Benjaminpwh/qwen2.5-7b-instruct-poly-adaLORA-6
Adapter vs unsloth/Qwen2.5-7B-Instruct. Benjaminpwh/qwen2.5-7b-instruct-poly-adaLORA-6 is a LoRA adapter, not full weights: rank 8, on q_proj, v_proj in 28 layers (56 adapted matrices). It modifies unsloth/Qwen2.5-7B-Instruct at load time.