ssc-dsai/gc-llm-apertus-70b-instruct-2509
LoRA-style vs swiss-ai/Apertus-70B-Instruct-2509. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 80 of 80 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/321, MLP 160/480, Attention 320/320, lm_head 0/1
1 derivatives checked
- ssc-dsai/gc-llm-apertus-70b-instruct-2509-4bit-mlx: Quantization