Skip to content

LanguageModelPolicyGradientLossConfig

Abstract

This class cannot be instantiated directly. Use one of the variants listed below.

Module: fast_llm.layers.language_model.loss.config

Inherits from: LanguageModelLossConfig

Fields

weightcore

Type: float    Default: 1.0

Weight for this loss in the total loss computation.

logits_scale_factorfeature

Type: float    Default: 1.0

Extra logits scale factor applied for this loss only, stacked on top of the model's logits_scale_factor.

epsilon_high

Type: float    Default: 0.2

Upper clip parameter for ratio of log probs

epsilon_low

Type: float    Default: 0.2

Lower clip parameter for ratio of log probs