djuna/L3.1-RPganoff-8B-B
LoRA-style vs ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.1. 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
3 derivatives checked
- mradermacher/L3.1-RPganoff-8B-B-i1-GGUF: Quantization
- featherless-ai-quants/djuna-L3.1-RPganoff-8B-B-GGUF: Quantization
- mradermacher/L3.1-RPganoff-8B-B-GGUF: Same weights, converted