llmfan46/GLM-4-32B-0414-uncensored-heretic-v2
Abliterated vs zai-org/GLM-4-32B-0414. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 27 of 61 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/245, MLP 15/122, Attention 26/244, lm_head 0/1
3 derivatives checked
- mradermacher/GLM-4-32B-0414-uncensored-heretic-v2-i1-GGUF: Quantization
- mradermacher/GLM-4-32B-0414-uncensored-heretic-v2-GGUF: Quantization
- llmfan46/GLM-4-32B-0414-uncensored-heretic-v2-GGUF: Same weights, converted