d0xin/Swift-1.5-Qwen3.8-27B-Uncensored-BF16
Abliterated vs ukisai/Swift-1.5-Qwen3.8-27b. Looks like abliteration (refusal removal): only the output projections (down_proj, embed_tokens, o_proj, out_proj) changed, in 64 of 64 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Norms 0/325, Attention 65/560, MLP 65/307, Embeddings 1/5, lm_head 0/1, other 0/1
5 derivatives checked
- mradermacher/Swift-1.5-Qwen3.8-27B-Uncensored-BF16-GGUF: Quantization
- mradermacher/Swift-1.5-Qwen3.8-27B-Uncensored-BF16-i1-GGUF: Quantization
- jakeatx/ATX-Swift-1.5-Qwen3.8-27B-Uncensored-IQ4_XS-M-GGUF: Quantization
- d0xin/Swift-1.5-Qwen3.8-27B-Uncensored-FP8: Quantization
- jakeatx/ATX-Swift-1.5-Qwen3.8-27B-Uncensored-2.3bpw-MTP-GGUF: Not comparable yet