Update README.md
Browse files
README.md
CHANGED
@@ -4,7 +4,7 @@ base_model: Qwen/Qwen2-1.5B
|
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
model-index:
|
7 |
-
- name:
|
8 |
results: []
|
9 |
---
|
10 |
|
@@ -12,82 +12,6 @@ model-index:
|
|
12 |
should probably proofread and complete it, then remove this comment. -->
|
13 |
|
14 |
[<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)
|
15 |
-
<details><summary>See axolotl config</summary>
|
16 |
-
|
17 |
-
axolotl version: `0.4.1`
|
18 |
-
```yaml
|
19 |
-
base_model: Qwen/Qwen2-1.5B
|
20 |
-
trust_remote_code:
|
21 |
-
|
22 |
-
load_in_8bit: false
|
23 |
-
load_in_4bit: false
|
24 |
-
strict: false
|
25 |
-
|
26 |
-
datasets:
|
27 |
-
- path: arcee-ai/eval_tome
|
28 |
-
type: sharegpt
|
29 |
-
conversation: chatml
|
30 |
-
dataset_prepared_path:
|
31 |
-
val_set_size: 0.0
|
32 |
-
output_dir: ./outputs/qwen2-1.5b-super
|
33 |
-
|
34 |
-
sequence_len: 8192
|
35 |
-
sample_packing: true
|
36 |
-
eval_sample_packing: false
|
37 |
-
pad_to_sequence_len: true
|
38 |
-
|
39 |
-
adapter:
|
40 |
-
lora_model_dir:
|
41 |
-
lora_r:
|
42 |
-
lora_alpha:
|
43 |
-
lora_dropout:
|
44 |
-
lora_target_linear:
|
45 |
-
lora_fan_in_fan_out:
|
46 |
-
|
47 |
-
wandb_project: qwen2-1.5b
|
48 |
-
wandb_entity:
|
49 |
-
wandb_watch:
|
50 |
-
wandb_name:
|
51 |
-
wandb_log_model:
|
52 |
-
|
53 |
-
gradient_accumulation_steps: 4
|
54 |
-
micro_batch_size: 4
|
55 |
-
num_epochs: 5
|
56 |
-
optimizer: adamw_torch
|
57 |
-
lr_scheduler: cosine
|
58 |
-
learning_rate: 0.00005
|
59 |
-
|
60 |
-
train_on_inputs: false
|
61 |
-
group_by_length: false
|
62 |
-
bf16: auto
|
63 |
-
fp16:
|
64 |
-
tf32: true
|
65 |
-
|
66 |
-
gradient_checkpointing: true
|
67 |
-
early_stopping_patience:
|
68 |
-
resume_from_checkpoint:
|
69 |
-
local_rank:
|
70 |
-
logging_steps: 1
|
71 |
-
xformers_attention:
|
72 |
-
flash_attention: true
|
73 |
-
|
74 |
-
warmup_steps: 50
|
75 |
-
evals_per_epoch:
|
76 |
-
saves_per_epoch: 1
|
77 |
-
debug:
|
78 |
-
deepspeed: zero3_bf16.json
|
79 |
-
weight_decay: 0.0
|
80 |
-
special_tokens:
|
81 |
-
eos_token: "<|im_end|>"
|
82 |
-
bos_token: "<|im_start|>"
|
83 |
-
|
84 |
-
```
|
85 |
-
|
86 |
-
</details><br>
|
87 |
-
|
88 |
-
# outputs/qwen2-1.5b-super
|
89 |
-
|
90 |
-
This model is a fine-tuned version of [Qwen/Qwen2-1.5B](https://huggingface.co/Qwen/Qwen2-1.5B) on the None dataset.
|
91 |
|
92 |
## Model description
|
93 |
|
|
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
model-index:
|
7 |
+
- name: qnguyen3/WitchLM-1.5B
|
8 |
results: []
|
9 |
---
|
10 |
|
|
|
12 |
should probably proofread and complete it, then remove this comment. -->
|
13 |
|
14 |
[<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)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
|
16 |
## Model description
|
17 |
|