初始化项目,由ModelHub XC社区提供模型

Model: hkr04/qwen3-4b-grpo-dapo17k
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-07-23 20:00:20 +08:00
commit b7312ceffb
14 changed files with 152323 additions and 0 deletions

14
README.md Normal file
View File

@@ -0,0 +1,14 @@
---
datasets:
- BytedTsinghua-SIA/DAPO-Math-17k
base_model:
- Qwen/Qwen3-4B
---
Trained on DAPO-Math-17k using GRPO implemented by VeRL.
- Batch Size: 32
- Group Size: 8
- Epoch: 1
- Step: 559
- Max Response Length: 8192