Commit History

PERMANENCE: reversibility-aware RL environment for training LLM agents
8f27137
verified

chane335 commited on

Run 6.1: env precondition fix (destructive DB ops on missing tables short-circuit)
5a06418
verified

chane335 commited on

Run 4: tech-only curriculum, 3B model, integrated deploy task
94bea2c
verified

chane335 commited on

Sync: new curriculum, 6 tasks, composable rubric, latent dynamics
d69dab0
verified

chane335 commited on

Upload folder using huggingface_hub
2e8a367
verified

chane335 commited on