tancho / evaluation /thresholds.json

Commit History

28 days ago

  1. Replace the root checkpoint with Observation Program v8 (Modal-trained) (part 2)

    Verified signature
    1d2b42a
  2. Replace the root checkpoint with Observation Program v7 (Modal-trained) and correct the v4 synthetic claims (part 2)

    Verified signature
    6019ca6
  3. Bundle the synthetic final pack, thresholds and raw outputs for offline scoring reproduction

    Verified signature
    f635a54