ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-MTP
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
11 derivatives checked
- ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-Dynamic-MTP-GGUF: Same weights, converted
- mradermacher/Swift-1.5-Qwen3.8-27B-Uncensored-MTP-GGUF: Quantization
- ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-NVFP4: Quantization
- mradermacher/Swift-1.5-Qwen3.8-27B-Uncensored-MTP-i1-GGUF: Quantization
- IsValorum/Swift-1.5-Qwen3.8-27B-Uncensored-MTP-VAL-APEX-I-NanoPlus-V3-GGUF: Quantization
- IsValorum/Swift-1.5-Qwen3.8-27B-Uncensored-MTP-VAL-APEX-I-MiniPlus-V3-GGUF: Quantization
- bjivanovich/ATX-Swift-1.5-Qwen3.8-27B-Uncensored-MTP-GGUF: Quantization
- greglechin/Swift-1.5-Qwen3.8-27B-Uncensored-GPTQ-Int4-sym-G128-MTP-BF16: Quantization
- ultimaterex/Swift-1.5-Qwen3.8-27B-Uncensored-W4A16-AutoRound: Quantization
- bjivanovich/swift-1.5-27b-coding-GGUF: Quantization
- vwdubb/Swift-Qwen3.8-27B-Uncensored-MTP-FP8: Quantization