Model save
Browse files- README.md +58 -0
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1050 -0
README.md
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen2.5-0.5B-Instruct
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: qwen-2.5-0.5B-instruct-sft-lora-countdown-o3-5k
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for qwen-2.5-0.5B-instruct-sft-lora-countdown-o3-5k
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-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="chloeli/qwen-2.5-0.5B-instruct-sft-lora-countdown-o3-5k", 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/chloeli/huggingface/runs/jxpdihqa)
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with SFT.
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.15.2
|
| 38 |
+
- Transformers: 4.49.0
|
| 39 |
+
- Pytorch: 2.6.0
|
| 40 |
+
- Datasets: 3.3.2
|
| 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": 1.5047606448160768e+16,
|
| 3 |
+
"train_loss": 0.3481092903137207,
|
| 4 |
+
"train_runtime": 972.7733,
|
| 5 |
+
"train_samples": 5000,
|
| 6 |
+
"train_samples_per_second": 5.14,
|
| 7 |
+
"train_steps_per_second": 0.642
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 1.5047606448160768e+16,
|
| 3 |
+
"train_loss": 0.3481092903137207,
|
| 4 |
+
"train_runtime": 972.7733,
|
| 5 |
+
"train_samples": 5000,
|
| 6 |
+
"train_samples_per_second": 5.14,
|
| 7 |
+
"train_steps_per_second": 0.642
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1050 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 625,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0016,
|
| 13 |
+
"grad_norm": 1.5717865228652954,
|
| 14 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 15 |
+
"loss": 2.7077,
|
| 16 |
+
"mean_token_accuracy": 0.459680899977684,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.008,
|
| 21 |
+
"grad_norm": 1.202968716621399,
|
| 22 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 23 |
+
"loss": 2.2671,
|
| 24 |
+
"mean_token_accuracy": 0.5242175050079823,
|
| 25 |
+
"step": 5
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.016,
|
| 29 |
+
"grad_norm": 1.0542172193527222,
|
| 30 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 31 |
+
"loss": 2.2698,
|
| 32 |
+
"mean_token_accuracy": 0.5283941477537155,
|
| 33 |
+
"step": 10
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.024,
|
| 37 |
+
"grad_norm": 0.7223343253135681,
|
| 38 |
+
"learning_rate": 4.761904761904762e-05,
|
| 39 |
+
"loss": 2.1156,
|
| 40 |
+
"mean_token_accuracy": 0.5514968127012253,
|
| 41 |
+
"step": 15
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.032,
|
| 45 |
+
"grad_norm": 0.6670629382133484,
|
| 46 |
+
"learning_rate": 6.349206349206349e-05,
|
| 47 |
+
"loss": 1.8644,
|
| 48 |
+
"mean_token_accuracy": 0.5947718620300293,
|
| 49 |
+
"step": 20
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.04,
|
| 53 |
+
"grad_norm": 0.8882656693458557,
|
| 54 |
+
"learning_rate": 7.936507936507937e-05,
|
| 55 |
+
"loss": 1.6563,
|
| 56 |
+
"mean_token_accuracy": 0.6210890173912048,
|
| 57 |
+
"step": 25
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.048,
|
| 61 |
+
"grad_norm": 1.0315715074539185,
|
| 62 |
+
"learning_rate": 9.523809523809524e-05,
|
| 63 |
+
"loss": 1.4437,
|
| 64 |
+
"mean_token_accuracy": 0.654322224855423,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.056,
|
| 69 |
+
"grad_norm": 1.4416899681091309,
|
| 70 |
+
"learning_rate": 0.00011111111111111112,
|
| 71 |
+
"loss": 1.0037,
|
| 72 |
+
"mean_token_accuracy": 0.7415849208831787,
|
| 73 |
+
"step": 35
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.064,
|
| 77 |
+
"grad_norm": 0.8182677626609802,
|
| 78 |
+
"learning_rate": 0.00012698412698412698,
|
| 79 |
+
"loss": 0.6626,
|
| 80 |
+
"mean_token_accuracy": 0.853333032131195,
|
| 81 |
+
"step": 40
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.072,
|
| 85 |
+
"grad_norm": 0.4539111256599426,
|
| 86 |
+
"learning_rate": 0.00014285714285714287,
|
| 87 |
+
"loss": 0.4946,
|
| 88 |
+
"mean_token_accuracy": 0.8977023303508759,
|
| 89 |
+
"step": 45
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.08,
|
| 93 |
+
"grad_norm": 0.3059958815574646,
|
| 94 |
+
"learning_rate": 0.00015873015873015873,
|
| 95 |
+
"loss": 0.4924,
|
| 96 |
+
"mean_token_accuracy": 0.8881445944309234,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.088,
|
| 101 |
+
"grad_norm": 0.20895916223526,
|
| 102 |
+
"learning_rate": 0.00017460317460317462,
|
| 103 |
+
"loss": 0.3816,
|
| 104 |
+
"mean_token_accuracy": 0.9159973561763763,
|
| 105 |
+
"step": 55
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.096,
|
| 109 |
+
"grad_norm": 0.241449773311615,
|
| 110 |
+
"learning_rate": 0.00019047619047619048,
|
| 111 |
+
"loss": 0.2424,
|
| 112 |
+
"mean_token_accuracy": 0.9461545169353485,
|
| 113 |
+
"step": 60
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.104,
|
| 117 |
+
"grad_norm": 0.13886623084545135,
|
| 118 |
+
"learning_rate": 0.00019999375039475277,
|
| 119 |
+
"loss": 0.2525,
|
| 120 |
+
"mean_token_accuracy": 0.9451982796192169,
|
| 121 |
+
"step": 65
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.112,
|
| 125 |
+
"grad_norm": 0.15808190405368805,
|
| 126 |
+
"learning_rate": 0.0001999234513064475,
|
| 127 |
+
"loss": 0.3108,
|
| 128 |
+
"mean_token_accuracy": 0.9311837553977966,
|
| 129 |
+
"step": 70
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.12,
|
| 133 |
+
"grad_norm": 0.1368260234594345,
|
| 134 |
+
"learning_rate": 0.00019977509622105233,
|
| 135 |
+
"loss": 0.2217,
|
| 136 |
+
"mean_token_accuracy": 0.9483723938465118,
|
| 137 |
+
"step": 75
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.128,
|
| 141 |
+
"grad_norm": 0.14972065389156342,
|
| 142 |
+
"learning_rate": 0.0001995488010273198,
|
| 143 |
+
"loss": 0.1707,
|
| 144 |
+
"mean_token_accuracy": 0.9621350407600403,
|
| 145 |
+
"step": 80
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.136,
|
| 149 |
+
"grad_norm": 0.2032196819782257,
|
| 150 |
+
"learning_rate": 0.00019924474249753655,
|
| 151 |
+
"loss": 0.3796,
|
| 152 |
+
"mean_token_accuracy": 0.9232820510864258,
|
| 153 |
+
"step": 85
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.144,
|
| 157 |
+
"grad_norm": 0.09374168515205383,
|
| 158 |
+
"learning_rate": 0.00019886315814943647,
|
| 159 |
+
"loss": 0.3032,
|
| 160 |
+
"mean_token_accuracy": 0.9280714929103852,
|
| 161 |
+
"step": 90
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.152,
|
| 165 |
+
"grad_norm": 0.1895170956850052,
|
| 166 |
+
"learning_rate": 0.0001984043460606618,
|
| 167 |
+
"loss": 0.2635,
|
| 168 |
+
"mean_token_accuracy": 0.9419675827026367,
|
| 169 |
+
"step": 95
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.16,
|
| 173 |
+
"grad_norm": 0.17207027971744537,
|
| 174 |
+
"learning_rate": 0.0001978686646359173,
|
| 175 |
+
"loss": 0.2961,
|
| 176 |
+
"mean_token_accuracy": 0.9313458263874054,
|
| 177 |
+
"step": 100
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.168,
|
| 181 |
+
"grad_norm": 0.1757289320230484,
|
| 182 |
+
"learning_rate": 0.0001972565323269996,
|
| 183 |
+
"loss": 0.2087,
|
| 184 |
+
"mean_token_accuracy": 0.9513824045658111,
|
| 185 |
+
"step": 105
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.176,
|
| 189 |
+
"grad_norm": 0.2260744869709015,
|
| 190 |
+
"learning_rate": 0.00019656842730592046,
|
| 191 |
+
"loss": 0.3184,
|
| 192 |
+
"mean_token_accuracy": 0.9232833862304688,
|
| 193 |
+
"step": 110
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.184,
|
| 197 |
+
"grad_norm": 0.052938517183065414,
|
| 198 |
+
"learning_rate": 0.0001958048870913786,
|
| 199 |
+
"loss": 0.1822,
|
| 200 |
+
"mean_token_accuracy": 0.9538747906684876,
|
| 201 |
+
"step": 115
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.192,
|
| 205 |
+
"grad_norm": 0.1941942721605301,
|
| 206 |
+
"learning_rate": 0.0001949665081288729,
|
| 207 |
+
"loss": 0.2273,
|
| 208 |
+
"mean_token_accuracy": 0.9453578472137452,
|
| 209 |
+
"step": 120
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.2,
|
| 213 |
+
"grad_norm": 0.2198331207036972,
|
| 214 |
+
"learning_rate": 0.00019405394532478424,
|
| 215 |
+
"loss": 0.2049,
|
| 216 |
+
"mean_token_accuracy": 0.9542803287506103,
|
| 217 |
+
"step": 125
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.208,
|
| 221 |
+
"grad_norm": 0.24331390857696533,
|
| 222 |
+
"learning_rate": 0.00019306791153479006,
|
| 223 |
+
"loss": 0.2061,
|
| 224 |
+
"mean_token_accuracy": 0.9513079404830933,
|
| 225 |
+
"step": 130
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.216,
|
| 229 |
+
"grad_norm": 0.2335432767868042,
|
| 230 |
+
"learning_rate": 0.00019200917700701176,
|
| 231 |
+
"loss": 0.1368,
|
| 232 |
+
"mean_token_accuracy": 0.9651266932487488,
|
| 233 |
+
"step": 135
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.224,
|
| 237 |
+
"grad_norm": 0.23610533773899078,
|
| 238 |
+
"learning_rate": 0.0001908785687803289,
|
| 239 |
+
"loss": 0.3442,
|
| 240 |
+
"mean_token_accuracy": 0.9188908874988556,
|
| 241 |
+
"step": 140
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.232,
|
| 245 |
+
"grad_norm": 0.22944550216197968,
|
| 246 |
+
"learning_rate": 0.00018967697003833157,
|
| 247 |
+
"loss": 0.337,
|
| 248 |
+
"mean_token_accuracy": 0.9214253067970276,
|
| 249 |
+
"step": 145
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.24,
|
| 253 |
+
"grad_norm": 0.21961958706378937,
|
| 254 |
+
"learning_rate": 0.0001884053194194142,
|
| 255 |
+
"loss": 0.223,
|
| 256 |
+
"mean_token_accuracy": 0.9437191367149353,
|
| 257 |
+
"step": 150
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.248,
|
| 261 |
+
"grad_norm": 0.20730167627334595,
|
| 262 |
+
"learning_rate": 0.00018706461028355104,
|
| 263 |
+
"loss": 0.2741,
|
| 264 |
+
"mean_token_accuracy": 0.938022255897522,
|
| 265 |
+
"step": 155
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.256,
|
| 269 |
+
"grad_norm": 0.16260486841201782,
|
| 270 |
+
"learning_rate": 0.00018565588993632487,
|
| 271 |
+
"loss": 0.2648,
|
| 272 |
+
"mean_token_accuracy": 0.9357350587844848,
|
| 273 |
+
"step": 160
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.264,
|
| 277 |
+
"grad_norm": 0.22049878537654877,
|
| 278 |
+
"learning_rate": 0.0001841802588108161,
|
| 279 |
+
"loss": 0.3412,
|
| 280 |
+
"mean_token_accuracy": 0.9166976153850556,
|
| 281 |
+
"step": 165
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.272,
|
| 285 |
+
"grad_norm": 0.19371896982192993,
|
| 286 |
+
"learning_rate": 0.00018263886960799062,
|
| 287 |
+
"loss": 0.2456,
|
| 288 |
+
"mean_token_accuracy": 0.945724493265152,
|
| 289 |
+
"step": 170
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.28,
|
| 293 |
+
"grad_norm": 0.22933930158615112,
|
| 294 |
+
"learning_rate": 0.00018103292639625837,
|
| 295 |
+
"loss": 0.3841,
|
| 296 |
+
"mean_token_accuracy": 0.9089005470275879,
|
| 297 |
+
"step": 175
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.288,
|
| 301 |
+
"grad_norm": 0.1793544441461563,
|
| 302 |
+
"learning_rate": 0.0001793636836709057,
|
| 303 |
+
"loss": 0.2999,
|
| 304 |
+
"mean_token_accuracy": 0.9297621667385101,
|
| 305 |
+
"step": 180
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.296,
|
| 309 |
+
"grad_norm": 0.0395968072116375,
|
| 310 |
+
"learning_rate": 0.0001776324453741365,
|
| 311 |
+
"loss": 0.1624,
|
| 312 |
+
"mean_token_accuracy": 0.9584607481956482,
|
| 313 |
+
"step": 185
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.304,
|
| 317 |
+
"grad_norm": 0.2473573386669159,
|
| 318 |
+
"learning_rate": 0.00017584056387648727,
|
| 319 |
+
"loss": 0.3053,
|
| 320 |
+
"mean_token_accuracy": 0.9382229685783386,
|
| 321 |
+
"step": 190
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.312,
|
| 325 |
+
"grad_norm": 0.04702218249440193,
|
| 326 |
+
"learning_rate": 0.0001739894389204122,
|
| 327 |
+
"loss": 0.172,
|
| 328 |
+
"mean_token_accuracy": 0.9594997644424439,
|
| 329 |
+
"step": 195
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.32,
|
| 333 |
+
"grad_norm": 0.23656697571277618,
|
| 334 |
+
"learning_rate": 0.00017208051652686335,
|
| 335 |
+
"loss": 0.2276,
|
| 336 |
+
"mean_token_accuracy": 0.9398577272891998,
|
| 337 |
+
"step": 200
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.328,
|
| 341 |
+
"grad_norm": 0.07044972479343414,
|
| 342 |
+
"learning_rate": 0.00017011528786571969,
|
| 343 |
+
"loss": 0.2854,
|
| 344 |
+
"mean_token_accuracy": 0.926360809803009,
|
| 345 |
+
"step": 205
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.336,
|
| 349 |
+
"grad_norm": 0.03222399577498436,
|
| 350 |
+
"learning_rate": 0.00016809528809094807,
|
| 351 |
+
"loss": 0.2172,
|
| 352 |
+
"mean_token_accuracy": 0.9438831508159637,
|
| 353 |
+
"step": 210
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.344,
|
| 357 |
+
"grad_norm": 0.243390753865242,
|
| 358 |
+
"learning_rate": 0.0001660220951414055,
|
| 359 |
+
"loss": 0.2852,
|
| 360 |
+
"mean_token_accuracy": 0.9389818787574769,
|
| 361 |
+
"step": 215
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.352,
|
| 365 |
+
"grad_norm": 0.23130089044570923,
|
| 366 |
+
"learning_rate": 0.00016389732850821966,
|
| 367 |
+
"loss": 0.2899,
|
| 368 |
+
"mean_token_accuracy": 0.9331403017044068,
|
| 369 |
+
"step": 220
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.36,
|
| 373 |
+
"grad_norm": 0.2532561123371124,
|
| 374 |
+
"learning_rate": 0.0001617226479697105,
|
| 375 |
+
"loss": 0.3069,
|
| 376 |
+
"mean_token_accuracy": 0.9253335416316986,
|
| 377 |
+
"step": 225
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.368,
|
| 381 |
+
"grad_norm": 0.20991413295269012,
|
| 382 |
+
"learning_rate": 0.00015949975229484134,
|
| 383 |
+
"loss": 0.3178,
|
| 384 |
+
"mean_token_accuracy": 0.9201959311962128,
|
| 385 |
+
"step": 230
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.376,
|
| 389 |
+
"grad_norm": 0.20208591222763062,
|
| 390 |
+
"learning_rate": 0.00015723037791621193,
|
| 391 |
+
"loss": 0.18,
|
| 392 |
+
"mean_token_accuracy": 0.9534633576869964,
|
| 393 |
+
"step": 235
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.384,
|
| 397 |
+
"grad_norm": 0.23704850673675537,
|
| 398 |
+
"learning_rate": 0.00015491629757363032,
|
| 399 |
+
"loss": 0.3864,
|
| 400 |
+
"mean_token_accuracy": 0.9041174411773681,
|
| 401 |
+
"step": 240
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.392,
|
| 405 |
+
"grad_norm": 0.22486881911754608,
|
| 406 |
+
"learning_rate": 0.00015255931892932333,
|
| 407 |
+
"loss": 0.3174,
|
| 408 |
+
"mean_token_accuracy": 0.9177540898323059,
|
| 409 |
+
"step": 245
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4,
|
| 413 |
+
"grad_norm": 0.2111813724040985,
|
| 414 |
+
"learning_rate": 0.0001501612831558664,
|
| 415 |
+
"loss": 0.3145,
|
| 416 |
+
"mean_token_accuracy": 0.9238011240959167,
|
| 417 |
+
"step": 250
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.408,
|
| 421 |
+
"grad_norm": 0.23434004187583923,
|
| 422 |
+
"learning_rate": 0.00014772406349793744,
|
| 423 |
+
"loss": 0.2928,
|
| 424 |
+
"mean_token_accuracy": 0.929435807466507,
|
| 425 |
+
"step": 255
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.416,
|
| 429 |
+
"grad_norm": 0.25533345341682434,
|
| 430 |
+
"learning_rate": 0.0001452495638090167,
|
| 431 |
+
"loss": 0.3171,
|
| 432 |
+
"mean_token_accuracy": 0.927526468038559,
|
| 433 |
+
"step": 260
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.424,
|
| 437 |
+
"grad_norm": 0.1922113448381424,
|
| 438 |
+
"learning_rate": 0.00014273971706417647,
|
| 439 |
+
"loss": 0.2162,
|
| 440 |
+
"mean_token_accuracy": 0.9465757727622985,
|
| 441 |
+
"step": 265
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.432,
|
| 445 |
+
"grad_norm": 0.2513563334941864,
|
| 446 |
+
"learning_rate": 0.00014019648385012244,
|
| 447 |
+
"loss": 0.3813,
|
| 448 |
+
"mean_token_accuracy": 0.908132141828537,
|
| 449 |
+
"step": 270
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.44,
|
| 453 |
+
"grad_norm": 0.25622016191482544,
|
| 454 |
+
"learning_rate": 0.00013762185083366556,
|
| 455 |
+
"loss": 0.3005,
|
| 456 |
+
"mean_token_accuracy": 0.9310987770557404,
|
| 457 |
+
"step": 275
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.448,
|
| 461 |
+
"grad_norm": 0.20694169402122498,
|
| 462 |
+
"learning_rate": 0.00013501782920982184,
|
| 463 |
+
"loss": 0.1721,
|
| 464 |
+
"mean_token_accuracy": 0.9563108265399933,
|
| 465 |
+
"step": 280
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.456,
|
| 469 |
+
"grad_norm": 0.21768316626548767,
|
| 470 |
+
"learning_rate": 0.00013238645313075104,
|
| 471 |
+
"loss": 0.2444,
|
| 472 |
+
"mean_token_accuracy": 0.9432082533836365,
|
| 473 |
+
"step": 285
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.464,
|
| 477 |
+
"grad_norm": 0.1706106811761856,
|
| 478 |
+
"learning_rate": 0.00012972977811676287,
|
| 479 |
+
"loss": 0.233,
|
| 480 |
+
"mean_token_accuracy": 0.9396327018737793,
|
| 481 |
+
"step": 290
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.472,
|
| 485 |
+
"grad_norm": 0.2342345267534256,
|
| 486 |
+
"learning_rate": 0.00012704987945063068,
|
| 487 |
+
"loss": 0.2106,
|
| 488 |
+
"mean_token_accuracy": 0.9517422139644622,
|
| 489 |
+
"step": 295
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.48,
|
| 493 |
+
"grad_norm": 0.2374894917011261,
|
| 494 |
+
"learning_rate": 0.00012434885055646823,
|
| 495 |
+
"loss": 0.2222,
|
| 496 |
+
"mean_token_accuracy": 0.9430564045906067,
|
| 497 |
+
"step": 300
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.488,
|
| 501 |
+
"grad_norm": 0.20393089950084686,
|
| 502 |
+
"learning_rate": 0.00012162880136443447,
|
| 503 |
+
"loss": 0.3069,
|
| 504 |
+
"mean_token_accuracy": 0.9208929121494294,
|
| 505 |
+
"step": 305
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.496,
|
| 509 |
+
"grad_norm": 0.2132725864648819,
|
| 510 |
+
"learning_rate": 0.00011889185666254506,
|
| 511 |
+
"loss": 0.2821,
|
| 512 |
+
"mean_token_accuracy": 0.9318405151367187,
|
| 513 |
+
"step": 310
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.504,
|
| 517 |
+
"grad_norm": 0.15664634108543396,
|
| 518 |
+
"learning_rate": 0.00011614015443687722,
|
| 519 |
+
"loss": 0.16,
|
| 520 |
+
"mean_token_accuracy": 0.9559766769409179,
|
| 521 |
+
"step": 315
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.512,
|
| 525 |
+
"grad_norm": 0.22459107637405396,
|
| 526 |
+
"learning_rate": 0.0001133758442014651,
|
| 527 |
+
"loss": 0.1543,
|
| 528 |
+
"mean_token_accuracy": 0.9638924479484559,
|
| 529 |
+
"step": 320
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.52,
|
| 533 |
+
"grad_norm": 0.2619342505931854,
|
| 534 |
+
"learning_rate": 0.00011060108531918971,
|
| 535 |
+
"loss": 0.2865,
|
| 536 |
+
"mean_token_accuracy": 0.9385590136051178,
|
| 537 |
+
"step": 325
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.528,
|
| 541 |
+
"grad_norm": 0.21467000246047974,
|
| 542 |
+
"learning_rate": 0.0001078180453149754,
|
| 543 |
+
"loss": 0.2918,
|
| 544 |
+
"mean_token_accuracy": 0.9376808702945709,
|
| 545 |
+
"step": 330
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.536,
|
| 549 |
+
"grad_norm": 0.2496771514415741,
|
| 550 |
+
"learning_rate": 0.00010502889818261075,
|
| 551 |
+
"loss": 0.4184,
|
| 552 |
+
"mean_token_accuracy": 0.8969966411590576,
|
| 553 |
+
"step": 335
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.544,
|
| 557 |
+
"grad_norm": 0.2137787938117981,
|
| 558 |
+
"learning_rate": 0.00010223582268651586,
|
| 559 |
+
"loss": 0.257,
|
| 560 |
+
"mean_token_accuracy": 0.9423645496368408,
|
| 561 |
+
"step": 340
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.552,
|
| 565 |
+
"grad_norm": 0.2176610231399536,
|
| 566 |
+
"learning_rate": 9.94410006597835e-05,
|
| 567 |
+
"loss": 0.3322,
|
| 568 |
+
"mean_token_accuracy": 0.9129085004329681,
|
| 569 |
+
"step": 345
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.56,
|
| 573 |
+
"grad_norm": 0.21382386982440948,
|
| 574 |
+
"learning_rate": 9.66466152998226e-05,
|
| 575 |
+
"loss": 0.3399,
|
| 576 |
+
"mean_token_accuracy": 0.9135319530963898,
|
| 577 |
+
"step": 350
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.568,
|
| 581 |
+
"grad_norm": 0.22343839704990387,
|
| 582 |
+
"learning_rate": 9.385484946293637e-05,
|
| 583 |
+
"loss": 0.2464,
|
| 584 |
+
"mean_token_accuracy": 0.9397193789482117,
|
| 585 |
+
"step": 355
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.576,
|
| 589 |
+
"grad_norm": 0.18257169425487518,
|
| 590 |
+
"learning_rate": 9.106788395916678e-05,
|
| 591 |
+
"loss": 0.1896,
|
| 592 |
+
"mean_token_accuracy": 0.951214075088501,
|
| 593 |
+
"step": 360
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.584,
|
| 597 |
+
"grad_norm": 0.04507756605744362,
|
| 598 |
+
"learning_rate": 8.828789584873754e-05,
|
| 599 |
+
"loss": 0.2693,
|
| 600 |
+
"mean_token_accuracy": 0.9345629513263702,
|
| 601 |
+
"step": 365
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.592,
|
| 605 |
+
"grad_norm": 0.23864690959453583,
|
| 606 |
+
"learning_rate": 8.551705674142617e-05,
|
| 607 |
+
"loss": 0.2743,
|
| 608 |
+
"mean_token_accuracy": 0.9325626492500305,
|
| 609 |
+
"step": 370
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.6,
|
| 613 |
+
"grad_norm": 0.035314541310071945,
|
| 614 |
+
"learning_rate": 8.275753110019367e-05,
|
| 615 |
+
"loss": 0.1799,
|
| 616 |
+
"mean_token_accuracy": 0.9588949978351593,
|
| 617 |
+
"step": 375
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.608,
|
| 621 |
+
"grad_norm": 0.19131699204444885,
|
| 622 |
+
"learning_rate": 8.001147455039735e-05,
|
| 623 |
+
"loss": 0.2339,
|
| 624 |
+
"mean_token_accuracy": 0.9403933227062226,
|
| 625 |
+
"step": 380
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.616,
|
| 629 |
+
"grad_norm": 0.20205603539943695,
|
| 630 |
+
"learning_rate": 7.728103219590681e-05,
|
| 631 |
+
"loss": 0.2315,
|
| 632 |
+
"mean_token_accuracy": 0.9407491445541382,
|
| 633 |
+
"step": 385
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.624,
|
| 637 |
+
"grad_norm": 0.23396767675876617,
|
| 638 |
+
"learning_rate": 7.456833694343906e-05,
|
| 639 |
+
"loss": 0.2248,
|
| 640 |
+
"mean_token_accuracy": 0.9434651434421539,
|
| 641 |
+
"step": 390
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.632,
|
| 645 |
+
"grad_norm": 0.05149530991911888,
|
| 646 |
+
"learning_rate": 7.18755078364214e-05,
|
| 647 |
+
"loss": 0.1834,
|
| 648 |
+
"mean_token_accuracy": 0.950703501701355,
|
| 649 |
+
"step": 395
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.64,
|
| 653 |
+
"grad_norm": 0.1885431706905365,
|
| 654 |
+
"learning_rate": 6.920464839968405e-05,
|
| 655 |
+
"loss": 0.2292,
|
| 656 |
+
"mean_token_accuracy": 0.9455470621585846,
|
| 657 |
+
"step": 400
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.648,
|
| 661 |
+
"grad_norm": 0.22768881916999817,
|
| 662 |
+
"learning_rate": 6.65578449962749e-05,
|
| 663 |
+
"loss": 0.3276,
|
| 664 |
+
"mean_token_accuracy": 0.9167207598686218,
|
| 665 |
+
"step": 405
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.656,
|
| 669 |
+
"grad_norm": 0.2584383189678192,
|
| 670 |
+
"learning_rate": 6.393716519768047e-05,
|
| 671 |
+
"loss": 0.2714,
|
| 672 |
+
"mean_token_accuracy": 0.9378273606300354,
|
| 673 |
+
"step": 410
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.664,
|
| 677 |
+
"grad_norm": 0.27143898606300354,
|
| 678 |
+
"learning_rate": 6.134465616872598e-05,
|
| 679 |
+
"loss": 0.3238,
|
| 680 |
+
"mean_token_accuracy": 0.9266472160816193,
|
| 681 |
+
"step": 415
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.672,
|
| 685 |
+
"grad_norm": 0.16849137842655182,
|
| 686 |
+
"learning_rate": 5.878234306841637e-05,
|
| 687 |
+
"loss": 0.1855,
|
| 688 |
+
"mean_token_accuracy": 0.9501271903514862,
|
| 689 |
+
"step": 420
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.68,
|
| 693 |
+
"grad_norm": 0.23237672448158264,
|
| 694 |
+
"learning_rate": 5.62522274679673e-05,
|
| 695 |
+
"loss": 0.2252,
|
| 696 |
+
"mean_token_accuracy": 0.9437941908836365,
|
| 697 |
+
"step": 425
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.688,
|
| 701 |
+
"grad_norm": 0.19916483759880066,
|
| 702 |
+
"learning_rate": 5.375628578726181e-05,
|
| 703 |
+
"loss": 0.1561,
|
| 704 |
+
"mean_token_accuracy": 0.9600933134555817,
|
| 705 |
+
"step": 430
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.696,
|
| 709 |
+
"grad_norm": 0.22922399640083313,
|
| 710 |
+
"learning_rate": 5.1296467750954314e-05,
|
| 711 |
+
"loss": 0.3147,
|
| 712 |
+
"mean_token_accuracy": 0.925593638420105,
|
| 713 |
+
"step": 435
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.704,
|
| 717 |
+
"grad_norm": 0.2341531664133072,
|
| 718 |
+
"learning_rate": 4.8874694865427676e-05,
|
| 719 |
+
"loss": 0.2745,
|
| 720 |
+
"mean_token_accuracy": 0.9313132464885712,
|
| 721 |
+
"step": 440
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.712,
|
| 725 |
+
"grad_norm": 0.24038153886795044,
|
| 726 |
+
"learning_rate": 4.649285891779327e-05,
|
| 727 |
+
"loss": 0.2954,
|
| 728 |
+
"mean_token_accuracy": 0.9240333676338196,
|
| 729 |
+
"step": 445
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.72,
|
| 733 |
+
"grad_norm": 0.2073904424905777,
|
| 734 |
+
"learning_rate": 4.415282049810644e-05,
|
| 735 |
+
"loss": 0.2418,
|
| 736 |
+
"mean_token_accuracy": 0.93941730260849,
|
| 737 |
+
"step": 450
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.728,
|
| 741 |
+
"grad_norm": 0.232681542634964,
|
| 742 |
+
"learning_rate": 4.1856407545951834e-05,
|
| 743 |
+
"loss": 0.3575,
|
| 744 |
+
"mean_token_accuracy": 0.9102162063121796,
|
| 745 |
+
"step": 455
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.736,
|
| 749 |
+
"grad_norm": 0.20729133486747742,
|
| 750 |
+
"learning_rate": 3.9605413922533874e-05,
|
| 751 |
+
"loss": 0.2844,
|
| 752 |
+
"mean_token_accuracy": 0.9322370827198029,
|
| 753 |
+
"step": 460
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.744,
|
| 757 |
+
"grad_norm": 0.20540155470371246,
|
| 758 |
+
"learning_rate": 3.740159800938784e-05,
|
| 759 |
+
"loss": 0.1801,
|
| 760 |
+
"mean_token_accuracy": 0.9571804940700531,
|
| 761 |
+
"step": 465
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.752,
|
| 765 |
+
"grad_norm": 0.27062955498695374,
|
| 766 |
+
"learning_rate": 3.5246681334806175e-05,
|
| 767 |
+
"loss": 0.1628,
|
| 768 |
+
"mean_token_accuracy": 0.9602212190628052,
|
| 769 |
+
"step": 470
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.76,
|
| 773 |
+
"grad_norm": 0.2563348710536957,
|
| 774 |
+
"learning_rate": 3.3142347229053015e-05,
|
| 775 |
+
"loss": 0.2317,
|
| 776 |
+
"mean_token_accuracy": 0.9395133197307587,
|
| 777 |
+
"step": 475
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.768,
|
| 781 |
+
"grad_norm": 0.23781529068946838,
|
| 782 |
+
"learning_rate": 3.109023950941736e-05,
|
| 783 |
+
"loss": 0.3099,
|
| 784 |
+
"mean_token_accuracy": 0.9263223648071289,
|
| 785 |
+
"step": 480
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.776,
|
| 789 |
+
"grad_norm": 0.04337846860289574,
|
| 790 |
+
"learning_rate": 2.909196119613218e-05,
|
| 791 |
+
"loss": 0.1425,
|
| 792 |
+
"mean_token_accuracy": 0.9617048621177673,
|
| 793 |
+
"step": 485
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.784,
|
| 797 |
+
"grad_norm": 0.20578506588935852,
|
| 798 |
+
"learning_rate": 2.7149073260162416e-05,
|
| 799 |
+
"loss": 0.2547,
|
| 800 |
+
"mean_token_accuracy": 0.9427634000778198,
|
| 801 |
+
"step": 490
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.792,
|
| 805 |
+
"grad_norm": 0.2657724618911743,
|
| 806 |
+
"learning_rate": 2.5263093403840142e-05,
|
| 807 |
+
"loss": 0.2759,
|
| 808 |
+
"mean_token_accuracy": 0.9276242256164551,
|
| 809 |
+
"step": 495
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.8,
|
| 813 |
+
"grad_norm": 0.20405757427215576,
|
| 814 |
+
"learning_rate": 2.3435494875299314e-05,
|
| 815 |
+
"loss": 0.1946,
|
| 816 |
+
"mean_token_accuracy": 0.9477591276168823,
|
| 817 |
+
"step": 500
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.808,
|
| 821 |
+
"grad_norm": 0.2120545208454132,
|
| 822 |
+
"learning_rate": 2.166770531763633e-05,
|
| 823 |
+
"loss": 0.1492,
|
| 824 |
+
"mean_token_accuracy": 0.9617019176483155,
|
| 825 |
+
"step": 505
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.816,
|
| 829 |
+
"grad_norm": 0.1790059208869934,
|
| 830 |
+
"learning_rate": 1.9961105653695266e-05,
|
| 831 |
+
"loss": 0.2097,
|
| 832 |
+
"mean_token_accuracy": 0.9475365877151489,
|
| 833 |
+
"step": 510
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.824,
|
| 837 |
+
"grad_norm": 0.23069578409194946,
|
| 838 |
+
"learning_rate": 1.8317029007349085e-05,
|
| 839 |
+
"loss": 0.3262,
|
| 840 |
+
"mean_token_accuracy": 0.9245599567890167,
|
| 841 |
+
"step": 515
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.832,
|
| 845 |
+
"grad_norm": 0.1952206939458847,
|
| 846 |
+
"learning_rate": 1.6736759662119183e-05,
|
| 847 |
+
"loss": 0.2593,
|
| 848 |
+
"mean_token_accuracy": 0.9349613726139069,
|
| 849 |
+
"step": 520
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.84,
|
| 853 |
+
"grad_norm": 0.16667886078357697,
|
| 854 |
+
"learning_rate": 1.5221532057947419e-05,
|
| 855 |
+
"loss": 0.1832,
|
| 856 |
+
"mean_token_accuracy": 0.9498234987258911,
|
| 857 |
+
"step": 525
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.848,
|
| 861 |
+
"grad_norm": 0.16531656682491302,
|
| 862 |
+
"learning_rate": 1.3772529826903269e-05,
|
| 863 |
+
"loss": 0.3003,
|
| 864 |
+
"mean_token_accuracy": 0.927387660741806,
|
| 865 |
+
"step": 530
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.856,
|
| 869 |
+
"grad_norm": 0.03590136766433716,
|
| 870 |
+
"learning_rate": 1.23908848685804e-05,
|
| 871 |
+
"loss": 0.209,
|
| 872 |
+
"mean_token_accuracy": 0.9517148792743683,
|
| 873 |
+
"step": 535
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.864,
|
| 877 |
+
"grad_norm": 0.20711013674736023,
|
| 878 |
+
"learning_rate": 1.1077676465904208e-05,
|
| 879 |
+
"loss": 0.229,
|
| 880 |
+
"mean_token_accuracy": 0.9443170964717865,
|
| 881 |
+
"step": 540
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.872,
|
| 885 |
+
"grad_norm": 0.2173730880022049,
|
| 886 |
+
"learning_rate": 9.833930442041506e-06,
|
| 887 |
+
"loss": 0.1711,
|
| 888 |
+
"mean_token_accuracy": 0.95446497797966,
|
| 889 |
+
"step": 545
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 0.88,
|
| 893 |
+
"grad_norm": 0.25314000248908997,
|
| 894 |
+
"learning_rate": 8.660618359070604e-06,
|
| 895 |
+
"loss": 0.2478,
|
| 896 |
+
"mean_token_accuracy": 0.9370902240276336,
|
| 897 |
+
"step": 550
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.888,
|
| 901 |
+
"grad_norm": 0.21840238571166992,
|
| 902 |
+
"learning_rate": 7.558656759037797e-06,
|
| 903 |
+
"loss": 0.184,
|
| 904 |
+
"mean_token_accuracy": 0.9521986067295074,
|
| 905 |
+
"step": 555
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.896,
|
| 909 |
+
"grad_norm": 0.21549563109874725,
|
| 910 |
+
"learning_rate": 6.528906447993288e-06,
|
| 911 |
+
"loss": 0.2443,
|
| 912 |
+
"mean_token_accuracy": 0.9441956281661987,
|
| 913 |
+
"step": 560
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.904,
|
| 917 |
+
"grad_norm": 0.1734277904033661,
|
| 918 |
+
"learning_rate": 5.572171823565797e-06,
|
| 919 |
+
"loss": 0.22,
|
| 920 |
+
"mean_token_accuracy": 0.9439444482326508,
|
| 921 |
+
"step": 565
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.912,
|
| 925 |
+
"grad_norm": 0.2154768407344818,
|
| 926 |
+
"learning_rate": 4.689200246600867e-06,
|
| 927 |
+
"loss": 0.1987,
|
| 928 |
+
"mean_token_accuracy": 0.9493344485759735,
|
| 929 |
+
"step": 570
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.92,
|
| 933 |
+
"grad_norm": 0.1548367738723755,
|
| 934 |
+
"learning_rate": 3.880681457354118e-06,
|
| 935 |
+
"loss": 0.1942,
|
| 936 |
+
"mean_token_accuracy": 0.9513484239578247,
|
| 937 |
+
"step": 575
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.928,
|
| 941 |
+
"grad_norm": 0.18485531210899353,
|
| 942 |
+
"learning_rate": 3.1472470366950334e-06,
|
| 943 |
+
"loss": 0.2624,
|
| 944 |
+
"mean_token_accuracy": 0.9301692008972168,
|
| 945 |
+
"step": 580
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.936,
|
| 949 |
+
"grad_norm": 0.24809648096561432,
|
| 950 |
+
"learning_rate": 2.4894699127426367e-06,
|
| 951 |
+
"loss": 0.2501,
|
| 952 |
+
"mean_token_accuracy": 0.9349741101264953,
|
| 953 |
+
"step": 585
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.944,
|
| 957 |
+
"grad_norm": 0.2469683289527893,
|
| 958 |
+
"learning_rate": 1.907863913318153e-06,
|
| 959 |
+
"loss": 0.1806,
|
| 960 |
+
"mean_token_accuracy": 0.9534947514533997,
|
| 961 |
+
"step": 590
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.952,
|
| 965 |
+
"grad_norm": 0.21619068086147308,
|
| 966 |
+
"learning_rate": 1.4028833645643113e-06,
|
| 967 |
+
"loss": 0.1114,
|
| 968 |
+
"mean_token_accuracy": 0.9707520604133606,
|
| 969 |
+
"step": 595
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.96,
|
| 973 |
+
"grad_norm": 0.20197683572769165,
|
| 974 |
+
"learning_rate": 9.749227360448143e-07,
|
| 975 |
+
"loss": 0.202,
|
| 976 |
+
"mean_token_accuracy": 0.947997921705246,
|
| 977 |
+
"step": 600
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.968,
|
| 981 |
+
"grad_norm": 0.21867972612380981,
|
| 982 |
+
"learning_rate": 6.243163326014267e-07,
|
| 983 |
+
"loss": 0.2394,
|
| 984 |
+
"mean_token_accuracy": 0.9438014030456543,
|
| 985 |
+
"step": 605
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.976,
|
| 989 |
+
"grad_norm": 0.2265222817659378,
|
| 990 |
+
"learning_rate": 3.5133803320896994e-07,
|
| 991 |
+
"loss": 0.2911,
|
| 992 |
+
"mean_token_accuracy": 0.9275119185447693,
|
| 993 |
+
"step": 610
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.984,
|
| 997 |
+
"grad_norm": 0.19685231149196625,
|
| 998 |
+
"learning_rate": 1.562010770326916e-07,
|
| 999 |
+
"loss": 0.2725,
|
| 1000 |
+
"mean_token_accuracy": 0.9320066690444946,
|
| 1001 |
+
"step": 615
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.992,
|
| 1005 |
+
"grad_norm": 0.15336309373378754,
|
| 1006 |
+
"learning_rate": 3.905789685471062e-08,
|
| 1007 |
+
"loss": 0.3007,
|
| 1008 |
+
"mean_token_accuracy": 0.9265518903732299,
|
| 1009 |
+
"step": 620
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.0,
|
| 1013 |
+
"grad_norm": 0.23042897880077362,
|
| 1014 |
+
"learning_rate": 0.0,
|
| 1015 |
+
"loss": 0.2386,
|
| 1016 |
+
"mean_token_accuracy": 0.9401108622550964,
|
| 1017 |
+
"step": 625
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.0,
|
| 1021 |
+
"step": 625,
|
| 1022 |
+
"total_flos": 1.5047606448160768e+16,
|
| 1023 |
+
"train_loss": 0.3481092903137207,
|
| 1024 |
+
"train_runtime": 972.7733,
|
| 1025 |
+
"train_samples_per_second": 5.14,
|
| 1026 |
+
"train_steps_per_second": 0.642
|
| 1027 |
+
}
|
| 1028 |
+
],
|
| 1029 |
+
"logging_steps": 5,
|
| 1030 |
+
"max_steps": 625,
|
| 1031 |
+
"num_input_tokens_seen": 0,
|
| 1032 |
+
"num_train_epochs": 1,
|
| 1033 |
+
"save_steps": 100,
|
| 1034 |
+
"stateful_callbacks": {
|
| 1035 |
+
"TrainerControl": {
|
| 1036 |
+
"args": {
|
| 1037 |
+
"should_epoch_stop": false,
|
| 1038 |
+
"should_evaluate": false,
|
| 1039 |
+
"should_log": false,
|
| 1040 |
+
"should_save": true,
|
| 1041 |
+
"should_training_stop": true
|
| 1042 |
+
},
|
| 1043 |
+
"attributes": {}
|
| 1044 |
+
}
|
| 1045 |
+
},
|
| 1046 |
+
"total_flos": 1.5047606448160768e+16,
|
| 1047 |
+
"train_batch_size": 4,
|
| 1048 |
+
"trial_name": null,
|
| 1049 |
+
"trial_params": null
|
| 1050 |
+
}
|