jgrusewski 8b44a8c9c5 docs: Phase 1 implementation plan — eliminate PER CPU roundtrips ...

7 tasks: Philox RNG kernel, pre-allocated buffers, GPU-native sampling,
adam_step cleanup, profiling validation, to_host deprecation.

Target: 37s → 12-15s/epoch on H100.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-03-21 10:26:11 +01:00

..

plans

docs: Phase 1 implementation plan — eliminate PER CPU roundtrips

2026-03-21 10:26:11 +01:00

specs

docs: H100 epoch optimization spec (37s → <5s) + phase profiling

2026-03-21 10:23:17 +01:00