NousResearch/atropos
1
0
Fork 0
mirror of https://github.com/NousResearch/atropos.git synced 2026-04-30 17:40:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Update instruction_following_algorithm_environment.py

Browse source
This commit is contained in:
VolodymyrBg 2026-01-29 11:22:05 +02:00 • committed by GitHub
parent 1eb0d72099
commit 39f3509965
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
environments/instruction_following_algorithm_environment.py
View file

@ -139,7 +139,7 @@ class InstructionFollowingEnv(BaseEnv):
eval_limit_ratio=0.1,
dataset_name="allenai/RLVR-IFeval", # Default dataset
dataset_config_name=None, # RLVR-IFeval doesn't have a specific config name, uses 'default'
test_set_ratio=0.05, # The ratio of the selelcted dataset in %
test_set_ratio=0.05, # The ratio of the selected dataset in %
dump_rollouts=False, # Enable data dumping if needed
dump_failed_rollouts=False, # Enable failed rollout dumping for debugging
rollout_save_score_threshold=0.7, # Save rollouts with score >= 0.7