{ "metadata": { "ParamSize": 339, "ParamBytes": 335372288.0, "BitsPerParam": 3.569535268985962 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 77791232, "records": [ { "name": "model.embed_tokens.q_weight", "shape": [ 151936, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 77791232, "byteOffset": 0 } ], "md5sum": "5eff35dad10a3b4e08a44b7c296533eb" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 32740608, "records": [ { "name": "model.embed_tokens.q_scale", "shape": [ 151936, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 9723904, "byteOffset": 0 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 9723904 }, { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 9725952 }, { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 11298816 }, { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 11495424 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 14641152 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 15034368 }, { "name": "model.layers.0.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 15036416 }, { "name": "model.layers.0.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 15036672 }, { "name": "model.layers.0.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 17133824 }, { "name": "model.layers.0.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 17395968 }, { "name": "model.layers.0.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 18444544 }, { "name": "model.layers.0.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 18575616 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 18575872 }, { "name": "model.layers.1.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 18577920 }, { "name": "model.layers.1.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 20150784 }, { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 20347392 }, { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 23493120 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 23886336 }, { "name": "model.layers.1.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 23888384 }, { "name": "model.layers.1.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23888640 }, { "name": "model.layers.1.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 25985792 }, { "name": "model.layers.1.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 26247936 }, { "name": "model.layers.1.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 27296512 }, { "name": "model.layers.1.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 27427584 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 27427840 }, { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 27429888 }, { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 29002752 }, { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 29199360 }, { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 32345088 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 32738304 }, { "name": "model.layers.10.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 32740352 } ], "md5sum": "aef46e898a7ac7cb7a517e0ffa0e3526" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 31866624, "records": [ { "name": "model.layers.10.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.10.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.10.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 2359296 }, { "name": "model.layers.10.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 3407872 }, { "name": "model.layers.10.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 3538944 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 3539200 }, { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 3541248 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 5114112 }, { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 5310720 }, { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 8456448 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 8849664 }, { "name": "model.layers.11.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 8851712 }, { "name": "model.layers.11.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 8851968 }, { "name": "model.layers.11.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 10949120 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 11211264 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 12259840 }, { "name": "model.layers.11.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 12390912 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 12391168 }, { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12393216 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 13966080 }, { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14162688 }, { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 17308416 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 17701632 }, { "name": "model.layers.12.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 17703680 }, { "name": "model.layers.12.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17703936 }, { "name": "model.layers.12.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 19801088 }, { "name": "model.layers.12.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 20063232 }, { "name": "model.layers.12.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 21111808 }, { "name": "model.layers.12.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 21242880 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 21243136 }, { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 21245184 }, { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 22818048 }, { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 23014656 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 26160384 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 26553600 }, { "name": "model.layers.13.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 26555648 }, { "name": "model.layers.13.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26555904 }, { "name": "model.layers.13.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28653056 }, { "name": "model.layers.13.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 28915200 }, { "name": "model.layers.13.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 29963776 }, { "name": "model.layers.13.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 30094848 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 30095104 }, { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 30097152 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 31670016 } ], "md5sum": "fd2da2194fa133a993750d535d6376c0" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 33505024, "records": [ { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 3538944 }, { "name": "model.layers.14.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 3540992 }, { "name": "model.layers.14.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3541248 }, { "name": "model.layers.14.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5638400 }, { "name": "model.layers.14.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 5900544 }, { "name": "model.layers.14.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 6949120 }, { "name": "model.layers.14.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 7080192 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 7080448 }, { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 7082496 }, { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 8655360 }, { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 8851968 }, { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 11997696 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 12390912 }, { "name": "model.layers.15.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 12392960 }, { "name": "model.layers.15.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 12393216 }, { "name": "model.layers.15.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 14490368 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 14752512 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 15801088 }, { "name": "model.layers.15.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 15932160 }, { "name": "model.layers.16.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 15932416 }, { "name": "model.layers.16.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 15934464 }, { "name": "model.layers.16.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 17507328 }, { "name": "model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 17703936 }, { "name": "model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 20849664 }, { "name": "model.layers.16.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 21242880 }, { "name": "model.layers.16.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 21244928 }, { "name": "model.layers.16.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 21245184 }, { "name": "model.layers.16.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 23342336 }, { "name": "model.layers.16.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 23604480 }, { "name": "model.layers.16.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 24653056 }, { "name": "model.layers.16.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 24784128 }, { "name": "model.layers.17.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 24784384 }, { "name": "model.layers.17.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 24786432 }, { "name": "model.layers.17.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 26359296 }, { "name": "model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 26555904 }, { "name": "model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 29701632 }, { "name": "model.layers.17.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 30094848 }, { "name": "model.layers.17.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 30096896 }, { "name": "model.layers.17.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 30097152 }, { "name": "model.layers.17.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 32194304 }, { "name": "model.layers.17.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 32456448 } ], "md5sum": "2b38eca275cc615db7aee1629b3876bb" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 32000000, "records": [ { "name": "model.layers.17.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 0 }, { "name": "model.layers.17.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 131072 }, { "name": "model.layers.18.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 131328 }, { "name": "model.layers.18.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 133376 }, { "name": "model.layers.18.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 1706240 }, { "name": "model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 1902848 }, { "name": "model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 5048576 }, { "name": "model.layers.18.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 5441792 }, { "name": "model.layers.18.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 5443840 }, { "name": "model.layers.18.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 5444096 }, { "name": "model.layers.18.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 7541248 }, { "name": "model.layers.18.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 7803392 }, { "name": "model.layers.18.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 8851968 }, { "name": "model.layers.18.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 8983040 }, { "name": "model.layers.19.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 8983296 }, { "name": "model.layers.19.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 8985344 }, { "name": "model.layers.19.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 10558208 }, { "name": "model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 10754816 }, { "name": "model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 13900544 }, { "name": "model.layers.19.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 14293760 }, { "name": "model.layers.19.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 14295808 }, { "name": "model.layers.19.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 14296064 }, { "name": "model.layers.19.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 16393216 }, { "name": "model.layers.19.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 16655360 }, { "name": "model.layers.19.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 17703936 }, { "name": "model.layers.19.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 17835008 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 17835264 }, { "name": "model.layers.2.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 17837312 }, { "name": "model.layers.2.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 19410176 }, { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 19606784 }, { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 22752512 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 23145728 }, { "name": "model.layers.2.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 23147776 }, { "name": "model.layers.2.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23148032 }, { "name": "model.layers.2.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 25245184 }, { "name": "model.layers.2.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 25507328 }, { "name": "model.layers.2.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 26555904 }, { "name": "model.layers.2.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 26686976 }, { "name": "model.layers.20.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 26687232 }, { "name": "model.layers.20.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 26689280 }, { "name": "model.layers.20.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 28262144 }, { "name": "model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28458752 }, { "name": "model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 31604480 }, { "name": "model.layers.20.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 31997696 }, { "name": "model.layers.20.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 31999744 } ], "md5sum": "cc62c647380104add520de048e60e31e" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 31866624, "records": [ { "name": "model.layers.20.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.20.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.20.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 2359296 }, { "name": "model.layers.20.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 3407872 }, { "name": "model.layers.20.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 3538944 }, { "name": "model.layers.21.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 3539200 }, { "name": "model.layers.21.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 3541248 }, { "name": "model.layers.21.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 5114112 }, { "name": "model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 5310720 }, { "name": "model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 8456448 }, { "name": "model.layers.21.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 8849664 }, { "name": "model.layers.21.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 8851712 }, { "name": "model.layers.21.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 8851968 }, { "name": "model.layers.21.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 10949120 }, { "name": "model.layers.21.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 11211264 }, { "name": "model.layers.21.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 12259840 }, { "name": "model.layers.21.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 12390912 }, { "name": "model.layers.22.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 12391168 }, { "name": "model.layers.22.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12393216 }, { "name": "model.layers.22.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 13966080 }, { "name": "model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14162688 }, { "name": "model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 17308416 }, { "name": "model.layers.22.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 17701632 }, { "name": "model.layers.22.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 17703680 }, { "name": "model.layers.22.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17703936 }, { "name": "model.layers.22.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 19801088 }, { "name": "model.layers.22.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 20063232 }, { "name": "model.layers.22.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 21111808 }, { "name": "model.layers.22.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 21242880 }, { "name": "model.layers.23.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 21243136 }, { "name": "model.layers.23.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 21245184 }, { "name": "model.layers.23.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 22818048 }, { "name": "model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 23014656 }, { "name": "model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 26160384 }, { "name": "model.layers.23.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 26553600 }, { "name": "model.layers.23.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 26555648 }, { "name": "model.layers.23.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26555904 }, { "name": "model.layers.23.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28653056 }, { "name": "model.layers.23.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 28915200 }, { "name": "model.layers.23.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 29963776 }, { "name": "model.layers.23.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 30094848 }, { "name": "model.layers.24.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 30095104 }, { "name": "model.layers.24.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 30097152 }, { "name": "model.layers.24.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 31670016 } ], "md5sum": "f0b34ca86ed9b9277191deb0e9791807" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 33505024, "records": [ { "name": "model.layers.24.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 0 }, { "name": "model.layers.24.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 3145728 }, { "name": "model.layers.24.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 3538944 }, { "name": "model.layers.24.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 3540992 }, { "name": "model.layers.24.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 3541248 }, { "name": "model.layers.24.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 5638400 }, { "name": "model.layers.24.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 5900544 }, { "name": "model.layers.24.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 6949120 }, { "name": "model.layers.24.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 7080192 }, { "name": "model.layers.25.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 7080448 }, { "name": "model.layers.25.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 7082496 }, { "name": "model.layers.25.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 8655360 }, { "name": "model.layers.25.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 8851968 }, { "name": "model.layers.25.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 11997696 }, { "name": "model.layers.25.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 12390912 }, { "name": "model.layers.25.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 12392960 }, { "name": "model.layers.25.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 12393216 }, { "name": "model.layers.25.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 14490368 }, { "name": "model.layers.25.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 14752512 }, { "name": "model.layers.25.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 15801088 }, { "name": "model.layers.25.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 15932160 }, { "name": "model.layers.26.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 15932416 }, { "name": "model.layers.26.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 15934464 }, { "name": "model.layers.26.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 17507328 }, { "name": "model.layers.26.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 17703936 }, { "name": "model.layers.26.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 20849664 }, { "name": "model.layers.26.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 21242880 }, { "name": "model.layers.26.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 21244928 }, { "name": "model.layers.26.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 21245184 }, { "name": "model.layers.26.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 23342336 }, { "name": "model.layers.26.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 23604480 }, { "name": "model.layers.26.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 24653056 }, { "name": "model.layers.26.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 24784128 }, { "name": "model.layers.27.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 24784384 }, { "name": "model.layers.27.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 24786432 }, { "name": "model.layers.27.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 26359296 }, { "name": "model.layers.27.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 26555904 }, { "name": "model.layers.27.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 29701632 }, { "name": "model.layers.27.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 30094848 }, { "name": "model.layers.27.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 30096896 }, { "name": "model.layers.27.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 30097152 }, { "name": "model.layers.27.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 32194304 }, { "name": "model.layers.27.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 32456448 } ], "md5sum": "5e0bac7e7bcfd430c0a8ff0bc583c029" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 32000000, "records": [ { "name": "model.layers.27.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 0 }, { "name": "model.layers.27.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 131072 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 131328 }, { "name": "model.layers.3.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 133376 }, { "name": "model.layers.3.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 1706240 }, { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 1902848 }, { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 5048576 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 5441792 }, { "name": "model.layers.3.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 5443840 }, { "name": "model.layers.3.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 5444096 }, { "name": "model.layers.3.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 7541248 }, { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 7803392 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 8851968 }, { "name": "model.layers.3.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 8983040 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 8983296 }, { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 8985344 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 10558208 }, { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 10754816 }, { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 13900544 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 14293760 }, { "name": "model.layers.4.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 14295808 }, { "name": "model.layers.4.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 14296064 }, { "name": "model.layers.4.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 16393216 }, { "name": "model.layers.4.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 16655360 }, { "name": "model.layers.4.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 17703936 }, { "name": "model.layers.4.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 17835008 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 17835264 }, { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 17837312 }, { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 19410176 }, { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 19606784 }, { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 22752512 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 23145728 }, { "name": "model.layers.5.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 23147776 }, { "name": "model.layers.5.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 23148032 }, { "name": "model.layers.5.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 25245184 }, { "name": "model.layers.5.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 25507328 }, { "name": "model.layers.5.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 26555904 }, { "name": "model.layers.5.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 26686976 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 26687232 }, { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 26689280 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 28262144 }, { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 28458752 }, { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 31604480 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 31997696 }, { "name": "model.layers.6.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 31999744 } ], "md5sum": "3e5c9c2a8afd4eb9c753b6ebec57bc5b" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 30097152, "records": [ { "name": "model.layers.6.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 0 }, { "name": "model.layers.6.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 2097152 }, { "name": "model.layers.6.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 2359296 }, { "name": "model.layers.6.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 3407872 }, { "name": "model.layers.6.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 3538944 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 3539200 }, { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 3541248 }, { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 5114112 }, { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 5310720 }, { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 8456448 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 8849664 }, { "name": "model.layers.7.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 8851712 }, { "name": "model.layers.7.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 8851968 }, { "name": "model.layers.7.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 10949120 }, { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 11211264 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 12259840 }, { "name": "model.layers.7.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 12390912 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 12391168 }, { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 12393216 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 13966080 }, { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 14162688 }, { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 17308416 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 17701632 }, { "name": "model.layers.8.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 17703680 }, { "name": "model.layers.8.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 17703936 }, { "name": "model.layers.8.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 19801088 }, { "name": "model.layers.8.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 20063232 }, { "name": "model.layers.8.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 21111808 }, { "name": "model.layers.8.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 21242880 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 21243136 }, { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 1024, 384 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 21245184 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 1024, 96 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 196608, "byteOffset": 22818048 }, { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 6144, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 3145728, "byteOffset": 23014656 }, { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 6144, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 393216, "byteOffset": 26160384 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 26553600 }, { "name": "model.layers.9.self_attn.k_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 26555648 }, { "name": "model.layers.9.self_attn.c_attn.q_weight", "shape": [ 4096, 128 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 26555904 }, { "name": "model.layers.9.self_attn.c_attn.q_scale", "shape": [ 4096, 32 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 28653056 }, { "name": "model.layers.9.self_attn.o_proj.q_weight", "shape": [ 1024, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1048576, "byteOffset": 28915200 }, { "name": "model.layers.9.self_attn.o_proj.q_scale", "shape": [ 1024, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 131072, "byteOffset": 29963776 }, { "name": "model.layers.9.self_attn.q_norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 30094848 }, { "name": "model.norm.weight", "shape": [ 1024 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 30095104 } ], "md5sum": "15e502db700855b8924d7aab98c21d4a" } ] }