NorahAlobaied/Mawrooth-ALLaM-7B-LoRA
Adapter vs humain-ai/ALLaM-7B-Instruct-preview. NorahAlobaied/Mawrooth-ALLaM-7B-LoRA is a LoRA adapter, not full weights: rank 16, on k_proj, o_proj, q_proj, v_proj in 32 layers (128 adapted matrices). It modifies humain-ai/ALLaM-7B-Instruct-preview at load time.