huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 12:24:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

Do not convert weight scale to e4m3fnuz on CUDA

Browse Source
This commit is contained in:
Daniël de Kok 2025-01-16 11:06:11 +00:00
parent 922cc38fbc
commit f951a8b48d
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
server/text_generation_server/layers/compressed_tensors/w8an_fp.py
View File

@ -147,7 +147,7 @@ class W8ANFpLoader(WeightsLoader):
else None else None
) )
if self.load_weight_scale or SYSTEM == "rocm": if self.load_weight_scale and SYSTEM == "rocm":
w, weight_scale, input_scale = normalize_e4m3fn_to_e4m3fnuz( w, weight_scale, input_scale = normalize_e4m3fn_to_e4m3fnuz(
w, weight_scale, input_scale w, weight_scale, input_scale
) )