fix a type in code comment (bitsandbytes-foundation#1063)

was pointing to wrong class
wkpark · Feb 14, 2024 · ceae150 · ceae150
1 parent 5b28fd3
commit ceae150
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/bitsandbytes/nn/modules.py b/bitsandbytes/nn/modules.py
@@ -275,7 +275,7 @@ class Linear4bit(nn.Linear):
     compute datatypes such as FP4 and NF4.
 
     In order to quantize a linear layer one should first load the original fp16 / bf16 weights into
-    the Linear8bitLt module, then call `quantized_module.to("cuda")` to quantize the fp16 / bf16 weights.
+    the Linear4bit module, then call `quantized_module.to("cuda")` to quantize the fp16 / bf16 weights.
 
     Example: