{ "schemaVersion": 1, "modelFamily": "ming_3_tts_tiny_language", "alignment": 16384, "file": "weights.bin", "quantizedLinears": { "model.model.layers.1.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.1.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.1.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.10.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.10.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.10.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.11.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.11.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.11.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.12.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.12.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.12.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.13.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.13.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.13.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.14.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.14.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.14.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.15.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.15.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.15.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.16.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.16.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.16.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.17.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.17.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.17.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.18.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.18.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.18.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.19.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.19.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.19.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.2.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.2.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.2.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.20.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.20.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.20.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.21.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.21.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.21.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.22.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.22.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.22.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.23.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.23.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.23.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.3.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.3.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.3.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.4.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.4.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.4.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.5.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.5.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.5.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.6.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.6.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.6.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.7.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.7.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.7.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.8.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.8.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.8.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.9.mlp.experts.down_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.9.mlp.experts.gate_proj": { "bits": 4, "groupSize": 64 }, "model.model.layers.9.mlp.experts.up_proj": { "bits": 4, "groupSize": 64 } }, "tensors": { "model.model.layers.1.mlp.experts.down_proj.weight": { "offset": 0, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.1.mlp.experts.down_proj.scales": { "offset": 50331648, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.1.mlp.experts.down_proj.biases": { "offset": 56623104, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.1.mlp.experts.gate_proj.weight": { "offset": 62914560, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.1.mlp.experts.gate_proj.scales": { "offset": 113246208, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.1.mlp.experts.gate_proj.biases": { "offset": 119537664, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.1.mlp.experts.up_proj.weight": { "offset": 125829120, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.1.mlp.experts.up_proj.scales": { "offset": 176160768, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.1.mlp.experts.up_proj.biases": { "offset": 182452224, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.10.mlp.experts.down_proj.weight": { "offset": 188743680, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.10.mlp.experts.down_proj.scales": { "offset": 239075328, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.10.mlp.experts.down_proj.biases": { "offset": 245366784, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.10.mlp.experts.gate_proj.weight": { "offset": 251658240, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.10.mlp.experts.gate_proj.scales": { "offset": 301989888, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.10.mlp.experts.gate_proj.biases": { "offset": 308281344, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.10.mlp.experts.up_proj.weight": { "offset": 314572800, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.10.mlp.experts.up_proj.scales": { "offset": 364904448, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.10.mlp.experts.up_proj.biases": { "offset": 371195904, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.11.mlp.experts.down_proj.weight": { "offset": 377487360, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.11.mlp.experts.down_proj.scales": { "offset": 427819008, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.11.mlp.experts.down_proj.biases": { "offset": 434110464, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.11.mlp.experts.gate_proj.weight": { "offset": 440401920, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.11.mlp.experts.gate_proj.scales": { "offset": 490733568, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.11.mlp.experts.gate_proj.biases": { "offset": 497025024, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.11.mlp.experts.up_proj.weight": { "offset": 503316480, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.11.mlp.experts.up_proj.scales": { "offset": 553648128, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.11.mlp.experts.up_proj.biases": { "offset": 559939584, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.12.mlp.experts.down_proj.weight": { "offset": 566231040, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.12.mlp.experts.down_proj.scales": { "offset": 616562688, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.12.mlp.experts.down_proj.biases": { "offset": 622854144, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.12.mlp.experts.gate_proj.weight": { "offset": 629145600, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.12.mlp.experts.gate_proj.scales": { "offset": 679477248, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.12.mlp.experts.gate_proj.biases": { "offset": 685768704, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.12.mlp.experts.up_proj.weight": { "offset": 692060160, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.12.mlp.experts.up_proj.scales": { "offset": 742391808, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.12.mlp.experts.up_proj.biases": { "offset": 748683264, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.13.mlp.experts.down_proj.weight": { "offset": 754974720, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.13.mlp.experts.down_proj.scales": { "offset": 805306368, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.13.mlp.experts.down_proj.biases": { "offset": 811597824, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.13.mlp.experts.gate_proj.weight": { "offset": 817889280, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.13.mlp.experts.gate_proj.scales": { "offset": 868220928, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.13.mlp.experts.gate_proj.biases": { "offset": 874512384, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.13.mlp.experts.up_proj.weight": { "offset": 880803840, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.13.mlp.experts.up_proj.scales": { "offset": 931135488, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.13.mlp.experts.up_proj.biases": { "offset": 937426944, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.14.mlp.experts.down_proj.weight": { "offset": 943718400, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.14.mlp.experts.down_proj.scales": { "offset": 994050048, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.14.mlp.experts.down_proj.biases": { "offset": 1000341504, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.14.mlp.experts.gate_proj.weight": { "offset": 1006632960, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.14.mlp.experts.gate_proj.scales": { "offset": 1056964608, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.14.mlp.experts.gate_proj.biases": { "offset": 1063256064, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.14.mlp.experts.up_proj.weight": { "offset": 1069547520, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.14.mlp.experts.up_proj.scales": { "offset": 1119879168, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.14.mlp.experts.up_proj.biases": { "offset": 1126170624, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.15.mlp.experts.down_proj.weight": { "offset": 1132462080, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.15.mlp.experts.down_proj.scales": { "offset": 1182793728, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.15.mlp.experts.down_proj.biases": { "offset": 1189085184, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.15.mlp.experts.gate_proj.weight": { "offset": 1195376640, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.15.mlp.experts.gate_proj.scales": { "offset": 1245708288, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.15.mlp.experts.gate_proj.biases": { "offset": 1251999744, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.15.mlp.experts.up_proj.weight": { "offset": 1258291200, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.15.mlp.experts.up_proj.scales": { "offset": 1308622848, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.15.mlp.experts.up_proj.biases": { "offset": 1314914304, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.16.mlp.experts.down_proj.weight": { "offset": 1321205760, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.16.mlp.experts.down_proj.scales": { "offset": 1371537408, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.16.mlp.experts.down_proj.biases": { "offset": 1377828864, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.16.mlp.experts.gate_proj.weight": { "offset": 1384120320, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.16.mlp.experts.gate_proj.scales": { "offset": 1434451968, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.16.mlp.experts.gate_proj.biases": { "offset": 1440743424, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.16.mlp.experts.up_proj.weight": { "offset": 1447034880, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.16.mlp.experts.up_proj.scales": { "offset": 1497366528, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.16.mlp.experts.up_proj.biases": { "offset": 1503657984, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.17.mlp.experts.down_proj.weight": { "offset": 1509949440, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.17.mlp.experts.down_proj.scales": { "offset": 1560281088, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.17.mlp.experts.down_proj.biases": { "offset": 1566572544, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.17.mlp.experts.gate_proj.weight": { "offset": 1572864000, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.17.mlp.experts.gate_proj.scales": { "offset": 1623195648, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.17.mlp.experts.gate_proj.biases": { "offset": 1629487104, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.17.mlp.experts.up_proj.weight": { "offset": 1635778560, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.17.mlp.experts.up_proj.scales": { "offset": 1686110208, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.17.mlp.experts.up_proj.biases": { "offset": 1692401664, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.18.mlp.experts.down_proj.weight": { "offset": 1698693120, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.18.mlp.experts.down_proj.scales": { "offset": 1749024768, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.18.mlp.experts.down_proj.biases": { "offset": 1755316224, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.18.mlp.experts.gate_proj.weight": { "offset": 1761607680, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.18.mlp.experts.gate_proj.scales": { "offset": 1811939328, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.18.mlp.experts.gate_proj.biases": { "offset": 1818230784, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.18.mlp.experts.up_proj.weight": { "offset": 1824522240, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.18.mlp.experts.up_proj.scales": { "offset": 1874853888, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.18.mlp.experts.up_proj.biases": { "offset": 1881145344, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.19.mlp.experts.down_proj.weight": { "offset": 1887436800, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.19.mlp.experts.down_proj.scales": { "offset": 1937768448, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.19.mlp.experts.down_proj.biases": { "offset": 1944059904, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.19.mlp.experts.gate_proj.weight": { "offset": 1950351360, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.19.mlp.experts.gate_proj.scales": { "offset": 2000683008, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.19.mlp.experts.gate_proj.biases": { "offset": 2006974464, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.19.mlp.experts.up_proj.weight": { "offset": 2013265920, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.19.mlp.experts.up_proj.scales": { "offset": 2063597568, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.19.mlp.experts.up_proj.biases": { "offset": 2069889024, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.2.mlp.experts.down_proj.weight": { "offset": 2076180480, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.2.mlp.experts.down_proj.scales": { "offset": 2126512128, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.2.mlp.experts.down_proj.biases": { "offset": 2132803584, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.2.mlp.experts.gate_proj.weight": { "offset": 2139095040, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.2.mlp.experts.gate_proj.scales": { "offset": 2189426688, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.2.mlp.experts.gate_proj.biases": { "offset": 2195718144, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.2.mlp.experts.up_proj.weight": { "offset": 2202009600, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.2.mlp.experts.up_proj.scales": { "offset": 2252341248, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.2.mlp.experts.up_proj.biases": { "offset": 2258632704, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.20.mlp.experts.down_proj.weight": { "offset": 2264924160, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.20.mlp.experts.down_proj.scales": { "offset": 2315255808, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.20.mlp.experts.down_proj.biases": { "offset": 2321547264, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.20.mlp.experts.gate_proj.weight": { "offset": 2327838720, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.20.mlp.experts.gate_proj.scales": { "offset": 2378170368, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.20.mlp.experts.gate_proj.biases": { "offset": 2384461824, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.20.mlp.experts.up_proj.weight": { "offset": 2390753280, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.20.mlp.experts.up_proj.scales": { "offset": 2441084928, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.20.mlp.experts.up_proj.biases": { "offset": 2447376384, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.21.mlp.experts.down_proj.weight": { "offset": 2453667840, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.21.mlp.experts.down_proj.scales": { "offset": 2503999488, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.21.mlp.experts.down_proj.biases": { "offset": 2510290944, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.21.mlp.experts.gate_proj.weight": { "offset": 2516582400, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.21.mlp.experts.gate_proj.scales": { "offset": 2566914048, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.21.mlp.experts.gate_proj.biases": { "offset": 2573205504, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.21.mlp.experts.up_proj.weight": { "offset": 2579496960, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.21.mlp.experts.up_proj.scales": { "offset": 2629828608, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.21.mlp.experts.up_proj.biases": { "offset": 2636120064, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.22.mlp.experts.down_proj.weight": { "offset": 2642411520, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.22.mlp.experts.down_proj.scales": { "offset": 2692743168, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.22.mlp.experts.down_proj.biases": { "offset": 2699034624, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.22.mlp.experts.gate_proj.weight": { "offset": 2705326080, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.22.mlp.experts.gate_proj.scales": { "offset": 2755657728, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.22.mlp.experts.gate_proj.biases": { "offset": 2761949184, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.22.mlp.experts.up_proj.weight": { "offset": 2768240640, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.22.mlp.experts.up_proj.scales": { "offset": 2818572288, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.22.mlp.experts.up_proj.biases": { "offset": 2824863744, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.23.mlp.experts.down_proj.weight": { "offset": 2831155200, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.23.mlp.experts.down_proj.scales": { "offset": 2881486848, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.23.mlp.experts.down_proj.biases": { "offset": 2887778304, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.23.mlp.experts.gate_proj.weight": { "offset": 2894069760, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.23.mlp.experts.gate_proj.scales": { "offset": 2944401408, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.23.mlp.experts.gate_proj.biases": { "offset": 2950692864, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.23.mlp.experts.up_proj.weight": { "offset": 2956984320, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.23.mlp.experts.up_proj.scales": { "offset": 3007315968, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.23.mlp.experts.up_proj.biases": { "offset": 3013607424, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.3.mlp.experts.down_proj.weight": { "offset": 3019898880, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.3.mlp.experts.down_proj.scales": { "offset": 3070230528, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.3.mlp.experts.down_proj.biases": { "offset": 3076521984, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.3.mlp.experts.gate_proj.weight": { "offset": 3082813440, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.3.mlp.experts.gate_proj.scales": { "offset": 3133145088, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.3.mlp.experts.gate_proj.biases": { "offset": 3139436544, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.3.mlp.experts.up_proj.weight": { "offset": 3145728000, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.3.mlp.experts.up_proj.scales": { "offset": 3196059648, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.3.mlp.experts.up_proj.biases": { "offset": 3202351104, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.4.mlp.experts.down_proj.weight": { "offset": 3208642560, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.4.mlp.experts.down_proj.scales": { "offset": 3258974208, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.4.mlp.experts.down_proj.biases": { "offset": 3265265664, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.4.mlp.experts.gate_proj.weight": { "offset": 3271557120, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.4.mlp.experts.gate_proj.scales": { "offset": 3321888768, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.4.mlp.experts.gate_proj.biases": { "offset": 3328180224, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.4.mlp.experts.up_proj.weight": { "offset": 3334471680, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.4.mlp.experts.up_proj.scales": { "offset": 3384803328, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.4.mlp.experts.up_proj.biases": { "offset": 3391094784, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.5.mlp.experts.down_proj.weight": { "offset": 3397386240, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.5.mlp.experts.down_proj.scales": { "offset": 3447717888, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.5.mlp.experts.down_proj.biases": { "offset": 3454009344, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.5.mlp.experts.gate_proj.weight": { "offset": 3460300800, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.5.mlp.experts.gate_proj.scales": { "offset": 3510632448, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.5.mlp.experts.gate_proj.biases": { "offset": 3516923904, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.5.mlp.experts.up_proj.weight": { "offset": 3523215360, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.5.mlp.experts.up_proj.scales": { "offset": 3573547008, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.5.mlp.experts.up_proj.biases": { "offset": 3579838464, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.6.mlp.experts.down_proj.weight": { "offset": 3586129920, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.6.mlp.experts.down_proj.scales": { "offset": 3636461568, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.6.mlp.experts.down_proj.biases": { "offset": 3642753024, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.6.mlp.experts.gate_proj.weight": { "offset": 3649044480, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.6.mlp.experts.gate_proj.scales": { "offset": 3699376128, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.6.mlp.experts.gate_proj.biases": { "offset": 3705667584, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.6.mlp.experts.up_proj.weight": { "offset": 3711959040, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.6.mlp.experts.up_proj.scales": { "offset": 3762290688, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.6.mlp.experts.up_proj.biases": { "offset": 3768582144, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.7.mlp.experts.down_proj.weight": { "offset": 3774873600, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.7.mlp.experts.down_proj.scales": { "offset": 3825205248, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.7.mlp.experts.down_proj.biases": { "offset": 3831496704, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.7.mlp.experts.gate_proj.weight": { "offset": 3837788160, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.7.mlp.experts.gate_proj.scales": { "offset": 3888119808, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.7.mlp.experts.gate_proj.biases": { "offset": 3894411264, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.7.mlp.experts.up_proj.weight": { "offset": 3900702720, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.7.mlp.experts.up_proj.scales": { "offset": 3951034368, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.7.mlp.experts.up_proj.biases": { "offset": 3957325824, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.8.mlp.experts.down_proj.weight": { "offset": 3963617280, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.8.mlp.experts.down_proj.scales": { "offset": 4013948928, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.8.mlp.experts.down_proj.biases": { "offset": 4020240384, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.8.mlp.experts.gate_proj.weight": { "offset": 4026531840, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.8.mlp.experts.gate_proj.scales": { "offset": 4076863488, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.8.mlp.experts.gate_proj.biases": { "offset": 4083154944, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.8.mlp.experts.up_proj.weight": { "offset": 4089446400, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.8.mlp.experts.up_proj.scales": { "offset": 4139778048, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.8.mlp.experts.up_proj.biases": { "offset": 4146069504, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.9.mlp.experts.down_proj.weight": { "offset": 4152360960, "bytes": 50331648, "shape": [ 128, 1536, 64 ], "dtype": "uint32" }, "model.model.layers.9.mlp.experts.down_proj.scales": { "offset": 4202692608, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.9.mlp.experts.down_proj.biases": { "offset": 4208984064, "bytes": 6291456, "shape": [ 128, 1536, 8 ], "dtype": "float32" }, "model.model.layers.9.mlp.experts.gate_proj.weight": { "offset": 4215275520, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.9.mlp.experts.gate_proj.scales": { "offset": 4265607168, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.9.mlp.experts.gate_proj.biases": { "offset": 4271898624, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.9.mlp.experts.up_proj.weight": { "offset": 4278190080, "bytes": 50331648, "shape": [ 128, 512, 192 ], "dtype": "uint32" }, "model.model.layers.9.mlp.experts.up_proj.scales": { "offset": 4328521728, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.9.mlp.experts.up_proj.biases": { "offset": 4334813184, "bytes": 6291456, "shape": [ 128, 512, 24 ], "dtype": "float32" }, "model.model.layers.0.attention.A_log": { "offset": 4341104640, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.0.attention.b_proj.weight": { "offset": 4341121024, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.dt_bias": { "offset": 4341170176, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.0.attention.f_proj.weight": { "offset": 4341186560, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.g_proj.weight": { "offset": 4347478016, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.k_conv1d.weight": { "offset": 4353769472, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.k_proj.weight": { "offset": 4353785856, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.o_norm.weight": { "offset": 4360077312, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.o_proj.weight": { "offset": 4360093696, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.q_conv1d.weight": { "offset": 4366385152, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.q_proj.weight": { "offset": 4366401536, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.v_conv1d.weight": { "offset": 4372692992, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.0.attention.v_proj.weight": { "offset": 4372709376, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.input_layernorm.weight": { "offset": 4379000832, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.mlp.down_proj.weight": { "offset": 4379017216, "bytes": 14155776, "shape": [ 1536, 4608 ], "dtype": "bfloat16" }, "model.model.layers.0.mlp.gate_proj.weight": { "offset": 4393172992, "bytes": 14155776, "shape": [ 4608, 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.mlp.up_proj.weight": { "offset": 4407328768, "bytes": 14155776, "shape": [ 4608, 1536 ], "dtype": "bfloat16" }, "model.model.layers.0.post_attention_layernorm.weight": { "offset": 4421484544, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.A_log": { "offset": 4421500928, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.1.attention.b_proj.weight": { "offset": 4421517312, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.dt_bias": { "offset": 4421566464, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.1.attention.f_proj.weight": { "offset": 4421582848, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.g_proj.weight": { "offset": 4427874304, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.k_conv1d.weight": { "offset": 4434165760, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.k_proj.weight": { "offset": 4434182144, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.o_norm.weight": { "offset": 4440473600, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.o_proj.weight": { "offset": 4440489984, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.q_conv1d.weight": { "offset": 4446781440, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.q_proj.weight": { "offset": 4446797824, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.v_conv1d.weight": { "offset": 4453089280, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.1.attention.v_proj.weight": { "offset": 4453105664, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.input_layernorm.weight": { "offset": 4459397120, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.mlp.gate.expert_bias": { "offset": 4459413504, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.1.mlp.gate.weight": { "offset": 4459429888, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.1.mlp.shared_experts.down_proj.weight": { "offset": 4460216320, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.1.mlp.shared_experts.gate_proj.weight": { "offset": 4461789184, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.mlp.shared_experts.up_proj.weight": { "offset": 4463362048, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.1.post_attention_layernorm.weight": { "offset": 4464934912, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.A_log": { "offset": 4464951296, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.10.attention.b_proj.weight": { "offset": 4464967680, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.dt_bias": { "offset": 4465016832, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.10.attention.f_proj.weight": { "offset": 4465033216, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.g_proj.weight": { "offset": 4471324672, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.k_conv1d.weight": { "offset": 4477616128, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.k_proj.weight": { "offset": 4477632512, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.o_norm.weight": { "offset": 4483923968, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.o_proj.weight": { "offset": 4483940352, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.q_conv1d.weight": { "offset": 4490231808, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.q_proj.weight": { "offset": 4490248192, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.v_conv1d.weight": { "offset": 4496539648, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.10.attention.v_proj.weight": { "offset": 4496556032, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.input_layernorm.weight": { "offset": 4502847488, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.mlp.gate.expert_bias": { "offset": 4502863872, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.10.mlp.gate.weight": { "offset": 4502880256, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.10.mlp.shared_experts.down_proj.weight": { "offset": 4503666688, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.10.mlp.shared_experts.gate_proj.weight": { "offset": 4505239552, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.mlp.shared_experts.up_proj.weight": { "offset": 4506812416, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.10.post_attention_layernorm.weight": { "offset": 4508385280, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.11.attention.dense.weight": { "offset": 4508401664, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.11.attention.g_proj.weight": { "offset": 4514693120, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.11.attention.kv_a_layernorm.weight": { "offset": 4514742272, "bytes": 1024, "shape": [ 512 ], "dtype": "bfloat16" }, "model.model.layers.11.attention.kv_a_proj_with_mqa.weight": { "offset": 4514758656, "bytes": 1769472, "shape": [ 576, 1536 ], "dtype": "bfloat16" }, "model.model.layers.11.attention.kv_b_proj.weight": { "offset": 4516528128, "bytes": 4194304, "shape": [ 4096, 512 ], "dtype": "bfloat16" }, "model.model.layers.11.attention.q_a_layernorm.weight": { "offset": 4520722432, "bytes": 512, "shape": [ 256 ], "dtype": "bfloat16" }, "model.model.layers.11.attention.q_a_proj.weight": { "offset": 4520738816, "bytes": 786432, "shape": [ 256, 1536 ], "dtype": "bfloat16" }, "model.model.layers.11.attention.q_b_proj.weight": { "offset": 4521525248, "bytes": 1572864, "shape": [ 3072, 256 ], "dtype": "bfloat16" }, "model.model.layers.11.input_layernorm.weight": { "offset": 4523098112, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.11.mlp.gate.expert_bias": { "offset": 4523114496, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.11.mlp.gate.weight": { "offset": 4523130880, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.11.mlp.shared_experts.down_proj.weight": { "offset": 4523917312, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.11.mlp.shared_experts.gate_proj.weight": { "offset": 4525490176, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.11.mlp.shared_experts.up_proj.weight": { "offset": 4527063040, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.11.post_attention_layernorm.weight": { "offset": 4528635904, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.A_log": { "offset": 4528652288, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.12.attention.b_proj.weight": { "offset": 4528668672, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.dt_bias": { "offset": 4528717824, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.12.attention.f_proj.weight": { "offset": 4528734208, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.g_proj.weight": { "offset": 4535025664, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.k_conv1d.weight": { "offset": 4541317120, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.k_proj.weight": { "offset": 4541333504, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.o_norm.weight": { "offset": 4547624960, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.o_proj.weight": { "offset": 4547641344, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.q_conv1d.weight": { "offset": 4553932800, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.q_proj.weight": { "offset": 4553949184, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.v_conv1d.weight": { "offset": 4560240640, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.12.attention.v_proj.weight": { "offset": 4560257024, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.input_layernorm.weight": { "offset": 4566548480, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.mlp.gate.expert_bias": { "offset": 4566564864, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.12.mlp.gate.weight": { "offset": 4566581248, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.12.mlp.shared_experts.down_proj.weight": { "offset": 4567367680, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.12.mlp.shared_experts.gate_proj.weight": { "offset": 4568940544, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.mlp.shared_experts.up_proj.weight": { "offset": 4570513408, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.12.post_attention_layernorm.weight": { "offset": 4572086272, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.A_log": { "offset": 4572102656, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.13.attention.b_proj.weight": { "offset": 4572119040, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.dt_bias": { "offset": 4572168192, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.13.attention.f_proj.weight": { "offset": 4572184576, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.g_proj.weight": { "offset": 4578476032, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.k_conv1d.weight": { "offset": 4584767488, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.k_proj.weight": { "offset": 4584783872, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.o_norm.weight": { "offset": 4591075328, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.o_proj.weight": { "offset": 4591091712, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.q_conv1d.weight": { "offset": 4597383168, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.q_proj.weight": { "offset": 4597399552, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.v_conv1d.weight": { "offset": 4603691008, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.13.attention.v_proj.weight": { "offset": 4603707392, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.input_layernorm.weight": { "offset": 4609998848, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.mlp.gate.expert_bias": { "offset": 4610015232, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.13.mlp.gate.weight": { "offset": 4610031616, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.13.mlp.shared_experts.down_proj.weight": { "offset": 4610818048, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.13.mlp.shared_experts.gate_proj.weight": { "offset": 4612390912, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.mlp.shared_experts.up_proj.weight": { "offset": 4613963776, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.13.post_attention_layernorm.weight": { "offset": 4615536640, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.A_log": { "offset": 4615553024, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.14.attention.b_proj.weight": { "offset": 4615569408, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.dt_bias": { "offset": 4615618560, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.14.attention.f_proj.weight": { "offset": 4615634944, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.g_proj.weight": { "offset": 4621926400, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.k_conv1d.weight": { "offset": 4628217856, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.k_proj.weight": { "offset": 4628234240, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.o_norm.weight": { "offset": 4634525696, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.o_proj.weight": { "offset": 4634542080, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.q_conv1d.weight": { "offset": 4640833536, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.q_proj.weight": { "offset": 4640849920, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.v_conv1d.weight": { "offset": 4647141376, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.14.attention.v_proj.weight": { "offset": 4647157760, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.input_layernorm.weight": { "offset": 4653449216, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.mlp.gate.expert_bias": { "offset": 4653465600, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.14.mlp.gate.weight": { "offset": 4653481984, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.14.mlp.shared_experts.down_proj.weight": { "offset": 4654268416, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.14.mlp.shared_experts.gate_proj.weight": { "offset": 4655841280, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.mlp.shared_experts.up_proj.weight": { "offset": 4657414144, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.14.post_attention_layernorm.weight": { "offset": 4658987008, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.15.attention.dense.weight": { "offset": 4659003392, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.15.attention.g_proj.weight": { "offset": 4665294848, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.15.attention.kv_a_layernorm.weight": { "offset": 4665344000, "bytes": 1024, "shape": [ 512 ], "dtype": "bfloat16" }, "model.model.layers.15.attention.kv_a_proj_with_mqa.weight": { "offset": 4665360384, "bytes": 1769472, "shape": [ 576, 1536 ], "dtype": "bfloat16" }, "model.model.layers.15.attention.kv_b_proj.weight": { "offset": 4667129856, "bytes": 4194304, "shape": [ 4096, 512 ], "dtype": "bfloat16" }, "model.model.layers.15.attention.q_a_layernorm.weight": { "offset": 4671324160, "bytes": 512, "shape": [ 256 ], "dtype": "bfloat16" }, "model.model.layers.15.attention.q_a_proj.weight": { "offset": 4671340544, "bytes": 786432, "shape": [ 256, 1536 ], "dtype": "bfloat16" }, "model.model.layers.15.attention.q_b_proj.weight": { "offset": 4672126976, "bytes": 1572864, "shape": [ 3072, 256 ], "dtype": "bfloat16" }, "model.model.layers.15.input_layernorm.weight": { "offset": 4673699840, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.15.mlp.gate.expert_bias": { "offset": 4673716224, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.15.mlp.gate.weight": { "offset": 4673732608, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.15.mlp.shared_experts.down_proj.weight": { "offset": 4674519040, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.15.mlp.shared_experts.gate_proj.weight": { "offset": 4676091904, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.15.mlp.shared_experts.up_proj.weight": { "offset": 4677664768, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.15.post_attention_layernorm.weight": { "offset": 4679237632, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.A_log": { "offset": 4679254016, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.16.attention.b_proj.weight": { "offset": 4679270400, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.dt_bias": { "offset": 4679319552, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.16.attention.f_proj.weight": { "offset": 4679335936, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.g_proj.weight": { "offset": 4685627392, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.k_conv1d.weight": { "offset": 4691918848, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.k_proj.weight": { "offset": 4691935232, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.o_norm.weight": { "offset": 4698226688, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.o_proj.weight": { "offset": 4698243072, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.q_conv1d.weight": { "offset": 4704534528, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.q_proj.weight": { "offset": 4704550912, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.v_conv1d.weight": { "offset": 4710842368, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.16.attention.v_proj.weight": { "offset": 4710858752, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.input_layernorm.weight": { "offset": 4717150208, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.mlp.gate.expert_bias": { "offset": 4717166592, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.16.mlp.gate.weight": { "offset": 4717182976, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.16.mlp.shared_experts.down_proj.weight": { "offset": 4717969408, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.16.mlp.shared_experts.gate_proj.weight": { "offset": 4719542272, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.mlp.shared_experts.up_proj.weight": { "offset": 4721115136, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.16.post_attention_layernorm.weight": { "offset": 4722688000, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.A_log": { "offset": 4722704384, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.17.attention.b_proj.weight": { "offset": 4722720768, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.dt_bias": { "offset": 4722769920, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.17.attention.f_proj.weight": { "offset": 4722786304, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.g_proj.weight": { "offset": 4729077760, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.k_conv1d.weight": { "offset": 4735369216, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.k_proj.weight": { "offset": 4735385600, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.o_norm.weight": { "offset": 4741677056, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.o_proj.weight": { "offset": 4741693440, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.q_conv1d.weight": { "offset": 4747984896, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.q_proj.weight": { "offset": 4748001280, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.v_conv1d.weight": { "offset": 4754292736, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.17.attention.v_proj.weight": { "offset": 4754309120, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.input_layernorm.weight": { "offset": 4760600576, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.mlp.gate.expert_bias": { "offset": 4760616960, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.17.mlp.gate.weight": { "offset": 4760633344, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.17.mlp.shared_experts.down_proj.weight": { "offset": 4761419776, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.17.mlp.shared_experts.gate_proj.weight": { "offset": 4762992640, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.mlp.shared_experts.up_proj.weight": { "offset": 4764565504, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.17.post_attention_layernorm.weight": { "offset": 4766138368, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.A_log": { "offset": 4766154752, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.18.attention.b_proj.weight": { "offset": 4766171136, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.dt_bias": { "offset": 4766220288, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.18.attention.f_proj.weight": { "offset": 4766236672, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.g_proj.weight": { "offset": 4772528128, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.k_conv1d.weight": { "offset": 4778819584, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.k_proj.weight": { "offset": 4778835968, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.o_norm.weight": { "offset": 4785127424, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.o_proj.weight": { "offset": 4785143808, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.q_conv1d.weight": { "offset": 4791435264, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.q_proj.weight": { "offset": 4791451648, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.v_conv1d.weight": { "offset": 4797743104, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.18.attention.v_proj.weight": { "offset": 4797759488, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.input_layernorm.weight": { "offset": 4804050944, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.mlp.gate.expert_bias": { "offset": 4804067328, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.18.mlp.gate.weight": { "offset": 4804083712, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.18.mlp.shared_experts.down_proj.weight": { "offset": 4804870144, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.18.mlp.shared_experts.gate_proj.weight": { "offset": 4806443008, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.mlp.shared_experts.up_proj.weight": { "offset": 4808015872, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.18.post_attention_layernorm.weight": { "offset": 4809588736, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.19.attention.dense.weight": { "offset": 4809605120, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.19.attention.g_proj.weight": { "offset": 4815896576, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.19.attention.kv_a_layernorm.weight": { "offset": 4815945728, "bytes": 1024, "shape": [ 512 ], "dtype": "bfloat16" }, "model.model.layers.19.attention.kv_a_proj_with_mqa.weight": { "offset": 4815962112, "bytes": 1769472, "shape": [ 576, 1536 ], "dtype": "bfloat16" }, "model.model.layers.19.attention.kv_b_proj.weight": { "offset": 4817731584, "bytes": 4194304, "shape": [ 4096, 512 ], "dtype": "bfloat16" }, "model.model.layers.19.attention.q_a_layernorm.weight": { "offset": 4821925888, "bytes": 512, "shape": [ 256 ], "dtype": "bfloat16" }, "model.model.layers.19.attention.q_a_proj.weight": { "offset": 4821942272, "bytes": 786432, "shape": [ 256, 1536 ], "dtype": "bfloat16" }, "model.model.layers.19.attention.q_b_proj.weight": { "offset": 4822728704, "bytes": 1572864, "shape": [ 3072, 256 ], "dtype": "bfloat16" }, "model.model.layers.19.input_layernorm.weight": { "offset": 4824301568, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.19.mlp.gate.expert_bias": { "offset": 4824317952, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.19.mlp.gate.weight": { "offset": 4824334336, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.19.mlp.shared_experts.down_proj.weight": { "offset": 4825120768, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.19.mlp.shared_experts.gate_proj.weight": { "offset": 4826693632, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.19.mlp.shared_experts.up_proj.weight": { "offset": 4828266496, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.19.post_attention_layernorm.weight": { "offset": 4829839360, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.A_log": { "offset": 4829855744, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.2.attention.b_proj.weight": { "offset": 4829872128, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.dt_bias": { "offset": 4829921280, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.2.attention.f_proj.weight": { "offset": 4829937664, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.g_proj.weight": { "offset": 4836229120, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.k_conv1d.weight": { "offset": 4842520576, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.k_proj.weight": { "offset": 4842536960, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.o_norm.weight": { "offset": 4848828416, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.o_proj.weight": { "offset": 4848844800, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.q_conv1d.weight": { "offset": 4855136256, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.q_proj.weight": { "offset": 4855152640, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.v_conv1d.weight": { "offset": 4861444096, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.2.attention.v_proj.weight": { "offset": 4861460480, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.input_layernorm.weight": { "offset": 4867751936, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.mlp.gate.expert_bias": { "offset": 4867768320, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.2.mlp.gate.weight": { "offset": 4867784704, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.2.mlp.shared_experts.down_proj.weight": { "offset": 4868571136, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.2.mlp.shared_experts.gate_proj.weight": { "offset": 4870144000, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.mlp.shared_experts.up_proj.weight": { "offset": 4871716864, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.2.post_attention_layernorm.weight": { "offset": 4873289728, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.A_log": { "offset": 4873306112, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.20.attention.b_proj.weight": { "offset": 4873322496, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.dt_bias": { "offset": 4873371648, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.20.attention.f_proj.weight": { "offset": 4873388032, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.g_proj.weight": { "offset": 4879679488, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.k_conv1d.weight": { "offset": 4885970944, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.k_proj.weight": { "offset": 4885987328, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.o_norm.weight": { "offset": 4892278784, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.o_proj.weight": { "offset": 4892295168, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.q_conv1d.weight": { "offset": 4898586624, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.q_proj.weight": { "offset": 4898603008, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.v_conv1d.weight": { "offset": 4904894464, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.20.attention.v_proj.weight": { "offset": 4904910848, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.input_layernorm.weight": { "offset": 4911202304, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.mlp.gate.expert_bias": { "offset": 4911218688, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.20.mlp.gate.weight": { "offset": 4911235072, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.20.mlp.shared_experts.down_proj.weight": { "offset": 4912021504, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.20.mlp.shared_experts.gate_proj.weight": { "offset": 4913594368, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.mlp.shared_experts.up_proj.weight": { "offset": 4915167232, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.20.post_attention_layernorm.weight": { "offset": 4916740096, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.A_log": { "offset": 4916756480, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.21.attention.b_proj.weight": { "offset": 4916772864, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.dt_bias": { "offset": 4916822016, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.21.attention.f_proj.weight": { "offset": 4916838400, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.g_proj.weight": { "offset": 4923129856, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.k_conv1d.weight": { "offset": 4929421312, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.k_proj.weight": { "offset": 4929437696, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.o_norm.weight": { "offset": 4935729152, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.o_proj.weight": { "offset": 4935745536, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.q_conv1d.weight": { "offset": 4942036992, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.q_proj.weight": { "offset": 4942053376, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.v_conv1d.weight": { "offset": 4948344832, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.21.attention.v_proj.weight": { "offset": 4948361216, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.input_layernorm.weight": { "offset": 4954652672, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.mlp.gate.expert_bias": { "offset": 4954669056, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.21.mlp.gate.weight": { "offset": 4954685440, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.21.mlp.shared_experts.down_proj.weight": { "offset": 4955471872, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.21.mlp.shared_experts.gate_proj.weight": { "offset": 4957044736, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.mlp.shared_experts.up_proj.weight": { "offset": 4958617600, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.21.post_attention_layernorm.weight": { "offset": 4960190464, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.A_log": { "offset": 4960206848, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.22.attention.b_proj.weight": { "offset": 4960223232, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.dt_bias": { "offset": 4960272384, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.22.attention.f_proj.weight": { "offset": 4960288768, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.g_proj.weight": { "offset": 4966580224, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.k_conv1d.weight": { "offset": 4972871680, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.k_proj.weight": { "offset": 4972888064, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.o_norm.weight": { "offset": 4979179520, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.o_proj.weight": { "offset": 4979195904, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.q_conv1d.weight": { "offset": 4985487360, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.q_proj.weight": { "offset": 4985503744, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.v_conv1d.weight": { "offset": 4991795200, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.22.attention.v_proj.weight": { "offset": 4991811584, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.input_layernorm.weight": { "offset": 4998103040, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.mlp.gate.expert_bias": { "offset": 4998119424, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.22.mlp.gate.weight": { "offset": 4998135808, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.22.mlp.shared_experts.down_proj.weight": { "offset": 4998922240, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.22.mlp.shared_experts.gate_proj.weight": { "offset": 5000495104, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.mlp.shared_experts.up_proj.weight": { "offset": 5002067968, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.22.post_attention_layernorm.weight": { "offset": 5003640832, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.23.attention.dense.weight": { "offset": 5003657216, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.23.attention.g_proj.weight": { "offset": 5009948672, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.23.attention.kv_a_layernorm.weight": { "offset": 5009997824, "bytes": 1024, "shape": [ 512 ], "dtype": "bfloat16" }, "model.model.layers.23.attention.kv_a_proj_with_mqa.weight": { "offset": 5010014208, "bytes": 1769472, "shape": [ 576, 1536 ], "dtype": "bfloat16" }, "model.model.layers.23.attention.kv_b_proj.weight": { "offset": 5011783680, "bytes": 4194304, "shape": [ 4096, 512 ], "dtype": "bfloat16" }, "model.model.layers.23.attention.q_a_layernorm.weight": { "offset": 5015977984, "bytes": 512, "shape": [ 256 ], "dtype": "bfloat16" }, "model.model.layers.23.attention.q_a_proj.weight": { "offset": 5015994368, "bytes": 786432, "shape": [ 256, 1536 ], "dtype": "bfloat16" }, "model.model.layers.23.attention.q_b_proj.weight": { "offset": 5016780800, "bytes": 1572864, "shape": [ 3072, 256 ], "dtype": "bfloat16" }, "model.model.layers.23.input_layernorm.weight": { "offset": 5018353664, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.23.mlp.gate.expert_bias": { "offset": 5018370048, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.23.mlp.gate.weight": { "offset": 5018386432, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.23.mlp.shared_experts.down_proj.weight": { "offset": 5019172864, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.23.mlp.shared_experts.gate_proj.weight": { "offset": 5020745728, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.23.mlp.shared_experts.up_proj.weight": { "offset": 5022318592, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.23.post_attention_layernorm.weight": { "offset": 5023891456, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.3.attention.dense.weight": { "offset": 5023907840, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.3.attention.g_proj.weight": { "offset": 5030199296, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.3.attention.kv_a_layernorm.weight": { "offset": 5030248448, "bytes": 1024, "shape": [ 512 ], "dtype": "bfloat16" }, "model.model.layers.3.attention.kv_a_proj_with_mqa.weight": { "offset": 5030264832, "bytes": 1769472, "shape": [ 576, 1536 ], "dtype": "bfloat16" }, "model.model.layers.3.attention.kv_b_proj.weight": { "offset": 5032034304, "bytes": 4194304, "shape": [ 4096, 512 ], "dtype": "bfloat16" }, "model.model.layers.3.attention.q_a_layernorm.weight": { "offset": 5036228608, "bytes": 512, "shape": [ 256 ], "dtype": "bfloat16" }, "model.model.layers.3.attention.q_a_proj.weight": { "offset": 5036244992, "bytes": 786432, "shape": [ 256, 1536 ], "dtype": "bfloat16" }, "model.model.layers.3.attention.q_b_proj.weight": { "offset": 5037031424, "bytes": 1572864, "shape": [ 3072, 256 ], "dtype": "bfloat16" }, "model.model.layers.3.input_layernorm.weight": { "offset": 5038604288, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.3.mlp.gate.expert_bias": { "offset": 5038620672, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.3.mlp.gate.weight": { "offset": 5038637056, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.3.mlp.shared_experts.down_proj.weight": { "offset": 5039423488, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.3.mlp.shared_experts.gate_proj.weight": { "offset": 5040996352, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.3.mlp.shared_experts.up_proj.weight": { "offset": 5042569216, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.3.post_attention_layernorm.weight": { "offset": 5044142080, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.A_log": { "offset": 5044158464, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.4.attention.b_proj.weight": { "offset": 5044174848, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.dt_bias": { "offset": 5044224000, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.4.attention.f_proj.weight": { "offset": 5044240384, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.g_proj.weight": { "offset": 5050531840, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.k_conv1d.weight": { "offset": 5056823296, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.k_proj.weight": { "offset": 5056839680, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.o_norm.weight": { "offset": 5063131136, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.o_proj.weight": { "offset": 5063147520, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.q_conv1d.weight": { "offset": 5069438976, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.q_proj.weight": { "offset": 5069455360, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.v_conv1d.weight": { "offset": 5075746816, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.4.attention.v_proj.weight": { "offset": 5075763200, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.input_layernorm.weight": { "offset": 5082054656, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.mlp.gate.expert_bias": { "offset": 5082071040, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.4.mlp.gate.weight": { "offset": 5082087424, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.4.mlp.shared_experts.down_proj.weight": { "offset": 5082873856, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.4.mlp.shared_experts.gate_proj.weight": { "offset": 5084446720, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.mlp.shared_experts.up_proj.weight": { "offset": 5086019584, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.4.post_attention_layernorm.weight": { "offset": 5087592448, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.A_log": { "offset": 5087608832, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.5.attention.b_proj.weight": { "offset": 5087625216, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.dt_bias": { "offset": 5087674368, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.5.attention.f_proj.weight": { "offset": 5087690752, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.g_proj.weight": { "offset": 5093982208, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.k_conv1d.weight": { "offset": 5100273664, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.k_proj.weight": { "offset": 5100290048, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.o_norm.weight": { "offset": 5106581504, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.o_proj.weight": { "offset": 5106597888, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.q_conv1d.weight": { "offset": 5112889344, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.q_proj.weight": { "offset": 5112905728, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.v_conv1d.weight": { "offset": 5119197184, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.5.attention.v_proj.weight": { "offset": 5119213568, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.input_layernorm.weight": { "offset": 5125505024, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.mlp.gate.expert_bias": { "offset": 5125521408, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.5.mlp.gate.weight": { "offset": 5125537792, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.5.mlp.shared_experts.down_proj.weight": { "offset": 5126324224, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.5.mlp.shared_experts.gate_proj.weight": { "offset": 5127897088, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.mlp.shared_experts.up_proj.weight": { "offset": 5129469952, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.5.post_attention_layernorm.weight": { "offset": 5131042816, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.A_log": { "offset": 5131059200, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.6.attention.b_proj.weight": { "offset": 5131075584, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.dt_bias": { "offset": 5131124736, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.6.attention.f_proj.weight": { "offset": 5131141120, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.g_proj.weight": { "offset": 5137432576, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.k_conv1d.weight": { "offset": 5143724032, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.k_proj.weight": { "offset": 5143740416, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.o_norm.weight": { "offset": 5150031872, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.o_proj.weight": { "offset": 5150048256, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.q_conv1d.weight": { "offset": 5156339712, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.q_proj.weight": { "offset": 5156356096, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.v_conv1d.weight": { "offset": 5162647552, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.6.attention.v_proj.weight": { "offset": 5162663936, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.input_layernorm.weight": { "offset": 5168955392, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.mlp.gate.expert_bias": { "offset": 5168971776, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.6.mlp.gate.weight": { "offset": 5168988160, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.6.mlp.shared_experts.down_proj.weight": { "offset": 5169774592, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.6.mlp.shared_experts.gate_proj.weight": { "offset": 5171347456, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.mlp.shared_experts.up_proj.weight": { "offset": 5172920320, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.6.post_attention_layernorm.weight": { "offset": 5174493184, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.7.attention.dense.weight": { "offset": 5174509568, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.7.attention.g_proj.weight": { "offset": 5180801024, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.7.attention.kv_a_layernorm.weight": { "offset": 5180850176, "bytes": 1024, "shape": [ 512 ], "dtype": "bfloat16" }, "model.model.layers.7.attention.kv_a_proj_with_mqa.weight": { "offset": 5180866560, "bytes": 1769472, "shape": [ 576, 1536 ], "dtype": "bfloat16" }, "model.model.layers.7.attention.kv_b_proj.weight": { "offset": 5182636032, "bytes": 4194304, "shape": [ 4096, 512 ], "dtype": "bfloat16" }, "model.model.layers.7.attention.q_a_layernorm.weight": { "offset": 5186830336, "bytes": 512, "shape": [ 256 ], "dtype": "bfloat16" }, "model.model.layers.7.attention.q_a_proj.weight": { "offset": 5186846720, "bytes": 786432, "shape": [ 256, 1536 ], "dtype": "bfloat16" }, "model.model.layers.7.attention.q_b_proj.weight": { "offset": 5187633152, "bytes": 1572864, "shape": [ 3072, 256 ], "dtype": "bfloat16" }, "model.model.layers.7.input_layernorm.weight": { "offset": 5189206016, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.7.mlp.gate.expert_bias": { "offset": 5189222400, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.7.mlp.gate.weight": { "offset": 5189238784, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.7.mlp.shared_experts.down_proj.weight": { "offset": 5190025216, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.7.mlp.shared_experts.gate_proj.weight": { "offset": 5191598080, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.7.mlp.shared_experts.up_proj.weight": { "offset": 5193170944, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.7.post_attention_layernorm.weight": { "offset": 5194743808, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.A_log": { "offset": 5194760192, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.8.attention.b_proj.weight": { "offset": 5194776576, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.dt_bias": { "offset": 5194825728, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.8.attention.f_proj.weight": { "offset": 5194842112, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.g_proj.weight": { "offset": 5201133568, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.k_conv1d.weight": { "offset": 5207425024, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.k_proj.weight": { "offset": 5207441408, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.o_norm.weight": { "offset": 5213732864, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.o_proj.weight": { "offset": 5213749248, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.q_conv1d.weight": { "offset": 5220040704, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.q_proj.weight": { "offset": 5220057088, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.v_conv1d.weight": { "offset": 5226348544, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.8.attention.v_proj.weight": { "offset": 5226364928, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.input_layernorm.weight": { "offset": 5232656384, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.mlp.gate.expert_bias": { "offset": 5232672768, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.8.mlp.gate.weight": { "offset": 5232689152, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.8.mlp.shared_experts.down_proj.weight": { "offset": 5233475584, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.8.mlp.shared_experts.gate_proj.weight": { "offset": 5235048448, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.mlp.shared_experts.up_proj.weight": { "offset": 5236621312, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.8.post_attention_layernorm.weight": { "offset": 5238194176, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.A_log": { "offset": 5238210560, "bytes": 64, "shape": [ 16 ], "dtype": "float32" }, "model.model.layers.9.attention.b_proj.weight": { "offset": 5238226944, "bytes": 49152, "shape": [ 16, 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.dt_bias": { "offset": 5238276096, "bytes": 8192, "shape": [ 2048 ], "dtype": "float32" }, "model.model.layers.9.attention.f_proj.weight": { "offset": 5238292480, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.g_proj.weight": { "offset": 5244583936, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.k_conv1d.weight": { "offset": 5250875392, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.k_proj.weight": { "offset": 5250891776, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.o_norm.weight": { "offset": 5257183232, "bytes": 256, "shape": [ 128 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.o_proj.weight": { "offset": 5257199616, "bytes": 6291456, "shape": [ 1536, 2048 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.q_conv1d.weight": { "offset": 5263491072, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.q_proj.weight": { "offset": 5263507456, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.v_conv1d.weight": { "offset": 5269798912, "bytes": 16384, "shape": [ 2048, 1, 4 ], "dtype": "bfloat16" }, "model.model.layers.9.attention.v_proj.weight": { "offset": 5269815296, "bytes": 6291456, "shape": [ 2048, 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.input_layernorm.weight": { "offset": 5276106752, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.mlp.gate.expert_bias": { "offset": 5276123136, "bytes": 512, "shape": [ 128 ], "dtype": "float32" }, "model.model.layers.9.mlp.gate.weight": { "offset": 5276139520, "bytes": 786432, "shape": [ 128, 1536 ], "dtype": "float32" }, "model.model.layers.9.mlp.shared_experts.down_proj.weight": { "offset": 5276925952, "bytes": 1572864, "shape": [ 1536, 512 ], "dtype": "bfloat16" }, "model.model.layers.9.mlp.shared_experts.gate_proj.weight": { "offset": 5278498816, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.mlp.shared_experts.up_proj.weight": { "offset": 5280071680, "bytes": 1572864, "shape": [ 512, 1536 ], "dtype": "bfloat16" }, "model.model.layers.9.post_attention_layernorm.weight": { "offset": 5281644544, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.norm.weight": { "offset": 5281660928, "bytes": 3072, "shape": [ 1536 ], "dtype": "bfloat16" }, "model.model.word_embeddings.weight": { "offset": 5281677312, "bytes": 482869248, "shape": [ 157184, 1536 ], "dtype": "bfloat16" } } }