diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,2863 @@ +{ + "metadata": { + "ParamSize": 245, + "ParamBytes": 756949504.0, + "BitsPerParam": 4.500857753037146 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 32770048, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 32002, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 32770048, + "byteOffset": 0 + } + ], + "md5sum": "d30f75ba0a31394156baf776c8fd6184" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 32770048, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 32002, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 32770048, + "byteOffset": 0 + } + ], + "md5sum": "194dce82783af03303b97a3ae88bf16b" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 32866816, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 32002, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096256, + "byteOffset": 0 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 4096256 + }, + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 4100352 + }, + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 9736448 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 10440960 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 21713152 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23122176 + }, + { + "name": "model.norm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 23126272 + }, + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 32002, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096256, + "byteOffset": 23130368 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 27226624 + }, + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 27230720 + } + ], + "md5sum": "292c2a085b0674e22fdcba62198451d5" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 29171712, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 0 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 704512 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11976704 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 13385728 + }, + { + "name": "model.layers.0.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 13389824 + }, + { + "name": "model.layers.0.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 19681280 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 20467712 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 22564864 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22827008 + }, + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22831104 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 28467200 + } + ], + "md5sum": "11619ac1748dd5977b1d30d61a0cd4f1" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.1.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.1.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "5f32e60145f1f50bb0e3f37e7080f93d" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.10.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.10.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "db95ac126069cb3aef3de3ca2434ae33" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.11.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.11.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "7a8944c2579db0d82562939d0126e492" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.12.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.12.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "8cd8e2a8e37c370fba032509ceb56dcb" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.13.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.13.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "9f70a86578d51a203c7907092ba599ac" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.14.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.14.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "ecd9a0367b8e76b08c2f22a7ba17ed6a" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.15.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.15.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "1a95d231b049b843ba00c3cd13bdca55" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.16.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.16.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "9d9f7cb31766f1f31a41a3b6667f7000" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.17.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.17.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "ad3a55a91da3709fd0e8ced03476fb53" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.18.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.18.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "a6b901242a9b5680ef957ade2d02a36a" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.19.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.19.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "3429075b27224afdd0f789a4b5d8c669" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.2.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.2.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "b1b429b102084606266e42e6343cd513" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.20.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.20.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "b4b2c3e9b08b0c2eac4d39d9025d760d" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.21.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.21.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 22122496 + }, + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 22126592 + }, + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 27762688 + } + ], + "md5sum": "90b1ae2e81cd31c14bc61095e025eea8" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 31563776, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 0 + }, + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 11272192 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 12681216 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 12685312 + }, + { + "name": "model.layers.22.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 18976768 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 19763200 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 21860352 + }, + { + "name": "model.layers.23.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 22122496 + }, + { + "name": "model.layers.23.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 28413952 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 29200384 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 31297536 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 31559680 + } + ], + "md5sum": "81fb7747abf8f4e71c8d26593150feea" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 5636096 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 6340608 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 17612800 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19021824 + }, + { + "name": "model.layers.3.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 19025920 + }, + { + "name": "model.layers.3.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 25317376 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26103808 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28200960 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28463104 + } + ], + "md5sum": "e5bf2fa0200f8dfdddc35f338e395d6d" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 5636096 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 6340608 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 17612800 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19021824 + }, + { + "name": "model.layers.4.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 19025920 + }, + { + "name": "model.layers.4.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 25317376 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26103808 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28200960 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28463104 + } + ], + "md5sum": "ee009fa8df59bb28feef9161d4d17a6d" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 5636096 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 6340608 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 17612800 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19021824 + }, + { + "name": "model.layers.5.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 19025920 + }, + { + "name": "model.layers.5.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 25317376 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26103808 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28200960 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28463104 + } + ], + "md5sum": "795a9ecce39984f2866d702e79f2593f" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 5636096 + }, + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 6340608 + }, + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 17612800 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19021824 + }, + { + "name": "model.layers.6.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 19025920 + }, + { + "name": "model.layers.6.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 25317376 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26103808 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28200960 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28463104 + } + ], + "md5sum": "65ff80178f24d2fc9863ddb90b46b103" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 5636096 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 6340608 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 17612800 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19021824 + }, + { + "name": "model.layers.7.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 19025920 + }, + { + "name": "model.layers.7.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 25317376 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26103808 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28200960 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28463104 + } + ], + "md5sum": "3066afb5e44106aae0f2e42cd3e9cdf4" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 28467200, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 5636096 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 6340608 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 17612800 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19021824 + }, + { + "name": "model.layers.8.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 19025920 + }, + { + "name": "model.layers.8.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 25317376 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26103808 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28200960 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 28463104 + } + ], + "md5sum": "0ee7fba9afd48d74ba86d5ec7b7ac88d" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 28463104, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 2048, + 688 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 2048, + 172 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 704512, + "byteOffset": 5636096 + }, + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 11008, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 11272192, + "byteOffset": 6340608 + }, + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 11008, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1409024, + "byteOffset": 17612800 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 2048 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 4096, + "byteOffset": 19021824 + }, + { + "name": "model.layers.9.self_attn.qkv_proj.q_weight", + "shape": [ + 6144, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 6291456, + "byteOffset": 19025920 + }, + { + "name": "model.layers.9.self_attn.qkv_proj.q_scale", + "shape": [ + 6144, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 786432, + "byteOffset": 25317376 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 2048, + 256 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 2097152, + "byteOffset": 26103808 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 2048, + 64 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 262144, + "byteOffset": 28200960 + } + ], + "md5sum": "3095d4180f5e64d4305d5528c8e1d718" + } + ] +} \ No newline at end of file