open-thought/reasoning-gym
1
0
Fork 0
mirror of https://github.com/open-thought/reasoning-gym.git synced 2026-04-30 17:40:45 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

impl conditional reward

Browse source
This commit is contained in:
Oliver 2025-04-24 19:36:30 +01:00
parent 450d3dcfa4
commit 830ac3e10a
13 changed files with 52 additions and 24 deletions
Download patch file Download diff file Expand all files Collapse all files

5
training/configs/external_generalisation/math_curriculum_qwen_7b.yaml
View file

@ -52,11 +52,8 @@ curriculum:
'*': 0
reward:
use_accuracy: True
conditional_reward: True # Only provide a reward at all if the response format is correct
secondary_rewards:
- name: format
scaling_factor: 0.2
kwargs:
preappend_thinking_token: False
- name: length
scaling_factor: 0.2