Commit History

  1. Add tailored from-scratch training stack (tokenizer + random-init LM) for math conjecture solving

    Verified signature
    e69a71a
  2. Retarget model to math-conjecture solver profile; remove DeepSeek configs and legacy sync snapshot

    Verified signature
    662bc81

  1. Keep training stack parity: CPU fallback path and mixed-precision safeguards.

    Verified signature
    aef97c9
  2. Upgrade SOTA curriculum: 4-stage training, post-eval metrics, and quality-gated promotion.

    Verified signature
    a86edac

  1. Add SOTA advancement pipeline: multi-stage weighted curriculum + pass@k eval harness.

    Verified signature
    4461ccc
  2. Rename model repo target to math-conjecture-model and upload pipeline.

    Verified signature
    90dacf5