Model save
Browse files- README.md +68 -0
- all_results.json +8 -0
- config.json +1 -1
- generation_config.json +6 -0
- train_results.json +8 -0
- trainer_state.json +1667 -0
README.md
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen2.5-Coder-7B
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Qwen2.5-Coder-7B
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- grpo
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Qwen2.5-Coder-7B
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-Coder-7B](https://huggingface.co/Qwen/Qwen2.5-Coder-7B).
|
| 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="zerozeroz/Qwen2.5-Coder-7B", 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 |
+
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.14.0
|
| 38 |
+
- Transformers: 4.48.1
|
| 39 |
+
- Pytorch: 2.5.1+cu121
|
| 40 |
+
- Datasets: 3.1.0
|
| 41 |
+
- Tokenizers: 0.21.0
|
| 42 |
+
|
| 43 |
+
## Citations
|
| 44 |
+
|
| 45 |
+
Cite GRPO as:
|
| 46 |
+
|
| 47 |
+
```bibtex
|
| 48 |
+
@article{zhihong2024deepseekmath,
|
| 49 |
+
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
|
| 50 |
+
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
|
| 51 |
+
year = 2024,
|
| 52 |
+
eprint = {arXiv:2402.03300},
|
| 53 |
+
}
|
| 54 |
+
|
| 55 |
+
```
|
| 56 |
+
|
| 57 |
+
Cite TRL as:
|
| 58 |
+
|
| 59 |
+
```bibtex
|
| 60 |
+
@misc{vonwerra2022trl,
|
| 61 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 62 |
+
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},
|
| 63 |
+
year = 2020,
|
| 64 |
+
journal = {GitHub repository},
|
| 65 |
+
publisher = {GitHub},
|
| 66 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 67 |
+
}
|
| 68 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 2.596636219443127e-05,
|
| 4 |
+
"train_runtime": 4869.9523,
|
| 5 |
+
"train_samples": 374,
|
| 6 |
+
"train_samples_per_second": 0.154,
|
| 7 |
+
"train_steps_per_second": 0.026
|
| 8 |
+
}
|
config.json
CHANGED
|
@@ -23,7 +23,7 @@
|
|
| 23 |
"tie_word_embeddings": false,
|
| 24 |
"torch_dtype": "bfloat16",
|
| 25 |
"transformers_version": "4.48.1",
|
| 26 |
-
"use_cache":
|
| 27 |
"use_sliding_window": false,
|
| 28 |
"vocab_size": 152064
|
| 29 |
}
|
|
|
|
| 23 |
"tie_word_embeddings": false,
|
| 24 |
"torch_dtype": "bfloat16",
|
| 25 |
"transformers_version": "4.48.1",
|
| 26 |
+
"use_cache": true,
|
| 27 |
"use_sliding_window": false,
|
| 28 |
"vocab_size": 152064
|
| 29 |
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.48.1"
|
| 6 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 2.596636219443127e-05,
|
| 4 |
+
"train_runtime": 4869.9523,
|
| 5 |
+
"train_samples": 374,
|
| 6 |
+
"train_samples_per_second": 0.154,
|
| 7 |
+
"train_steps_per_second": 0.026
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1667 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.992,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 125,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"completion_length": 142.08333587646484,
|
| 13 |
+
"epoch": 0.016,
|
| 14 |
+
"grad_norm": 0.9348842823066913,
|
| 15 |
+
"kl": 0.0,
|
| 16 |
+
"learning_rate": 1.25e-07,
|
| 17 |
+
"loss": -0.0,
|
| 18 |
+
"reward": 0.6988552212715149,
|
| 19 |
+
"reward_std": 0.5108021795749664,
|
| 20 |
+
"rewards/correct_code_reward_func": 0.4583333432674408,
|
| 21 |
+
"rewards/len_reward_func": 0.24052191525697708,
|
| 22 |
+
"step": 1
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"completion_length": 126.56250762939453,
|
| 26 |
+
"epoch": 0.032,
|
| 27 |
+
"grad_norm": 0.8243337506143691,
|
| 28 |
+
"kl": 0.0,
|
| 29 |
+
"learning_rate": 2.5e-07,
|
| 30 |
+
"loss": -0.0,
|
| 31 |
+
"reward": 0.6203328222036362,
|
| 32 |
+
"reward_std": 0.45215627551078796,
|
| 33 |
+
"rewards/correct_code_reward_func": 0.3541666865348816,
|
| 34 |
+
"rewards/len_reward_func": 0.26616617292165756,
|
| 35 |
+
"step": 2
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"completion_length": 108.79166793823242,
|
| 39 |
+
"epoch": 0.048,
|
| 40 |
+
"grad_norm": 0.9128499702856003,
|
| 41 |
+
"kl": 7.30752944946289e-05,
|
| 42 |
+
"learning_rate": 3.75e-07,
|
| 43 |
+
"loss": 0.0,
|
| 44 |
+
"reward": 0.6249927878379822,
|
| 45 |
+
"reward_std": 0.5788741409778595,
|
| 46 |
+
"rewards/correct_code_reward_func": 0.3958333432674408,
|
| 47 |
+
"rewards/len_reward_func": 0.22915946692228317,
|
| 48 |
+
"step": 3
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"completion_length": 142.2291717529297,
|
| 52 |
+
"epoch": 0.064,
|
| 53 |
+
"grad_norm": 0.9975440173663436,
|
| 54 |
+
"kl": 0.00012683868408203125,
|
| 55 |
+
"learning_rate": 5e-07,
|
| 56 |
+
"loss": 0.0,
|
| 57 |
+
"reward": 0.4923219382762909,
|
| 58 |
+
"reward_std": 0.5618958175182343,
|
| 59 |
+
"rewards/correct_code_reward_func": 0.2708333358168602,
|
| 60 |
+
"rewards/len_reward_func": 0.2214886099100113,
|
| 61 |
+
"step": 4
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"completion_length": 167.9166717529297,
|
| 65 |
+
"epoch": 0.08,
|
| 66 |
+
"grad_norm": 0.6875029348712502,
|
| 67 |
+
"kl": 7.176399230957031e-05,
|
| 68 |
+
"learning_rate": 4.999157413258781e-07,
|
| 69 |
+
"loss": 0.0,
|
| 70 |
+
"reward": 0.3392188400030136,
|
| 71 |
+
"reward_std": 0.3500474542379379,
|
| 72 |
+
"rewards/correct_code_reward_func": 0.0833333358168602,
|
| 73 |
+
"rewards/len_reward_func": 0.255885511636734,
|
| 74 |
+
"step": 5
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"completion_length": 132.9166717529297,
|
| 78 |
+
"epoch": 0.096,
|
| 79 |
+
"grad_norm": 0.7328221161754551,
|
| 80 |
+
"kl": 0.0001456737518310547,
|
| 81 |
+
"learning_rate": 4.996630220997057e-07,
|
| 82 |
+
"loss": 0.0,
|
| 83 |
+
"reward": 0.7516234815120697,
|
| 84 |
+
"reward_std": 0.5683842748403549,
|
| 85 |
+
"rewards/correct_code_reward_func": 0.4375000298023224,
|
| 86 |
+
"rewards/len_reward_func": 0.3141235262155533,
|
| 87 |
+
"step": 6
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"completion_length": 178.0416717529297,
|
| 91 |
+
"epoch": 0.112,
|
| 92 |
+
"grad_norm": 0.7294488262352424,
|
| 93 |
+
"kl": 0.00011301040649414062,
|
| 94 |
+
"learning_rate": 4.992420126717784e-07,
|
| 95 |
+
"loss": 0.0,
|
| 96 |
+
"reward": 0.4081832021474838,
|
| 97 |
+
"reward_std": 0.4984392523765564,
|
| 98 |
+
"rewards/correct_code_reward_func": 0.1666666716337204,
|
| 99 |
+
"rewards/len_reward_func": 0.24151653796434402,
|
| 100 |
+
"step": 7
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"completion_length": 190.81250762939453,
|
| 104 |
+
"epoch": 0.128,
|
| 105 |
+
"grad_norm": 0.774451422596215,
|
| 106 |
+
"kl": 0.00012731552124023438,
|
| 107 |
+
"learning_rate": 4.986529968316653e-07,
|
| 108 |
+
"loss": 0.0,
|
| 109 |
+
"reward": 0.5889585018157959,
|
| 110 |
+
"reward_std": 0.5029059052467346,
|
| 111 |
+
"rewards/correct_code_reward_func": 0.2500000149011612,
|
| 112 |
+
"rewards/len_reward_func": 0.3389585465192795,
|
| 113 |
+
"step": 8
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"completion_length": 203.1041717529297,
|
| 117 |
+
"epoch": 0.144,
|
| 118 |
+
"grad_norm": 0.7610759960843761,
|
| 119 |
+
"kl": 0.0001308917999267578,
|
| 120 |
+
"learning_rate": 4.978963716169165e-07,
|
| 121 |
+
"loss": 0.0,
|
| 122 |
+
"reward": 0.5821144282817841,
|
| 123 |
+
"reward_std": 0.5114115178585052,
|
| 124 |
+
"rewards/correct_code_reward_func": 0.3333333432674408,
|
| 125 |
+
"rewards/len_reward_func": 0.24878107011318207,
|
| 126 |
+
"step": 9
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"completion_length": 119.56250762939453,
|
| 130 |
+
"epoch": 0.16,
|
| 131 |
+
"grad_norm": 0.9290248509168798,
|
| 132 |
+
"kl": 0.00010132789611816406,
|
| 133 |
+
"learning_rate": 4.969726470454313e-07,
|
| 134 |
+
"loss": 0.0,
|
| 135 |
+
"reward": 0.7238170504570007,
|
| 136 |
+
"reward_std": 0.5344631671905518,
|
| 137 |
+
"rewards/correct_code_reward_func": 0.4166666716337204,
|
| 138 |
+
"rewards/len_reward_func": 0.30715033411979675,
|
| 139 |
+
"step": 10
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"completion_length": 150.9166717529297,
|
| 143 |
+
"epoch": 0.176,
|
| 144 |
+
"grad_norm": 1.3590266587813116,
|
| 145 |
+
"kl": 0.00018453598022460938,
|
| 146 |
+
"learning_rate": 4.958824457716706e-07,
|
| 147 |
+
"loss": 0.0,
|
| 148 |
+
"reward": 0.6649805009365082,
|
| 149 |
+
"reward_std": 0.4508441388607025,
|
| 150 |
+
"rewards/correct_code_reward_func": 0.375,
|
| 151 |
+
"rewards/len_reward_func": 0.289980486035347,
|
| 152 |
+
"step": 11
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"completion_length": 206.8541717529297,
|
| 156 |
+
"epoch": 0.192,
|
| 157 |
+
"grad_norm": 0.6744815448192079,
|
| 158 |
+
"kl": 0.000133514404296875,
|
| 159 |
+
"learning_rate": 4.946265026669454e-07,
|
| 160 |
+
"loss": 0.0,
|
| 161 |
+
"reward": 0.44046473503112793,
|
| 162 |
+
"reward_std": 0.4918256551027298,
|
| 163 |
+
"rewards/correct_code_reward_func": 0.2083333432674408,
|
| 164 |
+
"rewards/len_reward_func": 0.23213139921426773,
|
| 165 |
+
"step": 12
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"completion_length": 100.33333587646484,
|
| 169 |
+
"epoch": 0.208,
|
| 170 |
+
"grad_norm": 1.1473694035346043,
|
| 171 |
+
"kl": 0.00011897087097167969,
|
| 172 |
+
"learning_rate": 4.932056643240618e-07,
|
| 173 |
+
"loss": 0.0,
|
| 174 |
+
"reward": 0.8099721968173981,
|
| 175 |
+
"reward_std": 0.5177792608737946,
|
| 176 |
+
"rewards/correct_code_reward_func": 0.4791666716337204,
|
| 177 |
+
"rewards/len_reward_func": 0.3308054953813553,
|
| 178 |
+
"step": 13
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"completion_length": 174.52084350585938,
|
| 182 |
+
"epoch": 0.224,
|
| 183 |
+
"grad_norm": 0.7173952224852224,
|
| 184 |
+
"kl": 0.00014257431030273438,
|
| 185 |
+
"learning_rate": 4.916208884866592e-07,
|
| 186 |
+
"loss": 0.0,
|
| 187 |
+
"reward": 0.5219025313854218,
|
| 188 |
+
"reward_std": 0.43796107172966003,
|
| 189 |
+
"rewards/correct_code_reward_func": 0.229166679084301,
|
| 190 |
+
"rewards/len_reward_func": 0.29273584485054016,
|
| 191 |
+
"step": 14
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"completion_length": 118.60417175292969,
|
| 195 |
+
"epoch": 0.24,
|
| 196 |
+
"grad_norm": 1.2540281005004774,
|
| 197 |
+
"kl": 0.00018072128295898438,
|
| 198 |
+
"learning_rate": 4.898732434036243e-07,
|
| 199 |
+
"loss": 0.0,
|
| 200 |
+
"reward": 0.7924503684043884,
|
| 201 |
+
"reward_std": 0.506152406334877,
|
| 202 |
+
"rewards/correct_code_reward_func": 0.4375000149011612,
|
| 203 |
+
"rewards/len_reward_func": 0.3549504280090332,
|
| 204 |
+
"step": 15
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"completion_length": 105.31250381469727,
|
| 208 |
+
"epoch": 0.256,
|
| 209 |
+
"grad_norm": 0.9431489918373908,
|
| 210 |
+
"kl": 0.00018644332885742188,
|
| 211 |
+
"learning_rate": 4.879639071090173e-07,
|
| 212 |
+
"loss": 0.0,
|
| 213 |
+
"reward": 0.5001727938652039,
|
| 214 |
+
"reward_std": 0.45510660111904144,
|
| 215 |
+
"rewards/correct_code_reward_func": 0.2291666716337204,
|
| 216 |
+
"rewards/len_reward_func": 0.27100610733032227,
|
| 217 |
+
"step": 16
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"completion_length": 130.85416793823242,
|
| 221 |
+
"epoch": 0.272,
|
| 222 |
+
"grad_norm": 0.7237713131170656,
|
| 223 |
+
"kl": 0.00011801719665527344,
|
| 224 |
+
"learning_rate": 4.858941666279955e-07,
|
| 225 |
+
"loss": 0.0,
|
| 226 |
+
"reward": 0.7888750731945038,
|
| 227 |
+
"reward_std": 0.5659129619598389,
|
| 228 |
+
"rewards/correct_code_reward_func": 0.5208333432674408,
|
| 229 |
+
"rewards/len_reward_func": 0.2680417224764824,
|
| 230 |
+
"step": 17
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"completion_length": 134.89583587646484,
|
| 234 |
+
"epoch": 0.288,
|
| 235 |
+
"grad_norm": 0.8717728732040717,
|
| 236 |
+
"kl": 0.000164031982421875,
|
| 237 |
+
"learning_rate": 4.836654171092682e-07,
|
| 238 |
+
"loss": 0.0,
|
| 239 |
+
"reward": 0.5576134622097015,
|
| 240 |
+
"reward_std": 0.5418040752410889,
|
| 241 |
+
"rewards/correct_code_reward_func": 0.3125000149011612,
|
| 242 |
+
"rewards/len_reward_func": 0.24511344730854034,
|
| 243 |
+
"step": 18
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"completion_length": 116.54167175292969,
|
| 247 |
+
"epoch": 0.304,
|
| 248 |
+
"grad_norm": 0.9377250817808462,
|
| 249 |
+
"kl": 0.0002789497375488281,
|
| 250 |
+
"learning_rate": 4.812791608846709e-07,
|
| 251 |
+
"loss": 0.0,
|
| 252 |
+
"reward": 0.7832874357700348,
|
| 253 |
+
"reward_std": 0.5703642070293427,
|
| 254 |
+
"rewards/correct_code_reward_func": 0.4583333432674408,
|
| 255 |
+
"rewards/len_reward_func": 0.3249540776014328,
|
| 256 |
+
"step": 19
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"completion_length": 186.06250762939453,
|
| 260 |
+
"epoch": 0.32,
|
| 261 |
+
"grad_norm": 0.9812718853662215,
|
| 262 |
+
"kl": 0.00038433074951171875,
|
| 263 |
+
"learning_rate": 4.787370064564882e-07,
|
| 264 |
+
"loss": 0.0,
|
| 265 |
+
"reward": 0.5579635202884674,
|
| 266 |
+
"reward_std": 0.519572913646698,
|
| 267 |
+
"rewards/correct_code_reward_func": 0.3125,
|
| 268 |
+
"rewards/len_reward_func": 0.2454635202884674,
|
| 269 |
+
"step": 20
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"completion_length": 110.75,
|
| 273 |
+
"epoch": 0.336,
|
| 274 |
+
"grad_norm": 1.0782358974361461,
|
| 275 |
+
"kl": 0.00030231475830078125,
|
| 276 |
+
"learning_rate": 4.7604066741321253e-07,
|
| 277 |
+
"loss": 0.0,
|
| 278 |
+
"reward": 0.9209870100021362,
|
| 279 |
+
"reward_std": 0.5739021599292755,
|
| 280 |
+
"rewards/correct_code_reward_func": 0.6458333432674408,
|
| 281 |
+
"rewards/len_reward_func": 0.2751536965370178,
|
| 282 |
+
"step": 21
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"completion_length": 162.08334350585938,
|
| 286 |
+
"epoch": 0.352,
|
| 287 |
+
"grad_norm": 0.874006092063079,
|
| 288 |
+
"kl": 0.00022363662719726562,
|
| 289 |
+
"learning_rate": 4.731919612744659e-07,
|
| 290 |
+
"loss": 0.0,
|
| 291 |
+
"reward": 0.7241076529026031,
|
| 292 |
+
"reward_std": 0.5581964254379272,
|
| 293 |
+
"rewards/correct_code_reward_func": 0.4791666865348816,
|
| 294 |
+
"rewards/len_reward_func": 0.24494098126888275,
|
| 295 |
+
"step": 22
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"completion_length": 204.14584350585938,
|
| 299 |
+
"epoch": 0.368,
|
| 300 |
+
"grad_norm": 0.82204687952658,
|
| 301 |
+
"kl": 0.00034046173095703125,
|
| 302 |
+
"learning_rate": 4.7019280826586604e-07,
|
| 303 |
+
"loss": 0.0,
|
| 304 |
+
"reward": 0.532179206609726,
|
| 305 |
+
"reward_std": 0.34643039107322693,
|
| 306 |
+
"rewards/correct_code_reward_func": 0.1666666716337204,
|
| 307 |
+
"rewards/len_reward_func": 0.36551257967948914,
|
| 308 |
+
"step": 23
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"completion_length": 151.0208396911621,
|
| 312 |
+
"epoch": 0.384,
|
| 313 |
+
"grad_norm": 0.9960569412421229,
|
| 314 |
+
"kl": 0.00042057037353515625,
|
| 315 |
+
"learning_rate": 4.6704523002466094e-07,
|
| 316 |
+
"loss": 0.0,
|
| 317 |
+
"reward": 0.5193642377853394,
|
| 318 |
+
"reward_std": 0.40249721705913544,
|
| 319 |
+
"rewards/correct_code_reward_func": 0.2083333395421505,
|
| 320 |
+
"rewards/len_reward_func": 0.31103089451789856,
|
| 321 |
+
"step": 24
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"completion_length": 187.20834350585938,
|
| 325 |
+
"epoch": 0.4,
|
| 326 |
+
"grad_norm": 0.6829787161734556,
|
| 327 |
+
"kl": 0.0003147125244140625,
|
| 328 |
+
"learning_rate": 4.6375134823700503e-07,
|
| 329 |
+
"loss": 0.0,
|
| 330 |
+
"reward": 0.43366140127182007,
|
| 331 |
+
"reward_std": 0.4143451601266861,
|
| 332 |
+
"rewards/correct_code_reward_func": 0.1666666716337204,
|
| 333 |
+
"rewards/len_reward_func": 0.2669947147369385,
|
| 334 |
+
"step": 25
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"completion_length": 99.33333587646484,
|
| 338 |
+
"epoch": 0.416,
|
| 339 |
+
"grad_norm": 1.0546549364927,
|
| 340 |
+
"kl": 0.0007066726684570312,
|
| 341 |
+
"learning_rate": 4.603133832077953e-07,
|
| 342 |
+
"loss": 0.0,
|
| 343 |
+
"reward": 0.6213856935501099,
|
| 344 |
+
"reward_std": 0.5858824849128723,
|
| 345 |
+
"rewards/correct_code_reward_func": 0.395833358168602,
|
| 346 |
+
"rewards/len_reward_func": 0.22555235773324966,
|
| 347 |
+
"step": 26
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"completion_length": 119.04167175292969,
|
| 351 |
+
"epoch": 0.432,
|
| 352 |
+
"grad_norm": 1.0435674697004191,
|
| 353 |
+
"kl": 0.000743865966796875,
|
| 354 |
+
"learning_rate": 4.5673365236403216e-07,
|
| 355 |
+
"loss": 0.0,
|
| 356 |
+
"reward": 0.6226212680339813,
|
| 357 |
+
"reward_std": 0.515766441822052,
|
| 358 |
+
"rewards/correct_code_reward_func": 0.4791666865348816,
|
| 359 |
+
"rewards/len_reward_func": 0.14345459267497063,
|
| 360 |
+
"step": 27
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"completion_length": 204.02083587646484,
|
| 364 |
+
"epoch": 0.448,
|
| 365 |
+
"grad_norm": 1.1685217734586066,
|
| 366 |
+
"kl": 0.000659942626953125,
|
| 367 |
+
"learning_rate": 4.530145686927125e-07,
|
| 368 |
+
"loss": 0.0,
|
| 369 |
+
"reward": 0.6347978413105011,
|
| 370 |
+
"reward_std": 0.4856678545475006,
|
| 371 |
+
"rewards/correct_code_reward_func": 0.3333333432674408,
|
| 372 |
+
"rewards/len_reward_func": 0.3014644980430603,
|
| 373 |
+
"step": 28
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"completion_length": 149.87500762939453,
|
| 377 |
+
"epoch": 0.464,
|
| 378 |
+
"grad_norm": 0.7804112511261988,
|
| 379 |
+
"kl": 0.0006847381591796875,
|
| 380 |
+
"learning_rate": 4.4915863911430897e-07,
|
| 381 |
+
"loss": 0.0,
|
| 382 |
+
"reward": 0.4445287883281708,
|
| 383 |
+
"reward_std": 0.4996263086795807,
|
| 384 |
+
"rewards/correct_code_reward_func": 0.1458333358168602,
|
| 385 |
+
"rewards/len_reward_func": 0.29869547486305237,
|
| 386 |
+
"step": 29
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"completion_length": 132.0833396911621,
|
| 390 |
+
"epoch": 0.48,
|
| 391 |
+
"grad_norm": 0.8535635529221804,
|
| 392 |
+
"kl": 0.0006618499755859375,
|
| 393 |
+
"learning_rate": 4.45168462792932e-07,
|
| 394 |
+
"loss": 0.0,
|
| 395 |
+
"reward": 0.6259033381938934,
|
| 396 |
+
"reward_std": 0.4563398212194443,
|
| 397 |
+
"rewards/correct_code_reward_func": 0.3750000149011612,
|
| 398 |
+
"rewards/len_reward_func": 0.2509033679962158,
|
| 399 |
+
"step": 30
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"completion_length": 126.79166793823242,
|
| 403 |
+
"epoch": 0.496,
|
| 404 |
+
"grad_norm": 0.8761490464403178,
|
| 405 |
+
"kl": 0.0010280609130859375,
|
| 406 |
+
"learning_rate": 4.4104672938431223e-07,
|
| 407 |
+
"loss": 0.0,
|
| 408 |
+
"reward": 0.7820720672607422,
|
| 409 |
+
"reward_std": 0.4476567506790161,
|
| 410 |
+
"rewards/correct_code_reward_func": 0.583333358168602,
|
| 411 |
+
"rewards/len_reward_func": 0.1987387351691723,
|
| 412 |
+
"step": 31
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"completion_length": 95.87500381469727,
|
| 416 |
+
"epoch": 0.512,
|
| 417 |
+
"grad_norm": 1.0760263673729638,
|
| 418 |
+
"kl": 0.0013580322265625,
|
| 419 |
+
"learning_rate": 4.367962172227866e-07,
|
| 420 |
+
"loss": 0.0,
|
| 421 |
+
"reward": 0.7095580399036407,
|
| 422 |
+
"reward_std": 0.4921827018260956,
|
| 423 |
+
"rewards/correct_code_reward_func": 0.5000000149011612,
|
| 424 |
+
"rewards/len_reward_func": 0.20955805480480194,
|
| 425 |
+
"step": 32
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"completion_length": 155.02083587646484,
|
| 429 |
+
"epoch": 0.528,
|
| 430 |
+
"grad_norm": 0.9305216291102568,
|
| 431 |
+
"kl": 0.001430511474609375,
|
| 432 |
+
"learning_rate": 4.324197914485075e-07,
|
| 433 |
+
"loss": 0.0,
|
| 434 |
+
"reward": 0.6416721642017365,
|
| 435 |
+
"reward_std": 0.4772767722606659,
|
| 436 |
+
"rewards/correct_code_reward_func": 0.3958333432674408,
|
| 437 |
+
"rewards/len_reward_func": 0.24583880603313446,
|
| 438 |
+
"step": 33
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"completion_length": 215.0416717529297,
|
| 442 |
+
"epoch": 0.544,
|
| 443 |
+
"grad_norm": 0.8211396277152214,
|
| 444 |
+
"kl": 0.0010585784912109375,
|
| 445 |
+
"learning_rate": 4.2792040207614e-07,
|
| 446 |
+
"loss": 0.0,
|
| 447 |
+
"reward": 0.7173371911048889,
|
| 448 |
+
"reward_std": 0.5010073632001877,
|
| 449 |
+
"rewards/correct_code_reward_func": 0.3958333432674408,
|
| 450 |
+
"rewards/len_reward_func": 0.3215038478374481,
|
| 451 |
+
"step": 34
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"completion_length": 128.8958396911621,
|
| 455 |
+
"epoch": 0.56,
|
| 456 |
+
"grad_norm": 0.9266230852523272,
|
| 457 |
+
"kl": 0.001445770263671875,
|
| 458 |
+
"learning_rate": 4.2330108200634723e-07,
|
| 459 |
+
"loss": 0.0,
|
| 460 |
+
"reward": 0.6981382668018341,
|
| 461 |
+
"reward_std": 0.5356450974941254,
|
| 462 |
+
"rewards/correct_code_reward_func": 0.4166666716337204,
|
| 463 |
+
"rewards/len_reward_func": 0.2814715951681137,
|
| 464 |
+
"step": 35
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"completion_length": 174.62500762939453,
|
| 468 |
+
"epoch": 0.576,
|
| 469 |
+
"grad_norm": 0.8774350304368993,
|
| 470 |
+
"kl": 0.00160980224609375,
|
| 471 |
+
"learning_rate": 4.185649449814045e-07,
|
| 472 |
+
"loss": 0.0,
|
| 473 |
+
"reward": 0.819614589214325,
|
| 474 |
+
"reward_std": 0.5062630474567413,
|
| 475 |
+
"rewards/correct_code_reward_func": 0.5000000223517418,
|
| 476 |
+
"rewards/len_reward_func": 0.3196146488189697,
|
| 477 |
+
"step": 36
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"completion_length": 79.29166793823242,
|
| 481 |
+
"epoch": 0.592,
|
| 482 |
+
"grad_norm": 1.1529766368540975,
|
| 483 |
+
"kl": 0.0024871826171875,
|
| 484 |
+
"learning_rate": 4.137151834863213e-07,
|
| 485 |
+
"loss": 0.0,
|
| 486 |
+
"reward": 0.615891844034195,
|
| 487 |
+
"reward_std": 0.6233284175395966,
|
| 488 |
+
"rewards/correct_code_reward_func": 0.4166666716337204,
|
| 489 |
+
"rewards/len_reward_func": 0.19922512769699097,
|
| 490 |
+
"step": 37
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"completion_length": 85.97917175292969,
|
| 494 |
+
"epoch": 0.608,
|
| 495 |
+
"grad_norm": 1.2549480144397245,
|
| 496 |
+
"kl": 0.0028076171875,
|
| 497 |
+
"learning_rate": 4.087550665968846e-07,
|
| 498 |
+
"loss": 0.0,
|
| 499 |
+
"reward": 0.5825270563364029,
|
| 500 |
+
"reward_std": 0.45819054543972015,
|
| 501 |
+
"rewards/correct_code_reward_func": 0.35416667722165585,
|
| 502 |
+
"rewards/len_reward_func": 0.2283603698015213,
|
| 503 |
+
"step": 38
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"completion_length": 85.68750381469727,
|
| 507 |
+
"epoch": 0.624,
|
| 508 |
+
"grad_norm": 1.4005267604268767,
|
| 509 |
+
"kl": 0.003387451171875,
|
| 510 |
+
"learning_rate": 4.036879377760752e-07,
|
| 511 |
+
"loss": 0.0,
|
| 512 |
+
"reward": 0.8216443359851837,
|
| 513 |
+
"reward_std": 0.548437625169754,
|
| 514 |
+
"rewards/correct_code_reward_func": 0.5833333432674408,
|
| 515 |
+
"rewards/len_reward_func": 0.23831100016832352,
|
| 516 |
+
"step": 39
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"completion_length": 113.16667175292969,
|
| 520 |
+
"epoch": 0.64,
|
| 521 |
+
"grad_norm": 1.0341725323218647,
|
| 522 |
+
"kl": 0.003143310546875,
|
| 523 |
+
"learning_rate": 3.9851721262034157e-07,
|
| 524 |
+
"loss": 0.0,
|
| 525 |
+
"reward": 0.520989790558815,
|
| 526 |
+
"reward_std": 0.47860175371170044,
|
| 527 |
+
"rewards/correct_code_reward_func": 0.25,
|
| 528 |
+
"rewards/len_reward_func": 0.270989790558815,
|
| 529 |
+
"step": 40
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"completion_length": 109.70833587646484,
|
| 533 |
+
"epoch": 0.656,
|
| 534 |
+
"grad_norm": 1.0974308815666283,
|
| 535 |
+
"kl": 0.00255584716796875,
|
| 536 |
+
"learning_rate": 3.932463765572505e-07,
|
| 537 |
+
"loss": 0.0,
|
| 538 |
+
"reward": 0.5007437467575073,
|
| 539 |
+
"reward_std": 0.4586540758609772,
|
| 540 |
+
"rewards/correct_code_reward_func": 0.3333333432674408,
|
| 541 |
+
"rewards/len_reward_func": 0.16741041094064713,
|
| 542 |
+
"step": 41
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"completion_length": 113.50000381469727,
|
| 546 |
+
"epoch": 0.672,
|
| 547 |
+
"grad_norm": 0.8007467983812053,
|
| 548 |
+
"kl": 0.00237274169921875,
|
| 549 |
+
"learning_rate": 3.8787898249606767e-07,
|
| 550 |
+
"loss": 0.0,
|
| 551 |
+
"reward": 0.4927902817726135,
|
| 552 |
+
"reward_std": 0.4271247088909149,
|
| 553 |
+
"rewards/correct_code_reward_func": 0.2708333395421505,
|
| 554 |
+
"rewards/len_reward_func": 0.22195692360401154,
|
| 555 |
+
"step": 42
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"completion_length": 79.85416793823242,
|
| 559 |
+
"epoch": 0.688,
|
| 560 |
+
"grad_norm": 1.2179865851315288,
|
| 561 |
+
"kl": 0.00423431396484375,
|
| 562 |
+
"learning_rate": 3.8241864843284964e-07,
|
| 563 |
+
"loss": 0.0,
|
| 564 |
+
"reward": 0.8101656138896942,
|
| 565 |
+
"reward_std": 0.4996851086616516,
|
| 566 |
+
"rewards/correct_code_reward_func": 0.5416666716337204,
|
| 567 |
+
"rewards/len_reward_func": 0.2684989869594574,
|
| 568 |
+
"step": 43
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"completion_length": 91.54167175292969,
|
| 572 |
+
"epoch": 0.704,
|
| 573 |
+
"grad_norm": 1.240439347309207,
|
| 574 |
+
"kl": 0.005615234375,
|
| 575 |
+
"learning_rate": 3.768690550116639e-07,
|
| 576 |
+
"loss": 0.0,
|
| 577 |
+
"reward": 0.7196292579174042,
|
| 578 |
+
"reward_std": 0.5478431880474091,
|
| 579 |
+
"rewards/correct_code_reward_func": 0.4166666865348816,
|
| 580 |
+
"rewards/len_reward_func": 0.3029625713825226,
|
| 581 |
+
"step": 44
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"completion_length": 121.97916793823242,
|
| 585 |
+
"epoch": 0.72,
|
| 586 |
+
"grad_norm": 0.9464336788144472,
|
| 587 |
+
"kl": 0.0048980712890625,
|
| 588 |
+
"learning_rate": 3.712339430435792e-07,
|
| 589 |
+
"loss": 0.0,
|
| 590 |
+
"reward": 0.8251023292541504,
|
| 591 |
+
"reward_std": 0.5142593383789062,
|
| 592 |
+
"rewards/correct_code_reward_func": 0.4791666716337204,
|
| 593 |
+
"rewards/len_reward_func": 0.34593565762043,
|
| 594 |
+
"step": 45
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"completion_length": 88.20833587646484,
|
| 598 |
+
"epoch": 0.736,
|
| 599 |
+
"grad_norm": 0.944700095686709,
|
| 600 |
+
"kl": 0.0039215087890625,
|
| 601 |
+
"learning_rate": 3.65517110985099e-07,
|
| 602 |
+
"loss": 0.0,
|
| 603 |
+
"reward": 0.6876890659332275,
|
| 604 |
+
"reward_std": 0.4308091402053833,
|
| 605 |
+
"rewards/correct_code_reward_func": 0.4166666716337204,
|
| 606 |
+
"rewards/len_reward_func": 0.27102240175008774,
|
| 607 |
+
"step": 46
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"completion_length": 69.75000381469727,
|
| 611 |
+
"epoch": 0.752,
|
| 612 |
+
"grad_norm": 1.2034725464007063,
|
| 613 |
+
"kl": 0.0064544677734375,
|
| 614 |
+
"learning_rate": 3.597224123777389e-07,
|
| 615 |
+
"loss": 0.0,
|
| 616 |
+
"reward": 0.9030305743217468,
|
| 617 |
+
"reward_std": 0.5125944316387177,
|
| 618 |
+
"rewards/correct_code_reward_func": 0.6458333730697632,
|
| 619 |
+
"rewards/len_reward_func": 0.25719721615314484,
|
| 620 |
+
"step": 47
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"completion_length": 84.00000381469727,
|
| 624 |
+
"epoch": 0.768,
|
| 625 |
+
"grad_norm": 1.188785130607309,
|
| 626 |
+
"kl": 0.0072174072265625,
|
| 627 |
+
"learning_rate": 3.5385375325047163e-07,
|
| 628 |
+
"loss": 0.0,
|
| 629 |
+
"reward": 0.6972790956497192,
|
| 630 |
+
"reward_std": 0.6147326231002808,
|
| 631 |
+
"rewards/correct_code_reward_func": 0.3750000149011612,
|
| 632 |
+
"rewards/len_reward_func": 0.32227905094623566,
|
| 633 |
+
"step": 48
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"completion_length": 54.35416793823242,
|
| 637 |
+
"epoch": 0.784,
|
| 638 |
+
"grad_norm": 1.5392691471218216,
|
| 639 |
+
"kl": 0.011932373046875,
|
| 640 |
+
"learning_rate": 3.479150894867926e-07,
|
| 641 |
+
"loss": 0.0,
|
| 642 |
+
"reward": 0.7476979494094849,
|
| 643 |
+
"reward_std": 0.5232938826084137,
|
| 644 |
+
"rewards/correct_code_reward_func": 0.520833358168602,
|
| 645 |
+
"rewards/len_reward_func": 0.22686457633972168,
|
| 646 |
+
"step": 49
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"completion_length": 105.375,
|
| 650 |
+
"epoch": 0.8,
|
| 651 |
+
"grad_norm": 0.9867137272930855,
|
| 652 |
+
"kl": 0.0070037841796875,
|
| 653 |
+
"learning_rate": 3.4191042415818e-07,
|
| 654 |
+
"loss": 0.0,
|
| 655 |
+
"reward": 0.7631438374519348,
|
| 656 |
+
"reward_std": 0.5120529979467392,
|
| 657 |
+
"rewards/correct_code_reward_func": 0.4375,
|
| 658 |
+
"rewards/len_reward_func": 0.3256438076496124,
|
| 659 |
+
"step": 50
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"completion_length": 79.79166793823242,
|
| 663 |
+
"epoch": 0.816,
|
| 664 |
+
"grad_norm": 1.2765736460552886,
|
| 665 |
+
"kl": 0.009490966796875,
|
| 666 |
+
"learning_rate": 3.3584380482574717e-07,
|
| 667 |
+
"loss": 0.0,
|
| 668 |
+
"reward": 0.8190419375896454,
|
| 669 |
+
"reward_std": 0.5837846994400024,
|
| 670 |
+
"rewards/correct_code_reward_func": 0.583333358168602,
|
| 671 |
+
"rewards/len_reward_func": 0.23570860922336578,
|
| 672 |
+
"step": 51
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"completion_length": 66.62500381469727,
|
| 676 |
+
"epoch": 0.832,
|
| 677 |
+
"grad_norm": 1.2357424740774945,
|
| 678 |
+
"kl": 0.012359619140625,
|
| 679 |
+
"learning_rate": 3.297193208119047e-07,
|
| 680 |
+
"loss": 0.0,
|
| 681 |
+
"reward": 0.9992968440055847,
|
| 682 |
+
"reward_std": 0.38213877379894257,
|
| 683 |
+
"rewards/correct_code_reward_func": 0.7500000298023224,
|
| 684 |
+
"rewards/len_reward_func": 0.24929680675268173,
|
| 685 |
+
"step": 52
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"completion_length": 99.83333587646484,
|
| 689 |
+
"epoch": 0.848,
|
| 690 |
+
"grad_norm": 1.1189426445940507,
|
| 691 |
+
"kl": 0.01153564453125,
|
| 692 |
+
"learning_rate": 3.235411004438741e-07,
|
| 693 |
+
"loss": 0.0,
|
| 694 |
+
"reward": 0.8771131932735443,
|
| 695 |
+
"reward_std": 0.5585960447788239,
|
| 696 |
+
"rewards/correct_code_reward_func": 0.520833358168602,
|
| 697 |
+
"rewards/len_reward_func": 0.3562798500061035,
|
| 698 |
+
"step": 53
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"completion_length": 115.89583587646484,
|
| 702 |
+
"epoch": 0.864,
|
| 703 |
+
"grad_norm": 1.3030164084383453,
|
| 704 |
+
"kl": 0.013397216796875,
|
| 705 |
+
"learning_rate": 3.173133082709086e-07,
|
| 706 |
+
"loss": 0.0,
|
| 707 |
+
"reward": 0.9209796786308289,
|
| 708 |
+
"reward_std": 0.5127619951963425,
|
| 709 |
+
"rewards/correct_code_reward_func": 0.5416666865348816,
|
| 710 |
+
"rewards/len_reward_func": 0.37931299209594727,
|
| 711 |
+
"step": 54
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"completion_length": 45.81250190734863,
|
| 715 |
+
"epoch": 0.88,
|
| 716 |
+
"grad_norm": 1.2758138396906251,
|
| 717 |
+
"kl": 0.02093505859375,
|
| 718 |
+
"learning_rate": 3.1104014225709784e-07,
|
| 719 |
+
"loss": 0.0,
|
| 720 |
+
"reward": 1.014316976070404,
|
| 721 |
+
"reward_std": 0.5054647028446198,
|
| 722 |
+
"rewards/correct_code_reward_func": 0.6666666865348816,
|
| 723 |
+
"rewards/len_reward_func": 0.34765030443668365,
|
| 724 |
+
"step": 55
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"completion_length": 177.89584350585938,
|
| 728 |
+
"epoch": 0.896,
|
| 729 |
+
"grad_norm": 0.9790523295873852,
|
| 730 |
+
"kl": 0.0159149169921875,
|
| 731 |
+
"learning_rate": 3.0472583095164873e-07,
|
| 732 |
+
"loss": 0.0,
|
| 733 |
+
"reward": 0.5877984166145325,
|
| 734 |
+
"reward_std": 0.4042231887578964,
|
| 735 |
+
"rewards/correct_code_reward_func": 0.3125000149011612,
|
| 736 |
+
"rewards/len_reward_func": 0.2752983868122101,
|
| 737 |
+
"step": 56
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"completion_length": 46.97916793823242,
|
| 741 |
+
"epoch": 0.912,
|
| 742 |
+
"grad_norm": 2.352715010123481,
|
| 743 |
+
"kl": 0.0218505859375,
|
| 744 |
+
"learning_rate": 2.983746306385499e-07,
|
| 745 |
+
"loss": 0.0,
|
| 746 |
+
"reward": 0.9101540148258209,
|
| 747 |
+
"reward_std": 0.5364184379577637,
|
| 748 |
+
"rewards/correct_code_reward_func": 0.5625,
|
| 749 |
+
"rewards/len_reward_func": 0.3476540297269821,
|
| 750 |
+
"step": 57
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"completion_length": 86.5,
|
| 754 |
+
"epoch": 0.928,
|
| 755 |
+
"grad_norm": 1.3590745378537943,
|
| 756 |
+
"kl": 0.018341064453125,
|
| 757 |
+
"learning_rate": 2.919908224675412e-07,
|
| 758 |
+
"loss": 0.0,
|
| 759 |
+
"reward": 0.7466456890106201,
|
| 760 |
+
"reward_std": 0.4038194566965103,
|
| 761 |
+
"rewards/correct_code_reward_func": 0.458333358168602,
|
| 762 |
+
"rewards/len_reward_func": 0.2883123308420181,
|
| 763 |
+
"step": 58
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"completion_length": 93.35416793823242,
|
| 767 |
+
"epoch": 0.944,
|
| 768 |
+
"grad_norm": 1.0657115005336795,
|
| 769 |
+
"kl": 0.0169677734375,
|
| 770 |
+
"learning_rate": 2.8557870956832133e-07,
|
| 771 |
+
"loss": 0.0,
|
| 772 |
+
"reward": 0.8307068645954132,
|
| 773 |
+
"reward_std": 0.41568903625011444,
|
| 774 |
+
"rewards/correct_code_reward_func": 0.458333358168602,
|
| 775 |
+
"rewards/len_reward_func": 0.37237347662448883,
|
| 776 |
+
"step": 59
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"completion_length": 96.04166793823242,
|
| 780 |
+
"epoch": 0.96,
|
| 781 |
+
"grad_norm": 1.1894982274411716,
|
| 782 |
+
"kl": 0.0167236328125,
|
| 783 |
+
"learning_rate": 2.7914261414993976e-07,
|
| 784 |
+
"loss": 0.0,
|
| 785 |
+
"reward": 0.9781838655471802,
|
| 786 |
+
"reward_std": 0.5707030892372131,
|
| 787 |
+
"rewards/correct_code_reward_func": 0.6458333730697632,
|
| 788 |
+
"rewards/len_reward_func": 0.33235056698322296,
|
| 789 |
+
"step": 60
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"completion_length": 57.041669845581055,
|
| 793 |
+
"epoch": 0.976,
|
| 794 |
+
"grad_norm": 1.6373435351232113,
|
| 795 |
+
"kl": 0.02020263671875,
|
| 796 |
+
"learning_rate": 2.726868745873286e-07,
|
| 797 |
+
"loss": 0.0,
|
| 798 |
+
"reward": 0.7698030173778534,
|
| 799 |
+
"reward_std": 0.5640588998794556,
|
| 800 |
+
"rewards/correct_code_reward_func": 0.5208333432674408,
|
| 801 |
+
"rewards/len_reward_func": 0.2489696592092514,
|
| 802 |
+
"step": 61
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"completion_length": 84.89583587646484,
|
| 806 |
+
"epoch": 0.992,
|
| 807 |
+
"grad_norm": 1.223004703854309,
|
| 808 |
+
"kl": 0.017547607421875,
|
| 809 |
+
"learning_rate": 2.662158424969357e-07,
|
| 810 |
+
"loss": 0.0,
|
| 811 |
+
"reward": 0.8090977072715759,
|
| 812 |
+
"reward_std": 0.6418364942073822,
|
| 813 |
+
"rewards/correct_code_reward_func": 0.4791666716337204,
|
| 814 |
+
"rewards/len_reward_func": 0.32993103563785553,
|
| 815 |
+
"step": 62
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"completion_length": 52.958335876464844,
|
| 819 |
+
"epoch": 1.0,
|
| 820 |
+
"grad_norm": 1.223004703854309,
|
| 821 |
+
"kl": 0.03564453125,
|
| 822 |
+
"learning_rate": 2.597338798034344e-07,
|
| 823 |
+
"loss": 0.0,
|
| 824 |
+
"reward": 0.9149335622787476,
|
| 825 |
+
"reward_std": 0.3469616770744324,
|
| 826 |
+
"rewards/correct_code_reward_func": 0.5416666865348816,
|
| 827 |
+
"rewards/len_reward_func": 0.3732668161392212,
|
| 828 |
+
"step": 63
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"completion_length": 76.31250381469727,
|
| 832 |
+
"epoch": 1.016,
|
| 833 |
+
"grad_norm": 1.234231990060427,
|
| 834 |
+
"kl": 0.0224609375,
|
| 835 |
+
"learning_rate": 2.532453557994827e-07,
|
| 836 |
+
"loss": 0.0,
|
| 837 |
+
"reward": 0.7154572010040283,
|
| 838 |
+
"reward_std": 0.5053917020559311,
|
| 839 |
+
"rewards/correct_code_reward_func": 0.3958333432674408,
|
| 840 |
+
"rewards/len_reward_func": 0.31962384283542633,
|
| 841 |
+
"step": 64
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"completion_length": 116.52083587646484,
|
| 845 |
+
"epoch": 1.032,
|
| 846 |
+
"grad_norm": 1.0750099724368811,
|
| 847 |
+
"kl": 0.01617431640625,
|
| 848 |
+
"learning_rate": 2.467546442005173e-07,
|
| 849 |
+
"loss": 0.0,
|
| 850 |
+
"reward": 0.7226243913173676,
|
| 851 |
+
"reward_std": 0.5376951545476913,
|
| 852 |
+
"rewards/correct_code_reward_func": 0.3750000149011612,
|
| 853 |
+
"rewards/len_reward_func": 0.34762439131736755,
|
| 854 |
+
"step": 65
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"completion_length": 116.10416793823242,
|
| 858 |
+
"epoch": 1.048,
|
| 859 |
+
"grad_norm": 1.0133791335296118,
|
| 860 |
+
"kl": 0.01983642578125,
|
| 861 |
+
"learning_rate": 2.4026612019656556e-07,
|
| 862 |
+
"loss": 0.0,
|
| 863 |
+
"reward": 0.9590997993946075,
|
| 864 |
+
"reward_std": 0.4536292999982834,
|
| 865 |
+
"rewards/correct_code_reward_func": 0.6041666865348816,
|
| 866 |
+
"rewards/len_reward_func": 0.35493315756320953,
|
| 867 |
+
"step": 66
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"completion_length": 66.625,
|
| 871 |
+
"epoch": 1.064,
|
| 872 |
+
"grad_norm": 1.2625208133395418,
|
| 873 |
+
"kl": 0.030517578125,
|
| 874 |
+
"learning_rate": 2.337841575030642e-07,
|
| 875 |
+
"loss": 0.0,
|
| 876 |
+
"reward": 0.9440672397613525,
|
| 877 |
+
"reward_std": 0.42680656909942627,
|
| 878 |
+
"rewards/correct_code_reward_func": 0.583333358168602,
|
| 879 |
+
"rewards/len_reward_func": 0.36073388159275055,
|
| 880 |
+
"step": 67
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"completion_length": 60.458335876464844,
|
| 884 |
+
"epoch": 1.08,
|
| 885 |
+
"grad_norm": 1.174665706452754,
|
| 886 |
+
"kl": 0.02490234375,
|
| 887 |
+
"learning_rate": 2.2731312541267143e-07,
|
| 888 |
+
"loss": 0.0,
|
| 889 |
+
"reward": 0.5735488831996918,
|
| 890 |
+
"reward_std": 0.48749370872974396,
|
| 891 |
+
"rewards/correct_code_reward_func": 0.2500000149011612,
|
| 892 |
+
"rewards/len_reward_func": 0.3235488831996918,
|
| 893 |
+
"step": 68
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"completion_length": 74.10416984558105,
|
| 897 |
+
"epoch": 1.096,
|
| 898 |
+
"grad_norm": 1.5260410634207344,
|
| 899 |
+
"kl": 0.029052734375,
|
| 900 |
+
"learning_rate": 2.2085738585006021e-07,
|
| 901 |
+
"loss": 0.0,
|
| 902 |
+
"reward": 1.0659485459327698,
|
| 903 |
+
"reward_std": 0.3796464204788208,
|
| 904 |
+
"rewards/correct_code_reward_func": 0.6666666865348816,
|
| 905 |
+
"rewards/len_reward_func": 0.3992818146944046,
|
| 906 |
+
"step": 69
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"completion_length": 71.89583396911621,
|
| 910 |
+
"epoch": 1.112,
|
| 911 |
+
"grad_norm": 1.059976697715823,
|
| 912 |
+
"kl": 0.0418701171875,
|
| 913 |
+
"learning_rate": 2.1442129043167873e-07,
|
| 914 |
+
"loss": 0.0,
|
| 915 |
+
"reward": 0.7276312112808228,
|
| 916 |
+
"reward_std": 0.5470257103443146,
|
| 917 |
+
"rewards/correct_code_reward_func": 0.3958333432674408,
|
| 918 |
+
"rewards/len_reward_func": 0.33179786801338196,
|
| 919 |
+
"step": 70
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"completion_length": 96.29166793823242,
|
| 923 |
+
"epoch": 1.1280000000000001,
|
| 924 |
+
"grad_norm": 0.9278761590759935,
|
| 925 |
+
"kl": 0.02056884765625,
|
| 926 |
+
"learning_rate": 2.0800917753245875e-07,
|
| 927 |
+
"loss": 0.0,
|
| 928 |
+
"reward": 0.8413068056106567,
|
| 929 |
+
"reward_std": 0.3271617591381073,
|
| 930 |
+
"rewards/correct_code_reward_func": 0.4375000298023224,
|
| 931 |
+
"rewards/len_reward_func": 0.40380676090717316,
|
| 932 |
+
"step": 71
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"completion_length": 79.33333587646484,
|
| 936 |
+
"epoch": 1.144,
|
| 937 |
+
"grad_norm": 0.9890378359951959,
|
| 938 |
+
"kl": 0.025390625,
|
| 939 |
+
"learning_rate": 2.0162536936145008e-07,
|
| 940 |
+
"loss": 0.0,
|
| 941 |
+
"reward": 0.7865794003009796,
|
| 942 |
+
"reward_std": 0.3915296047925949,
|
| 943 |
+
"rewards/correct_code_reward_func": 0.4166666716337204,
|
| 944 |
+
"rewards/len_reward_func": 0.3699127733707428,
|
| 945 |
+
"step": 72
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"completion_length": 72.95833587646484,
|
| 949 |
+
"epoch": 1.16,
|
| 950 |
+
"grad_norm": 1.3690756427464768,
|
| 951 |
+
"kl": 0.0413818359375,
|
| 952 |
+
"learning_rate": 1.9527416904835132e-07,
|
| 953 |
+
"loss": 0.0,
|
| 954 |
+
"reward": 0.9756404757499695,
|
| 955 |
+
"reward_std": 0.3930533230304718,
|
| 956 |
+
"rewards/correct_code_reward_func": 0.5416666865348816,
|
| 957 |
+
"rewards/len_reward_func": 0.4339737892150879,
|
| 958 |
+
"step": 73
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"completion_length": 79.89583396911621,
|
| 962 |
+
"epoch": 1.176,
|
| 963 |
+
"grad_norm": 1.1734737313856272,
|
| 964 |
+
"kl": 0.02801513671875,
|
| 965 |
+
"learning_rate": 1.889598577429022e-07,
|
| 966 |
+
"loss": 0.0,
|
| 967 |
+
"reward": 0.7878330945968628,
|
| 968 |
+
"reward_std": 0.4340344965457916,
|
| 969 |
+
"rewards/correct_code_reward_func": 0.4166666716337204,
|
| 970 |
+
"rewards/len_reward_func": 0.3711664080619812,
|
| 971 |
+
"step": 74
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"completion_length": 58.270835876464844,
|
| 975 |
+
"epoch": 1.192,
|
| 976 |
+
"grad_norm": 1.2871892634157702,
|
| 977 |
+
"kl": 0.0377197265625,
|
| 978 |
+
"learning_rate": 1.8268669172909136e-07,
|
| 979 |
+
"loss": 0.0,
|
| 980 |
+
"reward": 1.1113883256912231,
|
| 981 |
+
"reward_std": 0.5605219900608063,
|
| 982 |
+
"rewards/correct_code_reward_func": 0.7291666865348816,
|
| 983 |
+
"rewards/len_reward_func": 0.3822215795516968,
|
| 984 |
+
"step": 75
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"completion_length": 97.25,
|
| 988 |
+
"epoch": 1.208,
|
| 989 |
+
"grad_norm": 2.007411904832845,
|
| 990 |
+
"kl": 0.04949951171875,
|
| 991 |
+
"learning_rate": 1.7645889955612592e-07,
|
| 992 |
+
"loss": 0.0,
|
| 993 |
+
"reward": 1.0362713038921356,
|
| 994 |
+
"reward_std": 0.3777136504650116,
|
| 995 |
+
"rewards/correct_code_reward_func": 0.6666666865348816,
|
| 996 |
+
"rewards/len_reward_func": 0.3696046322584152,
|
| 997 |
+
"step": 76
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"completion_length": 42.60416793823242,
|
| 1001 |
+
"epoch": 1.224,
|
| 1002 |
+
"grad_norm": 1.196008727333628,
|
| 1003 |
+
"kl": 0.0433349609375,
|
| 1004 |
+
"learning_rate": 1.7028067918809535e-07,
|
| 1005 |
+
"loss": 0.0,
|
| 1006 |
+
"reward": 0.8968790173530579,
|
| 1007 |
+
"reward_std": 0.47050511837005615,
|
| 1008 |
+
"rewards/correct_code_reward_func": 0.520833358168602,
|
| 1009 |
+
"rewards/len_reward_func": 0.3760456293821335,
|
| 1010 |
+
"step": 77
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"completion_length": 44.97916793823242,
|
| 1014 |
+
"epoch": 1.24,
|
| 1015 |
+
"grad_norm": 1.4303810146633773,
|
| 1016 |
+
"kl": 0.0574951171875,
|
| 1017 |
+
"learning_rate": 1.6415619517425294e-07,
|
| 1018 |
+
"loss": 0.0001,
|
| 1019 |
+
"reward": 0.9696877598762512,
|
| 1020 |
+
"reward_std": 0.33823370933532715,
|
| 1021 |
+
"rewards/correct_code_reward_func": 0.5208333432674408,
|
| 1022 |
+
"rewards/len_reward_func": 0.4488544166088104,
|
| 1023 |
+
"step": 78
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"completion_length": 66.0625,
|
| 1027 |
+
"epoch": 1.256,
|
| 1028 |
+
"grad_norm": 1.5543919802377633,
|
| 1029 |
+
"kl": 0.03875732421875,
|
| 1030 |
+
"learning_rate": 1.5808957584181994e-07,
|
| 1031 |
+
"loss": 0.0,
|
| 1032 |
+
"reward": 0.9006170630455017,
|
| 1033 |
+
"reward_std": 0.37997904419898987,
|
| 1034 |
+
"rewards/correct_code_reward_func": 0.4791666716337204,
|
| 1035 |
+
"rewards/len_reward_func": 0.4214503914117813,
|
| 1036 |
+
"step": 79
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"completion_length": 76.06250381469727,
|
| 1040 |
+
"epoch": 1.272,
|
| 1041 |
+
"grad_norm": 1.8800678172876588,
|
| 1042 |
+
"kl": 0.047607421875,
|
| 1043 |
+
"learning_rate": 1.5208491051320744e-07,
|
| 1044 |
+
"loss": 0.0,
|
| 1045 |
+
"reward": 0.9962214827537537,
|
| 1046 |
+
"reward_std": 0.3772743344306946,
|
| 1047 |
+
"rewards/correct_code_reward_func": 0.5625000149011612,
|
| 1048 |
+
"rewards/len_reward_func": 0.4337214529514313,
|
| 1049 |
+
"step": 80
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"completion_length": 43.27083396911621,
|
| 1053 |
+
"epoch": 1.288,
|
| 1054 |
+
"grad_norm": 1.3749781365688138,
|
| 1055 |
+
"kl": 0.055908203125,
|
| 1056 |
+
"learning_rate": 1.461462467495284e-07,
|
| 1057 |
+
"loss": 0.0001,
|
| 1058 |
+
"reward": 1.1482934355735779,
|
| 1059 |
+
"reward_std": 0.4654877036809921,
|
| 1060 |
+
"rewards/correct_code_reward_func": 0.7291666865348816,
|
| 1061 |
+
"rewards/len_reward_func": 0.41912680864334106,
|
| 1062 |
+
"step": 81
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"completion_length": 64.25000381469727,
|
| 1066 |
+
"epoch": 1.304,
|
| 1067 |
+
"grad_norm": 1.2603902864707686,
|
| 1068 |
+
"kl": 0.041259765625,
|
| 1069 |
+
"learning_rate": 1.4027758762226107e-07,
|
| 1070 |
+
"loss": 0.0,
|
| 1071 |
+
"reward": 0.8633248507976532,
|
| 1072 |
+
"reward_std": 0.44106219708919525,
|
| 1073 |
+
"rewards/correct_code_reward_func": 0.4583333432674408,
|
| 1074 |
+
"rewards/len_reward_func": 0.40499147772789,
|
| 1075 |
+
"step": 82
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"completion_length": 84.87500190734863,
|
| 1079 |
+
"epoch": 1.32,
|
| 1080 |
+
"grad_norm": 1.3553050400485256,
|
| 1081 |
+
"kl": 0.037109375,
|
| 1082 |
+
"learning_rate": 1.3448288901490092e-07,
|
| 1083 |
+
"loss": 0.0,
|
| 1084 |
+
"reward": 0.930885374546051,
|
| 1085 |
+
"reward_std": 0.28970160335302353,
|
| 1086 |
+
"rewards/correct_code_reward_func": 0.4791666865348816,
|
| 1087 |
+
"rewards/len_reward_func": 0.45171867311000824,
|
| 1088 |
+
"step": 83
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"completion_length": 44.12500190734863,
|
| 1092 |
+
"epoch": 1.336,
|
| 1093 |
+
"grad_norm": 1.3471283821112554,
|
| 1094 |
+
"kl": 0.05126953125,
|
| 1095 |
+
"learning_rate": 1.2876605695642084e-07,
|
| 1096 |
+
"loss": 0.0001,
|
| 1097 |
+
"reward": 0.8797213435173035,
|
| 1098 |
+
"reward_std": 0.4040851444005966,
|
| 1099 |
+
"rewards/correct_code_reward_func": 0.4375000149011612,
|
| 1100 |
+
"rewards/len_reward_func": 0.44222137331962585,
|
| 1101 |
+
"step": 84
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"completion_length": 77.27083587646484,
|
| 1105 |
+
"epoch": 1.3519999999999999,
|
| 1106 |
+
"grad_norm": 1.3204773489251114,
|
| 1107 |
+
"kl": 0.0401611328125,
|
| 1108 |
+
"learning_rate": 1.231309449883361e-07,
|
| 1109 |
+
"loss": 0.0,
|
| 1110 |
+
"reward": 0.9848110675811768,
|
| 1111 |
+
"reward_std": 0.5164197236299515,
|
| 1112 |
+
"rewards/correct_code_reward_func": 0.5833333432674408,
|
| 1113 |
+
"rewards/len_reward_func": 0.40147776901721954,
|
| 1114 |
+
"step": 85
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"completion_length": 58.729169845581055,
|
| 1118 |
+
"epoch": 1.3679999999999999,
|
| 1119 |
+
"grad_norm": 1.6764054808630413,
|
| 1120 |
+
"kl": 0.0513916015625,
|
| 1121 |
+
"learning_rate": 1.1758135156715041e-07,
|
| 1122 |
+
"loss": 0.0001,
|
| 1123 |
+
"reward": 1.1368677616119385,
|
| 1124 |
+
"reward_std": 0.3856023848056793,
|
| 1125 |
+
"rewards/correct_code_reward_func": 0.7708333432674408,
|
| 1126 |
+
"rewards/len_reward_func": 0.3660343587398529,
|
| 1127 |
+
"step": 86
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"completion_length": 90.02083396911621,
|
| 1131 |
+
"epoch": 1.384,
|
| 1132 |
+
"grad_norm": 1.1184416966735806,
|
| 1133 |
+
"kl": 0.03411865234375,
|
| 1134 |
+
"learning_rate": 1.1212101750393235e-07,
|
| 1135 |
+
"loss": 0.0,
|
| 1136 |
+
"reward": 1.00173819065094,
|
| 1137 |
+
"reward_std": 0.49202577769756317,
|
| 1138 |
+
"rewards/correct_code_reward_func": 0.5625000298023224,
|
| 1139 |
+
"rewards/len_reward_func": 0.43923819065093994,
|
| 1140 |
+
"step": 87
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"completion_length": 46.60416793823242,
|
| 1144 |
+
"epoch": 1.4,
|
| 1145 |
+
"grad_norm": 1.1504631747132115,
|
| 1146 |
+
"kl": 0.0496826171875,
|
| 1147 |
+
"learning_rate": 1.0675362344274952e-07,
|
| 1148 |
+
"loss": 0.0,
|
| 1149 |
+
"reward": 0.9823802709579468,
|
| 1150 |
+
"reward_std": 0.46341855823993683,
|
| 1151 |
+
"rewards/correct_code_reward_func": 0.6041666865348816,
|
| 1152 |
+
"rewards/len_reward_func": 0.3782135844230652,
|
| 1153 |
+
"step": 88
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"completion_length": 59.937503814697266,
|
| 1157 |
+
"epoch": 1.416,
|
| 1158 |
+
"grad_norm": 1.3546823329176554,
|
| 1159 |
+
"kl": 0.0450439453125,
|
| 1160 |
+
"learning_rate": 1.0148278737965844e-07,
|
| 1161 |
+
"loss": 0.0,
|
| 1162 |
+
"reward": 1.116898536682129,
|
| 1163 |
+
"reward_std": 0.3632017821073532,
|
| 1164 |
+
"rewards/correct_code_reward_func": 0.6250000298023224,
|
| 1165 |
+
"rewards/len_reward_func": 0.4918985068798065,
|
| 1166 |
+
"step": 89
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"completion_length": 38.08333396911621,
|
| 1170 |
+
"epoch": 1.432,
|
| 1171 |
+
"grad_norm": 1.4742606396398155,
|
| 1172 |
+
"kl": 0.06396484375,
|
| 1173 |
+
"learning_rate": 9.631206222392479e-08,
|
| 1174 |
+
"loss": 0.0001,
|
| 1175 |
+
"reward": 1.079964131116867,
|
| 1176 |
+
"reward_std": 0.47177985310554504,
|
| 1177 |
+
"rewards/correct_code_reward_func": 0.6666666865348816,
|
| 1178 |
+
"rewards/len_reward_func": 0.4132973849773407,
|
| 1179 |
+
"step": 90
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"completion_length": 72.18750190734863,
|
| 1183 |
+
"epoch": 1.448,
|
| 1184 |
+
"grad_norm": 1.4885040482346985,
|
| 1185 |
+
"kl": 0.0491943359375,
|
| 1186 |
+
"learning_rate": 9.124493340311537e-08,
|
| 1187 |
+
"loss": 0.0,
|
| 1188 |
+
"reward": 1.0042789578437805,
|
| 1189 |
+
"reward_std": 0.3533863425254822,
|
| 1190 |
+
"rewards/correct_code_reward_func": 0.5625000298023224,
|
| 1191 |
+
"rewards/len_reward_func": 0.44177892804145813,
|
| 1192 |
+
"step": 91
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"completion_length": 69.35416984558105,
|
| 1196 |
+
"epoch": 1.464,
|
| 1197 |
+
"grad_norm": 1.1925308190836468,
|
| 1198 |
+
"kl": 0.0478515625,
|
| 1199 |
+
"learning_rate": 8.628481651367875e-08,
|
| 1200 |
+
"loss": 0.0,
|
| 1201 |
+
"reward": 1.0887993574142456,
|
| 1202 |
+
"reward_std": 0.5487662255764008,
|
| 1203 |
+
"rewards/correct_code_reward_func": 0.6666666865348816,
|
| 1204 |
+
"rewards/len_reward_func": 0.422132670879364,
|
| 1205 |
+
"step": 92
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"completion_length": 48.41666793823242,
|
| 1209 |
+
"epoch": 1.48,
|
| 1210 |
+
"grad_norm": 1.6549689653260358,
|
| 1211 |
+
"kl": 0.051025390625,
|
| 1212 |
+
"learning_rate": 8.143505501859551e-08,
|
| 1213 |
+
"loss": 0.0001,
|
| 1214 |
+
"reward": 0.920165479183197,
|
| 1215 |
+
"reward_std": 0.415915310382843,
|
| 1216 |
+
"rewards/correct_code_reward_func": 0.5208333432674408,
|
| 1217 |
+
"rewards/len_reward_func": 0.39933212101459503,
|
| 1218 |
+
"step": 93
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"completion_length": 75.87500381469727,
|
| 1222 |
+
"epoch": 1.496,
|
| 1223 |
+
"grad_norm": 1.586423514577547,
|
| 1224 |
+
"kl": 0.053466796875,
|
| 1225 |
+
"learning_rate": 7.669891799365282e-08,
|
| 1226 |
+
"loss": 0.0001,
|
| 1227 |
+
"reward": 0.8828703761100769,
|
| 1228 |
+
"reward_std": 0.49832169711589813,
|
| 1229 |
+
"rewards/correct_code_reward_func": 0.4375000149011612,
|
| 1230 |
+
"rewards/len_reward_func": 0.4453703910112381,
|
| 1231 |
+
"step": 94
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"completion_length": 79.06250190734863,
|
| 1235 |
+
"epoch": 1.512,
|
| 1236 |
+
"grad_norm": 1.7029251090199986,
|
| 1237 |
+
"kl": 0.0489501953125,
|
| 1238 |
+
"learning_rate": 7.207959792385998e-08,
|
| 1239 |
+
"loss": 0.0,
|
| 1240 |
+
"reward": 1.0712128281593323,
|
| 1241 |
+
"reward_std": 0.39892764389514923,
|
| 1242 |
+
"rewards/correct_code_reward_func": 0.6041666865348816,
|
| 1243 |
+
"rewards/len_reward_func": 0.4670460820198059,
|
| 1244 |
+
"step": 95
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"completion_length": 77.66666984558105,
|
| 1248 |
+
"epoch": 1.528,
|
| 1249 |
+
"grad_norm": 1.068625601865938,
|
| 1250 |
+
"kl": 0.03729248046875,
|
| 1251 |
+
"learning_rate": 6.758020855149249e-08,
|
| 1252 |
+
"loss": 0.0,
|
| 1253 |
+
"reward": 0.839818924665451,
|
| 1254 |
+
"reward_std": 0.4449944496154785,
|
| 1255 |
+
"rewards/correct_code_reward_func": 0.3958333432674408,
|
| 1256 |
+
"rewards/len_reward_func": 0.44398559629917145,
|
| 1257 |
+
"step": 96
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"completion_length": 46.00000190734863,
|
| 1261 |
+
"epoch": 1.544,
|
| 1262 |
+
"grad_norm": 1.6059040022360391,
|
| 1263 |
+
"kl": 0.0543212890625,
|
| 1264 |
+
"learning_rate": 6.320378277721342e-08,
|
| 1265 |
+
"loss": 0.0001,
|
| 1266 |
+
"reward": 0.8210516273975372,
|
| 1267 |
+
"reward_std": 0.34706588089466095,
|
| 1268 |
+
"rewards/correct_code_reward_func": 0.3958333432674408,
|
| 1269 |
+
"rewards/len_reward_func": 0.42521825432777405,
|
| 1270 |
+
"step": 97
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"completion_length": 47.35416793823242,
|
| 1274 |
+
"epoch": 1.56,
|
| 1275 |
+
"grad_norm": 1.0536706533475677,
|
| 1276 |
+
"kl": 0.06640625,
|
| 1277 |
+
"learning_rate": 5.895327061568775e-08,
|
| 1278 |
+
"loss": 0.0001,
|
| 1279 |
+
"reward": 0.9913617670536041,
|
| 1280 |
+
"reward_std": 0.24820256233215332,
|
| 1281 |
+
"rewards/correct_code_reward_func": 0.5208333432674408,
|
| 1282 |
+
"rewards/len_reward_func": 0.47052840888500214,
|
| 1283 |
+
"step": 98
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"completion_length": 34.83333396911621,
|
| 1287 |
+
"epoch": 1.576,
|
| 1288 |
+
"grad_norm": 1.7119720585841143,
|
| 1289 |
+
"kl": 0.0626220703125,
|
| 1290 |
+
"learning_rate": 5.483153720706798e-08,
|
| 1291 |
+
"loss": 0.0001,
|
| 1292 |
+
"reward": 0.8738410770893097,
|
| 1293 |
+
"reward_std": 0.5085435211658478,
|
| 1294 |
+
"rewards/correct_code_reward_func": 0.4375000149011612,
|
| 1295 |
+
"rewards/len_reward_func": 0.4363410472869873,
|
| 1296 |
+
"step": 99
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"completion_length": 57.04166793823242,
|
| 1300 |
+
"epoch": 1.592,
|
| 1301 |
+
"grad_norm": 1.3955252614043234,
|
| 1302 |
+
"kl": 0.036865234375,
|
| 1303 |
+
"learning_rate": 5.0841360885690996e-08,
|
| 1304 |
+
"loss": 0.0,
|
| 1305 |
+
"reward": 1.1610032320022583,
|
| 1306 |
+
"reward_std": 0.3681405633687973,
|
| 1307 |
+
"rewards/correct_code_reward_func": 0.6666666865348816,
|
| 1308 |
+
"rewards/len_reward_func": 0.4943365752696991,
|
| 1309 |
+
"step": 100
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"completion_length": 71.41666984558105,
|
| 1313 |
+
"epoch": 1.608,
|
| 1314 |
+
"grad_norm": 0.9360155945219467,
|
| 1315 |
+
"kl": 0.055419921875,
|
| 1316 |
+
"learning_rate": 4.698543130728755e-08,
|
| 1317 |
+
"loss": 0.0001,
|
| 1318 |
+
"reward": 1.013959676027298,
|
| 1319 |
+
"reward_std": 0.39147183299064636,
|
| 1320 |
+
"rewards/correct_code_reward_func": 0.625,
|
| 1321 |
+
"rewards/len_reward_func": 0.3889596611261368,
|
| 1322 |
+
"step": 101
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"completion_length": 43.5625,
|
| 1326 |
+
"epoch": 1.624,
|
| 1327 |
+
"grad_norm": 1.620965651402336,
|
| 1328 |
+
"kl": 0.054443359375,
|
| 1329 |
+
"learning_rate": 4.326634763596784e-08,
|
| 1330 |
+
"loss": 0.0001,
|
| 1331 |
+
"reward": 0.8748385310173035,
|
| 1332 |
+
"reward_std": 0.3259390592575073,
|
| 1333 |
+
"rewards/correct_code_reward_func": 0.4166666679084301,
|
| 1334 |
+
"rewards/len_reward_func": 0.45817187428474426,
|
| 1335 |
+
"step": 102
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"completion_length": 54.125003814697266,
|
| 1339 |
+
"epoch": 1.6400000000000001,
|
| 1340 |
+
"grad_norm": 2.035180505859693,
|
| 1341 |
+
"kl": 0.056640625,
|
| 1342 |
+
"learning_rate": 3.968661679220467e-08,
|
| 1343 |
+
"loss": 0.0001,
|
| 1344 |
+
"reward": 1.2046028971672058,
|
| 1345 |
+
"reward_std": 0.4274601340293884,
|
| 1346 |
+
"rewards/correct_code_reward_func": 0.7708333730697632,
|
| 1347 |
+
"rewards/len_reward_func": 0.433769553899765,
|
| 1348 |
+
"step": 103
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"completion_length": 37.72916793823242,
|
| 1352 |
+
"epoch": 1.6560000000000001,
|
| 1353 |
+
"grad_norm": 1.3110502429273858,
|
| 1354 |
+
"kl": 0.07470703125,
|
| 1355 |
+
"learning_rate": 3.624865176299499e-08,
|
| 1356 |
+
"loss": 0.0001,
|
| 1357 |
+
"reward": 1.3125,
|
| 1358 |
+
"reward_std": 0.2893980145454407,
|
| 1359 |
+
"rewards/correct_code_reward_func": 0.8333333432674408,
|
| 1360 |
+
"rewards/len_reward_func": 0.4791666716337204,
|
| 1361 |
+
"step": 104
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"completion_length": 85.20833587646484,
|
| 1365 |
+
"epoch": 1.6720000000000002,
|
| 1366 |
+
"grad_norm": 0.7548512888732658,
|
| 1367 |
+
"kl": 0.0379638671875,
|
| 1368 |
+
"learning_rate": 3.295476997533905e-08,
|
| 1369 |
+
"loss": 0.0,
|
| 1370 |
+
"reward": 0.8693651854991913,
|
| 1371 |
+
"reward_std": 0.467289537191391,
|
| 1372 |
+
"rewards/correct_code_reward_func": 0.4166666865348816,
|
| 1373 |
+
"rewards/len_reward_func": 0.4526985138654709,
|
| 1374 |
+
"step": 105
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"completion_length": 60.72916793823242,
|
| 1378 |
+
"epoch": 1.688,
|
| 1379 |
+
"grad_norm": 0.7605372623815415,
|
| 1380 |
+
"kl": 0.03662109375,
|
| 1381 |
+
"learning_rate": 2.980719173413396e-08,
|
| 1382 |
+
"loss": 0.0,
|
| 1383 |
+
"reward": 1.058800995349884,
|
| 1384 |
+
"reward_std": 0.4301101863384247,
|
| 1385 |
+
"rewards/correct_code_reward_func": 0.5625000298023224,
|
| 1386 |
+
"rewards/len_reward_func": 0.49630098044872284,
|
| 1387 |
+
"step": 106
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"completion_length": 49.08333396911621,
|
| 1391 |
+
"epoch": 1.704,
|
| 1392 |
+
"grad_norm": 1.0207996418513696,
|
| 1393 |
+
"kl": 0.07421875,
|
| 1394 |
+
"learning_rate": 2.680803872553408e-08,
|
| 1395 |
+
"loss": 0.0001,
|
| 1396 |
+
"reward": 0.9703012108802795,
|
| 1397 |
+
"reward_std": 0.23036788403987885,
|
| 1398 |
+
"rewards/correct_code_reward_func": 0.5208333432674408,
|
| 1399 |
+
"rewards/len_reward_func": 0.44946780800819397,
|
| 1400 |
+
"step": 107
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"completion_length": 57.62500190734863,
|
| 1404 |
+
"epoch": 1.72,
|
| 1405 |
+
"grad_norm": 1.5035291383359601,
|
| 1406 |
+
"kl": 0.0594482421875,
|
| 1407 |
+
"learning_rate": 2.395933258678745e-08,
|
| 1408 |
+
"loss": 0.0001,
|
| 1409 |
+
"reward": 1.1501469016075134,
|
| 1410 |
+
"reward_std": 0.34404293447732925,
|
| 1411 |
+
"rewards/correct_code_reward_func": 0.7083333432674408,
|
| 1412 |
+
"rewards/len_reward_func": 0.44181351363658905,
|
| 1413 |
+
"step": 108
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"completion_length": 31.312501907348633,
|
| 1417 |
+
"epoch": 1.736,
|
| 1418 |
+
"grad_norm": 1.6167712080198857,
|
| 1419 |
+
"kl": 0.0712890625,
|
| 1420 |
+
"learning_rate": 2.1262993543511715e-08,
|
| 1421 |
+
"loss": 0.0001,
|
| 1422 |
+
"reward": 1.3109871745109558,
|
| 1423 |
+
"reward_std": 0.3160089999437332,
|
| 1424 |
+
"rewards/correct_code_reward_func": 0.875,
|
| 1425 |
+
"rewards/len_reward_func": 0.4359871447086334,
|
| 1426 |
+
"step": 109
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"completion_length": 87.22916793823242,
|
| 1430 |
+
"epoch": 1.752,
|
| 1431 |
+
"grad_norm": 1.2313006391246961,
|
| 1432 |
+
"kl": 0.0474853515625,
|
| 1433 |
+
"learning_rate": 1.872083911532907e-08,
|
| 1434 |
+
"loss": 0.0,
|
| 1435 |
+
"reward": 0.8871810734272003,
|
| 1436 |
+
"reward_std": 0.43447698652744293,
|
| 1437 |
+
"rewards/correct_code_reward_func": 0.4166666865348816,
|
| 1438 |
+
"rewards/len_reward_func": 0.4705143868923187,
|
| 1439 |
+
"step": 110
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"completion_length": 38.52083396911621,
|
| 1443 |
+
"epoch": 1.768,
|
| 1444 |
+
"grad_norm": 1.1333291307805236,
|
| 1445 |
+
"kl": 0.0531005859375,
|
| 1446 |
+
"learning_rate": 1.6334582890731697e-08,
|
| 1447 |
+
"loss": 0.0001,
|
| 1448 |
+
"reward": 1.3089049458503723,
|
| 1449 |
+
"reward_std": 0.39449335634708405,
|
| 1450 |
+
"rewards/correct_code_reward_func": 0.8541666865348816,
|
| 1451 |
+
"rewards/len_reward_func": 0.45473821461200714,
|
| 1452 |
+
"step": 111
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"completion_length": 99.85416984558105,
|
| 1456 |
+
"epoch": 1.784,
|
| 1457 |
+
"grad_norm": 1.1649171723265364,
|
| 1458 |
+
"kl": 0.04095458984375,
|
| 1459 |
+
"learning_rate": 1.4105833372004523e-08,
|
| 1460 |
+
"loss": 0.0,
|
| 1461 |
+
"reward": 0.8114411234855652,
|
| 1462 |
+
"reward_std": 0.3455437570810318,
|
| 1463 |
+
"rewards/correct_code_reward_func": 0.4166666716337204,
|
| 1464 |
+
"rewards/len_reward_func": 0.3947744071483612,
|
| 1465 |
+
"step": 112
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"completion_length": 62.97916793823242,
|
| 1469 |
+
"epoch": 1.8,
|
| 1470 |
+
"grad_norm": 1.3270422235967783,
|
| 1471 |
+
"kl": 0.042236328125,
|
| 1472 |
+
"learning_rate": 1.2036092890982619e-08,
|
| 1473 |
+
"loss": 0.0,
|
| 1474 |
+
"reward": 0.8704836964607239,
|
| 1475 |
+
"reward_std": 0.37642528116703033,
|
| 1476 |
+
"rewards/correct_code_reward_func": 0.3958333432674408,
|
| 1477 |
+
"rewards/len_reward_func": 0.4746503531932831,
|
| 1478 |
+
"step": 113
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"completion_length": 61.854169845581055,
|
| 1482 |
+
"epoch": 1.8159999999999998,
|
| 1483 |
+
"grad_norm": 1.47111563325612,
|
| 1484 |
+
"kl": 0.0511474609375,
|
| 1485 |
+
"learning_rate": 1.0126756596375685e-08,
|
| 1486 |
+
"loss": 0.0001,
|
| 1487 |
+
"reward": 1.0473325848579407,
|
| 1488 |
+
"reward_std": 0.5612048506736755,
|
| 1489 |
+
"rewards/correct_code_reward_func": 0.5833333730697632,
|
| 1490 |
+
"rewards/len_reward_func": 0.4639992117881775,
|
| 1491 |
+
"step": 114
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"completion_length": 44.22916793823242,
|
| 1495 |
+
"epoch": 1.8319999999999999,
|
| 1496 |
+
"grad_norm": 1.3437447624267829,
|
| 1497 |
+
"kl": 0.0599365234375,
|
| 1498 |
+
"learning_rate": 8.379111513340753e-09,
|
| 1499 |
+
"loss": 0.0001,
|
| 1500 |
+
"reward": 0.9941486120223999,
|
| 1501 |
+
"reward_std": 0.5003164112567902,
|
| 1502 |
+
"rewards/correct_code_reward_func": 0.5625000149011612,
|
| 1503 |
+
"rewards/len_reward_func": 0.4316485822200775,
|
| 1504 |
+
"step": 115
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"completion_length": 51.45833396911621,
|
| 1508 |
+
"epoch": 1.8479999999999999,
|
| 1509 |
+
"grad_norm": 1.3897204537445644,
|
| 1510 |
+
"kl": 0.0416259765625,
|
| 1511 |
+
"learning_rate": 6.7943356759381785e-09,
|
| 1512 |
+
"loss": 0.0,
|
| 1513 |
+
"reward": 1.0625000596046448,
|
| 1514 |
+
"reward_std": 0.4191845655441284,
|
| 1515 |
+
"rewards/correct_code_reward_func": 0.5833333730697632,
|
| 1516 |
+
"rewards/len_reward_func": 0.4791666716337204,
|
| 1517 |
+
"step": 116
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"completion_length": 64.45833587646484,
|
| 1521 |
+
"epoch": 1.8639999999999999,
|
| 1522 |
+
"grad_norm": 2.0280580958010224,
|
| 1523 |
+
"kl": 0.07666015625,
|
| 1524 |
+
"learning_rate": 5.373497333054616e-09,
|
| 1525 |
+
"loss": 0.0001,
|
| 1526 |
+
"reward": 0.9957386255264282,
|
| 1527 |
+
"reward_std": 0.49953845143318176,
|
| 1528 |
+
"rewards/correct_code_reward_func": 0.5,
|
| 1529 |
+
"rewards/len_reward_func": 0.4957386255264282,
|
| 1530 |
+
"step": 117
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"completion_length": 61.833335876464844,
|
| 1534 |
+
"epoch": 1.88,
|
| 1535 |
+
"grad_norm": 1.4575505392586763,
|
| 1536 |
+
"kl": 0.06689453125,
|
| 1537 |
+
"learning_rate": 4.117554228329406e-09,
|
| 1538 |
+
"loss": 0.0001,
|
| 1539 |
+
"reward": 1.2341400384902954,
|
| 1540 |
+
"reward_std": 0.472938671708107,
|
| 1541 |
+
"rewards/correct_code_reward_func": 0.8125000298023224,
|
| 1542 |
+
"rewards/len_reward_func": 0.421640083193779,
|
| 1543 |
+
"step": 118
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"completion_length": 54.43750286102295,
|
| 1547 |
+
"epoch": 1.896,
|
| 1548 |
+
"grad_norm": 1.6449474004844165,
|
| 1549 |
+
"kl": 0.0567626953125,
|
| 1550 |
+
"learning_rate": 3.0273529545687125e-09,
|
| 1551 |
+
"loss": 0.0001,
|
| 1552 |
+
"reward": 1.0164108276367188,
|
| 1553 |
+
"reward_std": 0.4117661267518997,
|
| 1554 |
+
"rewards/correct_code_reward_func": 0.5208333432674408,
|
| 1555 |
+
"rewards/len_reward_func": 0.49557754397392273,
|
| 1556 |
+
"step": 119
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"completion_length": 59.8125,
|
| 1560 |
+
"epoch": 1.912,
|
| 1561 |
+
"grad_norm": 1.2181441627902125,
|
| 1562 |
+
"kl": 0.052978515625,
|
| 1563 |
+
"learning_rate": 2.1036283830834224e-09,
|
| 1564 |
+
"loss": 0.0001,
|
| 1565 |
+
"reward": 1.1736499071121216,
|
| 1566 |
+
"reward_std": 0.41473129391670227,
|
| 1567 |
+
"rewards/correct_code_reward_func": 0.7083333432674408,
|
| 1568 |
+
"rewards/len_reward_func": 0.465316578745842,
|
| 1569 |
+
"step": 120
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"completion_length": 29.125001907348633,
|
| 1573 |
+
"epoch": 1.928,
|
| 1574 |
+
"grad_norm": 1.6515660045888947,
|
| 1575 |
+
"kl": 0.0712890625,
|
| 1576 |
+
"learning_rate": 1.347003168334665e-09,
|
| 1577 |
+
"loss": 0.0001,
|
| 1578 |
+
"reward": 1.2279411554336548,
|
| 1579 |
+
"reward_std": 0.212213896214962,
|
| 1580 |
+
"rewards/correct_code_reward_func": 0.7500000298023224,
|
| 1581 |
+
"rewards/len_reward_func": 0.477941170334816,
|
| 1582 |
+
"step": 121
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"completion_length": 40.29166793823242,
|
| 1586 |
+
"epoch": 1.944,
|
| 1587 |
+
"grad_norm": 2.3395660695095883,
|
| 1588 |
+
"kl": 0.0679931640625,
|
| 1589 |
+
"learning_rate": 7.579873282216598e-10,
|
| 1590 |
+
"loss": 0.0001,
|
| 1591 |
+
"reward": 0.9375000298023224,
|
| 1592 |
+
"reward_std": 0.37034808099269867,
|
| 1593 |
+
"rewards/correct_code_reward_func": 0.4583333432674408,
|
| 1594 |
+
"rewards/len_reward_func": 0.4791666716337204,
|
| 1595 |
+
"step": 122
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"completion_length": 70.83333396911621,
|
| 1599 |
+
"epoch": 1.96,
|
| 1600 |
+
"grad_norm": 1.119825823829922,
|
| 1601 |
+
"kl": 0.057861328125,
|
| 1602 |
+
"learning_rate": 3.3697790029424413e-10,
|
| 1603 |
+
"loss": 0.0001,
|
| 1604 |
+
"reward": 1.2083333730697632,
|
| 1605 |
+
"reward_std": 0.39485183358192444,
|
| 1606 |
+
"rewards/correct_code_reward_func": 0.7083333730697632,
|
| 1607 |
+
"rewards/len_reward_func": 0.5,
|
| 1608 |
+
"step": 123
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"completion_length": 52.750003814697266,
|
| 1612 |
+
"epoch": 1.976,
|
| 1613 |
+
"grad_norm": 2.4270573554119026,
|
| 1614 |
+
"kl": 0.0513916015625,
|
| 1615 |
+
"learning_rate": 8.425867412190091e-11,
|
| 1616 |
+
"loss": 0.0001,
|
| 1617 |
+
"reward": 1.1584753692150116,
|
| 1618 |
+
"reward_std": 0.37459391355514526,
|
| 1619 |
+
"rewards/correct_code_reward_func": 0.7083333432674408,
|
| 1620 |
+
"rewards/len_reward_func": 0.4501419961452484,
|
| 1621 |
+
"step": 124
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"completion_length": 95.9375,
|
| 1625 |
+
"epoch": 1.992,
|
| 1626 |
+
"grad_norm": 1.1107259696791534,
|
| 1627 |
+
"kl": 0.038330078125,
|
| 1628 |
+
"learning_rate": 0.0,
|
| 1629 |
+
"loss": 0.0,
|
| 1630 |
+
"reward": 1.0544261932373047,
|
| 1631 |
+
"reward_std": 0.4110799580812454,
|
| 1632 |
+
"rewards/correct_code_reward_func": 0.6041666865348816,
|
| 1633 |
+
"rewards/len_reward_func": 0.4502594470977783,
|
| 1634 |
+
"step": 125
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.992,
|
| 1638 |
+
"step": 125,
|
| 1639 |
+
"total_flos": 0.0,
|
| 1640 |
+
"train_loss": 2.596636219443127e-05,
|
| 1641 |
+
"train_runtime": 4869.9523,
|
| 1642 |
+
"train_samples_per_second": 0.154,
|
| 1643 |
+
"train_steps_per_second": 0.026
|
| 1644 |
+
}
|
| 1645 |
+
],
|
| 1646 |
+
"logging_steps": 1,
|
| 1647 |
+
"max_steps": 125,
|
| 1648 |
+
"num_input_tokens_seen": 0,
|
| 1649 |
+
"num_train_epochs": 3,
|
| 1650 |
+
"save_steps": 25,
|
| 1651 |
+
"stateful_callbacks": {
|
| 1652 |
+
"TrainerControl": {
|
| 1653 |
+
"args": {
|
| 1654 |
+
"should_epoch_stop": false,
|
| 1655 |
+
"should_evaluate": false,
|
| 1656 |
+
"should_log": false,
|
| 1657 |
+
"should_save": true,
|
| 1658 |
+
"should_training_stop": true
|
| 1659 |
+
},
|
| 1660 |
+
"attributes": {}
|
| 1661 |
+
}
|
| 1662 |
+
},
|
| 1663 |
+
"total_flos": 0.0,
|
| 1664 |
+
"train_batch_size": 1,
|
| 1665 |
+
"trial_name": null,
|
| 1666 |
+
"trial_params": null
|
| 1667 |
+
}
|