core.ssm.utils#
Module Contents#
Functions#
Builds a factory that splits a given ShardedTensor into several independent chunks. |
|
Reject headwise shapes unsupported by any scheduler-enabled CP size at startup. |
API#
- core.ssm.utils._split_tensor_factory(
- orig_sh_ten: megatron.core.dist_checkpointing.ShardedTensor,
- split_sections: list[int],
- split_names: list[str],
- split_dim: int,
Builds a factory that splits a given ShardedTensor into several independent chunks.
- core.ssm.utils._validate_dynamic_cp_head_geometry(
- config: megatron.core.model_parallel_config.ModelParallelConfig,
- pg_collection: megatron.core.process_groups_config.ProcessGroupCollection,
- *,
- head_counts: dict[str, int],
- group_count: int | None = None,
Reject headwise shapes unsupported by any scheduler-enabled CP size at startup.
Chunkwise GDP does not shard heads and must not call this validator. Mamba/GDP may repeat groups when CP exceeds their TP-local group count; GDN must shard both its key and value heads without repetition.