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-search-react-seq8k-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-search-react-seq8k-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-search-react-seq8k-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/fcgmyt9b)
|
| 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": 8.947190570627891e+16,
|
| 3 |
+
"train_loss": 0.03501472600698471,
|
| 4 |
+
"train_runtime": 1355.0569,
|
| 5 |
+
"train_samples": 5000,
|
| 6 |
+
"train_samples_per_second": 3.69,
|
| 7 |
+
"train_steps_per_second": 0.461
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 8.947190570627891e+16,
|
| 3 |
+
"train_loss": 0.03501472600698471,
|
| 4 |
+
"train_runtime": 1355.0569,
|
| 5 |
+
"train_samples": 5000,
|
| 6 |
+
"train_samples_per_second": 3.69,
|
| 7 |
+
"train_steps_per_second": 0.461
|
| 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": 0.35758575797080994,
|
| 14 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 15 |
+
"loss": 0.2918,
|
| 16 |
+
"mean_token_accuracy": 0.9317843317985535,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.008,
|
| 21 |
+
"grad_norm": 0.5407769680023193,
|
| 22 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 23 |
+
"loss": 0.334,
|
| 24 |
+
"mean_token_accuracy": 0.9228289499878883,
|
| 25 |
+
"step": 5
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.016,
|
| 29 |
+
"grad_norm": 0.2607511878013611,
|
| 30 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 31 |
+
"loss": 0.3393,
|
| 32 |
+
"mean_token_accuracy": 0.9221176445484162,
|
| 33 |
+
"step": 10
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.024,
|
| 37 |
+
"grad_norm": 0.1869797259569168,
|
| 38 |
+
"learning_rate": 4.761904761904762e-05,
|
| 39 |
+
"loss": 0.2987,
|
| 40 |
+
"mean_token_accuracy": 0.9277177572250366,
|
| 41 |
+
"step": 15
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.032,
|
| 45 |
+
"grad_norm": 0.13320446014404297,
|
| 46 |
+
"learning_rate": 6.349206349206349e-05,
|
| 47 |
+
"loss": 0.2496,
|
| 48 |
+
"mean_token_accuracy": 0.9362682342529297,
|
| 49 |
+
"step": 20
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.04,
|
| 53 |
+
"grad_norm": 0.1366184651851654,
|
| 54 |
+
"learning_rate": 7.936507936507937e-05,
|
| 55 |
+
"loss": 0.2242,
|
| 56 |
+
"mean_token_accuracy": 0.9423655033111572,
|
| 57 |
+
"step": 25
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.048,
|
| 61 |
+
"grad_norm": 0.16860996186733246,
|
| 62 |
+
"learning_rate": 9.523809523809524e-05,
|
| 63 |
+
"loss": 0.2088,
|
| 64 |
+
"mean_token_accuracy": 0.9416319727897644,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.056,
|
| 69 |
+
"grad_norm": 0.11776343733072281,
|
| 70 |
+
"learning_rate": 0.00011111111111111112,
|
| 71 |
+
"loss": 0.1694,
|
| 72 |
+
"mean_token_accuracy": 0.9534160196781158,
|
| 73 |
+
"step": 35
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.064,
|
| 77 |
+
"grad_norm": 0.16114415228366852,
|
| 78 |
+
"learning_rate": 0.00012698412698412698,
|
| 79 |
+
"loss": 0.122,
|
| 80 |
+
"mean_token_accuracy": 0.9644478380680084,
|
| 81 |
+
"step": 40
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.072,
|
| 85 |
+
"grad_norm": 0.12855438888072968,
|
| 86 |
+
"learning_rate": 0.00014285714285714287,
|
| 87 |
+
"loss": 0.0825,
|
| 88 |
+
"mean_token_accuracy": 0.9731343150138855,
|
| 89 |
+
"step": 45
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.08,
|
| 93 |
+
"grad_norm": 0.11165463179349899,
|
| 94 |
+
"learning_rate": 0.00015873015873015873,
|
| 95 |
+
"loss": 0.0556,
|
| 96 |
+
"mean_token_accuracy": 0.9801632344722748,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.088,
|
| 101 |
+
"grad_norm": 0.11061853170394897,
|
| 102 |
+
"learning_rate": 0.00017460317460317462,
|
| 103 |
+
"loss": 0.0411,
|
| 104 |
+
"mean_token_accuracy": 0.9841950356960296,
|
| 105 |
+
"step": 55
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.096,
|
| 109 |
+
"grad_norm": 0.08642830699682236,
|
| 110 |
+
"learning_rate": 0.00019047619047619048,
|
| 111 |
+
"loss": 0.0391,
|
| 112 |
+
"mean_token_accuracy": 0.9843509018421173,
|
| 113 |
+
"step": 60
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.104,
|
| 117 |
+
"grad_norm": 0.10681991279125214,
|
| 118 |
+
"learning_rate": 0.00019999375039475277,
|
| 119 |
+
"loss": 0.0353,
|
| 120 |
+
"mean_token_accuracy": 0.9855763733386993,
|
| 121 |
+
"step": 65
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.112,
|
| 125 |
+
"grad_norm": 0.09516558051109314,
|
| 126 |
+
"learning_rate": 0.0001999234513064475,
|
| 127 |
+
"loss": 0.0315,
|
| 128 |
+
"mean_token_accuracy": 0.9871177613735199,
|
| 129 |
+
"step": 70
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.12,
|
| 133 |
+
"grad_norm": 0.0980728268623352,
|
| 134 |
+
"learning_rate": 0.00019977509622105233,
|
| 135 |
+
"loss": 0.0297,
|
| 136 |
+
"mean_token_accuracy": 0.9878179967403412,
|
| 137 |
+
"step": 75
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.128,
|
| 141 |
+
"grad_norm": 0.06371135264635086,
|
| 142 |
+
"learning_rate": 0.0001995488010273198,
|
| 143 |
+
"loss": 0.0294,
|
| 144 |
+
"mean_token_accuracy": 0.9879972219467164,
|
| 145 |
+
"step": 80
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.136,
|
| 149 |
+
"grad_norm": 0.07989838719367981,
|
| 150 |
+
"learning_rate": 0.00019924474249753655,
|
| 151 |
+
"loss": 0.0283,
|
| 152 |
+
"mean_token_accuracy": 0.9881730616092682,
|
| 153 |
+
"step": 85
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.144,
|
| 157 |
+
"grad_norm": 0.07586675882339478,
|
| 158 |
+
"learning_rate": 0.00019886315814943647,
|
| 159 |
+
"loss": 0.0265,
|
| 160 |
+
"mean_token_accuracy": 0.9889779150485992,
|
| 161 |
+
"step": 90
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.152,
|
| 165 |
+
"grad_norm": 0.04547739028930664,
|
| 166 |
+
"learning_rate": 0.0001984043460606618,
|
| 167 |
+
"loss": 0.0273,
|
| 168 |
+
"mean_token_accuracy": 0.9885783493518829,
|
| 169 |
+
"step": 95
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.16,
|
| 173 |
+
"grad_norm": 0.06169603019952774,
|
| 174 |
+
"learning_rate": 0.0001978686646359173,
|
| 175 |
+
"loss": 0.0244,
|
| 176 |
+
"mean_token_accuracy": 0.9895990073680878,
|
| 177 |
+
"step": 100
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.168,
|
| 181 |
+
"grad_norm": 0.06537225097417831,
|
| 182 |
+
"learning_rate": 0.0001972565323269996,
|
| 183 |
+
"loss": 0.0257,
|
| 184 |
+
"mean_token_accuracy": 0.9891368567943573,
|
| 185 |
+
"step": 105
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.176,
|
| 189 |
+
"grad_norm": 0.0580211840569973,
|
| 190 |
+
"learning_rate": 0.00019656842730592046,
|
| 191 |
+
"loss": 0.0248,
|
| 192 |
+
"mean_token_accuracy": 0.9892465591430664,
|
| 193 |
+
"step": 110
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.184,
|
| 197 |
+
"grad_norm": 0.050916433334350586,
|
| 198 |
+
"learning_rate": 0.0001958048870913786,
|
| 199 |
+
"loss": 0.0242,
|
| 200 |
+
"mean_token_accuracy": 0.9895266950130462,
|
| 201 |
+
"step": 115
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.192,
|
| 205 |
+
"grad_norm": 0.05767924338579178,
|
| 206 |
+
"learning_rate": 0.0001949665081288729,
|
| 207 |
+
"loss": 0.0229,
|
| 208 |
+
"mean_token_accuracy": 0.9902631580829621,
|
| 209 |
+
"step": 120
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.2,
|
| 213 |
+
"grad_norm": 0.0532953217625618,
|
| 214 |
+
"learning_rate": 0.00019405394532478424,
|
| 215 |
+
"loss": 0.022,
|
| 216 |
+
"mean_token_accuracy": 0.9905503630638123,
|
| 217 |
+
"step": 125
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.208,
|
| 221 |
+
"grad_norm": 0.06565467268228531,
|
| 222 |
+
"learning_rate": 0.00019306791153479006,
|
| 223 |
+
"loss": 0.0237,
|
| 224 |
+
"mean_token_accuracy": 0.9896982431411743,
|
| 225 |
+
"step": 130
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.216,
|
| 229 |
+
"grad_norm": 0.07220505177974701,
|
| 230 |
+
"learning_rate": 0.00019200917700701176,
|
| 231 |
+
"loss": 0.0236,
|
| 232 |
+
"mean_token_accuracy": 0.9899338781833649,
|
| 233 |
+
"step": 135
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.224,
|
| 237 |
+
"grad_norm": 0.0682716816663742,
|
| 238 |
+
"learning_rate": 0.0001908785687803289,
|
| 239 |
+
"loss": 0.0229,
|
| 240 |
+
"mean_token_accuracy": 0.9898757636547089,
|
| 241 |
+
"step": 140
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.232,
|
| 245 |
+
"grad_norm": 0.04945652559399605,
|
| 246 |
+
"learning_rate": 0.00018967697003833157,
|
| 247 |
+
"loss": 0.0231,
|
| 248 |
+
"mean_token_accuracy": 0.9899633169174195,
|
| 249 |
+
"step": 145
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.24,
|
| 253 |
+
"grad_norm": 0.05297650769352913,
|
| 254 |
+
"learning_rate": 0.0001884053194194142,
|
| 255 |
+
"loss": 0.0203,
|
| 256 |
+
"mean_token_accuracy": 0.9912129640579224,
|
| 257 |
+
"step": 150
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.248,
|
| 261 |
+
"grad_norm": 0.05250829458236694,
|
| 262 |
+
"learning_rate": 0.00018706461028355104,
|
| 263 |
+
"loss": 0.0206,
|
| 264 |
+
"mean_token_accuracy": 0.9913576364517211,
|
| 265 |
+
"step": 155
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.256,
|
| 269 |
+
"grad_norm": 0.05076967924833298,
|
| 270 |
+
"learning_rate": 0.00018565588993632487,
|
| 271 |
+
"loss": 0.0252,
|
| 272 |
+
"mean_token_accuracy": 0.988889217376709,
|
| 273 |
+
"step": 160
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.264,
|
| 277 |
+
"grad_norm": 0.05826568230986595,
|
| 278 |
+
"learning_rate": 0.0001841802588108161,
|
| 279 |
+
"loss": 0.0216,
|
| 280 |
+
"mean_token_accuracy": 0.990450119972229,
|
| 281 |
+
"step": 165
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.272,
|
| 285 |
+
"grad_norm": 0.06373272091150284,
|
| 286 |
+
"learning_rate": 0.00018263886960799062,
|
| 287 |
+
"loss": 0.0226,
|
| 288 |
+
"mean_token_accuracy": 0.9901030540466309,
|
| 289 |
+
"step": 170
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.28,
|
| 293 |
+
"grad_norm": 0.058064743876457214,
|
| 294 |
+
"learning_rate": 0.00018103292639625837,
|
| 295 |
+
"loss": 0.0221,
|
| 296 |
+
"mean_token_accuracy": 0.9902714133262634,
|
| 297 |
+
"step": 175
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.288,
|
| 301 |
+
"grad_norm": 0.055855073034763336,
|
| 302 |
+
"learning_rate": 0.0001793636836709057,
|
| 303 |
+
"loss": 0.0216,
|
| 304 |
+
"mean_token_accuracy": 0.9907104849815369,
|
| 305 |
+
"step": 180
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.296,
|
| 309 |
+
"grad_norm": 0.04673806577920914,
|
| 310 |
+
"learning_rate": 0.0001776324453741365,
|
| 311 |
+
"loss": 0.0204,
|
| 312 |
+
"mean_token_accuracy": 0.9914953052997589,
|
| 313 |
+
"step": 185
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.304,
|
| 317 |
+
"grad_norm": 0.05335243418812752,
|
| 318 |
+
"learning_rate": 0.00017584056387648727,
|
| 319 |
+
"loss": 0.0225,
|
| 320 |
+
"mean_token_accuracy": 0.9900318086147308,
|
| 321 |
+
"step": 190
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.312,
|
| 325 |
+
"grad_norm": 0.0564313605427742,
|
| 326 |
+
"learning_rate": 0.0001739894389204122,
|
| 327 |
+
"loss": 0.0213,
|
| 328 |
+
"mean_token_accuracy": 0.9907339155673981,
|
| 329 |
+
"step": 195
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.32,
|
| 333 |
+
"grad_norm": 0.04711662605404854,
|
| 334 |
+
"learning_rate": 0.00017208051652686335,
|
| 335 |
+
"loss": 0.0214,
|
| 336 |
+
"mean_token_accuracy": 0.9910281419754028,
|
| 337 |
+
"step": 200
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.328,
|
| 341 |
+
"grad_norm": 0.04865030199289322,
|
| 342 |
+
"learning_rate": 0.00017011528786571969,
|
| 343 |
+
"loss": 0.0196,
|
| 344 |
+
"mean_token_accuracy": 0.9915309727191925,
|
| 345 |
+
"step": 205
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.336,
|
| 349 |
+
"grad_norm": 0.065005362033844,
|
| 350 |
+
"learning_rate": 0.00016809528809094807,
|
| 351 |
+
"loss": 0.0212,
|
| 352 |
+
"mean_token_accuracy": 0.9909046471118927,
|
| 353 |
+
"step": 210
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.344,
|
| 357 |
+
"grad_norm": 0.06859762966632843,
|
| 358 |
+
"learning_rate": 0.0001660220951414055,
|
| 359 |
+
"loss": 0.0208,
|
| 360 |
+
"mean_token_accuracy": 0.991067773103714,
|
| 361 |
+
"step": 215
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.352,
|
| 365 |
+
"grad_norm": 0.062335845082998276,
|
| 366 |
+
"learning_rate": 0.00016389732850821966,
|
| 367 |
+
"loss": 0.0202,
|
| 368 |
+
"mean_token_accuracy": 0.991184014081955,
|
| 369 |
+
"step": 220
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.36,
|
| 373 |
+
"grad_norm": 0.0483645424246788,
|
| 374 |
+
"learning_rate": 0.0001617226479697105,
|
| 375 |
+
"loss": 0.0193,
|
| 376 |
+
"mean_token_accuracy": 0.9917209327220917,
|
| 377 |
+
"step": 225
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.368,
|
| 381 |
+
"grad_norm": 0.04639686644077301,
|
| 382 |
+
"learning_rate": 0.00015949975229484134,
|
| 383 |
+
"loss": 0.0215,
|
| 384 |
+
"mean_token_accuracy": 0.9905673682689666,
|
| 385 |
+
"step": 230
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.376,
|
| 389 |
+
"grad_norm": 0.05585622414946556,
|
| 390 |
+
"learning_rate": 0.00015723037791621193,
|
| 391 |
+
"loss": 0.0194,
|
| 392 |
+
"mean_token_accuracy": 0.9916095018386841,
|
| 393 |
+
"step": 235
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.384,
|
| 397 |
+
"grad_norm": 0.053755246102809906,
|
| 398 |
+
"learning_rate": 0.00015491629757363032,
|
| 399 |
+
"loss": 0.0193,
|
| 400 |
+
"mean_token_accuracy": 0.9914996027946472,
|
| 401 |
+
"step": 240
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.392,
|
| 405 |
+
"grad_norm": 0.04922570288181305,
|
| 406 |
+
"learning_rate": 0.00015255931892932333,
|
| 407 |
+
"loss": 0.0196,
|
| 408 |
+
"mean_token_accuracy": 0.9919640719890594,
|
| 409 |
+
"step": 245
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4,
|
| 413 |
+
"grad_norm": 0.05820295959711075,
|
| 414 |
+
"learning_rate": 0.0001501612831558664,
|
| 415 |
+
"loss": 0.0213,
|
| 416 |
+
"mean_token_accuracy": 0.9906392514705658,
|
| 417 |
+
"step": 250
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.408,
|
| 421 |
+
"grad_norm": 0.04789372533559799,
|
| 422 |
+
"learning_rate": 0.00014772406349793744,
|
| 423 |
+
"loss": 0.0194,
|
| 424 |
+
"mean_token_accuracy": 0.9916912317276001,
|
| 425 |
+
"step": 255
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.416,
|
| 429 |
+
"grad_norm": 0.04380471259355545,
|
| 430 |
+
"learning_rate": 0.0001452495638090167,
|
| 431 |
+
"loss": 0.0185,
|
| 432 |
+
"mean_token_accuracy": 0.9918674886226654,
|
| 433 |
+
"step": 260
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.424,
|
| 437 |
+
"grad_norm": 0.04164614900946617,
|
| 438 |
+
"learning_rate": 0.00014273971706417647,
|
| 439 |
+
"loss": 0.0204,
|
| 440 |
+
"mean_token_accuracy": 0.990676760673523,
|
| 441 |
+
"step": 265
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.432,
|
| 445 |
+
"grad_norm": 0.0481300987303257,
|
| 446 |
+
"learning_rate": 0.00014019648385012244,
|
| 447 |
+
"loss": 0.0191,
|
| 448 |
+
"mean_token_accuracy": 0.9913807928562164,
|
| 449 |
+
"step": 270
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.44,
|
| 453 |
+
"grad_norm": 0.0358743742108345,
|
| 454 |
+
"learning_rate": 0.00013762185083366556,
|
| 455 |
+
"loss": 0.0193,
|
| 456 |
+
"mean_token_accuracy": 0.9915093362331391,
|
| 457 |
+
"step": 275
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.448,
|
| 461 |
+
"grad_norm": 0.04558572173118591,
|
| 462 |
+
"learning_rate": 0.00013501782920982184,
|
| 463 |
+
"loss": 0.021,
|
| 464 |
+
"mean_token_accuracy": 0.9905170977115632,
|
| 465 |
+
"step": 280
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.456,
|
| 469 |
+
"grad_norm": 0.050024427473545074,
|
| 470 |
+
"learning_rate": 0.00013238645313075104,
|
| 471 |
+
"loss": 0.0206,
|
| 472 |
+
"mean_token_accuracy": 0.9908523619174957,
|
| 473 |
+
"step": 285
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.464,
|
| 477 |
+
"grad_norm": 0.05017619952559471,
|
| 478 |
+
"learning_rate": 0.00012972977811676287,
|
| 479 |
+
"loss": 0.0176,
|
| 480 |
+
"mean_token_accuracy": 0.9921334087848663,
|
| 481 |
+
"step": 290
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.472,
|
| 485 |
+
"grad_norm": 0.03845011815428734,
|
| 486 |
+
"learning_rate": 0.00012704987945063068,
|
| 487 |
+
"loss": 0.0201,
|
| 488 |
+
"mean_token_accuracy": 0.9913134098052978,
|
| 489 |
+
"step": 295
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.48,
|
| 493 |
+
"grad_norm": 0.04298843815922737,
|
| 494 |
+
"learning_rate": 0.00012434885055646823,
|
| 495 |
+
"loss": 0.0175,
|
| 496 |
+
"mean_token_accuracy": 0.9922883987426758,
|
| 497 |
+
"step": 300
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.488,
|
| 501 |
+
"grad_norm": 0.05314357951283455,
|
| 502 |
+
"learning_rate": 0.00012162880136443447,
|
| 503 |
+
"loss": 0.0218,
|
| 504 |
+
"mean_token_accuracy": 0.990600174665451,
|
| 505 |
+
"step": 305
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.496,
|
| 509 |
+
"grad_norm": 0.042282719165086746,
|
| 510 |
+
"learning_rate": 0.00011889185666254506,
|
| 511 |
+
"loss": 0.0192,
|
| 512 |
+
"mean_token_accuracy": 0.991459196805954,
|
| 513 |
+
"step": 310
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.504,
|
| 517 |
+
"grad_norm": 0.05242412909865379,
|
| 518 |
+
"learning_rate": 0.00011614015443687722,
|
| 519 |
+
"loss": 0.0189,
|
| 520 |
+
"mean_token_accuracy": 0.9916914582252503,
|
| 521 |
+
"step": 315
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.512,
|
| 525 |
+
"grad_norm": 0.05002991855144501,
|
| 526 |
+
"learning_rate": 0.0001133758442014651,
|
| 527 |
+
"loss": 0.0181,
|
| 528 |
+
"mean_token_accuracy": 0.9920280575752258,
|
| 529 |
+
"step": 320
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.52,
|
| 533 |
+
"grad_norm": 0.04210834577679634,
|
| 534 |
+
"learning_rate": 0.00011060108531918971,
|
| 535 |
+
"loss": 0.0194,
|
| 536 |
+
"mean_token_accuracy": 0.9914444029331207,
|
| 537 |
+
"step": 325
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.528,
|
| 541 |
+
"grad_norm": 0.04244352877140045,
|
| 542 |
+
"learning_rate": 0.0001078180453149754,
|
| 543 |
+
"loss": 0.0185,
|
| 544 |
+
"mean_token_accuracy": 0.991835606098175,
|
| 545 |
+
"step": 330
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.536,
|
| 549 |
+
"grad_norm": 0.054087404161691666,
|
| 550 |
+
"learning_rate": 0.00010502889818261075,
|
| 551 |
+
"loss": 0.0205,
|
| 552 |
+
"mean_token_accuracy": 0.990981924533844,
|
| 553 |
+
"step": 335
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.544,
|
| 557 |
+
"grad_norm": 0.04749380052089691,
|
| 558 |
+
"learning_rate": 0.00010223582268651586,
|
| 559 |
+
"loss": 0.019,
|
| 560 |
+
"mean_token_accuracy": 0.9914271533489227,
|
| 561 |
+
"step": 340
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.552,
|
| 565 |
+
"grad_norm": 0.0451885461807251,
|
| 566 |
+
"learning_rate": 9.94410006597835e-05,
|
| 567 |
+
"loss": 0.0193,
|
| 568 |
+
"mean_token_accuracy": 0.99147869348526,
|
| 569 |
+
"step": 345
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.56,
|
| 573 |
+
"grad_norm": 0.04427170008420944,
|
| 574 |
+
"learning_rate": 9.66466152998226e-05,
|
| 575 |
+
"loss": 0.0171,
|
| 576 |
+
"mean_token_accuracy": 0.9927612245082855,
|
| 577 |
+
"step": 350
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.568,
|
| 581 |
+
"grad_norm": 0.04799705743789673,
|
| 582 |
+
"learning_rate": 9.385484946293637e-05,
|
| 583 |
+
"loss": 0.0206,
|
| 584 |
+
"mean_token_accuracy": 0.9906625807285309,
|
| 585 |
+
"step": 355
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.576,
|
| 589 |
+
"grad_norm": 0.04452817514538765,
|
| 590 |
+
"learning_rate": 9.106788395916678e-05,
|
| 591 |
+
"loss": 0.019,
|
| 592 |
+
"mean_token_accuracy": 0.9919753789901733,
|
| 593 |
+
"step": 360
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.584,
|
| 597 |
+
"grad_norm": 0.05037883296608925,
|
| 598 |
+
"learning_rate": 8.828789584873754e-05,
|
| 599 |
+
"loss": 0.0188,
|
| 600 |
+
"mean_token_accuracy": 0.9917163908481598,
|
| 601 |
+
"step": 365
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.592,
|
| 605 |
+
"grad_norm": 0.044616032391786575,
|
| 606 |
+
"learning_rate": 8.551705674142617e-05,
|
| 607 |
+
"loss": 0.0195,
|
| 608 |
+
"mean_token_accuracy": 0.9913720488548279,
|
| 609 |
+
"step": 370
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.6,
|
| 613 |
+
"grad_norm": 0.04669741541147232,
|
| 614 |
+
"learning_rate": 8.275753110019367e-05,
|
| 615 |
+
"loss": 0.0176,
|
| 616 |
+
"mean_token_accuracy": 0.9922231256961822,
|
| 617 |
+
"step": 375
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.608,
|
| 621 |
+
"grad_norm": 0.06772830337285995,
|
| 622 |
+
"learning_rate": 8.001147455039735e-05,
|
| 623 |
+
"loss": 0.0175,
|
| 624 |
+
"mean_token_accuracy": 0.9923631131649018,
|
| 625 |
+
"step": 380
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.616,
|
| 629 |
+
"grad_norm": 0.04503108188509941,
|
| 630 |
+
"learning_rate": 7.728103219590681e-05,
|
| 631 |
+
"loss": 0.0167,
|
| 632 |
+
"mean_token_accuracy": 0.9929349243640899,
|
| 633 |
+
"step": 385
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.624,
|
| 637 |
+
"grad_norm": 0.04398513212800026,
|
| 638 |
+
"learning_rate": 7.456833694343906e-05,
|
| 639 |
+
"loss": 0.0179,
|
| 640 |
+
"mean_token_accuracy": 0.9921640634536744,
|
| 641 |
+
"step": 390
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.632,
|
| 645 |
+
"grad_norm": 0.04398965463042259,
|
| 646 |
+
"learning_rate": 7.18755078364214e-05,
|
| 647 |
+
"loss": 0.0191,
|
| 648 |
+
"mean_token_accuracy": 0.9915706038475036,
|
| 649 |
+
"step": 395
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.64,
|
| 653 |
+
"grad_norm": 0.05448702350258827,
|
| 654 |
+
"learning_rate": 6.920464839968405e-05,
|
| 655 |
+
"loss": 0.017,
|
| 656 |
+
"mean_token_accuracy": 0.9924534082412719,
|
| 657 |
+
"step": 400
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.648,
|
| 661 |
+
"grad_norm": 0.050546955317258835,
|
| 662 |
+
"learning_rate": 6.65578449962749e-05,
|
| 663 |
+
"loss": 0.02,
|
| 664 |
+
"mean_token_accuracy": 0.9911172151565552,
|
| 665 |
+
"step": 405
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.656,
|
| 669 |
+
"grad_norm": 0.0460050068795681,
|
| 670 |
+
"learning_rate": 6.393716519768047e-05,
|
| 671 |
+
"loss": 0.0178,
|
| 672 |
+
"mean_token_accuracy": 0.9920965254306793,
|
| 673 |
+
"step": 410
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.664,
|
| 677 |
+
"grad_norm": 0.04446449503302574,
|
| 678 |
+
"learning_rate": 6.134465616872598e-05,
|
| 679 |
+
"loss": 0.0166,
|
| 680 |
+
"mean_token_accuracy": 0.9923894107341766,
|
| 681 |
+
"step": 415
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.672,
|
| 685 |
+
"grad_norm": 0.05464813858270645,
|
| 686 |
+
"learning_rate": 5.878234306841637e-05,
|
| 687 |
+
"loss": 0.0171,
|
| 688 |
+
"mean_token_accuracy": 0.9922880589962005,
|
| 689 |
+
"step": 420
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.68,
|
| 693 |
+
"grad_norm": 0.06052134186029434,
|
| 694 |
+
"learning_rate": 5.62522274679673e-05,
|
| 695 |
+
"loss": 0.0175,
|
| 696 |
+
"mean_token_accuracy": 0.9923413753509521,
|
| 697 |
+
"step": 425
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.688,
|
| 701 |
+
"grad_norm": 0.054314225912094116,
|
| 702 |
+
"learning_rate": 5.375628578726181e-05,
|
| 703 |
+
"loss": 0.0168,
|
| 704 |
+
"mean_token_accuracy": 0.9928416252136231,
|
| 705 |
+
"step": 430
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.696,
|
| 709 |
+
"grad_norm": 0.06336328387260437,
|
| 710 |
+
"learning_rate": 5.1296467750954314e-05,
|
| 711 |
+
"loss": 0.0172,
|
| 712 |
+
"mean_token_accuracy": 0.9924824059009552,
|
| 713 |
+
"step": 435
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.704,
|
| 717 |
+
"grad_norm": 0.060671087354421616,
|
| 718 |
+
"learning_rate": 4.8874694865427676e-05,
|
| 719 |
+
"loss": 0.0172,
|
| 720 |
+
"mean_token_accuracy": 0.9923681735992431,
|
| 721 |
+
"step": 440
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.712,
|
| 725 |
+
"grad_norm": 0.05438563972711563,
|
| 726 |
+
"learning_rate": 4.649285891779327e-05,
|
| 727 |
+
"loss": 0.0171,
|
| 728 |
+
"mean_token_accuracy": 0.9924930989742279,
|
| 729 |
+
"step": 445
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.72,
|
| 733 |
+
"grad_norm": 0.05433967337012291,
|
| 734 |
+
"learning_rate": 4.415282049810644e-05,
|
| 735 |
+
"loss": 0.0173,
|
| 736 |
+
"mean_token_accuracy": 0.9924486458301545,
|
| 737 |
+
"step": 450
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.728,
|
| 741 |
+
"grad_norm": 0.03932139277458191,
|
| 742 |
+
"learning_rate": 4.1856407545951834e-05,
|
| 743 |
+
"loss": 0.0182,
|
| 744 |
+
"mean_token_accuracy": 0.9917416453361512,
|
| 745 |
+
"step": 455
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.736,
|
| 749 |
+
"grad_norm": 0.07568760961294174,
|
| 750 |
+
"learning_rate": 3.9605413922533874e-05,
|
| 751 |
+
"loss": 0.0179,
|
| 752 |
+
"mean_token_accuracy": 0.9920516312122345,
|
| 753 |
+
"step": 460
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.744,
|
| 757 |
+
"grad_norm": 0.03854453191161156,
|
| 758 |
+
"learning_rate": 3.740159800938784e-05,
|
| 759 |
+
"loss": 0.0151,
|
| 760 |
+
"mean_token_accuracy": 0.9933602154254914,
|
| 761 |
+
"step": 465
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.752,
|
| 765 |
+
"grad_norm": 0.04567549750208855,
|
| 766 |
+
"learning_rate": 3.5246681334806175e-05,
|
| 767 |
+
"loss": 0.015,
|
| 768 |
+
"mean_token_accuracy": 0.9936224162578583,
|
| 769 |
+
"step": 470
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.76,
|
| 773 |
+
"grad_norm": 0.0444038100540638,
|
| 774 |
+
"learning_rate": 3.3142347229053015e-05,
|
| 775 |
+
"loss": 0.0182,
|
| 776 |
+
"mean_token_accuracy": 0.9919311761856079,
|
| 777 |
+
"step": 475
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.768,
|
| 781 |
+
"grad_norm": 0.04453401267528534,
|
| 782 |
+
"learning_rate": 3.109023950941736e-05,
|
| 783 |
+
"loss": 0.0162,
|
| 784 |
+
"mean_token_accuracy": 0.9927919328212738,
|
| 785 |
+
"step": 480
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.776,
|
| 789 |
+
"grad_norm": 0.050479914993047714,
|
| 790 |
+
"learning_rate": 2.909196119613218e-05,
|
| 791 |
+
"loss": 0.0178,
|
| 792 |
+
"mean_token_accuracy": 0.992197722196579,
|
| 793 |
+
"step": 485
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.784,
|
| 797 |
+
"grad_norm": 0.03968972712755203,
|
| 798 |
+
"learning_rate": 2.7149073260162416e-05,
|
| 799 |
+
"loss": 0.0174,
|
| 800 |
+
"mean_token_accuracy": 0.9922388553619385,
|
| 801 |
+
"step": 490
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.792,
|
| 805 |
+
"grad_norm": 0.0438266284763813,
|
| 806 |
+
"learning_rate": 2.5263093403840142e-05,
|
| 807 |
+
"loss": 0.0179,
|
| 808 |
+
"mean_token_accuracy": 0.9920966625213623,
|
| 809 |
+
"step": 495
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.8,
|
| 813 |
+
"grad_norm": 0.035676371306180954,
|
| 814 |
+
"learning_rate": 2.3435494875299314e-05,
|
| 815 |
+
"loss": 0.0169,
|
| 816 |
+
"mean_token_accuracy": 0.9925434589385986,
|
| 817 |
+
"step": 500
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.808,
|
| 821 |
+
"grad_norm": 0.04425378888845444,
|
| 822 |
+
"learning_rate": 2.166770531763633e-05,
|
| 823 |
+
"loss": 0.0164,
|
| 824 |
+
"mean_token_accuracy": 0.9927955329418182,
|
| 825 |
+
"step": 505
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.816,
|
| 829 |
+
"grad_norm": 0.03838867321610451,
|
| 830 |
+
"learning_rate": 1.9961105653695266e-05,
|
| 831 |
+
"loss": 0.0173,
|
| 832 |
+
"mean_token_accuracy": 0.9923167705535889,
|
| 833 |
+
"step": 510
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.824,
|
| 837 |
+
"grad_norm": 0.05161241069436073,
|
| 838 |
+
"learning_rate": 1.8317029007349085e-05,
|
| 839 |
+
"loss": 0.0179,
|
| 840 |
+
"mean_token_accuracy": 0.9918343544006347,
|
| 841 |
+
"step": 515
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.832,
|
| 845 |
+
"grad_norm": 0.04114331677556038,
|
| 846 |
+
"learning_rate": 1.6736759662119183e-05,
|
| 847 |
+
"loss": 0.0167,
|
| 848 |
+
"mean_token_accuracy": 0.9925995886325836,
|
| 849 |
+
"step": 520
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.84,
|
| 853 |
+
"grad_norm": 0.04665491729974747,
|
| 854 |
+
"learning_rate": 1.5221532057947419e-05,
|
| 855 |
+
"loss": 0.0169,
|
| 856 |
+
"mean_token_accuracy": 0.9925190150737763,
|
| 857 |
+
"step": 525
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.848,
|
| 861 |
+
"grad_norm": 0.0381510928273201,
|
| 862 |
+
"learning_rate": 1.3772529826903269e-05,
|
| 863 |
+
"loss": 0.0153,
|
| 864 |
+
"mean_token_accuracy": 0.9932702004909515,
|
| 865 |
+
"step": 530
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.856,
|
| 869 |
+
"grad_norm": 0.04139837622642517,
|
| 870 |
+
"learning_rate": 1.23908848685804e-05,
|
| 871 |
+
"loss": 0.016,
|
| 872 |
+
"mean_token_accuracy": 0.993154889345169,
|
| 873 |
+
"step": 535
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.864,
|
| 877 |
+
"grad_norm": 0.04805777966976166,
|
| 878 |
+
"learning_rate": 1.1077676465904208e-05,
|
| 879 |
+
"loss": 0.0172,
|
| 880 |
+
"mean_token_accuracy": 0.992431515455246,
|
| 881 |
+
"step": 540
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.872,
|
| 885 |
+
"grad_norm": 0.051602400839328766,
|
| 886 |
+
"learning_rate": 9.833930442041506e-06,
|
| 887 |
+
"loss": 0.0167,
|
| 888 |
+
"mean_token_accuracy": 0.9925341963768005,
|
| 889 |
+
"step": 545
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 0.88,
|
| 893 |
+
"grad_norm": 0.044222284108400345,
|
| 894 |
+
"learning_rate": 8.660618359070604e-06,
|
| 895 |
+
"loss": 0.0169,
|
| 896 |
+
"mean_token_accuracy": 0.99241544008255,
|
| 897 |
+
"step": 550
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.888,
|
| 901 |
+
"grad_norm": 0.038402166217565536,
|
| 902 |
+
"learning_rate": 7.558656759037797e-06,
|
| 903 |
+
"loss": 0.016,
|
| 904 |
+
"mean_token_accuracy": 0.9928807556629181,
|
| 905 |
+
"step": 555
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.896,
|
| 909 |
+
"grad_norm": 0.048669859766960144,
|
| 910 |
+
"learning_rate": 6.528906447993288e-06,
|
| 911 |
+
"loss": 0.0183,
|
| 912 |
+
"mean_token_accuracy": 0.9919456124305726,
|
| 913 |
+
"step": 560
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.904,
|
| 917 |
+
"grad_norm": 0.03222226724028587,
|
| 918 |
+
"learning_rate": 5.572171823565797e-06,
|
| 919 |
+
"loss": 0.0171,
|
| 920 |
+
"mean_token_accuracy": 0.9923150539398193,
|
| 921 |
+
"step": 565
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.912,
|
| 925 |
+
"grad_norm": 0.04830151051282883,
|
| 926 |
+
"learning_rate": 4.689200246600867e-06,
|
| 927 |
+
"loss": 0.0177,
|
| 928 |
+
"mean_token_accuracy": 0.991916960477829,
|
| 929 |
+
"step": 570
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.92,
|
| 933 |
+
"grad_norm": 0.038783151656389236,
|
| 934 |
+
"learning_rate": 3.880681457354118e-06,
|
| 935 |
+
"loss": 0.0158,
|
| 936 |
+
"mean_token_accuracy": 0.9929914236068725,
|
| 937 |
+
"step": 575
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.928,
|
| 941 |
+
"grad_norm": 0.03586906939744949,
|
| 942 |
+
"learning_rate": 3.1472470366950334e-06,
|
| 943 |
+
"loss": 0.0157,
|
| 944 |
+
"mean_token_accuracy": 0.9929950356483459,
|
| 945 |
+
"step": 580
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.936,
|
| 949 |
+
"grad_norm": 0.047617629170417786,
|
| 950 |
+
"learning_rate": 2.4894699127426367e-06,
|
| 951 |
+
"loss": 0.0167,
|
| 952 |
+
"mean_token_accuracy": 0.9928344428539276,
|
| 953 |
+
"step": 585
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.944,
|
| 957 |
+
"grad_norm": 0.03604697436094284,
|
| 958 |
+
"learning_rate": 1.907863913318153e-06,
|
| 959 |
+
"loss": 0.0161,
|
| 960 |
+
"mean_token_accuracy": 0.9927676558494568,
|
| 961 |
+
"step": 590
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.952,
|
| 965 |
+
"grad_norm": 0.04133192077279091,
|
| 966 |
+
"learning_rate": 1.4028833645643113e-06,
|
| 967 |
+
"loss": 0.0173,
|
| 968 |
+
"mean_token_accuracy": 0.9923523128032684,
|
| 969 |
+
"step": 595
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.96,
|
| 973 |
+
"grad_norm": 0.04649489372968674,
|
| 974 |
+
"learning_rate": 9.749227360448143e-07,
|
| 975 |
+
"loss": 0.0163,
|
| 976 |
+
"mean_token_accuracy": 0.9930159628391266,
|
| 977 |
+
"step": 600
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.968,
|
| 981 |
+
"grad_norm": 0.05105933919548988,
|
| 982 |
+
"learning_rate": 6.243163326014267e-07,
|
| 983 |
+
"loss": 0.0167,
|
| 984 |
+
"mean_token_accuracy": 0.9926475822925568,
|
| 985 |
+
"step": 605
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.976,
|
| 989 |
+
"grad_norm": 0.0360567681491375,
|
| 990 |
+
"learning_rate": 3.5133803320896994e-07,
|
| 991 |
+
"loss": 0.0162,
|
| 992 |
+
"mean_token_accuracy": 0.9927246868610382,
|
| 993 |
+
"step": 610
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.984,
|
| 997 |
+
"grad_norm": 0.050918541848659515,
|
| 998 |
+
"learning_rate": 1.562010770326916e-07,
|
| 999 |
+
"loss": 0.0184,
|
| 1000 |
+
"mean_token_accuracy": 0.9916555285453796,
|
| 1001 |
+
"step": 615
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.992,
|
| 1005 |
+
"grad_norm": 0.04419026896357536,
|
| 1006 |
+
"learning_rate": 3.905789685471062e-08,
|
| 1007 |
+
"loss": 0.0177,
|
| 1008 |
+
"mean_token_accuracy": 0.9921371281147003,
|
| 1009 |
+
"step": 620
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.0,
|
| 1013 |
+
"grad_norm": 0.04455499351024628,
|
| 1014 |
+
"learning_rate": 0.0,
|
| 1015 |
+
"loss": 0.0164,
|
| 1016 |
+
"mean_token_accuracy": 0.9929368138313294,
|
| 1017 |
+
"step": 625
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.0,
|
| 1021 |
+
"step": 625,
|
| 1022 |
+
"total_flos": 8.947190570627891e+16,
|
| 1023 |
+
"train_loss": 0.03501472600698471,
|
| 1024 |
+
"train_runtime": 1355.0569,
|
| 1025 |
+
"train_samples_per_second": 3.69,
|
| 1026 |
+
"train_steps_per_second": 0.461
|
| 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": 8.947190570627891e+16,
|
| 1047 |
+
"train_batch_size": 4,
|
| 1048 |
+
"trial_name": null,
|
| 1049 |
+
"trial_params": null
|
| 1050 |
+
}
|