djuna/L3.1-Suze-Vume-calc
LoRA-style vs Orenguteng/Llama-3.1-8B-Lexi-Uncensored. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, o_proj, up_proj, v_proj changed, in 24 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 72/96, Attention 48/128, lm_head 0/1
2 derivatives checked
- mradermacher/L3.1-Suze-Vume-calc-GGUF: Same weights, converted
- featherless-ai-quants/djuna-L3.1-Suze-Vume-calc-GGUF: Quantization