Model evaluation mode

Evaluates another model using Heretic's built-in refusal and KL-divergence functionality.

Parent repository GitHub

Type: CLI workflow
Repository: p-e-w/heretic
Readiness: Usable beta for technically equipped local-model researchers, with substantial compute, validation, and safety-review requirements.
Keywords: 5

Location

Repository path

README.md

Invocation

heretic --model ORIGINAL_MODEL --evaluate-model EVALUATED_MODEL

Setup

Installation / activation

Pass both the original model and evaluation target options to the Heretic CLI.

Keywords

evaluationrefusalsKL divergencebenchmarkmodel

Repository context

p-e-w/heretic

AGPL-3.0-or-later Python 3.10+ GPU-oriented CLI and research package that automatically applies optimized directional ablation to transformer language models to suppress refusal behavior while minimizing measured KL divergence; it is standalone model-modification tooling, not a documented Claude Code extension.

Transformer language models.Model alignment modification.Mechanistic interpretability.GPU-based machine learning.Local model evaluation.Hugging Face model workflows.

Open full repository research