diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,4311 @@ +{ + "metadata": { + "ParamSize": 325, + "ParamBytes": 4140998656.0, + "BitsPerParam": 4.070120983102826 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 262668288, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 128256, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 262668288, + "byteOffset": 0 + } + ], + "md5sum": "e9760ed0239ad2fd7c8932b0d2992265" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "d136552e1de68d90d24285ecda481187" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "15377d2872cae337dce4a8e98292178f" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 262668288, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 128256, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 262668288, + "byteOffset": 0 + } + ], + "md5sum": "3bf27fd33fb7254c278f24b9667bd095" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "7daa6eedba929b286b1dd182502ba705" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "cce03dfdf4f05dda1ca16631b55e8007" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 21962752, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 128256, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8208384, + "byteOffset": 0 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8208384 + }, + { + "name": "model.layers.31.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 8216576 + }, + { + "name": "model.layers.31.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 9134080 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 10969088 + }, + { + "name": "model.norm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 10977280 + }, + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 128256, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8208384, + "byteOffset": 10985472 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 19193856 + }, + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 19202048 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 20119552 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21954560 + } + ], + "md5sum": "7778e58059bac7c67e029d88be587327" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "0ceaee69b0026ea336f65412ce035106" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "a37e3ba81d7dcb644576f167b03266eb" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.0.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "6921fbe7e5f6478b79487bc020fdc40e" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "5a460be6b84d5c7c125064cd325c195f" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "0abdd094bad8e126432c734fe7cbbc73" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.1.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "0f3bae3b110217770956ca543b9e9afd" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "d8c88c567b36b11e78d5734b91a10b2d" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "e9569c584133ae6a3cadac077f9442cc" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.2.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "46342cf290690cf2f7de5951d3d198c8" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "7df069bc8e9796ff5a15b422f23b9525" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "d7bf9d6d4861a479ec9ef71e2cd456eb" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.3.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "84cba3a1e2f4ad5fcacaf40bde46d31c" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "a6fbce23ea8c5c5c22a0cfd769b2116d" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "55fb624de7ccd3750a52d2cfab9286f8" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.4.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "16ca47f3674736ce7770f5ff971d6aa9" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "d3631d8499f38c23bd1e8f16666154f4" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "eb099eaa16bae58c279ccc9a42615d20" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.5.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "c6ab900419b4fa0bc061f6a0e0ea578d" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "447921e68316419b872e432069c0f211" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "42c2f32d5468255e0d7abf402758fc27" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.6.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "11d558453a8e89c8ed3f7319004b535e" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "74aed988bb5fc657a67756aef6eed14f" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "2598d7d553059ba94cd633eb70a9627e" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 24379392, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.7.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21626880 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22544384 + } + ], + "md5sum": "a29a176f1617655032eb0cbb6368e68e" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "c57ee77ecdd3a5bd967356fac569b43d" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "40fd04f591dd3ce20977e9eb46108234" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.8.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "5310f2b1c90b6f92356736e66a5acbac" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "5170f799e7b3cc2b8ff50040ae86c9eb" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "94d3f05e68f93541d6b190fd02b0b6a3" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.10.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "cbce05883eef0a88d6026b58022cf3c9" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "1a35091d5b3c8b7645d9d15992310862" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "434a12e7dadbbbd57669fb67dcad7eab" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.11.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "d21fd2b89203b92c478794d9f95f6967" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "e1f34c4f56d8e83b8bda26ca8124b5ca" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "c2b243c90d148994f7446d359752835c" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.12.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "8362a5d5bb7bef0b6db656f654b4fe35" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "78b9fe587a89397d33a9a6e07f8c38e3" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "aa50f6627fd56f5029efd835f6c5c146" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.13.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "835e2820efb0f5c93cd9524b1f73c843" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "4d1751741c461a1bd2cead1c382b5982" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "1689df7efc9f72341e69d5776b584909" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.14.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "77b395a03232237685303813238ce8df" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "bc79fb99be4b32332e28396a3efcfbfa" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "aa7ca3fd56d59a8bd7303b2db9886db1" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.15.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "a3edb07954162fdba4d14fb6bc401dab" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "1e507a4a1381f8f5cbd0e95c8ea86cb0" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "99d33086b69377adf68ef5bcabaaa3c6" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.16.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "761a510ac0802016a1e7786464ce0f33" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "cb49c15348e350613d48ea3e84a64638" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "ce5cc68e7818b6b5cf7a6ca46c23381f" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.17.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "3529453db20b309c206ea03774ade88e" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "961fc357336df1210f3d2f9db06abcd2" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "159d50980d5574ad3acd1fcf0c3261ee" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.18.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "0c69369661397b6d60b36b01716c2206" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "cf35df634e29847635618d37a2a81749" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "7c104eb9e817f08a67ab72876a862d95" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 24412160, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.19.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21635072 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21643264 + }, + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21651456 + }, + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22568960 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24403968 + } + ], + "md5sum": "f2fe9c7bead383d4b92f84375fedcd30" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "3579cf83c2b9bd351c1bde70675305a2" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "2bad1393e84d0e877716580ad8b41222" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.9.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "b14ff04b9a3d1c9ab8ec0357596e723f" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "32b9768b819a9ba42aa9b0e40bfe37e7" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "6a2ed31fe40c4ac0f37544f93d361277" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.20.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "a87188721b2bc7258040ac00c533c9a4" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "a6f573d2493bb4be346fbaf35fdcb8a7" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "e9154a1791e5da142329854582b3eb64" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.21.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "9e828d63704e4adb3ddae594e96e1619" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "5265cc7dbd77de652019145804d9022a" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "ff1fdc19972faf5e30c873b608c052c0" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "1b43a61b29a20ab73b1d78c982bdb039" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "36cd2656bf44acabcbb8825645377e0d" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "04e5cf443a76b82763b82d439ea2c363" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.23.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "15858bf9a6ad9e02af8dacd1254f3d01" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "e38cd0a0b12fb84254d602c803820864" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "51bb2fcda5624ae2570918b148f56ea9" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.24.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "8bdd7e59f80111d4b630f2eccc89deae" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "ccfac67b0223fecf91d85d420481b6a7" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "65814d84cd08ca5486cb9aa5b796771d" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.25.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "b9e407a9cedf3363a1f6607ca44115b0" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "10332d0b8248e23b6e82400ff761a3f6" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "c8bcbaf88872e012105799e311244ae2" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.26.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "06b1ec87e182dac75eb1d85522266be8" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "8db588b6147b3153ffb69d63cf3ee098" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "81f8f048f2c8f2636ba078cf3c80eafb" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.27.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "b1034fe2684b4895917f8fe6f5bf625c" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "17d79772207b44bb866b588a1a1da412" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "74f6debac243f1b0292017108a2506ce" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.28.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "3ad4cdd053e200c3415d7380352473a8" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 29360128, + "records": [ + { + "name": "model.layers.30.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 29360128, + "byteOffset": 0 + } + ], + "md5sum": "b063b90612d9d25ba7aa62241d49c96c" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_weight", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 58720256, + "byteOffset": 0 + } + ], + "md5sum": "59e7f9a4b5d043393ff6ec82f2352f78" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 24395776, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.29.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21626880 + }, + { + "name": "model.layers.30.mlp.down_proj.q_scale", + "shape": [ + 4096, + 112 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 917504, + "byteOffset": 21635072 + }, + { + "name": "model.layers.30.mlp.gate_up_proj.q_scale", + "shape": [ + 28672, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1835008, + "byteOffset": 22552576 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 24387584 + } + ], + "md5sum": "52758a555516872f5a616708c05ccf2c" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 21626880, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.30.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + } + ], + "md5sum": "88e312c2b578622103def851587b79e2" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 21626880, + "records": [ + { + "name": "model.layers.31.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "model.layers.31.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 393216, + "byteOffset": 12582912 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 12976128 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 32 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21364736 + } + ], + "md5sum": "c501859941772739a68dbe87dceb0397" + } + ] +} \ No newline at end of file