sequential quantization even within a single Transformer block
if args.true_sequential: sequential = [['self_attn.k_proj', 'self_attn.v_proj', 'self_attn.q_proj'], ['self_attn.o_proj'], ['mlp.up_proj', 'mlp.gate_proj'], ['mlp.down_proj']]
if args.true_sequential: sequential = [['self_attn.k_proj', 'self_attn.v_proj', 'self_attn.q_proj'], ['self_attn.o_proj'], ['mlp.up_proj', 'mlp.gate_proj'], ['mlp.down_proj']]