mlabonne/Qwen3-14B-abliterated
LoRA-style vs Qwen/Qwen3-14B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 32 of 40 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/161, MLP 96/120, Attention 128/160, lm_head 0/1
5 derivatives checked
- bartowski/mlabonne_Qwen3-14B-abliterated-GGUF: Same weights, converted
- mradermacher/Qwen3-14B-abliterated-GGUF: Full fine-tune
- mradermacher/Qwen3-14B-abliterated-i1-GGUF: Full fine-tune
- kevinhh1111/Qwen3-14B-abliterated-Q4_K_M-GGUF: Quantization
- DevQuasar/mlabonne.Qwen3-14B-abliterated-GGUF: Same weights, converted