Commit History

  1. Add PyTorch weights (.pt) converted from JLD2 checkpoint

    Verified signature
    a7ea2a7
  2. Fix model card: match actual HF checkpoint (d=512, 8L, 8Q/2KV, ~23M params, ctx=256, FFN=1344)

    Verified signature
    afa692e
  3. Fix model card: actual trained model is d=256, 4 layers, 4Q/2KV, ~4M params (was incorrectly listed as 10M)

    Verified signature
    287076b

  1. Fix model card: context_length=256 (not 512), dropout=0.1 (not 0.0) per checkpoint

    Verified signature
    5907abe

  1. Add model card with architecture details, provenance, and training metrics

    Verified signature
    9c956d0
  2. Delete julia-slm/5m-chinchilla/config.toml with huggingface_hub

    Verified signature
    91a8ddf
  3. Delete julia-slm/5m-chinchilla/step_12000.jld2 with huggingface_hub

    Verified signature
    36bde8c
  4. Delete julia-slm/5m-chinchilla/final.jld2 with huggingface_hub

    Verified signature
    ac4afd0
  5. Upload julia-slm/5m-chinchilla/step_12000.jld2 with huggingface_hub

    Verified signature
    c0014cd
  6. Upload julia-slm/5m-chinchilla/final.jld2 with huggingface_hub

    Verified signature
    5667a2a
  7. Upload julia-slm/5m-chinchilla/config.toml with huggingface_hub

    Verified signature
    01972e5
  8. Fix tokenizer: trim to 2000 vocab to match trained model

    Verified signature
    db0e784

  1. Upload checkpoint_interrupted.jld2 (261.1 MB)

    Verified signature
    b61a07c
  2. Upload best_model.jld2 (261.1 MB)

    Verified signature
    3b4a52a
  3. Upload checkpoint_interrupted.jld2 (261.1 MB)

    Verified signature
    319a7a5
  4. Upload final_model.jld2 (261.1 MB)

    Verified signature
    2440491
  5. Upload best_model.jld2 (261.1 MB)

    Verified signature
    20cb6be
  6. Upload checkpoint_latest.jld2 (261.2 MB)

    Verified signature
    e825dce
  7. Upload checkpoint_latest.jld2 (261.2 MB)

    Verified signature
    b8fddc2
  8. Upload checkpoint_latest.jld2 (261.2 MB)

    Verified signature
    1e78279
  9. Upload checkpoint_latest.jld2 (261.2 MB)

    Verified signature
    51d7933
  10. Upload checkpoint_latest.jld2 (261.2 MB)

    Verified signature
    12aa2f0
  11. Upload checkpoint_latest.jld2 (261.2 MB)

    Verified signature
    2e78061
  12. Upload checkpoint_latest.jld2 (261.2 MB)

    Verified signature
    1667120
  13. Upload checkpoint_latest.jld2 (261.2 MB)

    Verified signature
    8c2cfb2
  14. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    cc7b028
  15. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    d755d74
  16. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    7a9da3e
  17. Upload final_model.jld2 (261.1 MB)

    Verified signature
    5805138
  18. Upload best_model.jld2 (261.1 MB)

    Verified signature
    b742ef3
  19. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    37a2041
  20. Upload final_model.jld2 (261.1 MB)

    Verified signature
    f8a550d
  21. Upload best_model.jld2 (261.1 MB)

    Verified signature
    50d8cf1

  1. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    78c0464
  2. Upload final_model.jld2 (261.1 MB)

    Verified signature
    4b86ceb
  3. Upload best_model.jld2 (261.1 MB)

    Verified signature
    fcc671a
  4. Upload final_model.jld2 (261.1 MB)

    Verified signature
    d4e408c
  5. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    107ab06
  6. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    b931fa4
  7. Upload best_model.jld2 (261.1 MB)

    Verified signature
    3ca123f
  8. Upload best_model.jld2 (261.1 MB)

    Verified signature
    1ec1fff
  9. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    1397976
  10. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    586f0bb
  11. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    b994abe
  12. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    cd6666c
  13. Upload best_model.jld2 (261.1 MB)

    Verified signature
    d31166a
  14. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    bf24214
  15. Upload best_model.jld2 (261.1 MB)

    Verified signature
    0e8955c
  16. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    b1f2137
  17. Upload checkpoint_latest.jld2 (261.1 MB)

    Verified signature
    a615aaa