ademeure 322eb6b86b Add profiling infrastructure (env-var controlled, nsys/ncu/torch profiler) ... - base_train.py: CUDA profiler + PyTorch profiler hooks gated by NANOCHAT_PROFILE_* env vars - profile_step.py: standalone single-step profiler with NVTX ranges and phase selection - LOCAL_STATE.md: documents local branch/file state before machine teardown Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>		2026-04-09 11:29:04 +00:00
..
base_eval.py	delete autocast, an unnecessary thorn in my side, manage dtypes directly	2026-03-04 23:55:30 +00:00
base_train.py	Add profiling infrastructure (env-var controlled, nsys/ncu/torch profiler)	2026-04-09 11:29:04 +00:00
chat_cli.py	delete autocast, an unnecessary thorn in my side, manage dtypes directly	2026-03-04 23:55:30 +00:00
chat_eval.py	delete autocast, an unnecessary thorn in my side, manage dtypes directly	2026-03-04 23:55:30 +00:00
chat_rl.py	delete autocast, an unnecessary thorn in my side, manage dtypes directly	2026-03-04 23:55:30 +00:00
chat_sft.py	FA3/FlexAttention/SDPA attention + PyTorch 2.11/CUDA 13.0	2026-04-08 21:38:29 +00:00
chat_web.py	delete autocast, an unnecessary thorn in my side, manage dtypes directly	2026-03-04 23:55:30 +00:00
profile_step.py	Add profiling infrastructure (env-var controlled, nsys/ncu/torch profiler)	2026-04-09 11:29:04 +00:00
tok_eval.py	initial commit	2025-10-13 06:49:24 -07:00
tok_train.py	fix: correct minor typos in help text, README, and comments	2026-03-12 17:03:26 +08:00