Open HarryWu99 opened 2 months ago
https://github.com/FasterDecoding/SnapKV/blob/ea655b18061313e088879bd2b4a3e3c0c2dc2e21/snapkv_utils.py#L50
In update_kv function, instead of using the function's arguments attention_mask, this variable is overridden.
update_kv
attention_mask
Just refactored the codebase. I think the arg is dummy and will remove it in the future update. You can test to see if there is any bug! Thx!
https://github.com/FasterDecoding/SnapKV/blob/ea655b18061313e088879bd2b4a3e3c0c2dc2e21/snapkv_utils.py#L50
In
update_kv
function, instead of using the function's argumentsattention_mask
, this variable is overridden.