3MPER0RR/Qwen2.5-Coder-32B-Instruct-3MPER0RR-abliterated

Abliterated vs Qwen/Qwen2.5-Coder-32B-Instruct. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 58 of 64 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.

Tensors changed: Embeddings 0/1, Norms 0/129, MLP 57/192, Attention 58/448, lm_head 0/1

3 derivatives checked

• mradermacher/Qwen2.5-Coder-32B-Instruct-3MPER0RR-abliterated-i1-GGUF: Quantization
• mradermacher/Qwen2.5-Coder-32B-Instruct-3MPER0RR-abliterated-GGUF: Quantization
• 3MPER0RR/Qwen2.5-Coder-32B-Instruct-3MPER0RR-abliterated-Q8_0-GGUF: Quantization

On the Hub · Downloads on Model Pulse