{ "metadata": { "ParamSize": 435, "ParamBytes": 1059316480.0, "BitsPerParam": 4.5033578806236 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 254279680, "records": [ { "name": "model.embed_tokens.q_weight", "shape": [ 248320, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 254279680, "byteOffset": 0 } ], "md5sum": "ced008d1ffa7c2a1eef6ee98bcc6bf79" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 31875680, "records": [ { "name": "model.layers.0.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 0 }, { "name": "model.layers.1.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 32 }, { "name": "model.layers.10.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 64 }, { "name": "model.layers.12.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 96 }, { "name": "model.layers.13.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 128 }, { "name": "model.layers.14.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 160 }, { "name": "model.layers.16.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 192 }, { "name": "model.layers.17.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 224 }, { "name": "model.layers.18.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 256 }, { "name": "model.layers.2.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 288 }, { "name": "model.layers.20.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 320 }, { "name": "model.layers.21.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 352 }, { "name": "model.layers.22.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 384 }, { "name": "model.layers.4.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 416 }, { "name": "model.layers.5.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 448 }, { "name": "model.layers.6.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 480 }, { "name": "model.layers.8.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 512 }, { "name": "model.layers.9.linear_attn.A_log", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 544 }, { "name": "model.embed_tokens.q_scale", "shape": [ 248320, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 31784960, "byteOffset": 576 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 31785536 }, { "name": "model.layers.0.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 31789632 }, { "name": "model.layers.0.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 31838784 }, { "name": "model.layers.0.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 31838816 }, { "name": "model.layers.0.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 31855200 }, { "name": "model.layers.0.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 31857248 }, { "name": "model.layers.0.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 31873632 } ], "md5sum": "0bdce1b4b869453a8142fc752bb7be7f" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.0.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.0.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.0.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.0.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.0.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.0.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.0.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.1.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.1.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.1.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.1.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.1.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.1.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "09b9683b2236355373ee72257d99cec9" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.1.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.1.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.1.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.1.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.1.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.1.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.1.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.1.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.1.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.10.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.10.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.10.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.10.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.10.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.10.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "296162c3cbe3bf1fc0624fef116729b3" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 33038592, "records": [ { "name": "model.layers.10.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.10.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.10.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.10.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.10.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.10.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.10.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 } ], "md5sum": "4c57b023961510c442e51341be49ee3c" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 29586464, "records": [ { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 7077888 }, { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 19660800 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 21233664 }, { "name": "model.layers.11.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 21237760 }, { "name": "model.layers.11.self_attn.c_attn.q_weight", "shape": [ 5120, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 21238272 }, { "name": "model.layers.11.self_attn.c_attn.q_scale", "shape": [ 5120, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 655360, "byteOffset": 26481152 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 27136512 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 29233664 }, { "name": "model.layers.11.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29495808 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 29496320 }, { "name": "model.layers.12.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 29500416 }, { "name": "model.layers.12.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 29549568 }, { "name": "model.layers.12.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29549600 }, { "name": "model.layers.12.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29565984 }, { "name": "model.layers.12.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29568032 }, { "name": "model.layers.12.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29584416 } ], "md5sum": "681b6f945c6a71f6496b0bd05daa3b65" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.12.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.12.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.12.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.12.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.12.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.12.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.12.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.13.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.13.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.13.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.13.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.13.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.13.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "0eddc23be842b46956f0b9d639d01018" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.13.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.13.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.13.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.13.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.13.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.13.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.13.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.14.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.14.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.14.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.14.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.14.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.14.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "190eec3af426e28d7a9749e56b389487" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 33038592, "records": [ { "name": "model.layers.14.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.14.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.14.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.14.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.14.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.14.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.14.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 } ], "md5sum": "f10e1905c4eec3362985648810373b5a" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 29586464, "records": [ { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 7077888 }, { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 19660800 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 21233664 }, { "name": "model.layers.15.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 21237760 }, { "name": "model.layers.15.self_attn.c_attn.q_weight", "shape": [ 5120, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 21238272 }, { "name": "model.layers.15.self_attn.c_attn.q_scale", "shape": [ 5120, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 655360, "byteOffset": 26481152 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 27136512 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 29233664 }, { "name": "model.layers.15.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29495808 }, { "name": "model.layers.16.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 29496320 }, { "name": "model.layers.16.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 29500416 }, { "name": "model.layers.16.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 29549568 }, { "name": "model.layers.16.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29549600 }, { "name": "model.layers.16.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29565984 }, { "name": "model.layers.16.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29568032 }, { "name": "model.layers.16.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29584416 } ], "md5sum": "30df5e96f8d596fca9a8c896e4738db3" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.16.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.16.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.16.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.16.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.16.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.16.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.16.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.16.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.16.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.16.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.17.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.17.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.17.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.17.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.17.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.17.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.17.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "edaea7c758549e85870fa34dfb17501a" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.17.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.17.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.17.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.17.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.17.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.17.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.17.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.17.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.17.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.17.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.18.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.18.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.18.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.18.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.18.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.18.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.18.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "6fb9dbd845045d658ce3e037d37c5e7f" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 33038592, "records": [ { "name": "model.layers.18.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.18.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.18.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.18.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.18.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.18.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.18.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.18.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.18.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.18.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.19.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 } ], "md5sum": "d20122400f192883554a6291a9ff8bf1" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 29586464, "records": [ { "name": "model.layers.19.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.19.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 7077888 }, { "name": "model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 19660800 }, { "name": "model.layers.19.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 21233664 }, { "name": "model.layers.19.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 21237760 }, { "name": "model.layers.19.self_attn.c_attn.q_weight", "shape": [ 5120, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 21238272 }, { "name": "model.layers.19.self_attn.c_attn.q_scale", "shape": [ 5120, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 655360, "byteOffset": 26481152 }, { "name": "model.layers.19.self_attn.o_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 27136512 }, { "name": "model.layers.19.self_attn.o_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 29233664 }, { "name": "model.layers.19.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29495808 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 29496320 }, { "name": "model.layers.2.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 29500416 }, { "name": "model.layers.2.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 29549568 }, { "name": "model.layers.2.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29549600 }, { "name": "model.layers.2.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29565984 }, { "name": "model.layers.2.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29568032 }, { "name": "model.layers.2.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29584416 } ], "md5sum": "697f6246252dcbcf22d9ab291c87ac0a" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.2.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.2.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.2.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.2.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.2.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.2.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.2.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.2.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.2.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.20.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.20.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.20.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.20.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.20.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.20.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.20.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "1c7a4814c21593fe9bbc5b7d2b34764a" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.20.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.20.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.20.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.20.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.20.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.20.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.20.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.20.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.20.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.20.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.21.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.21.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.21.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.21.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.21.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.21.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.21.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "9a9ec269064628f3831f1da1d8e24e9f" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.21.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.21.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.21.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.21.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.21.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.21.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.21.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.21.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.21.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.21.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.22.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.22.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.22.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.22.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.22.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.22.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.22.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "61f932b8413ec1513bd1589d3da302d0" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 33038592, "records": [ { "name": "model.layers.22.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.22.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.22.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.22.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.22.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.22.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.22.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.22.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.22.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.22.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.23.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 } ], "md5sum": "b6649f7fe9f23dd31de99238a2a004f8" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 29500416, "records": [ { "name": "model.layers.23.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.23.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 7077888 }, { "name": "model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 19660800 }, { "name": "model.layers.23.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 21233664 }, { "name": "model.layers.23.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 21237760 }, { "name": "model.layers.23.self_attn.c_attn.q_weight", "shape": [ 5120, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 21238272 }, { "name": "model.layers.23.self_attn.c_attn.q_scale", "shape": [ 5120, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 655360, "byteOffset": 26481152 }, { "name": "model.layers.23.self_attn.o_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 27136512 }, { "name": "model.layers.23.self_attn.o_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 29233664 }, { "name": "model.layers.23.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29495808 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 29496320 } ], "md5sum": "b1a705ef2ec5af107f058e1cab431b62" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 29586464, "records": [ { "name": "model.layers.3.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.3.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 7077888 }, { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 19660800 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 21233664 }, { "name": "model.layers.3.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 21237760 }, { "name": "model.layers.3.self_attn.c_attn.q_weight", "shape": [ 5120, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 21238272 }, { "name": "model.layers.3.self_attn.c_attn.q_scale", "shape": [ 5120, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 655360, "byteOffset": 26481152 }, { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 27136512 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 29233664 }, { "name": "model.layers.3.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29495808 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 29496320 }, { "name": "model.layers.4.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 29500416 }, { "name": "model.layers.4.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 29549568 }, { "name": "model.layers.4.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29549600 }, { "name": "model.layers.4.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29565984 }, { "name": "model.layers.4.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29568032 }, { "name": "model.layers.4.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29584416 } ], "md5sum": "84762f7c0f38ae037ba6b6c7d6760dab" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.4.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.4.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.4.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.4.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.4.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.4.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.4.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.5.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.5.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.5.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.5.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.5.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.5.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "581f76c0908b09c8244bce9014ef2430" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.5.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.5.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.5.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.5.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.5.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.5.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.5.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.6.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.6.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.6.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.6.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.6.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.6.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "5c16586b69c5ac7c5a38f0ce3952fd85" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 33038592, "records": [ { "name": "model.layers.6.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.6.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.6.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.6.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.6.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.6.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.6.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 } ], "md5sum": "f98a1cbd17a12be7a5eb114a338098e2" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 29586464, "records": [ { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 7077888 }, { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 19660800 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 21233664 }, { "name": "model.layers.7.self_attn.k_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 21237760 }, { "name": "model.layers.7.self_attn.c_attn.q_weight", "shape": [ 5120, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 5242880, "byteOffset": 21238272 }, { "name": "model.layers.7.self_attn.c_attn.q_scale", "shape": [ 5120, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 655360, "byteOffset": 26481152 }, { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 27136512 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 29233664 }, { "name": "model.layers.7.self_attn.q_norm.weight", "shape": [ 256 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 512, "byteOffset": 29495808 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 29496320 }, { "name": "model.layers.8.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 29500416 }, { "name": "model.layers.8.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 29549568 }, { "name": "model.layers.8.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29549600 }, { "name": "model.layers.8.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29565984 }, { "name": "model.layers.8.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 29568032 }, { "name": "model.layers.8.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 29584416 } ], "md5sum": "847caaca475c44d6cb8c571bafb9843b" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 33124640, "records": [ { "name": "model.layers.8.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.8.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.8.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.8.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.8.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.8.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.8.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 }, { "name": "model.layers.9.linear_attn.conv1d_weight", "shape": [ 6144, 1, 4 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 49152, "byteOffset": 33038592 }, { "name": "model.layers.9.linear_attn.dt_bias", "shape": [ 16 ], "dtype": "bfloat16", "format": "raw", "nbytes": 32, "byteOffset": 33087744 }, { "name": "model.layers.9.linear_attn.in_proj_a.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33087776 }, { "name": "model.layers.9.linear_attn.in_proj_a.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33104160 }, { "name": "model.layers.9.linear_attn.in_proj_b.q_weight", "shape": [ 16, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 16384, "byteOffset": 33106208 }, { "name": "model.layers.9.linear_attn.in_proj_b.q_scale", "shape": [ 16, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 2048, "byteOffset": 33122592 } ], "md5sum": "4abcfafe2b2fab328feb4907da72184e" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 33038592, "records": [ { "name": "model.layers.9.linear_attn.in_proj_qkv.q_weight", "shape": [ 6144, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 0 }, { "name": "model.layers.9.linear_attn.in_proj_qkv.q_scale", "shape": [ 6144, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 6291456 }, { "name": "model.layers.9.linear_attn.in_proj_z.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 7077888 }, { "name": "model.layers.9.linear_attn.in_proj_z.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 9175040 }, { "name": "model.layers.9.linear_attn.norm.weight", "shape": [ 128 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 256, "byteOffset": 9437184 }, { "name": "model.layers.9.linear_attn.out_proj.q_weight", "shape": [ 2048, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 2097152, "byteOffset": 9437440 }, { "name": "model.layers.9.linear_attn.out_proj.q_scale", "shape": [ 2048, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 262144, "byteOffset": 11534592 }, { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 2048, 768 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 6291456, "byteOffset": 11796736 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 2048, 192 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 786432, "byteOffset": 18088192 }, { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 12288, 256 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 12582912, "byteOffset": 18874624 }, { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 12288, 64 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1572864, "byteOffset": 31457536 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33030400 }, { "name": "model.norm.weight", "shape": [ 2048 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 4096, "byteOffset": 33034496 } ], "md5sum": "97d9b02452f397b1c307caaf54feb546" } ] }