ReadyArt/Forgotten-Abomination-36B-v4.1
LoRA-style vs ReadyArt/Forgotten-Safeword-36B-4.1. 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 64 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/129, MLP 192/192, Attention 256/256, lm_head 0/1