Skip to content

GitLab

Commit 44b3187a authored by Casper Hansen's avatar Casper Hansen
Browse files

Fix comment

parent 0f39807b
Hide whitespace changes
Inline Side-by-side
Showing with 1 addition and 1 deletion
awq/quantize/quantizer.py
View file @ 44b3187a
...@@ -75,7 +75,7 @@ class AwqQuantizer: ...@@ -75,7 +75,7 @@ class AwqQuantizer:
# [STEP 4]: Quantize weights # [STEP 4]: Quantize weights
for name, linear_layer in named_linears.items(): for name, linear_layer in named_linears.items():
# NOTE: small regression in perplexity if linear layer you use .cpu().float() # NOTE: small regression in perplexity if linear layer uses .cpu().float()
linear_layer = linear_layer.cuda().half() linear_layer = linear_layer.cuda().half()
linear_layer.weight.data, scales, zeros = self.pseudo_quantize_tensor( linear_layer.weight.data, scales, zeros = self.pseudo_quantize_tensor(
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment