ClaudioItaly/FimbMoister-11Range
LoRA-style vs Sao10K/Fimbulvetr-11B-v2. 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 46 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/97, MLP 138/144, Attention 184/192, lm_head 0/1
2 derivatives checked
- mradermacher/FimbMoister-11Range-i1-GGUF: Quantization
- mradermacher/FimbMoister-11Range-GGUF: Same weights, converted