Model save
Browse files- README.md +58 -0
- all_results.json +8 -0
- generation_config.json +14 -0
- train_results.json +8 -0
- trainer_state.json +560 -0
README.md
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen2.5-3B-Instruct
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Qwen2.5-3B-WebArena-Lite-SFT-epoch-5
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Qwen2.5-3B-WebArena-Lite-SFT-epoch-5
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct).
|
| 15 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
+
|
| 17 |
+
## Quick start
|
| 18 |
+
|
| 19 |
+
```python
|
| 20 |
+
from transformers import pipeline
|
| 21 |
+
|
| 22 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 23 |
+
generator = pipeline("text-generation", model="weizhepei/Qwen2.5-3B-WebArena-Lite-SFT-epoch-5", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/uva-llm/huggingface/runs/qhullwv6)
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with SFT.
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.16.0.dev0
|
| 38 |
+
- Transformers: 4.49.0
|
| 39 |
+
- Pytorch: 2.5.1
|
| 40 |
+
- Datasets: 3.4.1
|
| 41 |
+
- Tokenizers: 0.21.1
|
| 42 |
+
|
| 43 |
+
## Citations
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
Cite TRL as:
|
| 48 |
+
|
| 49 |
+
```bibtex
|
| 50 |
+
@misc{vonwerra2022trl,
|
| 51 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 52 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
|
| 53 |
+
year = 2020,
|
| 54 |
+
journal = {GitHub repository},
|
| 55 |
+
publisher = {GitHub},
|
| 56 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 57 |
+
}
|
| 58 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 626617246416896.0,
|
| 3 |
+
"train_loss": 0.08566334958414773,
|
| 4 |
+
"train_runtime": 7937.4817,
|
| 5 |
+
"train_samples": 9460,
|
| 6 |
+
"train_samples_per_second": 5.959,
|
| 7 |
+
"train_steps_per_second": 0.047
|
| 8 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.05,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.49.0"
|
| 14 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 626617246416896.0,
|
| 3 |
+
"train_loss": 0.08566334958414773,
|
| 4 |
+
"train_runtime": 7937.4817,
|
| 5 |
+
"train_samples": 9460,
|
| 6 |
+
"train_samples_per_second": 5.959,
|
| 7 |
+
"train_steps_per_second": 0.047
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,560 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 370,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06756756756756757,
|
| 13 |
+
"grad_norm": 4.227170409911558,
|
| 14 |
+
"learning_rate": 1.3157894736842106e-05,
|
| 15 |
+
"loss": 0.575,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.13513513513513514,
|
| 20 |
+
"grad_norm": 2.2776126764932063,
|
| 21 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 22 |
+
"loss": 0.378,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.20270270270270271,
|
| 27 |
+
"grad_norm": 1.8654646175129226,
|
| 28 |
+
"learning_rate": 3.9473684210526316e-05,
|
| 29 |
+
"loss": 0.2857,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2702702702702703,
|
| 34 |
+
"grad_norm": 1.380928609001772,
|
| 35 |
+
"learning_rate": 4.9999098771046674e-05,
|
| 36 |
+
"loss": 0.2447,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.33783783783783783,
|
| 41 |
+
"grad_norm": 3.241471282418704,
|
| 42 |
+
"learning_rate": 4.996756333769319e-05,
|
| 43 |
+
"loss": 0.2235,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.40540540540540543,
|
| 48 |
+
"grad_norm": 1.2985992616488145,
|
| 49 |
+
"learning_rate": 4.989103862744732e-05,
|
| 50 |
+
"loss": 0.2145,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.47297297297297297,
|
| 55 |
+
"grad_norm": 1.1067383831026958,
|
| 56 |
+
"learning_rate": 4.976967787416565e-05,
|
| 57 |
+
"loss": 0.1899,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.5405405405405406,
|
| 62 |
+
"grad_norm": 1.049570235777466,
|
| 63 |
+
"learning_rate": 4.9603724091852456e-05,
|
| 64 |
+
"loss": 0.1987,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.6081081081081081,
|
| 69 |
+
"grad_norm": 1.0331209710899183,
|
| 70 |
+
"learning_rate": 4.9393509588046036e-05,
|
| 71 |
+
"loss": 0.1912,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.6756756756756757,
|
| 76 |
+
"grad_norm": 0.925233682955553,
|
| 77 |
+
"learning_rate": 4.9139455298402656e-05,
|
| 78 |
+
"loss": 0.1746,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.7432432432432432,
|
| 83 |
+
"grad_norm": 0.7958783153604793,
|
| 84 |
+
"learning_rate": 4.884206994381078e-05,
|
| 85 |
+
"loss": 0.1822,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.8108108108108109,
|
| 90 |
+
"grad_norm": 0.9751386844667708,
|
| 91 |
+
"learning_rate": 4.8501949011723144e-05,
|
| 92 |
+
"loss": 0.1876,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.8783783783783784,
|
| 97 |
+
"grad_norm": 0.8402293921317799,
|
| 98 |
+
"learning_rate": 4.811977356374667e-05,
|
| 99 |
+
"loss": 0.1872,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.9459459459459459,
|
| 104 |
+
"grad_norm": 0.718701514552068,
|
| 105 |
+
"learning_rate": 4.769630887187782e-05,
|
| 106 |
+
"loss": 0.1676,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.0135135135135136,
|
| 111 |
+
"grad_norm": 0.5398531301606878,
|
| 112 |
+
"learning_rate": 4.72324028861141e-05,
|
| 113 |
+
"loss": 0.158,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.0810810810810811,
|
| 118 |
+
"grad_norm": 0.6189607559877081,
|
| 119 |
+
"learning_rate": 4.6728984536510454e-05,
|
| 120 |
+
"loss": 0.103,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.1486486486486487,
|
| 125 |
+
"grad_norm": 0.6977698037888556,
|
| 126 |
+
"learning_rate": 4.6187061873080264e-05,
|
| 127 |
+
"loss": 0.11,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.2162162162162162,
|
| 132 |
+
"grad_norm": 0.6702191594025242,
|
| 133 |
+
"learning_rate": 4.560772004726575e-05,
|
| 134 |
+
"loss": 0.0974,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.2837837837837838,
|
| 139 |
+
"grad_norm": 0.8375137882275626,
|
| 140 |
+
"learning_rate": 4.499211913901974e-05,
|
| 141 |
+
"loss": 0.0962,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.3513513513513513,
|
| 146 |
+
"grad_norm": 0.5623841444139801,
|
| 147 |
+
"learning_rate": 4.434149183384977e-05,
|
| 148 |
+
"loss": 0.0963,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.4189189189189189,
|
| 153 |
+
"grad_norm": 0.746823712074494,
|
| 154 |
+
"learning_rate": 4.3657140954476165e-05,
|
| 155 |
+
"loss": 0.0978,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.4864864864864864,
|
| 160 |
+
"grad_norm": 0.7241042591726438,
|
| 161 |
+
"learning_rate": 4.294043685204651e-05,
|
| 162 |
+
"loss": 0.0943,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.554054054054054,
|
| 167 |
+
"grad_norm": 0.8309567209936259,
|
| 168 |
+
"learning_rate": 4.219281466213066e-05,
|
| 169 |
+
"loss": 0.0997,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.6216216216216215,
|
| 174 |
+
"grad_norm": 0.7135108874545476,
|
| 175 |
+
"learning_rate": 4.141577143099074e-05,
|
| 176 |
+
"loss": 0.0954,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.689189189189189,
|
| 181 |
+
"grad_norm": 0.4901288548943104,
|
| 182 |
+
"learning_rate": 4.06108631178804e-05,
|
| 183 |
+
"loss": 0.0911,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.7567567567567568,
|
| 188 |
+
"grad_norm": 0.5126638837154042,
|
| 189 |
+
"learning_rate": 3.977970147937635e-05,
|
| 190 |
+
"loss": 0.0981,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 1.8243243243243243,
|
| 195 |
+
"grad_norm": 0.4361129957181847,
|
| 196 |
+
"learning_rate": 3.892395084198053e-05,
|
| 197 |
+
"loss": 0.0866,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.8918918918918919,
|
| 202 |
+
"grad_norm": 0.45933479640398944,
|
| 203 |
+
"learning_rate": 3.8045324769455834e-05,
|
| 204 |
+
"loss": 0.0815,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.9594594594594594,
|
| 209 |
+
"grad_norm": 0.6180056798090116,
|
| 210 |
+
"learning_rate": 3.714558263156872e-05,
|
| 211 |
+
"loss": 0.0853,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.027027027027027,
|
| 216 |
+
"grad_norm": 0.8867749784287353,
|
| 217 |
+
"learning_rate": 3.622652608110924e-05,
|
| 218 |
+
"loss": 0.0714,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.0945945945945947,
|
| 223 |
+
"grad_norm": 0.41595529690592503,
|
| 224 |
+
"learning_rate": 3.528999544624324e-05,
|
| 225 |
+
"loss": 0.0582,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 2.1621621621621623,
|
| 230 |
+
"grad_norm": 0.46173444521128193,
|
| 231 |
+
"learning_rate": 3.433786604542056e-05,
|
| 232 |
+
"loss": 0.0546,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.22972972972973,
|
| 237 |
+
"grad_norm": 0.8285015328803698,
|
| 238 |
+
"learning_rate": 3.337204443221834e-05,
|
| 239 |
+
"loss": 0.0459,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.2972972972972974,
|
| 244 |
+
"grad_norm": 0.40330243772956614,
|
| 245 |
+
"learning_rate": 3.2394464577638765e-05,
|
| 246 |
+
"loss": 0.058,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 2.364864864864865,
|
| 251 |
+
"grad_norm": 0.38712423069839835,
|
| 252 |
+
"learning_rate": 3.140708399750594e-05,
|
| 253 |
+
"loss": 0.0524,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.4324324324324325,
|
| 258 |
+
"grad_norm": 0.4789835728870257,
|
| 259 |
+
"learning_rate": 3.0411879832716246e-05,
|
| 260 |
+
"loss": 0.0526,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.5,
|
| 265 |
+
"grad_norm": 0.37205307961706024,
|
| 266 |
+
"learning_rate": 2.9410844890191458e-05,
|
| 267 |
+
"loss": 0.0507,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 2.5675675675675675,
|
| 272 |
+
"grad_norm": 0.6004132935282005,
|
| 273 |
+
"learning_rate": 2.840598365246184e-05,
|
| 274 |
+
"loss": 0.0527,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 2.635135135135135,
|
| 279 |
+
"grad_norm": 0.4560434366486544,
|
| 280 |
+
"learning_rate": 2.7399308263870027e-05,
|
| 281 |
+
"loss": 0.052,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 2.7027027027027026,
|
| 286 |
+
"grad_norm": 0.4624335921203337,
|
| 287 |
+
"learning_rate": 2.6392834501432746e-05,
|
| 288 |
+
"loss": 0.0462,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 2.77027027027027,
|
| 293 |
+
"grad_norm": 0.5498423698020634,
|
| 294 |
+
"learning_rate": 2.538857773842847e-05,
|
| 295 |
+
"loss": 0.0532,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 2.8378378378378377,
|
| 300 |
+
"grad_norm": 0.3705618200299343,
|
| 301 |
+
"learning_rate": 2.438854890879351e-05,
|
| 302 |
+
"loss": 0.0486,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 2.9054054054054053,
|
| 307 |
+
"grad_norm": 0.43411865881569756,
|
| 308 |
+
"learning_rate": 2.3394750480407467e-05,
|
| 309 |
+
"loss": 0.0522,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 2.972972972972973,
|
| 314 |
+
"grad_norm": 0.34856818385336974,
|
| 315 |
+
"learning_rate": 2.2409172445331257e-05,
|
| 316 |
+
"loss": 0.0433,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.0405405405405403,
|
| 321 |
+
"grad_norm": 0.3056885229191675,
|
| 322 |
+
"learning_rate": 2.143378833502677e-05,
|
| 323 |
+
"loss": 0.0367,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 3.108108108108108,
|
| 328 |
+
"grad_norm": 0.3437713870681847,
|
| 329 |
+
"learning_rate": 2.0470551268537457e-05,
|
| 330 |
+
"loss": 0.0264,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 3.175675675675676,
|
| 335 |
+
"grad_norm": 0.336480003243831,
|
| 336 |
+
"learning_rate": 1.9521390041542946e-05,
|
| 337 |
+
"loss": 0.0311,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.2432432432432434,
|
| 342 |
+
"grad_norm": 0.3411706759863995,
|
| 343 |
+
"learning_rate": 1.8588205264118974e-05,
|
| 344 |
+
"loss": 0.0269,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.310810810810811,
|
| 349 |
+
"grad_norm": 0.34062033026640975,
|
| 350 |
+
"learning_rate": 1.7672865554936467e-05,
|
| 351 |
+
"loss": 0.0281,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 3.3783783783783785,
|
| 356 |
+
"grad_norm": 0.3695748741767915,
|
| 357 |
+
"learning_rate": 1.6777203799520573e-05,
|
| 358 |
+
"loss": 0.027,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 3.445945945945946,
|
| 363 |
+
"grad_norm": 0.3601998041543281,
|
| 364 |
+
"learning_rate": 1.5903013480062085e-05,
|
| 365 |
+
"loss": 0.0282,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 3.5135135135135136,
|
| 370 |
+
"grad_norm": 0.45887851685959974,
|
| 371 |
+
"learning_rate": 1.5052045084130445e-05,
|
| 372 |
+
"loss": 0.0273,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 3.581081081081081,
|
| 377 |
+
"grad_norm": 0.35229536201704065,
|
| 378 |
+
"learning_rate": 1.4226002599479743e-05,
|
| 379 |
+
"loss": 0.0295,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 3.6486486486486487,
|
| 384 |
+
"grad_norm": 0.3353310971308658,
|
| 385 |
+
"learning_rate": 1.3426540101966257e-05,
|
| 386 |
+
"loss": 0.0322,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.7162162162162162,
|
| 391 |
+
"grad_norm": 0.41483094692292416,
|
| 392 |
+
"learning_rate": 1.2655258443410268e-05,
|
| 393 |
+
"loss": 0.0311,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 3.7837837837837838,
|
| 398 |
+
"grad_norm": 0.36712000215058754,
|
| 399 |
+
"learning_rate": 1.1913702046034016e-05,
|
| 400 |
+
"loss": 0.0289,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 3.8513513513513513,
|
| 405 |
+
"grad_norm": 0.5646368352969865,
|
| 406 |
+
"learning_rate": 1.1203355809895042e-05,
|
| 407 |
+
"loss": 0.0274,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 3.918918918918919,
|
| 412 |
+
"grad_norm": 0.4158288527920725,
|
| 413 |
+
"learning_rate": 1.0525642139507222e-05,
|
| 414 |
+
"loss": 0.028,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 3.9864864864864864,
|
| 419 |
+
"grad_norm": 0.3573187154947232,
|
| 420 |
+
"learning_rate": 9.881918095603501e-06,
|
| 421 |
+
"loss": 0.0276,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.054054054054054,
|
| 426 |
+
"grad_norm": 0.2472522738677183,
|
| 427 |
+
"learning_rate": 9.273472677743755e-06,
|
| 428 |
+
"loss": 0.0179,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.121621621621622,
|
| 433 |
+
"grad_norm": 0.20424848415230237,
|
| 434 |
+
"learning_rate": 8.701524243208935e-06,
|
| 435 |
+
"loss": 0.0142,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 4.1891891891891895,
|
| 440 |
+
"grad_norm": 0.2706498097595218,
|
| 441 |
+
"learning_rate": 8.167218067350122e-06,
|
| 442 |
+
"loss": 0.0174,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 4.256756756756757,
|
| 447 |
+
"grad_norm": 0.228170366357337,
|
| 448 |
+
"learning_rate": 7.67162405027753e-06,
|
| 449 |
+
"loss": 0.0149,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 4.324324324324325,
|
| 454 |
+
"grad_norm": 0.2817486193345113,
|
| 455 |
+
"learning_rate": 7.21573457448164e-06,
|
| 456 |
+
"loss": 0.0148,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 4.391891891891892,
|
| 461 |
+
"grad_norm": 0.22979041725844782,
|
| 462 |
+
"learning_rate": 6.800462517676456e-06,
|
| 463 |
+
"loss": 0.0133,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 4.45945945945946,
|
| 468 |
+
"grad_norm": 0.25690480860866444,
|
| 469 |
+
"learning_rate": 6.426639424843892e-06,
|
| 470 |
+
"loss": 0.0155,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 4.527027027027027,
|
| 475 |
+
"grad_norm": 0.30615645251349044,
|
| 476 |
+
"learning_rate": 6.095013843139662e-06,
|
| 477 |
+
"loss": 0.0142,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 4.594594594594595,
|
| 482 |
+
"grad_norm": 0.24160715474139485,
|
| 483 |
+
"learning_rate": 5.806249822994849e-06,
|
| 484 |
+
"loss": 0.0134,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 4.662162162162162,
|
| 489 |
+
"grad_norm": 0.24906080295211588,
|
| 490 |
+
"learning_rate": 5.560925588414595e-06,
|
| 491 |
+
"loss": 0.0133,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 4.72972972972973,
|
| 496 |
+
"grad_norm": 0.4021706429563905,
|
| 497 |
+
"learning_rate": 5.359532379136472e-06,
|
| 498 |
+
"loss": 0.0137,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 4.797297297297297,
|
| 503 |
+
"grad_norm": 0.29394464912604296,
|
| 504 |
+
"learning_rate": 5.20247346696706e-06,
|
| 505 |
+
"loss": 0.0139,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 4.864864864864865,
|
| 510 |
+
"grad_norm": 0.31733228546671827,
|
| 511 |
+
"learning_rate": 5.090063348266363e-06,
|
| 512 |
+
"loss": 0.0158,
|
| 513 |
+
"step": 360
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 4.9324324324324325,
|
| 517 |
+
"grad_norm": 0.33778908257293244,
|
| 518 |
+
"learning_rate": 5.022527114197126e-06,
|
| 519 |
+
"loss": 0.0143,
|
| 520 |
+
"step": 365
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 5.0,
|
| 524 |
+
"grad_norm": 0.28233495289671173,
|
| 525 |
+
"learning_rate": 5e-06,
|
| 526 |
+
"loss": 0.0131,
|
| 527 |
+
"step": 370
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 5.0,
|
| 531 |
+
"step": 370,
|
| 532 |
+
"total_flos": 626617246416896.0,
|
| 533 |
+
"train_loss": 0.08566334958414773,
|
| 534 |
+
"train_runtime": 7937.4817,
|
| 535 |
+
"train_samples_per_second": 5.959,
|
| 536 |
+
"train_steps_per_second": 0.047
|
| 537 |
+
}
|
| 538 |
+
],
|
| 539 |
+
"logging_steps": 5,
|
| 540 |
+
"max_steps": 370,
|
| 541 |
+
"num_input_tokens_seen": 0,
|
| 542 |
+
"num_train_epochs": 5,
|
| 543 |
+
"save_steps": 500,
|
| 544 |
+
"stateful_callbacks": {
|
| 545 |
+
"TrainerControl": {
|
| 546 |
+
"args": {
|
| 547 |
+
"should_epoch_stop": false,
|
| 548 |
+
"should_evaluate": false,
|
| 549 |
+
"should_log": false,
|
| 550 |
+
"should_save": true,
|
| 551 |
+
"should_training_stop": true
|
| 552 |
+
},
|
| 553 |
+
"attributes": {}
|
| 554 |
+
}
|
| 555 |
+
},
|
| 556 |
+
"total_flos": 626617246416896.0,
|
| 557 |
+
"train_batch_size": 16,
|
| 558 |
+
"trial_name": null,
|
| 559 |
+
"trial_params": null
|
| 560 |
+
}
|