Casual-Autopsy/G4-MeroMero-31B-StyleSwap
LoRA-style vs Gryphe/Gemma-4-31B-StyleTune. 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 60 of 60 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/4, Norms 0/523, other 0/62, MLP 180/261, Attention 230/338
4 derivatives checked
- mradermacher/G4-MeroMero-31B-StyleSwap-i1-GGUF: Quantization (fewer layers)
- mradermacher/G4-MeroMero-31B-StyleSwap-GGUF: Quantization (fewer layers)
- Casual-Autopsy/G4-MeroMero-31B-StyleSwap-GGUFs: Same weights, converted (fewer layers)
- Casual-Autopsy/G4-MeroMero-31B-StyleSwap-imat-GGUFs: Quantization (fewer layers)