Model save
Browse files- README.md +58 -0
- all_results.json +8 -0
- generation_config.json +9 -0
- train_results.json +8 -0
- trainer_state.json +1540 -0
README.md
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: open-r1/OpenR1-Qwen-7B
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: OpenR1-Qwen-7B-Sparse
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for OpenR1-Qwen-7B-Sparse
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [open-r1/OpenR1-Qwen-7B](https://huggingface.co/open-r1/OpenR1-Qwen-7B).
|
| 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="ZMC2019/OpenR1-Qwen-7B-Sparse", 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/chenzhuoming911/huggingface/runs/mharh69t)
|
| 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: 3.5.0
|
| 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": 9.186429923093381e+17,
|
| 3 |
+
"train_loss": 0.27417567097351914,
|
| 4 |
+
"train_runtime": 29088.9913,
|
| 5 |
+
"train_samples": 93733,
|
| 6 |
+
"train_samples_per_second": 0.59,
|
| 7 |
+
"train_steps_per_second": 0.037
|
| 8 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"transformers_version": "4.49.0"
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 9.186429923093381e+17,
|
| 3 |
+
"train_loss": 0.27417567097351914,
|
| 4 |
+
"train_runtime": 29088.9913,
|
| 5 |
+
"train_samples": 93733,
|
| 6 |
+
"train_samples_per_second": 0.59,
|
| 7 |
+
"train_steps_per_second": 0.037
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1540 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1073,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004659832246039142,
|
| 13 |
+
"grad_norm": 0.2480848187526065,
|
| 14 |
+
"learning_rate": 9.259259259259259e-07,
|
| 15 |
+
"loss": 0.2856,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.009319664492078284,
|
| 20 |
+
"grad_norm": 0.27846440569740516,
|
| 21 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 22 |
+
"loss": 0.2743,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.013979496738117428,
|
| 27 |
+
"grad_norm": 0.1775956763008417,
|
| 28 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 29 |
+
"loss": 0.2735,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01863932898415657,
|
| 34 |
+
"grad_norm": 0.2097019213694641,
|
| 35 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 36 |
+
"loss": 0.2783,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.023299161230195712,
|
| 41 |
+
"grad_norm": 0.2017973119570326,
|
| 42 |
+
"learning_rate": 4.62962962962963e-06,
|
| 43 |
+
"loss": 0.2623,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.027958993476234855,
|
| 48 |
+
"grad_norm": 0.19396546589386038,
|
| 49 |
+
"learning_rate": 5.555555555555557e-06,
|
| 50 |
+
"loss": 0.2588,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.032618825722273995,
|
| 55 |
+
"grad_norm": 0.1833475348357745,
|
| 56 |
+
"learning_rate": 6.481481481481482e-06,
|
| 57 |
+
"loss": 0.2629,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03727865796831314,
|
| 62 |
+
"grad_norm": 0.19103036464826775,
|
| 63 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 64 |
+
"loss": 0.2639,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.04193849021435228,
|
| 69 |
+
"grad_norm": 0.1903046005080379,
|
| 70 |
+
"learning_rate": 8.333333333333334e-06,
|
| 71 |
+
"loss": 0.2673,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.046598322460391424,
|
| 76 |
+
"grad_norm": 0.18460973869272648,
|
| 77 |
+
"learning_rate": 9.25925925925926e-06,
|
| 78 |
+
"loss": 0.2761,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.05125815470643057,
|
| 83 |
+
"grad_norm": 0.19245235339515152,
|
| 84 |
+
"learning_rate": 1.0185185185185186e-05,
|
| 85 |
+
"loss": 0.2712,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05591798695246971,
|
| 90 |
+
"grad_norm": 0.187838588087584,
|
| 91 |
+
"learning_rate": 1.1111111111111113e-05,
|
| 92 |
+
"loss": 0.2723,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.06057781919850885,
|
| 97 |
+
"grad_norm": 0.17781539091622867,
|
| 98 |
+
"learning_rate": 1.2037037037037039e-05,
|
| 99 |
+
"loss": 0.2667,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.06523765144454799,
|
| 104 |
+
"grad_norm": 0.20027656085731357,
|
| 105 |
+
"learning_rate": 1.2962962962962964e-05,
|
| 106 |
+
"loss": 0.2736,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.06989748369058714,
|
| 111 |
+
"grad_norm": 0.20349467354700868,
|
| 112 |
+
"learning_rate": 1.388888888888889e-05,
|
| 113 |
+
"loss": 0.2696,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.07455731593662628,
|
| 118 |
+
"grad_norm": 0.20858032960565945,
|
| 119 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 120 |
+
"loss": 0.2659,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.07921714818266543,
|
| 125 |
+
"grad_norm": 0.21178392393672066,
|
| 126 |
+
"learning_rate": 1.5740740740740744e-05,
|
| 127 |
+
"loss": 0.2715,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.08387698042870456,
|
| 132 |
+
"grad_norm": 0.19683247262126224,
|
| 133 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 134 |
+
"loss": 0.276,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.08853681267474371,
|
| 139 |
+
"grad_norm": 0.19763192488794942,
|
| 140 |
+
"learning_rate": 1.7592592592592595e-05,
|
| 141 |
+
"loss": 0.2704,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.09319664492078285,
|
| 146 |
+
"grad_norm": 0.1956972064059814,
|
| 147 |
+
"learning_rate": 1.851851851851852e-05,
|
| 148 |
+
"loss": 0.2682,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.097856477166822,
|
| 153 |
+
"grad_norm": 0.20066192797959884,
|
| 154 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 155 |
+
"loss": 0.2694,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.10251630941286113,
|
| 160 |
+
"grad_norm": 0.22173843606276825,
|
| 161 |
+
"learning_rate": 1.995854922279793e-05,
|
| 162 |
+
"loss": 0.2691,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.10717614165890028,
|
| 167 |
+
"grad_norm": 0.22661606789228167,
|
| 168 |
+
"learning_rate": 1.985492227979275e-05,
|
| 169 |
+
"loss": 0.2707,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.11183597390493942,
|
| 174 |
+
"grad_norm": 0.25878304515306677,
|
| 175 |
+
"learning_rate": 1.9751295336787565e-05,
|
| 176 |
+
"loss": 0.2712,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.11649580615097857,
|
| 181 |
+
"grad_norm": 0.20769568193046992,
|
| 182 |
+
"learning_rate": 1.9647668393782386e-05,
|
| 183 |
+
"loss": 0.2662,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.1211556383970177,
|
| 188 |
+
"grad_norm": 0.2339523798236711,
|
| 189 |
+
"learning_rate": 1.9544041450777206e-05,
|
| 190 |
+
"loss": 0.2741,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.12581547064305684,
|
| 195 |
+
"grad_norm": 0.21510297906435297,
|
| 196 |
+
"learning_rate": 1.9440414507772023e-05,
|
| 197 |
+
"loss": 0.2684,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.13047530288909598,
|
| 202 |
+
"grad_norm": 0.23732427940490297,
|
| 203 |
+
"learning_rate": 1.9336787564766843e-05,
|
| 204 |
+
"loss": 0.2685,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.13513513513513514,
|
| 209 |
+
"grad_norm": 0.2123608724012592,
|
| 210 |
+
"learning_rate": 1.923316062176166e-05,
|
| 211 |
+
"loss": 0.2808,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.13979496738117428,
|
| 216 |
+
"grad_norm": 0.22764321452420136,
|
| 217 |
+
"learning_rate": 1.9129533678756477e-05,
|
| 218 |
+
"loss": 0.2726,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.14445479962721341,
|
| 223 |
+
"grad_norm": 0.20310601163154524,
|
| 224 |
+
"learning_rate": 1.9025906735751297e-05,
|
| 225 |
+
"loss": 0.2852,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.14911463187325255,
|
| 230 |
+
"grad_norm": 0.2298242877800892,
|
| 231 |
+
"learning_rate": 1.8922279792746117e-05,
|
| 232 |
+
"loss": 0.2757,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.15377446411929171,
|
| 237 |
+
"grad_norm": 0.20978701600081326,
|
| 238 |
+
"learning_rate": 1.8818652849740934e-05,
|
| 239 |
+
"loss": 0.2733,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.15843429636533085,
|
| 244 |
+
"grad_norm": 0.1998319044566166,
|
| 245 |
+
"learning_rate": 1.8715025906735754e-05,
|
| 246 |
+
"loss": 0.2559,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.16309412861137,
|
| 251 |
+
"grad_norm": 0.20450079549144815,
|
| 252 |
+
"learning_rate": 1.861139896373057e-05,
|
| 253 |
+
"loss": 0.2743,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.16775396085740912,
|
| 258 |
+
"grad_norm": 0.19585540202223178,
|
| 259 |
+
"learning_rate": 1.850777202072539e-05,
|
| 260 |
+
"loss": 0.2731,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1724137931034483,
|
| 265 |
+
"grad_norm": 0.20114655528014272,
|
| 266 |
+
"learning_rate": 1.8404145077720208e-05,
|
| 267 |
+
"loss": 0.2676,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.17707362534948742,
|
| 272 |
+
"grad_norm": 0.20898866856201928,
|
| 273 |
+
"learning_rate": 1.8300518134715028e-05,
|
| 274 |
+
"loss": 0.2746,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.18173345759552656,
|
| 279 |
+
"grad_norm": 0.23418307676215824,
|
| 280 |
+
"learning_rate": 1.8196891191709845e-05,
|
| 281 |
+
"loss": 0.2738,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1863932898415657,
|
| 286 |
+
"grad_norm": 0.21381181093698814,
|
| 287 |
+
"learning_rate": 1.8093264248704665e-05,
|
| 288 |
+
"loss": 0.2755,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.19105312208760486,
|
| 293 |
+
"grad_norm": 0.1945831655656839,
|
| 294 |
+
"learning_rate": 1.7989637305699482e-05,
|
| 295 |
+
"loss": 0.2729,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.195712954333644,
|
| 300 |
+
"grad_norm": 0.25188414879097026,
|
| 301 |
+
"learning_rate": 1.7886010362694302e-05,
|
| 302 |
+
"loss": 0.2747,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.20037278657968313,
|
| 307 |
+
"grad_norm": 0.2034438567116008,
|
| 308 |
+
"learning_rate": 1.778238341968912e-05,
|
| 309 |
+
"loss": 0.2707,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.20503261882572227,
|
| 314 |
+
"grad_norm": 0.19342833535436457,
|
| 315 |
+
"learning_rate": 1.767875647668394e-05,
|
| 316 |
+
"loss": 0.2761,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.2096924510717614,
|
| 321 |
+
"grad_norm": 0.2043835725939126,
|
| 322 |
+
"learning_rate": 1.757512953367876e-05,
|
| 323 |
+
"loss": 0.2681,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.21435228331780057,
|
| 328 |
+
"grad_norm": 0.2074491682435995,
|
| 329 |
+
"learning_rate": 1.7471502590673576e-05,
|
| 330 |
+
"loss": 0.2743,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.2190121155638397,
|
| 335 |
+
"grad_norm": 0.19431836721424317,
|
| 336 |
+
"learning_rate": 1.7367875647668397e-05,
|
| 337 |
+
"loss": 0.2696,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.22367194780987884,
|
| 342 |
+
"grad_norm": 0.23535008907014365,
|
| 343 |
+
"learning_rate": 1.7264248704663214e-05,
|
| 344 |
+
"loss": 0.2756,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.22833178005591798,
|
| 349 |
+
"grad_norm": 0.22992274931044096,
|
| 350 |
+
"learning_rate": 1.716062176165803e-05,
|
| 351 |
+
"loss": 0.2747,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.23299161230195714,
|
| 356 |
+
"grad_norm": 0.23642332032502936,
|
| 357 |
+
"learning_rate": 1.705699481865285e-05,
|
| 358 |
+
"loss": 0.2726,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.23765144454799628,
|
| 363 |
+
"grad_norm": 0.20748084991662222,
|
| 364 |
+
"learning_rate": 1.695336787564767e-05,
|
| 365 |
+
"loss": 0.2716,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2423112767940354,
|
| 370 |
+
"grad_norm": 0.21945114618793227,
|
| 371 |
+
"learning_rate": 1.6849740932642488e-05,
|
| 372 |
+
"loss": 0.2771,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.24697110904007455,
|
| 377 |
+
"grad_norm": 0.22645209579957054,
|
| 378 |
+
"learning_rate": 1.6746113989637308e-05,
|
| 379 |
+
"loss": 0.2765,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2516309412861137,
|
| 384 |
+
"grad_norm": 0.19551387120009264,
|
| 385 |
+
"learning_rate": 1.6642487046632125e-05,
|
| 386 |
+
"loss": 0.2723,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.25629077353215285,
|
| 391 |
+
"grad_norm": 0.20635102879128378,
|
| 392 |
+
"learning_rate": 1.6538860103626945e-05,
|
| 393 |
+
"loss": 0.2725,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.26095060577819196,
|
| 398 |
+
"grad_norm": 0.19198952075337267,
|
| 399 |
+
"learning_rate": 1.6435233160621765e-05,
|
| 400 |
+
"loss": 0.2709,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2656104380242311,
|
| 405 |
+
"grad_norm": 0.2183101530827561,
|
| 406 |
+
"learning_rate": 1.6331606217616582e-05,
|
| 407 |
+
"loss": 0.2718,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2702702702702703,
|
| 412 |
+
"grad_norm": 0.20425797133671955,
|
| 413 |
+
"learning_rate": 1.6227979274611402e-05,
|
| 414 |
+
"loss": 0.2788,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2749301025163094,
|
| 419 |
+
"grad_norm": 0.23011000314623464,
|
| 420 |
+
"learning_rate": 1.612435233160622e-05,
|
| 421 |
+
"loss": 0.2905,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.27958993476234856,
|
| 426 |
+
"grad_norm": 0.2155480458012789,
|
| 427 |
+
"learning_rate": 1.6020725388601036e-05,
|
| 428 |
+
"loss": 0.2848,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.2842497670083877,
|
| 433 |
+
"grad_norm": 0.2197010179321984,
|
| 434 |
+
"learning_rate": 1.5917098445595856e-05,
|
| 435 |
+
"loss": 0.2747,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.28890959925442683,
|
| 440 |
+
"grad_norm": 0.22179807332039045,
|
| 441 |
+
"learning_rate": 1.5813471502590673e-05,
|
| 442 |
+
"loss": 0.2728,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.293569431500466,
|
| 447 |
+
"grad_norm": 0.2120751835986257,
|
| 448 |
+
"learning_rate": 1.5709844559585493e-05,
|
| 449 |
+
"loss": 0.2787,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.2982292637465051,
|
| 454 |
+
"grad_norm": 0.226069231034194,
|
| 455 |
+
"learning_rate": 1.5606217616580313e-05,
|
| 456 |
+
"loss": 0.2651,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.30288909599254427,
|
| 461 |
+
"grad_norm": 0.20857741942191318,
|
| 462 |
+
"learning_rate": 1.550259067357513e-05,
|
| 463 |
+
"loss": 0.2672,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.30754892823858343,
|
| 468 |
+
"grad_norm": 0.2027127864635231,
|
| 469 |
+
"learning_rate": 1.539896373056995e-05,
|
| 470 |
+
"loss": 0.2758,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.31220876048462254,
|
| 475 |
+
"grad_norm": 0.21391079816437977,
|
| 476 |
+
"learning_rate": 1.5295336787564767e-05,
|
| 477 |
+
"loss": 0.2737,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3168685927306617,
|
| 482 |
+
"grad_norm": 0.1949931649207275,
|
| 483 |
+
"learning_rate": 1.5191709844559586e-05,
|
| 484 |
+
"loss": 0.2709,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.32152842497670087,
|
| 489 |
+
"grad_norm": 0.203272659243948,
|
| 490 |
+
"learning_rate": 1.5088082901554406e-05,
|
| 491 |
+
"loss": 0.2744,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.32618825722274,
|
| 496 |
+
"grad_norm": 0.23386254140039578,
|
| 497 |
+
"learning_rate": 1.4984455958549225e-05,
|
| 498 |
+
"loss": 0.2829,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.33084808946877914,
|
| 503 |
+
"grad_norm": 0.20956086579279376,
|
| 504 |
+
"learning_rate": 1.4880829015544043e-05,
|
| 505 |
+
"loss": 0.282,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.33550792171481825,
|
| 510 |
+
"grad_norm": 0.19915707487576464,
|
| 511 |
+
"learning_rate": 1.4777202072538862e-05,
|
| 512 |
+
"loss": 0.2728,
|
| 513 |
+
"step": 360
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3401677539608574,
|
| 517 |
+
"grad_norm": 0.19888362106152468,
|
| 518 |
+
"learning_rate": 1.4673575129533678e-05,
|
| 519 |
+
"loss": 0.2711,
|
| 520 |
+
"step": 365
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3448275862068966,
|
| 524 |
+
"grad_norm": 0.20565614821988315,
|
| 525 |
+
"learning_rate": 1.4569948186528497e-05,
|
| 526 |
+
"loss": 0.2771,
|
| 527 |
+
"step": 370
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3494874184529357,
|
| 531 |
+
"grad_norm": 0.22007541774769399,
|
| 532 |
+
"learning_rate": 1.4466321243523317e-05,
|
| 533 |
+
"loss": 0.2727,
|
| 534 |
+
"step": 375
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.35414725069897485,
|
| 538 |
+
"grad_norm": 0.21285035462598179,
|
| 539 |
+
"learning_rate": 1.4362694300518136e-05,
|
| 540 |
+
"loss": 0.2813,
|
| 541 |
+
"step": 380
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.35880708294501396,
|
| 545 |
+
"grad_norm": 0.2074859169658842,
|
| 546 |
+
"learning_rate": 1.4259067357512954e-05,
|
| 547 |
+
"loss": 0.2755,
|
| 548 |
+
"step": 385
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3634669151910531,
|
| 552 |
+
"grad_norm": 0.20518630624764556,
|
| 553 |
+
"learning_rate": 1.4155440414507773e-05,
|
| 554 |
+
"loss": 0.2751,
|
| 555 |
+
"step": 390
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.3681267474370923,
|
| 559 |
+
"grad_norm": 0.24631525396138984,
|
| 560 |
+
"learning_rate": 1.4051813471502591e-05,
|
| 561 |
+
"loss": 0.2758,
|
| 562 |
+
"step": 395
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3727865796831314,
|
| 566 |
+
"grad_norm": 0.2072164144473051,
|
| 567 |
+
"learning_rate": 1.394818652849741e-05,
|
| 568 |
+
"loss": 0.2679,
|
| 569 |
+
"step": 400
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.37744641192917056,
|
| 573 |
+
"grad_norm": 0.20130515204148264,
|
| 574 |
+
"learning_rate": 1.384455958549223e-05,
|
| 575 |
+
"loss": 0.288,
|
| 576 |
+
"step": 405
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.3821062441752097,
|
| 580 |
+
"grad_norm": 0.21318795834604967,
|
| 581 |
+
"learning_rate": 1.3740932642487049e-05,
|
| 582 |
+
"loss": 0.2774,
|
| 583 |
+
"step": 410
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.38676607642124883,
|
| 587 |
+
"grad_norm": 0.19837831244344498,
|
| 588 |
+
"learning_rate": 1.3637305699481867e-05,
|
| 589 |
+
"loss": 0.2732,
|
| 590 |
+
"step": 415
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.391425908667288,
|
| 594 |
+
"grad_norm": 0.21706421514687568,
|
| 595 |
+
"learning_rate": 1.3533678756476684e-05,
|
| 596 |
+
"loss": 0.2805,
|
| 597 |
+
"step": 420
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3960857409133271,
|
| 601 |
+
"grad_norm": 0.21316576729012615,
|
| 602 |
+
"learning_rate": 1.3430051813471503e-05,
|
| 603 |
+
"loss": 0.2844,
|
| 604 |
+
"step": 425
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.40074557315936626,
|
| 608 |
+
"grad_norm": 0.2014337720595473,
|
| 609 |
+
"learning_rate": 1.3326424870466321e-05,
|
| 610 |
+
"loss": 0.2828,
|
| 611 |
+
"step": 430
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.40540540540540543,
|
| 615 |
+
"grad_norm": 0.2072514307470887,
|
| 616 |
+
"learning_rate": 1.3222797927461141e-05,
|
| 617 |
+
"loss": 0.2852,
|
| 618 |
+
"step": 435
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.41006523765144454,
|
| 622 |
+
"grad_norm": 0.24227887513487076,
|
| 623 |
+
"learning_rate": 1.311917098445596e-05,
|
| 624 |
+
"loss": 0.2755,
|
| 625 |
+
"step": 440
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.4147250698974837,
|
| 629 |
+
"grad_norm": 0.20801065806848365,
|
| 630 |
+
"learning_rate": 1.3015544041450778e-05,
|
| 631 |
+
"loss": 0.2764,
|
| 632 |
+
"step": 445
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.4193849021435228,
|
| 636 |
+
"grad_norm": 0.1942439454612911,
|
| 637 |
+
"learning_rate": 1.2911917098445597e-05,
|
| 638 |
+
"loss": 0.2691,
|
| 639 |
+
"step": 450
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.424044734389562,
|
| 643 |
+
"grad_norm": 0.2076864298032534,
|
| 644 |
+
"learning_rate": 1.2808290155440415e-05,
|
| 645 |
+
"loss": 0.2771,
|
| 646 |
+
"step": 455
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.42870456663560114,
|
| 650 |
+
"grad_norm": 0.20740530807254287,
|
| 651 |
+
"learning_rate": 1.2704663212435234e-05,
|
| 652 |
+
"loss": 0.2749,
|
| 653 |
+
"step": 460
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.43336439888164024,
|
| 657 |
+
"grad_norm": 0.2226181167276175,
|
| 658 |
+
"learning_rate": 1.2601036269430054e-05,
|
| 659 |
+
"loss": 0.2726,
|
| 660 |
+
"step": 465
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.4380242311276794,
|
| 664 |
+
"grad_norm": 0.2163332314393917,
|
| 665 |
+
"learning_rate": 1.2497409326424873e-05,
|
| 666 |
+
"loss": 0.2801,
|
| 667 |
+
"step": 470
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.4426840633737186,
|
| 671 |
+
"grad_norm": 0.22697445952775339,
|
| 672 |
+
"learning_rate": 1.239378238341969e-05,
|
| 673 |
+
"loss": 0.2721,
|
| 674 |
+
"step": 475
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.4473438956197577,
|
| 678 |
+
"grad_norm": 0.21459489713651447,
|
| 679 |
+
"learning_rate": 1.2290155440414508e-05,
|
| 680 |
+
"loss": 0.2876,
|
| 681 |
+
"step": 480
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.45200372786579684,
|
| 685 |
+
"grad_norm": 0.212728336485677,
|
| 686 |
+
"learning_rate": 1.2186528497409327e-05,
|
| 687 |
+
"loss": 0.2775,
|
| 688 |
+
"step": 485
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.45666356011183595,
|
| 692 |
+
"grad_norm": 0.18956662975107208,
|
| 693 |
+
"learning_rate": 1.2082901554404145e-05,
|
| 694 |
+
"loss": 0.2717,
|
| 695 |
+
"step": 490
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.4613233923578751,
|
| 699 |
+
"grad_norm": 0.19911793563914765,
|
| 700 |
+
"learning_rate": 1.1979274611398965e-05,
|
| 701 |
+
"loss": 0.2761,
|
| 702 |
+
"step": 495
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4659832246039143,
|
| 706 |
+
"grad_norm": 0.21638725379085677,
|
| 707 |
+
"learning_rate": 1.1875647668393784e-05,
|
| 708 |
+
"loss": 0.2876,
|
| 709 |
+
"step": 500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.4706430568499534,
|
| 713 |
+
"grad_norm": 0.21417364115903229,
|
| 714 |
+
"learning_rate": 1.1772020725388602e-05,
|
| 715 |
+
"loss": 0.2768,
|
| 716 |
+
"step": 505
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.47530288909599255,
|
| 720 |
+
"grad_norm": 0.19378036957825795,
|
| 721 |
+
"learning_rate": 1.1668393782383421e-05,
|
| 722 |
+
"loss": 0.2757,
|
| 723 |
+
"step": 510
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.47996272134203166,
|
| 727 |
+
"grad_norm": 0.20348312570963792,
|
| 728 |
+
"learning_rate": 1.1564766839378238e-05,
|
| 729 |
+
"loss": 0.2754,
|
| 730 |
+
"step": 515
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.4846225535880708,
|
| 734 |
+
"grad_norm": 0.20078805057918098,
|
| 735 |
+
"learning_rate": 1.1461139896373056e-05,
|
| 736 |
+
"loss": 0.2801,
|
| 737 |
+
"step": 520
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.48928238583411,
|
| 741 |
+
"grad_norm": 0.22711582634492097,
|
| 742 |
+
"learning_rate": 1.1357512953367878e-05,
|
| 743 |
+
"loss": 0.2805,
|
| 744 |
+
"step": 525
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.4939422180801491,
|
| 748 |
+
"grad_norm": 0.19833314424388493,
|
| 749 |
+
"learning_rate": 1.1253886010362695e-05,
|
| 750 |
+
"loss": 0.2749,
|
| 751 |
+
"step": 530
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.49860205032618826,
|
| 755 |
+
"grad_norm": 0.20627918163276826,
|
| 756 |
+
"learning_rate": 1.1150259067357514e-05,
|
| 757 |
+
"loss": 0.2807,
|
| 758 |
+
"step": 535
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.5032618825722274,
|
| 762 |
+
"grad_norm": 0.22111449652544438,
|
| 763 |
+
"learning_rate": 1.1046632124352332e-05,
|
| 764 |
+
"loss": 0.2672,
|
| 765 |
+
"step": 540
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5079217148182665,
|
| 769 |
+
"grad_norm": 0.2245796946555213,
|
| 770 |
+
"learning_rate": 1.094300518134715e-05,
|
| 771 |
+
"loss": 0.2745,
|
| 772 |
+
"step": 545
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5125815470643057,
|
| 776 |
+
"grad_norm": 0.21913989463492667,
|
| 777 |
+
"learning_rate": 1.083937823834197e-05,
|
| 778 |
+
"loss": 0.2808,
|
| 779 |
+
"step": 550
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5172413793103449,
|
| 783 |
+
"grad_norm": 0.21743987494816502,
|
| 784 |
+
"learning_rate": 1.073575129533679e-05,
|
| 785 |
+
"loss": 0.2805,
|
| 786 |
+
"step": 555
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5219012115563839,
|
| 790 |
+
"grad_norm": 0.2223044993066891,
|
| 791 |
+
"learning_rate": 1.0632124352331608e-05,
|
| 792 |
+
"loss": 0.2875,
|
| 793 |
+
"step": 560
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.5265610438024231,
|
| 797 |
+
"grad_norm": 0.18695066177700756,
|
| 798 |
+
"learning_rate": 1.0528497409326426e-05,
|
| 799 |
+
"loss": 0.2643,
|
| 800 |
+
"step": 565
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.5312208760484622,
|
| 804 |
+
"grad_norm": 0.21093529744919742,
|
| 805 |
+
"learning_rate": 1.0424870466321243e-05,
|
| 806 |
+
"loss": 0.2699,
|
| 807 |
+
"step": 570
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.5358807082945014,
|
| 811 |
+
"grad_norm": 0.19752807407672002,
|
| 812 |
+
"learning_rate": 1.0321243523316062e-05,
|
| 813 |
+
"loss": 0.2734,
|
| 814 |
+
"step": 575
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.5405405405405406,
|
| 818 |
+
"grad_norm": 0.20088984513783537,
|
| 819 |
+
"learning_rate": 1.021761658031088e-05,
|
| 820 |
+
"loss": 0.2712,
|
| 821 |
+
"step": 580
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.5452003727865797,
|
| 825 |
+
"grad_norm": 0.20325055999916153,
|
| 826 |
+
"learning_rate": 1.01139896373057e-05,
|
| 827 |
+
"loss": 0.2774,
|
| 828 |
+
"step": 585
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.5498602050326188,
|
| 832 |
+
"grad_norm": 0.22174441086027497,
|
| 833 |
+
"learning_rate": 1.0010362694300519e-05,
|
| 834 |
+
"loss": 0.2689,
|
| 835 |
+
"step": 590
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.554520037278658,
|
| 839 |
+
"grad_norm": 0.20061271709209516,
|
| 840 |
+
"learning_rate": 9.906735751295338e-06,
|
| 841 |
+
"loss": 0.273,
|
| 842 |
+
"step": 595
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.5591798695246971,
|
| 846 |
+
"grad_norm": 0.2226596622045658,
|
| 847 |
+
"learning_rate": 9.803108808290156e-06,
|
| 848 |
+
"loss": 0.2756,
|
| 849 |
+
"step": 600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.5638397017707363,
|
| 853 |
+
"grad_norm": 0.19697076210420275,
|
| 854 |
+
"learning_rate": 9.699481865284975e-06,
|
| 855 |
+
"loss": 0.2732,
|
| 856 |
+
"step": 605
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5684995340167754,
|
| 860 |
+
"grad_norm": 0.22175232192708388,
|
| 861 |
+
"learning_rate": 9.595854922279793e-06,
|
| 862 |
+
"loss": 0.2812,
|
| 863 |
+
"step": 610
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5731593662628145,
|
| 867 |
+
"grad_norm": 0.21147846013410135,
|
| 868 |
+
"learning_rate": 9.492227979274612e-06,
|
| 869 |
+
"loss": 0.2729,
|
| 870 |
+
"step": 615
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5778191985088537,
|
| 874 |
+
"grad_norm": 0.20160869332931788,
|
| 875 |
+
"learning_rate": 9.388601036269432e-06,
|
| 876 |
+
"loss": 0.2663,
|
| 877 |
+
"step": 620
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5824790307548928,
|
| 881 |
+
"grad_norm": 0.20814998856523514,
|
| 882 |
+
"learning_rate": 9.284974093264249e-06,
|
| 883 |
+
"loss": 0.2787,
|
| 884 |
+
"step": 625
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.587138863000932,
|
| 888 |
+
"grad_norm": 0.1873990471712869,
|
| 889 |
+
"learning_rate": 9.181347150259067e-06,
|
| 890 |
+
"loss": 0.2784,
|
| 891 |
+
"step": 630
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.5917986952469712,
|
| 895 |
+
"grad_norm": 0.21324250407573406,
|
| 896 |
+
"learning_rate": 9.077720207253888e-06,
|
| 897 |
+
"loss": 0.2791,
|
| 898 |
+
"step": 635
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5964585274930102,
|
| 902 |
+
"grad_norm": 0.19205913704094654,
|
| 903 |
+
"learning_rate": 8.974093264248706e-06,
|
| 904 |
+
"loss": 0.2808,
|
| 905 |
+
"step": 640
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6011183597390494,
|
| 909 |
+
"grad_norm": 0.20657296059347666,
|
| 910 |
+
"learning_rate": 8.870466321243523e-06,
|
| 911 |
+
"loss": 0.2757,
|
| 912 |
+
"step": 645
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.6057781919850885,
|
| 916 |
+
"grad_norm": 0.20355030156034296,
|
| 917 |
+
"learning_rate": 8.766839378238343e-06,
|
| 918 |
+
"loss": 0.2774,
|
| 919 |
+
"step": 650
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6104380242311277,
|
| 923 |
+
"grad_norm": 0.210129052042712,
|
| 924 |
+
"learning_rate": 8.663212435233162e-06,
|
| 925 |
+
"loss": 0.2813,
|
| 926 |
+
"step": 655
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.6150978564771669,
|
| 930 |
+
"grad_norm": 0.20693376106919167,
|
| 931 |
+
"learning_rate": 8.55958549222798e-06,
|
| 932 |
+
"loss": 0.2654,
|
| 933 |
+
"step": 660
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.6197576887232059,
|
| 937 |
+
"grad_norm": 0.20048258208742983,
|
| 938 |
+
"learning_rate": 8.455958549222799e-06,
|
| 939 |
+
"loss": 0.2686,
|
| 940 |
+
"step": 665
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.6244175209692451,
|
| 944 |
+
"grad_norm": 0.19402957660001943,
|
| 945 |
+
"learning_rate": 8.352331606217617e-06,
|
| 946 |
+
"loss": 0.2695,
|
| 947 |
+
"step": 670
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.6290773532152842,
|
| 951 |
+
"grad_norm": 0.18920776684229643,
|
| 952 |
+
"learning_rate": 8.248704663212436e-06,
|
| 953 |
+
"loss": 0.2673,
|
| 954 |
+
"step": 675
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.6337371854613234,
|
| 958 |
+
"grad_norm": 0.21815092086301652,
|
| 959 |
+
"learning_rate": 8.145077720207254e-06,
|
| 960 |
+
"loss": 0.2788,
|
| 961 |
+
"step": 680
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6383970177073626,
|
| 965 |
+
"grad_norm": 0.18249459905289228,
|
| 966 |
+
"learning_rate": 8.041450777202073e-06,
|
| 967 |
+
"loss": 0.2794,
|
| 968 |
+
"step": 685
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.6430568499534017,
|
| 972 |
+
"grad_norm": 0.19591809096684484,
|
| 973 |
+
"learning_rate": 7.937823834196891e-06,
|
| 974 |
+
"loss": 0.2715,
|
| 975 |
+
"step": 690
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.6477166821994408,
|
| 979 |
+
"grad_norm": 0.19514031116444766,
|
| 980 |
+
"learning_rate": 7.834196891191712e-06,
|
| 981 |
+
"loss": 0.2795,
|
| 982 |
+
"step": 695
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.65237651444548,
|
| 986 |
+
"grad_norm": 0.19204390954009534,
|
| 987 |
+
"learning_rate": 7.730569948186528e-06,
|
| 988 |
+
"loss": 0.2837,
|
| 989 |
+
"step": 700
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.6570363466915191,
|
| 993 |
+
"grad_norm": 0.1814003327176538,
|
| 994 |
+
"learning_rate": 7.626943005181348e-06,
|
| 995 |
+
"loss": 0.27,
|
| 996 |
+
"step": 705
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.6616961789375583,
|
| 1000 |
+
"grad_norm": 0.20946585680553592,
|
| 1001 |
+
"learning_rate": 7.523316062176167e-06,
|
| 1002 |
+
"loss": 0.2704,
|
| 1003 |
+
"step": 710
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.6663560111835974,
|
| 1007 |
+
"grad_norm": 0.1921568693032343,
|
| 1008 |
+
"learning_rate": 7.419689119170985e-06,
|
| 1009 |
+
"loss": 0.2677,
|
| 1010 |
+
"step": 715
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.6710158434296365,
|
| 1014 |
+
"grad_norm": 0.19912449557751416,
|
| 1015 |
+
"learning_rate": 7.3160621761658035e-06,
|
| 1016 |
+
"loss": 0.2703,
|
| 1017 |
+
"step": 720
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.6756756756756757,
|
| 1021 |
+
"grad_norm": 0.20566000197011958,
|
| 1022 |
+
"learning_rate": 7.212435233160623e-06,
|
| 1023 |
+
"loss": 0.2766,
|
| 1024 |
+
"step": 725
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6803355079217148,
|
| 1028 |
+
"grad_norm": 0.20652162303920849,
|
| 1029 |
+
"learning_rate": 7.108808290155441e-06,
|
| 1030 |
+
"loss": 0.2727,
|
| 1031 |
+
"step": 730
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.684995340167754,
|
| 1035 |
+
"grad_norm": 0.19038258519772516,
|
| 1036 |
+
"learning_rate": 7.005181347150259e-06,
|
| 1037 |
+
"loss": 0.2724,
|
| 1038 |
+
"step": 735
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6896551724137931,
|
| 1042 |
+
"grad_norm": 0.19490385886001102,
|
| 1043 |
+
"learning_rate": 6.9015544041450784e-06,
|
| 1044 |
+
"loss": 0.2675,
|
| 1045 |
+
"step": 740
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.6943150046598322,
|
| 1049 |
+
"grad_norm": 0.18560068669657961,
|
| 1050 |
+
"learning_rate": 6.797927461139897e-06,
|
| 1051 |
+
"loss": 0.2702,
|
| 1052 |
+
"step": 745
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.6989748369058714,
|
| 1056 |
+
"grad_norm": 0.19621768587568514,
|
| 1057 |
+
"learning_rate": 6.6943005181347155e-06,
|
| 1058 |
+
"loss": 0.2692,
|
| 1059 |
+
"step": 750
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7036346691519105,
|
| 1063 |
+
"grad_norm": 0.2112507969969134,
|
| 1064 |
+
"learning_rate": 6.590673575129535e-06,
|
| 1065 |
+
"loss": 0.2743,
|
| 1066 |
+
"step": 755
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7082945013979497,
|
| 1070 |
+
"grad_norm": 0.18981955629490066,
|
| 1071 |
+
"learning_rate": 6.487046632124353e-06,
|
| 1072 |
+
"loss": 0.2723,
|
| 1073 |
+
"step": 760
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.7129543336439889,
|
| 1077 |
+
"grad_norm": 0.18268361672429612,
|
| 1078 |
+
"learning_rate": 6.383419689119171e-06,
|
| 1079 |
+
"loss": 0.2689,
|
| 1080 |
+
"step": 765
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.7176141658900279,
|
| 1084 |
+
"grad_norm": 0.21759778168797214,
|
| 1085 |
+
"learning_rate": 6.2797927461139905e-06,
|
| 1086 |
+
"loss": 0.275,
|
| 1087 |
+
"step": 770
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.7222739981360671,
|
| 1091 |
+
"grad_norm": 0.19607867923889052,
|
| 1092 |
+
"learning_rate": 6.176165803108809e-06,
|
| 1093 |
+
"loss": 0.2784,
|
| 1094 |
+
"step": 775
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.7269338303821062,
|
| 1098 |
+
"grad_norm": 0.17816386997362446,
|
| 1099 |
+
"learning_rate": 6.0725388601036275e-06,
|
| 1100 |
+
"loss": 0.267,
|
| 1101 |
+
"step": 780
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.7315936626281454,
|
| 1105 |
+
"grad_norm": 0.21117162602410225,
|
| 1106 |
+
"learning_rate": 5.968911917098445e-06,
|
| 1107 |
+
"loss": 0.2698,
|
| 1108 |
+
"step": 785
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.7362534948741846,
|
| 1112 |
+
"grad_norm": 0.21060183599085328,
|
| 1113 |
+
"learning_rate": 5.865284974093265e-06,
|
| 1114 |
+
"loss": 0.2743,
|
| 1115 |
+
"step": 790
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.7409133271202236,
|
| 1119 |
+
"grad_norm": 0.19400476019142068,
|
| 1120 |
+
"learning_rate": 5.761658031088083e-06,
|
| 1121 |
+
"loss": 0.2748,
|
| 1122 |
+
"step": 795
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.7455731593662628,
|
| 1126 |
+
"grad_norm": 0.20052345073381453,
|
| 1127 |
+
"learning_rate": 5.658031088082902e-06,
|
| 1128 |
+
"loss": 0.2736,
|
| 1129 |
+
"step": 800
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.750232991612302,
|
| 1133 |
+
"grad_norm": 0.20728807092698712,
|
| 1134 |
+
"learning_rate": 5.554404145077721e-06,
|
| 1135 |
+
"loss": 0.2743,
|
| 1136 |
+
"step": 805
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.7548928238583411,
|
| 1140 |
+
"grad_norm": 0.19885878147551592,
|
| 1141 |
+
"learning_rate": 5.4507772020725395e-06,
|
| 1142 |
+
"loss": 0.2806,
|
| 1143 |
+
"step": 810
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.7595526561043803,
|
| 1147 |
+
"grad_norm": 0.18112546061809712,
|
| 1148 |
+
"learning_rate": 5.347150259067357e-06,
|
| 1149 |
+
"loss": 0.2721,
|
| 1150 |
+
"step": 815
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.7642124883504194,
|
| 1154 |
+
"grad_norm": 0.18692172501667362,
|
| 1155 |
+
"learning_rate": 5.243523316062177e-06,
|
| 1156 |
+
"loss": 0.2772,
|
| 1157 |
+
"step": 820
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.7688723205964585,
|
| 1161 |
+
"grad_norm": 0.18580108350557434,
|
| 1162 |
+
"learning_rate": 5.139896373056995e-06,
|
| 1163 |
+
"loss": 0.2691,
|
| 1164 |
+
"step": 825
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.7735321528424977,
|
| 1168 |
+
"grad_norm": 0.20217401036904592,
|
| 1169 |
+
"learning_rate": 5.036269430051814e-06,
|
| 1170 |
+
"loss": 0.2693,
|
| 1171 |
+
"step": 830
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.7781919850885368,
|
| 1175 |
+
"grad_norm": 0.1943652264496117,
|
| 1176 |
+
"learning_rate": 4.932642487046633e-06,
|
| 1177 |
+
"loss": 0.2706,
|
| 1178 |
+
"step": 835
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.782851817334576,
|
| 1182 |
+
"grad_norm": 0.2198072283577463,
|
| 1183 |
+
"learning_rate": 4.829015544041451e-06,
|
| 1184 |
+
"loss": 0.2767,
|
| 1185 |
+
"step": 840
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.7875116495806151,
|
| 1189 |
+
"grad_norm": 0.18668864356982678,
|
| 1190 |
+
"learning_rate": 4.72538860103627e-06,
|
| 1191 |
+
"loss": 0.2841,
|
| 1192 |
+
"step": 845
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.7921714818266542,
|
| 1196 |
+
"grad_norm": 0.2046695245569858,
|
| 1197 |
+
"learning_rate": 4.621761658031089e-06,
|
| 1198 |
+
"loss": 0.2691,
|
| 1199 |
+
"step": 850
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.7968313140726934,
|
| 1203 |
+
"grad_norm": 0.20500691517005445,
|
| 1204 |
+
"learning_rate": 4.518134715025907e-06,
|
| 1205 |
+
"loss": 0.2776,
|
| 1206 |
+
"step": 855
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.8014911463187325,
|
| 1210 |
+
"grad_norm": 0.19167582804792827,
|
| 1211 |
+
"learning_rate": 4.414507772020726e-06,
|
| 1212 |
+
"loss": 0.283,
|
| 1213 |
+
"step": 860
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.8061509785647717,
|
| 1217 |
+
"grad_norm": 0.19270689356754245,
|
| 1218 |
+
"learning_rate": 4.310880829015544e-06,
|
| 1219 |
+
"loss": 0.2731,
|
| 1220 |
+
"step": 865
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.8108108108108109,
|
| 1224 |
+
"grad_norm": 0.20242561543530732,
|
| 1225 |
+
"learning_rate": 4.207253886010363e-06,
|
| 1226 |
+
"loss": 0.276,
|
| 1227 |
+
"step": 870
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.8154706430568499,
|
| 1231 |
+
"grad_norm": 0.18482911338240235,
|
| 1232 |
+
"learning_rate": 4.103626943005182e-06,
|
| 1233 |
+
"loss": 0.274,
|
| 1234 |
+
"step": 875
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.8201304753028891,
|
| 1238 |
+
"grad_norm": 0.19023311518057515,
|
| 1239 |
+
"learning_rate": 4.000000000000001e-06,
|
| 1240 |
+
"loss": 0.2744,
|
| 1241 |
+
"step": 880
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.8247903075489282,
|
| 1245 |
+
"grad_norm": 0.17956559274009018,
|
| 1246 |
+
"learning_rate": 3.896373056994819e-06,
|
| 1247 |
+
"loss": 0.2753,
|
| 1248 |
+
"step": 885
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.8294501397949674,
|
| 1252 |
+
"grad_norm": 0.18533025491324542,
|
| 1253 |
+
"learning_rate": 3.7927461139896377e-06,
|
| 1254 |
+
"loss": 0.2716,
|
| 1255 |
+
"step": 890
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.8341099720410066,
|
| 1259 |
+
"grad_norm": 0.18868829919298247,
|
| 1260 |
+
"learning_rate": 3.6891191709844567e-06,
|
| 1261 |
+
"loss": 0.285,
|
| 1262 |
+
"step": 895
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.8387698042870456,
|
| 1266 |
+
"grad_norm": 0.1821294998845929,
|
| 1267 |
+
"learning_rate": 3.5854922279792748e-06,
|
| 1268 |
+
"loss": 0.275,
|
| 1269 |
+
"step": 900
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.8434296365330848,
|
| 1273 |
+
"grad_norm": 0.18362042280215524,
|
| 1274 |
+
"learning_rate": 3.4818652849740937e-06,
|
| 1275 |
+
"loss": 0.2717,
|
| 1276 |
+
"step": 905
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.848089468779124,
|
| 1280 |
+
"grad_norm": 0.18125344385824224,
|
| 1281 |
+
"learning_rate": 3.3782383419689123e-06,
|
| 1282 |
+
"loss": 0.275,
|
| 1283 |
+
"step": 910
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.8527493010251631,
|
| 1287 |
+
"grad_norm": 0.20049284941029782,
|
| 1288 |
+
"learning_rate": 3.274611398963731e-06,
|
| 1289 |
+
"loss": 0.2797,
|
| 1290 |
+
"step": 915
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.8574091332712023,
|
| 1294 |
+
"grad_norm": 0.19284600312397066,
|
| 1295 |
+
"learning_rate": 3.1709844559585493e-06,
|
| 1296 |
+
"loss": 0.2695,
|
| 1297 |
+
"step": 920
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.8620689655172413,
|
| 1301 |
+
"grad_norm": 0.18752491075216704,
|
| 1302 |
+
"learning_rate": 3.0673575129533683e-06,
|
| 1303 |
+
"loss": 0.2664,
|
| 1304 |
+
"step": 925
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.8667287977632805,
|
| 1308 |
+
"grad_norm": 0.1891493054852102,
|
| 1309 |
+
"learning_rate": 2.963730569948187e-06,
|
| 1310 |
+
"loss": 0.2675,
|
| 1311 |
+
"step": 930
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.8713886300093197,
|
| 1315 |
+
"grad_norm": 0.19851493019299982,
|
| 1316 |
+
"learning_rate": 2.8601036269430053e-06,
|
| 1317 |
+
"loss": 0.2699,
|
| 1318 |
+
"step": 935
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.8760484622553588,
|
| 1322 |
+
"grad_norm": 0.19015168393839105,
|
| 1323 |
+
"learning_rate": 2.7564766839378243e-06,
|
| 1324 |
+
"loss": 0.2741,
|
| 1325 |
+
"step": 940
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.880708294501398,
|
| 1329 |
+
"grad_norm": 0.19130488968167667,
|
| 1330 |
+
"learning_rate": 2.6528497409326424e-06,
|
| 1331 |
+
"loss": 0.2698,
|
| 1332 |
+
"step": 945
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.8853681267474371,
|
| 1336 |
+
"grad_norm": 0.19002010995722263,
|
| 1337 |
+
"learning_rate": 2.5492227979274614e-06,
|
| 1338 |
+
"loss": 0.2818,
|
| 1339 |
+
"step": 950
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.8900279589934762,
|
| 1343 |
+
"grad_norm": 0.18897353773835285,
|
| 1344 |
+
"learning_rate": 2.44559585492228e-06,
|
| 1345 |
+
"loss": 0.2745,
|
| 1346 |
+
"step": 955
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.8946877912395154,
|
| 1350 |
+
"grad_norm": 0.1902154880962153,
|
| 1351 |
+
"learning_rate": 2.3419689119170984e-06,
|
| 1352 |
+
"loss": 0.2798,
|
| 1353 |
+
"step": 960
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.8993476234855545,
|
| 1357 |
+
"grad_norm": 0.20192915350132004,
|
| 1358 |
+
"learning_rate": 2.2383419689119174e-06,
|
| 1359 |
+
"loss": 0.2771,
|
| 1360 |
+
"step": 965
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.9040074557315937,
|
| 1364 |
+
"grad_norm": 0.18079604071938268,
|
| 1365 |
+
"learning_rate": 2.134715025906736e-06,
|
| 1366 |
+
"loss": 0.265,
|
| 1367 |
+
"step": 970
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.9086672879776329,
|
| 1371 |
+
"grad_norm": 0.1843476030030155,
|
| 1372 |
+
"learning_rate": 2.0310880829015544e-06,
|
| 1373 |
+
"loss": 0.2807,
|
| 1374 |
+
"step": 975
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.9133271202236719,
|
| 1378 |
+
"grad_norm": 0.2150486826165296,
|
| 1379 |
+
"learning_rate": 1.9274611398963734e-06,
|
| 1380 |
+
"loss": 0.2727,
|
| 1381 |
+
"step": 980
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.9179869524697111,
|
| 1385 |
+
"grad_norm": 0.19640451524415894,
|
| 1386 |
+
"learning_rate": 1.823834196891192e-06,
|
| 1387 |
+
"loss": 0.2705,
|
| 1388 |
+
"step": 985
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.9226467847157502,
|
| 1392 |
+
"grad_norm": 0.1881146695565526,
|
| 1393 |
+
"learning_rate": 1.7202072538860104e-06,
|
| 1394 |
+
"loss": 0.2689,
|
| 1395 |
+
"step": 990
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.9273066169617894,
|
| 1399 |
+
"grad_norm": 0.1919966422921075,
|
| 1400 |
+
"learning_rate": 1.6165803108808292e-06,
|
| 1401 |
+
"loss": 0.2739,
|
| 1402 |
+
"step": 995
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.9319664492078286,
|
| 1406 |
+
"grad_norm": 0.18379110079996336,
|
| 1407 |
+
"learning_rate": 1.5129533678756477e-06,
|
| 1408 |
+
"loss": 0.2726,
|
| 1409 |
+
"step": 1000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.9366262814538676,
|
| 1413 |
+
"grad_norm": 0.18038958121343104,
|
| 1414 |
+
"learning_rate": 1.4093264248704663e-06,
|
| 1415 |
+
"loss": 0.2756,
|
| 1416 |
+
"step": 1005
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.9412861136999068,
|
| 1420 |
+
"grad_norm": 0.18912068564324505,
|
| 1421 |
+
"learning_rate": 1.3056994818652852e-06,
|
| 1422 |
+
"loss": 0.2757,
|
| 1423 |
+
"step": 1010
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.9459459459459459,
|
| 1427 |
+
"grad_norm": 0.19447222147289886,
|
| 1428 |
+
"learning_rate": 1.2020725388601037e-06,
|
| 1429 |
+
"loss": 0.2773,
|
| 1430 |
+
"step": 1015
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.9506057781919851,
|
| 1434 |
+
"grad_norm": 0.17799555766267627,
|
| 1435 |
+
"learning_rate": 1.0984455958549225e-06,
|
| 1436 |
+
"loss": 0.2736,
|
| 1437 |
+
"step": 1020
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.9552656104380243,
|
| 1441 |
+
"grad_norm": 0.1879289390938795,
|
| 1442 |
+
"learning_rate": 9.94818652849741e-07,
|
| 1443 |
+
"loss": 0.2829,
|
| 1444 |
+
"step": 1025
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.9599254426840633,
|
| 1448 |
+
"grad_norm": 0.18465306001205795,
|
| 1449 |
+
"learning_rate": 8.911917098445596e-07,
|
| 1450 |
+
"loss": 0.2712,
|
| 1451 |
+
"step": 1030
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.9645852749301025,
|
| 1455 |
+
"grad_norm": 0.18684975369398882,
|
| 1456 |
+
"learning_rate": 7.875647668393784e-07,
|
| 1457 |
+
"loss": 0.2762,
|
| 1458 |
+
"step": 1035
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.9692451071761417,
|
| 1462 |
+
"grad_norm": 0.18354813003667567,
|
| 1463 |
+
"learning_rate": 6.839378238341969e-07,
|
| 1464 |
+
"loss": 0.2723,
|
| 1465 |
+
"step": 1040
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.9739049394221808,
|
| 1469 |
+
"grad_norm": 0.1790853955275003,
|
| 1470 |
+
"learning_rate": 5.803108808290156e-07,
|
| 1471 |
+
"loss": 0.2686,
|
| 1472 |
+
"step": 1045
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.97856477166822,
|
| 1476 |
+
"grad_norm": 0.18958296444822564,
|
| 1477 |
+
"learning_rate": 4.7668393782383424e-07,
|
| 1478 |
+
"loss": 0.286,
|
| 1479 |
+
"step": 1050
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.983224603914259,
|
| 1483 |
+
"grad_norm": 0.1857121657317691,
|
| 1484 |
+
"learning_rate": 3.730569948186528e-07,
|
| 1485 |
+
"loss": 0.2674,
|
| 1486 |
+
"step": 1055
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.9878844361602982,
|
| 1490 |
+
"grad_norm": 0.17831325037974058,
|
| 1491 |
+
"learning_rate": 2.694300518134715e-07,
|
| 1492 |
+
"loss": 0.272,
|
| 1493 |
+
"step": 1060
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.9925442684063374,
|
| 1497 |
+
"grad_norm": 0.18156458650747095,
|
| 1498 |
+
"learning_rate": 1.6580310880829015e-07,
|
| 1499 |
+
"loss": 0.2746,
|
| 1500 |
+
"step": 1065
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.9972041006523765,
|
| 1504 |
+
"grad_norm": 0.18194645895560613,
|
| 1505 |
+
"learning_rate": 6.217616580310881e-08,
|
| 1506 |
+
"loss": 0.2705,
|
| 1507 |
+
"step": 1070
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 1.0,
|
| 1511 |
+
"step": 1073,
|
| 1512 |
+
"total_flos": 9.186429923093381e+17,
|
| 1513 |
+
"train_loss": 0.27417567097351914,
|
| 1514 |
+
"train_runtime": 29088.9913,
|
| 1515 |
+
"train_samples_per_second": 0.59,
|
| 1516 |
+
"train_steps_per_second": 0.037
|
| 1517 |
+
}
|
| 1518 |
+
],
|
| 1519 |
+
"logging_steps": 5,
|
| 1520 |
+
"max_steps": 1073,
|
| 1521 |
+
"num_input_tokens_seen": 0,
|
| 1522 |
+
"num_train_epochs": 1,
|
| 1523 |
+
"save_steps": 500,
|
| 1524 |
+
"stateful_callbacks": {
|
| 1525 |
+
"TrainerControl": {
|
| 1526 |
+
"args": {
|
| 1527 |
+
"should_epoch_stop": false,
|
| 1528 |
+
"should_evaluate": false,
|
| 1529 |
+
"should_log": false,
|
| 1530 |
+
"should_save": true,
|
| 1531 |
+
"should_training_stop": true
|
| 1532 |
+
},
|
| 1533 |
+
"attributes": {}
|
| 1534 |
+
}
|
| 1535 |
+
},
|
| 1536 |
+
"total_flos": 9.186429923093381e+17,
|
| 1537 |
+
"train_batch_size": 1,
|
| 1538 |
+
"trial_name": null,
|
| 1539 |
+
"trial_params": null
|
| 1540 |
+
}
|