Skip to content

v0.9.1 — Scaling study & a harder benchmark

Choose a tag to compare

View all tags
@cattolatte cattolatte released this 06 Jul 18:13
v0.9.1
f0959e8
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

A measured scaling study (bits/char vs model size, 0.6M → 10.7M — a clean diminishing-returns curve into tiny-shakespeare's data floor) and a text8 subset benchmark (1.78 bits/char, coherent out-of-domain prose), both reported honestly. Reproducible via scripts/scaling_study.py and scripts/download_text8.py.

Also backfills design docs for phases 8 & 9 and ADR-0009 (modern architecture), and hardens CI against PyTorch mirror outages. Benchmarks/scripts/docs/CI only — no library API change.

Assets 2
Loading