b1tg 45e2f916a3 add quantize fp8 in llama3 (#12893) ...

* add quantize fp8 in llama3

* don't truncate fp8 alu result

* cast to float32 before matmul

* --model weights/LLaMA-3/8B-SF-DPO/

---------

Co-authored-by: chenyu <chenyu@fastmail.com>

2025-10-27 10:22:57 -04:00

..

autogen.yml

Mesa NIR backend (NAK/LLVMpipe) (#12089)

2025-10-15 17:38:33 +08:00

benchmark_search.yml

add timeout to benchmark_search and mlperf action (#11058)

2025-07-02 14:17:34 -04:00

benchmark.yml

add quantize fp8 in llama3 (#12893)

2025-10-27 10:22:57 -04:00

docs.yml

add strict mkdocs check (#5497)

2024-07-15 14:21:37 -07:00

mlperf.yml

increase timeout of resnet cron (#12693)

2025-10-15 06:08:58 -04:00

python-publish.yml

update gh actions (#3033)

2024-01-09 17:52:22 -08:00

szdiff.yml

update gh actions (#3033)

2024-01-09 17:52:22 -08:00

test.yml

check SPEC=2 in CI (#12945)

2025-10-27 21:53:57 +08:00