Larxel/HealthSafetyLM-BR
LoRA-style vs nvidia/Nemotron-3.5-Content-Safety. 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 34 of 34 layers; norms, biases and embeddings are untouched.
Tensors changed: other 0/1, Norms 0/316, Embeddings 0/4, MLP 102/210, Attention 136/352
1 derivatives checked
- mradermacher/MedSafetyLM-1-GGUF: Same weights, converted (fewer layers)