ProCreations/grug-v2-9b
LoRA-style vs ProCreations/grug-9b. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, o_proj, out_proj, q_proj, up_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/4, Norms 0/215, Attention 152/332, MLP 96/208
1 derivatives checked
- ProCreations/grug-v2-9b-gguf: Quantization