VisionEncoderConfig¶

Module: fast_llm.layers.vision.config

Inherits from: BlockConfig, ModuleConfig

Fields¶

adapter — architecture

Type: MLPBaseConfig Default: (sub-fields optional)

Configuration for the adapter layer.

embeddings — architecture

Type: PatchEmbeddingsConfig Default: (sub-fields optional)

Configuration for the patch convolution layer.

encoder — architecture

Type: BlockSequenceConfig Default: (sub-fields optional)

Configuration for the vision decoder.

hidden_size — architecture

Type: int Default: 1024

Size of the vision encoder main hidden dimension.

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 and child layers, if applicable.

normalization — feature

Type: ImageNormalizationConfig Default: (sub-fields optional)

Configuration for image normalization during preprocessing.