fix: remove max_new_tokens from GRPOConfig, set batch_size=2 (multiple of num_generations) e99351e
muskan singh Claude Opus 4.7 commited on
Spaces:
Runtime error
Runtime error
Commit History
fix: pin trl<=0.24, multi-step reward, lower LR, reduce NUM_GEN 2ab0fe0
muskan singh Claude Opus 4.7 commited on
fix: remove max_new_tokens from GRPOConfig (not supported by Unsloth patched version) 7a0b2ce
muskan singh commited on
x f4bbbf8
muskan singh commited on
context 2921ab6
muskan singh commited on
context error b06a0de
muskan singh commited on
fix: add missing server modules and unsloth import order 7557a2b
muskan singh Claude Sonnet 4.6 commited on
fix f328529
muskan singh commited on
Add OrgOS training pipeline a9b2cb6
muskan singh commited on