3 repos
FailSpy/abliterator
Simple Python library/structure to ablate features in LLMs which are supported by TransformerLens
701
31 commits
Tsadoq/ErisForge
Dead Simple LLM Abliteration
277
81 commits
Sumandora/remove-refusals-with-transformers
Implements harmful/harmless refusal removal using pure HF Transformers
2,184
11 commits
