diff options
| author | Lynx-Chen <garbage-collect@outlook.com> | 2025-05-28 01:18:37 +0800 |
|---|---|---|
| committer | Lynx-Chen <garbage-collect@outlook.com> | 2025-05-28 01:18:37 +0800 |
| commit | ad7ac69b6c31c290f27dd1dc6b2e6efb42e89142 (patch) | |
| tree | a0a8876ba94b1f9d2799e23899f96c876afc07a1 /README.md | |
| parent | 8a7e0e9e47488e7f16d6aa59de37a637b1cdaa8a (diff) | |
feat: update readme
Diffstat (limited to 'README.md')
| -rw-r--r-- | README.md | 9 |
1 files changed, 0 insertions, 9 deletions
@@ -2,7 +2,6 @@ <a href='https://arxiv.org/abs/2505.20282'><img src='https://img.shields.io/badge/arXiv-2505.20282-b31b1b.svg'></a> ---- ### Installation @@ -19,14 +18,10 @@ accelerate launch train.py \ --model_name Qwen2.5-Math-7B \ --model_path /path/to/Qwen2.5-Math-7B \ --train_data dataset/1shot_rlvr/pi1_r1280.parquet \ - --eval_data dataset/1shot_rlvr/pi1_r1280.parquet \ --effective_batch 64 \ --micro_batch_size auto \ --temperature 0.5 \ --learning_rate 2e-5 \ - --eval_steps 5 \ - --eval_batch_size 4 \ - --eval_size 10 \ --max_steps 1000 \ --log_steps 1 \ --save_steps 1 \ @@ -43,14 +38,10 @@ accelerate launch train.py \ --model_name Qwen2.5-Math-7B \ --model_path /path/to/Qwen2.5-Math-7B \ --train_data dataset/numina/numina_00.parquet \ - --eval_data dataset/numina/numina_01.parquet \ --effective_batch 64 \ --micro_batch_size auto \ --temperature 0.5 \ --learning_rate 2e-5 \ - --eval_steps 5 \ - --eval_batch_size 4 \ - --eval_size 10 \ --max_steps 1000 \ --log_steps 1 \ --save_steps 1 \ |