Skip to content

CausalConv1dConfig

Module: fast_llm.layers.common.linear.config

Inherits from: AffineLinearBaseConfig, LinearBaseConfig, ModuleConfig

Fields

activation — architecture

Type: ActivationType or None    Default: None

Activation function applied after the convolution. None means no activation.

bias — architecture

Type: OptionalParameterConfig    Default: (sub-fields optional)

Use bias.

kernel_size — architecture

Type: int    Default: 4

Convolution kernel size.

weight — architecture

Type: ParameterConfig    Default: (sub-fields optional)

Configuration for the weight.

lr_scale — feature

Type: float or None    Default: None

Scaling factor for the layer learning rate. Combines multiplicatively with the scale set by the parent layer and individual parameters, if applicable.