Harsha901/Qwen3_4B-GRPO-Math
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: Harsha901/Qwen3_4B-GRPO-Math
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-04 05:34:37 +08:00
commit db2a248d72
13 changed files with 152234 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
chat_template.jinja Normal file
View File

@@ -0,0 +1,4 @@
{% if messages[0]['role'] == 'system' %}{{ messages[0]['content'] + eos_token }}{% set loop_messages = messages[1:] %}{% else %}{{ 'You are given a problem.
Think about the problem and provide your working out.
Place it between <start_working_out> and <end_working_out>.
Then, provide your solution between <SOLUTION></SOLUTION>' + eos_token }}{% set loop_messages = messages %}{% endif %}{% for message in loop_messages %}{% if message['role'] == 'user' %}{{ message['content'] }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<start_working_out>' }}{% endif %}