---
base_model: EleutherAI/pythia-160m-deduped
library_name: peft
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: outputs/lora-alpaca-pythia
  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. -->

[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>

axolotl version: `0.4.1`
```yaml
base_model: EleutherAI/pythia-160m-deduped
load_in_8bit: false
datasets:
  - path: teknium/GPT4-LLM-Cleaned
    type: alpaca
dataset_prepared_path:
val_set_size: 0.05
adapter: lora
lora_model_dir:
sequence_len: 512
lora_r: 16
lora_alpha: 32
lora_dropout: 0.05
lora_target_modules:
  - query_key_value
  - dense
  - dense_h_to_4h
  - dense_4h_to_h
lora_target_linear:
lora_fan_in_fan_out: true  # pythia/GPTNeoX lora specific
output_dir: ./outputs/lora-alpaca-pythia
gradient_accumulation_steps: 1
micro_batch_size: 16
num_epochs: 4
learning_rate: 0.000005
train_on_inputs: false
group_by_length: false
bf16: false
tf32: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
weight_decay: 0.1
evals_per_epoch: 4
logging_steps: 1
push_to_hub: tommyp111/pythia-160m-deduped-alpaca-lora
wandb_project: pythia-alpaca-lora
wandb_name: pythia-160m-grad-norm

optimizer: adamw_torch
adam_beta2: 0.95
adam_epsilon: 0.00001
max_grad_norm: 1.0
gradient_checkpointing: true
warmup_steps: 10000

```

</details><br>

# outputs/lora-alpaca-pythia

This model is a fine-tuned version of [EleutherAI/pythia-160m-deduped](https://huggingface.co/EleutherAI/pythia-160m-deduped) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.2758

## 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-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 10000
- num_epochs: 4

### Training results

| Training Loss | Epoch  | Step  | Validation Loss |
|:-------------:|:------:|:-----:|:---------------:|
| 8.3121        | 0.0003 | 1     | 28.8947         |
| 8.51          | 0.2502 | 798   | 28.8493         |
| 7.1252        | 0.5003 | 1596  | 28.6938         |
| 11.0054       | 0.7505 | 2394  | 27.8628         |
| 2.7374        | 1.0006 | 3192  | 5.7286          |
| 3.3225        | 1.2508 | 3990  | 3.8328          |
| 2.8093        | 1.5009 | 4788  | 3.0960          |
| 2.5311        | 1.7511 | 5586  | 2.7825          |
| 1.9888        | 2.0013 | 6384  | 2.6022          |
| 2.1802        | 2.2514 | 7182  | 2.4945          |
| 2.3964        | 2.5016 | 7980  | 2.3910          |
| 2.1141        | 2.7517 | 8778  | 2.3618          |
| 2.7874        | 3.0019 | 9576  | 2.3030          |
| 2.2354        | 3.2520 | 10374 | 2.2600          |
| 2.0795        | 3.5022 | 11172 | 2.2918          |
| 2.2697        | 3.7524 | 11970 | 2.2758          |


### Framework versions

- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1