Danielbrdz/Barcenas-31b-Fable
LoRA-style vs google/gemma-4-31B-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 60 of 60 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, Norms 0/523, other 0/62, MLP 180/261, Attention 230/338
3 derivatives checked
- mradermacher/Barcenas-31b-Fable-i1-GGUF: Quantization (fewer layers)
- mradermacher/Barcenas-31b-Fable-GGUF: Quantization (fewer layers)
- Iloqt/Barcenas-StyleTune-31B-Fable: Same weights, repacked
1 other repos hold byte-for-byte identical weight files.