{ "activations": "bfloat16", "granularity": "per-output-channel", "quant_method": "booper_symmetric_int8", "source": "/home/user/models/Booper-Big-Chat", "tensors": { "lm_head.weight": { "axis": 1, "dtype": "int8", "scale": "lm_head.weight.scale" }, "model.embed_tokens.weight": { "axis": 1, "dtype": "int8", "scale": "model.embed_tokens.weight.scale" }, "model.layers.0.block_sparse_moe.experts.0.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.0.w1.weight.scale" }, "model.layers.0.block_sparse_moe.experts.0.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.0.w2.weight.scale" }, "model.layers.0.block_sparse_moe.experts.0.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.0.w3.weight.scale" }, "model.layers.0.block_sparse_moe.experts.1.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.1.w1.weight.scale" }, "model.layers.0.block_sparse_moe.experts.1.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.1.w2.weight.scale" }, "model.layers.0.block_sparse_moe.experts.1.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.1.w3.weight.scale" }, "model.layers.0.block_sparse_moe.experts.2.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.2.w1.weight.scale" }, "model.layers.0.block_sparse_moe.experts.2.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.2.w2.weight.scale" }, "model.layers.0.block_sparse_moe.experts.2.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.2.w3.weight.scale" }, "model.layers.0.block_sparse_moe.experts.3.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.3.w1.weight.scale" }, "model.layers.0.block_sparse_moe.experts.3.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.3.w2.weight.scale" }, "model.layers.0.block_sparse_moe.experts.3.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.3.w3.weight.scale" }, "model.layers.0.block_sparse_moe.experts.4.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.4.w1.weight.scale" }, "model.layers.0.block_sparse_moe.experts.4.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.4.w2.weight.scale" }, "model.layers.0.block_sparse_moe.experts.4.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.4.w3.weight.scale" }, "model.layers.0.block_sparse_moe.experts.5.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.5.w1.weight.scale" }, "model.layers.0.block_sparse_moe.experts.5.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.5.w2.weight.scale" }, "model.layers.0.block_sparse_moe.experts.5.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.5.w3.weight.scale" }, "model.layers.0.block_sparse_moe.experts.6.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.6.w1.weight.scale" }, "model.layers.0.block_sparse_moe.experts.6.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.6.w2.weight.scale" }, "model.layers.0.block_sparse_moe.experts.6.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.experts.6.w3.weight.scale" }, "model.layers.0.block_sparse_moe.gate.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.block_sparse_moe.gate.weight.scale" }, "model.layers.0.input_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.0.post_attention_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.0.self_attn.k_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.self_attn.k_proj.weight.scale" }, "model.layers.0.self_attn.o_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.self_attn.o_proj.weight.scale" }, "model.layers.0.self_attn.q_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.self_attn.q_proj.weight.scale" }, "model.layers.0.self_attn.v_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.0.self_attn.v_proj.weight.scale" }, "model.layers.1.block_sparse_moe.experts.0.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.0.w1.weight.scale" }, "model.layers.1.block_sparse_moe.experts.0.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.0.w2.weight.scale" }, "model.layers.1.block_sparse_moe.experts.0.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.0.w3.weight.scale" }, "model.layers.1.block_sparse_moe.experts.1.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.1.w1.weight.scale" }, "model.layers.1.block_sparse_moe.experts.1.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.1.w2.weight.scale" }, "model.layers.1.block_sparse_moe.experts.1.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.1.w3.weight.scale" }, "model.layers.1.block_sparse_moe.experts.2.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.2.w1.weight.scale" }, "model.layers.1.block_sparse_moe.experts.2.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.2.w2.weight.scale" }, "model.layers.1.block_sparse_moe.experts.2.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.2.w3.weight.scale" }, "model.layers.1.block_sparse_moe.experts.3.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.3.w1.weight.scale" }, "model.layers.1.block_sparse_moe.experts.3.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.3.w2.weight.scale" }, "model.layers.1.block_sparse_moe.experts.3.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.3.w3.weight.scale" }, "model.layers.1.block_sparse_moe.experts.4.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.4.w1.weight.scale" }, "model.layers.1.block_sparse_moe.experts.4.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.4.w2.weight.scale" }, "model.layers.1.block_sparse_moe.experts.4.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.4.w3.weight.scale" }, "model.layers.1.block_sparse_moe.experts.5.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.5.w1.weight.scale" }, "model.layers.1.block_sparse_moe.experts.5.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.5.w2.weight.scale" }, "model.layers.1.block_sparse_moe.experts.5.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.5.w3.weight.scale" }, "model.layers.1.block_sparse_moe.experts.6.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.6.w1.weight.scale" }, "model.layers.1.block_sparse_moe.experts.6.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.6.w2.weight.scale" }, "model.layers.1.block_sparse_moe.experts.6.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.experts.6.w3.weight.scale" }, "model.layers.1.block_sparse_moe.gate.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.block_sparse_moe.gate.weight.scale" }, "model.layers.1.input_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.1.post_attention_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.1.self_attn.k_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.self_attn.k_proj.weight.scale" }, "model.layers.1.self_attn.o_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.self_attn.o_proj.weight.scale" }, "model.layers.1.self_attn.q_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.self_attn.q_proj.weight.scale" }, "model.layers.1.self_attn.v_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.1.self_attn.v_proj.weight.scale" }, "model.layers.2.block_sparse_moe.experts.0.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.0.w1.weight.scale" }, "model.layers.2.block_sparse_moe.experts.0.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.0.w2.weight.scale" }, "model.layers.2.block_sparse_moe.experts.0.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.0.w3.weight.scale" }, "model.layers.2.block_sparse_moe.experts.1.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.1.w1.weight.scale" }, "model.layers.2.block_sparse_moe.experts.1.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.1.w2.weight.scale" }, "model.layers.2.block_sparse_moe.experts.1.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.1.w3.weight.scale" }, "model.layers.2.block_sparse_moe.experts.2.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.2.w1.weight.scale" }, "model.layers.2.block_sparse_moe.experts.2.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.2.w2.weight.scale" }, "model.layers.2.block_sparse_moe.experts.2.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.2.w3.weight.scale" }, "model.layers.2.block_sparse_moe.experts.3.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.3.w1.weight.scale" }, "model.layers.2.block_sparse_moe.experts.3.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.3.w2.weight.scale" }, "model.layers.2.block_sparse_moe.experts.3.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.3.w3.weight.scale" }, "model.layers.2.block_sparse_moe.experts.4.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.4.w1.weight.scale" }, "model.layers.2.block_sparse_moe.experts.4.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.4.w2.weight.scale" }, "model.layers.2.block_sparse_moe.experts.4.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.4.w3.weight.scale" }, "model.layers.2.block_sparse_moe.experts.5.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.5.w1.weight.scale" }, "model.layers.2.block_sparse_moe.experts.5.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.5.w2.weight.scale" }, "model.layers.2.block_sparse_moe.experts.5.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.5.w3.weight.scale" }, "model.layers.2.block_sparse_moe.experts.6.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.6.w1.weight.scale" }, "model.layers.2.block_sparse_moe.experts.6.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.6.w2.weight.scale" }, "model.layers.2.block_sparse_moe.experts.6.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.experts.6.w3.weight.scale" }, "model.layers.2.block_sparse_moe.gate.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.block_sparse_moe.gate.weight.scale" }, "model.layers.2.input_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.2.post_attention_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.2.self_attn.k_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.self_attn.k_proj.weight.scale" }, "model.layers.2.self_attn.o_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.self_attn.o_proj.weight.scale" }, "model.layers.2.self_attn.q_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.self_attn.q_proj.weight.scale" }, "model.layers.2.self_attn.v_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.2.self_attn.v_proj.weight.scale" }, "model.layers.3.block_sparse_moe.experts.0.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.0.w1.weight.scale" }, "model.layers.3.block_sparse_moe.experts.0.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.0.w2.weight.scale" }, "model.layers.3.block_sparse_moe.experts.0.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.0.w3.weight.scale" }, "model.layers.3.block_sparse_moe.experts.1.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.1.w1.weight.scale" }, "model.layers.3.block_sparse_moe.experts.1.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.1.w2.weight.scale" }, "model.layers.3.block_sparse_moe.experts.1.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.1.w3.weight.scale" }, "model.layers.3.block_sparse_moe.experts.2.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.2.w1.weight.scale" }, "model.layers.3.block_sparse_moe.experts.2.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.2.w2.weight.scale" }, "model.layers.3.block_sparse_moe.experts.2.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.2.w3.weight.scale" }, "model.layers.3.block_sparse_moe.experts.3.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.3.w1.weight.scale" }, "model.layers.3.block_sparse_moe.experts.3.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.3.w2.weight.scale" }, "model.layers.3.block_sparse_moe.experts.3.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.3.w3.weight.scale" }, "model.layers.3.block_sparse_moe.experts.4.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.4.w1.weight.scale" }, "model.layers.3.block_sparse_moe.experts.4.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.4.w2.weight.scale" }, "model.layers.3.block_sparse_moe.experts.4.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.4.w3.weight.scale" }, "model.layers.3.block_sparse_moe.experts.5.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.5.w1.weight.scale" }, "model.layers.3.block_sparse_moe.experts.5.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.5.w2.weight.scale" }, "model.layers.3.block_sparse_moe.experts.5.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.5.w3.weight.scale" }, "model.layers.3.block_sparse_moe.experts.6.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.6.w1.weight.scale" }, "model.layers.3.block_sparse_moe.experts.6.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.6.w2.weight.scale" }, "model.layers.3.block_sparse_moe.experts.6.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.experts.6.w3.weight.scale" }, "model.layers.3.block_sparse_moe.gate.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.block_sparse_moe.gate.weight.scale" }, "model.layers.3.input_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.3.post_attention_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.3.self_attn.k_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.self_attn.k_proj.weight.scale" }, "model.layers.3.self_attn.o_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.self_attn.o_proj.weight.scale" }, "model.layers.3.self_attn.q_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.self_attn.q_proj.weight.scale" }, "model.layers.3.self_attn.v_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.3.self_attn.v_proj.weight.scale" }, "model.layers.4.block_sparse_moe.experts.0.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.0.w1.weight.scale" }, "model.layers.4.block_sparse_moe.experts.0.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.0.w2.weight.scale" }, "model.layers.4.block_sparse_moe.experts.0.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.0.w3.weight.scale" }, "model.layers.4.block_sparse_moe.experts.1.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.1.w1.weight.scale" }, "model.layers.4.block_sparse_moe.experts.1.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.1.w2.weight.scale" }, "model.layers.4.block_sparse_moe.experts.1.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.1.w3.weight.scale" }, "model.layers.4.block_sparse_moe.experts.2.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.2.w1.weight.scale" }, "model.layers.4.block_sparse_moe.experts.2.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.2.w2.weight.scale" }, "model.layers.4.block_sparse_moe.experts.2.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.2.w3.weight.scale" }, "model.layers.4.block_sparse_moe.experts.3.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.3.w1.weight.scale" }, "model.layers.4.block_sparse_moe.experts.3.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.3.w2.weight.scale" }, "model.layers.4.block_sparse_moe.experts.3.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.3.w3.weight.scale" }, "model.layers.4.block_sparse_moe.experts.4.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.4.w1.weight.scale" }, "model.layers.4.block_sparse_moe.experts.4.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.4.w2.weight.scale" }, "model.layers.4.block_sparse_moe.experts.4.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.4.w3.weight.scale" }, "model.layers.4.block_sparse_moe.experts.5.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.5.w1.weight.scale" }, "model.layers.4.block_sparse_moe.experts.5.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.5.w2.weight.scale" }, "model.layers.4.block_sparse_moe.experts.5.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.5.w3.weight.scale" }, "model.layers.4.block_sparse_moe.experts.6.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.6.w1.weight.scale" }, "model.layers.4.block_sparse_moe.experts.6.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.6.w2.weight.scale" }, "model.layers.4.block_sparse_moe.experts.6.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.experts.6.w3.weight.scale" }, "model.layers.4.block_sparse_moe.gate.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.block_sparse_moe.gate.weight.scale" }, "model.layers.4.input_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.4.post_attention_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.4.self_attn.k_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.self_attn.k_proj.weight.scale" }, "model.layers.4.self_attn.o_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.self_attn.o_proj.weight.scale" }, "model.layers.4.self_attn.q_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.self_attn.q_proj.weight.scale" }, "model.layers.4.self_attn.v_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.4.self_attn.v_proj.weight.scale" }, "model.layers.5.block_sparse_moe.experts.0.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.0.w1.weight.scale" }, "model.layers.5.block_sparse_moe.experts.0.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.0.w2.weight.scale" }, "model.layers.5.block_sparse_moe.experts.0.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.0.w3.weight.scale" }, "model.layers.5.block_sparse_moe.experts.1.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.1.w1.weight.scale" }, "model.layers.5.block_sparse_moe.experts.1.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.1.w2.weight.scale" }, "model.layers.5.block_sparse_moe.experts.1.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.1.w3.weight.scale" }, "model.layers.5.block_sparse_moe.experts.2.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.2.w1.weight.scale" }, "model.layers.5.block_sparse_moe.experts.2.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.2.w2.weight.scale" }, "model.layers.5.block_sparse_moe.experts.2.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.2.w3.weight.scale" }, "model.layers.5.block_sparse_moe.experts.3.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.3.w1.weight.scale" }, "model.layers.5.block_sparse_moe.experts.3.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.3.w2.weight.scale" }, "model.layers.5.block_sparse_moe.experts.3.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.3.w3.weight.scale" }, "model.layers.5.block_sparse_moe.experts.4.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.4.w1.weight.scale" }, "model.layers.5.block_sparse_moe.experts.4.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.4.w2.weight.scale" }, "model.layers.5.block_sparse_moe.experts.4.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.4.w3.weight.scale" }, "model.layers.5.block_sparse_moe.experts.5.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.5.w1.weight.scale" }, "model.layers.5.block_sparse_moe.experts.5.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.5.w2.weight.scale" }, "model.layers.5.block_sparse_moe.experts.5.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.5.w3.weight.scale" }, "model.layers.5.block_sparse_moe.experts.6.w1.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.6.w1.weight.scale" }, "model.layers.5.block_sparse_moe.experts.6.w2.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.6.w2.weight.scale" }, "model.layers.5.block_sparse_moe.experts.6.w3.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.experts.6.w3.weight.scale" }, "model.layers.5.block_sparse_moe.gate.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.block_sparse_moe.gate.weight.scale" }, "model.layers.5.input_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.5.post_attention_layernorm.weight": { "dtype": "bfloat16" }, "model.layers.5.self_attn.k_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.self_attn.k_proj.weight.scale" }, "model.layers.5.self_attn.o_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.self_attn.o_proj.weight.scale" }, "model.layers.5.self_attn.q_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.self_attn.q_proj.weight.scale" }, "model.layers.5.self_attn.v_proj.weight": { "axis": 1, "dtype": "int8", "scale": "model.layers.5.self_attn.v_proj.weight.scale" }, "model.norm.weight": { "dtype": "bfloat16" } }, "weights": "int8" }