |
| 0.9781 | 179 | 0.3149 |
| 0.9836 | 180 | 0.3437 |
| 0.9891 | 181 | 0.3259 |
| 0.9945 | 182 | 0.3329 |
| 1.0 | 183 | 0.3135 |
</details>
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.2.1
- Transformers: 4.44.2
- PyTorch: 2.3.1+cu121
- Accelerate: 1.1.1
- Datasets: 2.21.0
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### CachedMultipleNegativesRankingLoss
```bibtex
@misc{gao2021scaling,
title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
year={2021},
eprint={2101.06983},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
{"base_model": "BAAI/bge-m3", "library_name": "sentence-transformers", "pipeline_tag": "sentence-similarity", "tags": ["sentence-transformers", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:451949", "loss:CachedMultipleNegativesRankingLoss"], "widget": [{"source_sentence": "어느 정도의 인원을 대상으로 25일에 도착한 예방주사를 접종할 수 있어?", "sentences": ["“길고 길었던 ‘코로나19 터널’을 빠져 나오는 신호탄이 되길 간절히 기원합니다.” 코로나19 종식 염원을 담은 아스트라제네카(AZ) 백신이 26일부터 접종에 들어간다. 접종이 하루 앞으로 다가오면서 일선 보건당국도 비상 근무체제에 돌입했다. 25일 낮 12시10분께 AZ 백신을 실은 운반차량이 군경의 호위 속에 안양시 동안구보건소 앞에 도착하자, 보건소 직원들이 건물 앞까지 나와 백신을 맞았다. ‘생물학적 제제 출하증명서’가 오가고 아이스박스가 열리자, 투명한 5밀리리터 통 위에 빨간 캡이 씌워진 AZ가 그 모습을 드러냈다. 보건소 직원들은 조심스럽게 백신을 하나 하나 들추며 혹시라도 깨지거나 파손된 물건은 없는지 상태를 살폈다. 점검을 마친 백신이 곧장 보건소 3층에 마련된 전용 냉장고로 들어가면서 이날 백신 상륙 작전은 약 5분 만에 마무리됐다. 이날 들어온 1차 물량은 30바이알, 300명분이다. 한숨 돌릴 틈도 없이 입고식을 마친 직원들은 보건소장실에 모여 ‘예방접종 후 이상반응 관리’ 교육을 받았다. 26일 오전 9시부터는 동안구 9개 요양시설 종사자 중 일부인 1개소, 20명이 보건소를 방문해 첫 접종을 받을 예정이다. 동안구보건소 관계자는 “접종을 차질 없이 진행하기 위해 만반의 준비를 마쳤다”면서 “가장 중요한 백신을 보관하는 전용 냉장고는 영상 2~8도를 항시 유지하고 있으며 정전에도 대비하기 위해 무정전 전원장치(UPS)까지 갖춰져 있다”고 말했다. 이와 함께 수원시 영통구보건소에도 20바이알의 AZ 백신이 무사히 도착했다. 보건소 1층에 마련된 접종실엔 안내문이 비치됐고, 직원들은 예진표 작성부터 신분증과 체온 확인, 예진 후 접종, 이상반응 모니터링, 확인증 발급까지 방문자가 거쳐야 할 동선을 한 번 더 꼼꼼히 점검했다. 영통구보건소 관계자는 “소방서와 연계한 긴급 출동 체계가 마련돼 있다”면서 “만일의 상황을 대비해 자동제세동기(AED) 점검을 마쳤으며 응급키트도 구비해놨다”고 밝혔다.", "사법농단 의혹 재판에서 6연속 무죄를 끊고 ‘1호 유죄’가 나온 배경에는 “대법원장과 법원행정처는 특정 재판에서 명백한 잘못이 있을 경우 판사 지적권한이 있다”는 재판부의 독특한 판시가 있다. 앞서 임성근 전 부산고법 부장판사가 1심에서 ‘위헌적이지만, 남용할 직권 자체가 없었다’며 무죄 선고를 받은 것과는 상반된다. 서울중앙지법 형사32부(부장판사 윤종섭)는 23일 대법원장과 법원행정처에는 재판사무의 핵심영역에 대해 명백한 잘못이 있을 경우 담당 판사의 잘못을 ‘지적’할 수 있는 권한이 있다고 밝혔다. 이민걸 전 법원행정처 기획조정실장과 이규진 전 대법원 양형위원회 상임위원의 일부 재판개입 등 행위(직권남용)를 유죄 판단하면서다. 사법농단 관련 재판에서는 처음 나온 판시다. 재판부는 전제로 “직업적으로 충분히 단련하지 못하거나 나태해 쟁점이 별로 없는 것만 우선 처리해 장기미제를 만드는 판사가 있을 수 있다”고 말했다. 그러면서 “이런 경우에도 재판 독립을 이유로 누구도 지적을 못한다면 국민의 재판청구권 보장과 충돌할 수 있다”고 했다. 헌법 103조에 따른 법관 독립이 ‘신성불가침’은 아니라는 얘기다. 재판부는 이를 뒷받침할 명시적 규정은 없지만 헌법과 법원조직법 해석상 명백한 잘못을 전제로 사법행정사무를 총괄하는 대법원장과 행정처에 판사를 지적할 권한이 있다고 봤다. 선뜻 이해하기 어려운 이 법리는 경계가 애매모호한 직권남용죄의 특성 때문에 도출된 것이다. 직권남용은 ‘남용될 직권’의 존재를 전제한다. 표면상 이유로는 수긍할 수 있지만 실제로 따져보면 위법하다는 까다로운 조건을 만족해야 한다. 임 전 부장판사의 경우 “각 재판관여행위가 위헌적”이라는 판단에도 불구하고 “(서울중앙지법 형사수석부장판사 당시의) 일반적 직무권한에 속한다고 해석될 여지가 없다”며 1심에서 무죄를 선고 받았다. 아울러 재판부는 “공무원의 직무범위를 벗어난 ‘월권행위’도 직무와 상당한 관련성이 인정된다면 직권남용이 성립될 수 있다”며 직권남용 범위를 상대적으로 넓게 판단했다. 직권의 범주를 형식적으로만 봐선 안 된다는 취지다. 이 전 실장 등은 이를 엄격하게 판단해야 한다며 검찰과 치열한 공방을 벌였지만 1심에서 받아들여지지 않았다. 재판부는 이 같은 법리에 따라 헌법재판소 파견 법관을 통한 내부 기밀 수집, 한정위헌 취지의 위헌제청 결정 사건과 옛 통합진보당 국회의원·비례대표 지방의회의원 관련 행정소송에 대한 재판개입 등에 대한 일부 직권남용 혐의를 유죄로 판단했다. 검찰은 이번 재판을 양승태 전 대법원장 등 ‘수뇌부’를 단죄할 여지를 열어줄 계기로 보고 있다. 검찰 관계자는 “재판부가 앞선 사법농단 관련 재판과 달리 직권남용 범위를 넓게 봤다”며 “사법부 윗선의 유죄 가능성을 높인 판결”이라고 평가했다.", "65세 이상 73%, 6-59개월 54% 인플루엔자 접종 마쳐, 가능하면 11월 15일 이전 완료하세요~ □ 또한, 남은 물량은 충분하나, 일시적으로 지역적 부족상황이 발생할 수 있어 그에 대한 백신수급 운영 대책도 준비하고 있다고 덧붙였다. ○ 올해 처음으로 부족한 공급에 신속히 대처하기 위해 시도 및 질병관리본부에서 직접 공급할 수 있는 추가 물량*을 확보하였으며, 상황 발생 시 신속히 조치할 예정이다. * 추가확보물량: 시·도0.25㎖5,220 도즈, 0.5㎖80,680 도즈/ 본부0.25㎖40,000 도즈, 0.5㎖40,000 도즈\n□ *** 예방접종관리과장은 “아직까지 접종 받지 않은 어르신 및 어린이 보호자들은 지정의료기관 방문 전에 꼭 관할 보건소, 질병관리본부 콜센터(☎1339) 및 보건복지콜센터(☎129)로 문의하여 당일 접종이 가능한 의료기관을 안내받거나, 예방접종 도우미 홈페이지 및 스마트폰 앱*에서 확인 후 방문” 해달라고 당부했다. * 질병관리본부 예방접종도우미 홈페이지((https://nip.cdc.go.kr) 접속> 일반인(자세히 보기)> 예방접종도우미 연결), 예방접종도우미 스마트폰 앱의 의료기관 검색\n○ 또한, “인플루엔자 예방접종의 효과는 균주, 유행 특성 등에 따라 건강한 성인에서 70-90%, 소아에서 50-95% 범위 내로 다양한 것으로 알려져 있어,\n- 30초 이상 손씻기, 기침이나 재채기를 할 때 휴지나 옷깃으로 입을 가리는 기침예절 준수 등의 예방 수칙 실천을 병행하고,\n- 기침, 발열, 근육통 등 인플루엔자 의심 증상이 발생할 경우 사람이 많이 모인 곳을 피하고 의료기관을 늦지 않게 방문하여 진료받는 것이 중요하다”며 덧붙였다. ○ 아울러, “무료접종 대상자 외에 임신부, 만성질환자, 50~64세 인구 등의 인플루엔자 고위험군도 감염 예방을 위해 본격적인 인플루엔자 유행 전에 예방접종이 필요하다”고 밝혔다."]}, {"source_sentence": "길어지는 코로나를 준비하기 위해 조선대학교가 시행하고 있는 과정은 뭐야?", "sentences": ["조선대학교병원(병원장 정종훈)은 코로나19 장기화 및 재확산 대비를 위한 중증환자 전담 간호사 양성 교육을 시작했다고 2일 밝혔다. 보건복지부와 국립중앙의료원이 주최하고 조선대병원이 주관하는 이번 양성 교육 과정은 체계적인 간호교육 시스템을 통해 중증 환자 담당 간호사 양성을 목표로 하고 있다. 지난 1일 병원 7층 세미나A실에서 시작된 '중증환자 전담 간호사' 양성 교육은 4주간 진행되며 교육 내용은 중환자 간호 이론교육 및 모의실습과 중환자실 임상실습으로 구성된다. 이론교육 및 모의실습에는 △중환자의 신체사정 및 기본간호 △호흡기계 중재 및 간호 △특수처치 및 응급간호 등 중환자 간호 공통교육 이론강의 등을 비롯해 개인보호구 착탈의 방법 등이 포함됐다. 중환자실 실습을 통해서는 실제 중환자실 현장에서 개인별로 1:1 프리셉터가 배정되어 습득한 내용을 바탕으로 밀착 교육을 진행한다. 유윤정 조선대병원 간호부장은 \"코로나19 장기화 및 재확산 대비를 위해 진행하는 이번 양성 교육을 통해 전문성 있는 중환자 전담 인력을 양성하고 간호역량을 강화할 수 있을 것으로 보인다\"며 \"중환자 간호교육 인프라와 인적 자원 확보를 위해 최선을 다하겠다\"고 밝혔다.", "조선대학교는 지난 25일 총장실에서 우수 행정서비스 운영 사례를 발굴해 구성원들에게 공유, 확산시키고자 '대학행정서비스 우수사례 및 인증 시상식'을 진행했다고 30일 밝혔다. 지난 2019년부터 부서컨설팅을 수행 중인 조선대는 올해 18개 부서(누적 45개)에 대한 컨설팅을 수행한다. 5년 주기로 대학 내 전 부서에 대한 컨설팅을 진행하는 것이 목표다. 조선대가 지난 2월 수행한 대학행정서비스 컨설팅 결과, 교수학습지원센터의 'E-Learning지원팀'이 코로나19에 적극적으로 대응한 우수사례로 선정됐다. E-Learning지원팀은 코로나19 속 온라인수업에 대한 수요 및 요구사항을 수집해 반영했다. 또 온라인수업 콘텐츠를 제작해 수시로 안내하고 우수콘텐츠 공모전 시행으로 우수사례를 공유해 온라인 수업의 전반적인 질을 향상시키는 데 기여했다. 또 온라인시험 및 수업 운영 모니터링단을 꾸려 수업의 질 관리에 적극 노력했다. 조선대학교는 E-Learning지원팀이 진행한 온라인 수업의 질 관리 결과를 교육혁신원 홈페이지 및 업무포털을 활용해 '2020학년도 대학 교육행정 서비스 CQI를 위한 부서 컨설팅 결과 보고서'로 공유할 예정이다.", "김천상무가 최하위 부천과 원정 경기서 승점 1점을 보태는 데 그쳐 순위상승 기회를 놓쳤다. 김천은 지난 8일 부천종합운동장에서 열린 ‘하나원큐 K리그2 2021’ 11R 부천FC와의 경기에서 득점없이 0-0무승부를 기록했다. 김천은 조규성-오현규-강지훈이 쓰리톱, 심상민-정현철-박용우-안태현 중원, 이명재-정승현-하창래가 쓰리백에 나서는 3-4-3전술로 부천을 맞았다. 특히 이날 골키퍼에는 구성윤이 처음으로 장갑을 꼈다. 양팀은 경기 휘슬과 함께 치열한 중원다툼에 나서 전반 7분 부천 안재준이 먼저 날카로운 슛을 날리자 김천도 오현규가 12분과 18분 잇따라 위협적인 슛으로 맞받아쳤다. 오현규의 슛으로 조금씩 공세의 강도를 높이기 시작한 김천은 32분 이명재가 회심의 슛을 쐈으나 골대를 살짝 넘어갔으며, 39분에는 박용우가 프리킥 슛을 쐈지만 부천 수비수 강의빈에 막히면서 전반을 0-0으로 마쳤다. 김천은 후반들어서도 공격이 여의치 않자 7분 강지훈 대신 정재를 투입해 측면 공세를 높였고, 11분과 15분 안태현과 정재희 잇따라 부천 골문을 노렸지만 무위에 그쳤다. 좀처럼 득점 찬스를 만들지 못하던 김천은 31분 부천 한지호에게 골을 허용했으나 VAR 판독으로 무효가 되면서 위기를 넘겼다. 김천은 후반 44분 이명재가 결정적인 문전 슈팅을 날렸지만 부천 골키퍼 전종혁의 선방에 막히는 등 막판 파상적인 공세를 펼쳤지만 득점을 만들지 못했다."]}, {"source_sentence": "반도체 양성에 대한 정부의 후원은 언제 끝날 예정이야?", "sentences": ["반도체 산업의 정치적 요인을 분석한 결과, 초기에는 반도체 산업의 활성화를 위해 대형 국책사업 추진 등 전폭적인 지원사업이 이루어져왔으나, 현재 반도체와 관련된 예산 등 정부지원은 상대적으로 감소하고 있는 것으로 나타났다 (한국산업기술진흥원 2017). 그러나 정부는 반도체 산업의 성장을 지원하기위해 다각적인 측면에서 노력 중에 있다. 관련 업종별 ․ 수준별 특성을 고려한 차별화된 스마트공장 지원 강화를 통하여 뿌리 등 16개 등의 업종별 ‘대표 스마트공장’(30개, ~’22년) 및 고도화를 통한 스마트공장 프리미엄 5,000개(’22년) 구축을 추진하고자 하였으며, SW ․ 제어기기 등의 스마트 제조 플랫폼 개발을 통하여 반도체 산업의 재도약을 추진하고 있다(산업통상자원부 2017). 뿐만 아니라, 지능형반도체산업과 융합 가능한 임베디드 SW 산업 발전을 위해 다양한 정책 사업을 추진하고 있다(한국산업기술진흥원 2017).", "정부가 인공지능(AI), 전기차 등에 들어가는 핵심 부품인 차세대 전력 반도체 육성에 본격적으로 나선다. 2025년까지 차세대 전력 반도체 상용화 제품을 5개 이상 개발하고, 6∼8인치 파운드리(위탁생산) 인프라 구축도 추진한다. 정부는 1일 제7차 혁신성장 빅3 추진회의를 열고 이런 내용을 담은 '차세대 전력 반도체 기술개발 및 생산역량 확충 방안'을 발표했다. 전력 반도체란 전력의 변환·변압·안정·분배·제어를 수행하는 반도체를 말하며, 차세대 전력 반도체는 실리콘보다 전력 효율과 내구성이 뛰어난 3대 신소재 웨이퍼로 제작된 반도체를 뜻한다. 인공지능(AI), 5G나 고온·고압 내구성이 필요한 전기차, 신재생에너지 발전의 인버터 등에 적용되는 필수 부품이다. 앞으로 전자기기 수요가 확대되고 전력 소비가 늘 것으로 예상되면서 미래 성장 가능성이 높은 분야로 꼽힌다. 산업부에 따르면 글로벌 전력 반도체 시장 규모는 2019년 450억 달러에서 2023년 530억 달러 규모로 성장할 것으로 전망됐다. 이에 따라 미국과 중국, 일본 등도 다양한 정책적 지원을 추진 중이다. 우리 정부도 2017년부터 총 830억 원 규모의 '파워반도체 상용화 사업'을 진행 중이다. 그러나 국내 전력 반도체 시장 규모는 약 20억 달러 규모로, 기술력 부족과 해외 기업의 특허 선점으로 수요의 90% 이상을 수입에 의존하는 실정이다. 이에 정부는 2025년까지 글로벌 수준의 경쟁력을 확보하기 위해 상용화 제품 개발, 기반 기술 강화, 미래 제조공정 확보 등을 본격적으로 지원하기로 했다. 국내 수요기업과 연계한 과제를 발굴해 차세대 전력 반도체 상용화 제품을 5개 이상 개발할 방침이다. 현재 실리콘 소재의 한계를 극복하기 위해 실리콘카바이드(SiC), 질화갈륨(GaN), 갈륨 옥사이드(Ga2O3) 등 3대 신소재 응용기술을 개발하고, 국내 기업의 소재·웨이퍼 기술 확보를 지원할 방침이다. 고집적·고성능 차세대 전력 반도체 개발을 위해 파워 집적회로(IC) 설계 기술개발을 지원한다. 아울러 차세대 전력 반도체 관련 파운드리 서비스를 확대하기 위해 국내 파운드리와 6∼8인치 기반의 양산 공정을 구축하는 방안도 추진한다. 성윤모 산업부 장관은 \"AI, 5G 등 신기술 구현과 자율차, 신재생 등 미래 성장 분야의 활성화를 위해선 전력의 효율적인 관리가 필수이며, 차세대 전력 반도체는 이를 위한 핵심 부품\"이라며 \"아직 초기인 이 분야 시장을 선점하고 미래 경쟁력을 확보하기 위해 R＆D, 인프라 등을 적극 지원할 계획\"이라고 말했다.", "근로시간은 짧지만 국민 소득은 높은 국가들을 분석해 보니 노동 유연성과 생산성이 높다는 연구결과가 나왔다. 전국경제인연합회 산하 한국경제연구원은 경제협력개발기구(OECD) 통계와 통계청 데이터를 이용해 우리나라와 덴마크ㆍ노르웨이ㆍ독일ㆍ네덜란드 4개국의 노동생산성 등을 비교한 결과 이같이 나타났다고 29일 밝혔다. 한경연에 따르면 이들 국가의 평균 연간 근로시간은 1,396시간이었고, 평균 1인당 국민총소득은 6만187달러로 집계됐다. 반면 한국은 이들 국가보다 1.4배(1,967시간) 더 길게 일했지만, 소득은 3만2,115달러로 절반 수준이었다. 한경연은 이들 유럽 4국의 특징을 △높은 고용률 △높은 노동생산성 △높은 노동유연성 △시간제 근로 활성화 △높은 수준의 인적자원이라고 정리했다. 이들 국가의 평균 고용률은 76.4%로 한국(66.8%)보다 9.6%포인트 높았다. 특히 네덜란드와는 11.4%포인트 차이가 났다. 우리나라가 네덜란드 수준의 고용률을 달성하려면 418만6,000명의 일자리가 더 만들어져야 한다고 한경연은 분석했다. 시간당 노동생산성도 한국(40.5달러)은 이들 국가 평균(73.3달러)에 크게 못 미쳤다. 노르웨이는 84.3달러로 한국보다 2배 이상 높았다. 세계경제포럼(WEF) 노동시장 유연성 평가에서도 한국 점수는 크게 낮았다. 한국은 54.1점을 받아 OECD 37개국 중 35위에 그친 반면 4개 국가의 평균 점수는 68.9점이었다. 가장 높은 덴마크(71.4점)는 OECD 3위에 올랐다. 아울러 이들 국가는 시간제 근로 비율이 높았는데 네덜란드는 37.0%에 달해 한국(14.0%)보다 2.6배 높은 것으로 나타났다. 한경연은 이들 국가가 적게 일하면서 많이 벌 수 있었던 비결은 노동시장 개혁을 통해 노동 유연성을 확보했기 때문이라고 분석했다. 추광호 한경연 경제정책실장은 “적게 일하고 많이 버는 국가들은 시간제 근로 활성화, 노동 규제 완화 등을 통해 고용률을 높이고, 높은 생산성을 토대로 소득을 높였다”고 말했다. 한편 한경연이 우리나라 산업별 소득과 근로시간을 비교한 결과 적게 일하고 많이 버는 산업은 금융 및 보험업으로 나타났다. 한경연에 따르면 금융 및 보험업은 전체 월 임금 총액인 372만원보다 1.6배 높은 593만원을 벌었지만, 월 근로시간은 9시간 짧았다. 반면 숙박 및 음식점업은 이보다 적은 234만원을 벌지만 12시간 더 일하는 것으로 나타났다."]}, {"source_sentence": "E-GMP의 세번째 배터리 공급 회사로 어느 곳이 선발됐어?", "sentences": ["손흥민(손흥민)의 골이 30일째 침묵한 토트넘이 시즌 첫 3연패에 빠지며 8위로 추락했다. 토트넘은 5일(한국시간) 영국 런던의 토트넘 홋스퍼 스타디움에서 열린 첼시와 2020-2021 잉글랜드 프로축구 프리미어리그 22라운드 홈경기에서 전반 24분 조르지뉴에게 내준 페널티킥 결승 골을 만회하지 못하고 0-1로 패했다. 해리 케인의 부상 결장에 손흥민 골침묵까지 이어진 토트넘은 이번 시즌 첫 3연패이자 홈 경기 2연패로 승점 33에 머물면서 8위로 한 계단 내려앉았다. 지난달 6일 브렌트퍼드(2부)와 EFL컵 준결승전에서 득점을 맛본 손흥민은 이후 30일째 득점 신고를 하지 못하며 6경기(정규리그 5경기ㆍFA컵 1경기) 연속 무득점 행진을 이어갔다. 손흥민의 이번 시즌 공격포인트는 16골 10도움(EPL 12골 6도움ㆍ유로파리그 3골 3도움·리그컵 1골ㆍFA컵 1도움)이다. 아울러 통계전문 옵타에 따르면 토트넘의 조제 모리뉴 감독은 2000년 9월 벤피카(포르투갈)에서 감독직을 시작한 이후 자신의 사령탑 커리어 통산 327번째 홈 경기에서 처음으로 ‘홈경기 2연패’를 당하는 아쉬움까지 곱씹어야 했다. 반면 첼시는 승점 36(골 득실+13)으로 에버턴(승점 36ㆍ골 득실+6)을 골득실차로 따돌리고 8위에서 6위로 올라섰다.", "‘포항 배터리 규제자유특구’가 중소벤처기업부의 2020년 규제자유특구 운영성과 평가에서 전국 유일 2년 연속 최고등급인 ‘우수’ 특구로 21일 선정됐다. 이번 중소벤처기업부의 규제자유특구 운영성과 평가는 전국 14개 규제자유특구를 대상으로 성과 달성도, 파급효과, 규제특례 활용실적 등을 토대로 종합평가를 실시됐으며, 포항 배터리·대구 스마트웰니스·전북 친환경자동차 규제자유특구 3개 특구가 우수 평가를 받았다. 2019년 7월 영일만 일반산업단지 및 블루밸리 국가산업단지 일원(556,694.22㎡, 17만평)을 배터리 규제자유특구로 지정받은 이후 포항시는 배터리 전후방 기업들에 대한 적극적인 투자유치 활동으로 대규모 투자를 이끌어냈다. GS건설의 배터리 리사이클링 공장 건립, 포스코케미칼의 이차전지 인조흑연 음극재 생산공장 건립, 에코프로비엠 등 에코프로 6개 자회사의 양극재 생산공장 건립 등의 투자로 이차전지 밸류체인 구축과 이를 바탕으로 배터리 규제자유특구를 미래 유망산업인 배터리산업 육성의 거점으로서의 기반을 마련한 것이 높은 평가를 받았다. 뿐만 아니라, 규제자유특구 실증 연구개발 수행과 전기차 배터리 산업 육성 전초 기지 역할을 감당할 이차전지종합관리센터를 107억 원의 사업비를 투입해 블루밸리 국가산업단지 규제자유특구 내 건립 하는 등 특구 활성화를 위한 지자체의 적극적인 의지도 높은 평가를 받았다. 이차전지종합관리센터는 이차전지 고도화와 고부가 가치화를 통해 배터리 산업을 체계적으로 육성하기 위한 국내 최초, 최대 규모의 인프라이며, 이곳을 배터리 산업 허브로 육성해 배터리 소재, 부품, 장비 상용화 기술개발 및 우수 기업 유치 등의 마중물 역할로 지역경제 활성화에 크게 기여할 것이다. 더 나아가 포항시는 국가 전기차 사용후 배터리 순환경제 기반 조성 및 배터리 산업 컨트롤 타워 구축을 목표로 추진하고 있는 환경부의 ‘배터리 자원순환 클러스터 구축사업’에도 참여해 총력을 기울이고 있다. 포항시는 철강산업의 뒤를 이어 지역의 새로운 먹거리 산업으로 이차전지산업을 육성하기 위해 중소벤처기업부, 환경부 등 중앙 부처와 긴밀한 협력체계를 유지하면서 이차전지분야 기술개발, 기술실증, 기업육성의 전주기적 기업지원 인프라 구축으로 포항을 우리나라의 배터리 메카도시로 육성해나갈 계획이다. 이강덕 포항시장은 “중앙부처뿐만 아니라 배터리소재기업인 포스코케미칼과 에코프로, GS건설 등과 적극적인 협업으로 이차전지산업 육성을 통해 포항의 발전를 넘어 우리나라 배터리산업 육성에도 큰 기여를 할 수 있도록 행정력을 집중하겠다”라고 전했다.", "중국 전기차용 배터리 업체들이 탈중국화에 속도를 내며 ‘K배터리’를 위협하고 있다. 텃밭인 한국과 유럽 시장에서도 K배터리는 ‘C배터리’(중국산 배터리)의 거센 도전에 직면했다. 21일 업계에 따르면 중국 배터리 선두 업체인 CATL은 현대자동차그룹이 최근 선정한 전기차 전용 플랫폼 E-GMP 3차 물량 배터리 공급사로 낙찰됐다. 현대차그룹이 2023년 이후 출시 예정인 3개 차종에 대한 배터리 공급사를 선정했는데 SK이노베이션이 1종, CATL이 2종을 따낸 것이다. 총 9조원 규모 물량 중 절반 이상을 CATL이 차지한 것으로 알려졌다. 업계 관계자는 “CATL이 국내 시장까지 파고든 것은 가격·성능·생산력 측면에서 K배터리와 견줘 충분한 경쟁력을 지니고 있다는 뜻”이라며 “안정적 수급을 위해 배터리 공급사를 다변화해야하는 현대차 입장에서 스펙이 맞지 않는 삼성SDI, 코나EV 등 화재로 안전 이슈가 불거진 LG에너지솔루션의 상황을 감안하면 CATL을 선택할 수밖에 없었을 것”이라고 말했다. 다만 당초 3차 물량의 핵심으로 꼽혔던 스포츠유틸리티(SUV) 전기차 ‘아이오닉7’에 탑재될 배터리 선정은 미뤄졌다. 업계에서는 현대차와 LG에너지솔루션이 인도네시아에 건설을 추진 중인 합작법인이 아이오닉7에 탑재할 배터리를 생산할 것이라는 관측이 우세하다. C배터리의 영토 확장은 한국 시장을 넘어 유럽을 정조준하고 있다. 메르세데스 벤츠를 생산하는 독일 완성차 업체 다임러 그룹은 지난해 CATL과 배터리 공급은 물론 연구 분야에서도 협력을 강화키로 했다. 업계에 따르면 양사는 향후 다임러 그룹에서 생산하는 모든 전기차에 사용되는 배터리를 CATL에 우선 발주하고 CATL이 수주를 거부하는 경우에만 다른 배터리 업체를 통해 공급받는 형태의 계약을 체결한 것으로 알려졌다. 이에 대해 업계 관계자는 “차종별로 배터리 업체를 선정하는 기존 방식에서 벗어난 파격적인 형태”라며 “말 그대로 CATL이 ‘수퍼 을’(갑을 뛰어넘는 영향력을 지닌 을)로 성장했음을 보여주는 대목”이라고 말했다. ‘세계의 자동차 시장’이라 불리는 중국의 막대한 전기차 생산량에다 정부의 전폭적인 지원을 바탕으로 성장한 CATL은 지난해 비중국 시장 점유율을 대폭 늘렸다. 배터리 시장조사 기관인 SNE리서치에 따르면 2019년 중국을 제외한 글로벌 전기차 시장에서 CATL 배터리 사용량은 0.2GWh(기가와트시)로 LG에너지솔루션(12.3GWh), 삼성SDI(4.3GWh), SK이노베이션(2.1GWh)에 비해 초라한 수준이었다. 하지만 지난해 5.3GWh로 무려 2,457.1%의 성장률을 기록했다. 점유율 역시 0.4%에서 6.5%로 껑충 뛰었다. CATL은 또 상상을 초월하는 생산능력을 갖춰 전기차 배터리 시장의 최상위 포식자로 군림하겠다는 계획을 세우고 있다. CATL이 현재 국내·외에서 진행 중인 배터리 공장 건설 프로젝트를 종합하면 자체 생산과 합작사를 합쳐 2025년엔 486.8GWh, 2030년엔 591.8GWh의 생산능력을 보유하게 된다. 2020년 글로벌 시장 전체 전기차용 배터리 사용량이 142.8GWh인 점을 감안하면 어마어마한 규모다. 업계 관계자는 “올해 예정된 글로벌 자동차 회사들의 배터리 공급사 입찰 규모는 약 1.4TWh(테라와트시)로 지난해 판매된 전기차 총 배터리 용량의 10배에 달한다”며 “수주전 결과에 따라 배터리 업체들의 향후 수년간 순위표에 지각 변동이 생길 수 있다는 뜻”이라고 말했다. 이어 “전기차 시장을 선도하는 유럽까지 C배터리가 침투한 만큼 K배터리는 신흥 시장이자 중국의 진출이 제한되는 미국 시장을 선점할 필요가 있다”면서 “LG와 SK가 빠른 시일 내에 합의를 하고 선의의 경쟁을 통해 K배터리의 경쟁력을 키워야 할 때”라고 강조했다."]}, {"source_sentence": "지역 예술인을 위한 공모를 시행했던 프로젝트는 어떤 것에 일부분이야?", "sentences": ["충남 공주시는 제민천과 국고개 일원을 대상으로 지난해 8월부터 추진한 공공미술 프로젝트 사업이 완료됐다고 21일 밝혔다. 문화체육관광부가 주최 공주시가 주관한 공공미술 프로젝트는 지역예술인 일자리 제공 및 주민 문화 향유 증진이라는 취지로 전국 228개 지자체 대상으로 추진됐다. 이에 시는 지난해 8월 공모를 통해 한국미술협회(공주지부)와 충남공예협동조합 2개팀을 선정 지역작가 37명의 참여를 통해 진행해 '중동교~대통교'사이에 조명과 벤치, 포토존 그리고 주민참여로 만들어진 송사리 조형물을 설치해 주민들에게 휴식 공간을 제공했다. 그리고 '국고개 흐르는 효심'은 도조작품, 화분조형, 이복이야기 동화책 및 국고개 가이드북 만들기 등 커뮤니티 프로그램 운영 역사.문화.예술 중심지의 특성을 살리는데 중점을 뒀다. 황의정 문화체육과장은 \"앞으로도 시민들이 언제 어디서나 문화예술을 누리고, 지역예술인에게는 일자리를 제공할 수 있는 다양한 사업을 진행하겠다\"고 말했다.", "배우 신성록이 여권 사진을 공개해 시선을 모았다. 18일 오후 방송된 SBS 예능 프로그램 ‘집사부일체’에는 김종국이 사부로 출연했다. 사부를 만나기 전, 출연진은 가고 싶었던 바캉스 장소에 대해 이야기했다. 차은우는 “하와이에 가고 싶다”고 했고, 이승기는 “제주도 여행이 간절하다”고 말했다. 양세형은 “나도 제주도”라고 외쳤다. 양세형은 “은우의 여권 사진이 궁금하다”며 호기심을 드러냈다. 차은우의 여권 사진이 공개되자, 이승기는 “프로필 사진이냐”며 감탄했다. 양세형은 “모델 같다”고 말했다. 신성록은 “동네 사진관에서 여권 사진을 찍었다. 보정이 심해서 AI 같다”며 자신의 사진을 공개했다. 그의 여권 사진을 본 양세형은 “실물이 더 잘생겼다”고 말했다. 한편 ‘집사부일체’는 매주 일요일 오후 6시 25분에 방송된다.", "마산의 원도심 성호동의 역사가 예술로 기록되고 되살아난다. 시대를 살아낸 사람들의 애틋한 이야기가 아름다운 소리, 몸짓, 조형, 화첩에 담겨 시민을 만난다. 창원시와 창원시문화도시지원센터는 오는 30일 오후 2시 마산합포구 문신길 일대에서 성호동과 그 주변 지역주민들의 삶, 지리적, 환경적 문화를 예술에 담은 〈사회참여예술창작지원사업〉 ‘예술로 성호창발(城湖創發)’ 창작 발표회를 갖는다고 밝혔다. 문화특화지역조성사업의 일환으로 진행하는 이 사업은 생성과 소멸의 과정을 거쳐 성장해온 도심 속 흩어져 있는 다양한 문화를 예술로 기록하고 후대에 물려주기 위한 취지로 순수예술 창작물을 구현하고 예술의 사회적 책임에 관해 고찰해 보는 사업이다. 앞으로 지역 예술인들과 뜻을 모아 소중한 원도심의 문화가 한순간 덧없이 사라지는 일이 없도록 매년 확대해 나갈 계획이다. 성호동을 대상으로 올해 처음 시행하는 ‘예술로 성호창발(城湖創發)’ 사업은 지난해 10월 지역 예술인들을 대상으로 참가자를 공개 모집한 결과, 사업 취지를 이해하고 창작 능력과 작품의 완성도가 기대되는 4개 단체를 선정하였다. 선정된 4개 단체 작가들은 스토리 개발을 위해 성호동 주민들을 만나 세밀한 대화를 나누기도 하고, 장르와 특성에 맞는 창작품을 만들기 위해 성호동 구석구석을 누비는 열정을 보였다. 발표회는 1월 30일 오후 2시부터 4시 30분까지 ‘성호동 이바구 투어’라는 이름으로 마산박물관 앞에 모여 문신 길을 따라 추산창작공작소 4933을 거쳐 임항선 그린웨이, 그리고 성호동 생활문화센터를 지나 정법사로 이동하면서 펼쳐진다. 〈(사)경남국악관현악단 ‘휴’〉가 ‘추산창작공작소 4933’에서 동네 이야기를 판소리로 엮어 국악 관현악과 경서도 소리를 통해 성호동의 옛이야기를 펼치고, 는 ‘성호동의 역사와 추억을 담다’란 주제로 도시가 발달하면서 퇴색되어가는 성호동 동네잔치의 상징적 의미를 담은 조형물을 설치했다. 〈신미란 작가〉는 ‘성호동 생활문화센터 2층 전시장’에서 성호동 사람과 삶의 이야기(話)와 그림(畵)을 표현한 ‘성호화첩(成湖話畵)’을 발표하며, 〈박은혜 춤패〉는 ‘일란 이필이의 춤을 찾아가는 소소한 춤 여행’이란 주제로 이필이 선생의 춤을 추며 다큐멘터리 시사회를 가지는 것으로 전체 일정을 마친다. 공식 행사 외의 진행은 성호동 토박이 김경년 창원시도시재생지원센터 해설사가 맡아 행사장을 이동하는 중간중간 맛깔나는 이바구로 성호동의 역사를 들려준다. 심재욱 창원시 문화체육관광국장은 “도시는 사람의 역사를 품은 곳으로 이제 사람이 도시를 품어야 문화예술로 지속 성장의 가능성을 보여줄 수 있다”라며 문화의 힘을 강조했다. 한편 행사 참가는 사회적 거리두기를 위해 참석 인원을 20명으로 제한하여 운영한다."]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,258 |
gokuls/hbertv1-emotion-logit_KD-small
|
gokuls
|
text-classification
|
[
"transformers",
"tensorboard",
"safetensors",
"hybridbert",
"text-classification",
"generated_from_trainer",
"dataset:emotion",
"base_model:gokuls/model_v1_complete_training_wt_init_48_small_freeze_new",
"base_model:finetune:gokuls/model_v1_complete_training_wt_init_48_small_freeze_new",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | 2023-11-26T00:56:49Z |
2023-11-26T01:00:53+00:00
| 7 | 0 |
---
base_model: gokuls/model_v1_complete_training_wt_init_48_small_freeze_new
datasets:
- emotion
metrics:
- accuracy
tags:
- generated_from_trainer
model-index:
- name: hbertv1-emotion-logit_KD-small
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: emotion
type: emotion
config: split
split: validation
args: split
metrics:
- type: accuracy
value: 0.9335
name: Accuracy
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hbertv1-emotion-logit_KD-small
This model is a fine-tuned version of [gokuls/model_v1_complete_training_wt_init_48_small_freeze_new](https://huggingface.co/gokuls/model_v1_complete_training_wt_init_48_small_freeze_new) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2473
- Accuracy: 0.9335
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 1.4023 | 1.0 | 250 | 0.5204 | 0.8825 |
| 0.3903 | 2.0 | 500 | 0.3014 | 0.91 |
| 0.2438 | 3.0 | 750 | 0.2849 | 0.9185 |
| 0.1778 | 4.0 | 1000 | 0.2489 | 0.9265 |
| 0.1394 | 5.0 | 1250 | 0.2878 | 0.9205 |
| 0.1218 | 6.0 | 1500 | 0.2887 | 0.923 |
| 0.1083 | 7.0 | 1750 | 0.2788 | 0.9285 |
| 0.1019 | 8.0 | 2000 | 0.2373 | 0.928 |
| 0.0898 | 9.0 | 2250 | 0.2473 | 0.9335 |
| 0.0817 | 10.0 | 2500 | 0.2822 | 0.926 |
| 0.0827 | 11.0 | 2750 | 0.2474 | 0.926 |
| 0.0733 | 12.0 | 3000 | 0.2329 | 0.9285 |
| 0.0631 | 13.0 | 3250 | 0.2301 | 0.929 |
| 0.06 | 14.0 | 3500 | 0.2565 | 0.9295 |
### Framework versions
- Transformers 4.35.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.15.0
- Tokenizers 0.15.0
| null |
Non_BioNLP
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# hbertv1-emotion-logit_KD-small
This model is a fine-tuned version of [gokuls/model_v1_complete_training_wt_init_48_small_freeze_new](https://huggingface.co/gokuls/model_v1_complete_training_wt_init_48_small_freeze_new) on the emotion dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2473
- Accuracy: 0.9335
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 1.4023 | 1.0 | 250 | 0.5204 | 0.8825 |
| 0.3903 | 2.0 | 500 | 0.3014 | 0.91 |
| 0.2438 | 3.0 | 750 | 0.2849 | 0.9185 |
| 0.1778 | 4.0 | 1000 | 0.2489 | 0.9265 |
| 0.1394 | 5.0 | 1250 | 0.2878 | 0.9205 |
| 0.1218 | 6.0 | 1500 | 0.2887 | 0.923 |
| 0.1083 | 7.0 | 1750 | 0.2788 | 0.9285 |
| 0.1019 | 8.0 | 2000 | 0.2373 | 0.928 |
| 0.0898 | 9.0 | 2250 | 0.2473 | 0.9335 |
| 0.0817 | 10.0 | 2500 | 0.2822 | 0.926 |
| 0.0827 | 11.0 | 2750 | 0.2474 | 0.926 |
| 0.0733 | 12.0 | 3000 | 0.2329 | 0.9285 |
| 0.0631 | 13.0 | 3250 | 0.2301 | 0.929 |
| 0.06 | 14.0 | 3500 | 0.2565 | 0.9295 |
### Framework versions
- Transformers 4.35.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.15.0
- Tokenizers 0.15.0
|
{"base_model": "gokuls/model_v1_complete_training_wt_init_48_small_freeze_new", "datasets": ["emotion"], "metrics": ["accuracy"], "tags": ["generated_from_trainer"], "model-index": [{"name": "hbertv1-emotion-logit_KD-small", "results": [{"task": {"type": "text-classification", "name": "Text Classification"}, "dataset": {"name": "emotion", "type": "emotion", "config": "split", "split": "validation", "args": "split"}, "metrics": [{"type": "accuracy", "value": 0.9335, "name": "Accuracy"}]}]}]}
|
task
|
[
"TEXT_CLASSIFICATION"
] | 45,259 |
koalajun/Gemma-2-9b-it-Ko-Crypto-Translate
|
koalajun
|
text-generation
|
[
"transformers",
"safetensors",
"gemma2",
"text-generation",
"conversational",
"en",
"ko",
"base_model:google/gemma-2-9b-it",
"base_model:finetune:google/gemma-2-9b-it",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | 2024-10-03T12:09:00Z |
2024-10-03T12:41:05+00:00
| 7 | 1 |
---
base_model:
- google/gemma-2-9b-it
language:
- en
- ko
library_name: transformers
license: mit
---
# Model Card for Gemma-2-9b-it-Ko-Crypto-Translate
This model has been fine-tuned on a crypto news translation task. It is designed to translate English crypto news into Korean, leveraging the Gemma-2-9b-it architecture. The model is intended for natural language processing (NLP) tasks, specifically translation, within the crypto news domain.
## Model Details
### Model Description
This fine-tuned model is based on the **Gemma-2-9b-it** architecture and has been specifically trained to translate English crypto news into Korean. Fine-tuning was performed using a custom dataset focused on cryptocurrency news articles, ensuring the model's output is accurate in both language translation and crypto-specific terminology.
- **Developed by:** Hyoun Jun Lee
- **Model type:** Gemma-2-9b-it
- **Language(s) (NLP):** English, Korean
### Model Sources
- **Repository:** [Hugging Face: koalajun/Gemma-2-9b-it-Ko-Crypto-Translate](https://huggingface.co/koalajun/Gemma-2-9b-it-Ko-Crypto-Translate)
## Uses
### Direct Use
This model can be used for translating English cryptocurrency news articles into Korean. It can be integrated into applications such as financial platforms or news websites to provide real-time translation of crypto news.
### Downstream Use
The model can be further fine-tuned for more specific translation tasks in the financial or legal domains. Additionally, it can be used as a basis for other translation or language generation tasks that require bilingual capabilities in English and Korean.
### Out-of-Scope Use
This model is not intended for general translation tasks outside the financial/crypto domain. It may not perform well in non-financial contexts, as it was fine-tuned with specialized crypto-related datasets.
## Bias, Risks, and Limitations
Given the specific nature of the dataset (crypto news), the model may introduce biases related to the financial or crypto sector. The translation might also be less effective for general or non-financial text, and there could be inaccuracies in domain-specific terms.
### Recommendations
Users should validate the model's output in critical applications, especially when used in real-time financial decision-making or for publications where accuracy is paramount.
## How to Get Started with the Model
To use this model for inference, you can load it using the Hugging Face `transformers` library as follows:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model_name = "koalajun/Gemma-2-9b-it-Ko-Crypto-Translate"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
device = "cuda" if torch.cuda.is_available() else "cpu"
model = model.to(device)
# Define the input prompt for testing
prompt = "Translate the latest crypto news from English to Korean: Bitcoin prices continue to rise, surpassing $30,000 this week."
# Tokenize the input prompt
inputs = tokenizer(prompt, return_tensors="pt").to(device)
# Generate response from the model
outputs = model.generate(inputs.input_ids, max_length=200, num_return_sequences=1)
# Decode and print the generated text (translation)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print("Translation:", response)
| null |
Non_BioNLP
|
# Model Card for Gemma-2-9b-it-Ko-Crypto-Translate
This model has been fine-tuned on a crypto news translation task. It is designed to translate English crypto news into Korean, leveraging the Gemma-2-9b-it architecture. The model is intended for natural language processing (NLP) tasks, specifically translation, within the crypto news domain.
## Model Details
### Model Description
This fine-tuned model is based on the **Gemma-2-9b-it** architecture and has been specifically trained to translate English crypto news into Korean. Fine-tuning was performed using a custom dataset focused on cryptocurrency news articles, ensuring the model's output is accurate in both language translation and crypto-specific terminology.
- **Developed by:** Hyoun Jun Lee
- **Model type:** Gemma-2-9b-it
- **Language(s) (NLP):** English, Korean
### Model Sources
- **Repository:** [Hugging Face: koalajun/Gemma-2-9b-it-Ko-Crypto-Translate](https://huggingface.co/koalajun/Gemma-2-9b-it-Ko-Crypto-Translate)
## Uses
### Direct Use
This model can be used for translating English cryptocurrency news articles into Korean. It can be integrated into applications such as financial platforms or news websites to provide real-time translation of crypto news.
### Downstream Use
The model can be further fine-tuned for more specific translation tasks in the financial or legal domains. Additionally, it can be used as a basis for other translation or language generation tasks that require bilingual capabilities in English and Korean.
### Out-of-Scope Use
This model is not intended for general translation tasks outside the financial/crypto domain. It may not perform well in non-financial contexts, as it was fine-tuned with specialized crypto-related datasets.
## Bias, Risks, and Limitations
Given the specific nature of the dataset (crypto news), the model may introduce biases related to the financial or crypto sector. The translation might also be less effective for general or non-financial text, and there could be inaccuracies in domain-specific terms.
### Recommendations
Users should validate the model's output in critical applications, especially when used in real-time financial decision-making or for publications where accuracy is paramount.
## How to Get Started with the Model
To use this model for inference, you can load it using the Hugging Face `transformers` library as follows:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model_name = "koalajun/Gemma-2-9b-it-Ko-Crypto-Translate"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
device = "cuda" if torch.cuda.is_available() else "cpu"
model = model.to(device)
# Define the input prompt for testing
prompt = "Translate the latest crypto news from English to Korean: Bitcoin prices continue to rise, surpassing $30,000 this week."
# Tokenize the input prompt
inputs = tokenizer(prompt, return_tensors="pt").to(device)
# Generate response from the model
outputs = model.generate(inputs.input_ids, max_length=200, num_return_sequences=1)
# Decode and print the generated text (translation)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print("Translation:", response)
|
{"base_model": ["google/gemma-2-9b-it"], "language": ["en", "ko"], "library_name": "transformers", "license": "mit"}
|
task
|
[
"TRANSLATION"
] | 45,260 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.