Mawdistical/Kuwutu-7B
LoRA-style vs Aurore-Reveil/Koto-Small-7B-IT. 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 36 of 36 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/73, MLP 108/108, Attention 144/252, lm_head 0/1
5 derivatives checked
- Mawdistical/Kuwutu-7B-GGUF: Quantization
- KeinNiemand/Kuwutu-7B-CYOA-GGUF: LoRA-style
- Mawdistical-S1/Infracelestial-7B: Full fine-tune
- KeinNiemand/Kuwutu-7B-CYOA-v2: LoRA-style
- KeinNiemand/Kuwutu-7B-CYOA: LoRA-style