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-correct-seq10k-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-correct-seq10k-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-correct-seq10k-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/5i8iiba5)
|
| 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": 9.533672832735642e+16,
|
| 3 |
+
"train_loss": 0.036805511808395384,
|
| 4 |
+
"train_runtime": 1380.4804,
|
| 5 |
+
"train_samples": 5000,
|
| 6 |
+
"train_samples_per_second": 3.622,
|
| 7 |
+
"train_steps_per_second": 0.453
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 9.533672832735642e+16,
|
| 3 |
+
"train_loss": 0.036805511808395384,
|
| 4 |
+
"train_runtime": 1380.4804,
|
| 5 |
+
"train_samples": 5000,
|
| 6 |
+
"train_samples_per_second": 3.622,
|
| 7 |
+
"train_steps_per_second": 0.453
|
| 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": 0.4117720127105713,
|
| 14 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 15 |
+
"loss": 0.3207,
|
| 16 |
+
"mean_token_accuracy": 0.9301324486732483,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.008,
|
| 21 |
+
"grad_norm": 0.4662795662879944,
|
| 22 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 23 |
+
"loss": 0.3764,
|
| 24 |
+
"mean_token_accuracy": 0.9112910404801369,
|
| 25 |
+
"step": 5
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.016,
|
| 29 |
+
"grad_norm": 0.3393407166004181,
|
| 30 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 31 |
+
"loss": 0.3561,
|
| 32 |
+
"mean_token_accuracy": 0.9210117816925049,
|
| 33 |
+
"step": 10
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.024,
|
| 37 |
+
"grad_norm": 0.20423269271850586,
|
| 38 |
+
"learning_rate": 4.761904761904762e-05,
|
| 39 |
+
"loss": 0.3355,
|
| 40 |
+
"mean_token_accuracy": 0.9193181097507477,
|
| 41 |
+
"step": 15
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.032,
|
| 45 |
+
"grad_norm": 0.19864961504936218,
|
| 46 |
+
"learning_rate": 6.349206349206349e-05,
|
| 47 |
+
"loss": 0.3258,
|
| 48 |
+
"mean_token_accuracy": 0.9193494200706482,
|
| 49 |
+
"step": 20
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.04,
|
| 53 |
+
"grad_norm": 0.12974287569522858,
|
| 54 |
+
"learning_rate": 7.936507936507937e-05,
|
| 55 |
+
"loss": 0.2594,
|
| 56 |
+
"mean_token_accuracy": 0.9317782700061799,
|
| 57 |
+
"step": 25
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.048,
|
| 61 |
+
"grad_norm": 0.15414533019065857,
|
| 62 |
+
"learning_rate": 9.523809523809524e-05,
|
| 63 |
+
"loss": 0.1998,
|
| 64 |
+
"mean_token_accuracy": 0.9463495194911957,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.056,
|
| 69 |
+
"grad_norm": 0.16060014069080353,
|
| 70 |
+
"learning_rate": 0.00011111111111111112,
|
| 71 |
+
"loss": 0.179,
|
| 72 |
+
"mean_token_accuracy": 0.9507739901542663,
|
| 73 |
+
"step": 35
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.064,
|
| 77 |
+
"grad_norm": 0.18361693620681763,
|
| 78 |
+
"learning_rate": 0.00012698412698412698,
|
| 79 |
+
"loss": 0.1489,
|
| 80 |
+
"mean_token_accuracy": 0.9504757285118103,
|
| 81 |
+
"step": 40
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.072,
|
| 85 |
+
"grad_norm": 0.11368059366941452,
|
| 86 |
+
"learning_rate": 0.00014285714285714287,
|
| 87 |
+
"loss": 0.0746,
|
| 88 |
+
"mean_token_accuracy": 0.9756224453449249,
|
| 89 |
+
"step": 45
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.08,
|
| 93 |
+
"grad_norm": 0.14619940519332886,
|
| 94 |
+
"learning_rate": 0.00015873015873015873,
|
| 95 |
+
"loss": 0.0563,
|
| 96 |
+
"mean_token_accuracy": 0.9806519865989685,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.088,
|
| 101 |
+
"grad_norm": 0.12718865275382996,
|
| 102 |
+
"learning_rate": 0.00017460317460317462,
|
| 103 |
+
"loss": 0.0531,
|
| 104 |
+
"mean_token_accuracy": 0.9808475732803345,
|
| 105 |
+
"step": 55
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.096,
|
| 109 |
+
"grad_norm": 0.0965471938252449,
|
| 110 |
+
"learning_rate": 0.00019047619047619048,
|
| 111 |
+
"loss": 0.0378,
|
| 112 |
+
"mean_token_accuracy": 0.9849684774875641,
|
| 113 |
+
"step": 60
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.104,
|
| 117 |
+
"grad_norm": 0.100978784263134,
|
| 118 |
+
"learning_rate": 0.00019999375039475277,
|
| 119 |
+
"loss": 0.03,
|
| 120 |
+
"mean_token_accuracy": 0.9881078660488128,
|
| 121 |
+
"step": 65
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.112,
|
| 125 |
+
"grad_norm": 0.07947579771280289,
|
| 126 |
+
"learning_rate": 0.0001999234513064475,
|
| 127 |
+
"loss": 0.0319,
|
| 128 |
+
"mean_token_accuracy": 0.9873642146587371,
|
| 129 |
+
"step": 70
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.12,
|
| 133 |
+
"grad_norm": 0.10587678849697113,
|
| 134 |
+
"learning_rate": 0.00019977509622105233,
|
| 135 |
+
"loss": 0.0316,
|
| 136 |
+
"mean_token_accuracy": 0.9869580447673798,
|
| 137 |
+
"step": 75
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.128,
|
| 141 |
+
"grad_norm": 0.07462529093027115,
|
| 142 |
+
"learning_rate": 0.0001995488010273198,
|
| 143 |
+
"loss": 0.0273,
|
| 144 |
+
"mean_token_accuracy": 0.9888291597366333,
|
| 145 |
+
"step": 80
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.136,
|
| 149 |
+
"grad_norm": 0.06621429324150085,
|
| 150 |
+
"learning_rate": 0.00019924474249753655,
|
| 151 |
+
"loss": 0.0273,
|
| 152 |
+
"mean_token_accuracy": 0.9885290026664734,
|
| 153 |
+
"step": 85
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.144,
|
| 157 |
+
"grad_norm": 0.07394655793905258,
|
| 158 |
+
"learning_rate": 0.00019886315814943647,
|
| 159 |
+
"loss": 0.0243,
|
| 160 |
+
"mean_token_accuracy": 0.9901373028755188,
|
| 161 |
+
"step": 90
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.152,
|
| 165 |
+
"grad_norm": 0.08377687633037567,
|
| 166 |
+
"learning_rate": 0.0001984043460606618,
|
| 167 |
+
"loss": 0.0246,
|
| 168 |
+
"mean_token_accuracy": 0.9900334179401398,
|
| 169 |
+
"step": 95
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.16,
|
| 173 |
+
"grad_norm": 0.09071408957242966,
|
| 174 |
+
"learning_rate": 0.0001978686646359173,
|
| 175 |
+
"loss": 0.0244,
|
| 176 |
+
"mean_token_accuracy": 0.9898565173149109,
|
| 177 |
+
"step": 100
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.168,
|
| 181 |
+
"grad_norm": 0.07183797657489777,
|
| 182 |
+
"learning_rate": 0.0001972565323269996,
|
| 183 |
+
"loss": 0.0236,
|
| 184 |
+
"mean_token_accuracy": 0.9900844812393188,
|
| 185 |
+
"step": 105
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.176,
|
| 189 |
+
"grad_norm": 0.08181659877300262,
|
| 190 |
+
"learning_rate": 0.00019656842730592046,
|
| 191 |
+
"loss": 0.0266,
|
| 192 |
+
"mean_token_accuracy": 0.9888815939426422,
|
| 193 |
+
"step": 110
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.184,
|
| 197 |
+
"grad_norm": 0.06551086157560349,
|
| 198 |
+
"learning_rate": 0.0001958048870913786,
|
| 199 |
+
"loss": 0.0249,
|
| 200 |
+
"mean_token_accuracy": 0.9896153032779693,
|
| 201 |
+
"step": 115
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.192,
|
| 205 |
+
"grad_norm": 0.09750600904226303,
|
| 206 |
+
"learning_rate": 0.0001949665081288729,
|
| 207 |
+
"loss": 0.0233,
|
| 208 |
+
"mean_token_accuracy": 0.9905028760433197,
|
| 209 |
+
"step": 120
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.2,
|
| 213 |
+
"grad_norm": 0.06609735637903214,
|
| 214 |
+
"learning_rate": 0.00019405394532478424,
|
| 215 |
+
"loss": 0.0225,
|
| 216 |
+
"mean_token_accuracy": 0.9906948924064636,
|
| 217 |
+
"step": 125
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.208,
|
| 221 |
+
"grad_norm": 0.05661633983254433,
|
| 222 |
+
"learning_rate": 0.00019306791153479006,
|
| 223 |
+
"loss": 0.0239,
|
| 224 |
+
"mean_token_accuracy": 0.9897183120250702,
|
| 225 |
+
"step": 130
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.216,
|
| 229 |
+
"grad_norm": 0.05774804577231407,
|
| 230 |
+
"learning_rate": 0.00019200917700701176,
|
| 231 |
+
"loss": 0.0226,
|
| 232 |
+
"mean_token_accuracy": 0.9906117141246795,
|
| 233 |
+
"step": 135
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.224,
|
| 237 |
+
"grad_norm": 0.058209504932165146,
|
| 238 |
+
"learning_rate": 0.0001908785687803289,
|
| 239 |
+
"loss": 0.025,
|
| 240 |
+
"mean_token_accuracy": 0.9893945038318634,
|
| 241 |
+
"step": 140
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.232,
|
| 245 |
+
"grad_norm": 0.07567816972732544,
|
| 246 |
+
"learning_rate": 0.00018967697003833157,
|
| 247 |
+
"loss": 0.0242,
|
| 248 |
+
"mean_token_accuracy": 0.9893796682357788,
|
| 249 |
+
"step": 145
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.24,
|
| 253 |
+
"grad_norm": 0.05945500358939171,
|
| 254 |
+
"learning_rate": 0.0001884053194194142,
|
| 255 |
+
"loss": 0.0213,
|
| 256 |
+
"mean_token_accuracy": 0.9913579225540161,
|
| 257 |
+
"step": 150
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.248,
|
| 261 |
+
"grad_norm": 0.0682721808552742,
|
| 262 |
+
"learning_rate": 0.00018706461028355104,
|
| 263 |
+
"loss": 0.0203,
|
| 264 |
+
"mean_token_accuracy": 0.9914861619472504,
|
| 265 |
+
"step": 155
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.256,
|
| 269 |
+
"grad_norm": 0.0625675767660141,
|
| 270 |
+
"learning_rate": 0.00018565588993632487,
|
| 271 |
+
"loss": 0.0202,
|
| 272 |
+
"mean_token_accuracy": 0.9916122198104859,
|
| 273 |
+
"step": 160
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.264,
|
| 277 |
+
"grad_norm": 0.06037646159529686,
|
| 278 |
+
"learning_rate": 0.0001841802588108161,
|
| 279 |
+
"loss": 0.0247,
|
| 280 |
+
"mean_token_accuracy": 0.9895594000816346,
|
| 281 |
+
"step": 165
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.272,
|
| 285 |
+
"grad_norm": 0.062324173748493195,
|
| 286 |
+
"learning_rate": 0.00018263886960799062,
|
| 287 |
+
"loss": 0.0198,
|
| 288 |
+
"mean_token_accuracy": 0.9916119515895844,
|
| 289 |
+
"step": 170
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.28,
|
| 293 |
+
"grad_norm": 0.06946054100990295,
|
| 294 |
+
"learning_rate": 0.00018103292639625837,
|
| 295 |
+
"loss": 0.0208,
|
| 296 |
+
"mean_token_accuracy": 0.9909637570381165,
|
| 297 |
+
"step": 175
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.288,
|
| 301 |
+
"grad_norm": 0.06305631995201111,
|
| 302 |
+
"learning_rate": 0.0001793636836709057,
|
| 303 |
+
"loss": 0.0197,
|
| 304 |
+
"mean_token_accuracy": 0.9915975213050843,
|
| 305 |
+
"step": 180
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.296,
|
| 309 |
+
"grad_norm": 0.06307779997587204,
|
| 310 |
+
"learning_rate": 0.0001776324453741365,
|
| 311 |
+
"loss": 0.0231,
|
| 312 |
+
"mean_token_accuracy": 0.9900566816329956,
|
| 313 |
+
"step": 185
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.304,
|
| 317 |
+
"grad_norm": 0.06790298223495483,
|
| 318 |
+
"learning_rate": 0.00017584056387648727,
|
| 319 |
+
"loss": 0.0209,
|
| 320 |
+
"mean_token_accuracy": 0.9912635624408722,
|
| 321 |
+
"step": 190
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.312,
|
| 325 |
+
"grad_norm": 0.06551709771156311,
|
| 326 |
+
"learning_rate": 0.0001739894389204122,
|
| 327 |
+
"loss": 0.0216,
|
| 328 |
+
"mean_token_accuracy": 0.9909388244152069,
|
| 329 |
+
"step": 195
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.32,
|
| 333 |
+
"grad_norm": 0.08183348178863525,
|
| 334 |
+
"learning_rate": 0.00017208051652686335,
|
| 335 |
+
"loss": 0.0192,
|
| 336 |
+
"mean_token_accuracy": 0.9920693576335907,
|
| 337 |
+
"step": 200
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.328,
|
| 341 |
+
"grad_norm": 0.06281828880310059,
|
| 342 |
+
"learning_rate": 0.00017011528786571969,
|
| 343 |
+
"loss": 0.021,
|
| 344 |
+
"mean_token_accuracy": 0.9912043333053588,
|
| 345 |
+
"step": 205
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.336,
|
| 349 |
+
"grad_norm": 0.041762325912714005,
|
| 350 |
+
"learning_rate": 0.00016809528809094807,
|
| 351 |
+
"loss": 0.0228,
|
| 352 |
+
"mean_token_accuracy": 0.9899637579917908,
|
| 353 |
+
"step": 210
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.344,
|
| 357 |
+
"grad_norm": 0.07002817094326019,
|
| 358 |
+
"learning_rate": 0.0001660220951414055,
|
| 359 |
+
"loss": 0.0195,
|
| 360 |
+
"mean_token_accuracy": 0.9913987815380096,
|
| 361 |
+
"step": 215
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.352,
|
| 365 |
+
"grad_norm": 0.054525118321180344,
|
| 366 |
+
"learning_rate": 0.00016389732850821966,
|
| 367 |
+
"loss": 0.023,
|
| 368 |
+
"mean_token_accuracy": 0.9904501914978028,
|
| 369 |
+
"step": 220
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.36,
|
| 373 |
+
"grad_norm": 0.10583920031785965,
|
| 374 |
+
"learning_rate": 0.0001617226479697105,
|
| 375 |
+
"loss": 0.0191,
|
| 376 |
+
"mean_token_accuracy": 0.991842246055603,
|
| 377 |
+
"step": 225
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.368,
|
| 381 |
+
"grad_norm": 0.06435126811265945,
|
| 382 |
+
"learning_rate": 0.00015949975229484134,
|
| 383 |
+
"loss": 0.0193,
|
| 384 |
+
"mean_token_accuracy": 0.9917158544063568,
|
| 385 |
+
"step": 230
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.376,
|
| 389 |
+
"grad_norm": 0.046176522970199585,
|
| 390 |
+
"learning_rate": 0.00015723037791621193,
|
| 391 |
+
"loss": 0.0207,
|
| 392 |
+
"mean_token_accuracy": 0.9910112500190735,
|
| 393 |
+
"step": 235
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.384,
|
| 397 |
+
"grad_norm": 0.0548773854970932,
|
| 398 |
+
"learning_rate": 0.00015491629757363032,
|
| 399 |
+
"loss": 0.0185,
|
| 400 |
+
"mean_token_accuracy": 0.9921935558319092,
|
| 401 |
+
"step": 240
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.392,
|
| 405 |
+
"grad_norm": 0.07670185714960098,
|
| 406 |
+
"learning_rate": 0.00015255931892932333,
|
| 407 |
+
"loss": 0.0215,
|
| 408 |
+
"mean_token_accuracy": 0.9907633364200592,
|
| 409 |
+
"step": 245
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4,
|
| 413 |
+
"grad_norm": 0.056587159633636475,
|
| 414 |
+
"learning_rate": 0.0001501612831558664,
|
| 415 |
+
"loss": 0.0206,
|
| 416 |
+
"mean_token_accuracy": 0.9911744177341462,
|
| 417 |
+
"step": 250
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.408,
|
| 421 |
+
"grad_norm": 0.04901133477687836,
|
| 422 |
+
"learning_rate": 0.00014772406349793744,
|
| 423 |
+
"loss": 0.0202,
|
| 424 |
+
"mean_token_accuracy": 0.991132628917694,
|
| 425 |
+
"step": 255
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.416,
|
| 429 |
+
"grad_norm": 0.056713420897722244,
|
| 430 |
+
"learning_rate": 0.0001452495638090167,
|
| 431 |
+
"loss": 0.0196,
|
| 432 |
+
"mean_token_accuracy": 0.991290944814682,
|
| 433 |
+
"step": 260
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.424,
|
| 437 |
+
"grad_norm": 0.04639092832803726,
|
| 438 |
+
"learning_rate": 0.00014273971706417647,
|
| 439 |
+
"loss": 0.0228,
|
| 440 |
+
"mean_token_accuracy": 0.9901349723339081,
|
| 441 |
+
"step": 265
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.432,
|
| 445 |
+
"grad_norm": 0.05444953590631485,
|
| 446 |
+
"learning_rate": 0.00014019648385012244,
|
| 447 |
+
"loss": 0.0212,
|
| 448 |
+
"mean_token_accuracy": 0.9904214441776276,
|
| 449 |
+
"step": 270
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.44,
|
| 453 |
+
"grad_norm": 0.046870455145835876,
|
| 454 |
+
"learning_rate": 0.00013762185083366556,
|
| 455 |
+
"loss": 0.0195,
|
| 456 |
+
"mean_token_accuracy": 0.9917024731636047,
|
| 457 |
+
"step": 275
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.448,
|
| 461 |
+
"grad_norm": 0.05092842876911163,
|
| 462 |
+
"learning_rate": 0.00013501782920982184,
|
| 463 |
+
"loss": 0.02,
|
| 464 |
+
"mean_token_accuracy": 0.9912722766399383,
|
| 465 |
+
"step": 280
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.456,
|
| 469 |
+
"grad_norm": 0.03894677385687828,
|
| 470 |
+
"learning_rate": 0.00013238645313075104,
|
| 471 |
+
"loss": 0.0182,
|
| 472 |
+
"mean_token_accuracy": 0.992118376493454,
|
| 473 |
+
"step": 285
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.464,
|
| 477 |
+
"grad_norm": 0.047839269042015076,
|
| 478 |
+
"learning_rate": 0.00012972977811676287,
|
| 479 |
+
"loss": 0.0188,
|
| 480 |
+
"mean_token_accuracy": 0.9918950438499451,
|
| 481 |
+
"step": 290
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.472,
|
| 485 |
+
"grad_norm": 0.0430167056620121,
|
| 486 |
+
"learning_rate": 0.00012704987945063068,
|
| 487 |
+
"loss": 0.0185,
|
| 488 |
+
"mean_token_accuracy": 0.992196011543274,
|
| 489 |
+
"step": 295
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.48,
|
| 493 |
+
"grad_norm": 0.05636301264166832,
|
| 494 |
+
"learning_rate": 0.00012434885055646823,
|
| 495 |
+
"loss": 0.0203,
|
| 496 |
+
"mean_token_accuracy": 0.9910727739334106,
|
| 497 |
+
"step": 300
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.488,
|
| 501 |
+
"grad_norm": 0.04330025613307953,
|
| 502 |
+
"learning_rate": 0.00012162880136443447,
|
| 503 |
+
"loss": 0.0183,
|
| 504 |
+
"mean_token_accuracy": 0.9922338902950287,
|
| 505 |
+
"step": 305
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.496,
|
| 509 |
+
"grad_norm": 0.05060559883713722,
|
| 510 |
+
"learning_rate": 0.00011889185666254506,
|
| 511 |
+
"loss": 0.0186,
|
| 512 |
+
"mean_token_accuracy": 0.9921160638332367,
|
| 513 |
+
"step": 310
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.504,
|
| 517 |
+
"grad_norm": 0.047406505793333054,
|
| 518 |
+
"learning_rate": 0.00011614015443687722,
|
| 519 |
+
"loss": 0.0163,
|
| 520 |
+
"mean_token_accuracy": 0.9927721679210663,
|
| 521 |
+
"step": 315
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.512,
|
| 525 |
+
"grad_norm": 0.04101146012544632,
|
| 526 |
+
"learning_rate": 0.0001133758442014651,
|
| 527 |
+
"loss": 0.0184,
|
| 528 |
+
"mean_token_accuracy": 0.9921581447124481,
|
| 529 |
+
"step": 320
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.52,
|
| 533 |
+
"grad_norm": 0.05084632709622383,
|
| 534 |
+
"learning_rate": 0.00011060108531918971,
|
| 535 |
+
"loss": 0.0185,
|
| 536 |
+
"mean_token_accuracy": 0.9916965663433075,
|
| 537 |
+
"step": 325
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.528,
|
| 541 |
+
"grad_norm": 0.08331380039453506,
|
| 542 |
+
"learning_rate": 0.0001078180453149754,
|
| 543 |
+
"loss": 0.019,
|
| 544 |
+
"mean_token_accuracy": 0.9916886448860168,
|
| 545 |
+
"step": 330
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.536,
|
| 549 |
+
"grad_norm": 0.0668216124176979,
|
| 550 |
+
"learning_rate": 0.00010502889818261075,
|
| 551 |
+
"loss": 0.02,
|
| 552 |
+
"mean_token_accuracy": 0.9912381112575531,
|
| 553 |
+
"step": 335
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.544,
|
| 557 |
+
"grad_norm": 0.04514472931623459,
|
| 558 |
+
"learning_rate": 0.00010223582268651586,
|
| 559 |
+
"loss": 0.0179,
|
| 560 |
+
"mean_token_accuracy": 0.9919449985027313,
|
| 561 |
+
"step": 340
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.552,
|
| 565 |
+
"grad_norm": 0.07452940195798874,
|
| 566 |
+
"learning_rate": 9.94410006597835e-05,
|
| 567 |
+
"loss": 0.0189,
|
| 568 |
+
"mean_token_accuracy": 0.9916715919971466,
|
| 569 |
+
"step": 345
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.56,
|
| 573 |
+
"grad_norm": 0.10710987448692322,
|
| 574 |
+
"learning_rate": 9.66466152998226e-05,
|
| 575 |
+
"loss": 0.0168,
|
| 576 |
+
"mean_token_accuracy": 0.9927741289138794,
|
| 577 |
+
"step": 350
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.568,
|
| 581 |
+
"grad_norm": 0.04447026550769806,
|
| 582 |
+
"learning_rate": 9.385484946293637e-05,
|
| 583 |
+
"loss": 0.019,
|
| 584 |
+
"mean_token_accuracy": 0.9916839182376862,
|
| 585 |
+
"step": 355
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.576,
|
| 589 |
+
"grad_norm": 0.03214811906218529,
|
| 590 |
+
"learning_rate": 9.106788395916678e-05,
|
| 591 |
+
"loss": 0.0154,
|
| 592 |
+
"mean_token_accuracy": 0.9934532225131989,
|
| 593 |
+
"step": 360
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.584,
|
| 597 |
+
"grad_norm": 0.039207059890031815,
|
| 598 |
+
"learning_rate": 8.828789584873754e-05,
|
| 599 |
+
"loss": 0.0191,
|
| 600 |
+
"mean_token_accuracy": 0.9917472124099731,
|
| 601 |
+
"step": 365
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.592,
|
| 605 |
+
"grad_norm": 0.04347281903028488,
|
| 606 |
+
"learning_rate": 8.551705674142617e-05,
|
| 607 |
+
"loss": 0.019,
|
| 608 |
+
"mean_token_accuracy": 0.9917202949523926,
|
| 609 |
+
"step": 370
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.6,
|
| 613 |
+
"grad_norm": 0.049683503806591034,
|
| 614 |
+
"learning_rate": 8.275753110019367e-05,
|
| 615 |
+
"loss": 0.0164,
|
| 616 |
+
"mean_token_accuracy": 0.9931604623794555,
|
| 617 |
+
"step": 375
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.608,
|
| 621 |
+
"grad_norm": 0.045603714883327484,
|
| 622 |
+
"learning_rate": 8.001147455039735e-05,
|
| 623 |
+
"loss": 0.0201,
|
| 624 |
+
"mean_token_accuracy": 0.990942370891571,
|
| 625 |
+
"step": 380
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.616,
|
| 629 |
+
"grad_norm": 0.048600271344184875,
|
| 630 |
+
"learning_rate": 7.728103219590681e-05,
|
| 631 |
+
"loss": 0.0167,
|
| 632 |
+
"mean_token_accuracy": 0.9930158495903015,
|
| 633 |
+
"step": 385
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.624,
|
| 637 |
+
"grad_norm": 0.05142870917916298,
|
| 638 |
+
"learning_rate": 7.456833694343906e-05,
|
| 639 |
+
"loss": 0.0195,
|
| 640 |
+
"mean_token_accuracy": 0.991544383764267,
|
| 641 |
+
"step": 390
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.632,
|
| 645 |
+
"grad_norm": 0.053054407238960266,
|
| 646 |
+
"learning_rate": 7.18755078364214e-05,
|
| 647 |
+
"loss": 0.0175,
|
| 648 |
+
"mean_token_accuracy": 0.9922033488750458,
|
| 649 |
+
"step": 395
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.64,
|
| 653 |
+
"grad_norm": 0.04487873613834381,
|
| 654 |
+
"learning_rate": 6.920464839968405e-05,
|
| 655 |
+
"loss": 0.0149,
|
| 656 |
+
"mean_token_accuracy": 0.9939690411090851,
|
| 657 |
+
"step": 400
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.648,
|
| 661 |
+
"grad_norm": 0.07014092803001404,
|
| 662 |
+
"learning_rate": 6.65578449962749e-05,
|
| 663 |
+
"loss": 0.0179,
|
| 664 |
+
"mean_token_accuracy": 0.9924103438854217,
|
| 665 |
+
"step": 405
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.656,
|
| 669 |
+
"grad_norm": 0.0414702370762825,
|
| 670 |
+
"learning_rate": 6.393716519768047e-05,
|
| 671 |
+
"loss": 0.0185,
|
| 672 |
+
"mean_token_accuracy": 0.9919926702976227,
|
| 673 |
+
"step": 410
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.664,
|
| 677 |
+
"grad_norm": 0.03618001565337181,
|
| 678 |
+
"learning_rate": 6.134465616872598e-05,
|
| 679 |
+
"loss": 0.0162,
|
| 680 |
+
"mean_token_accuracy": 0.9932132542133332,
|
| 681 |
+
"step": 415
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.672,
|
| 685 |
+
"grad_norm": 0.03933430090546608,
|
| 686 |
+
"learning_rate": 5.878234306841637e-05,
|
| 687 |
+
"loss": 0.0178,
|
| 688 |
+
"mean_token_accuracy": 0.9921888768672943,
|
| 689 |
+
"step": 420
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.68,
|
| 693 |
+
"grad_norm": 0.04353130981326103,
|
| 694 |
+
"learning_rate": 5.62522274679673e-05,
|
| 695 |
+
"loss": 0.0162,
|
| 696 |
+
"mean_token_accuracy": 0.993138599395752,
|
| 697 |
+
"step": 425
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.688,
|
| 701 |
+
"grad_norm": 0.03876515105366707,
|
| 702 |
+
"learning_rate": 5.375628578726181e-05,
|
| 703 |
+
"loss": 0.0174,
|
| 704 |
+
"mean_token_accuracy": 0.9924117088317871,
|
| 705 |
+
"step": 430
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.696,
|
| 709 |
+
"grad_norm": 0.05640017241239548,
|
| 710 |
+
"learning_rate": 5.1296467750954314e-05,
|
| 711 |
+
"loss": 0.0181,
|
| 712 |
+
"mean_token_accuracy": 0.992024153470993,
|
| 713 |
+
"step": 435
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.704,
|
| 717 |
+
"grad_norm": 0.043328192085027695,
|
| 718 |
+
"learning_rate": 4.8874694865427676e-05,
|
| 719 |
+
"loss": 0.0176,
|
| 720 |
+
"mean_token_accuracy": 0.9921295821666718,
|
| 721 |
+
"step": 440
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.712,
|
| 725 |
+
"grad_norm": 0.06356772780418396,
|
| 726 |
+
"learning_rate": 4.649285891779327e-05,
|
| 727 |
+
"loss": 0.0173,
|
| 728 |
+
"mean_token_accuracy": 0.992793595790863,
|
| 729 |
+
"step": 445
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.72,
|
| 733 |
+
"grad_norm": 0.04712694138288498,
|
| 734 |
+
"learning_rate": 4.415282049810644e-05,
|
| 735 |
+
"loss": 0.0179,
|
| 736 |
+
"mean_token_accuracy": 0.9918626189231873,
|
| 737 |
+
"step": 450
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.728,
|
| 741 |
+
"grad_norm": 0.04961908608675003,
|
| 742 |
+
"learning_rate": 4.1856407545951834e-05,
|
| 743 |
+
"loss": 0.0168,
|
| 744 |
+
"mean_token_accuracy": 0.9926211416721344,
|
| 745 |
+
"step": 455
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.736,
|
| 749 |
+
"grad_norm": 0.05269668623805046,
|
| 750 |
+
"learning_rate": 3.9605413922533874e-05,
|
| 751 |
+
"loss": 0.0187,
|
| 752 |
+
"mean_token_accuracy": 0.9921265006065368,
|
| 753 |
+
"step": 460
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.744,
|
| 757 |
+
"grad_norm": 0.05327823385596275,
|
| 758 |
+
"learning_rate": 3.740159800938784e-05,
|
| 759 |
+
"loss": 0.0202,
|
| 760 |
+
"mean_token_accuracy": 0.9910061419010162,
|
| 761 |
+
"step": 465
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.752,
|
| 765 |
+
"grad_norm": 0.03653375431895256,
|
| 766 |
+
"learning_rate": 3.5246681334806175e-05,
|
| 767 |
+
"loss": 0.017,
|
| 768 |
+
"mean_token_accuracy": 0.9925118863582612,
|
| 769 |
+
"step": 470
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.76,
|
| 773 |
+
"grad_norm": 0.039025988429784775,
|
| 774 |
+
"learning_rate": 3.3142347229053015e-05,
|
| 775 |
+
"loss": 0.0168,
|
| 776 |
+
"mean_token_accuracy": 0.9928381443023682,
|
| 777 |
+
"step": 475
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.768,
|
| 781 |
+
"grad_norm": 0.047734424471855164,
|
| 782 |
+
"learning_rate": 3.109023950941736e-05,
|
| 783 |
+
"loss": 0.0169,
|
| 784 |
+
"mean_token_accuracy": 0.9921681106090545,
|
| 785 |
+
"step": 480
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.776,
|
| 789 |
+
"grad_norm": 0.03646542876958847,
|
| 790 |
+
"learning_rate": 2.909196119613218e-05,
|
| 791 |
+
"loss": 0.0161,
|
| 792 |
+
"mean_token_accuracy": 0.9926591336727142,
|
| 793 |
+
"step": 485
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.784,
|
| 797 |
+
"grad_norm": 0.059171825647354126,
|
| 798 |
+
"learning_rate": 2.7149073260162416e-05,
|
| 799 |
+
"loss": 0.0167,
|
| 800 |
+
"mean_token_accuracy": 0.9927067875862121,
|
| 801 |
+
"step": 490
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.792,
|
| 805 |
+
"grad_norm": 0.03208303824067116,
|
| 806 |
+
"learning_rate": 2.5263093403840142e-05,
|
| 807 |
+
"loss": 0.0163,
|
| 808 |
+
"mean_token_accuracy": 0.9924599826335907,
|
| 809 |
+
"step": 495
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.8,
|
| 813 |
+
"grad_norm": 0.04300253465771675,
|
| 814 |
+
"learning_rate": 2.3435494875299314e-05,
|
| 815 |
+
"loss": 0.0163,
|
| 816 |
+
"mean_token_accuracy": 0.9925986528396606,
|
| 817 |
+
"step": 500
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.808,
|
| 821 |
+
"grad_norm": 0.05904829129576683,
|
| 822 |
+
"learning_rate": 2.166770531763633e-05,
|
| 823 |
+
"loss": 0.0181,
|
| 824 |
+
"mean_token_accuracy": 0.9921526372432709,
|
| 825 |
+
"step": 505
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.816,
|
| 829 |
+
"grad_norm": 0.04240652918815613,
|
| 830 |
+
"learning_rate": 1.9961105653695266e-05,
|
| 831 |
+
"loss": 0.0173,
|
| 832 |
+
"mean_token_accuracy": 0.9926676869392395,
|
| 833 |
+
"step": 510
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.824,
|
| 837 |
+
"grad_norm": 0.062428414821624756,
|
| 838 |
+
"learning_rate": 1.8317029007349085e-05,
|
| 839 |
+
"loss": 0.0175,
|
| 840 |
+
"mean_token_accuracy": 0.9925188779830932,
|
| 841 |
+
"step": 515
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.832,
|
| 845 |
+
"grad_norm": 0.037988096475601196,
|
| 846 |
+
"learning_rate": 1.6736759662119183e-05,
|
| 847 |
+
"loss": 0.0147,
|
| 848 |
+
"mean_token_accuracy": 0.9932997107505799,
|
| 849 |
+
"step": 520
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.84,
|
| 853 |
+
"grad_norm": 0.03455980122089386,
|
| 854 |
+
"learning_rate": 1.5221532057947419e-05,
|
| 855 |
+
"loss": 0.0157,
|
| 856 |
+
"mean_token_accuracy": 0.9931821465492249,
|
| 857 |
+
"step": 525
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.848,
|
| 861 |
+
"grad_norm": 0.05244341120123863,
|
| 862 |
+
"learning_rate": 1.3772529826903269e-05,
|
| 863 |
+
"loss": 0.017,
|
| 864 |
+
"mean_token_accuracy": 0.9925680100917816,
|
| 865 |
+
"step": 530
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.856,
|
| 869 |
+
"grad_norm": 0.06056361645460129,
|
| 870 |
+
"learning_rate": 1.23908848685804e-05,
|
| 871 |
+
"loss": 0.0175,
|
| 872 |
+
"mean_token_accuracy": 0.9920696496963501,
|
| 873 |
+
"step": 535
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.864,
|
| 877 |
+
"grad_norm": 0.030792532488703728,
|
| 878 |
+
"learning_rate": 1.1077676465904208e-05,
|
| 879 |
+
"loss": 0.0148,
|
| 880 |
+
"mean_token_accuracy": 0.993705016374588,
|
| 881 |
+
"step": 540
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.872,
|
| 885 |
+
"grad_norm": 0.03837195783853531,
|
| 886 |
+
"learning_rate": 9.833930442041506e-06,
|
| 887 |
+
"loss": 0.0153,
|
| 888 |
+
"mean_token_accuracy": 0.993491017818451,
|
| 889 |
+
"step": 545
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 0.88,
|
| 893 |
+
"grad_norm": 0.04586377367377281,
|
| 894 |
+
"learning_rate": 8.660618359070604e-06,
|
| 895 |
+
"loss": 0.0177,
|
| 896 |
+
"mean_token_accuracy": 0.9921986818313598,
|
| 897 |
+
"step": 550
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.888,
|
| 901 |
+
"grad_norm": 0.0402282290160656,
|
| 902 |
+
"learning_rate": 7.558656759037797e-06,
|
| 903 |
+
"loss": 0.0169,
|
| 904 |
+
"mean_token_accuracy": 0.9928896963596344,
|
| 905 |
+
"step": 555
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.896,
|
| 909 |
+
"grad_norm": 0.04727677255868912,
|
| 910 |
+
"learning_rate": 6.528906447993288e-06,
|
| 911 |
+
"loss": 0.0181,
|
| 912 |
+
"mean_token_accuracy": 0.99213707447052,
|
| 913 |
+
"step": 560
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.904,
|
| 917 |
+
"grad_norm": 0.04939686879515648,
|
| 918 |
+
"learning_rate": 5.572171823565797e-06,
|
| 919 |
+
"loss": 0.0172,
|
| 920 |
+
"mean_token_accuracy": 0.9930021226406097,
|
| 921 |
+
"step": 565
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.912,
|
| 925 |
+
"grad_norm": 0.04446263611316681,
|
| 926 |
+
"learning_rate": 4.689200246600867e-06,
|
| 927 |
+
"loss": 0.0179,
|
| 928 |
+
"mean_token_accuracy": 0.992498242855072,
|
| 929 |
+
"step": 570
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.92,
|
| 933 |
+
"grad_norm": 0.050484731793403625,
|
| 934 |
+
"learning_rate": 3.880681457354118e-06,
|
| 935 |
+
"loss": 0.0186,
|
| 936 |
+
"mean_token_accuracy": 0.9920108139514923,
|
| 937 |
+
"step": 575
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.928,
|
| 941 |
+
"grad_norm": 0.0397685170173645,
|
| 942 |
+
"learning_rate": 3.1472470366950334e-06,
|
| 943 |
+
"loss": 0.0196,
|
| 944 |
+
"mean_token_accuracy": 0.9915927886962891,
|
| 945 |
+
"step": 580
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.936,
|
| 949 |
+
"grad_norm": 0.04176720976829529,
|
| 950 |
+
"learning_rate": 2.4894699127426367e-06,
|
| 951 |
+
"loss": 0.0174,
|
| 952 |
+
"mean_token_accuracy": 0.9924736022949219,
|
| 953 |
+
"step": 585
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.944,
|
| 957 |
+
"grad_norm": 0.048351410776376724,
|
| 958 |
+
"learning_rate": 1.907863913318153e-06,
|
| 959 |
+
"loss": 0.0166,
|
| 960 |
+
"mean_token_accuracy": 0.992824774980545,
|
| 961 |
+
"step": 590
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.952,
|
| 965 |
+
"grad_norm": 0.034784235060214996,
|
| 966 |
+
"learning_rate": 1.4028833645643113e-06,
|
| 967 |
+
"loss": 0.0167,
|
| 968 |
+
"mean_token_accuracy": 0.9926623046398163,
|
| 969 |
+
"step": 595
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.96,
|
| 973 |
+
"grad_norm": 0.03560305014252663,
|
| 974 |
+
"learning_rate": 9.749227360448143e-07,
|
| 975 |
+
"loss": 0.0156,
|
| 976 |
+
"mean_token_accuracy": 0.9932109773159027,
|
| 977 |
+
"step": 600
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.968,
|
| 981 |
+
"grad_norm": 0.04202045872807503,
|
| 982 |
+
"learning_rate": 6.243163326014267e-07,
|
| 983 |
+
"loss": 0.0188,
|
| 984 |
+
"mean_token_accuracy": 0.9912631869316101,
|
| 985 |
+
"step": 605
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.976,
|
| 989 |
+
"grad_norm": 0.059313174337148666,
|
| 990 |
+
"learning_rate": 3.5133803320896994e-07,
|
| 991 |
+
"loss": 0.0183,
|
| 992 |
+
"mean_token_accuracy": 0.9916300714015961,
|
| 993 |
+
"step": 610
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.984,
|
| 997 |
+
"grad_norm": 0.03327292576432228,
|
| 998 |
+
"learning_rate": 1.562010770326916e-07,
|
| 999 |
+
"loss": 0.0187,
|
| 1000 |
+
"mean_token_accuracy": 0.9919956386089325,
|
| 1001 |
+
"step": 615
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.992,
|
| 1005 |
+
"grad_norm": 0.05758478492498398,
|
| 1006 |
+
"learning_rate": 3.905789685471062e-08,
|
| 1007 |
+
"loss": 0.0155,
|
| 1008 |
+
"mean_token_accuracy": 0.9931742966175079,
|
| 1009 |
+
"step": 620
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.0,
|
| 1013 |
+
"grad_norm": 0.036703914403915405,
|
| 1014 |
+
"learning_rate": 0.0,
|
| 1015 |
+
"loss": 0.0158,
|
| 1016 |
+
"mean_token_accuracy": 0.9930949747562409,
|
| 1017 |
+
"step": 625
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.0,
|
| 1021 |
+
"step": 625,
|
| 1022 |
+
"total_flos": 9.533672832735642e+16,
|
| 1023 |
+
"train_loss": 0.036805511808395384,
|
| 1024 |
+
"train_runtime": 1380.4804,
|
| 1025 |
+
"train_samples_per_second": 3.622,
|
| 1026 |
+
"train_steps_per_second": 0.453
|
| 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": 9.533672832735642e+16,
|
| 1047 |
+
"train_batch_size": 4,
|
| 1048 |
+
"trial_name": null,
|
| 1049 |
+
"trial_params": null
|
| 1050 |
+
}
|