{ "metadata": { "ParamSize": 163, "ParamBytes": 637317120.0, "BitsPerParam": 4.075508022440898 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 131334144, "records": [ { "name": "model.embed_tokens.q_weight", "shape": [ 128256, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 131334144, "byteOffset": 0 } ], "md5sum": "92bb5f9401a7ad51edeecddb6684a8f6" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 33308672, "records": [ { "name": "model.embed_tokens.q_scale", "shape": [ 128256, 16 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4104192, "byteOffset": 0 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 4104192 }, { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 4108288 }, { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 12496896 }, { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 12759040 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 29536256 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 30060544 }, { "name": "model.layers.0.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 30064640 }, { "name": "model.layers.0.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 33210368 } ], "md5sum": "40498abbac7335953ab3ec3db08633f5" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 33533952, "records": [ { "name": "model.layers.0.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.0.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 2097152 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 2162688 }, { "name": "model.layers.1.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 2166784 }, { "name": "model.layers.1.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 10555392 }, { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 10817536 }, { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 27594752 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 28119040 }, { "name": "model.layers.1.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28123136 }, { "name": "model.layers.1.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 31268864 }, { "name": "model.layers.1.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 31367168 }, { "name": "model.layers.1.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 33464320 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33529856 } ], "md5sum": "4cddfb2be501154fbe99d1938c647912" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.10.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.10.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.10.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.10.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "f973021f1bbc4d9658101a6eae5c9d8e" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.11.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.11.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "c7cf82c8e367f7f383dbabb002199102" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.12.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.12.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.12.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.12.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "cb096a8d59d2b017e6a172bacd76187c" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.13.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.13.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.13.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.13.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "34c3726813e5bf56f29b3a5fed8f9b15" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.14.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.14.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.14.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.14.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "e822da2390fdf1a2b7d3636e01a304d8" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.15.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.15.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "2a7c645a41a011c4401b9dcb9525242c" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.2.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.2.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.2.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.2.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.2.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.2.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "c3fbb200a438b231801adb018ec5f8cd" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.3.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.3.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.3.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.3.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "d97437ed324eb22c73aebb6797210d9f" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.4.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.4.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.4.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.4.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "2f527c2e1953d4531e51d8d6c7e51275" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.5.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.5.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.5.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.5.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "70247abdefd9909cc5f01a6a517bde3e" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.6.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.6.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.6.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.6.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "a4855443d6c96e38e1e36abc0a16f4ff" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.7.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.7.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "6668f78b37686d334b790dc22ab421bc" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.8.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.8.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.8.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.8.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "2a879b4de8c222fc42d89bd5d77db652" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 31367168, "records": [ { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 1024, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 8388608, "byteOffset": 0 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 64, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 8388608 }, { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 256, 16384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16777216, "byteOffset": 8650752 }, { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 16, 16384 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 524288, "byteOffset": 25427968 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 25952256 }, { "name": "model.layers.9.self_attn.qkv_proj.q_weight", "shape": [ 256, 3072 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 25956352 }, { "name": "model.layers.9.self_attn.qkv_proj.q_scale", "shape": [ 16, 3072 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 98304, "byteOffset": 29102080 }, { "name": "model.layers.9.self_attn.o_proj.q_weight", "shape": [ 256, 2048 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 29200384 }, { "name": "model.layers.9.self_attn.o_proj.q_scale", "shape": [ 16, 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 65536, "byteOffset": 31297536 }, { "name": "model.norm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31363072 } ], "md5sum": "a8b067a7b591a810534b095474f4a76b" } ] }