ilotoki_thu/alpaca_lora_4bit
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

fix bug

Browse Source
This commit is contained in:
John Smith
2023-04-13 10:36:57 +08:00
parent 6aab31bd73
commit 76d7963dff
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
inference.py
View File

@@ -14,7 +14,7 @@ print('Fitting 4bit scales and zeros to half')
model.half() model.half()
for n, m in model.named_modules(): for n, m in model.named_modules():
if isinstance(m, Autograd4bitQuantLinear): if isinstance(m, Autograd4bitQuantLinear):
if m.groupsize == -1: if m.is_v1_model:
m.zeros = m.zeros.half() m.zeros = m.zeros.half()
m.scales = m.scales.half() m.scales = m.scales.half()
m.bias = m.bias.half() m.bias = m.bias.half()