MagicalAlchemist/Qwen-SEA-LION-v4-8B-VL-Magic_decensored
Abliterated vs aisingapore/Qwen-SEA-LION-v4-8B-VL. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 26 of 36 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/4, Norms 0/261, MLP 26/232, Attention 26/252, lm_head 0/1
2 derivatives checked
- mradermacher/Qwen-SEA-LION-v4-8B-VL-Magic_decensored-i1-GGUF: Quantization (fewer layers)
- mradermacher/Qwen-SEA-LION-v4-8B-VL-Magic_decensored-GGUF: Same weights, converted (fewer layers)