biggy/DeepSeek-V3
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepSeek-V3.git synced 2025-05-03 13:46:29 +02:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Compare commits

base: biggy:b612a1cc105b17ad353107c148637903e2ddeb5d
Branches Tags
biggy:main
...
compare: biggy:05be48bd930ed5f5ed2dce48bdbab39992ea14ae
Branches Tags
biggy:main

2 Commits
b612a1cc10 ... 05be48bd93

Author SHA1 Message Date
codlifedz22
05be48bd93
Merge 4a65fd9221103ff03864337453c238e65d1f4a1b into f09f5fa321f5a421704136c0463b1eaca6557712 2025-02-18 20:50:19 +08:00
oyzh
4a65fd9221 fix an args description. 2025-02-15 11:02:28 +08:00
1 changed files with 1 additions and 1 deletions
Show Stats Expand all files Collapse all files

2
inference/kernel.py
View File

@ -87,7 +87,7 @@ def weight_dequant(x: torch.Tensor, s: torch.Tensor, block_size: int = 128) -> t
Args: Args:
x (torch.Tensor): The quantized weight tensor of shape (M, N). x (torch.Tensor): The quantized weight tensor of shape (M, N).
s (torch.Tensor): The scale tensor of shape (M, N). s (torch.Tensor): The scale tensor of shape (M//block_size, N//block_size).
block_size (int, optional): The block size to use for dequantization. Defaults to 128. block_size (int, optional): The block size to use for dequantization. Defaults to 128.
Returns: Returns: