Skip to content

slat_encoder checkpoint mismatch with slat_encoder.yaml #191

@Hallucinat1on

Description

@Hallucinat1on

Hi, thanks for the great work.
In the HF Repo, slat_encoder.ckpt mismatch with slat_encoder.yaml
It seems like a decoder weight:
input_layer.weight: model (768, 1024), ckpt (768, 8)
out_layer.weight: model (16, 768), ckpt (448, 768)
out_layer.bias: model (16,), ckpt (448,)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions