Miladsaeedi70/smollm2-135m-scientific-sft-lora
Adapter vs HuggingFaceTB/SmolLM2-135M-Instruct. Miladsaeedi70/smollm2-135m-scientific-sft-lora is a LoRA adapter, not full weights: rank 16, on down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj in 30 layers (210 adapted matrices). It modifies HuggingFaceTB/SmolLM2-135M-Instruct at load time.