wassname b6f24cac1c add human labels

2026-05-02 15:58:08 +08:00

..

.gitkeep

init

2026-04-30 17:10:09 +08:00

consistency_airisk.jsonl

fix airisk self_violate actor archetype + simplify to 2 conditions

2026-04-30 21:29:38 +08:00

consistency_scifi.jsonl

fix airisk self_violate actor archetype + simplify to 2 conditions

2026-04-30 21:29:38 +08:00

consistency.jsonl

fix airisk self_violate actor archetype + simplify to 2 conditions

2026-04-30 21:29:38 +08:00

validation_scifi.jsonl

decent

2026-04-30 21:22:07 +08:00

validation.jsonl

decent

2026-04-30 21:22:07 +08:00

vignettes_airisk_other_violate.jsonl

add human labels

2026-05-02 15:58:08 +08:00

vignettes_airisk_self_violate.jsonl

add human labels

2026-05-02 15:58:08 +08:00

vignettes_other_violate.jsonl

Implement N-token evaluation with guided rollouts

2026-05-01 21:44:14 +08:00

vignettes_scifi_other_violate.jsonl

add human labels

2026-05-02 15:58:08 +08:00

vignettes_scifi_self_violate.jsonl

add human labels

2026-05-02 15:58:08 +08:00

vignettes_self_violate.jsonl

Implement N-token evaluation with guided rollouts

2026-05-01 21:44:14 +08:00