{ "metadata": { "ParamSize": 405, "ParamBytes": 12155136000.0, "BitsPerParam": 2.8867256128728958 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 335544320, "records": [ { "name": "lm_head.q_weight", "shape": [ 131072, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 335544320, "byteOffset": 0 } ], "md5sum": "4cd8cb6089c8f84b9ae8298495b751db" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.37.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "2f8eda3bf5f1604f0be49bb22434b419" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.38.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "7e1e5de3294ebea88c7b4e74484ca92c" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.38.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "bd5825881dda4650ab4dad690476805e" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 23633920, "records": [ { "name": "lm_head.q_scale", "shape": [ 131072, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.37.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10485760 }, { "name": "model.layers.37.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 10496000 }, { "name": "model.layers.37.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 15738880 }, { "name": "model.layers.38.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 15749120 }, { "name": "model.layers.38.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 15759360 }, { "name": "model.layers.38.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 18380800 }, { "name": "model.layers.38.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 23623680 } ], "md5sum": "c672d49a7f9ef575720546ab633f4c7a" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.39.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "da3871ac540552167f9273e54b996ea1" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.39.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "10f733a449915f4b4eac86c6b80e77fd" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.38.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.38.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.38.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.38.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.39.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.39.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "763d6b65eb37255616eff47263e976a1" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 335544320, "records": [ { "name": "model.embed_tokens.q_weight", "shape": [ 131072, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 335544320, "byteOffset": 0 } ], "md5sum": "8018983cc2a6d29059ee6898ebba278a" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 32296960, "records": [ { "name": "model.layers.39.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.39.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.39.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.39.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.39.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.39.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 }, { "name": "model.norm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 32286720 } ], "md5sum": "a41732cd115c5993a47617990515ce9d" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "09ea6aef6d7bb96e2426f7fd4820d5df" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "0985928bfdaf0fe6df57ab5ef39973ae" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 18370560, "records": [ { "name": "model.embed_tokens.q_scale", "shape": [ 131072, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10485760 }, { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10496000 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13117440 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18360320 } ], "md5sum": "43cf3cbeacf7df85d8bdd051ec21e7ed" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.1.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "8c1bfa08b6ae763240e9cadcfb3a6c24" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "43aeba404cc0adc2177cb5318bf7b554" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.0.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.0.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.0.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.0.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.1.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "cdaf6336c4a6096b2b8f4138d8af980f" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.2.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "711163266f50e8349004a5799e095573" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 32296960, "records": [ { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.1.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.1.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.1.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.1.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 32286720 } ], "md5sum": "c57cdfbbdc41d66dd3cea86f66c0fad5" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "f86cac128f1211248ee7ecdaed038bda" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 24094720, "records": [ { "name": "model.layers.2.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 2621440 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 7864320 }, { "name": "model.layers.2.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 7874560 }, { "name": "model.layers.2.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 23603200 } ], "md5sum": "655c091e1f26d452a31f7a729b81de0f" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 27033600, "records": [ { "name": "model.layers.2.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.2.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.3.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 10813440 }, { "name": "model.layers.3.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 26542080 } ], "md5sum": "a81b96a6458fb66bf4d2f762792637e2" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "f9d878d3c0411630a12001778a9b661c" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "fb9f74a4b125043c9c5895573996946f" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "34b81cbe48ebb4bb3a2b88ec9ff6f243" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "b9b995fddf6cfd2793bf40f1aa955f73" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "6ba83d0c97a745e90e2e52677be29e05" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 29655040, "records": [ { "name": "model.layers.10.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.10.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.10.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.10.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27033600 } ], "md5sum": "6fedd66b3c50feb2b546275aee20c9f2" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "d4a452aa0f6e09b37d5e5644a54da786" }, { "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 32286720, "records": [ { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.11.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5242880 }, { "name": "model.layers.11.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20971520 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21463040 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31948800 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 32276480 } ], "md5sum": "b1e4a6a448d3875d1c3f00b3dd8b3002" }, { "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "aed1676b229586c983df9b7776e32f5f" }, { "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "fef215cc9f4b7e0763e0519e91ed9070" }, { "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "6d1425dc789a9b24298eecbb44739a5a" }, { "dataPath": "params_shard_32.bin", "format": "raw-shard", "nbytes": 31979520, "records": [ { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 2621440 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 7864320 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 7874560 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 7884800 }, { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 10506240 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 15749120 }, { "name": "model.layers.8.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 15759360 }, { "name": "model.layers.8.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 31488000 } ], "md5sum": "148164530a12b76e30524c6faf674d12" }, { "dataPath": "params_shard_33.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "8320ba35e95f5eb62483ccb4364a6ad3" }, { "dataPath": "params_shard_34.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "550520b2ab7b8e7705e8d3e897aacf3c" }, { "dataPath": "params_shard_35.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.8.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.8.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "b2516bb96ecf7a00b310e6eec1e7f88e" }, { "dataPath": "params_shard_36.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "7e3b26f63038b44ad210b138edb74bfa" }, { "dataPath": "params_shard_37.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "57afd4ddaf656991786f5c7bf8789d90" }, { "dataPath": "params_shard_38.bin", "format": "raw-shard", "nbytes": 29685760, "records": [ { "name": "model.layers.9.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.9.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.9.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.9.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27043840 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27054080 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27064320 } ], "md5sum": "77eb2186230ccba76ac8743c87f75071" }, { "dataPath": "params_shard_39.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "2154d56e2a31e348b4c22bacddb27132" }, { "dataPath": "params_shard_40.bin", "format": "raw-shard", "nbytes": 32296960, "records": [ { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.12.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.12.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.12.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.12.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 32286720 } ], "md5sum": "251113b64a7cd0ff8ad01fab4cc39d28" }, { "dataPath": "params_shard_41.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "99f227c77c5eb4c8d0c033218a729e66" }, { "dataPath": "params_shard_42.bin", "format": "raw-shard", "nbytes": 24094720, "records": [ { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 2621440 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 7864320 }, { "name": "model.layers.13.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 7874560 }, { "name": "model.layers.13.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 23603200 } ], "md5sum": "b275d0db28e88bae3173f295591fdf9c" }, { "dataPath": "params_shard_43.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "4a09adff4b64dc48bea339337a3449a3" }, { "dataPath": "params_shard_44.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "08522c5042b73cdc0e58b5a8f070ef76" }, { "dataPath": "params_shard_45.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.13.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.13.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "6379e3adb67901d356efcb58362fc887" }, { "dataPath": "params_shard_46.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "2df96e4aa6163c4f0e70dd29a58f2e91" }, { "dataPath": "params_shard_47.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "28e20e1115b6e26b148e8b60939d53c3" }, { "dataPath": "params_shard_48.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.14.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.14.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.14.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.14.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "f69a164114c38fba1c8b2f60c8287a34" }, { "dataPath": "params_shard_49.bin", "format": "raw-shard", "nbytes": 32286720, "records": [ { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.15.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.15.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 } ], "md5sum": "613b4a50aa529383a1c473fc3435400f" }, { "dataPath": "params_shard_50.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.16.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "4dfd5f74a1b1078e8c2b79ca87ac72d4" }, { "dataPath": "params_shard_51.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "7a41ef0c64801a33c3d4cdc9e596eb05" }, { "dataPath": "params_shard_52.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.16.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.16.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.16.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.16.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.16.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.16.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "f3c7f6e8794169044f53c6cb5a36d729" }, { "dataPath": "params_shard_53.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.17.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "1b54adac49af41f4615462fe7fa1594d" }, { "dataPath": "params_shard_54.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "aa3c0a69955052c5a74cfc6a1d1cefa8" }, { "dataPath": "params_shard_55.bin", "format": "raw-shard", "nbytes": 29358080, "records": [ { "name": "model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.16.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.17.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5253120 }, { "name": "model.layers.17.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 5263360 }, { "name": "model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 7884800 }, { "name": "model.layers.17.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13127680 }, { "name": "model.layers.17.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 13137920 }, { "name": "model.layers.17.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 28866560 } ], "md5sum": "1beab04934dfd88b02a9ec31e54d212d" }, { "dataPath": "params_shard_56.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.18.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "781bb152888be2e56be947dca236cc72" }, { "dataPath": "params_shard_57.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "b1fc7db0def4b612379859795235975e" }, { "dataPath": "params_shard_58.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.17.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.17.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.18.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.18.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.18.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "16b3aebd895ba73e1367a2c32705f57f" }, { "dataPath": "params_shard_59.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.19.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "172f88509c3a27e5de0d26c070e88e63" }, { "dataPath": "params_shard_60.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "51959c950f4ea4be7e5e8ceb2bfa0117" }, { "dataPath": "params_shard_61.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.18.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.18.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.18.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.18.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.19.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.19.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "115b689f3f09ac83946980a875d9e0ae" }, { "dataPath": "params_shard_62.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "ca2dcbc97ce3e42e4456375a7afcbbde" }, { "dataPath": "params_shard_63.bin", "format": "raw-shard", "nbytes": 32286720, "records": [ { "name": "model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.19.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.19.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.19.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.19.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.19.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 } ], "md5sum": "a9f418ebd6753b8cccca692f62858625" }, { "dataPath": "params_shard_64.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.20.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "50f13bf85419d7e2dbe990b3985bfdf1" }, { "dataPath": "params_shard_65.bin", "format": "raw-shard", "nbytes": 32286720, "records": [ { "name": "model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.20.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5242880 }, { "name": "model.layers.20.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20971520 }, { "name": "model.layers.20.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21463040 }, { "name": "model.layers.20.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31948800 }, { "name": "model.layers.20.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 32276480 } ], "md5sum": "d3c5a9efa44632b185e6110d5557c706" }, { "dataPath": "params_shard_66.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.21.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "a5cd1eddc046d9e48672f4db3695a9db" }, { "dataPath": "params_shard_67.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "2cb77689ac01c0b5a4c75afa307f06dd" }, { "dataPath": "params_shard_68.bin", "format": "raw-shard", "nbytes": 26736640, "records": [ { "name": "model.layers.20.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "model.layers.20.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2621440 }, { "name": "model.layers.21.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 2631680 }, { "name": "model.layers.21.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 2641920 }, { "name": "model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 5263360 }, { "name": "model.layers.21.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10506240 }, { "name": "model.layers.21.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 10516480 }, { "name": "model.layers.21.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 26245120 } ], "md5sum": "a1ecb9dce5a9b083720cc8c9a325f4f1" }, { "dataPath": "params_shard_69.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.22.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "7f8dcfd0a806db539c209a2c63aecfa4" }, { "dataPath": "params_shard_70.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "bdddfa35af03608f5780c8b376101f41" }, { "dataPath": "params_shard_71.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.21.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.21.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.22.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.22.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.22.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "e6d238ee8c8ef09a70e734261d03a664" }, { "dataPath": "params_shard_72.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.23.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "ceadfc7c624d5730792b4080be182f13" }, { "dataPath": "params_shard_73.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "69e5cafb9ea462c88be925a5addcbaed" }, { "dataPath": "params_shard_74.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.22.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.22.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.22.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.22.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.23.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.23.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "34f8e0458f60001f0a271d7ad2ca5796" }, { "dataPath": "params_shard_75.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.24.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "d8a7ed9c0d2e873d1cc3aa2a8e848981" }, { "dataPath": "params_shard_76.bin", "format": "raw-shard", "nbytes": 32286720, "records": [ { "name": "model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.23.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.23.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.23.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.23.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.23.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 } ], "md5sum": "6eb0a334996eddfac82341212590457a" }, { "dataPath": "params_shard_77.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.24.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "8359e63b839e36c08128bf12168c1651" }, { "dataPath": "params_shard_78.bin", "format": "raw-shard", "nbytes": 24084480, "records": [ { "name": "model.layers.24.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "model.layers.24.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 2621440 }, { "name": "model.layers.24.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 7864320 }, { "name": "model.layers.24.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 23592960 } ], "md5sum": "9c443ad70f74d39690e968b2ce7fb3bf" }, { "dataPath": "params_shard_79.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.25.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "a61df88e75fce7953180fc21d1ca964e" }, { "dataPath": "params_shard_80.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.25.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "854922a9e907f6905f8033122e151548" }, { "dataPath": "params_shard_81.bin", "format": "raw-shard", "nbytes": 18718720, "records": [ { "name": "model.layers.24.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.24.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.24.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.24.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10823680 }, { "name": "model.layers.25.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10833920 }, { "name": "model.layers.25.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10844160 }, { "name": "model.layers.25.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13465600 }, { "name": "model.layers.25.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18708480 } ], "md5sum": "e0bdc3ad05833ec4cdd08796cc7c4d09" }, { "dataPath": "params_shard_82.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.26.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "d7d42ca962087bc97c4808503552adf9" }, { "dataPath": "params_shard_83.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.26.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "596fae1985c506ee9346ae72eb191bd8" }, { "dataPath": "params_shard_84.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.25.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.25.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.25.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.25.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.26.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.26.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "dda4cd3e3e2c88b84f9e874191095896" }, { "dataPath": "params_shard_85.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.27.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "17318d5f3afa3658b46cad3db41eceb1" }, { "dataPath": "params_shard_86.bin", "format": "raw-shard", "nbytes": 32296960, "records": [ { "name": "model.layers.26.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.26.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.26.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.26.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.26.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.26.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 }, { "name": "model.layers.27.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 32286720 } ], "md5sum": "3d53a96594b907fed6dc092f14997148" }, { "dataPath": "params_shard_87.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.27.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "18cff1b300fcace0c4d448d98f66db61" }, { "dataPath": "params_shard_88.bin", "format": "raw-shard", "nbytes": 24094720, "records": [ { "name": "model.layers.27.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "model.layers.27.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 2621440 }, { "name": "model.layers.27.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 7864320 }, { "name": "model.layers.27.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 7874560 }, { "name": "model.layers.27.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 23603200 } ], "md5sum": "91c6a5552a33ca4bf8bd2eb1da6f497c" }, { "dataPath": "params_shard_89.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.28.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "10df9f8fb233142f9bd67f36905d1c2c" }, { "dataPath": "params_shard_90.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.28.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "64b0019cc1bc5b11e0c7d3580752b66e" }, { "dataPath": "params_shard_91.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.27.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.27.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.28.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.28.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.28.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.28.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "64c2d0022f787c8fc355435674c151a6" }, { "dataPath": "params_shard_92.bin", "format": "raw-shard", "nbytes": 27033600, "records": [ { "name": "model.layers.28.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.28.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.28.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.28.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 } ], "md5sum": "a9209c9c26af021a9021c3ffe260c35d" }, { "dataPath": "params_shard_93.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.29.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "e55c83d56816efc7526c53035dbdbd51" }, { "dataPath": "params_shard_94.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.29.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "3cfd6df9cd917b44de5c33cc915af78f" }, { "dataPath": "params_shard_95.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.29.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.29.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.29.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.29.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.29.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.29.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "652499ad8ead7b6dda98aa3cf2141479" }, { "dataPath": "params_shard_96.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.30.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "b6b10a6eed26f58ed21b92c8af927242" }, { "dataPath": "params_shard_97.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.30.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "dfbf67d1327cb93722aa41e10f6adef8" }, { "dataPath": "params_shard_98.bin", "format": "raw-shard", "nbytes": 29358080, "records": [ { "name": "model.layers.29.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.29.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.30.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5253120 }, { "name": "model.layers.30.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 5263360 }, { "name": "model.layers.30.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 7884800 }, { "name": "model.layers.30.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 13127680 }, { "name": "model.layers.30.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 13137920 }, { "name": "model.layers.30.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 28866560 } ], "md5sum": "6483f57960e59407dc03a01e07555416" }, { "dataPath": "params_shard_99.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.31.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "5ed9c7fefd995d98aea58b8436a11840" }, { "dataPath": "params_shard_100.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.31.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "e7b31811ded52465c34b159af9cc4dfc" }, { "dataPath": "params_shard_101.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.30.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.30.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.31.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.31.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.31.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.31.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "a6e9c076b21b790e833e0ac446b6f218" }, { "dataPath": "params_shard_102.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.32.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "56c544a718f4dcb9201ba1c02c661d18" }, { "dataPath": "params_shard_103.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.32.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "f12a8ce483b4f043502b6f90c28c4216" }, { "dataPath": "params_shard_104.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.31.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.31.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.31.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.31.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.32.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.32.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "40b5bd1867b936bca69a7ecad18d6b1f" }, { "dataPath": "params_shard_105.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.33.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "335acfe32ee9fa7dd5d3ed533458ccab" }, { "dataPath": "params_shard_106.bin", "format": "raw-shard", "nbytes": 32286720, "records": [ { "name": "model.layers.32.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.32.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.32.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.32.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.32.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.32.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 } ], "md5sum": "b4131aa2ba3803339fe2c49d12b99140" }, { "dataPath": "params_shard_107.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.3.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "b04b8818684dd98f4abc14654ae1d614" }, { "dataPath": "params_shard_108.bin", "format": "raw-shard", "nbytes": 32286720, "records": [ { "name": "model.layers.33.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.33.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5242880 }, { "name": "model.layers.33.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20971520 }, { "name": "model.layers.33.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21463040 }, { "name": "model.layers.33.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31948800 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 32276480 } ], "md5sum": "564281c91fc298234f5ee18a92777e2d" }, { "dataPath": "params_shard_109.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "cdf31d7039e7358223c79f3393d61b83" }, { "dataPath": "params_shard_110.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "0c9e5a814e31b611c33a7092b70d48d8" }, { "dataPath": "params_shard_111.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "ca215bcba2071927bde25b74605a8add" }, { "dataPath": "params_shard_112.bin", "format": "raw-shard", "nbytes": 31979520, "records": [ { "name": "model.layers.3.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 2621440 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 7864320 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 7874560 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 7884800 }, { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 10506240 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 15749120 }, { "name": "model.layers.4.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 15759360 }, { "name": "model.layers.4.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 31488000 } ], "md5sum": "a2868a77df21d19db9d733bfde2e91e0" }, { "dataPath": "params_shard_113.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "ea003d55666cfe0c81a63abe73f5d28a" }, { "dataPath": "params_shard_114.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "b76cbf71c822bc6c401621e67309ab41" }, { "dataPath": "params_shard_115.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.4.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.4.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "58b86f3ca23c2323634f8f58d2683df8" }, { "dataPath": "params_shard_116.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "394977a497dcd100e78bea2f46408c1c" }, { "dataPath": "params_shard_117.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "7408d7ef20d5f3638c691d3823bf3cc6" }, { "dataPath": "params_shard_118.bin", "format": "raw-shard", "nbytes": 29665280, "records": [ { "name": "model.layers.5.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.5.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.5.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.5.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 } ], "md5sum": "b83e462058b6b987bea8e528d7157e21" }, { "dataPath": "params_shard_119.bin", "format": "raw-shard", "nbytes": 32286720, "records": [ { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.6.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.6.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.6.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.6.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 } ], "md5sum": "de4bce489ec31ec65347e495fad19dbb" }, { "dataPath": "params_shard_120.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.33.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "02376d5b3b959b612c0f868b1846248c" }, { "dataPath": "params_shard_121.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.34.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "49e83f7c0d9d02b24e25612b1a7ea5fa" }, { "dataPath": "params_shard_122.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.34.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "1103f2b3297845f8ef4afc608aa3ad09" }, { "dataPath": "params_shard_123.bin", "format": "raw-shard", "nbytes": 32307200, "records": [ { "name": "model.layers.7.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.7.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.33.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 27033600 }, { "name": "model.layers.33.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27043840 }, { "name": "model.layers.33.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 29665280 }, { "name": "model.layers.34.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 29675520 }, { "name": "model.layers.34.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 29685760 } ], "md5sum": "eda6479cfe3f12e22b33138ba9715d03" }, { "dataPath": "params_shard_124.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.35.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "96be3c0f90227fceea22381202dfec95" }, { "dataPath": "params_shard_125.bin", "format": "raw-shard", "nbytes": 32296960, "records": [ { "name": "model.layers.34.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 0 }, { "name": "model.layers.34.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 5242880 }, { "name": "model.layers.34.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 5253120 }, { "name": "model.layers.34.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 20981760 }, { "name": "model.layers.34.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 21473280 }, { "name": "model.layers.34.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 31959040 }, { "name": "model.layers.35.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 32286720 } ], "md5sum": "e6c3f692c1c80b723feab5a7084be650" }, { "dataPath": "params_shard_126.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.35.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "f16190b02cc87c1b44fdcf0ec3a05211" }, { "dataPath": "params_shard_127.bin", "format": "raw-shard", "nbytes": 24094720, "records": [ { "name": "model.layers.35.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 0 }, { "name": "model.layers.35.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 2621440 }, { "name": "model.layers.35.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 7864320 }, { "name": "model.layers.35.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 7874560 }, { "name": "model.layers.35.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 23603200 } ], "md5sum": "9de5f577c23b5c9890d7a32afa08d8b6" }, { "dataPath": "params_shard_128.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.36.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "badd2f3fbfb02b1e3ebf3c2c37c0bbb1" }, { "dataPath": "params_shard_129.bin", "format": "raw-shard", "nbytes": 167772160, "records": [ { "name": "model.layers.36.mlp.gate_up_proj.q_weight", "shape": [ 65536, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 167772160, "byteOffset": 0 } ], "md5sum": "d4a469a6047496941f7cb0de08a8c154" }, { "dataPath": "params_shard_130.bin", "format": "raw-shard", "nbytes": 18698240, "records": [ { "name": "model.layers.35.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 0 }, { "name": "model.layers.35.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 10485760 }, { "name": "model.layers.36.input_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 10813440 }, { "name": "model.layers.36.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 10823680 }, { "name": "model.layers.36.mlp.gate_up_proj.q_scale", "shape": [ 65536, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 13445120 }, { "name": "model.layers.36.post_attention_layernorm.weight", "shape": [ 5120 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 10240, "byteOffset": 18688000 } ], "md5sum": "3fbf3500f95a83e10f39db006591812a" }, { "dataPath": "params_shard_131.bin", "format": "raw-shard", "nbytes": 83886080, "records": [ { "name": "model.layers.37.mlp.down_proj.q_weight", "shape": [ 5120, 4096 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 83886080, "byteOffset": 0 } ], "md5sum": "bbff5f01311083e0315eb941feb16a5a" }, { "dataPath": "params_shard_132.bin", "format": "raw-shard", "nbytes": 29655040, "records": [ { "name": "model.layers.36.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.36.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.36.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.36.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 }, { "name": "model.layers.37.mlp.down_proj.q_scale", "shape": [ 5120, 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2621440, "byteOffset": 27033600 } ], "md5sum": "24751078dfc0a6ab7008ee8ce7353cb1" }, { "dataPath": "params_shard_133.bin", "format": "raw-shard", "nbytes": 27033600, "records": [ { "name": "model.layers.37.self_attn.qkv_proj.q_weight", "shape": [ 6144, 640 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 15728640, "byteOffset": 0 }, { "name": "model.layers.37.self_attn.qkv_proj.q_scale", "shape": [ 6144, 40 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 491520, "byteOffset": 15728640 }, { "name": "model.layers.37.self_attn.o_proj.q_weight", "shape": [ 5120, 512 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 10485760, "byteOffset": 16220160 }, { "name": "model.layers.37.self_attn.o_proj.q_scale", "shape": [ 5120, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 327680, "byteOffset": 26705920 } ], "md5sum": "12f46e36b4fde5b14178ac65d070729c" } ] }