diff --git "a/ndarray-cache-b16.json" "b/ndarray-cache-b16.json" --- "a/ndarray-cache-b16.json" +++ "b/ndarray-cache-b16.json" @@ -1,136 +1,92 @@ { "metadata": { - "ParamSize": 313, - "ParamBytes": 1111169024.0, - "BitsPerParam": 5.002201462167321 + "ParamSize": 199, + "ParamBytes": 7108352000.0, + "BitsPerParam": 32.0 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", - "nbytes": 116686848, + "nbytes": 466747392, "records": [ { - "name": "lm_head.q_weight", + "name": "lm_head.weight", "shape": [ 151936, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 116686848, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 466747392, "byteOffset": 0 } ], - "md5sum": "3eba7944239eea590b63e87f00016a6b" + "md5sum": "e376b8cb28d4376590f01ad535d6071d" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", - "nbytes": 116686848, + "nbytes": 466747392, "records": [ { - "name": "model.embed_tokens.q_weight", + "name": "model.embed_tokens.weight", "shape": [ 151936, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 116686848, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 466747392, "byteOffset": 0 } ], - "md5sum": "be8d442fc5234135620521550a8148fa" + "md5sum": "6a2ea36ede6c428672115e3a3d8ff0d2" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", - "nbytes": 29174784, + "nbytes": 55050240, "records": [ { - "name": "lm_head.q_scale", - "shape": [ - 151936, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 14585856, - "byteOffset": 0 - }, - { - "name": "model.embed_tokens.q_scale", - "shape": [ - 151936, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 14585856, - "byteOffset": 14585856 - }, - { - "name": "model.layers.0.input_layernorm.weight", + "name": "model.layers.0.mlp.gate_up_proj.weight", "shape": [ + 17920, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 29171712 + "nbytes": 55050240, + "byteOffset": 0 } ], - "md5sum": "51e65235b4e65bd14f51a603e8ef7b49" + "md5sum": "9463e4903f97fd312a8f057495aa1239" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", - "nbytes": 33212416, + "nbytes": 27535360, "records": [ { - "name": "model.layers.0.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 0 - }, - { - "name": "model.layers.0.mlp.down_proj.q_scale", + "name": "model.layers.0.input_layernorm.weight", "shape": [ - 1536, - 280 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 6881280 - }, - { - "name": "model.layers.0.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 7741440 + "nbytes": 3072, + "byteOffset": 0 }, { - "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "name": "model.layers.0.mlp.down_proj.weight", "shape": [ - 17920, - 48 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 21504000 + "nbytes": 27525120, + "byteOffset": 3072 }, { "name": "model.layers.0.post_attention_layernorm.weight", @@ -140,7 +96,7 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 23224320 + "byteOffset": 27528192 }, { "name": "model.layers.0.self_attn.c_attn.bias", @@ -150,113 +106,161 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 23227392 - }, + "byteOffset": 27531264 + } + ], + "md5sum": "1431994b9042b3fd664eae27a814f0dd" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.0.self_attn.c_attn.q_weight", + "name": "model.layers.1.mlp.down_proj.weight", "shape": [ - 2048, - 192 + 1536, + 8960 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 23231488 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "22267d86aa57d6448950d6b1fd5d3a1b" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.0.self_attn.c_attn.q_scale", + "name": "model.layers.1.mlp.gate_up_proj.weight", "shape": [ - 2048, - 48 + 17920, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 24804352 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "e545b60a4a590e0d1c5e3b9140db35f3" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.0.self_attn.o_proj.q_weight", + "name": "model.layers.10.mlp.down_proj.weight", "shape": [ 1536, - 192 + 8960 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 25000960 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "8e38f424f7c0d358cd84cfd6d524534b" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.0.self_attn.o_proj.q_scale", + "name": "model.layers.10.mlp.gate_up_proj.weight", "shape": [ - 1536, - 48 + 17920, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 26180608 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "631ad4762294e6034a11b0b66bae13b5" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.1.input_layernorm.weight", + "name": "model.layers.11.mlp.down_proj.weight", "shape": [ - 1536 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 26328064 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "4c542b91b3dc431e62e87024c4b79779" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.1.mlp.down_proj.q_weight", + "name": "model.layers.11.mlp.gate_up_proj.weight", "shape": [ - 1536, - 1120 + 17920, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 26331136 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, + "byteOffset": 0 } ], - "md5sum": "59cea6a143c3e50522e6a7ab56fd1302" + "md5sum": "da04ac4500d98575c04bc5229b7fcafb" }, { - "dataPath": "params_shard_4.bin", + "dataPath": "params_shard_10.bin", "format": "raw-shard", - "nbytes": 27191296, + "nbytes": 33060864, "records": [ { - "name": "model.layers.1.mlp.down_proj.q_scale", + "name": "model.layers.0.self_attn.c_attn.weight", "shape": [ - 1536, - 280 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, + "nbytes": 6291456, "byteOffset": 0 }, { - "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "name": "model.layers.0.self_attn.o_proj.weight", "shape": [ - 17920, - 192 + 1536, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 860160 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "name": "model.layers.1.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 14622720 + "nbytes": 3072, + "byteOffset": 11010048 }, { "name": "model.layers.1.post_attention_layernorm.weight", @@ -266,7 +270,7 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 16343040 + "byteOffset": 11013120 }, { "name": "model.layers.1.self_attn.c_attn.bias", @@ -276,51 +280,29 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 16346112 - }, - { - "name": "model.layers.1.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 16350208 + "byteOffset": 11016192 }, { - "name": "model.layers.1.self_attn.c_attn.q_scale", + "name": "model.layers.1.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17923072 - }, - { - "name": "model.layers.1.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 18119680 + "nbytes": 6291456, + "byteOffset": 11020288 }, { - "name": "model.layers.1.self_attn.o_proj.q_scale", + "name": "model.layers.1.self_attn.o_proj.weight", "shape": [ 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 19299328 + "nbytes": 4718592, + "byteOffset": 17311744 }, { "name": "model.layers.10.input_layernorm.weight", @@ -330,59 +312,7 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 19446784 - }, - { - "name": "model.layers.10.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 19449856 - }, - { - "name": "model.layers.10.mlp.down_proj.q_scale", - "shape": [ - 1536, - 280 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 860160, - "byteOffset": 26331136 - } - ], - "md5sum": "13283f546f5e1640e0d179f91c737060" - }, - { - "dataPath": "params_shard_5.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ - { - "name": "model.layers.10.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.10.mlp.gate_up_proj.q_scale", - "shape": [ - 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "byteOffset": 22030336 }, { "name": "model.layers.10.post_attention_layernorm.weight", @@ -392,7 +322,7 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 22033408 }, { "name": "model.layers.10.self_attn.c_attn.bias", @@ -402,2759 +332,1985 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.10.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 + "byteOffset": 22036480 }, { - "name": "model.layers.10.self_attn.c_attn.q_scale", + "name": "model.layers.10.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 22040576 }, { - "name": "model.layers.10.self_attn.o_proj.q_weight", + "name": "model.layers.10.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 28332032 }, { - "name": "model.layers.10.self_attn.o_proj.q_scale", + "name": "model.layers.11.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 33050624 }, { - "name": "model.layers.11.input_layernorm.weight", + "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 33053696 }, { - "name": "model.layers.11.mlp.down_proj.q_weight", + "name": "model.layers.11.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 33056768 + } + ], + "md5sum": "f1dfff1bdbfc338078cd03dcad4f25ad" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.11.mlp.down_proj.q_scale", + "name": "model.layers.12.mlp.down_proj.weight", "shape": [ 1536, - 280 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 27525120, + "byteOffset": 0 } ], - "md5sum": "f80a070c11451222f8169d453336a19c" + "md5sum": "b8e570d8afef1127f3e3bd074074d74f" }, { - "dataPath": "params_shard_6.bin", + "dataPath": "params_shard_12.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 55050240, "records": [ { - "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "name": "model.layers.12.mlp.gate_up_proj.weight", "shape": [ 17920, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, "byteOffset": 0 - }, + } + ], + "md5sum": "e7a70f4c903318c16326ef7bcd6181d6" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "name": "model.layers.13.mlp.down_proj.weight", "shape": [ - 17920, - 48 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "a4eab6c623efb1460907f9e4bfbc9ee5" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.11.post_attention_layernorm.weight", + "name": "model.layers.13.mlp.gate_up_proj.weight", "shape": [ + 17920, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "127d9f5e010997c546676c086891a0ad" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.11.self_attn.c_attn.bias", + "name": "model.layers.14.mlp.down_proj.weight", "shape": [ - 2048 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "64c5a74a548ff008fdb5d95636e4fd63" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.11.self_attn.c_attn.q_weight", + "name": "model.layers.14.mlp.gate_up_proj.weight", "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, + 17920, + 1536 + ], + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "99334446e606608b2aaa2c25815f3f04" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 33060864, + "records": [ { - "name": "model.layers.11.self_attn.c_attn.q_scale", + "name": "model.layers.11.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 0 }, { - "name": "model.layers.11.self_attn.o_proj.q_weight", + "name": "model.layers.11.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.11.self_attn.o_proj.q_scale", + "name": "model.layers.12.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 11010048 }, { - "name": "model.layers.12.input_layernorm.weight", + "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 11013120 }, { - "name": "model.layers.12.mlp.down_proj.q_weight", + "name": "model.layers.12.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 11016192 }, { - "name": "model.layers.12.mlp.down_proj.q_scale", + "name": "model.layers.12.self_attn.c_attn.weight", "shape": [ - 1536, - 280 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "114d2917404e99a37c39aacfa57a9a50" - }, - { - "dataPath": "params_shard_7.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ + "nbytes": 6291456, + "byteOffset": 11020288 + }, { - "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "name": "model.layers.12.self_attn.o_proj.weight", "shape": [ - 17920, - 192 + 1536, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 17311744 }, { - "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "name": "model.layers.13.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 22030336 }, { - "name": "model.layers.12.post_attention_layernorm.weight", + "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 22033408 }, { - "name": "model.layers.12.self_attn.c_attn.bias", + "name": "model.layers.13.self_attn.c_attn.bias", "shape": [ 2048 ], "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 + "byteOffset": 22036480 }, { - "name": "model.layers.12.self_attn.c_attn.q_weight", + "name": "model.layers.13.self_attn.c_attn.weight", "shape": [ 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.12.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 22040576 }, { - "name": "model.layers.12.self_attn.o_proj.q_weight", + "name": "model.layers.13.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 28332032 }, { - "name": "model.layers.12.self_attn.o_proj.q_scale", + "name": "model.layers.14.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 33050624 }, { - "name": "model.layers.13.input_layernorm.weight", + "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 33053696 }, { - "name": "model.layers.13.mlp.down_proj.q_weight", + "name": "model.layers.14.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 33056768 + } + ], + "md5sum": "f94fe101baadd653c719760c39936366" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.13.mlp.down_proj.q_scale", + "name": "model.layers.15.mlp.down_proj.weight", "shape": [ 1536, - 280 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 27525120, + "byteOffset": 0 } ], - "md5sum": "dc3b5e347d2aba0026bc6cc0dcf4b005" + "md5sum": "6a3fc846bb0b5a13f659c6a1dc4c4694" }, { - "dataPath": "params_shard_8.bin", + "dataPath": "params_shard_19.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 55050240, "records": [ { - "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "name": "model.layers.15.mlp.gate_up_proj.weight", "shape": [ 17920, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, "byteOffset": 0 - }, + } + ], + "md5sum": "ad4b8262ad73b103f7d24ae8a372da04" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "name": "model.layers.16.mlp.down_proj.weight", "shape": [ - 17920, - 48 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "8a98a1a048e006dd076822db1ae73846" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.13.post_attention_layernorm.weight", + "name": "model.layers.16.mlp.gate_up_proj.weight", "shape": [ + 17920, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "72543b4e31346428357244f70b615d12" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.13.self_attn.c_attn.bias", + "name": "model.layers.17.mlp.down_proj.weight", "shape": [ - 2048 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "284d2e9aa13b996ea3297594b10df1f1" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.13.self_attn.c_attn.q_weight", + "name": "model.layers.17.mlp.gate_up_proj.weight", "shape": [ - 2048, - 192 + 17920, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "a150d9ebc7ec5c2915ef543348a130cf" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 33060864, + "records": [ { - "name": "model.layers.13.self_attn.c_attn.q_scale", + "name": "model.layers.14.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 0 }, { - "name": "model.layers.13.self_attn.o_proj.q_weight", + "name": "model.layers.14.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.13.self_attn.o_proj.q_scale", + "name": "model.layers.15.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 11010048 }, { - "name": "model.layers.14.input_layernorm.weight", + "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 11013120 }, { - "name": "model.layers.14.mlp.down_proj.q_weight", + "name": "model.layers.15.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 11016192 }, { - "name": "model.layers.14.mlp.down_proj.q_scale", + "name": "model.layers.15.self_attn.c_attn.weight", "shape": [ - 1536, - 280 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "60e7284515bb4c2999432289d45839c4" - }, - { - "dataPath": "params_shard_9.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ + "nbytes": 6291456, + "byteOffset": 11020288 + }, { - "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "name": "model.layers.15.self_attn.o_proj.weight", "shape": [ - 17920, - 192 + 1536, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 17311744 }, { - "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "name": "model.layers.16.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 22030336 }, { - "name": "model.layers.14.post_attention_layernorm.weight", + "name": "model.layers.16.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 22033408 }, { - "name": "model.layers.14.self_attn.c_attn.bias", + "name": "model.layers.16.self_attn.c_attn.bias", "shape": [ 2048 ], "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.14.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 + "byteOffset": 22036480 }, { - "name": "model.layers.14.self_attn.c_attn.q_scale", + "name": "model.layers.16.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.14.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "nbytes": 6291456, + "byteOffset": 22040576 }, { - "name": "model.layers.14.self_attn.o_proj.q_scale", + "name": "model.layers.16.self_attn.o_proj.weight", "shape": [ 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 4718592, + "byteOffset": 28332032 }, { - "name": "model.layers.15.input_layernorm.weight", + "name": "model.layers.17.input_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 33050624 }, { - "name": "model.layers.15.mlp.down_proj.q_weight", + "name": "model.layers.17.post_attention_layernorm.weight", "shape": [ - 1536, - 1120 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 3072, + "byteOffset": 33053696 }, { - "name": "model.layers.15.mlp.down_proj.q_scale", + "name": "model.layers.17.self_attn.c_attn.bias", "shape": [ - 1536, - 280 + 2048 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 4096, + "byteOffset": 33056768 } ], - "md5sum": "f48e561b8d2b76519c23c5db3e8f6f16" + "md5sum": "eccb4e34a4523e8f5ddad7512bce96cd" }, { - "dataPath": "params_shard_10.bin", + "dataPath": "params_shard_25.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 27525120, "records": [ { - "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "name": "model.layers.18.mlp.down_proj.weight", "shape": [ - 17920, - 192 + 1536, + 8960 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 27525120, "byteOffset": 0 - }, + } + ], + "md5sum": "98c6b2f2b0a3fbdaa97b34ff1c44f19f" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "name": "model.layers.18.mlp.gate_up_proj.weight", "shape": [ 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "a27c1d063ffa5747651e4090e48e7b4c" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.15.post_attention_layernorm.weight", + "name": "model.layers.19.mlp.down_proj.weight", "shape": [ - 1536 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "bce8361e10f3eba7353b8f498d779676" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.15.self_attn.c_attn.bias", + "name": "model.layers.19.mlp.gate_up_proj.weight", "shape": [ - 2048 + 17920, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.15.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.15.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.15.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, - { - "name": "model.layers.15.self_attn.o_proj.q_scale", - "shape": [ - 1536, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, - { - "name": "model.layers.16.input_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.16.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, - { - "name": "model.layers.16.mlp.down_proj.q_scale", - "shape": [ - 1536, - 280 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "252c5013974cfddd0dbba30e029f877e" - }, - { - "dataPath": "params_shard_11.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ - { - "name": "model.layers.16.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.16.mlp.gate_up_proj.q_scale", - "shape": [ - 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, - { - "name": "model.layers.16.post_attention_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, - { - "name": "model.layers.16.self_attn.c_attn.bias", - "shape": [ - 2048 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.16.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.16.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.16.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, - { - "name": "model.layers.16.self_attn.o_proj.q_scale", - "shape": [ - 1536, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, - { - "name": "model.layers.17.input_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.17.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, - { - "name": "model.layers.17.mlp.down_proj.q_scale", - "shape": [ - 1536, - 280 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "390cb82a590d93b12129b8d3c50f73f1" - }, - { - "dataPath": "params_shard_12.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ - { - "name": "model.layers.17.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.17.mlp.gate_up_proj.q_scale", - "shape": [ - 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, - { - "name": "model.layers.17.post_attention_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, - { - "name": "model.layers.17.self_attn.c_attn.bias", - "shape": [ - 2048 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.17.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.17.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.17.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, - { - "name": "model.layers.17.self_attn.o_proj.q_scale", - "shape": [ - 1536, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, - { - "name": "model.layers.18.input_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.18.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, - { - "name": "model.layers.18.mlp.down_proj.q_scale", - "shape": [ - 1536, - 280 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "a3495d635ea3c53c9b0bbbcef749ceb8" - }, - { - "dataPath": "params_shard_13.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ - { - "name": "model.layers.18.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.18.mlp.gate_up_proj.q_scale", - "shape": [ - 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, - { - "name": "model.layers.18.post_attention_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, - { - "name": "model.layers.18.self_attn.c_attn.bias", - "shape": [ - 2048 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.18.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.18.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.18.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, - { - "name": "model.layers.18.self_attn.o_proj.q_scale", - "shape": [ - 1536, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, - { - "name": "model.layers.19.input_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.19.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, - { - "name": "model.layers.19.mlp.down_proj.q_scale", - "shape": [ - 1536, - 280 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "35700f4804c01da98c6241cee2ff5732" - }, - { - "dataPath": "params_shard_14.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ - { - "name": "model.layers.19.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.19.mlp.gate_up_proj.q_scale", - "shape": [ - 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, - { - "name": "model.layers.19.post_attention_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, - { - "name": "model.layers.19.self_attn.c_attn.bias", - "shape": [ - 2048 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.19.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.19.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.19.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, - { - "name": "model.layers.19.self_attn.o_proj.q_scale", - "shape": [ - 1536, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, - { - "name": "model.layers.2.input_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.2.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, - { - "name": "model.layers.2.mlp.down_proj.q_scale", - "shape": [ - 1536, - 280 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 55050240, + "byteOffset": 0 } ], - "md5sum": "fc44a49e8c67d1e47c7d9f9b47c32cb5" + "md5sum": "8528ddaa61b8703c89aa81d92afff1cc" }, { - "dataPath": "params_shard_15.bin", + "dataPath": "params_shard_29.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 27525120, "records": [ { - "name": "model.layers.2.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.2.mlp.gate_up_proj.q_scale", - "shape": [ - 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, - { - "name": "model.layers.2.post_attention_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, - { - "name": "model.layers.2.self_attn.c_attn.bias", - "shape": [ - 2048 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.2.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.2.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.2.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, - { - "name": "model.layers.2.self_attn.o_proj.q_scale", - "shape": [ - 1536, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, - { - "name": "model.layers.20.input_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.20.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, - { - "name": "model.layers.20.mlp.down_proj.q_scale", + "name": "model.layers.2.mlp.down_proj.weight", "shape": [ 1536, - 280 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 27525120, + "byteOffset": 0 } ], - "md5sum": "66952571c7809812a74701771112d5ac" + "md5sum": "d33d62ce5f4294e21b34df2db479f5d3" }, { - "dataPath": "params_shard_16.bin", + "dataPath": "params_shard_30.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 55050240, "records": [ { - "name": "model.layers.20.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "name": "model.layers.2.mlp.gate_up_proj.weight", "shape": [ 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, - { - "name": "model.layers.20.post_attention_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, - { - "name": "model.layers.20.self_attn.c_attn.bias", - "shape": [ - 2048 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.20.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.20.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.20.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, - { - "name": "model.layers.20.self_attn.o_proj.q_scale", - "shape": [ - 1536, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, - { - "name": "model.layers.21.input_layernorm.weight", - "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.21.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, - { - "name": "model.layers.21.mlp.down_proj.q_scale", - "shape": [ - 1536, - 280 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 55050240, + "byteOffset": 0 } ], - "md5sum": "451ddd6918f481774a5ddcae93fcc19a" + "md5sum": "2dd687b44fa54cc3489a523d9797e2f0" }, { - "dataPath": "params_shard_17.bin", + "dataPath": "params_shard_31.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 33060864, "records": [ { - "name": "model.layers.21.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.21.mlp.gate_up_proj.q_scale", - "shape": [ - 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 - }, - { - "name": "model.layers.21.post_attention_layernorm.weight", - "shape": [ - 1536 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 - }, - { - "name": "model.layers.21.self_attn.c_attn.bias", - "shape": [ - 2048 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.21.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.21.self_attn.c_attn.q_scale", + "name": "model.layers.17.self_attn.c_attn.weight", "shape": [ 2048, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.21.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, - { - "name": "model.layers.21.self_attn.o_proj.q_scale", - "shape": [ - 1536, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, - { - "name": "model.layers.22.input_layernorm.weight", - "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.22.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "nbytes": 6291456, + "byteOffset": 0 }, { - "name": "model.layers.22.mlp.down_proj.q_scale", + "name": "model.layers.17.self_attn.o_proj.weight", "shape": [ 1536, - 280 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "6d6e530a4f2c516481ef459c432eab73" - }, - { - "dataPath": "params_shard_18.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ - { - "name": "model.layers.22.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "name": "model.layers.18.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 11010048 }, { - "name": "model.layers.22.post_attention_layernorm.weight", + "name": "model.layers.18.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 11013120 }, { - "name": "model.layers.22.self_attn.c_attn.bias", + "name": "model.layers.18.self_attn.c_attn.bias", "shape": [ 2048 ], "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 + "byteOffset": 11016192 }, { - "name": "model.layers.22.self_attn.c_attn.q_weight", + "name": "model.layers.18.self_attn.c_attn.weight", "shape": [ 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, - { - "name": "model.layers.22.self_attn.c_attn.q_scale", - "shape": [ - 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 11020288 }, { - "name": "model.layers.22.self_attn.o_proj.q_weight", + "name": "model.layers.18.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 17311744 }, { - "name": "model.layers.22.self_attn.o_proj.q_scale", + "name": "model.layers.19.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 22030336 }, { - "name": "model.layers.23.input_layernorm.weight", + "name": "model.layers.19.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 22033408 }, { - "name": "model.layers.23.mlp.down_proj.q_weight", + "name": "model.layers.19.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 22036480 }, { - "name": "model.layers.23.mlp.down_proj.q_scale", + "name": "model.layers.19.self_attn.c_attn.weight", "shape": [ - 1536, - 280 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "0e25e76a52534e27e009f1966b60e908" - }, - { - "dataPath": "params_shard_19.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ + "nbytes": 6291456, + "byteOffset": 22040576 + }, { - "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "name": "model.layers.19.self_attn.o_proj.weight", "shape": [ - 17920, - 192 + 1536, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 28332032 }, { - "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "name": "model.layers.2.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 33050624 }, { - "name": "model.layers.23.post_attention_layernorm.weight", + "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 33053696 }, { - "name": "model.layers.23.self_attn.c_attn.bias", + "name": "model.layers.2.self_attn.c_attn.bias", "shape": [ 2048 ], "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.23.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, + "byteOffset": 33056768 + } + ], + "md5sum": "7cc1e4b5ba621f3e55b0dcc20acdbc09" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.23.self_attn.c_attn.q_scale", + "name": "model.layers.20.mlp.down_proj.weight", "shape": [ - 2048, - 48 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "7050d538a01ea8409da478ecfe4b7308" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.23.self_attn.o_proj.q_weight", + "name": "model.layers.20.mlp.gate_up_proj.weight", "shape": [ - 1536, - 192 + 17920, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "a1a18612a0019a522efe32bbf85b09d3" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.23.self_attn.o_proj.q_scale", + "name": "model.layers.21.mlp.down_proj.weight", "shape": [ 1536, - 48 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "449087c60ab28cead04f4d648e9c115b" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.24.input_layernorm.weight", + "name": "model.layers.21.mlp.gate_up_proj.weight", "shape": [ + 17920, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "349c72f8fe7053102705ccc87332b90d" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.24.mlp.down_proj.q_weight", + "name": "model.layers.22.mlp.down_proj.weight", "shape": [ 1536, - 1120 + 8960 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "b5995018e8c8653dff6649c8deb76b4b" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.24.mlp.down_proj.q_scale", + "name": "model.layers.22.mlp.gate_up_proj.weight", "shape": [ - 1536, - 280 + 17920, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 55050240, + "byteOffset": 0 } ], - "md5sum": "0b2f3521eb647a178bf4fe69ebbefca4" + "md5sum": "c9f851e6d38fbb3193e8401ceb8a1d43" }, { - "dataPath": "params_shard_20.bin", + "dataPath": "params_shard_38.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 33060864, "records": [ { - "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "name": "model.layers.2.self_attn.c_attn.weight", "shape": [ - 17920, - 192 + 2048, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 6291456, "byteOffset": 0 }, { - "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "name": "model.layers.2.self_attn.o_proj.weight", "shape": [ - 17920, - 48 + 1536, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.24.post_attention_layernorm.weight", + "name": "model.layers.20.input_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 11010048 }, { - "name": "model.layers.24.self_attn.c_attn.bias", + "name": "model.layers.20.post_attention_layernorm.weight", "shape": [ - 2048 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 + "nbytes": 3072, + "byteOffset": 11013120 }, { - "name": "model.layers.24.self_attn.c_attn.q_weight", + "name": "model.layers.20.self_attn.c_attn.bias", "shape": [ - 2048, - 192 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 11016192 }, { - "name": "model.layers.24.self_attn.c_attn.q_scale", + "name": "model.layers.20.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 11020288 }, { - "name": "model.layers.24.self_attn.o_proj.q_weight", + "name": "model.layers.20.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 17311744 }, { - "name": "model.layers.24.self_attn.o_proj.q_scale", + "name": "model.layers.21.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 22030336 }, { - "name": "model.layers.25.input_layernorm.weight", + "name": "model.layers.21.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 22033408 }, { - "name": "model.layers.25.mlp.down_proj.q_weight", + "name": "model.layers.21.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 22036480 }, { - "name": "model.layers.25.mlp.down_proj.q_scale", + "name": "model.layers.21.self_attn.c_attn.weight", "shape": [ - 1536, - 280 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "176c5666e37994d7f23960de639e90cc" - }, - { - "dataPath": "params_shard_21.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ + "nbytes": 6291456, + "byteOffset": 22040576 + }, { - "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "name": "model.layers.21.self_attn.o_proj.weight", "shape": [ - 17920, - 192 + 1536, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 28332032 }, { - "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "name": "model.layers.22.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 33050624 }, { - "name": "model.layers.25.post_attention_layernorm.weight", + "name": "model.layers.22.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 33053696 }, { - "name": "model.layers.25.self_attn.c_attn.bias", + "name": "model.layers.22.self_attn.c_attn.bias", "shape": [ 2048 ], "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.25.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, + "byteOffset": 33056768 + } + ], + "md5sum": "49f162c562d76a1aa323b681ae8bf59b" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.25.self_attn.c_attn.q_scale", + "name": "model.layers.23.mlp.down_proj.weight", "shape": [ - 2048, - 48 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "3de22bdb23771bfc14d719fa125c6be6" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.25.self_attn.o_proj.q_weight", + "name": "model.layers.23.mlp.gate_up_proj.weight", "shape": [ - 1536, - 192 + 17920, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "6edf9c36cfa4e3b3b5afaa14b62fa842" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.25.self_attn.o_proj.q_scale", + "name": "model.layers.24.mlp.down_proj.weight", "shape": [ 1536, - 48 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "52a49b8afd861c3081f11803f008dba7" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.26.input_layernorm.weight", + "name": "model.layers.24.mlp.gate_up_proj.weight", "shape": [ + 17920, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "07cad64074e88e27ed0ab89a91052581" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.26.mlp.down_proj.q_weight", + "name": "model.layers.25.mlp.down_proj.weight", "shape": [ 1536, - 1120 + 8960 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "f373b934ab548fb2b04a729c5f22dd0b" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.26.mlp.down_proj.q_scale", + "name": "model.layers.25.mlp.gate_up_proj.weight", "shape": [ - 1536, - 280 + 17920, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 55050240, + "byteOffset": 0 } ], - "md5sum": "ceb962974aa3ffd06a616a6c52270097" + "md5sum": "496023d69be445e33b091efe17662be2" }, { - "dataPath": "params_shard_22.bin", + "dataPath": "params_shard_45.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 33060864, "records": [ { - "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "name": "model.layers.22.self_attn.c_attn.weight", "shape": [ - 17920, - 192 + 2048, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 6291456, "byteOffset": 0 }, { - "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "name": "model.layers.22.self_attn.o_proj.weight", "shape": [ - 17920, - 48 + 1536, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.26.post_attention_layernorm.weight", + "name": "model.layers.23.input_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 11010048 }, { - "name": "model.layers.26.self_attn.c_attn.bias", + "name": "model.layers.23.post_attention_layernorm.weight", "shape": [ - 2048 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 + "nbytes": 3072, + "byteOffset": 11013120 }, { - "name": "model.layers.26.self_attn.c_attn.q_weight", + "name": "model.layers.23.self_attn.c_attn.bias", "shape": [ - 2048, - 192 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 11016192 }, { - "name": "model.layers.26.self_attn.c_attn.q_scale", + "name": "model.layers.23.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 11020288 }, { - "name": "model.layers.26.self_attn.o_proj.q_weight", + "name": "model.layers.23.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 17311744 }, { - "name": "model.layers.26.self_attn.o_proj.q_scale", + "name": "model.layers.24.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 22030336 }, { - "name": "model.layers.27.input_layernorm.weight", + "name": "model.layers.24.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 22033408 }, { - "name": "model.layers.27.mlp.down_proj.q_weight", + "name": "model.layers.24.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 22036480 }, { - "name": "model.layers.27.mlp.down_proj.q_scale", + "name": "model.layers.24.self_attn.c_attn.weight", "shape": [ - 1536, - 280 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "ab9d7f443673b7890cfbf49008594ab9" - }, - { - "dataPath": "params_shard_23.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ + "nbytes": 6291456, + "byteOffset": 22040576 + }, { - "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "name": "model.layers.24.self_attn.o_proj.weight", "shape": [ - 17920, - 192 + 1536, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 28332032 }, { - "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "name": "model.layers.25.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 33050624 }, { - "name": "model.layers.27.post_attention_layernorm.weight", + "name": "model.layers.25.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 33053696 }, { - "name": "model.layers.27.self_attn.c_attn.bias", + "name": "model.layers.25.self_attn.c_attn.bias", "shape": [ 2048 ], "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.27.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, + "byteOffset": 33056768 + } + ], + "md5sum": "b162a58ff40747786153f9fbf3562eb4" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.27.self_attn.c_attn.q_scale", + "name": "model.layers.26.mlp.down_proj.weight", "shape": [ - 2048, - 48 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "caa76d72478fe6666accec3e8334136d" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.27.self_attn.o_proj.q_weight", + "name": "model.layers.26.mlp.gate_up_proj.weight", "shape": [ - 1536, - 192 + 17920, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "fbfea3bf12262f3f8f987854e1902f72" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.27.self_attn.o_proj.q_scale", + "name": "model.layers.27.mlp.down_proj.weight", "shape": [ 1536, - 48 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "5ecb3464368f9a2575552ed93f8f800e" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.3.input_layernorm.weight", + "name": "model.layers.27.mlp.gate_up_proj.weight", "shape": [ + 17920, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "702514b36ca983e41199b4370005eb76" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.3.mlp.down_proj.q_weight", + "name": "model.layers.3.mlp.down_proj.weight", "shape": [ 1536, - 1120 + 8960 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "093b34e27d9159a6433c31b0ed2b2c39" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.3.mlp.down_proj.q_scale", + "name": "model.layers.3.mlp.gate_up_proj.weight", "shape": [ - 1536, - 280 + 17920, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 55050240, + "byteOffset": 0 } ], - "md5sum": "61f19df02bf82992f18b0a6ec1370776" + "md5sum": "c3dea40fde4fba6e8b5ec5702df0519d" }, { - "dataPath": "params_shard_24.bin", + "dataPath": "params_shard_52.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 33060864, "records": [ { - "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "name": "model.layers.25.self_attn.c_attn.weight", "shape": [ - 17920, - 192 + 2048, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 6291456, "byteOffset": 0 }, { - "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "name": "model.layers.25.self_attn.o_proj.weight", "shape": [ - 17920, - 48 + 1536, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.3.post_attention_layernorm.weight", + "name": "model.layers.26.input_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 11010048 }, { - "name": "model.layers.3.self_attn.c_attn.bias", + "name": "model.layers.26.post_attention_layernorm.weight", "shape": [ - 2048 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 + "nbytes": 3072, + "byteOffset": 11013120 }, { - "name": "model.layers.3.self_attn.c_attn.q_weight", + "name": "model.layers.26.self_attn.c_attn.bias", "shape": [ - 2048, - 192 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 11016192 }, { - "name": "model.layers.3.self_attn.c_attn.q_scale", + "name": "model.layers.26.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 11020288 }, { - "name": "model.layers.3.self_attn.o_proj.q_weight", + "name": "model.layers.26.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 17311744 }, { - "name": "model.layers.3.self_attn.o_proj.q_scale", + "name": "model.layers.27.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 22030336 }, { - "name": "model.layers.4.input_layernorm.weight", + "name": "model.layers.27.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 22033408 }, { - "name": "model.layers.4.mlp.down_proj.q_weight", + "name": "model.layers.27.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 22036480 }, { - "name": "model.layers.4.mlp.down_proj.q_scale", + "name": "model.layers.27.self_attn.c_attn.weight", "shape": [ - 1536, - 280 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "9bd46c03febaca7936dd5c670e87c9db" - }, - { - "dataPath": "params_shard_25.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ + "nbytes": 6291456, + "byteOffset": 22040576 + }, { - "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "name": "model.layers.27.self_attn.o_proj.weight", "shape": [ - 17920, - 192 + 1536, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 28332032 }, { - "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "name": "model.layers.3.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 33050624 }, { - "name": "model.layers.4.post_attention_layernorm.weight", + "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 33053696 }, { - "name": "model.layers.4.self_attn.c_attn.bias", + "name": "model.layers.3.self_attn.c_attn.bias", "shape": [ 2048 ], "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.4.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, + "byteOffset": 33056768 + } + ], + "md5sum": "3e354febc906d34d54536ea7885c5889" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.4.self_attn.c_attn.q_scale", + "name": "model.layers.4.mlp.down_proj.weight", "shape": [ - 2048, - 48 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "b6824e2ce69ee5ac3b231da74405c28d" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.4.self_attn.o_proj.q_weight", + "name": "model.layers.4.mlp.gate_up_proj.weight", "shape": [ - 1536, - 192 + 17920, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "577fbe9d5221ad319404760818265f2f" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.4.self_attn.o_proj.q_scale", + "name": "model.layers.5.mlp.down_proj.weight", "shape": [ 1536, - 48 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "dc97e694cd68d26939b3fa7a68b6437e" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.5.input_layernorm.weight", + "name": "model.layers.5.mlp.gate_up_proj.weight", "shape": [ + 17920, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "e029aaef82109dae100ed0d76abf03f4" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.5.mlp.down_proj.q_weight", + "name": "model.layers.6.mlp.down_proj.weight", "shape": [ 1536, - 1120 + 8960 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "0cebedd6e22c5f3c4ad9de9a4a301b0c" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.5.mlp.down_proj.q_scale", + "name": "model.layers.6.mlp.gate_up_proj.weight", "shape": [ - 1536, - 280 + 17920, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 55050240, + "byteOffset": 0 } ], - "md5sum": "a967b6249d494149ba4d36888fb580be" + "md5sum": "d021ceb96965eac2d6acd5040e31ff2b" }, { - "dataPath": "params_shard_26.bin", + "dataPath": "params_shard_59.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 33060864, "records": [ { - "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "name": "model.layers.3.self_attn.c_attn.weight", "shape": [ - 17920, - 192 + 2048, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 6291456, "byteOffset": 0 }, { - "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "name": "model.layers.3.self_attn.o_proj.weight", "shape": [ - 17920, - 48 + 1536, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.5.post_attention_layernorm.weight", + "name": "model.layers.4.input_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 11010048 }, { - "name": "model.layers.5.self_attn.c_attn.bias", + "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ - 2048 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 + "nbytes": 3072, + "byteOffset": 11013120 }, { - "name": "model.layers.5.self_attn.c_attn.q_weight", + "name": "model.layers.4.self_attn.c_attn.bias", "shape": [ - 2048, - 192 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 11016192 }, { - "name": "model.layers.5.self_attn.c_attn.q_scale", + "name": "model.layers.4.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 + "nbytes": 6291456, + "byteOffset": 11020288 }, { - "name": "model.layers.5.self_attn.o_proj.q_weight", + "name": "model.layers.4.self_attn.o_proj.weight", "shape": [ 1536, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 17311744 }, { - "name": "model.layers.5.self_attn.o_proj.q_scale", + "name": "model.layers.5.input_layernorm.weight", "shape": [ - 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 3072, + "byteOffset": 22030336 }, { - "name": "model.layers.6.input_layernorm.weight", + "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 1536 ], "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 22033408 }, { - "name": "model.layers.6.mlp.down_proj.q_weight", + "name": "model.layers.5.self_attn.c_attn.bias", "shape": [ - 1536, - 1120 + 2048 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4096, + "byteOffset": 22036480 }, { - "name": "model.layers.6.mlp.down_proj.q_scale", + "name": "model.layers.5.self_attn.c_attn.weight", "shape": [ - 1536, - 280 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "c456c605796669a1a035b5509af8b906" - }, - { - "dataPath": "params_shard_27.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ + "nbytes": 6291456, + "byteOffset": 22040576 + }, { - "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "name": "model.layers.5.self_attn.o_proj.weight", "shape": [ - 17920, - 192 + 1536, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "dtype": "bfloat16", + "format": "raw", + "nbytes": 4718592, + "byteOffset": 28332032 }, { - "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "name": "model.layers.6.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 33050624 }, { "name": "model.layers.6.post_attention_layernorm.weight", @@ -3164,7 +2320,7 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 33053696 }, { "name": "model.layers.6.self_attn.c_attn.bias", @@ -3174,239 +2330,213 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.6.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, + "byteOffset": 33056768 + } + ], + "md5sum": "ae36f42102ecf32968c4b9a5494a5632" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.6.self_attn.c_attn.q_scale", + "name": "model.layers.7.mlp.down_proj.weight", "shape": [ - 2048, - 48 + 1536, + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "4f8b37541d2669f35474558755ad3595" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.6.self_attn.o_proj.q_weight", + "name": "model.layers.7.mlp.gate_up_proj.weight", "shape": [ - 1536, - 192 + 17920, + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 - }, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "301e207ed6aae4c43a940a2cb0041422" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.6.self_attn.o_proj.q_scale", + "name": "model.layers.8.mlp.down_proj.weight", "shape": [ 1536, - 48 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 - }, + "nbytes": 27525120, + "byteOffset": 0 + } + ], + "md5sum": "94c308f7fab6e6806497419fb2bddd14" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 55050240, + "records": [ { - "name": "model.layers.7.input_layernorm.weight", + "name": "model.layers.8.mlp.gate_up_proj.weight", "shape": [ + 17920, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.7.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, + "nbytes": 55050240, + "byteOffset": 0 + } + ], + "md5sum": "125d1635c4fb0ef278b46c2b6f644c0a" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 27525120, + "records": [ { - "name": "model.layers.7.mlp.down_proj.q_scale", + "name": "model.layers.9.mlp.down_proj.weight", "shape": [ 1536, - 280 + 8960 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 + "nbytes": 27525120, + "byteOffset": 0 } ], - "md5sum": "4932c078feff0216f175050579e08a9a" + "md5sum": "e77e14c76d670413b391d98877a76d86" }, { - "dataPath": "params_shard_28.bin", + "dataPath": "params_shard_65.bin", "format": "raw-shard", - "nbytes": 26331136, + "nbytes": 55050240, "records": [ { - "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "name": "model.layers.9.mlp.gate_up_proj.weight", "shape": [ 17920, - 192 + 1536 ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, + "dtype": "bfloat16", + "format": "raw", + "nbytes": 55050240, "byteOffset": 0 - }, + } + ], + "md5sum": "7ade211e79fd496efc5cbfcad1c50060" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 33060864, + "records": [ { - "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "name": "model.layers.6.self_attn.c_attn.weight", "shape": [ - 17920, - 48 + 2048, + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 6291456, + "byteOffset": 0 }, { - "name": "model.layers.7.post_attention_layernorm.weight", + "name": "model.layers.6.self_attn.o_proj.weight", "shape": [ + 1536, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 15482880 + "nbytes": 4718592, + "byteOffset": 6291456 }, { - "name": "model.layers.7.self_attn.c_attn.bias", + "name": "model.layers.7.input_layernorm.weight", "shape": [ - 2048 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.7.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 + "nbytes": 3072, + "byteOffset": 11010048 }, { - "name": "model.layers.7.self_attn.c_attn.q_scale", + "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ - 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.7.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "nbytes": 3072, + "byteOffset": 11013120 }, { - "name": "model.layers.7.self_attn.o_proj.q_scale", + "name": "model.layers.7.self_attn.c_attn.bias", "shape": [ - 1536, - 48 + 2048 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 4096, + "byteOffset": 11016192 }, { - "name": "model.layers.8.input_layernorm.weight", + "name": "model.layers.7.self_attn.c_attn.weight", "shape": [ + 2048, 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.8.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 + "nbytes": 6291456, + "byteOffset": 11020288 }, { - "name": "model.layers.8.mlp.down_proj.q_scale", + "name": "model.layers.7.self_attn.o_proj.weight", "shape": [ 1536, - 280 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "8686b64ff10449b640afc32ed7439a65" - }, - { - "dataPath": "params_shard_29.bin", - "format": "raw-shard", - "nbytes": 26331136, - "records": [ - { - "name": "model.layers.8.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 + "nbytes": 4718592, + "byteOffset": 17311744 }, { - "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "name": "model.layers.8.input_layernorm.weight", "shape": [ - 17920, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "nbytes": 3072, + "byteOffset": 22030336 }, { "name": "model.layers.8.post_attention_layernorm.weight", @@ -3416,7 +2546,7 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 22033408 }, { "name": "model.layers.8.self_attn.c_attn.bias", @@ -3426,51 +2556,29 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.8.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 + "byteOffset": 22036480 }, { - "name": "model.layers.8.self_attn.c_attn.q_scale", + "name": "model.layers.8.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.8.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "nbytes": 6291456, + "byteOffset": 22040576 }, { - "name": "model.layers.8.self_attn.o_proj.q_scale", + "name": "model.layers.8.self_attn.o_proj.weight", "shape": [ 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 4718592, + "byteOffset": 28332032 }, { "name": "model.layers.9.input_layernorm.weight", @@ -3480,59 +2588,7 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 - }, - { - "name": "model.layers.9.mlp.down_proj.q_weight", - "shape": [ - 1536, - 1120 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 6881280, - "byteOffset": 18589696 - }, - { - "name": "model.layers.9.mlp.down_proj.q_scale", - "shape": [ - 1536, - 280 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 860160, - "byteOffset": 25470976 - } - ], - "md5sum": "037698a9234e2d563a950ced66a97bf3" - }, - { - "dataPath": "params_shard_30.bin", - "format": "raw-shard", - "nbytes": 18589696, - "records": [ - { - "name": "model.layers.9.mlp.gate_up_proj.q_weight", - "shape": [ - 17920, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 13762560, - "byteOffset": 0 - }, - { - "name": "model.layers.9.mlp.gate_up_proj.q_scale", - "shape": [ - 17920, - 48 - ], - "dtype": "bfloat16", - "format": "raw", - "nbytes": 1720320, - "byteOffset": 13762560 + "byteOffset": 33050624 }, { "name": "model.layers.9.post_attention_layernorm.weight", @@ -3542,7 +2598,7 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 15482880 + "byteOffset": 33053696 }, { "name": "model.layers.9.self_attn.c_attn.bias", @@ -3552,51 +2608,37 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 4096, - "byteOffset": 15485952 - }, - { - "name": "model.layers.9.self_attn.c_attn.q_weight", - "shape": [ - 2048, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1572864, - "byteOffset": 15490048 - }, + "byteOffset": 33056768 + } + ], + "md5sum": "5471237681d93c50e780f7a581d37dce" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 11013120, + "records": [ { - "name": "model.layers.9.self_attn.c_attn.q_scale", + "name": "model.layers.9.self_attn.c_attn.weight", "shape": [ 2048, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 196608, - "byteOffset": 17062912 - }, - { - "name": "model.layers.9.self_attn.o_proj.q_weight", - "shape": [ - 1536, - 192 - ], - "dtype": "uint32", - "format": "f32-to-bf16", - "nbytes": 1179648, - "byteOffset": 17259520 + "nbytes": 6291456, + "byteOffset": 0 }, { - "name": "model.layers.9.self_attn.o_proj.q_scale", + "name": "model.layers.9.self_attn.o_proj.weight", "shape": [ 1536, - 48 + 1536 ], "dtype": "bfloat16", "format": "raw", - "nbytes": 147456, - "byteOffset": 18439168 + "nbytes": 4718592, + "byteOffset": 6291456 }, { "name": "model.norm.weight", @@ -3606,10 +2648,10 @@ "dtype": "bfloat16", "format": "raw", "nbytes": 3072, - "byteOffset": 18586624 + "byteOffset": 11010048 } ], - "md5sum": "e5c2ee400f5502dbba790e85967f0990" + "md5sum": "f709c941037554e6d4ae56ecba08fc03" } ] } \ No newline at end of file