diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" new file mode 100644--- /dev/null +++ "b/ndarray-cache.json" @@ -0,0 +1,4431 @@ +{ + "metadata": { + "ParamSize": 325, + "ParamBytes": 4212490240.0, + "BitsPerParam": 5.001066770081567 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 65568768, + "records": [ + { + "name": "lm_head.q_weight", + "shape": [ + 32016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 65568768, + "byteOffset": 0 + } + ], + "md5sum": "0d2a47a0c4687331365f7c308194cdf6" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 30748672, + "records": [ + { + "name": "lm_head.q_scale", + "shape": [ + 32016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8196096, + "byteOffset": 0 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8196096 + }, + { + "name": "model.layers.23.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 8204288 + } + ], + "md5sum": "6a79c4f8cddbd0947e5787b34fe60f6e" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "ebb3bf010586bbf4c9290c2a3a1beec1" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 28196864, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2818048 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2826240 + }, + { + "name": "model.layers.24.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 2834432 + }, + { + "name": "model.layers.24.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 25378816 + } + ], + "md5sum": "ccb7c662503976e4364af9eca71b140c" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 30810112, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 5636096 + }, + { + "name": "model.layers.24.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 5644288 + } + ], + "md5sum": "9a6be9f13c90ad3f1f03db173a930ca7" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "5dbb633d9f598d7bf693a8c54b690ac9" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "d313a96f20af4b1908aefb54cae8a474" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "00d59147847f690dc609ae129b0a7672" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.24.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.25.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.25.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.25.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.25.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "9564667bb38f61e19540e8505d0ffacf" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "1c9eb1381e5d8077de55d42496368de8" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "f2be7b7f7b3d83d371f139aa14a3e18c" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.25.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.26.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.26.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.26.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "69f2e668e3b2cd1b10b5c38ffbaa8386" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "bcc6bade5c067a9e524e6b3ca3ce88ad" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "477bc475b9a67aaeffb392055780b652" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "00674b4872329bfe10b9e3e427da4e62" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.26.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.27.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.27.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.27.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.27.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "be991c4e8975d9480ac60ff9b893e165" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.28.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "1e2150c37f3fa055aea28f139c18a2b0" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "882b640b021e693f4b28b9facee130d6" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.27.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.28.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.28.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.28.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.28.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.28.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "8dc59d22d267eb865c221128f5af7c28" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.29.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "584dbb1372234ef5fd4294c11e88039b" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.29.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "3361b6ea6a9917fa266d27095fcfe8d6" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.29.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "188856da72d666ea2ca3adcfdb072c00" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.28.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.28.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.29.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.29.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.29.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.29.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.29.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.29.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "c787ba9daf0949315c929d2e2711e17a" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.30.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "3cadccac583c3810b96fed3e2a6d8fdd" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "900a239828d108eca7334b8c5e0b64fe" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.29.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.30.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.30.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.30.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.30.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.30.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "e8006bf1d3da7144dbb2d53cb97b35f4" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.31.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "f4a6901f85e878f50996343d0cd5dbcb" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.31.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "aa8ce9c339d2eab6c90fffaf6a7a2b28" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.31.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "eb54e366fe084f308628cf0d63263165" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.30.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.30.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.31.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.31.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.31.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.31.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.31.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.31.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "42d1011e1f331210aa5fc6972dbc7d32" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 65568768, + "records": [ + { + "name": "model.embed_tokens.q_weight", + "shape": [ + 32016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 65568768, + "byteOffset": 0 + } + ], + "md5sum": "df470470c319fac0664b5d831d02cc5f" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 31805440, + "records": [ + { + "name": "model.layers.31.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.norm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.embed_tokens.q_scale", + "shape": [ + 32016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8196096, + "byteOffset": 1056768 + }, + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9252864 + }, + { + "name": "model.layers.0.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9261056 + } + ], + "md5sum": "8885f841a5fc2755774de25216cbc296" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "b7b2fc735275f027638772fc5b19a07a" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "e57a6193d604bf7ee6eb742a2941f66e" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "687f2528834514a17b14fd2ab81b7d2b" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "87594d02f0f5cde6f99772ca3209be1b" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "32ece1565b2bcb7df3e1353d6d20d1cf" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 32661504, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.0.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.0.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.0.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.1.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 21053440 + }, + { + "name": "model.layers.1.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 23871488 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 29507584 + }, + { + "name": "model.layers.1.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 29515776 + } + ], + "md5sum": "08d4046966e42836cdd07bf73c9c829a" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.1.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.10.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "32a77f5866adec17c31464f1b189a4de" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "c196b14e44e578b97356e9b1c512ea77" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "45dc579fe5e0a997a17bed7abe52465f" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "5e814bda36ae4aaeb815d94d75a3a351" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "c27486fb0a80869086289c09a7c8e1c2" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 29827072, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.10.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.10.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.10.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.11.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 21045248 + }, + { + "name": "model.layers.11.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 26681344 + } + ], + "md5sum": "a214e17e2b75612b715ea8ad53fd0f2b" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.11.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.11.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.2.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "438c6dd3edc846c1fb1dfc78b39afd34" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "19ab6e5268ec5a6b1bcc0c9a6606457a" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "a52f6fb032489a19a9f7bf94d95b9c9f" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "cdc4529052853e723adb2d02dbb6c091" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "a5d5602df778fd8da8edd402a1ad3d9c" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "214286650b913f8ce57f5255bf72de4f" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 32661504, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.2.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.2.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.2.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.3.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 21053440 + }, + { + "name": "model.layers.3.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 23871488 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 29507584 + }, + { + "name": "model.layers.3.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 29515776 + } + ], + "md5sum": "411a81f9248239e6d9b3c7046cb66ea7" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.3.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.3.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.4.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "1105c8149a08cba1518672d5a85c7cd9" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "aad21fefac6b20bdae9ab21b302141c0" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "9b95dd594bb056dc243f9c835dfdbd40" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "6ee44a40864a6fa26b98a79a3b6faaf3" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "54ee89d7b75a760e02a14ea7e71b1eef" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "1977b9aa9d1fa6b25793cbc949f1659f" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 32661504, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.4.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.4.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.4.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.5.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 21053440 + }, + { + "name": "model.layers.5.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 23871488 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 29507584 + }, + { + "name": "model.layers.5.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 29515776 + } + ], + "md5sum": "bf8e2c6f2574b47b4f5e5688ad6a9b42" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.5.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.5.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.6.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "653918eadced2bfdba5795a425f3088c" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "3670b10e0b995078046207eee0966d99" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "efd5054ce903b93a37f49199aaf9d34b" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "7eaf55d760e003bea81dafc7c11b1bc1" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "f06f6fdd834b222ebafb673c241abd12" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "cefc90b43792196a4975637252d542c5" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 32661504, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.6.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.6.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.6.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.7.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 21053440 + }, + { + "name": "model.layers.7.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 23871488 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 29507584 + }, + { + "name": "model.layers.7.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 29515776 + } + ], + "md5sum": "1a8a1769f5b596f302a15d5220e2b115" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.7.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.7.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.8.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "be4017dfb03f862d9800ae6e9ec65930" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "52ad0f8ef746470f1a4e02485f8c74c0" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "df83bd6364c1a99b050412d02b9e7957" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "cf80bee988cc7678f41fb3bed9b63327" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "2bc29ed59a66a77ec2cba28c7be6a7bb" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "646fd358247bebdb4fd285faa2da32cb" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 32661504, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.8.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 2818048 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 8454144 + }, + { + "name": "model.layers.8.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 8462336 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 11608064 + }, + { + "name": "model.layers.8.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 19996672 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.9.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 21053440 + }, + { + "name": "model.layers.9.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 23871488 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 29507584 + }, + { + "name": "model.layers.9.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 29515776 + } + ], + "md5sum": "4478e16e341c7d1935df1519e2ca25a9" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 31989760, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 0 + }, + { + "name": "model.layers.9.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 8388608 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 9437184 + }, + { + "name": "model.layers.11.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 9445376 + } + ], + "md5sum": "0d8878242ec06de128e619489fb0494f" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "0792e52e0bb622f147e88799da010180" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 28196864, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 0 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2818048 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 2826240 + }, + { + "name": "model.layers.12.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 2834432 + }, + { + "name": "model.layers.12.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 25378816 + } + ], + "md5sum": "dd2549f8d102762ee5b4b5e2e0b349ef" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 30810112, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 0 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 5636096 + }, + { + "name": "model.layers.12.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 5644288 + } + ], + "md5sum": "c4eab47b6fab02657f1e2a92c3cc3157" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "6d159ea8819c1a73a1d2ef20ba77b98d" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "3592814a47269090993beb0ce510de04" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "e54ef8e098bb1ae06b3819b576cf6d1a" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.12.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.13.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.13.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.13.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.13.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "5c880babc24c48e8d837615259c1c9ec" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "e84771cfcb0ca494f5ed81f17bf9934e" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "7ecf039a9721f1e58ae55f569299e3d0" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.13.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.14.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.14.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.14.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "9e7314e9d2fb21f1531ced994d061c07" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "16652f3f22ae817f5b5707b780a593aa" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "df2be6859b302973c1acfcef8c04c332" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "d529ddc9f0524c6d17e09ce7459cf135" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.14.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.15.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.15.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.15.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.15.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "10e99b9aebdbe715c09f73f993a082e4" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "a53b85c344c00d94054df1911e621a34" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "44edec086ab92d4095047faa36834b71" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.15.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.16.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.16.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.16.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "18111cfd9ac34e8f69e205971590a94b" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "3a9d198e58cac4b4674661b25a606f9a" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "8ef63b95507cb8d9640eeab82e44fe80" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "e2aa78dd4a8386b8511039869e35f34f" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.16.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.17.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.17.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.17.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.17.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "1f4edaf2b5883221061c5b1999e0117d" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "049c45acbd0b2bd918b246a5ac74b767" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "127286cde8319b56b5bc45822e7d532f" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.17.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.18.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.18.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.18.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "d40c3ec34f545b6c1b18b9f9fc147805" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "4dbfe34bd62390c00aadbedb86e73c76" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "793fb5e9b0836916158749de35acd819" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "28dc7ec986949cf83cc760f9cc6d7103" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.18.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.19.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.19.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.19.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.19.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "ce7f4614a9013e2edd2496e6c878f2b9" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "c7bb08a9bf169b67804ce0ec22782895" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "c3ee708e67ddf10c4e331d71455f7614" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.20.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.20.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.20.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "3a39133a2359b8f9037598d6666cf515" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 22544384, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 0 + } + ], + "md5sum": "33f7650ae3797d33a0da72c04b394b5b" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "f295e0102144fac772c28b669d699c54" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "898c817ad296fbc07c3657c05b1864c1" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 32587776, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.20.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 12582912 + }, + { + "name": "model.layers.21.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 12591104 + }, + { + "name": "model.layers.21.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 15409152 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 21045248 + }, + { + "name": "model.layers.21.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 21053440 + }, + { + "name": "model.layers.21.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 24199168 + } + ], + "md5sum": "d09cba7adad966edd7a0328aa44c1705" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "42179f6d64e783d2596d6bc8c77b9fad" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "0e3b0417d9df97c25c0692e091f93be2" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 32063488, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 0 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 1048576 + }, + { + "name": "model.layers.22.mlp.down_proj.q_weight", + "shape": [ + 4096, + 1376 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 22544384, + "byteOffset": 1056768 + }, + { + "name": "model.layers.22.mlp.down_proj.q_scale", + "shape": [ + 4096, + 344 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 2818048, + "byteOffset": 23601152 + }, + { + "name": "model.layers.22.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 26419200 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 4096 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 8192, + "byteOffset": 32055296 + } + ], + "md5sum": "2161eac94139e068f8c3194b132870f2" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 45088768, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.q_weight", + "shape": [ + 22016, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 45088768, + "byteOffset": 0 + } + ], + "md5sum": "48df17cd2f304bae79c5077d9ba19932" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.q_weight", + "shape": [ + 12288, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "d42172c540026a76a8169fd2855e3e7f" + }, + { + "dataPath": "params_shard_113.bin", + "format": "raw-shard", + "nbytes": 30801920, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 0 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 3145728 + }, + { + "name": "model.layers.22.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 11534336 + }, + { + "name": "model.layers.23.mlp.gate_up_proj.q_scale", + "shape": [ + 22016, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 5636096, + "byteOffset": 12582912 + }, + { + "name": "model.layers.23.self_attn.qkv_proj.q_scale", + "shape": [ + 12288, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 3145728, + "byteOffset": 18219008 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_weight", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "f32-to-bf16", + "nbytes": 8388608, + "byteOffset": 21364736 + }, + { + "name": "model.layers.23.self_attn.o_proj.q_scale", + "shape": [ + 4096, + 128 + ], + "dtype": "float32", + "format": "f32-to-bf16", + "nbytes": 1048576, + "byteOffset": 29753344 + } + ], + "md5sum": "73f31a06126c71de8d1ba9066179c1c9" + } + ] +} \ No newline at end of file