decoderstack-d24 / convert_ckpt_to_nanochat.py

Commit History

  1. Converter now emits nanochat optimizer shards for an SFT warm-start

    Verified signature
    412ea12
  2. Correct optimizer-state notes: SFT needs only the weights; W_O axis is a no-op at d24

    Verified signature
    162df34
  3. Add model card, meta.json, logs, tokenizer, training source, and nanochat converter

    Verified signature
    11f07f9