Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- train_results.json +9 -0
- trainer_state.json +2418 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
model_name: Qwen2.5-7B-Open-R1-Distill
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
licence: license
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# Model Card for Qwen2.5-7B-Open-R1-Distill
|
| 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="howardzhou/Qwen2.5-7B-Open-R1-Distill", 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/howardzhou92-nw/huggingface/runs/6rgf4qfj)
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.15.0.dev0
|
| 37 |
+
- Transformers: 4.49.0.dev0
|
| 38 |
+
- Pytorch: 2.5.1
|
| 39 |
+
- Datasets: 3.2.0
|
| 40 |
+
- Tokenizers: 0.21.0
|
| 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 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.996662958843159,
|
| 3 |
+
"total_flos": 352364519227392.0,
|
| 4 |
+
"train_loss": 0.5676420692886625,
|
| 5 |
+
"train_runtime": 10418.0052,
|
| 6 |
+
"train_samples": 16610,
|
| 7 |
+
"train_samples_per_second": 4.141,
|
| 8 |
+
"train_steps_per_second": 0.032
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.49.0.dev0"
|
| 6 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.996662958843159,
|
| 3 |
+
"total_flos": 352364519227392.0,
|
| 4 |
+
"train_loss": 0.5676420692886625,
|
| 5 |
+
"train_runtime": 10418.0052,
|
| 6 |
+
"train_samples": 16610,
|
| 7 |
+
"train_samples_per_second": 4.141,
|
| 8 |
+
"train_steps_per_second": 0.032
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2418 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.996662958843159,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 336,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005932517612161661,
|
| 13 |
+
"grad_norm": 5.324305785932815,
|
| 14 |
+
"learning_rate": 1.999956289272873e-05,
|
| 15 |
+
"loss": 0.9594,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.011865035224323322,
|
| 20 |
+
"grad_norm": 22.84012317380655,
|
| 21 |
+
"learning_rate": 1.9998251609127465e-05,
|
| 22 |
+
"loss": 0.9271,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.017797552836484983,
|
| 27 |
+
"grad_norm": 4.084146062025191,
|
| 28 |
+
"learning_rate": 1.9996066263830533e-05,
|
| 29 |
+
"loss": 0.885,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.023730070448646643,
|
| 34 |
+
"grad_norm": 5.3602743742788626,
|
| 35 |
+
"learning_rate": 1.9993007047883988e-05,
|
| 36 |
+
"loss": 0.8686,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.029662588060808306,
|
| 41 |
+
"grad_norm": 3.8319494768533953,
|
| 42 |
+
"learning_rate": 1.9989074228728942e-05,
|
| 43 |
+
"loss": 0.8172,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.035595105672969966,
|
| 48 |
+
"grad_norm": 2.175190126654642,
|
| 49 |
+
"learning_rate": 1.998426815017817e-05,
|
| 50 |
+
"loss": 0.7451,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.041527623285131626,
|
| 55 |
+
"grad_norm": 2.199115330911503,
|
| 56 |
+
"learning_rate": 1.9978589232386036e-05,
|
| 57 |
+
"loss": 0.7646,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.047460140897293286,
|
| 62 |
+
"grad_norm": 2.1181147808123333,
|
| 63 |
+
"learning_rate": 1.9972037971811802e-05,
|
| 64 |
+
"loss": 0.7291,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.05339265850945495,
|
| 69 |
+
"grad_norm": 1.7260889336074359,
|
| 70 |
+
"learning_rate": 1.9964614941176194e-05,
|
| 71 |
+
"loss": 0.7191,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.05932517612161661,
|
| 76 |
+
"grad_norm": 1.1637147793554257,
|
| 77 |
+
"learning_rate": 1.9956320789411338e-05,
|
| 78 |
+
"loss": 0.7203,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.06525769373377827,
|
| 83 |
+
"grad_norm": 1.4450484122253298,
|
| 84 |
+
"learning_rate": 1.994715624160405e-05,
|
| 85 |
+
"loss": 0.6792,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.07119021134593993,
|
| 90 |
+
"grad_norm": 1.1620631343638403,
|
| 91 |
+
"learning_rate": 1.9937122098932428e-05,
|
| 92 |
+
"loss": 0.7178,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0771227289581016,
|
| 97 |
+
"grad_norm": 0.9627269401137271,
|
| 98 |
+
"learning_rate": 1.992621923859581e-05,
|
| 99 |
+
"loss": 0.6847,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.08305524657026325,
|
| 104 |
+
"grad_norm": 0.9961377846813068,
|
| 105 |
+
"learning_rate": 1.9914448613738107e-05,
|
| 106 |
+
"loss": 0.6762,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.08898776418242492,
|
| 111 |
+
"grad_norm": 0.8096822959197216,
|
| 112 |
+
"learning_rate": 1.9901811253364458e-05,
|
| 113 |
+
"loss": 0.6961,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.09492028179458657,
|
| 118 |
+
"grad_norm": 0.856064052820252,
|
| 119 |
+
"learning_rate": 1.9888308262251286e-05,
|
| 120 |
+
"loss": 0.6797,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.10085279940674824,
|
| 125 |
+
"grad_norm": 1.0455477055796631,
|
| 126 |
+
"learning_rate": 1.9873940820849714e-05,
|
| 127 |
+
"loss": 0.6807,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.1067853170189099,
|
| 132 |
+
"grad_norm": 0.7952535188433517,
|
| 133 |
+
"learning_rate": 1.985871018518236e-05,
|
| 134 |
+
"loss": 0.6792,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.11271783463107156,
|
| 139 |
+
"grad_norm": 0.912745561968978,
|
| 140 |
+
"learning_rate": 1.9842617686733546e-05,
|
| 141 |
+
"loss": 0.6581,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.11865035224323323,
|
| 146 |
+
"grad_norm": 0.9048458674371787,
|
| 147 |
+
"learning_rate": 1.9825664732332886e-05,
|
| 148 |
+
"loss": 0.6593,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.12458286985539488,
|
| 153 |
+
"grad_norm": 0.9891444883072399,
|
| 154 |
+
"learning_rate": 1.9807852804032306e-05,
|
| 155 |
+
"loss": 0.6646,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.13051538746755653,
|
| 160 |
+
"grad_norm": 0.8380876682823516,
|
| 161 |
+
"learning_rate": 1.9789183458976485e-05,
|
| 162 |
+
"loss": 0.6555,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.1364479050797182,
|
| 167 |
+
"grad_norm": 0.7752881193795825,
|
| 168 |
+
"learning_rate": 1.9769658329266718e-05,
|
| 169 |
+
"loss": 0.6423,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.14238042269187987,
|
| 174 |
+
"grad_norm": 0.6671431472869794,
|
| 175 |
+
"learning_rate": 1.9749279121818235e-05,
|
| 176 |
+
"loss": 0.6663,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.14831294030404152,
|
| 181 |
+
"grad_norm": 0.8415919815133858,
|
| 182 |
+
"learning_rate": 1.9728047618210995e-05,
|
| 183 |
+
"loss": 0.6716,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.1542454579162032,
|
| 188 |
+
"grad_norm": 0.7245934639150007,
|
| 189 |
+
"learning_rate": 1.970596567453391e-05,
|
| 190 |
+
"loss": 0.6312,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.16017797552836485,
|
| 195 |
+
"grad_norm": 0.9130270576949818,
|
| 196 |
+
"learning_rate": 1.9683035221222617e-05,
|
| 197 |
+
"loss": 0.6449,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.1661104931405265,
|
| 202 |
+
"grad_norm": 0.6519157743634046,
|
| 203 |
+
"learning_rate": 1.9659258262890683e-05,
|
| 204 |
+
"loss": 0.6382,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.17204301075268819,
|
| 209 |
+
"grad_norm": 0.831521850696097,
|
| 210 |
+
"learning_rate": 1.9634636878154393e-05,
|
| 211 |
+
"loss": 0.6489,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.17797552836484984,
|
| 216 |
+
"grad_norm": 1.4176318143795756,
|
| 217 |
+
"learning_rate": 1.9609173219450998e-05,
|
| 218 |
+
"loss": 0.6103,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1839080459770115,
|
| 223 |
+
"grad_norm": 0.7923365048789488,
|
| 224 |
+
"learning_rate": 1.9582869512850576e-05,
|
| 225 |
+
"loss": 0.6577,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.18984056358917314,
|
| 230 |
+
"grad_norm": 0.8145126868212449,
|
| 231 |
+
"learning_rate": 1.955572805786141e-05,
|
| 232 |
+
"loss": 0.6601,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.19577308120133483,
|
| 237 |
+
"grad_norm": 0.7417273199883386,
|
| 238 |
+
"learning_rate": 1.9527751227228964e-05,
|
| 239 |
+
"loss": 0.6357,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.20170559881349648,
|
| 244 |
+
"grad_norm": 0.9812326390343167,
|
| 245 |
+
"learning_rate": 1.9498941466728462e-05,
|
| 246 |
+
"loss": 0.6435,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.20763811642565813,
|
| 251 |
+
"grad_norm": 0.7122362598498115,
|
| 252 |
+
"learning_rate": 1.946930129495106e-05,
|
| 253 |
+
"loss": 0.6444,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.2135706340378198,
|
| 258 |
+
"grad_norm": 0.6999496527556666,
|
| 259 |
+
"learning_rate": 1.9438833303083677e-05,
|
| 260 |
+
"loss": 0.6216,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.21950315164998146,
|
| 265 |
+
"grad_norm": 0.9616816358092815,
|
| 266 |
+
"learning_rate": 1.9407540154682473e-05,
|
| 267 |
+
"loss": 0.6579,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.22543566926214312,
|
| 272 |
+
"grad_norm": 0.7213515842394086,
|
| 273 |
+
"learning_rate": 1.9375424585439994e-05,
|
| 274 |
+
"loss": 0.6281,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.23136818687430477,
|
| 279 |
+
"grad_norm": 0.9878068930431408,
|
| 280 |
+
"learning_rate": 1.9342489402945997e-05,
|
| 281 |
+
"loss": 0.633,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.23730070448646645,
|
| 286 |
+
"grad_norm": 0.6631768147177095,
|
| 287 |
+
"learning_rate": 1.9308737486442045e-05,
|
| 288 |
+
"loss": 0.6093,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.2432332220986281,
|
| 293 |
+
"grad_norm": 0.8581774181959311,
|
| 294 |
+
"learning_rate": 1.927417178656975e-05,
|
| 295 |
+
"loss": 0.6579,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.24916573971078976,
|
| 300 |
+
"grad_norm": 0.6410107561811785,
|
| 301 |
+
"learning_rate": 1.9238795325112867e-05,
|
| 302 |
+
"loss": 0.6203,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.25509825732295144,
|
| 307 |
+
"grad_norm": 0.8809363552180719,
|
| 308 |
+
"learning_rate": 1.9202611194733107e-05,
|
| 309 |
+
"loss": 0.6034,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.26103077493511306,
|
| 314 |
+
"grad_norm": 0.6867316845306615,
|
| 315 |
+
"learning_rate": 1.9165622558699763e-05,
|
| 316 |
+
"loss": 0.6516,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.26696329254727474,
|
| 321 |
+
"grad_norm": 1.0648968501842426,
|
| 322 |
+
"learning_rate": 1.912783265061319e-05,
|
| 323 |
+
"loss": 0.6446,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.2728958101594364,
|
| 328 |
+
"grad_norm": 0.7208316977945584,
|
| 329 |
+
"learning_rate": 1.908924477412211e-05,
|
| 330 |
+
"loss": 0.6351,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.27882832777159805,
|
| 335 |
+
"grad_norm": 0.8089443201366949,
|
| 336 |
+
"learning_rate": 1.90498623026348e-05,
|
| 337 |
+
"loss": 0.6386,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.28476084538375973,
|
| 342 |
+
"grad_norm": 0.6427781719286114,
|
| 343 |
+
"learning_rate": 1.900968867902419e-05,
|
| 344 |
+
"loss": 0.6243,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.2906933629959214,
|
| 349 |
+
"grad_norm": 0.7797049642173535,
|
| 350 |
+
"learning_rate": 1.8968727415326885e-05,
|
| 351 |
+
"loss": 0.6124,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.29662588060808304,
|
| 356 |
+
"grad_norm": 0.5348618501721483,
|
| 357 |
+
"learning_rate": 1.8926982092436117e-05,
|
| 358 |
+
"loss": 0.6176,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.3025583982202447,
|
| 363 |
+
"grad_norm": 0.8550876198086238,
|
| 364 |
+
"learning_rate": 1.8884456359788725e-05,
|
| 365 |
+
"loss": 0.6271,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.3084909158324064,
|
| 370 |
+
"grad_norm": 0.6138520011172943,
|
| 371 |
+
"learning_rate": 1.8841153935046098e-05,
|
| 372 |
+
"loss": 0.625,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.314423433444568,
|
| 377 |
+
"grad_norm": 0.7544109212965426,
|
| 378 |
+
"learning_rate": 1.8797078603769184e-05,
|
| 379 |
+
"loss": 0.6451,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.3203559510567297,
|
| 384 |
+
"grad_norm": 0.6909872317657667,
|
| 385 |
+
"learning_rate": 1.8752234219087538e-05,
|
| 386 |
+
"loss": 0.623,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.3262884686688914,
|
| 391 |
+
"grad_norm": 0.6057588839577427,
|
| 392 |
+
"learning_rate": 1.8706624701362485e-05,
|
| 393 |
+
"loss": 0.6173,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.332220986281053,
|
| 398 |
+
"grad_norm": 0.7753788330958152,
|
| 399 |
+
"learning_rate": 1.866025403784439e-05,
|
| 400 |
+
"loss": 0.5968,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3381535038932147,
|
| 405 |
+
"grad_norm": 0.5677646570795517,
|
| 406 |
+
"learning_rate": 1.8613126282324092e-05,
|
| 407 |
+
"loss": 0.6083,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.34408602150537637,
|
| 412 |
+
"grad_norm": 0.8456419916175821,
|
| 413 |
+
"learning_rate": 1.8565245554778516e-05,
|
| 414 |
+
"loss": 0.6222,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.350018539117538,
|
| 419 |
+
"grad_norm": 0.7548049715006433,
|
| 420 |
+
"learning_rate": 1.8516616041010495e-05,
|
| 421 |
+
"loss": 0.6328,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.3559510567296997,
|
| 426 |
+
"grad_norm": 0.7884744995171575,
|
| 427 |
+
"learning_rate": 1.8467241992282842e-05,
|
| 428 |
+
"loss": 0.5983,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.3618835743418613,
|
| 433 |
+
"grad_norm": 0.7405638018271332,
|
| 434 |
+
"learning_rate": 1.84171277249467e-05,
|
| 435 |
+
"loss": 0.6158,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.367816091954023,
|
| 440 |
+
"grad_norm": 0.7437562413628,
|
| 441 |
+
"learning_rate": 1.83662776200642e-05,
|
| 442 |
+
"loss": 0.6083,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.37374860956618466,
|
| 447 |
+
"grad_norm": 0.6809427452609553,
|
| 448 |
+
"learning_rate": 1.8314696123025456e-05,
|
| 449 |
+
"loss": 0.6384,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.3796811271783463,
|
| 454 |
+
"grad_norm": 0.7150552234310213,
|
| 455 |
+
"learning_rate": 1.826238774315995e-05,
|
| 456 |
+
"loss": 0.5962,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.38561364479050797,
|
| 461 |
+
"grad_norm": 0.7499268266093695,
|
| 462 |
+
"learning_rate": 1.8209357053342325e-05,
|
| 463 |
+
"loss": 0.5992,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.39154616240266965,
|
| 468 |
+
"grad_norm": 0.701144986922023,
|
| 469 |
+
"learning_rate": 1.8155608689592604e-05,
|
| 470 |
+
"loss": 0.6224,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.3974786800148313,
|
| 475 |
+
"grad_norm": 0.7848864580999906,
|
| 476 |
+
"learning_rate": 1.8101147350670905e-05,
|
| 477 |
+
"loss": 0.6063,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.40341119762699296,
|
| 482 |
+
"grad_norm": 0.5906752612381765,
|
| 483 |
+
"learning_rate": 1.8045977797666685e-05,
|
| 484 |
+
"loss": 0.5999,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.40934371523915464,
|
| 489 |
+
"grad_norm": 0.8877108267943582,
|
| 490 |
+
"learning_rate": 1.7990104853582494e-05,
|
| 491 |
+
"loss": 0.6236,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.41527623285131626,
|
| 496 |
+
"grad_norm": 0.6701464198767606,
|
| 497 |
+
"learning_rate": 1.7933533402912354e-05,
|
| 498 |
+
"loss": 0.6158,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.42120875046347794,
|
| 503 |
+
"grad_norm": 0.8978968992213687,
|
| 504 |
+
"learning_rate": 1.7876268391214756e-05,
|
| 505 |
+
"loss": 0.6139,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.4271412680756396,
|
| 510 |
+
"grad_norm": 0.5561189587112576,
|
| 511 |
+
"learning_rate": 1.78183148246803e-05,
|
| 512 |
+
"loss": 0.5855,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.43307378568780125,
|
| 517 |
+
"grad_norm": 0.9537599050583528,
|
| 518 |
+
"learning_rate": 1.775967776969405e-05,
|
| 519 |
+
"loss": 0.6143,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.43900630329996293,
|
| 524 |
+
"grad_norm": 0.7065253211082851,
|
| 525 |
+
"learning_rate": 1.7700362352392632e-05,
|
| 526 |
+
"loss": 0.6316,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.44493882091212456,
|
| 531 |
+
"grad_norm": 0.8079873624667518,
|
| 532 |
+
"learning_rate": 1.7640373758216075e-05,
|
| 533 |
+
"loss": 0.5903,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.45087133852428624,
|
| 538 |
+
"grad_norm": 0.7300425305669899,
|
| 539 |
+
"learning_rate": 1.757971723145453e-05,
|
| 540 |
+
"loss": 0.5841,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.4568038561364479,
|
| 545 |
+
"grad_norm": 0.6134926917670087,
|
| 546 |
+
"learning_rate": 1.7518398074789776e-05,
|
| 547 |
+
"loss": 0.613,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.46273637374860954,
|
| 552 |
+
"grad_norm": 0.6589828304739275,
|
| 553 |
+
"learning_rate": 1.7456421648831658e-05,
|
| 554 |
+
"loss": 0.595,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.4686688913607712,
|
| 559 |
+
"grad_norm": 0.6394731073863064,
|
| 560 |
+
"learning_rate": 1.739379337164946e-05,
|
| 561 |
+
"loss": 0.6126,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.4746014089729329,
|
| 566 |
+
"grad_norm": 0.615495452837106,
|
| 567 |
+
"learning_rate": 1.7330518718298263e-05,
|
| 568 |
+
"loss": 0.5948,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.48053392658509453,
|
| 573 |
+
"grad_norm": 0.6010913241719915,
|
| 574 |
+
"learning_rate": 1.7266603220340273e-05,
|
| 575 |
+
"loss": 0.6127,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.4864664441972562,
|
| 580 |
+
"grad_norm": 0.6447864666738669,
|
| 581 |
+
"learning_rate": 1.7202052465361268e-05,
|
| 582 |
+
"loss": 0.5975,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.4923989618094179,
|
| 587 |
+
"grad_norm": 0.5299790811179832,
|
| 588 |
+
"learning_rate": 1.7136872096482123e-05,
|
| 589 |
+
"loss": 0.6191,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.4983314794215795,
|
| 594 |
+
"grad_norm": 0.6051833501420879,
|
| 595 |
+
"learning_rate": 1.7071067811865477e-05,
|
| 596 |
+
"loss": 0.616,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.5042639970337411,
|
| 601 |
+
"grad_norm": 0.545917313248304,
|
| 602 |
+
"learning_rate": 1.7004645364217584e-05,
|
| 603 |
+
"loss": 0.6186,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5101965146459029,
|
| 608 |
+
"grad_norm": 0.6742615641272589,
|
| 609 |
+
"learning_rate": 1.693761056028542e-05,
|
| 610 |
+
"loss": 0.6218,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.5161290322580645,
|
| 615 |
+
"grad_norm": 0.56943138685776,
|
| 616 |
+
"learning_rate": 1.686996926034902e-05,
|
| 617 |
+
"loss": 0.6113,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.5220615498702261,
|
| 622 |
+
"grad_norm": 0.5709110972034852,
|
| 623 |
+
"learning_rate": 1.6801727377709195e-05,
|
| 624 |
+
"loss": 0.6125,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5279940674823879,
|
| 629 |
+
"grad_norm": 0.5196995804675383,
|
| 630 |
+
"learning_rate": 1.6732890878170573e-05,
|
| 631 |
+
"loss": 0.5862,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5339265850945495,
|
| 636 |
+
"grad_norm": 0.5096600514404106,
|
| 637 |
+
"learning_rate": 1.6663465779520042e-05,
|
| 638 |
+
"loss": 0.592,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.5398591027067111,
|
| 643 |
+
"grad_norm": 0.5781695292378604,
|
| 644 |
+
"learning_rate": 1.659345815100069e-05,
|
| 645 |
+
"loss": 0.5741,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.5457916203188728,
|
| 650 |
+
"grad_norm": 0.5603815789576477,
|
| 651 |
+
"learning_rate": 1.6522874112781213e-05,
|
| 652 |
+
"loss": 0.6247,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.5517241379310345,
|
| 657 |
+
"grad_norm": 0.5795708352012364,
|
| 658 |
+
"learning_rate": 1.645171983542088e-05,
|
| 659 |
+
"loss": 0.6023,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.5576566555431961,
|
| 664 |
+
"grad_norm": 0.543427974530907,
|
| 665 |
+
"learning_rate": 1.6380001539330088e-05,
|
| 666 |
+
"loss": 0.5991,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.5635891731553578,
|
| 671 |
+
"grad_norm": 0.6346825164978813,
|
| 672 |
+
"learning_rate": 1.6307725494226586e-05,
|
| 673 |
+
"loss": 0.6374,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.5695216907675195,
|
| 678 |
+
"grad_norm": 0.5799055100493153,
|
| 679 |
+
"learning_rate": 1.6234898018587336e-05,
|
| 680 |
+
"loss": 0.6267,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5754542083796811,
|
| 685 |
+
"grad_norm": 0.5921423473197417,
|
| 686 |
+
"learning_rate": 1.616152547909618e-05,
|
| 687 |
+
"loss": 0.6164,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.5813867259918428,
|
| 692 |
+
"grad_norm": 0.49534355256808843,
|
| 693 |
+
"learning_rate": 1.608761429008721e-05,
|
| 694 |
+
"loss": 0.578,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5873192436040044,
|
| 699 |
+
"grad_norm": 0.6107346169575619,
|
| 700 |
+
"learning_rate": 1.601317091298406e-05,
|
| 701 |
+
"loss": 0.5756,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.5932517612161661,
|
| 706 |
+
"grad_norm": 0.5446618742898293,
|
| 707 |
+
"learning_rate": 1.5938201855735017e-05,
|
| 708 |
+
"loss": 0.6025,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5932517612161661,
|
| 713 |
+
"eval_loss": 0.6248904466629028,
|
| 714 |
+
"eval_runtime": 6.4597,
|
| 715 |
+
"eval_samples_per_second": 19.66,
|
| 716 |
+
"eval_steps_per_second": 2.477,
|
| 717 |
+
"step": 100
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.5991842788283278,
|
| 721 |
+
"grad_norm": 0.5859755046139327,
|
| 722 |
+
"learning_rate": 1.5862713672244092e-05,
|
| 723 |
+
"loss": 0.5881,
|
| 724 |
+
"step": 101
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.6051167964404894,
|
| 728 |
+
"grad_norm": 0.6191852673376101,
|
| 729 |
+
"learning_rate": 1.578671296179806e-05,
|
| 730 |
+
"loss": 0.605,
|
| 731 |
+
"step": 102
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.6110493140526511,
|
| 735 |
+
"grad_norm": 0.7094358933361382,
|
| 736 |
+
"learning_rate": 1.5710206368489555e-05,
|
| 737 |
+
"loss": 0.6173,
|
| 738 |
+
"step": 103
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.6169818316648128,
|
| 742 |
+
"grad_norm": 0.5352134780488959,
|
| 743 |
+
"learning_rate": 1.563320058063622e-05,
|
| 744 |
+
"loss": 0.5986,
|
| 745 |
+
"step": 104
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.6229143492769744,
|
| 749 |
+
"grad_norm": 0.6198887975123263,
|
| 750 |
+
"learning_rate": 1.5555702330196024e-05,
|
| 751 |
+
"loss": 0.5871,
|
| 752 |
+
"step": 105
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.628846866889136,
|
| 756 |
+
"grad_norm": 0.6049898880261826,
|
| 757 |
+
"learning_rate": 1.5477718392178716e-05,
|
| 758 |
+
"loss": 0.6041,
|
| 759 |
+
"step": 106
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.6347793845012978,
|
| 763 |
+
"grad_norm": 0.7958720800000618,
|
| 764 |
+
"learning_rate": 1.5399255584053568e-05,
|
| 765 |
+
"loss": 0.5951,
|
| 766 |
+
"step": 107
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.6407119021134594,
|
| 770 |
+
"grad_norm": 0.6616316291827746,
|
| 771 |
+
"learning_rate": 1.5320320765153367e-05,
|
| 772 |
+
"loss": 0.6084,
|
| 773 |
+
"step": 108
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.646644419725621,
|
| 777 |
+
"grad_norm": 0.5322657642988358,
|
| 778 |
+
"learning_rate": 1.5240920836074777e-05,
|
| 779 |
+
"loss": 0.6065,
|
| 780 |
+
"step": 109
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.6525769373377828,
|
| 784 |
+
"grad_norm": 0.6975717651431369,
|
| 785 |
+
"learning_rate": 1.5161062738075068e-05,
|
| 786 |
+
"loss": 0.6123,
|
| 787 |
+
"step": 110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.6585094549499444,
|
| 791 |
+
"grad_norm": 0.5314930484389285,
|
| 792 |
+
"learning_rate": 1.5080753452465296e-05,
|
| 793 |
+
"loss": 0.5953,
|
| 794 |
+
"step": 111
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.664441972562106,
|
| 798 |
+
"grad_norm": 0.6439755660279854,
|
| 799 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 800 |
+
"loss": 0.5794,
|
| 801 |
+
"step": 112
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.6703744901742678,
|
| 805 |
+
"grad_norm": 0.6572591013210453,
|
| 806 |
+
"learning_rate": 1.4918809440263435e-05,
|
| 807 |
+
"loss": 0.6116,
|
| 808 |
+
"step": 113
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.6763070077864294,
|
| 812 |
+
"grad_norm": 0.7834266438815685,
|
| 813 |
+
"learning_rate": 1.4837188871052399e-05,
|
| 814 |
+
"loss": 0.5871,
|
| 815 |
+
"step": 114
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.682239525398591,
|
| 819 |
+
"grad_norm": 0.7160570880780417,
|
| 820 |
+
"learning_rate": 1.4755145427755755e-05,
|
| 821 |
+
"loss": 0.5677,
|
| 822 |
+
"step": 115
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.6881720430107527,
|
| 826 |
+
"grad_norm": 0.6176131655448136,
|
| 827 |
+
"learning_rate": 1.4672686282730622e-05,
|
| 828 |
+
"loss": 0.5804,
|
| 829 |
+
"step": 116
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.6941045606229144,
|
| 833 |
+
"grad_norm": 0.7328033360980243,
|
| 834 |
+
"learning_rate": 1.4589818644675378e-05,
|
| 835 |
+
"loss": 0.5827,
|
| 836 |
+
"step": 117
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.700037078235076,
|
| 840 |
+
"grad_norm": 0.5454029774533725,
|
| 841 |
+
"learning_rate": 1.4506549757999456e-05,
|
| 842 |
+
"loss": 0.585,
|
| 843 |
+
"step": 118
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.7059695958472376,
|
| 847 |
+
"grad_norm": 0.6681806820396734,
|
| 848 |
+
"learning_rate": 1.4422886902190014e-05,
|
| 849 |
+
"loss": 0.5805,
|
| 850 |
+
"step": 119
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.7119021134593994,
|
| 854 |
+
"grad_norm": 0.550161826800762,
|
| 855 |
+
"learning_rate": 1.4338837391175582e-05,
|
| 856 |
+
"loss": 0.5798,
|
| 857 |
+
"step": 120
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.717834631071561,
|
| 861 |
+
"grad_norm": 0.5611088668582199,
|
| 862 |
+
"learning_rate": 1.4254408572686642e-05,
|
| 863 |
+
"loss": 0.5861,
|
| 864 |
+
"step": 121
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.7237671486837226,
|
| 868 |
+
"grad_norm": 0.5366392258292604,
|
| 869 |
+
"learning_rate": 1.4169607827613284e-05,
|
| 870 |
+
"loss": 0.5758,
|
| 871 |
+
"step": 122
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.7296996662958843,
|
| 875 |
+
"grad_norm": 0.6344536914740986,
|
| 876 |
+
"learning_rate": 1.4084442569359964e-05,
|
| 877 |
+
"loss": 0.5786,
|
| 878 |
+
"step": 123
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.735632183908046,
|
| 882 |
+
"grad_norm": 0.5387616042555342,
|
| 883 |
+
"learning_rate": 1.3998920243197408e-05,
|
| 884 |
+
"loss": 0.5722,
|
| 885 |
+
"step": 124
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.7415647015202076,
|
| 889 |
+
"grad_norm": 0.5579694394513077,
|
| 890 |
+
"learning_rate": 1.391304832561175e-05,
|
| 891 |
+
"loss": 0.5907,
|
| 892 |
+
"step": 125
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.7474972191323693,
|
| 896 |
+
"grad_norm": 0.5370248165446521,
|
| 897 |
+
"learning_rate": 1.3826834323650899e-05,
|
| 898 |
+
"loss": 0.5838,
|
| 899 |
+
"step": 126
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.753429736744531,
|
| 903 |
+
"grad_norm": 0.503594155073456,
|
| 904 |
+
"learning_rate": 1.3740285774268282e-05,
|
| 905 |
+
"loss": 0.6,
|
| 906 |
+
"step": 127
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.7593622543566926,
|
| 910 |
+
"grad_norm": 0.5282433463387846,
|
| 911 |
+
"learning_rate": 1.3653410243663953e-05,
|
| 912 |
+
"loss": 0.6192,
|
| 913 |
+
"step": 128
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.7652947719688543,
|
| 917 |
+
"grad_norm": 0.5148036620174011,
|
| 918 |
+
"learning_rate": 1.3566215326623131e-05,
|
| 919 |
+
"loss": 0.5844,
|
| 920 |
+
"step": 129
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.7712272895810159,
|
| 924 |
+
"grad_norm": 0.47585872492395137,
|
| 925 |
+
"learning_rate": 1.3478708645852272e-05,
|
| 926 |
+
"loss": 0.5903,
|
| 927 |
+
"step": 130
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.7771598071931776,
|
| 931 |
+
"grad_norm": 0.5284323921284109,
|
| 932 |
+
"learning_rate": 1.3390897851312667e-05,
|
| 933 |
+
"loss": 0.5822,
|
| 934 |
+
"step": 131
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.7830923248053393,
|
| 938 |
+
"grad_norm": 0.5114134262850306,
|
| 939 |
+
"learning_rate": 1.3302790619551673e-05,
|
| 940 |
+
"loss": 0.5914,
|
| 941 |
+
"step": 132
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.7890248424175009,
|
| 945 |
+
"grad_norm": 0.5033307503491072,
|
| 946 |
+
"learning_rate": 1.3214394653031616e-05,
|
| 947 |
+
"loss": 0.5822,
|
| 948 |
+
"step": 133
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.7949573600296626,
|
| 952 |
+
"grad_norm": 0.5585716796647099,
|
| 953 |
+
"learning_rate": 1.3125717679456447e-05,
|
| 954 |
+
"loss": 0.6147,
|
| 955 |
+
"step": 134
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.8008898776418243,
|
| 959 |
+
"grad_norm": 0.4893700313257534,
|
| 960 |
+
"learning_rate": 1.3036767451096148e-05,
|
| 961 |
+
"loss": 0.5923,
|
| 962 |
+
"step": 135
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.8068223952539859,
|
| 966 |
+
"grad_norm": 0.5257982159346182,
|
| 967 |
+
"learning_rate": 1.2947551744109044e-05,
|
| 968 |
+
"loss": 0.5715,
|
| 969 |
+
"step": 136
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.8127549128661475,
|
| 973 |
+
"grad_norm": 0.6395398302691117,
|
| 974 |
+
"learning_rate": 1.2858078357861979e-05,
|
| 975 |
+
"loss": 0.6105,
|
| 976 |
+
"step": 137
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.8186874304783093,
|
| 980 |
+
"grad_norm": 0.5283010642253179,
|
| 981 |
+
"learning_rate": 1.2768355114248493e-05,
|
| 982 |
+
"loss": 0.5826,
|
| 983 |
+
"step": 138
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.8246199480904709,
|
| 987 |
+
"grad_norm": 0.6035790420033149,
|
| 988 |
+
"learning_rate": 1.2678389857005033e-05,
|
| 989 |
+
"loss": 0.5748,
|
| 990 |
+
"step": 139
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.8305524657026325,
|
| 994 |
+
"grad_norm": 0.5955335369154928,
|
| 995 |
+
"learning_rate": 1.2588190451025209e-05,
|
| 996 |
+
"loss": 0.603,
|
| 997 |
+
"step": 140
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.8364849833147943,
|
| 1001 |
+
"grad_norm": 0.5214414900773817,
|
| 1002 |
+
"learning_rate": 1.249776478167227e-05,
|
| 1003 |
+
"loss": 0.5791,
|
| 1004 |
+
"step": 141
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.8424175009269559,
|
| 1008 |
+
"grad_norm": 0.6417186166595917,
|
| 1009 |
+
"learning_rate": 1.2407120754089733e-05,
|
| 1010 |
+
"loss": 0.5969,
|
| 1011 |
+
"step": 142
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.8483500185391175,
|
| 1015 |
+
"grad_norm": 0.5136915661544744,
|
| 1016 |
+
"learning_rate": 1.2316266292510305e-05,
|
| 1017 |
+
"loss": 0.5983,
|
| 1018 |
+
"step": 143
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.8542825361512792,
|
| 1022 |
+
"grad_norm": 0.6533160354469787,
|
| 1023 |
+
"learning_rate": 1.2225209339563144e-05,
|
| 1024 |
+
"loss": 0.594,
|
| 1025 |
+
"step": 144
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.8602150537634409,
|
| 1029 |
+
"grad_norm": 0.47221905497667216,
|
| 1030 |
+
"learning_rate": 1.2133957855579501e-05,
|
| 1031 |
+
"loss": 0.581,
|
| 1032 |
+
"step": 145
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.8661475713756025,
|
| 1036 |
+
"grad_norm": 0.6131552653676784,
|
| 1037 |
+
"learning_rate": 1.2042519817896805e-05,
|
| 1038 |
+
"loss": 0.6023,
|
| 1039 |
+
"step": 146
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.8720800889877642,
|
| 1043 |
+
"grad_norm": 0.5846657268915201,
|
| 1044 |
+
"learning_rate": 1.1950903220161286e-05,
|
| 1045 |
+
"loss": 0.5961,
|
| 1046 |
+
"step": 147
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.8780126065999259,
|
| 1050 |
+
"grad_norm": 0.5541734985583223,
|
| 1051 |
+
"learning_rate": 1.1859116071629148e-05,
|
| 1052 |
+
"loss": 0.5781,
|
| 1053 |
+
"step": 148
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.8839451242120875,
|
| 1057 |
+
"grad_norm": 0.5275638237960486,
|
| 1058 |
+
"learning_rate": 1.1767166396466404e-05,
|
| 1059 |
+
"loss": 0.6061,
|
| 1060 |
+
"step": 149
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.8898776418242491,
|
| 1064 |
+
"grad_norm": 0.48482325077647975,
|
| 1065 |
+
"learning_rate": 1.1675062233047365e-05,
|
| 1066 |
+
"loss": 0.5905,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.8958101594364108,
|
| 1071 |
+
"grad_norm": 0.42738134564467045,
|
| 1072 |
+
"learning_rate": 1.1582811633251949e-05,
|
| 1073 |
+
"loss": 0.595,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.9017426770485725,
|
| 1078 |
+
"grad_norm": 0.5452867174600223,
|
| 1079 |
+
"learning_rate": 1.1490422661761744e-05,
|
| 1080 |
+
"loss": 0.6139,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.9076751946607341,
|
| 1085 |
+
"grad_norm": 0.45866736271381936,
|
| 1086 |
+
"learning_rate": 1.1397903395354996e-05,
|
| 1087 |
+
"loss": 0.5539,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.9136077122728958,
|
| 1092 |
+
"grad_norm": 0.5301598351347845,
|
| 1093 |
+
"learning_rate": 1.130526192220052e-05,
|
| 1094 |
+
"loss": 0.5902,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.9195402298850575,
|
| 1099 |
+
"grad_norm": 0.4744630476971996,
|
| 1100 |
+
"learning_rate": 1.1212506341150615e-05,
|
| 1101 |
+
"loss": 0.5791,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.9254727474972191,
|
| 1106 |
+
"grad_norm": 0.5400171376545928,
|
| 1107 |
+
"learning_rate": 1.1119644761033079e-05,
|
| 1108 |
+
"loss": 0.5879,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.9314052651093808,
|
| 1113 |
+
"grad_norm": 0.49031695283723725,
|
| 1114 |
+
"learning_rate": 1.1026685299942286e-05,
|
| 1115 |
+
"loss": 0.5786,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.9373377827215424,
|
| 1120 |
+
"grad_norm": 0.541456950819932,
|
| 1121 |
+
"learning_rate": 1.0933636084529507e-05,
|
| 1122 |
+
"loss": 0.5934,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.9432703003337041,
|
| 1127 |
+
"grad_norm": 0.5201441726217624,
|
| 1128 |
+
"learning_rate": 1.0840505249292477e-05,
|
| 1129 |
+
"loss": 0.5931,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.9492028179458658,
|
| 1134 |
+
"grad_norm": 0.5197234596906991,
|
| 1135 |
+
"learning_rate": 1.0747300935864245e-05,
|
| 1136 |
+
"loss": 0.5867,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.9551353355580274,
|
| 1141 |
+
"grad_norm": 0.6456281680986811,
|
| 1142 |
+
"learning_rate": 1.0654031292301432e-05,
|
| 1143 |
+
"loss": 0.5887,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.9610678531701891,
|
| 1148 |
+
"grad_norm": 0.498657726991286,
|
| 1149 |
+
"learning_rate": 1.0560704472371919e-05,
|
| 1150 |
+
"loss": 0.596,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.9670003707823508,
|
| 1155 |
+
"grad_norm": 0.6139669635009651,
|
| 1156 |
+
"learning_rate": 1.0467328634842024e-05,
|
| 1157 |
+
"loss": 0.5827,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.9729328883945124,
|
| 1162 |
+
"grad_norm": 0.6135223890559531,
|
| 1163 |
+
"learning_rate": 1.037391194276326e-05,
|
| 1164 |
+
"loss": 0.596,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.978865406006674,
|
| 1169 |
+
"grad_norm": 0.4774291756729163,
|
| 1170 |
+
"learning_rate": 1.028046256275869e-05,
|
| 1171 |
+
"loss": 0.585,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.9847979236188358,
|
| 1176 |
+
"grad_norm": 0.5242543425803347,
|
| 1177 |
+
"learning_rate": 1.0186988664309023e-05,
|
| 1178 |
+
"loss": 0.577,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.9907304412309974,
|
| 1183 |
+
"grad_norm": 0.45667298643197274,
|
| 1184 |
+
"learning_rate": 1.0093498419038394e-05,
|
| 1185 |
+
"loss": 0.5936,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.996662958843159,
|
| 1190 |
+
"grad_norm": 0.5025893794233189,
|
| 1191 |
+
"learning_rate": 1e-05,
|
| 1192 |
+
"loss": 0.5854,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.0059325176121616,
|
| 1197 |
+
"grad_norm": 1.1546218989986659,
|
| 1198 |
+
"learning_rate": 9.90650158096161e-06,
|
| 1199 |
+
"loss": 1.133,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.0118650352243233,
|
| 1204 |
+
"grad_norm": 0.640747017349487,
|
| 1205 |
+
"learning_rate": 9.81301133569098e-06,
|
| 1206 |
+
"loss": 0.5353,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.0177975528364849,
|
| 1211 |
+
"grad_norm": 0.6151332494210663,
|
| 1212 |
+
"learning_rate": 9.719537437241311e-06,
|
| 1213 |
+
"loss": 0.5355,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.0237300704486467,
|
| 1218 |
+
"grad_norm": 0.6267164652952573,
|
| 1219 |
+
"learning_rate": 9.626088057236745e-06,
|
| 1220 |
+
"loss": 0.5172,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.0296625880608083,
|
| 1225 |
+
"grad_norm": 0.6416046210551611,
|
| 1226 |
+
"learning_rate": 9.532671365157979e-06,
|
| 1227 |
+
"loss": 0.5105,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.03559510567297,
|
| 1232 |
+
"grad_norm": 0.7006960546020584,
|
| 1233 |
+
"learning_rate": 9.439295527628083e-06,
|
| 1234 |
+
"loss": 0.5282,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.0415276232851316,
|
| 1239 |
+
"grad_norm": 0.5780366560749366,
|
| 1240 |
+
"learning_rate": 9.34596870769857e-06,
|
| 1241 |
+
"loss": 0.4976,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.0474601408972932,
|
| 1246 |
+
"grad_norm": 0.5470594926347001,
|
| 1247 |
+
"learning_rate": 9.252699064135759e-06,
|
| 1248 |
+
"loss": 0.5112,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.0533926585094548,
|
| 1253 |
+
"grad_norm": 0.6201320684584469,
|
| 1254 |
+
"learning_rate": 9.159494750707527e-06,
|
| 1255 |
+
"loss": 0.5234,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.0593251761216167,
|
| 1260 |
+
"grad_norm": 0.6107579623832626,
|
| 1261 |
+
"learning_rate": 9.066363915470494e-06,
|
| 1262 |
+
"loss": 0.5137,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.0652576937337783,
|
| 1267 |
+
"grad_norm": 0.5656928235926393,
|
| 1268 |
+
"learning_rate": 8.973314700057717e-06,
|
| 1269 |
+
"loss": 0.5202,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.07119021134594,
|
| 1274 |
+
"grad_norm": 0.6419908684973982,
|
| 1275 |
+
"learning_rate": 8.880355238966923e-06,
|
| 1276 |
+
"loss": 0.5234,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.0771227289581016,
|
| 1281 |
+
"grad_norm": 0.5987046465110973,
|
| 1282 |
+
"learning_rate": 8.787493658849387e-06,
|
| 1283 |
+
"loss": 0.5226,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.0830552465702632,
|
| 1288 |
+
"grad_norm": 0.6073135501636805,
|
| 1289 |
+
"learning_rate": 8.694738077799487e-06,
|
| 1290 |
+
"loss": 0.5237,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.0889877641824248,
|
| 1295 |
+
"grad_norm": 0.523955265242911,
|
| 1296 |
+
"learning_rate": 8.602096604645009e-06,
|
| 1297 |
+
"loss": 0.5128,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.0949202817945867,
|
| 1302 |
+
"grad_norm": 0.5238418004483752,
|
| 1303 |
+
"learning_rate": 8.509577338238255e-06,
|
| 1304 |
+
"loss": 0.5299,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.1008527994067483,
|
| 1309 |
+
"grad_norm": 0.47473612794789644,
|
| 1310 |
+
"learning_rate": 8.417188366748051e-06,
|
| 1311 |
+
"loss": 0.5087,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.10678531701891,
|
| 1316 |
+
"grad_norm": 0.48708620030904176,
|
| 1317 |
+
"learning_rate": 8.324937766952638e-06,
|
| 1318 |
+
"loss": 0.4937,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.1127178346310715,
|
| 1323 |
+
"grad_norm": 0.503979194337666,
|
| 1324 |
+
"learning_rate": 8.232833603533601e-06,
|
| 1325 |
+
"loss": 0.5107,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.1186503522432332,
|
| 1330 |
+
"grad_norm": 0.4721181859728738,
|
| 1331 |
+
"learning_rate": 8.140883928370855e-06,
|
| 1332 |
+
"loss": 0.5092,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.1245828698553948,
|
| 1337 |
+
"grad_norm": 0.47519673374496596,
|
| 1338 |
+
"learning_rate": 8.04909677983872e-06,
|
| 1339 |
+
"loss": 0.5093,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.1305153874675566,
|
| 1344 |
+
"grad_norm": 0.48532470072787237,
|
| 1345 |
+
"learning_rate": 7.957480182103198e-06,
|
| 1346 |
+
"loss": 0.514,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.1364479050797183,
|
| 1351 |
+
"grad_norm": 0.5558277109069734,
|
| 1352 |
+
"learning_rate": 7.866042144420502e-06,
|
| 1353 |
+
"loss": 0.5114,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.14238042269188,
|
| 1358 |
+
"grad_norm": 0.5033467701152651,
|
| 1359 |
+
"learning_rate": 7.774790660436857e-06,
|
| 1360 |
+
"loss": 0.5044,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.1483129403040415,
|
| 1365 |
+
"grad_norm": 0.5253895227761279,
|
| 1366 |
+
"learning_rate": 7.6837337074897e-06,
|
| 1367 |
+
"loss": 0.528,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.1542454579162031,
|
| 1372 |
+
"grad_norm": 0.5839549683192119,
|
| 1373 |
+
"learning_rate": 7.592879245910273e-06,
|
| 1374 |
+
"loss": 0.5143,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.1601779755283648,
|
| 1379 |
+
"grad_norm": 0.5025007441342745,
|
| 1380 |
+
"learning_rate": 7.50223521832773e-06,
|
| 1381 |
+
"loss": 0.5431,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.1661104931405264,
|
| 1386 |
+
"grad_norm": 0.4459742958112407,
|
| 1387 |
+
"learning_rate": 7.411809548974792e-06,
|
| 1388 |
+
"loss": 0.5164,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.1720430107526882,
|
| 1393 |
+
"grad_norm": 0.49959227007299145,
|
| 1394 |
+
"learning_rate": 7.321610142994971e-06,
|
| 1395 |
+
"loss": 0.504,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.1779755283648499,
|
| 1400 |
+
"grad_norm": 0.434451715252235,
|
| 1401 |
+
"learning_rate": 7.2316448857515076e-06,
|
| 1402 |
+
"loss": 0.5146,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.1839080459770115,
|
| 1407 |
+
"grad_norm": 0.41274799335831514,
|
| 1408 |
+
"learning_rate": 7.141921642138025e-06,
|
| 1409 |
+
"loss": 0.4913,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.1898405635891731,
|
| 1414 |
+
"grad_norm": 0.4537854495083879,
|
| 1415 |
+
"learning_rate": 7.052448255890958e-06,
|
| 1416 |
+
"loss": 0.5147,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.1898405635891731,
|
| 1421 |
+
"eval_loss": 0.6126826405525208,
|
| 1422 |
+
"eval_runtime": 6.5201,
|
| 1423 |
+
"eval_samples_per_second": 19.478,
|
| 1424 |
+
"eval_steps_per_second": 2.454,
|
| 1425 |
+
"step": 200
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.1957730812013347,
|
| 1429 |
+
"grad_norm": 0.46234781827027166,
|
| 1430 |
+
"learning_rate": 6.963232548903853e-06,
|
| 1431 |
+
"loss": 0.4987,
|
| 1432 |
+
"step": 201
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.2017055988134966,
|
| 1436 |
+
"grad_norm": 0.45869290271578933,
|
| 1437 |
+
"learning_rate": 6.874282320543557e-06,
|
| 1438 |
+
"loss": 0.4966,
|
| 1439 |
+
"step": 202
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.2076381164256582,
|
| 1443 |
+
"grad_norm": 0.460091111016698,
|
| 1444 |
+
"learning_rate": 6.785605346968387e-06,
|
| 1445 |
+
"loss": 0.5141,
|
| 1446 |
+
"step": 203
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 1.2135706340378198,
|
| 1450 |
+
"grad_norm": 0.4117798491221172,
|
| 1451 |
+
"learning_rate": 6.697209380448333e-06,
|
| 1452 |
+
"loss": 0.5058,
|
| 1453 |
+
"step": 204
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 1.2195031516499815,
|
| 1457 |
+
"grad_norm": 0.45529452974398477,
|
| 1458 |
+
"learning_rate": 6.609102148687333e-06,
|
| 1459 |
+
"loss": 0.5191,
|
| 1460 |
+
"step": 205
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 1.225435669262143,
|
| 1464 |
+
"grad_norm": 0.4406074150768268,
|
| 1465 |
+
"learning_rate": 6.521291354147727e-06,
|
| 1466 |
+
"loss": 0.5158,
|
| 1467 |
+
"step": 206
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 1.2313681868743047,
|
| 1471 |
+
"grad_norm": 0.4265027031097921,
|
| 1472 |
+
"learning_rate": 6.43378467337687e-06,
|
| 1473 |
+
"loss": 0.492,
|
| 1474 |
+
"step": 207
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 1.2373007044864663,
|
| 1478 |
+
"grad_norm": 0.4092683556762198,
|
| 1479 |
+
"learning_rate": 6.34658975633605e-06,
|
| 1480 |
+
"loss": 0.5293,
|
| 1481 |
+
"step": 208
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 1.2432332220986282,
|
| 1485 |
+
"grad_norm": 0.4228580623573835,
|
| 1486 |
+
"learning_rate": 6.2597142257317185e-06,
|
| 1487 |
+
"loss": 0.5081,
|
| 1488 |
+
"step": 209
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 1.2491657397107898,
|
| 1492 |
+
"grad_norm": 0.44825386735657413,
|
| 1493 |
+
"learning_rate": 6.173165676349103e-06,
|
| 1494 |
+
"loss": 0.5116,
|
| 1495 |
+
"step": 210
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 1.2550982573229514,
|
| 1499 |
+
"grad_norm": 0.41296899816759075,
|
| 1500 |
+
"learning_rate": 6.086951674388252e-06,
|
| 1501 |
+
"loss": 0.5091,
|
| 1502 |
+
"step": 211
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 1.261030774935113,
|
| 1506 |
+
"grad_norm": 0.4086723147124949,
|
| 1507 |
+
"learning_rate": 6.001079756802592e-06,
|
| 1508 |
+
"loss": 0.5246,
|
| 1509 |
+
"step": 212
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 1.2669632925472747,
|
| 1513 |
+
"grad_norm": 0.3966806792909518,
|
| 1514 |
+
"learning_rate": 5.91555743064004e-06,
|
| 1515 |
+
"loss": 0.4979,
|
| 1516 |
+
"step": 213
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 1.2728958101594365,
|
| 1520 |
+
"grad_norm": 0.39802032371072127,
|
| 1521 |
+
"learning_rate": 5.830392172386723e-06,
|
| 1522 |
+
"loss": 0.5107,
|
| 1523 |
+
"step": 214
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 1.2788283277715982,
|
| 1527 |
+
"grad_norm": 0.4164302108778062,
|
| 1528 |
+
"learning_rate": 5.745591427313365e-06,
|
| 1529 |
+
"loss": 0.5052,
|
| 1530 |
+
"step": 215
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 1.2847608453837598,
|
| 1534 |
+
"grad_norm": 0.4289140325941008,
|
| 1535 |
+
"learning_rate": 5.66116260882442e-06,
|
| 1536 |
+
"loss": 0.5002,
|
| 1537 |
+
"step": 216
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 1.2906933629959214,
|
| 1541 |
+
"grad_norm": 0.3815632104177572,
|
| 1542 |
+
"learning_rate": 5.5771130978099896e-06,
|
| 1543 |
+
"loss": 0.4984,
|
| 1544 |
+
"step": 217
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 1.296625880608083,
|
| 1548 |
+
"grad_norm": 0.43125912042578773,
|
| 1549 |
+
"learning_rate": 5.493450242000546e-06,
|
| 1550 |
+
"loss": 0.512,
|
| 1551 |
+
"step": 218
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 1.3025583982202447,
|
| 1555 |
+
"grad_norm": 0.36576046723610733,
|
| 1556 |
+
"learning_rate": 5.410181355324622e-06,
|
| 1557 |
+
"loss": 0.5028,
|
| 1558 |
+
"step": 219
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 1.3084909158324063,
|
| 1562 |
+
"grad_norm": 0.41196608301028825,
|
| 1563 |
+
"learning_rate": 5.32731371726938e-06,
|
| 1564 |
+
"loss": 0.5153,
|
| 1565 |
+
"step": 220
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 1.314423433444568,
|
| 1569 |
+
"grad_norm": 0.4106672248690276,
|
| 1570 |
+
"learning_rate": 5.244854572244249e-06,
|
| 1571 |
+
"loss": 0.4975,
|
| 1572 |
+
"step": 221
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 1.3203559510567298,
|
| 1576 |
+
"grad_norm": 0.407121465820589,
|
| 1577 |
+
"learning_rate": 5.1628111289476025e-06,
|
| 1578 |
+
"loss": 0.526,
|
| 1579 |
+
"step": 222
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 1.3262884686688914,
|
| 1583 |
+
"grad_norm": 0.36895605030103884,
|
| 1584 |
+
"learning_rate": 5.081190559736569e-06,
|
| 1585 |
+
"loss": 0.4965,
|
| 1586 |
+
"step": 223
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 1.332220986281053,
|
| 1590 |
+
"grad_norm": 0.39832627810457305,
|
| 1591 |
+
"learning_rate": 5.000000000000003e-06,
|
| 1592 |
+
"loss": 0.5171,
|
| 1593 |
+
"step": 224
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 1.3381535038932146,
|
| 1597 |
+
"grad_norm": 0.4133707378550398,
|
| 1598 |
+
"learning_rate": 4.919246547534709e-06,
|
| 1599 |
+
"loss": 0.4996,
|
| 1600 |
+
"step": 225
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 1.3440860215053765,
|
| 1604 |
+
"grad_norm": 0.35880827901259854,
|
| 1605 |
+
"learning_rate": 4.838937261924933e-06,
|
| 1606 |
+
"loss": 0.5026,
|
| 1607 |
+
"step": 226
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 1.350018539117538,
|
| 1611 |
+
"grad_norm": 0.4229197427258153,
|
| 1612 |
+
"learning_rate": 4.759079163925223e-06,
|
| 1613 |
+
"loss": 0.5132,
|
| 1614 |
+
"step": 227
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 1.3559510567296997,
|
| 1618 |
+
"grad_norm": 0.4347343479516533,
|
| 1619 |
+
"learning_rate": 4.679679234846636e-06,
|
| 1620 |
+
"loss": 0.5124,
|
| 1621 |
+
"step": 228
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 1.3618835743418614,
|
| 1625 |
+
"grad_norm": 0.3927261283383854,
|
| 1626 |
+
"learning_rate": 4.600744415946438e-06,
|
| 1627 |
+
"loss": 0.4997,
|
| 1628 |
+
"step": 229
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 1.367816091954023,
|
| 1632 |
+
"grad_norm": 0.39262400622403115,
|
| 1633 |
+
"learning_rate": 4.522281607821288e-06,
|
| 1634 |
+
"loss": 0.4967,
|
| 1635 |
+
"step": 230
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 1.3737486095661846,
|
| 1639 |
+
"grad_norm": 0.4223266363182741,
|
| 1640 |
+
"learning_rate": 4.444297669803981e-06,
|
| 1641 |
+
"loss": 0.4837,
|
| 1642 |
+
"step": 231
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 1.3796811271783462,
|
| 1646 |
+
"grad_norm": 0.4490330791545468,
|
| 1647 |
+
"learning_rate": 4.3667994193637794e-06,
|
| 1648 |
+
"loss": 0.5067,
|
| 1649 |
+
"step": 232
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 1.3856136447905079,
|
| 1653 |
+
"grad_norm": 0.39266902412489757,
|
| 1654 |
+
"learning_rate": 4.289793631510449e-06,
|
| 1655 |
+
"loss": 0.5253,
|
| 1656 |
+
"step": 233
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 1.3915461624026697,
|
| 1660 |
+
"grad_norm": 0.4576356901813089,
|
| 1661 |
+
"learning_rate": 4.213287038201943e-06,
|
| 1662 |
+
"loss": 0.5094,
|
| 1663 |
+
"step": 234
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 1.3974786800148313,
|
| 1667 |
+
"grad_norm": 0.43913157401185643,
|
| 1668 |
+
"learning_rate": 4.137286327755913e-06,
|
| 1669 |
+
"loss": 0.5074,
|
| 1670 |
+
"step": 235
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 1.403411197626993,
|
| 1674 |
+
"grad_norm": 0.3800958684968872,
|
| 1675 |
+
"learning_rate": 4.061798144264986e-06,
|
| 1676 |
+
"loss": 0.5161,
|
| 1677 |
+
"step": 236
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 1.4093437152391546,
|
| 1681 |
+
"grad_norm": 0.3966235644138789,
|
| 1682 |
+
"learning_rate": 3.986829087015941e-06,
|
| 1683 |
+
"loss": 0.5135,
|
| 1684 |
+
"step": 237
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 1.4152762328513162,
|
| 1688 |
+
"grad_norm": 0.39361049332745907,
|
| 1689 |
+
"learning_rate": 3.912385709912794e-06,
|
| 1690 |
+
"loss": 0.5043,
|
| 1691 |
+
"step": 238
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 1.421208750463478,
|
| 1695 |
+
"grad_norm": 0.42148715667996,
|
| 1696 |
+
"learning_rate": 3.838474520903825e-06,
|
| 1697 |
+
"loss": 0.4986,
|
| 1698 |
+
"step": 239
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 1.4271412680756397,
|
| 1702 |
+
"grad_norm": 0.39414223898095996,
|
| 1703 |
+
"learning_rate": 3.7651019814126656e-06,
|
| 1704 |
+
"loss": 0.5216,
|
| 1705 |
+
"step": 240
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 1.4330737856878013,
|
| 1709 |
+
"grad_norm": 0.37253637577722365,
|
| 1710 |
+
"learning_rate": 3.692274505773419e-06,
|
| 1711 |
+
"loss": 0.5253,
|
| 1712 |
+
"step": 241
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 1.439006303299963,
|
| 1716 |
+
"grad_norm": 0.41704046440564047,
|
| 1717 |
+
"learning_rate": 3.619998460669916e-06,
|
| 1718 |
+
"loss": 0.5364,
|
| 1719 |
+
"step": 242
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 1.4449388209121246,
|
| 1723 |
+
"grad_norm": 0.3653364189332755,
|
| 1724 |
+
"learning_rate": 3.5482801645791266e-06,
|
| 1725 |
+
"loss": 0.4916,
|
| 1726 |
+
"step": 243
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 1.4508713385242862,
|
| 1730 |
+
"grad_norm": 0.35349867698633664,
|
| 1731 |
+
"learning_rate": 3.4771258872187917e-06,
|
| 1732 |
+
"loss": 0.5275,
|
| 1733 |
+
"step": 244
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 1.4568038561364478,
|
| 1737 |
+
"grad_norm": 0.41828803136166764,
|
| 1738 |
+
"learning_rate": 3.4065418489993118e-06,
|
| 1739 |
+
"loss": 0.5139,
|
| 1740 |
+
"step": 245
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 1.4627363737486094,
|
| 1744 |
+
"grad_norm": 0.35821753699509107,
|
| 1745 |
+
"learning_rate": 3.3365342204799613e-06,
|
| 1746 |
+
"loss": 0.4831,
|
| 1747 |
+
"step": 246
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 1.4686688913607713,
|
| 1751 |
+
"grad_norm": 0.38538972239171304,
|
| 1752 |
+
"learning_rate": 3.267109121829428e-06,
|
| 1753 |
+
"loss": 0.5177,
|
| 1754 |
+
"step": 247
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 1.474601408972933,
|
| 1758 |
+
"grad_norm": 0.36434237943436254,
|
| 1759 |
+
"learning_rate": 3.1982726222908046e-06,
|
| 1760 |
+
"loss": 0.5134,
|
| 1761 |
+
"step": 248
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 1.4805339265850945,
|
| 1765 |
+
"grad_norm": 0.3706860313727835,
|
| 1766 |
+
"learning_rate": 3.1300307396509833e-06,
|
| 1767 |
+
"loss": 0.508,
|
| 1768 |
+
"step": 249
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 1.4864664441972562,
|
| 1772 |
+
"grad_norm": 0.38985446887032604,
|
| 1773 |
+
"learning_rate": 3.0623894397145837e-06,
|
| 1774 |
+
"loss": 0.5026,
|
| 1775 |
+
"step": 250
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 1.492398961809418,
|
| 1779 |
+
"grad_norm": 0.3921469360946988,
|
| 1780 |
+
"learning_rate": 2.995354635782417e-06,
|
| 1781 |
+
"loss": 0.5165,
|
| 1782 |
+
"step": 251
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 1.4983314794215796,
|
| 1786 |
+
"grad_norm": 0.35486752662951043,
|
| 1787 |
+
"learning_rate": 2.9289321881345257e-06,
|
| 1788 |
+
"loss": 0.5148,
|
| 1789 |
+
"step": 252
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 1.5042639970337413,
|
| 1793 |
+
"grad_norm": 0.3454065133594964,
|
| 1794 |
+
"learning_rate": 2.8631279035178796e-06,
|
| 1795 |
+
"loss": 0.4874,
|
| 1796 |
+
"step": 253
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 1.5101965146459029,
|
| 1800 |
+
"grad_norm": 0.3930097863801726,
|
| 1801 |
+
"learning_rate": 2.7979475346387363e-06,
|
| 1802 |
+
"loss": 0.4974,
|
| 1803 |
+
"step": 254
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 1.5161290322580645,
|
| 1807 |
+
"grad_norm": 0.3932165819653655,
|
| 1808 |
+
"learning_rate": 2.7333967796597317e-06,
|
| 1809 |
+
"loss": 0.5134,
|
| 1810 |
+
"step": 255
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 1.5220615498702261,
|
| 1814 |
+
"grad_norm": 0.3320234783166823,
|
| 1815 |
+
"learning_rate": 2.669481281701739e-06,
|
| 1816 |
+
"loss": 0.5121,
|
| 1817 |
+
"step": 256
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 1.5279940674823878,
|
| 1821 |
+
"grad_norm": 0.3535467409588918,
|
| 1822 |
+
"learning_rate": 2.6062066283505404e-06,
|
| 1823 |
+
"loss": 0.5111,
|
| 1824 |
+
"step": 257
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 1.5339265850945494,
|
| 1828 |
+
"grad_norm": 0.3770042051140451,
|
| 1829 |
+
"learning_rate": 2.5435783511683444e-06,
|
| 1830 |
+
"loss": 0.4873,
|
| 1831 |
+
"step": 258
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 1.539859102706711,
|
| 1835 |
+
"grad_norm": 0.367146329985044,
|
| 1836 |
+
"learning_rate": 2.4816019252102274e-06,
|
| 1837 |
+
"loss": 0.5083,
|
| 1838 |
+
"step": 259
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 1.5457916203188728,
|
| 1842 |
+
"grad_norm": 0.3394325003492519,
|
| 1843 |
+
"learning_rate": 2.420282768545469e-06,
|
| 1844 |
+
"loss": 0.4824,
|
| 1845 |
+
"step": 260
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 1.5517241379310345,
|
| 1849 |
+
"grad_norm": 0.3866827025758643,
|
| 1850 |
+
"learning_rate": 2.3596262417839256e-06,
|
| 1851 |
+
"loss": 0.507,
|
| 1852 |
+
"step": 261
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 1.557656655543196,
|
| 1856 |
+
"grad_norm": 0.33725647495532407,
|
| 1857 |
+
"learning_rate": 2.2996376476073724e-06,
|
| 1858 |
+
"loss": 0.4927,
|
| 1859 |
+
"step": 262
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 1.563589173155358,
|
| 1863 |
+
"grad_norm": 0.34863076052644143,
|
| 1864 |
+
"learning_rate": 2.240322230305951e-06,
|
| 1865 |
+
"loss": 0.4967,
|
| 1866 |
+
"step": 263
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 1.5695216907675196,
|
| 1870 |
+
"grad_norm": 0.3588480566227521,
|
| 1871 |
+
"learning_rate": 2.1816851753197023e-06,
|
| 1872 |
+
"loss": 0.4969,
|
| 1873 |
+
"step": 264
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 1.5754542083796812,
|
| 1877 |
+
"grad_norm": 0.339836588579076,
|
| 1878 |
+
"learning_rate": 2.1237316087852465e-06,
|
| 1879 |
+
"loss": 0.5104,
|
| 1880 |
+
"step": 265
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 1.5813867259918428,
|
| 1884 |
+
"grad_norm": 0.3549261249693355,
|
| 1885 |
+
"learning_rate": 2.0664665970876496e-06,
|
| 1886 |
+
"loss": 0.5063,
|
| 1887 |
+
"step": 266
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 1.5873192436040044,
|
| 1891 |
+
"grad_norm": 0.3776247577793457,
|
| 1892 |
+
"learning_rate": 2.009895146417512e-06,
|
| 1893 |
+
"loss": 0.5007,
|
| 1894 |
+
"step": 267
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 1.593251761216166,
|
| 1898 |
+
"grad_norm": 0.34878518596843994,
|
| 1899 |
+
"learning_rate": 1.9540222023333165e-06,
|
| 1900 |
+
"loss": 0.477,
|
| 1901 |
+
"step": 268
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 1.5991842788283277,
|
| 1905 |
+
"grad_norm": 0.3347286639188288,
|
| 1906 |
+
"learning_rate": 1.8988526493290948e-06,
|
| 1907 |
+
"loss": 0.4978,
|
| 1908 |
+
"step": 269
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 1.6051167964404893,
|
| 1912 |
+
"grad_norm": 0.3697740517807442,
|
| 1913 |
+
"learning_rate": 1.8443913104073984e-06,
|
| 1914 |
+
"loss": 0.5155,
|
| 1915 |
+
"step": 270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 1.611049314052651,
|
| 1919 |
+
"grad_norm": 0.34233058020715335,
|
| 1920 |
+
"learning_rate": 1.7906429466576768e-06,
|
| 1921 |
+
"loss": 0.5012,
|
| 1922 |
+
"step": 271
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 1.6169818316648128,
|
| 1926 |
+
"grad_norm": 0.3415900695134301,
|
| 1927 |
+
"learning_rate": 1.7376122568400533e-06,
|
| 1928 |
+
"loss": 0.4992,
|
| 1929 |
+
"step": 272
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 1.6229143492769744,
|
| 1933 |
+
"grad_norm": 0.3415274161298902,
|
| 1934 |
+
"learning_rate": 1.6853038769745466e-06,
|
| 1935 |
+
"loss": 0.4859,
|
| 1936 |
+
"step": 273
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 1.628846866889136,
|
| 1940 |
+
"grad_norm": 0.36778032976021363,
|
| 1941 |
+
"learning_rate": 1.6337223799358025e-06,
|
| 1942 |
+
"loss": 0.4846,
|
| 1943 |
+
"step": 274
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 1.634779384501298,
|
| 1947 |
+
"grad_norm": 0.38185068578401754,
|
| 1948 |
+
"learning_rate": 1.582872275053301e-06,
|
| 1949 |
+
"loss": 0.5154,
|
| 1950 |
+
"step": 275
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 1.6407119021134595,
|
| 1954 |
+
"grad_norm": 0.33140851685955686,
|
| 1955 |
+
"learning_rate": 1.5327580077171589e-06,
|
| 1956 |
+
"loss": 0.5011,
|
| 1957 |
+
"step": 276
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 1.6466444197256211,
|
| 1961 |
+
"grad_norm": 0.36885492758442245,
|
| 1962 |
+
"learning_rate": 1.4833839589895072e-06,
|
| 1963 |
+
"loss": 0.5092,
|
| 1964 |
+
"step": 277
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 1.6525769373377828,
|
| 1968 |
+
"grad_norm": 0.36549072606819394,
|
| 1969 |
+
"learning_rate": 1.4347544452214869e-06,
|
| 1970 |
+
"loss": 0.5252,
|
| 1971 |
+
"step": 278
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 1.6585094549499444,
|
| 1975 |
+
"grad_norm": 0.36738791918152014,
|
| 1976 |
+
"learning_rate": 1.3868737176759105e-06,
|
| 1977 |
+
"loss": 0.5281,
|
| 1978 |
+
"step": 279
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 1.664441972562106,
|
| 1982 |
+
"grad_norm": 0.326412556386957,
|
| 1983 |
+
"learning_rate": 1.339745962155613e-06,
|
| 1984 |
+
"loss": 0.5045,
|
| 1985 |
+
"step": 280
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 1.6703744901742676,
|
| 1989 |
+
"grad_norm": 0.3415508112126587,
|
| 1990 |
+
"learning_rate": 1.293375298637518e-06,
|
| 1991 |
+
"loss": 0.4895,
|
| 1992 |
+
"step": 281
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 1.6763070077864293,
|
| 1996 |
+
"grad_norm": 0.34722556354132184,
|
| 1997 |
+
"learning_rate": 1.2477657809124632e-06,
|
| 1998 |
+
"loss": 0.4946,
|
| 1999 |
+
"step": 282
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 1.682239525398591,
|
| 2003 |
+
"grad_norm": 0.3310077749742758,
|
| 2004 |
+
"learning_rate": 1.2029213962308172e-06,
|
| 2005 |
+
"loss": 0.5023,
|
| 2006 |
+
"step": 283
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 1.6881720430107527,
|
| 2010 |
+
"grad_norm": 0.3348390877359583,
|
| 2011 |
+
"learning_rate": 1.1588460649539036e-06,
|
| 2012 |
+
"loss": 0.4781,
|
| 2013 |
+
"step": 284
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 1.6941045606229144,
|
| 2017 |
+
"grad_norm": 0.34619840766115634,
|
| 2018 |
+
"learning_rate": 1.1155436402112785e-06,
|
| 2019 |
+
"loss": 0.5277,
|
| 2020 |
+
"step": 285
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 1.700037078235076,
|
| 2024 |
+
"grad_norm": 0.33543629443771605,
|
| 2025 |
+
"learning_rate": 1.073017907563887e-06,
|
| 2026 |
+
"loss": 0.5147,
|
| 2027 |
+
"step": 286
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 1.7059695958472376,
|
| 2031 |
+
"grad_norm": 0.34460699457893246,
|
| 2032 |
+
"learning_rate": 1.0312725846731174e-06,
|
| 2033 |
+
"loss": 0.5109,
|
| 2034 |
+
"step": 287
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 1.7119021134593995,
|
| 2038 |
+
"grad_norm": 0.363368163191836,
|
| 2039 |
+
"learning_rate": 9.903113209758098e-07,
|
| 2040 |
+
"loss": 0.4809,
|
| 2041 |
+
"step": 288
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 1.717834631071561,
|
| 2045 |
+
"grad_norm": 0.34297466357856504,
|
| 2046 |
+
"learning_rate": 9.501376973651999e-07,
|
| 2047 |
+
"loss": 0.495,
|
| 2048 |
+
"step": 289
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 1.7237671486837227,
|
| 2052 |
+
"grad_norm": 0.35425940666511757,
|
| 2053 |
+
"learning_rate": 9.107552258778907e-07,
|
| 2054 |
+
"loss": 0.5322,
|
| 2055 |
+
"step": 290
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 1.7296996662958843,
|
| 2059 |
+
"grad_norm": 0.37065578497405705,
|
| 2060 |
+
"learning_rate": 8.721673493868111e-07,
|
| 2061 |
+
"loss": 0.5168,
|
| 2062 |
+
"step": 291
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 1.735632183908046,
|
| 2066 |
+
"grad_norm": 0.34680893408384794,
|
| 2067 |
+
"learning_rate": 8.343774413002382e-07,
|
| 2068 |
+
"loss": 0.5024,
|
| 2069 |
+
"step": 292
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 1.7415647015202076,
|
| 2073 |
+
"grad_norm": 0.34059992527417754,
|
| 2074 |
+
"learning_rate": 7.973888052668943e-07,
|
| 2075 |
+
"loss": 0.5185,
|
| 2076 |
+
"step": 293
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 1.7474972191323692,
|
| 2080 |
+
"grad_norm": 0.35068499752190946,
|
| 2081 |
+
"learning_rate": 7.612046748871327e-07,
|
| 2082 |
+
"loss": 0.5142,
|
| 2083 |
+
"step": 294
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 1.7534297367445308,
|
| 2087 |
+
"grad_norm": 0.3518003616621483,
|
| 2088 |
+
"learning_rate": 7.258282134302519e-07,
|
| 2089 |
+
"loss": 0.5055,
|
| 2090 |
+
"step": 295
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 1.7593622543566925,
|
| 2094 |
+
"grad_norm": 0.343801973358475,
|
| 2095 |
+
"learning_rate": 6.912625135579587e-07,
|
| 2096 |
+
"loss": 0.4978,
|
| 2097 |
+
"step": 296
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 1.7652947719688543,
|
| 2101 |
+
"grad_norm": 0.3568376912330781,
|
| 2102 |
+
"learning_rate": 6.57510597054003e-07,
|
| 2103 |
+
"loss": 0.5359,
|
| 2104 |
+
"step": 297
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 1.771227289581016,
|
| 2108 |
+
"grad_norm": 0.33791018323113514,
|
| 2109 |
+
"learning_rate": 6.245754145600091e-07,
|
| 2110 |
+
"loss": 0.5025,
|
| 2111 |
+
"step": 298
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 1.7771598071931776,
|
| 2115 |
+
"grad_norm": 0.33742679787508273,
|
| 2116 |
+
"learning_rate": 5.924598453175278e-07,
|
| 2117 |
+
"loss": 0.4985,
|
| 2118 |
+
"step": 299
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 1.7830923248053394,
|
| 2122 |
+
"grad_norm": 0.33858341861628327,
|
| 2123 |
+
"learning_rate": 5.611666969163243e-07,
|
| 2124 |
+
"loss": 0.513,
|
| 2125 |
+
"step": 300
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 1.7830923248053394,
|
| 2129 |
+
"eval_loss": 0.6072779297828674,
|
| 2130 |
+
"eval_runtime": 6.923,
|
| 2131 |
+
"eval_samples_per_second": 18.345,
|
| 2132 |
+
"eval_steps_per_second": 2.311,
|
| 2133 |
+
"step": 300
|
| 2134 |
+
},
|
| 2135 |
+
{
|
| 2136 |
+
"epoch": 1.789024842417501,
|
| 2137 |
+
"grad_norm": 0.35027770808668635,
|
| 2138 |
+
"learning_rate": 5.306987050489442e-07,
|
| 2139 |
+
"loss": 0.5016,
|
| 2140 |
+
"step": 301
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 1.7949573600296627,
|
| 2144 |
+
"grad_norm": 0.33886481172097505,
|
| 2145 |
+
"learning_rate": 5.010585332715401e-07,
|
| 2146 |
+
"loss": 0.5075,
|
| 2147 |
+
"step": 302
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 1.8008898776418243,
|
| 2151 |
+
"grad_norm": 0.361905746068308,
|
| 2152 |
+
"learning_rate": 4.7224877277103673e-07,
|
| 2153 |
+
"loss": 0.5092,
|
| 2154 |
+
"step": 303
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 1.806822395253986,
|
| 2158 |
+
"grad_norm": 0.3347249716755872,
|
| 2159 |
+
"learning_rate": 4.4427194213859216e-07,
|
| 2160 |
+
"loss": 0.5221,
|
| 2161 |
+
"step": 304
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 1.8127549128661475,
|
| 2165 |
+
"grad_norm": 0.3384027231629407,
|
| 2166 |
+
"learning_rate": 4.171304871494264e-07,
|
| 2167 |
+
"loss": 0.5164,
|
| 2168 |
+
"step": 305
|
| 2169 |
+
},
|
| 2170 |
+
{
|
| 2171 |
+
"epoch": 1.8186874304783092,
|
| 2172 |
+
"grad_norm": 0.3304521483560128,
|
| 2173 |
+
"learning_rate": 3.908267805490051e-07,
|
| 2174 |
+
"loss": 0.51,
|
| 2175 |
+
"step": 306
|
| 2176 |
+
},
|
| 2177 |
+
{
|
| 2178 |
+
"epoch": 1.8246199480904708,
|
| 2179 |
+
"grad_norm": 0.32865263259804217,
|
| 2180 |
+
"learning_rate": 3.6536312184560996e-07,
|
| 2181 |
+
"loss": 0.5185,
|
| 2182 |
+
"step": 307
|
| 2183 |
+
},
|
| 2184 |
+
{
|
| 2185 |
+
"epoch": 1.8305524657026324,
|
| 2186 |
+
"grad_norm": 0.3193148808383708,
|
| 2187 |
+
"learning_rate": 3.4074173710931804e-07,
|
| 2188 |
+
"loss": 0.5031,
|
| 2189 |
+
"step": 308
|
| 2190 |
+
},
|
| 2191 |
+
{
|
| 2192 |
+
"epoch": 1.8364849833147943,
|
| 2193 |
+
"grad_norm": 0.32050879286411965,
|
| 2194 |
+
"learning_rate": 3.1696477877738664e-07,
|
| 2195 |
+
"loss": 0.5087,
|
| 2196 |
+
"step": 309
|
| 2197 |
+
},
|
| 2198 |
+
{
|
| 2199 |
+
"epoch": 1.8424175009269559,
|
| 2200 |
+
"grad_norm": 0.3442859949879679,
|
| 2201 |
+
"learning_rate": 2.940343254660905e-07,
|
| 2202 |
+
"loss": 0.517,
|
| 2203 |
+
"step": 310
|
| 2204 |
+
},
|
| 2205 |
+
{
|
| 2206 |
+
"epoch": 1.8483500185391175,
|
| 2207 |
+
"grad_norm": 0.3273773858356578,
|
| 2208 |
+
"learning_rate": 2.7195238178900685e-07,
|
| 2209 |
+
"loss": 0.4948,
|
| 2210 |
+
"step": 311
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 1.8542825361512794,
|
| 2214 |
+
"grad_norm": 0.33321328950736606,
|
| 2215 |
+
"learning_rate": 2.507208781817638e-07,
|
| 2216 |
+
"loss": 0.5236,
|
| 2217 |
+
"step": 312
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 1.860215053763441,
|
| 2221 |
+
"grad_norm": 0.3386415168984448,
|
| 2222 |
+
"learning_rate": 2.3034167073328283e-07,
|
| 2223 |
+
"loss": 0.5116,
|
| 2224 |
+
"step": 313
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 1.8661475713756026,
|
| 2228 |
+
"grad_norm": 0.32943898290159496,
|
| 2229 |
+
"learning_rate": 2.1081654102351634e-07,
|
| 2230 |
+
"loss": 0.4981,
|
| 2231 |
+
"step": 314
|
| 2232 |
+
},
|
| 2233 |
+
{
|
| 2234 |
+
"epoch": 1.8720800889877642,
|
| 2235 |
+
"grad_norm": 0.31450631617292496,
|
| 2236 |
+
"learning_rate": 1.921471959676957e-07,
|
| 2237 |
+
"loss": 0.4972,
|
| 2238 |
+
"step": 315
|
| 2239 |
+
},
|
| 2240 |
+
{
|
| 2241 |
+
"epoch": 1.8780126065999259,
|
| 2242 |
+
"grad_norm": 0.33778423189464707,
|
| 2243 |
+
"learning_rate": 1.7433526766711727e-07,
|
| 2244 |
+
"loss": 0.5025,
|
| 2245 |
+
"step": 316
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 1.8839451242120875,
|
| 2249 |
+
"grad_norm": 0.32927984851113423,
|
| 2250 |
+
"learning_rate": 1.5738231326645758e-07,
|
| 2251 |
+
"loss": 0.5154,
|
| 2252 |
+
"step": 317
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 1.889877641824249,
|
| 2256 |
+
"grad_norm": 0.33696222369048556,
|
| 2257 |
+
"learning_rate": 1.4128981481764115e-07,
|
| 2258 |
+
"loss": 0.4804,
|
| 2259 |
+
"step": 318
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 1.8958101594364107,
|
| 2263 |
+
"grad_norm": 0.33119042978822105,
|
| 2264 |
+
"learning_rate": 1.2605917915028743e-07,
|
| 2265 |
+
"loss": 0.5099,
|
| 2266 |
+
"step": 319
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 1.9017426770485724,
|
| 2270 |
+
"grad_norm": 0.3369916299734094,
|
| 2271 |
+
"learning_rate": 1.1169173774871478e-07,
|
| 2272 |
+
"loss": 0.5009,
|
| 2273 |
+
"step": 320
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 1.907675194660734,
|
| 2277 |
+
"grad_norm": 0.32908222754161875,
|
| 2278 |
+
"learning_rate": 9.818874663554356e-08,
|
| 2279 |
+
"loss": 0.5085,
|
| 2280 |
+
"step": 321
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 1.9136077122728958,
|
| 2284 |
+
"grad_norm": 0.3305867681142081,
|
| 2285 |
+
"learning_rate": 8.555138626189619e-08,
|
| 2286 |
+
"loss": 0.4843,
|
| 2287 |
+
"step": 322
|
| 2288 |
+
},
|
| 2289 |
+
{
|
| 2290 |
+
"epoch": 1.9195402298850575,
|
| 2291 |
+
"grad_norm": 0.338494816075747,
|
| 2292 |
+
"learning_rate": 7.378076140419188e-08,
|
| 2293 |
+
"loss": 0.5247,
|
| 2294 |
+
"step": 323
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 1.925472747497219,
|
| 2298 |
+
"grad_norm": 0.3464955361533102,
|
| 2299 |
+
"learning_rate": 6.287790106757396e-08,
|
| 2300 |
+
"loss": 0.4982,
|
| 2301 |
+
"step": 324
|
| 2302 |
+
},
|
| 2303 |
+
{
|
| 2304 |
+
"epoch": 1.931405265109381,
|
| 2305 |
+
"grad_norm": 0.32358285120905184,
|
| 2306 |
+
"learning_rate": 5.284375839594958e-08,
|
| 2307 |
+
"loss": 0.5097,
|
| 2308 |
+
"step": 325
|
| 2309 |
+
},
|
| 2310 |
+
{
|
| 2311 |
+
"epoch": 1.9373377827215426,
|
| 2312 |
+
"grad_norm": 0.32591557161362456,
|
| 2313 |
+
"learning_rate": 4.367921058866187e-08,
|
| 2314 |
+
"loss": 0.4852,
|
| 2315 |
+
"step": 326
|
| 2316 |
+
},
|
| 2317 |
+
{
|
| 2318 |
+
"epoch": 1.9432703003337042,
|
| 2319 |
+
"grad_norm": 0.3114294363115202,
|
| 2320 |
+
"learning_rate": 3.538505882380916e-08,
|
| 2321 |
+
"loss": 0.482,
|
| 2322 |
+
"step": 327
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 1.9492028179458658,
|
| 2326 |
+
"grad_norm": 0.3149973184867936,
|
| 2327 |
+
"learning_rate": 2.796202818819871e-08,
|
| 2328 |
+
"loss": 0.4844,
|
| 2329 |
+
"step": 328
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 1.9551353355580274,
|
| 2333 |
+
"grad_norm": 0.3452282568361082,
|
| 2334 |
+
"learning_rate": 2.1410767613965212e-08,
|
| 2335 |
+
"loss": 0.5069,
|
| 2336 |
+
"step": 329
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 1.961067853170189,
|
| 2340 |
+
"grad_norm": 0.347397146126861,
|
| 2341 |
+
"learning_rate": 1.5731849821833955e-08,
|
| 2342 |
+
"loss": 0.5167,
|
| 2343 |
+
"step": 330
|
| 2344 |
+
},
|
| 2345 |
+
{
|
| 2346 |
+
"epoch": 1.9670003707823507,
|
| 2347 |
+
"grad_norm": 0.32204047912115363,
|
| 2348 |
+
"learning_rate": 1.0925771271058649e-08,
|
| 2349 |
+
"loss": 0.4928,
|
| 2350 |
+
"step": 331
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 1.9729328883945123,
|
| 2354 |
+
"grad_norm": 0.3287263087544649,
|
| 2355 |
+
"learning_rate": 6.992952116013918e-09,
|
| 2356 |
+
"loss": 0.4838,
|
| 2357 |
+
"step": 332
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 1.978865406006674,
|
| 2361 |
+
"grad_norm": 0.3269223168333791,
|
| 2362 |
+
"learning_rate": 3.933736169471347e-09,
|
| 2363 |
+
"loss": 0.5086,
|
| 2364 |
+
"step": 333
|
| 2365 |
+
},
|
| 2366 |
+
{
|
| 2367 |
+
"epoch": 1.9847979236188358,
|
| 2368 |
+
"grad_norm": 0.3469969747818369,
|
| 2369 |
+
"learning_rate": 1.7483908725357546e-09,
|
| 2370 |
+
"loss": 0.5039,
|
| 2371 |
+
"step": 334
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 1.9907304412309974,
|
| 2375 |
+
"grad_norm": 0.3226354438244203,
|
| 2376 |
+
"learning_rate": 4.3710727127277417e-10,
|
| 2377 |
+
"loss": 0.5337,
|
| 2378 |
+
"step": 335
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"epoch": 1.996662958843159,
|
| 2382 |
+
"grad_norm": 0.3421484877624497,
|
| 2383 |
+
"learning_rate": 0.0,
|
| 2384 |
+
"loss": 0.5223,
|
| 2385 |
+
"step": 336
|
| 2386 |
+
},
|
| 2387 |
+
{
|
| 2388 |
+
"epoch": 1.996662958843159,
|
| 2389 |
+
"step": 336,
|
| 2390 |
+
"total_flos": 352364519227392.0,
|
| 2391 |
+
"train_loss": 0.5676420692886625,
|
| 2392 |
+
"train_runtime": 10418.0052,
|
| 2393 |
+
"train_samples_per_second": 4.141,
|
| 2394 |
+
"train_steps_per_second": 0.032
|
| 2395 |
+
}
|
| 2396 |
+
],
|
| 2397 |
+
"logging_steps": 1,
|
| 2398 |
+
"max_steps": 336,
|
| 2399 |
+
"num_input_tokens_seen": 0,
|
| 2400 |
+
"num_train_epochs": 2,
|
| 2401 |
+
"save_steps": 50,
|
| 2402 |
+
"stateful_callbacks": {
|
| 2403 |
+
"TrainerControl": {
|
| 2404 |
+
"args": {
|
| 2405 |
+
"should_epoch_stop": false,
|
| 2406 |
+
"should_evaluate": false,
|
| 2407 |
+
"should_log": false,
|
| 2408 |
+
"should_save": true,
|
| 2409 |
+
"should_training_stop": true
|
| 2410 |
+
},
|
| 2411 |
+
"attributes": {}
|
| 2412 |
+
}
|
| 2413 |
+
},
|
| 2414 |
+
"total_flos": 352364519227392.0,
|
| 2415 |
+
"train_batch_size": 1,
|
| 2416 |
+
"trial_name": null,
|
| 2417 |
+
"trial_params": null
|
| 2418 |
+
}
|