File train_fast.py

train_fast.py:None–None · view source on GitHub ↗

Source from the content-addressed store, hash-verified

1	from llmzoo.models.llama.llama_flash_attn_monkey_patch import replace_llama_attn_with_flash_attn
2
3	replace_llama_attn_with_flash_attn()
4

nothing calls this directly

replace_llama_attn_with_flash_attnFunction · 0.90

trainFunction · 0.90

no test coverage detected