hiyouga fda2cf677b bf16 by default, gemma2 attns ...

Gemma2 finetuning cannot work until merging https://github.com/huggingface/transformers/pull/31674


Former-commit-id: da66c32c7be0adc28d2185b23e9f62d56acb961c

2024-06-28 06:00:26 +08:00

..

__init__.py

rename files

2024-06-07 00:09:06 +08:00

attention.py

bf16 by default, gemma2 attns

2024-06-28 06:00:26 +08:00

checkpointing.py

update patcher

2024-06-19 21:27:00 +08:00

embedding.py

add license

2024-06-15 17:54:33 +08:00

longlora.py

tiny fix

2024-06-16 01:06:41 +08:00

misc.py

add license

2024-06-15 17:54:33 +08:00

mod.py

add license

2024-06-15 17:54:33 +08:00

moe.py

add license

2024-06-15 17:54:33 +08:00

quantization.py

add quant checks

2024-06-27 01:12:25 +08:00

rope.py

fix #4357

2024-06-18 22:42:45 +08:00

unsloth.py

add license

2024-06-15 17:54:33 +08:00

valuehead.py

add license

2024-06-15 17:54:33 +08:00

visual.py

tiny fix

2024-06-16 01:06:41 +08:00