Add 68k capped-200 scores below the 40k/full tables (do not replace)

Add 68k capped-200 scores below the 40k/full tables (do not replace)

Add 68k capped-200 scores below the 40k/full tables (do not replace)

Note 15M was the Ultra-MC smoke test

This 15M run was the toy/smoke-test for Ultra-MC

Full uncapped benchmark on the card (JSON + complete tables)

Full uncapped benchmark on the card (JSON + complete tables)

Byrne-15M-Looped: inference package, scores, safetensors for leaderboard

initial commit