core.ssm.mlp_layer_config#

Module Contents#

Classes#

MLPLayerConfig

Configuration for a dense MLP layer in a hybrid stack.

API#

class core.ssm.mlp_layer_config.MLPLayerConfig#

Bases: megatron.core.transformer.transformer_config.TransformerConfig

Configuration for a dense MLP layer in a hybrid stack.

Due to backwards-compatibility, this config’s arguments are defined in TransformerConfig.