k-program-environment / data /train.jsonl

Commit History

25 days ago

  1. Store every target as a list (target_is_scalar flag) so all splits load; to_task() in k_prompts.py

    Verified signature
    fdb7d60
  2. K Program Environment: train (2,900), eval (220), eval_depth4 (100), interpreter, prompts

    Verified signature
    f7729c8