github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-08-03 17:19:51 +00:00
Code Issues Packages Projects Releases Wiki Activity

update reward

Browse Source
This commit is contained in:
Tong Li 2025-03-10 14:19:10 +08:00
parent 754b16dfbf
commit 71a0181fce
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
applications/ColossalChat/coati/distributed/reward/verifiable_reward.py
View File

@ -21,7 +21,7 @@ class VerifiableReward:
# Get batch size # Get batch size
bs = input_ids.size(0) bs = input_ids.size(0)
# Initialize reward # Initialize reward
rewards = torch.zeros(bs, device=input_ids.device) rewards = torch.zeros((bs, 3), device=input_ids.device)
# Loop through reward functions # Loop through reward functions
for reward_fn in self.reward_fns: for reward_fn in self.reward_fns: