Commit History

  1. Dedupe echoed user text in predictions

    Verified signature
    b132b97
  2. Map run id to deployed checkpoint and add reward distributions

    Verified signature
    c6c6f2b
  3. Add grouped result plots to technical report

    Verified signature
    ac245dc
  4. Fix checkpoint id for Interactive-RL default

    Verified signature
    dfb59d7
  5. Use final LoRA checkpoint in RL default

    Verified signature
    e4e0cb9
  6. Initial poor mans interaction models demo

    Verified signature
    128f88c