ErisForge is a Python library designed to modify Large Language Models (LLMs) by applying transformations to their internal layers. Named after Eris, the goddess of strife and discord, ErisForge allows you to alter model behavior in a controlled manner, creating both ablated and augmented versions of LLMs that respond differently to specific types of input.
It is also quite useful to perform studies on propaganda and bias in LLMs (planning to experiment with deepseek).
Features - Modify internal layers of LLMs to produce altered behaviors. - Ablate or enhance model responses with the AblationDecoderLayer and AdditionDecoderLayer classes. - Measure refusal expressions in model responses using the ExpressionRefusalScorer. - Supports custom behavior directions for applying specific types of transformations.
Comments URL: https://news.ycombinator.com/item?id=42842123
Points: 5
# Comments: 0
Login to add comment
Other posts in this group
I've been working on this alternative for Plex for almost 3 years now. It's main selling point is that it correctly handles multiple versions of albums and songs. As of today, it only has a web cl
Article URL: https://alexeystar.com/blog/slopify/
Comments URL: https://news.ycomb
Article URL: https://jimmyhmiller.github.io/discovery-coding
Article URL: https://axalatar.github.io/slicing-the-fourth/
Article URL: https://www.promptfoo.dev/blog/deepseek-censorship/
