Model save
Browse files- README.md +58 -0
- all_results.json +8 -0
- generation_config.json +14 -0
- train_results.json +8 -0
- trainer_state.json +1442 -0
README.md
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen2.5-1.5B-Instruct
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Qwen2.5-1.5B-Open-R1-Distill
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Qwen2.5-1.5B-Open-R1-Distill
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct).
|
| 15 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
+
|
| 17 |
+
## Quick start
|
| 18 |
+
|
| 19 |
+
```python
|
| 20 |
+
from transformers import pipeline
|
| 21 |
+
|
| 22 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 23 |
+
generator = pipeline("text-generation", model="yjh00/Qwen2.5-1.5B-Open-R1-Distill", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/liangzizhangluo-hohai-university/huggingface/runs/4cxrwklk)
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with SFT.
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.16.0.dev0
|
| 38 |
+
- Transformers: 4.49.0.dev0
|
| 39 |
+
- Pytorch: 2.5.1
|
| 40 |
+
- Datasets: 3.3.0
|
| 41 |
+
- Tokenizers: 0.21.0
|
| 42 |
+
|
| 43 |
+
## Citations
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
Cite TRL as:
|
| 48 |
+
|
| 49 |
+
```bibtex
|
| 50 |
+
@misc{vonwerra2022trl,
|
| 51 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 52 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
|
| 53 |
+
year = 2020,
|
| 54 |
+
journal = {GitHub repository},
|
| 55 |
+
publisher = {GitHub},
|
| 56 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 57 |
+
}
|
| 58 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 28487712768000.0,
|
| 3 |
+
"train_loss": 0.920774197101593,
|
| 4 |
+
"train_runtime": 1759.2244,
|
| 5 |
+
"train_samples": 16610,
|
| 6 |
+
"train_samples_per_second": 18.19,
|
| 7 |
+
"train_steps_per_second": 0.568
|
| 8 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.49.0.dev0"
|
| 14 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 28487712768000.0,
|
| 3 |
+
"train_loss": 0.920774197101593,
|
| 4 |
+
"train_runtime": 1759.2244,
|
| 5 |
+
"train_samples": 16610,
|
| 6 |
+
"train_samples_per_second": 18.19,
|
| 7 |
+
"train_steps_per_second": 0.568
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1442 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.3699593044765076,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.001849796522382538,
|
| 13 |
+
"grad_norm": 3.3279617254334535,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.3195,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003699593044765076,
|
| 20 |
+
"grad_norm": 1.9430817685675177,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.2699,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005549389567147614,
|
| 27 |
+
"grad_norm": 2.1226078391974967,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 1.1007,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.007399186089530152,
|
| 34 |
+
"grad_norm": 1.864202847273668,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 1.0631,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.00924898261191269,
|
| 41 |
+
"grad_norm": 1.5790964667081508,
|
| 42 |
+
"learning_rate": 2.5e-05,
|
| 43 |
+
"loss": 1.0944,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.011098779134295227,
|
| 48 |
+
"grad_norm": 1.3616177321523328,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 0.9914,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.012948575656677765,
|
| 55 |
+
"grad_norm": 1.404161383781241,
|
| 56 |
+
"learning_rate": 3.5e-05,
|
| 57 |
+
"loss": 0.9922,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.014798372179060304,
|
| 62 |
+
"grad_norm": 1.4817289036292525,
|
| 63 |
+
"learning_rate": 4e-05,
|
| 64 |
+
"loss": 0.992,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.016648168701442843,
|
| 69 |
+
"grad_norm": 1.4532039393449283,
|
| 70 |
+
"learning_rate": 4.5e-05,
|
| 71 |
+
"loss": 0.9724,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.01849796522382538,
|
| 76 |
+
"grad_norm": 1.459454732453025,
|
| 77 |
+
"learning_rate": 5e-05,
|
| 78 |
+
"loss": 0.9584,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.020347761746207917,
|
| 83 |
+
"grad_norm": 1.3435610444029484,
|
| 84 |
+
"learning_rate": 4.9996924362330555e-05,
|
| 85 |
+
"loss": 0.9809,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.022197558268590455,
|
| 90 |
+
"grad_norm": 1.507556200878351,
|
| 91 |
+
"learning_rate": 4.998769829017084e-05,
|
| 92 |
+
"loss": 1.0205,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.024047354790972992,
|
| 97 |
+
"grad_norm": 1.74928123764404,
|
| 98 |
+
"learning_rate": 4.997232430583686e-05,
|
| 99 |
+
"loss": 1.0078,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.02589715131335553,
|
| 104 |
+
"grad_norm": 1.3343301345399712,
|
| 105 |
+
"learning_rate": 4.995080661242243e-05,
|
| 106 |
+
"loss": 0.9691,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.02774694783573807,
|
| 111 |
+
"grad_norm": 1.383358561734251,
|
| 112 |
+
"learning_rate": 4.992315109265007e-05,
|
| 113 |
+
"loss": 1.0215,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.029596744358120607,
|
| 118 |
+
"grad_norm": 1.220791806067286,
|
| 119 |
+
"learning_rate": 4.988936530726276e-05,
|
| 120 |
+
"loss": 0.9797,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.031446540880503145,
|
| 125 |
+
"grad_norm": 1.7288941628239176,
|
| 126 |
+
"learning_rate": 4.984945849295686e-05,
|
| 127 |
+
"loss": 0.9667,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.033296337402885685,
|
| 132 |
+
"grad_norm": 1.3374822253676133,
|
| 133 |
+
"learning_rate": 4.980344155985695e-05,
|
| 134 |
+
"loss": 0.9725,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.03514613392526822,
|
| 139 |
+
"grad_norm": 1.258960837768452,
|
| 140 |
+
"learning_rate": 4.975132708853304e-05,
|
| 141 |
+
"loss": 0.9877,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.03699593044765076,
|
| 146 |
+
"grad_norm": 1.438691891865278,
|
| 147 |
+
"learning_rate": 4.9693129326561254e-05,
|
| 148 |
+
"loss": 1.0251,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.038845726970033294,
|
| 153 |
+
"grad_norm": 1.3123491113835215,
|
| 154 |
+
"learning_rate": 4.96288641846286e-05,
|
| 155 |
+
"loss": 0.9805,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.040695523492415835,
|
| 160 |
+
"grad_norm": 1.3868311724825364,
|
| 161 |
+
"learning_rate": 4.955854923218321e-05,
|
| 162 |
+
"loss": 0.9866,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.042545320014798375,
|
| 167 |
+
"grad_norm": 1.1867492456163233,
|
| 168 |
+
"learning_rate": 4.948220369263099e-05,
|
| 169 |
+
"loss": 0.9671,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.04439511653718091,
|
| 174 |
+
"grad_norm": 1.3516961645638472,
|
| 175 |
+
"learning_rate": 4.939984843808013e-05,
|
| 176 |
+
"loss": 0.9712,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.04624491305956345,
|
| 181 |
+
"grad_norm": 1.3818877123258775,
|
| 182 |
+
"learning_rate": 4.931150598363494e-05,
|
| 183 |
+
"loss": 0.9973,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.048094709581945984,
|
| 188 |
+
"grad_norm": 1.2461243722970794,
|
| 189 |
+
"learning_rate": 4.921720048124035e-05,
|
| 190 |
+
"loss": 0.9721,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.049944506104328525,
|
| 195 |
+
"grad_norm": 1.3531497904983025,
|
| 196 |
+
"learning_rate": 4.9116957713079084e-05,
|
| 197 |
+
"loss": 0.9947,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.05179430262671106,
|
| 202 |
+
"grad_norm": 1.3909365555868392,
|
| 203 |
+
"learning_rate": 4.901080508452305e-05,
|
| 204 |
+
"loss": 0.9886,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.0536440991490936,
|
| 209 |
+
"grad_norm": 1.3632639733726144,
|
| 210 |
+
"learning_rate": 4.889877161664096e-05,
|
| 211 |
+
"loss": 0.9856,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.05549389567147614,
|
| 216 |
+
"grad_norm": 1.376746573930021,
|
| 217 |
+
"learning_rate": 4.878088793826428e-05,
|
| 218 |
+
"loss": 1.0124,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.057343692193858674,
|
| 223 |
+
"grad_norm": 1.2051488416511116,
|
| 224 |
+
"learning_rate": 4.865718627761363e-05,
|
| 225 |
+
"loss": 0.9502,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.059193488716241215,
|
| 230 |
+
"grad_norm": 1.4444748706679968,
|
| 231 |
+
"learning_rate": 4.8527700453487873e-05,
|
| 232 |
+
"loss": 0.976,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.06104328523862375,
|
| 237 |
+
"grad_norm": 1.5080951955787747,
|
| 238 |
+
"learning_rate": 4.839246586601841e-05,
|
| 239 |
+
"loss": 0.9858,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.06289308176100629,
|
| 244 |
+
"grad_norm": 1.218432538840472,
|
| 245 |
+
"learning_rate": 4.825151948699116e-05,
|
| 246 |
+
"loss": 0.985,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.06474287828338883,
|
| 251 |
+
"grad_norm": 1.1799037097271943,
|
| 252 |
+
"learning_rate": 4.8104899849738795e-05,
|
| 253 |
+
"loss": 0.9378,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.06659267480577137,
|
| 258 |
+
"grad_norm": 1.2779121353402405,
|
| 259 |
+
"learning_rate": 4.795264703860616e-05,
|
| 260 |
+
"loss": 0.9683,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.0684424713281539,
|
| 265 |
+
"grad_norm": 1.4722942379840678,
|
| 266 |
+
"learning_rate": 4.779480267799158e-05,
|
| 267 |
+
"loss": 0.9904,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.07029226785053644,
|
| 272 |
+
"grad_norm": 1.3984538759925134,
|
| 273 |
+
"learning_rate": 4.763140992096718e-05,
|
| 274 |
+
"loss": 0.9702,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.07214206437291898,
|
| 279 |
+
"grad_norm": 1.2519064402447997,
|
| 280 |
+
"learning_rate": 4.7462513437481266e-05,
|
| 281 |
+
"loss": 0.9671,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.07399186089530152,
|
| 286 |
+
"grad_norm": 1.281644209465442,
|
| 287 |
+
"learning_rate": 4.7288159402146e-05,
|
| 288 |
+
"loss": 0.9653,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.07584165741768406,
|
| 293 |
+
"grad_norm": 1.3861942696753027,
|
| 294 |
+
"learning_rate": 4.7108395481613736e-05,
|
| 295 |
+
"loss": 0.989,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.07769145394006659,
|
| 300 |
+
"grad_norm": 1.2692148251690907,
|
| 301 |
+
"learning_rate": 4.692327082154542e-05,
|
| 302 |
+
"loss": 0.9695,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.07954125046244913,
|
| 307 |
+
"grad_norm": 1.2503784332879924,
|
| 308 |
+
"learning_rate": 4.6732836033174634e-05,
|
| 309 |
+
"loss": 0.9442,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.08139104698483167,
|
| 314 |
+
"grad_norm": 1.2071851181770694,
|
| 315 |
+
"learning_rate": 4.653714317947105e-05,
|
| 316 |
+
"loss": 0.9664,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.08324084350721421,
|
| 321 |
+
"grad_norm": 1.3973898482595015,
|
| 322 |
+
"learning_rate": 4.6336245760906896e-05,
|
| 323 |
+
"loss": 1.0037,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.08509064002959675,
|
| 328 |
+
"grad_norm": 1.1885748014842101,
|
| 329 |
+
"learning_rate": 4.6130198700830455e-05,
|
| 330 |
+
"loss": 0.9523,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.08694043655197928,
|
| 335 |
+
"grad_norm": 1.1676588738483402,
|
| 336 |
+
"learning_rate": 4.591905833045059e-05,
|
| 337 |
+
"loss": 0.9317,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.08879023307436182,
|
| 342 |
+
"grad_norm": 1.2482898560322035,
|
| 343 |
+
"learning_rate": 4.570288237343632e-05,
|
| 344 |
+
"loss": 0.9315,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.09064002959674436,
|
| 349 |
+
"grad_norm": 1.122744743080674,
|
| 350 |
+
"learning_rate": 4.5481729930135805e-05,
|
| 351 |
+
"loss": 0.9582,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.0924898261191269,
|
| 356 |
+
"grad_norm": 1.1964791201941039,
|
| 357 |
+
"learning_rate": 4.5255661461418854e-05,
|
| 358 |
+
"loss": 0.9613,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.09433962264150944,
|
| 363 |
+
"grad_norm": 1.313613853504735,
|
| 364 |
+
"learning_rate": 4.502473877214754e-05,
|
| 365 |
+
"loss": 0.9627,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.09618941916389197,
|
| 370 |
+
"grad_norm": 1.2998325386052085,
|
| 371 |
+
"learning_rate": 4.478902499427933e-05,
|
| 372 |
+
"loss": 0.9435,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.09803921568627451,
|
| 377 |
+
"grad_norm": 1.1188167692403637,
|
| 378 |
+
"learning_rate": 4.454858456960754e-05,
|
| 379 |
+
"loss": 0.9531,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.09988901220865705,
|
| 384 |
+
"grad_norm": 1.245499023500759,
|
| 385 |
+
"learning_rate": 4.430348323214347e-05,
|
| 386 |
+
"loss": 0.9615,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.10173880873103959,
|
| 391 |
+
"grad_norm": 1.2731826201754,
|
| 392 |
+
"learning_rate": 4.4053787990145465e-05,
|
| 393 |
+
"loss": 0.9852,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.10358860525342212,
|
| 398 |
+
"grad_norm": 1.0779984323652068,
|
| 399 |
+
"learning_rate": 4.379956710779951e-05,
|
| 400 |
+
"loss": 0.9649,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.10543840177580466,
|
| 405 |
+
"grad_norm": 1.3267240330921417,
|
| 406 |
+
"learning_rate": 4.3540890086556435e-05,
|
| 407 |
+
"loss": 0.967,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1072881982981872,
|
| 412 |
+
"grad_norm": 1.194869135121013,
|
| 413 |
+
"learning_rate": 4.327782764613099e-05,
|
| 414 |
+
"loss": 0.9745,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.10913799482056974,
|
| 419 |
+
"grad_norm": 1.092064798384306,
|
| 420 |
+
"learning_rate": 4.301045170516773e-05,
|
| 421 |
+
"loss": 0.9354,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.11098779134295228,
|
| 426 |
+
"grad_norm": 1.1564826438676266,
|
| 427 |
+
"learning_rate": 4.2738835361579175e-05,
|
| 428 |
+
"loss": 0.9301,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1128375878653348,
|
| 433 |
+
"grad_norm": 1.1034286954430845,
|
| 434 |
+
"learning_rate": 4.2463052872561584e-05,
|
| 435 |
+
"loss": 0.9536,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.11468738438771735,
|
| 440 |
+
"grad_norm": 1.1101601059988588,
|
| 441 |
+
"learning_rate": 4.2183179634293794e-05,
|
| 442 |
+
"loss": 0.9302,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.11653718091009989,
|
| 447 |
+
"grad_norm": 1.2757853825149674,
|
| 448 |
+
"learning_rate": 4.1899292161324627e-05,
|
| 449 |
+
"loss": 0.9547,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.11838697743248243,
|
| 454 |
+
"grad_norm": 1.1087176065384894,
|
| 455 |
+
"learning_rate": 4.1611468065654586e-05,
|
| 456 |
+
"loss": 0.9764,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.12023677395486497,
|
| 461 |
+
"grad_norm": 1.2270496050163748,
|
| 462 |
+
"learning_rate": 4.1319786035517534e-05,
|
| 463 |
+
"loss": 0.9551,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.1220865704772475,
|
| 468 |
+
"grad_norm": 1.2025590067136398,
|
| 469 |
+
"learning_rate": 4.1024325813868065e-05,
|
| 470 |
+
"loss": 0.9238,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.12393636699963004,
|
| 475 |
+
"grad_norm": 1.1631866711130052,
|
| 476 |
+
"learning_rate": 4.072516817658065e-05,
|
| 477 |
+
"loss": 0.9573,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.12578616352201258,
|
| 482 |
+
"grad_norm": 1.064949988724441,
|
| 483 |
+
"learning_rate": 4.0422394910366236e-05,
|
| 484 |
+
"loss": 0.9184,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.12763596004439512,
|
| 489 |
+
"grad_norm": 1.4077051494514279,
|
| 490 |
+
"learning_rate": 4.0116088790412645e-05,
|
| 491 |
+
"loss": 0.9518,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.12948575656677766,
|
| 496 |
+
"grad_norm": 1.2486218937453641,
|
| 497 |
+
"learning_rate": 3.980633355775461e-05,
|
| 498 |
+
"loss": 0.9518,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.1313355530891602,
|
| 503 |
+
"grad_norm": 1.2416341494672163,
|
| 504 |
+
"learning_rate": 3.949321389637986e-05,
|
| 505 |
+
"loss": 0.9238,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.13318534961154274,
|
| 510 |
+
"grad_norm": 1.18527481990195,
|
| 511 |
+
"learning_rate": 3.917681541007734e-05,
|
| 512 |
+
"loss": 0.9319,
|
| 513 |
+
"step": 360
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.13503514613392525,
|
| 517 |
+
"grad_norm": 1.1856076287802888,
|
| 518 |
+
"learning_rate": 3.885722459903399e-05,
|
| 519 |
+
"loss": 0.941,
|
| 520 |
+
"step": 365
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.1368849426563078,
|
| 524 |
+
"grad_norm": 1.3096185432838516,
|
| 525 |
+
"learning_rate": 3.853452883618644e-05,
|
| 526 |
+
"loss": 0.9531,
|
| 527 |
+
"step": 370
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.13873473917869034,
|
| 531 |
+
"grad_norm": 1.0667136184374695,
|
| 532 |
+
"learning_rate": 3.8208816343334156e-05,
|
| 533 |
+
"loss": 0.9648,
|
| 534 |
+
"step": 375
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.14058453570107288,
|
| 538 |
+
"grad_norm": 1.3556343012265513,
|
| 539 |
+
"learning_rate": 3.788017616702048e-05,
|
| 540 |
+
"loss": 0.9622,
|
| 541 |
+
"step": 380
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.14243433222345542,
|
| 545 |
+
"grad_norm": 1.2826346418353032,
|
| 546 |
+
"learning_rate": 3.754869815418815e-05,
|
| 547 |
+
"loss": 0.9243,
|
| 548 |
+
"step": 385
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.14428412874583796,
|
| 552 |
+
"grad_norm": 1.4908193033444779,
|
| 553 |
+
"learning_rate": 3.721447292761609e-05,
|
| 554 |
+
"loss": 0.9402,
|
| 555 |
+
"step": 390
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.1461339252682205,
|
| 559 |
+
"grad_norm": 1.2092388640077134,
|
| 560 |
+
"learning_rate": 3.687759186114403e-05,
|
| 561 |
+
"loss": 0.8881,
|
| 562 |
+
"step": 395
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.14798372179060304,
|
| 566 |
+
"grad_norm": 1.2289784753028945,
|
| 567 |
+
"learning_rate": 3.6538147054691817e-05,
|
| 568 |
+
"loss": 0.9171,
|
| 569 |
+
"step": 400
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.14983351831298558,
|
| 573 |
+
"grad_norm": 1.2686814293232442,
|
| 574 |
+
"learning_rate": 3.619623130908018e-05,
|
| 575 |
+
"loss": 0.9487,
|
| 576 |
+
"step": 405
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.15168331483536812,
|
| 580 |
+
"grad_norm": 1.0701718673112672,
|
| 581 |
+
"learning_rate": 3.5851938100659964e-05,
|
| 582 |
+
"loss": 0.9099,
|
| 583 |
+
"step": 410
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.15353311135775063,
|
| 587 |
+
"grad_norm": 1.090800531105658,
|
| 588 |
+
"learning_rate": 3.550536155575662e-05,
|
| 589 |
+
"loss": 0.8853,
|
| 590 |
+
"step": 415
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.15538290788013318,
|
| 594 |
+
"grad_norm": 1.2313388220061225,
|
| 595 |
+
"learning_rate": 3.515659642493697e-05,
|
| 596 |
+
"loss": 0.9385,
|
| 597 |
+
"step": 420
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.15723270440251572,
|
| 601 |
+
"grad_norm": 1.2536700830777894,
|
| 602 |
+
"learning_rate": 3.480573805710538e-05,
|
| 603 |
+
"loss": 0.9143,
|
| 604 |
+
"step": 425
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.15908250092489826,
|
| 608 |
+
"grad_norm": 1.1514860580661626,
|
| 609 |
+
"learning_rate": 3.4452882373436316e-05,
|
| 610 |
+
"loss": 0.8996,
|
| 611 |
+
"step": 430
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.1609322974472808,
|
| 615 |
+
"grad_norm": 1.3731197382175357,
|
| 616 |
+
"learning_rate": 3.4098125841150466e-05,
|
| 617 |
+
"loss": 0.9243,
|
| 618 |
+
"step": 435
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.16278209396966334,
|
| 622 |
+
"grad_norm": 1.0232827661891992,
|
| 623 |
+
"learning_rate": 3.37415654471415e-05,
|
| 624 |
+
"loss": 0.9369,
|
| 625 |
+
"step": 440
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.16463189049204588,
|
| 629 |
+
"grad_norm": 1.1305694231647365,
|
| 630 |
+
"learning_rate": 3.3383298671460944e-05,
|
| 631 |
+
"loss": 0.9142,
|
| 632 |
+
"step": 445
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.16648168701442842,
|
| 636 |
+
"grad_norm": 1.1398463628627697,
|
| 637 |
+
"learning_rate": 3.3023423460667985e-05,
|
| 638 |
+
"loss": 0.9094,
|
| 639 |
+
"step": 450
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.16833148353681096,
|
| 643 |
+
"grad_norm": 1.0590670030120253,
|
| 644 |
+
"learning_rate": 3.2662038201051914e-05,
|
| 645 |
+
"loss": 0.9013,
|
| 646 |
+
"step": 455
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.1701812800591935,
|
| 650 |
+
"grad_norm": 1.1536632313067792,
|
| 651 |
+
"learning_rate": 3.2299241691734304e-05,
|
| 652 |
+
"loss": 0.9233,
|
| 653 |
+
"step": 460
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.17203107658157601,
|
| 657 |
+
"grad_norm": 1.167439193543511,
|
| 658 |
+
"learning_rate": 3.1935133117658306e-05,
|
| 659 |
+
"loss": 0.952,
|
| 660 |
+
"step": 465
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.17388087310395856,
|
| 664 |
+
"grad_norm": 1.1956939577700672,
|
| 665 |
+
"learning_rate": 3.156981202247248e-05,
|
| 666 |
+
"loss": 0.9247,
|
| 667 |
+
"step": 470
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.1757306696263411,
|
| 671 |
+
"grad_norm": 1.1593001767390334,
|
| 672 |
+
"learning_rate": 3.1203378281316515e-05,
|
| 673 |
+
"loss": 0.9012,
|
| 674 |
+
"step": 475
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.17758046614872364,
|
| 678 |
+
"grad_norm": 1.0435734696221646,
|
| 679 |
+
"learning_rate": 3.0835932073516444e-05,
|
| 680 |
+
"loss": 0.9154,
|
| 681 |
+
"step": 480
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.17943026267110618,
|
| 685 |
+
"grad_norm": 1.1108156220083871,
|
| 686 |
+
"learning_rate": 3.0467573855196558e-05,
|
| 687 |
+
"loss": 0.899,
|
| 688 |
+
"step": 485
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.18128005919348872,
|
| 692 |
+
"grad_norm": 1.0167505348335453,
|
| 693 |
+
"learning_rate": 3.0098404331815695e-05,
|
| 694 |
+
"loss": 0.9174,
|
| 695 |
+
"step": 490
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.18312985571587126,
|
| 699 |
+
"grad_norm": 0.965632547470374,
|
| 700 |
+
"learning_rate": 2.9728524430635417e-05,
|
| 701 |
+
"loss": 0.9143,
|
| 702 |
+
"step": 495
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.1849796522382538,
|
| 706 |
+
"grad_norm": 1.1035315117198525,
|
| 707 |
+
"learning_rate": 2.9358035273127483e-05,
|
| 708 |
+
"loss": 0.9261,
|
| 709 |
+
"step": 500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.18682944876063634,
|
| 713 |
+
"grad_norm": 1.1169822035484065,
|
| 714 |
+
"learning_rate": 2.8987038147328238e-05,
|
| 715 |
+
"loss": 0.9313,
|
| 716 |
+
"step": 505
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.18867924528301888,
|
| 720 |
+
"grad_norm": 0.9769048796586226,
|
| 721 |
+
"learning_rate": 2.86156344801475e-05,
|
| 722 |
+
"loss": 0.9132,
|
| 723 |
+
"step": 510
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.1905290418054014,
|
| 727 |
+
"grad_norm": 1.030443130117634,
|
| 728 |
+
"learning_rate": 2.824392580963944e-05,
|
| 729 |
+
"loss": 0.9115,
|
| 730 |
+
"step": 515
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.19237883832778394,
|
| 734 |
+
"grad_norm": 1.1062439088724974,
|
| 735 |
+
"learning_rate": 2.787201375724307e-05,
|
| 736 |
+
"loss": 0.884,
|
| 737 |
+
"step": 520
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.19422863485016648,
|
| 741 |
+
"grad_norm": 1.0597749922719453,
|
| 742 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 743 |
+
"loss": 0.8783,
|
| 744 |
+
"step": 525
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.19607843137254902,
|
| 748 |
+
"grad_norm": 1.1259858455874245,
|
| 749 |
+
"learning_rate": 2.7127986242756936e-05,
|
| 750 |
+
"loss": 0.8982,
|
| 751 |
+
"step": 530
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.19792822789493156,
|
| 755 |
+
"grad_norm": 1.1500074530646092,
|
| 756 |
+
"learning_rate": 2.6756074190360563e-05,
|
| 757 |
+
"loss": 0.9127,
|
| 758 |
+
"step": 535
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.1997780244173141,
|
| 762 |
+
"grad_norm": 1.0395549080430184,
|
| 763 |
+
"learning_rate": 2.63843655198525e-05,
|
| 764 |
+
"loss": 0.8975,
|
| 765 |
+
"step": 540
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.20162782093969664,
|
| 769 |
+
"grad_norm": 1.083025795794125,
|
| 770 |
+
"learning_rate": 2.6012961852671767e-05,
|
| 771 |
+
"loss": 0.9231,
|
| 772 |
+
"step": 545
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.20347761746207918,
|
| 776 |
+
"grad_norm": 1.0839550256360933,
|
| 777 |
+
"learning_rate": 2.5641964726872526e-05,
|
| 778 |
+
"loss": 0.9459,
|
| 779 |
+
"step": 550
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.20532741398446172,
|
| 783 |
+
"grad_norm": 1.0874256330778362,
|
| 784 |
+
"learning_rate": 2.527147556936459e-05,
|
| 785 |
+
"loss": 0.891,
|
| 786 |
+
"step": 555
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.20717721050684423,
|
| 790 |
+
"grad_norm": 1.19750486880985,
|
| 791 |
+
"learning_rate": 2.4901595668184314e-05,
|
| 792 |
+
"loss": 0.8838,
|
| 793 |
+
"step": 560
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.20902700702922677,
|
| 797 |
+
"grad_norm": 1.092576941889689,
|
| 798 |
+
"learning_rate": 2.453242614480345e-05,
|
| 799 |
+
"loss": 0.9442,
|
| 800 |
+
"step": 565
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.21087680355160932,
|
| 804 |
+
"grad_norm": 1.227755705501534,
|
| 805 |
+
"learning_rate": 2.416406792648355e-05,
|
| 806 |
+
"loss": 0.8934,
|
| 807 |
+
"step": 570
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.21272660007399186,
|
| 811 |
+
"grad_norm": 1.2087319750759349,
|
| 812 |
+
"learning_rate": 2.3796621718683487e-05,
|
| 813 |
+
"loss": 0.9035,
|
| 814 |
+
"step": 575
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.2145763965963744,
|
| 818 |
+
"grad_norm": 1.1471923615461788,
|
| 819 |
+
"learning_rate": 2.3430187977527533e-05,
|
| 820 |
+
"loss": 0.8942,
|
| 821 |
+
"step": 580
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.21642619311875694,
|
| 825 |
+
"grad_norm": 0.9623686085619546,
|
| 826 |
+
"learning_rate": 2.3064866882341696e-05,
|
| 827 |
+
"loss": 0.8907,
|
| 828 |
+
"step": 585
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.21827598964113948,
|
| 832 |
+
"grad_norm": 1.1021625170926432,
|
| 833 |
+
"learning_rate": 2.27007583082657e-05,
|
| 834 |
+
"loss": 0.8488,
|
| 835 |
+
"step": 590
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.22012578616352202,
|
| 839 |
+
"grad_norm": 1.0507739548200454,
|
| 840 |
+
"learning_rate": 2.233796179894809e-05,
|
| 841 |
+
"loss": 0.9243,
|
| 842 |
+
"step": 595
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.22197558268590456,
|
| 846 |
+
"grad_norm": 1.0241361415980714,
|
| 847 |
+
"learning_rate": 2.1976576539332024e-05,
|
| 848 |
+
"loss": 0.8746,
|
| 849 |
+
"step": 600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.2238253792082871,
|
| 853 |
+
"grad_norm": 1.1308505070134895,
|
| 854 |
+
"learning_rate": 2.1616701328539057e-05,
|
| 855 |
+
"loss": 0.9094,
|
| 856 |
+
"step": 605
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.2256751757306696,
|
| 860 |
+
"grad_norm": 1.0945099014034436,
|
| 861 |
+
"learning_rate": 2.1258434552858502e-05,
|
| 862 |
+
"loss": 0.9125,
|
| 863 |
+
"step": 610
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.22752497225305215,
|
| 867 |
+
"grad_norm": 1.116483983505738,
|
| 868 |
+
"learning_rate": 2.090187415884955e-05,
|
| 869 |
+
"loss": 0.9148,
|
| 870 |
+
"step": 615
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.2293747687754347,
|
| 874 |
+
"grad_norm": 0.9728451267819282,
|
| 875 |
+
"learning_rate": 2.054711762656369e-05,
|
| 876 |
+
"loss": 0.8596,
|
| 877 |
+
"step": 620
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.23122456529781724,
|
| 881 |
+
"grad_norm": 1.0747467852239778,
|
| 882 |
+
"learning_rate": 2.0194261942894628e-05,
|
| 883 |
+
"loss": 0.8733,
|
| 884 |
+
"step": 625
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.23307436182019978,
|
| 888 |
+
"grad_norm": 1.0108414553651262,
|
| 889 |
+
"learning_rate": 1.984340357506303e-05,
|
| 890 |
+
"loss": 0.8476,
|
| 891 |
+
"step": 630
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.23492415834258232,
|
| 895 |
+
"grad_norm": 1.0393062726091873,
|
| 896 |
+
"learning_rate": 1.949463844424338e-05,
|
| 897 |
+
"loss": 0.8926,
|
| 898 |
+
"step": 635
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.23677395486496486,
|
| 902 |
+
"grad_norm": 1.017203795362952,
|
| 903 |
+
"learning_rate": 1.914806189934003e-05,
|
| 904 |
+
"loss": 0.8905,
|
| 905 |
+
"step": 640
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.2386237513873474,
|
| 909 |
+
"grad_norm": 1.0799681224310813,
|
| 910 |
+
"learning_rate": 1.8803768690919832e-05,
|
| 911 |
+
"loss": 0.8897,
|
| 912 |
+
"step": 645
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.24047354790972994,
|
| 916 |
+
"grad_norm": 1.0781213573400381,
|
| 917 |
+
"learning_rate": 1.8461852945308196e-05,
|
| 918 |
+
"loss": 0.8637,
|
| 919 |
+
"step": 650
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.24232334443211248,
|
| 923 |
+
"grad_norm": 1.1249251204928792,
|
| 924 |
+
"learning_rate": 1.8122408138855974e-05,
|
| 925 |
+
"loss": 0.8802,
|
| 926 |
+
"step": 655
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.244173140954495,
|
| 930 |
+
"grad_norm": 1.1923220257639287,
|
| 931 |
+
"learning_rate": 1.778552707238391e-05,
|
| 932 |
+
"loss": 0.9033,
|
| 933 |
+
"step": 660
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.24602293747687753,
|
| 937 |
+
"grad_norm": 1.0225563377110887,
|
| 938 |
+
"learning_rate": 1.7451301845811857e-05,
|
| 939 |
+
"loss": 0.8605,
|
| 940 |
+
"step": 665
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.24787273399926008,
|
| 944 |
+
"grad_norm": 1.0757446768121959,
|
| 945 |
+
"learning_rate": 1.711982383297953e-05,
|
| 946 |
+
"loss": 0.8906,
|
| 947 |
+
"step": 670
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.24972253052164262,
|
| 951 |
+
"grad_norm": 1.0589738260259443,
|
| 952 |
+
"learning_rate": 1.6791183656665846e-05,
|
| 953 |
+
"loss": 0.8938,
|
| 954 |
+
"step": 675
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.25157232704402516,
|
| 958 |
+
"grad_norm": 0.9884710814924109,
|
| 959 |
+
"learning_rate": 1.6465471163813574e-05,
|
| 960 |
+
"loss": 0.9306,
|
| 961 |
+
"step": 680
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.2534221235664077,
|
| 965 |
+
"grad_norm": 1.0526270648654439,
|
| 966 |
+
"learning_rate": 1.6142775400966023e-05,
|
| 967 |
+
"loss": 0.9038,
|
| 968 |
+
"step": 685
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.25527192008879024,
|
| 972 |
+
"grad_norm": 1.0276777351490343,
|
| 973 |
+
"learning_rate": 1.582318458992267e-05,
|
| 974 |
+
"loss": 0.8701,
|
| 975 |
+
"step": 690
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.2571217166111728,
|
| 979 |
+
"grad_norm": 1.1308701537231916,
|
| 980 |
+
"learning_rate": 1.5506786103620146e-05,
|
| 981 |
+
"loss": 0.9146,
|
| 982 |
+
"step": 695
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.2589715131335553,
|
| 986 |
+
"grad_norm": 1.1220390082042924,
|
| 987 |
+
"learning_rate": 1.5193666442245402e-05,
|
| 988 |
+
"loss": 0.894,
|
| 989 |
+
"step": 700
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.26082130965593786,
|
| 993 |
+
"grad_norm": 0.9872150958917795,
|
| 994 |
+
"learning_rate": 1.4883911209587368e-05,
|
| 995 |
+
"loss": 0.8846,
|
| 996 |
+
"step": 705
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.2626711061783204,
|
| 1000 |
+
"grad_norm": 1.0255612887775873,
|
| 1001 |
+
"learning_rate": 1.4577605089633773e-05,
|
| 1002 |
+
"loss": 0.8751,
|
| 1003 |
+
"step": 710
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.26452090270070294,
|
| 1007 |
+
"grad_norm": 1.0397487279123176,
|
| 1008 |
+
"learning_rate": 1.427483182341936e-05,
|
| 1009 |
+
"loss": 0.8553,
|
| 1010 |
+
"step": 715
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.2663706992230855,
|
| 1014 |
+
"grad_norm": 0.9811002147501025,
|
| 1015 |
+
"learning_rate": 1.3975674186131937e-05,
|
| 1016 |
+
"loss": 0.8811,
|
| 1017 |
+
"step": 720
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.268220495745468,
|
| 1021 |
+
"grad_norm": 0.9912452879027507,
|
| 1022 |
+
"learning_rate": 1.3680213964482475e-05,
|
| 1023 |
+
"loss": 0.882,
|
| 1024 |
+
"step": 725
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.2700702922678505,
|
| 1028 |
+
"grad_norm": 1.085598582193152,
|
| 1029 |
+
"learning_rate": 1.3388531934345416e-05,
|
| 1030 |
+
"loss": 0.8663,
|
| 1031 |
+
"step": 730
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.27192008879023305,
|
| 1035 |
+
"grad_norm": 1.0976908092393123,
|
| 1036 |
+
"learning_rate": 1.3100707838675377e-05,
|
| 1037 |
+
"loss": 0.8568,
|
| 1038 |
+
"step": 735
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.2737698853126156,
|
| 1042 |
+
"grad_norm": 1.110399579286176,
|
| 1043 |
+
"learning_rate": 1.2816820365706206e-05,
|
| 1044 |
+
"loss": 0.861,
|
| 1045 |
+
"step": 740
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.27561968183499813,
|
| 1049 |
+
"grad_norm": 0.9972527883665196,
|
| 1050 |
+
"learning_rate": 1.2536947127438415e-05,
|
| 1051 |
+
"loss": 0.8583,
|
| 1052 |
+
"step": 745
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.27746947835738067,
|
| 1056 |
+
"grad_norm": 0.9846092199040459,
|
| 1057 |
+
"learning_rate": 1.2261164638420832e-05,
|
| 1058 |
+
"loss": 0.8869,
|
| 1059 |
+
"step": 750
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.2793192748797632,
|
| 1063 |
+
"grad_norm": 1.0377584915303817,
|
| 1064 |
+
"learning_rate": 1.198954829483227e-05,
|
| 1065 |
+
"loss": 0.8689,
|
| 1066 |
+
"step": 755
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.28116907140214575,
|
| 1070 |
+
"grad_norm": 1.0080689260720659,
|
| 1071 |
+
"learning_rate": 1.1722172353869008e-05,
|
| 1072 |
+
"loss": 0.8494,
|
| 1073 |
+
"step": 760
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.2830188679245283,
|
| 1077 |
+
"grad_norm": 1.1377623343596663,
|
| 1078 |
+
"learning_rate": 1.1459109913443567e-05,
|
| 1079 |
+
"loss": 0.8789,
|
| 1080 |
+
"step": 765
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.28486866444691084,
|
| 1084 |
+
"grad_norm": 1.0844127152295038,
|
| 1085 |
+
"learning_rate": 1.12004328922005e-05,
|
| 1086 |
+
"loss": 0.8714,
|
| 1087 |
+
"step": 770
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.2867184609692934,
|
| 1091 |
+
"grad_norm": 0.9818291742586884,
|
| 1092 |
+
"learning_rate": 1.094621200985454e-05,
|
| 1093 |
+
"loss": 0.8762,
|
| 1094 |
+
"step": 775
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.2885682574916759,
|
| 1098 |
+
"grad_norm": 1.0561402094886105,
|
| 1099 |
+
"learning_rate": 1.0696516767856546e-05,
|
| 1100 |
+
"loss": 0.8602,
|
| 1101 |
+
"step": 780
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.29041805401405846,
|
| 1105 |
+
"grad_norm": 0.9507833621657122,
|
| 1106 |
+
"learning_rate": 1.0451415430392474e-05,
|
| 1107 |
+
"loss": 0.8493,
|
| 1108 |
+
"step": 785
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.292267850536441,
|
| 1112 |
+
"grad_norm": 1.0142419832874243,
|
| 1113 |
+
"learning_rate": 1.0210975005720677e-05,
|
| 1114 |
+
"loss": 0.8621,
|
| 1115 |
+
"step": 790
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.29411764705882354,
|
| 1119 |
+
"grad_norm": 1.0414075410194514,
|
| 1120 |
+
"learning_rate": 9.975261227852472e-06,
|
| 1121 |
+
"loss": 0.8867,
|
| 1122 |
+
"step": 795
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.2959674435812061,
|
| 1126 |
+
"grad_norm": 1.0172601224221116,
|
| 1127 |
+
"learning_rate": 9.744338538581147e-06,
|
| 1128 |
+
"loss": 0.861,
|
| 1129 |
+
"step": 800
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.2978172401035886,
|
| 1133 |
+
"grad_norm": 1.0520376994102576,
|
| 1134 |
+
"learning_rate": 9.518270069864195e-06,
|
| 1135 |
+
"loss": 0.8693,
|
| 1136 |
+
"step": 805
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.29966703662597116,
|
| 1140 |
+
"grad_norm": 1.0297669417176623,
|
| 1141 |
+
"learning_rate": 9.297117626563687e-06,
|
| 1142 |
+
"loss": 0.8556,
|
| 1143 |
+
"step": 810
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.3015168331483537,
|
| 1147 |
+
"grad_norm": 1.1288942237451884,
|
| 1148 |
+
"learning_rate": 9.080941669549423e-06,
|
| 1149 |
+
"loss": 0.8636,
|
| 1150 |
+
"step": 815
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.30336662967073624,
|
| 1154 |
+
"grad_norm": 1.0700035713287315,
|
| 1155 |
+
"learning_rate": 8.86980129916955e-06,
|
| 1156 |
+
"loss": 0.8735,
|
| 1157 |
+
"step": 820
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.3052164261931188,
|
| 1161 |
+
"grad_norm": 1.0433713434057506,
|
| 1162 |
+
"learning_rate": 8.663754239093109e-06,
|
| 1163 |
+
"loss": 0.8745,
|
| 1164 |
+
"step": 825
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.30706622271550127,
|
| 1168 |
+
"grad_norm": 0.9399395136256045,
|
| 1169 |
+
"learning_rate": 8.462856820528952e-06,
|
| 1170 |
+
"loss": 0.8625,
|
| 1171 |
+
"step": 830
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.3089160192378838,
|
| 1175 |
+
"grad_norm": 1.012084107494069,
|
| 1176 |
+
"learning_rate": 8.26716396682537e-06,
|
| 1177 |
+
"loss": 0.8575,
|
| 1178 |
+
"step": 835
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.31076581576026635,
|
| 1182 |
+
"grad_norm": 1.0479666608180018,
|
| 1183 |
+
"learning_rate": 8.076729178454588e-06,
|
| 1184 |
+
"loss": 0.8766,
|
| 1185 |
+
"step": 840
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.3126156122826489,
|
| 1189 |
+
"grad_norm": 1.137450242844205,
|
| 1190 |
+
"learning_rate": 7.89160451838626e-06,
|
| 1191 |
+
"loss": 0.8544,
|
| 1192 |
+
"step": 845
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.31446540880503143,
|
| 1196 |
+
"grad_norm": 1.0197993824595155,
|
| 1197 |
+
"learning_rate": 7.711840597853998e-06,
|
| 1198 |
+
"loss": 0.8583,
|
| 1199 |
+
"step": 850
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.316315205327414,
|
| 1203 |
+
"grad_norm": 1.0686381595688783,
|
| 1204 |
+
"learning_rate": 7.537486562518735e-06,
|
| 1205 |
+
"loss": 0.8518,
|
| 1206 |
+
"step": 855
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.3181650018497965,
|
| 1210 |
+
"grad_norm": 0.9978231071140236,
|
| 1211 |
+
"learning_rate": 7.368590079032822e-06,
|
| 1212 |
+
"loss": 0.8648,
|
| 1213 |
+
"step": 860
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.32001479837217905,
|
| 1217 |
+
"grad_norm": 1.0560393356842295,
|
| 1218 |
+
"learning_rate": 7.205197322008425e-06,
|
| 1219 |
+
"loss": 0.8349,
|
| 1220 |
+
"step": 865
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.3218645948945616,
|
| 1224 |
+
"grad_norm": 0.9842318437238298,
|
| 1225 |
+
"learning_rate": 7.047352961393844e-06,
|
| 1226 |
+
"loss": 0.8595,
|
| 1227 |
+
"step": 870
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.32371439141694414,
|
| 1231 |
+
"grad_norm": 0.9952118512469338,
|
| 1232 |
+
"learning_rate": 6.8951001502612065e-06,
|
| 1233 |
+
"loss": 0.869,
|
| 1234 |
+
"step": 875
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.3255641879393267,
|
| 1238 |
+
"grad_norm": 0.9424925973460067,
|
| 1239 |
+
"learning_rate": 6.748480513008844e-06,
|
| 1240 |
+
"loss": 0.8745,
|
| 1241 |
+
"step": 880
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.3274139844617092,
|
| 1245 |
+
"grad_norm": 1.0535831291700837,
|
| 1246 |
+
"learning_rate": 6.607534133981594e-06,
|
| 1247 |
+
"loss": 0.8748,
|
| 1248 |
+
"step": 885
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.32926378098409176,
|
| 1252 |
+
"grad_norm": 1.0723567866928634,
|
| 1253 |
+
"learning_rate": 6.472299546512134e-06,
|
| 1254 |
+
"loss": 0.8815,
|
| 1255 |
+
"step": 890
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.3311135775064743,
|
| 1259 |
+
"grad_norm": 1.0482717572477003,
|
| 1260 |
+
"learning_rate": 6.342813722386374e-06,
|
| 1261 |
+
"loss": 0.8543,
|
| 1262 |
+
"step": 895
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.33296337402885684,
|
| 1266 |
+
"grad_norm": 1.046401991122096,
|
| 1267 |
+
"learning_rate": 6.219112061735721e-06,
|
| 1268 |
+
"loss": 0.8785,
|
| 1269 |
+
"step": 900
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.3348131705512394,
|
| 1273 |
+
"grad_norm": 1.0211208893623536,
|
| 1274 |
+
"learning_rate": 6.1012283833590465e-06,
|
| 1275 |
+
"loss": 0.8757,
|
| 1276 |
+
"step": 905
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.3366629670736219,
|
| 1280 |
+
"grad_norm": 1.0035173287590444,
|
| 1281 |
+
"learning_rate": 5.989194915476954e-06,
|
| 1282 |
+
"loss": 0.8442,
|
| 1283 |
+
"step": 910
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.33851276359600446,
|
| 1287 |
+
"grad_norm": 0.9878066592821019,
|
| 1288 |
+
"learning_rate": 5.883042286920918e-06,
|
| 1289 |
+
"loss": 0.8638,
|
| 1290 |
+
"step": 915
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.340362560118387,
|
| 1294 |
+
"grad_norm": 1.05735602894325,
|
| 1295 |
+
"learning_rate": 5.782799518759658e-06,
|
| 1296 |
+
"loss": 0.8593,
|
| 1297 |
+
"step": 920
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.3422123566407695,
|
| 1301 |
+
"grad_norm": 1.0649111496758787,
|
| 1302 |
+
"learning_rate": 5.688494016365067e-06,
|
| 1303 |
+
"loss": 0.845,
|
| 1304 |
+
"step": 925
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.34406215316315203,
|
| 1308 |
+
"grad_norm": 0.957225274905496,
|
| 1309 |
+
"learning_rate": 5.600151561919871e-06,
|
| 1310 |
+
"loss": 0.8134,
|
| 1311 |
+
"step": 930
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.34591194968553457,
|
| 1315 |
+
"grad_norm": 1.0690464205878594,
|
| 1316 |
+
"learning_rate": 5.517796307369017e-06,
|
| 1317 |
+
"loss": 0.8364,
|
| 1318 |
+
"step": 935
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.3477617462079171,
|
| 1322 |
+
"grad_norm": 1.0520479367896434,
|
| 1323 |
+
"learning_rate": 5.44145076781679e-06,
|
| 1324 |
+
"loss": 0.8697,
|
| 1325 |
+
"step": 940
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.34961154273029965,
|
| 1329 |
+
"grad_norm": 0.9648256295717741,
|
| 1330 |
+
"learning_rate": 5.371135815371398e-06,
|
| 1331 |
+
"loss": 0.8334,
|
| 1332 |
+
"step": 945
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.3514613392526822,
|
| 1336 |
+
"grad_norm": 1.0090319357515105,
|
| 1337 |
+
"learning_rate": 5.3068706734387484e-06,
|
| 1338 |
+
"loss": 0.866,
|
| 1339 |
+
"step": 950
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.35331113577506473,
|
| 1343 |
+
"grad_norm": 1.0792234217081147,
|
| 1344 |
+
"learning_rate": 5.248672911466959e-06,
|
| 1345 |
+
"loss": 0.8796,
|
| 1346 |
+
"step": 955
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.3551609322974473,
|
| 1350 |
+
"grad_norm": 0.9703249482629162,
|
| 1351 |
+
"learning_rate": 5.196558440143059e-06,
|
| 1352 |
+
"loss": 0.8329,
|
| 1353 |
+
"step": 960
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.3570107288198298,
|
| 1357 |
+
"grad_norm": 1.0041999264211245,
|
| 1358 |
+
"learning_rate": 5.150541507043143e-06,
|
| 1359 |
+
"loss": 0.8871,
|
| 1360 |
+
"step": 965
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.35886052534221236,
|
| 1364 |
+
"grad_norm": 0.9816980969927207,
|
| 1365 |
+
"learning_rate": 5.110634692737244e-06,
|
| 1366 |
+
"loss": 0.8391,
|
| 1367 |
+
"step": 970
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.3607103218645949,
|
| 1371 |
+
"grad_norm": 0.9903601458320098,
|
| 1372 |
+
"learning_rate": 5.0768489073499295e-06,
|
| 1373 |
+
"loss": 0.8707,
|
| 1374 |
+
"step": 975
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.36256011838697744,
|
| 1378 |
+
"grad_norm": 0.9605358480816072,
|
| 1379 |
+
"learning_rate": 5.049193387577574e-06,
|
| 1380 |
+
"loss": 0.8677,
|
| 1381 |
+
"step": 980
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.36440991490936,
|
| 1385 |
+
"grad_norm": 1.0149824944311472,
|
| 1386 |
+
"learning_rate": 5.027675694163144e-06,
|
| 1387 |
+
"loss": 0.8252,
|
| 1388 |
+
"step": 985
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.3662597114317425,
|
| 1392 |
+
"grad_norm": 1.0514282882320594,
|
| 1393 |
+
"learning_rate": 5.012301709829164e-06,
|
| 1394 |
+
"loss": 0.8361,
|
| 1395 |
+
"step": 990
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.36810950795412506,
|
| 1399 |
+
"grad_norm": 1.0144426282975678,
|
| 1400 |
+
"learning_rate": 5.003075637669448e-06,
|
| 1401 |
+
"loss": 0.8604,
|
| 1402 |
+
"step": 995
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.3699593044765076,
|
| 1406 |
+
"grad_norm": 1.041222716272992,
|
| 1407 |
+
"learning_rate": 5e-06,
|
| 1408 |
+
"loss": 0.8502,
|
| 1409 |
+
"step": 1000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.3699593044765076,
|
| 1413 |
+
"step": 1000,
|
| 1414 |
+
"total_flos": 28487712768000.0,
|
| 1415 |
+
"train_loss": 0.920774197101593,
|
| 1416 |
+
"train_runtime": 1759.2244,
|
| 1417 |
+
"train_samples_per_second": 18.19,
|
| 1418 |
+
"train_steps_per_second": 0.568
|
| 1419 |
+
}
|
| 1420 |
+
],
|
| 1421 |
+
"logging_steps": 5,
|
| 1422 |
+
"max_steps": 1000,
|
| 1423 |
+
"num_input_tokens_seen": 0,
|
| 1424 |
+
"num_train_epochs": 1,
|
| 1425 |
+
"save_steps": 100,
|
| 1426 |
+
"stateful_callbacks": {
|
| 1427 |
+
"TrainerControl": {
|
| 1428 |
+
"args": {
|
| 1429 |
+
"should_epoch_stop": false,
|
| 1430 |
+
"should_evaluate": false,
|
| 1431 |
+
"should_log": false,
|
| 1432 |
+
"should_save": true,
|
| 1433 |
+
"should_training_stop": true
|
| 1434 |
+
},
|
| 1435 |
+
"attributes": {}
|
| 1436 |
+
}
|
| 1437 |
+
},
|
| 1438 |
+
"total_flos": 28487712768000.0,
|
| 1439 |
+
"train_batch_size": 16,
|
| 1440 |
+
"trial_name": null,
|
| 1441 |
+
"trial_params": null
|
| 1442 |
+
}
|