zhuangxialie commited on
Model save
Browse files- README.md +57 -0
- all_results.json +8 -0
- generation_config.json +14 -0
- train_results.json +8 -0
- trainer_state.json +714 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
model_name: Qwen-code-7B-SFT
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
licence: license
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# Model Card for Qwen-code-7B-SFT
|
| 12 |
+
|
| 13 |
+
This model is a fine-tuned version of [None](https://huggingface.co/None).
|
| 14 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 15 |
+
|
| 16 |
+
## Quick start
|
| 17 |
+
|
| 18 |
+
```python
|
| 19 |
+
from transformers import pipeline
|
| 20 |
+
|
| 21 |
+
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?"
|
| 22 |
+
generator = pipeline("text-generation", model="ZhuangXialie/Qwen-code-7B-SFT", device="cuda")
|
| 23 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 24 |
+
print(output["generated_text"])
|
| 25 |
+
```
|
| 26 |
+
|
| 27 |
+
## Training procedure
|
| 28 |
+
|
| 29 |
+
[<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/zhuangxialie-ubiquant-investment/huggingface/runs/w7fzacdr)
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.16.0.dev0
|
| 37 |
+
- Transformers: 4.49.0
|
| 38 |
+
- Pytorch: 2.4.0
|
| 39 |
+
- Datasets: 3.5.0
|
| 40 |
+
- Tokenizers: 0.21.1
|
| 41 |
+
|
| 42 |
+
## Citations
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
Cite TRL as:
|
| 47 |
+
|
| 48 |
+
```bibtex
|
| 49 |
+
@misc{vonwerra2022trl,
|
| 50 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 51 |
+
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},
|
| 52 |
+
year = 2020,
|
| 53 |
+
journal = {GitHub repository},
|
| 54 |
+
publisher = {GitHub},
|
| 55 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 56 |
+
}
|
| 57 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 216761967837184.0,
|
| 3 |
+
"train_loss": 0.20915592369579133,
|
| 4 |
+
"train_runtime": 2440.0132,
|
| 5 |
+
"train_samples": 19815,
|
| 6 |
+
"train_samples_per_second": 2.749,
|
| 7 |
+
"train_steps_per_second": 0.172
|
| 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": 1.1,
|
| 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": 216761967837184.0,
|
| 3 |
+
"train_loss": 0.20915592369579133,
|
| 4 |
+
"train_runtime": 2440.0132,
|
| 5 |
+
"train_samples": 19815,
|
| 6 |
+
"train_samples_per_second": 2.749,
|
| 7 |
+
"train_steps_per_second": 0.172
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,714 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 420,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.047619047619047616,
|
| 13 |
+
"grad_norm": 1.4257669918070304,
|
| 14 |
+
"learning_rate": 5.9523809523809525e-06,
|
| 15 |
+
"loss": 0.5346,
|
| 16 |
+
"mean_token_accuracy": 0.8659344732761383,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.09523809523809523,
|
| 21 |
+
"grad_norm": 0.7751540276308181,
|
| 22 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 23 |
+
"loss": 0.4535,
|
| 24 |
+
"mean_token_accuracy": 0.8758046269416809,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.14285714285714285,
|
| 29 |
+
"grad_norm": 0.3688156015884892,
|
| 30 |
+
"learning_rate": 1.785714285714286e-05,
|
| 31 |
+
"loss": 0.4071,
|
| 32 |
+
"mean_token_accuracy": 0.8853001713752746,
|
| 33 |
+
"step": 15
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.19047619047619047,
|
| 37 |
+
"grad_norm": 0.2571090114484444,
|
| 38 |
+
"learning_rate": 2.380952380952381e-05,
|
| 39 |
+
"loss": 0.3811,
|
| 40 |
+
"mean_token_accuracy": 0.8898558735847473,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.23809523809523808,
|
| 45 |
+
"grad_norm": 0.2445335136114927,
|
| 46 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 47 |
+
"loss": 0.3714,
|
| 48 |
+
"mean_token_accuracy": 0.8922576308250427,
|
| 49 |
+
"step": 25
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.2857142857142857,
|
| 53 |
+
"grad_norm": 0.25497240173448055,
|
| 54 |
+
"learning_rate": 3.571428571428572e-05,
|
| 55 |
+
"loss": 0.3621,
|
| 56 |
+
"mean_token_accuracy": 0.8938062608242034,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.3333333333333333,
|
| 61 |
+
"grad_norm": 0.22758565353000912,
|
| 62 |
+
"learning_rate": 4.166666666666667e-05,
|
| 63 |
+
"loss": 0.3572,
|
| 64 |
+
"mean_token_accuracy": 0.8950600028038025,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.38095238095238093,
|
| 69 |
+
"grad_norm": 0.21108211135784025,
|
| 70 |
+
"learning_rate": 4.761904761904762e-05,
|
| 71 |
+
"loss": 0.3463,
|
| 72 |
+
"mean_token_accuracy": 0.8972749292850495,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.42857142857142855,
|
| 77 |
+
"grad_norm": 0.2207672207804811,
|
| 78 |
+
"learning_rate": 4.999300659501837e-05,
|
| 79 |
+
"loss": 0.3443,
|
| 80 |
+
"mean_token_accuracy": 0.8975539267063141,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.47619047619047616,
|
| 85 |
+
"grad_norm": 0.21926886992248296,
|
| 86 |
+
"learning_rate": 4.995028486261366e-05,
|
| 87 |
+
"loss": 0.3378,
|
| 88 |
+
"mean_token_accuracy": 0.899033111333847,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.5238095238095238,
|
| 93 |
+
"grad_norm": 0.224235174049935,
|
| 94 |
+
"learning_rate": 4.986880029700503e-05,
|
| 95 |
+
"loss": 0.3403,
|
| 96 |
+
"mean_token_accuracy": 0.8979385972023011,
|
| 97 |
+
"step": 55
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.5714285714285714,
|
| 101 |
+
"grad_norm": 0.21434998232271477,
|
| 102 |
+
"learning_rate": 4.974869359006539e-05,
|
| 103 |
+
"loss": 0.3297,
|
| 104 |
+
"mean_token_accuracy": 0.9011363804340362,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.6190476190476191,
|
| 109 |
+
"grad_norm": 0.22126533026421416,
|
| 110 |
+
"learning_rate": 4.959017211895173e-05,
|
| 111 |
+
"loss": 0.3329,
|
| 112 |
+
"mean_token_accuracy": 0.8999340653419494,
|
| 113 |
+
"step": 65
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.6666666666666666,
|
| 117 |
+
"grad_norm": 0.2122647717862151,
|
| 118 |
+
"learning_rate": 4.9393509588046036e-05,
|
| 119 |
+
"loss": 0.3306,
|
| 120 |
+
"mean_token_accuracy": 0.9009318828582764,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7142857142857143,
|
| 125 |
+
"grad_norm": 0.20764297035844217,
|
| 126 |
+
"learning_rate": 4.915904555637527e-05,
|
| 127 |
+
"loss": 0.33,
|
| 128 |
+
"mean_token_accuracy": 0.9004346609115601,
|
| 129 |
+
"step": 75
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.7619047619047619,
|
| 133 |
+
"grad_norm": 0.20816077188373128,
|
| 134 |
+
"learning_rate": 4.8887184851326155e-05,
|
| 135 |
+
"loss": 0.3202,
|
| 136 |
+
"mean_token_accuracy": 0.9034851253032684,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.8095238095238095,
|
| 141 |
+
"grad_norm": 0.22148078953488418,
|
| 142 |
+
"learning_rate": 4.8578396869667095e-05,
|
| 143 |
+
"loss": 0.3252,
|
| 144 |
+
"mean_token_accuracy": 0.9017885386943817,
|
| 145 |
+
"step": 85
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.8571428571428571,
|
| 149 |
+
"grad_norm": 0.21406123617550696,
|
| 150 |
+
"learning_rate": 4.823321476708417e-05,
|
| 151 |
+
"loss": 0.3207,
|
| 152 |
+
"mean_token_accuracy": 0.9030977070331574,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.9047619047619048,
|
| 157 |
+
"grad_norm": 0.20063414494959075,
|
| 158 |
+
"learning_rate": 4.7852234537630435e-05,
|
| 159 |
+
"loss": 0.3151,
|
| 160 |
+
"mean_token_accuracy": 0.9040939033031463,
|
| 161 |
+
"step": 95
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.9523809523809523,
|
| 165 |
+
"grad_norm": 0.20157178980104304,
|
| 166 |
+
"learning_rate": 4.7436113984678024e-05,
|
| 167 |
+
"loss": 0.3133,
|
| 168 |
+
"mean_token_accuracy": 0.9044515311717987,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.0,
|
| 173 |
+
"grad_norm": 0.21946571056086206,
|
| 174 |
+
"learning_rate": 4.6985571585149876e-05,
|
| 175 |
+
"loss": 0.3088,
|
| 176 |
+
"mean_token_accuracy": 0.9057298839092255,
|
| 177 |
+
"step": 105
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.0476190476190477,
|
| 181 |
+
"grad_norm": 0.22243550466680892,
|
| 182 |
+
"learning_rate": 4.650138524899193e-05,
|
| 183 |
+
"loss": 0.2589,
|
| 184 |
+
"mean_token_accuracy": 0.9184723615646362,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.0952380952380953,
|
| 189 |
+
"grad_norm": 0.21782596690450573,
|
| 190 |
+
"learning_rate": 4.5984390976027946e-05,
|
| 191 |
+
"loss": 0.2467,
|
| 192 |
+
"mean_token_accuracy": 0.9217934429645538,
|
| 193 |
+
"step": 115
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.1428571428571428,
|
| 197 |
+
"grad_norm": 0.20047942498513227,
|
| 198 |
+
"learning_rate": 4.5435481412515755e-05,
|
| 199 |
+
"loss": 0.2458,
|
| 200 |
+
"mean_token_accuracy": 0.9213689804077149,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.1904761904761905,
|
| 205 |
+
"grad_norm": 0.20323781517968,
|
| 206 |
+
"learning_rate": 4.4855604309897496e-05,
|
| 207 |
+
"loss": 0.2533,
|
| 208 |
+
"mean_token_accuracy": 0.9193067252635956,
|
| 209 |
+
"step": 125
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.2380952380952381,
|
| 213 |
+
"grad_norm": 0.2089213813676996,
|
| 214 |
+
"learning_rate": 4.4245760888404665e-05,
|
| 215 |
+
"loss": 0.2451,
|
| 216 |
+
"mean_token_accuracy": 0.921397477388382,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 1.2857142857142856,
|
| 221 |
+
"grad_norm": 0.20871576941556397,
|
| 222 |
+
"learning_rate": 4.360700410834367e-05,
|
| 223 |
+
"loss": 0.2487,
|
| 224 |
+
"mean_token_accuracy": 0.9207526803016662,
|
| 225 |
+
"step": 135
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.3333333333333333,
|
| 229 |
+
"grad_norm": 0.20056126493311283,
|
| 230 |
+
"learning_rate": 4.294043685204651e-05,
|
| 231 |
+
"loss": 0.2361,
|
| 232 |
+
"mean_token_accuracy": 0.924414473772049,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.380952380952381,
|
| 237 |
+
"grad_norm": 0.2115275835238231,
|
| 238 |
+
"learning_rate": 4.224721001962573e-05,
|
| 239 |
+
"loss": 0.2409,
|
| 240 |
+
"mean_token_accuracy": 0.9230122208595276,
|
| 241 |
+
"step": 145
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.4285714285714286,
|
| 245 |
+
"grad_norm": 0.20541500240177765,
|
| 246 |
+
"learning_rate": 4.1528520541821506e-05,
|
| 247 |
+
"loss": 0.2398,
|
| 248 |
+
"mean_token_accuracy": 0.9232278406620026,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.4761904761904763,
|
| 253 |
+
"grad_norm": 0.20798560704125774,
|
| 254 |
+
"learning_rate": 4.078560931337187e-05,
|
| 255 |
+
"loss": 0.2471,
|
| 256 |
+
"mean_token_accuracy": 0.9215823113918304,
|
| 257 |
+
"step": 155
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.5238095238095237,
|
| 261 |
+
"grad_norm": 0.1939929705798728,
|
| 262 |
+
"learning_rate": 4.001975905047442e-05,
|
| 263 |
+
"loss": 0.2401,
|
| 264 |
+
"mean_token_accuracy": 0.923470401763916,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.5714285714285714,
|
| 269 |
+
"grad_norm": 0.21344717934526472,
|
| 270 |
+
"learning_rate": 3.923229207603871e-05,
|
| 271 |
+
"loss": 0.2396,
|
| 272 |
+
"mean_token_accuracy": 0.9238272488117218,
|
| 273 |
+
"step": 165
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.619047619047619,
|
| 277 |
+
"grad_norm": 0.2191275580698787,
|
| 278 |
+
"learning_rate": 3.842456803655342e-05,
|
| 279 |
+
"loss": 0.2406,
|
| 280 |
+
"mean_token_accuracy": 0.923265916109085,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.6666666666666665,
|
| 285 |
+
"grad_norm": 0.19421469398044988,
|
| 286 |
+
"learning_rate": 3.75979815545104e-05,
|
| 287 |
+
"loss": 0.2359,
|
| 288 |
+
"mean_token_accuracy": 0.924458909034729,
|
| 289 |
+
"step": 175
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.7142857142857144,
|
| 293 |
+
"grad_norm": 0.19470949781451338,
|
| 294 |
+
"learning_rate": 3.6753959820438764e-05,
|
| 295 |
+
"loss": 0.2321,
|
| 296 |
+
"mean_token_accuracy": 0.9256431758403778,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.7619047619047619,
|
| 301 |
+
"grad_norm": 0.1876708972166408,
|
| 302 |
+
"learning_rate": 3.589396012870687e-05,
|
| 303 |
+
"loss": 0.2291,
|
| 304 |
+
"mean_token_accuracy": 0.9266167640686035,
|
| 305 |
+
"step": 185
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.8095238095238095,
|
| 309 |
+
"grad_norm": 0.2147218447176946,
|
| 310 |
+
"learning_rate": 3.5019467361346724e-05,
|
| 311 |
+
"loss": 0.2329,
|
| 312 |
+
"mean_token_accuracy": 0.925212299823761,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.8571428571428572,
|
| 317 |
+
"grad_norm": 0.19962238819866412,
|
| 318 |
+
"learning_rate": 3.413199142424535e-05,
|
| 319 |
+
"loss": 0.2296,
|
| 320 |
+
"mean_token_accuracy": 0.9259015142917633,
|
| 321 |
+
"step": 195
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.9047619047619047,
|
| 325 |
+
"grad_norm": 0.19475591470139603,
|
| 326 |
+
"learning_rate": 3.3233064640129734e-05,
|
| 327 |
+
"loss": 0.2277,
|
| 328 |
+
"mean_token_accuracy": 0.92720667719841,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.9523809523809523,
|
| 333 |
+
"grad_norm": 0.20861501443310124,
|
| 334 |
+
"learning_rate": 3.232423910284672e-05,
|
| 335 |
+
"loss": 0.2284,
|
| 336 |
+
"mean_token_accuracy": 0.9266561150550843,
|
| 337 |
+
"step": 205
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 2.0,
|
| 341 |
+
"grad_norm": 0.19761114102987562,
|
| 342 |
+
"learning_rate": 3.140708399750594e-05,
|
| 343 |
+
"loss": 0.2291,
|
| 344 |
+
"mean_token_accuracy": 0.9258988976478577,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.0476190476190474,
|
| 349 |
+
"grad_norm": 0.24523566793415144,
|
| 350 |
+
"learning_rate": 3.048318289111279e-05,
|
| 351 |
+
"loss": 0.1563,
|
| 352 |
+
"mean_token_accuracy": 0.9481804072856903,
|
| 353 |
+
"step": 215
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 2.0952380952380953,
|
| 357 |
+
"grad_norm": 0.204067402376235,
|
| 358 |
+
"learning_rate": 2.955413099836959e-05,
|
| 359 |
+
"loss": 0.1542,
|
| 360 |
+
"mean_token_accuracy": 0.9486050844192505,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.142857142857143,
|
| 365 |
+
"grad_norm": 0.19593572116438304,
|
| 366 |
+
"learning_rate": 2.8621532427365687e-05,
|
| 367 |
+
"loss": 0.1553,
|
| 368 |
+
"mean_token_accuracy": 0.9480521976947784,
|
| 369 |
+
"step": 225
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.1904761904761907,
|
| 373 |
+
"grad_norm": 0.19800587405132705,
|
| 374 |
+
"learning_rate": 2.7686997409912192e-05,
|
| 375 |
+
"loss": 0.1566,
|
| 376 |
+
"mean_token_accuracy": 0.9476682603359222,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.238095238095238,
|
| 381 |
+
"grad_norm": 0.19599017566764934,
|
| 382 |
+
"learning_rate": 2.6752139521303403e-05,
|
| 383 |
+
"loss": 0.1525,
|
| 384 |
+
"mean_token_accuracy": 0.9489890992641449,
|
| 385 |
+
"step": 235
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 2.2857142857142856,
|
| 389 |
+
"grad_norm": 0.1957327582988075,
|
| 390 |
+
"learning_rate": 2.5818572894305453e-05,
|
| 391 |
+
"loss": 0.1515,
|
| 392 |
+
"mean_token_accuracy": 0.9491177141666413,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.3333333333333335,
|
| 397 |
+
"grad_norm": 0.19017281050653162,
|
| 398 |
+
"learning_rate": 2.4887909432182316e-05,
|
| 399 |
+
"loss": 0.1512,
|
| 400 |
+
"mean_token_accuracy": 0.9494254469871521,
|
| 401 |
+
"step": 245
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 2.380952380952381,
|
| 405 |
+
"grad_norm": 0.19356138909333898,
|
| 406 |
+
"learning_rate": 2.3961756025571336e-05,
|
| 407 |
+
"loss": 0.1535,
|
| 408 |
+
"mean_token_accuracy": 0.9488085567951202,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 2.4285714285714284,
|
| 413 |
+
"grad_norm": 0.21519917588614698,
|
| 414 |
+
"learning_rate": 2.304171177801356e-05,
|
| 415 |
+
"loss": 0.1521,
|
| 416 |
+
"mean_token_accuracy": 0.9488667666912078,
|
| 417 |
+
"step": 255
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.4761904761904763,
|
| 421 |
+
"grad_norm": 0.18356462457323924,
|
| 422 |
+
"learning_rate": 2.2129365244929402e-05,
|
| 423 |
+
"loss": 0.1503,
|
| 424 |
+
"mean_token_accuracy": 0.9497200846672058,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 2.5238095238095237,
|
| 429 |
+
"grad_norm": 0.19727737940368512,
|
| 430 |
+
"learning_rate": 2.1226291690806715e-05,
|
| 431 |
+
"loss": 0.1502,
|
| 432 |
+
"mean_token_accuracy": 0.9498027801513672,
|
| 433 |
+
"step": 265
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 2.571428571428571,
|
| 437 |
+
"grad_norm": 0.18491161543383636,
|
| 438 |
+
"learning_rate": 2.0334050369337104e-05,
|
| 439 |
+
"loss": 0.1493,
|
| 440 |
+
"mean_token_accuracy": 0.9500907003879547,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.619047619047619,
|
| 445 |
+
"grad_norm": 0.180689936763116,
|
| 446 |
+
"learning_rate": 1.945418183119656e-05,
|
| 447 |
+
"loss": 0.1461,
|
| 448 |
+
"mean_token_accuracy": 0.9510073781013488,
|
| 449 |
+
"step": 275
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 2.6666666666666665,
|
| 453 |
+
"grad_norm": 0.18357579016167094,
|
| 454 |
+
"learning_rate": 1.8588205264118974e-05,
|
| 455 |
+
"loss": 0.1522,
|
| 456 |
+
"mean_token_accuracy": 0.9489753544330597,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 2.7142857142857144,
|
| 461 |
+
"grad_norm": 0.17455870908089519,
|
| 462 |
+
"learning_rate": 1.7737615869854944e-05,
|
| 463 |
+
"loss": 0.1476,
|
| 464 |
+
"mean_token_accuracy": 0.9504713833332061,
|
| 465 |
+
"step": 285
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.761904761904762,
|
| 469 |
+
"grad_norm": 0.1895734172134673,
|
| 470 |
+
"learning_rate": 1.6903882282545055e-05,
|
| 471 |
+
"loss": 0.1471,
|
| 472 |
+
"mean_token_accuracy": 0.9508952736854553,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.8095238095238093,
|
| 477 |
+
"grad_norm": 0.1814138039625087,
|
| 478 |
+
"learning_rate": 1.6088444032964923e-05,
|
| 479 |
+
"loss": 0.1471,
|
| 480 |
+
"mean_token_accuracy": 0.9506540060043335,
|
| 481 |
+
"step": 295
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 2.857142857142857,
|
| 485 |
+
"grad_norm": 0.18505343239566618,
|
| 486 |
+
"learning_rate": 1.5292709063020415e-05,
|
| 487 |
+
"loss": 0.1453,
|
| 488 |
+
"mean_token_accuracy": 0.9513262331485748,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.9047619047619047,
|
| 493 |
+
"grad_norm": 0.1818729025239862,
|
| 494 |
+
"learning_rate": 1.4518051294784384e-05,
|
| 495 |
+
"loss": 0.1427,
|
| 496 |
+
"mean_token_accuracy": 0.9522208392620086,
|
| 497 |
+
"step": 305
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 2.9523809523809526,
|
| 501 |
+
"grad_norm": 0.18282477277115136,
|
| 502 |
+
"learning_rate": 1.3765808258272334e-05,
|
| 503 |
+
"loss": 0.1457,
|
| 504 |
+
"mean_token_accuracy": 0.9513007164001465,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 3.0,
|
| 509 |
+
"grad_norm": 0.17250893276801252,
|
| 510 |
+
"learning_rate": 1.3037278782052863e-05,
|
| 511 |
+
"loss": 0.1448,
|
| 512 |
+
"mean_token_accuracy": 0.9508862257003784,
|
| 513 |
+
"step": 315
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 3.0476190476190474,
|
| 517 |
+
"grad_norm": 0.21843020721408246,
|
| 518 |
+
"learning_rate": 1.2333720750680403e-05,
|
| 519 |
+
"loss": 0.0972,
|
| 520 |
+
"mean_token_accuracy": 0.9678010582923889,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 3.0952380952380953,
|
| 525 |
+
"grad_norm": 0.18713956774840773,
|
| 526 |
+
"learning_rate": 1.1656348932822086e-05,
|
| 527 |
+
"loss": 0.0951,
|
| 528 |
+
"mean_token_accuracy": 0.9680294811725616,
|
| 529 |
+
"step": 325
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 3.142857142857143,
|
| 533 |
+
"grad_norm": 0.17099224072656777,
|
| 534 |
+
"learning_rate": 1.1006332883828913e-05,
|
| 535 |
+
"loss": 0.0928,
|
| 536 |
+
"mean_token_accuracy": 0.9688285231590271,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.1904761904761907,
|
| 541 |
+
"grad_norm": 0.1889152084893897,
|
| 542 |
+
"learning_rate": 1.0384794926372563e-05,
|
| 543 |
+
"loss": 0.0945,
|
| 544 |
+
"mean_token_accuracy": 0.9682107090950012,
|
| 545 |
+
"step": 335
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 3.238095238095238,
|
| 549 |
+
"grad_norm": 0.16909157911032757,
|
| 550 |
+
"learning_rate": 9.792808212634502e-06,
|
| 551 |
+
"loss": 0.0911,
|
| 552 |
+
"mean_token_accuracy": 0.9692863464355469,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 3.2857142857142856,
|
| 557 |
+
"grad_norm": 0.16462195486055214,
|
| 558 |
+
"learning_rate": 9.231394871393228e-06,
|
| 559 |
+
"loss": 0.0926,
|
| 560 |
+
"mean_token_accuracy": 0.9688502609729767,
|
| 561 |
+
"step": 345
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.3333333333333335,
|
| 565 |
+
"grad_norm": 0.16694354195202218,
|
| 566 |
+
"learning_rate": 8.701524243208935e-06,
|
| 567 |
+
"loss": 0.0941,
|
| 568 |
+
"mean_token_accuracy": 0.968300586938858,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 3.380952380952381,
|
| 573 |
+
"grad_norm": 0.1692222626696623,
|
| 574 |
+
"learning_rate": 8.204111206752663e-06,
|
| 575 |
+
"loss": 0.0918,
|
| 576 |
+
"mean_token_accuracy": 0.9689784705638885,
|
| 577 |
+
"step": 355
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 3.4285714285714284,
|
| 581 |
+
"grad_norm": 0.17221192973282243,
|
| 582 |
+
"learning_rate": 7.740014599169857e-06,
|
| 583 |
+
"loss": 0.0904,
|
| 584 |
+
"mean_token_accuracy": 0.9694355607032776,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 3.4761904761904763,
|
| 589 |
+
"grad_norm": 0.1741810830817914,
|
| 590 |
+
"learning_rate": 7.3100357332055624e-06,
|
| 591 |
+
"loss": 0.0896,
|
| 592 |
+
"mean_token_accuracy": 0.9696675717830658,
|
| 593 |
+
"step": 365
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 3.5238095238095237,
|
| 597 |
+
"grad_norm": 0.17501221405018258,
|
| 598 |
+
"learning_rate": 6.914917013651723e-06,
|
| 599 |
+
"loss": 0.092,
|
| 600 |
+
"mean_token_accuracy": 0.968971711397171,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 3.571428571428571,
|
| 605 |
+
"grad_norm": 0.16570291902290668,
|
| 606 |
+
"learning_rate": 6.555340655505407e-06,
|
| 607 |
+
"loss": 0.0917,
|
| 608 |
+
"mean_token_accuracy": 0.9693442165851593,
|
| 609 |
+
"step": 375
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 3.619047619047619,
|
| 613 |
+
"grad_norm": 0.15923829992676775,
|
| 614 |
+
"learning_rate": 6.231927506051192e-06,
|
| 615 |
+
"loss": 0.0915,
|
| 616 |
+
"mean_token_accuracy": 0.969234949350357,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 3.6666666666666665,
|
| 621 |
+
"grad_norm": 0.17338149522120277,
|
| 622 |
+
"learning_rate": 5.9452359729015004e-06,
|
| 623 |
+
"loss": 0.0907,
|
| 624 |
+
"mean_token_accuracy": 0.9692964613437652,
|
| 625 |
+
"step": 385
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 3.7142857142857144,
|
| 629 |
+
"grad_norm": 0.17158993366626976,
|
| 630 |
+
"learning_rate": 5.695761059845749e-06,
|
| 631 |
+
"loss": 0.0902,
|
| 632 |
+
"mean_token_accuracy": 0.9696565389633178,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 3.761904761904762,
|
| 637 |
+
"grad_norm": 0.17688545350709856,
|
| 638 |
+
"learning_rate": 5.483933512173022e-06,
|
| 639 |
+
"loss": 0.0918,
|
| 640 |
+
"mean_token_accuracy": 0.9692005813121796,
|
| 641 |
+
"step": 395
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 3.8095238095238093,
|
| 645 |
+
"grad_norm": 0.1621556558547238,
|
| 646 |
+
"learning_rate": 5.310119072943991e-06,
|
| 647 |
+
"loss": 0.0892,
|
| 648 |
+
"mean_token_accuracy": 0.9700611054897308,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 3.857142857142857,
|
| 653 |
+
"grad_norm": 0.16839822104636895,
|
| 654 |
+
"learning_rate": 5.174617851496128e-06,
|
| 655 |
+
"loss": 0.0904,
|
| 656 |
+
"mean_token_accuracy": 0.9694457828998566,
|
| 657 |
+
"step": 405
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 3.9047619047619047,
|
| 661 |
+
"grad_norm": 0.16305557552247782,
|
| 662 |
+
"learning_rate": 5.077663805272652e-06,
|
| 663 |
+
"loss": 0.0902,
|
| 664 |
+
"mean_token_accuracy": 0.9695923388004303,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 3.9523809523809526,
|
| 669 |
+
"grad_norm": 0.16067720695857812,
|
| 670 |
+
"learning_rate": 5.019424335869808e-06,
|
| 671 |
+
"loss": 0.0903,
|
| 672 |
+
"mean_token_accuracy": 0.9695243299007416,
|
| 673 |
+
"step": 415
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 4.0,
|
| 677 |
+
"grad_norm": 0.1603132269708069,
|
| 678 |
+
"learning_rate": 5e-06,
|
| 679 |
+
"loss": 0.0903,
|
| 680 |
+
"mean_token_accuracy": 0.9692174196243286,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 4.0,
|
| 685 |
+
"step": 420,
|
| 686 |
+
"total_flos": 216761967837184.0,
|
| 687 |
+
"train_loss": 0.20915592369579133,
|
| 688 |
+
"train_runtime": 2440.0132,
|
| 689 |
+
"train_samples_per_second": 2.749,
|
| 690 |
+
"train_steps_per_second": 0.172
|
| 691 |
+
}
|
| 692 |
+
],
|
| 693 |
+
"logging_steps": 5,
|
| 694 |
+
"max_steps": 420,
|
| 695 |
+
"num_input_tokens_seen": 0,
|
| 696 |
+
"num_train_epochs": 4,
|
| 697 |
+
"save_steps": 500,
|
| 698 |
+
"stateful_callbacks": {
|
| 699 |
+
"TrainerControl": {
|
| 700 |
+
"args": {
|
| 701 |
+
"should_epoch_stop": false,
|
| 702 |
+
"should_evaluate": false,
|
| 703 |
+
"should_log": false,
|
| 704 |
+
"should_save": true,
|
| 705 |
+
"should_training_stop": true
|
| 706 |
+
},
|
| 707 |
+
"attributes": {}
|
| 708 |
+
}
|
| 709 |
+
},
|
| 710 |
+
"total_flos": 216761967837184.0,
|
| 711 |
+
"train_batch_size": 1,
|
| 712 |
+
"trial_name": null,
|
| 713 |
+
"trial_params": null
|
| 714 |
+
}
|