bridge.models.glm.glm_moe_mappings

GLM MoE mapping helpers for fused expert weights.

These are thin aliases around the shared FusedExpertMapping / FusedGatedExpertMapping classes in param_mapping.py. Kept for backwards compatibility with existing imports.

Module Contents

Classes

GLMExpertDownProjMapping

FusedExpertMapping for GLM down-projection expert weights.

API

class bridge.models.glm.glm_moe_mappings.GLMExpertDownProjMapping(

megatron_param: str,

hf_param: str,

permute_dims: tuple[int, ...] | None = None,

transpose_on_export: bool = False,

)

Bases: megatron.bridge.models.conversion.param_mapping.FusedExpertMapping

FusedExpertMapping for GLM down-projection expert weights.

GLM down-projection weights are stored transposed relative to Megatron’s layout, so transpose_on_export is always enabled.

Initialization