fix(docs): fix typos in comments and docstrings (#34836)
Signed-off-by: machov <mv1742@nyu.edu>
This commit is contained in:
@@ -420,7 +420,7 @@ def make_fp8_moe_quant_config(
|
||||
per_out_ch_quant: bool = False,
|
||||
) -> FusedMoEQuantConfig | None:
|
||||
"""
|
||||
Create FusedMoEQuantConfig for the specifed FP8 Backend.
|
||||
Create FusedMoEQuantConfig for the specified FP8 Backend.
|
||||
The FusedMoEQuantConfig holds the scales that are used
|
||||
at runtime by the Modular Kernel abstraction.
|
||||
|
||||
|
||||
@@ -151,7 +151,7 @@ class DefaultMoERunner(MoERunner):
|
||||
kernels for different parallel execution modes.
|
||||
|
||||
Eventually, this class will be split up and specialized for different
|
||||
configurations, e.g. the presense or absence of shared experts, a gate, etc.
|
||||
configurations, e.g. the presence or absence of shared experts, a gate, etc.
|
||||
"""
|
||||
|
||||
def __init__(
|
||||
|
||||
@@ -586,7 +586,7 @@ class GptOssModel(nn.Module):
|
||||
parts = name.split(".")
|
||||
ids = [s for s in parts if s.isdigit()]
|
||||
|
||||
# for amd-quark format that each expert is seperated
|
||||
# for amd-quark format that each expert is separated
|
||||
# need to extract the parameter name with experts fused.
|
||||
# example model: amd/gpt-oss-20b-MoE-Quant-W-MXFP4-A-FP8-KV-FP8
|
||||
if len(ids) == 2:
|
||||
|
||||
Reference in New Issue
Block a user