RL trajectory recording and export (Stage 14): trajectory data model, boundary capture decorators, pluggable reward, sampling, JSONL contract, replay, human feedback / DPO pairs

Latest Versions

6 versions โ†’
Version โ–ผVulnerabilitiesUsagesDate
0.1.x
0.1.5
1
Sep 18, 2026
0.1.4
1
Sep 18, 2026
0.1.3
1
Sep 13, 2026
Show 3 more from 0.1.x
6 versions โ†’