netcat420/MFANNv0.22.1
LoRA-style vs netcat420/MFANN-llama3.1-abliterated-v2. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of gate_proj, k_proj, o_proj, q_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/65, MLP 32/96, Attention 128/128
2 derivatives checked
- mradermacher/MFANNv0.22.1-i1-GGUF: Quantization
- mradermacher/MFANNv0.22.1-GGUF: Same weights, converted