integrations.cut_cross_entropy

integrations.cut_cross_entropy

Module for the Plugin for Cut Cross Entropy integration with Axolotl.

Cut Cross Entropy is an optimized implementation of cross entropy loss from Apple’s ML team.

Classes

Name Description
CutCrossEntropyPlugin Plugin for Cut Cross Entropy integration with Axolotl.

CutCrossEntropyPlugin

integrations.cut_cross_entropy.CutCrossEntropyPlugin()

Plugin for Cut Cross Entropy integration with Axolotl.

Methods

Name Description
patch_llama_like Generic patch for model architectures with causal lm similar to llama
pre_model_load Apply cut cross entropy before model loading if enabled.
patch_llama_like
integrations.cut_cross_entropy.CutCrossEntropyPlugin.patch_llama_like(
    model_type_to_patch,
)

Generic patch for model architectures with causal lm similar to llama

pre_model_load
integrations.cut_cross_entropy.CutCrossEntropyPlugin.pre_model_load(cfg)

Apply cut cross entropy before model loading if enabled.