Lambent/Cascade-rlvr75-12B
LoRA-style vs Lambent/IsoC-Gemma-3-12B. 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 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, Norms 0/400, MLP 140/252, Attention 186/408, other 0/1
2 derivatives checked
- mradermacher/Cascade-rlvr75-12B-i1-GGUF: Quantization (fewer layers)
- mradermacher/Cascade-rlvr75-12B-GGUF: Quantization (fewer layers)