diff --git "a/ndarray-cache.json" "b/ndarray-cache.json" --- "a/ndarray-cache.json" +++ "b/ndarray-cache.json" @@ -1,22 +1,22 @@ { "metadata": { - "ParamSize": 327 + "ParamSize": 325 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", - "nbytes": 65536000, + "nbytes": 262144000, "records": [ { "name": "param_0", "shape": [ 32000, - 512 + 4096 ], - "dtype": "uint32", + "dtype": "float16", "format": "raw", - "nbytes": 65536000, + "nbytes": 262144000, "byteOffset": 0 } ] @@ -27,7 +27,7 @@ "nbytes": 58720256, "records": [ { - "name": "param_6", + "name": "param_5", "shape": [ 28672, 512 @@ -42,21 +42,139 @@ { "dataPath": "params_shard_2.bin", "format": "raw-shard", - "nbytes": 31784960, + "nbytes": 29360128, + "records": [ + { + "name": "param_7", + "shape": [ + 4096, + 1792 + ], + "dtype": "uint32", + "format": "raw", + "nbytes": 29360128, + "byteOffset": 0 + } + ] + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 30932992, "records": [ { "name": "param_1", "shape": [ - 32000, + 6144, + 512 + ], + "dtype": "uint32", + "format": "raw", + "nbytes": 12582912, + "byteOffset": 0 + }, + { + "name": "param_2", + "shape": [ + 6144, + 128 + ], + "dtype": "float16", + "format": "raw", + "nbytes": 1572864, + "byteOffset": 12582912 + }, + { + "name": "param_3", + "shape": [ + 4096, + 512 + ], + "dtype": "uint32", + "format": "raw", + "nbytes": 8388608, + "byteOffset": 14155776 + }, + { + "name": "param_4", + "shape": [ + 4096, + 128 + ], + "dtype": "float16", + "format": "raw", + "nbytes": 1048576, + "byteOffset": 22544384 + }, + { + "name": "param_6", + "shape": [ + 28672, 128 ], "dtype": "float16", "format": "raw", - "nbytes": 8192000, + "nbytes": 7340032, + "byteOffset": 23592960 + } + ] + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 58720256, + "records": [ + { + "name": "param_15", + "shape": [ + 28672, + 512 + ], + "dtype": "uint32", + "format": "raw", + "nbytes": 58720256, + "byteOffset": 0 + } + ] + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 27279360, + "records": [ + { + "name": "param_8", + "shape": [ + 4096, + 448 + ], + "dtype": "float16", + "format": "raw", + "nbytes": 3670016, "byteOffset": 0 }, { - "name": "param_2", + "name": "param_9", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "raw", + "nbytes": 8192, + "byteOffset": 3670016 + }, + { + "name": "param_10", + "shape": [ + 4096 + ], + "dtype": "float16", + "format": "raw", + "nbytes": 8192, + "byteOffset": 3678208 + }, + { + "name": "param_11", "shape": [ 6144, 512 @@ -64,10 +182,10 @@ "dtype": "uint32", "format": "raw", "nbytes": 12582912, - "byteOffset": 8192000 + "byteOffset": 3686400 }, { - "name": "param_3", + "name": "param_12", "shape": [ 6144, 128 @@ -75,10 +193,10 @@ "dtype": "float16", "format": "raw", "nbytes": 1572864, - "byteOffset": 20774912 + "byteOffset": 16269312 }, { - "name": "param_4", + "name": "param_13", "shape": [ 4096, 512 @@ -86,10 +204,10 @@ "dtype": "uint32", "format": "raw", "nbytes": 8388608, - "byteOffset": 22347776 + "byteOffset": 17842176 }, { - "name": "param_5", + "name": "param_14", "shape": [ 4096, 128 @@ -97,17 +215,17 @@ "dtype": "float16", "format": "raw", "nbytes": 1048576, - "byteOffset": 30736384 + "byteOffset": 26230784 } ] }, { - "dataPath": "params_shard_3.bin", + "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_8", + "name": "param_17", "shape": [ 4096, 1792 @@ -120,12 +238,12 @@ ] }, { - "dataPath": "params_shard_4.bin", + "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 25182208, "records": [ { - "name": "param_7", + "name": "param_16", "shape": [ 28672, 128 @@ -136,7 +254,7 @@ "byteOffset": 0 }, { - "name": "param_9", + "name": "param_18", "shape": [ 4096, 448 @@ -147,7 +265,7 @@ "byteOffset": 7340032 }, { - "name": "param_10", + "name": "param_19", "shape": [ 4096 ], @@ -157,7 +275,7 @@ "byteOffset": 11010048 }, { - "name": "param_11", + "name": "param_20", "shape": [ 4096 ], @@ -167,7 +285,7 @@ "byteOffset": 11018240 }, { - "name": "param_12", + "name": "param_21", "shape": [ 6144, 512 @@ -178,7 +296,7 @@ "byteOffset": 11026432 }, { - "name": "param_13", + "name": "param_22", "shape": [ 6144, 128 @@ -191,12 +309,12 @@ ] }, { - "dataPath": "params_shard_5.bin", + "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_16", + "name": "param_25", "shape": [ 28672, 512 @@ -209,12 +327,12 @@ ] }, { - "dataPath": "params_shard_6.bin", + "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_18", + "name": "param_27", "shape": [ 4096, 1792 @@ -227,12 +345,12 @@ ] }, { - "dataPath": "params_shard_7.bin", + "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 33046528, "records": [ { - "name": "param_14", + "name": "param_23", "shape": [ 4096, 512 @@ -243,7 +361,7 @@ "byteOffset": 0 }, { - "name": "param_15", + "name": "param_24", "shape": [ 4096, 128 @@ -254,7 +372,7 @@ "byteOffset": 8388608 }, { - "name": "param_17", + "name": "param_26", "shape": [ 28672, 128 @@ -265,7 +383,7 @@ "byteOffset": 9437184 }, { - "name": "param_19", + "name": "param_28", "shape": [ 4096, 448 @@ -276,7 +394,7 @@ "byteOffset": 16777216 }, { - "name": "param_20", + "name": "param_29", "shape": [ 4096 ], @@ -286,7 +404,7 @@ "byteOffset": 20447232 }, { - "name": "param_21", + "name": "param_30", "shape": [ 4096 ], @@ -296,7 +414,7 @@ "byteOffset": 20455424 }, { - "name": "param_22", + "name": "param_31", "shape": [ 6144, 512 @@ -309,12 +427,12 @@ ] }, { - "dataPath": "params_shard_8.bin", + "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_26", + "name": "param_35", "shape": [ 28672, 512 @@ -327,12 +445,12 @@ ] }, { - "dataPath": "params_shard_9.bin", + "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_28", + "name": "param_37", "shape": [ 4096, 1792 @@ -345,12 +463,12 @@ ] }, { - "dataPath": "params_shard_10.bin", + "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 22036480, "records": [ { - "name": "param_23", + "name": "param_32", "shape": [ 6144, 128 @@ -361,7 +479,7 @@ "byteOffset": 0 }, { - "name": "param_24", + "name": "param_33", "shape": [ 4096, 512 @@ -372,7 +490,7 @@ "byteOffset": 1572864 }, { - "name": "param_25", + "name": "param_34", "shape": [ 4096, 128 @@ -383,7 +501,7 @@ "byteOffset": 9961472 }, { - "name": "param_27", + "name": "param_36", "shape": [ 28672, 128 @@ -394,7 +512,7 @@ "byteOffset": 11010048 }, { - "name": "param_29", + "name": "param_38", "shape": [ 4096, 448 @@ -405,7 +523,7 @@ "byteOffset": 18350080 }, { - "name": "param_30", + "name": "param_39", "shape": [ 4096 ], @@ -415,7 +533,7 @@ "byteOffset": 22020096 }, { - "name": "param_31", + "name": "param_40", "shape": [ 4096 ], @@ -427,12 +545,12 @@ ] }, { - "dataPath": "params_shard_11.bin", + "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_36", + "name": "param_45", "shape": [ 28672, 512 @@ -445,12 +563,12 @@ ] }, { - "dataPath": "params_shard_12.bin", + "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_38", + "name": "param_47", "shape": [ 4096, 1792 @@ -463,12 +581,12 @@ ] }, { - "dataPath": "params_shard_13.bin", + "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 30932992, "records": [ { - "name": "param_32", + "name": "param_41", "shape": [ 6144, 512 @@ -479,7 +597,7 @@ "byteOffset": 0 }, { - "name": "param_33", + "name": "param_42", "shape": [ 6144, 128 @@ -490,7 +608,7 @@ "byteOffset": 12582912 }, { - "name": "param_34", + "name": "param_43", "shape": [ 4096, 512 @@ -501,7 +619,7 @@ "byteOffset": 14155776 }, { - "name": "param_35", + "name": "param_44", "shape": [ 4096, 128 @@ -512,7 +630,7 @@ "byteOffset": 22544384 }, { - "name": "param_37", + "name": "param_46", "shape": [ 28672, 128 @@ -525,12 +643,12 @@ ] }, { - "dataPath": "params_shard_14.bin", + "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_46", + "name": "param_55", "shape": [ 28672, 512 @@ -543,12 +661,12 @@ ] }, { - "dataPath": "params_shard_15.bin", + "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 27279360, "records": [ { - "name": "param_39", + "name": "param_48", "shape": [ 4096, 448 @@ -559,7 +677,7 @@ "byteOffset": 0 }, { - "name": "param_40", + "name": "param_49", "shape": [ 4096 ], @@ -569,7 +687,7 @@ "byteOffset": 3670016 }, { - "name": "param_41", + "name": "param_50", "shape": [ 4096 ], @@ -579,7 +697,7 @@ "byteOffset": 3678208 }, { - "name": "param_42", + "name": "param_51", "shape": [ 6144, 512 @@ -590,7 +708,7 @@ "byteOffset": 3686400 }, { - "name": "param_43", + "name": "param_52", "shape": [ 6144, 128 @@ -601,7 +719,7 @@ "byteOffset": 16269312 }, { - "name": "param_44", + "name": "param_53", "shape": [ 4096, 512 @@ -612,7 +730,7 @@ "byteOffset": 17842176 }, { - "name": "param_45", + "name": "param_54", "shape": [ 4096, 128 @@ -625,12 +743,12 @@ ] }, { - "dataPath": "params_shard_16.bin", + "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_48", + "name": "param_57", "shape": [ 4096, 1792 @@ -643,12 +761,12 @@ ] }, { - "dataPath": "params_shard_17.bin", + "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 25182208, "records": [ { - "name": "param_47", + "name": "param_56", "shape": [ 28672, 128 @@ -659,7 +777,7 @@ "byteOffset": 0 }, { - "name": "param_49", + "name": "param_58", "shape": [ 4096, 448 @@ -670,7 +788,7 @@ "byteOffset": 7340032 }, { - "name": "param_50", + "name": "param_59", "shape": [ 4096 ], @@ -680,7 +798,7 @@ "byteOffset": 11010048 }, { - "name": "param_51", + "name": "param_60", "shape": [ 4096 ], @@ -690,7 +808,7 @@ "byteOffset": 11018240 }, { - "name": "param_52", + "name": "param_61", "shape": [ 6144, 512 @@ -701,7 +819,7 @@ "byteOffset": 11026432 }, { - "name": "param_53", + "name": "param_62", "shape": [ 6144, 128 @@ -714,12 +832,12 @@ ] }, { - "dataPath": "params_shard_18.bin", + "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_56", + "name": "param_65", "shape": [ 28672, 512 @@ -732,12 +850,12 @@ ] }, { - "dataPath": "params_shard_19.bin", + "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_58", + "name": "param_67", "shape": [ 4096, 1792 @@ -750,12 +868,12 @@ ] }, { - "dataPath": "params_shard_20.bin", + "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 33046528, "records": [ { - "name": "param_54", + "name": "param_63", "shape": [ 4096, 512 @@ -766,7 +884,7 @@ "byteOffset": 0 }, { - "name": "param_55", + "name": "param_64", "shape": [ 4096, 128 @@ -777,7 +895,7 @@ "byteOffset": 8388608 }, { - "name": "param_57", + "name": "param_66", "shape": [ 28672, 128 @@ -788,7 +906,7 @@ "byteOffset": 9437184 }, { - "name": "param_59", + "name": "param_68", "shape": [ 4096, 448 @@ -799,7 +917,7 @@ "byteOffset": 16777216 }, { - "name": "param_60", + "name": "param_69", "shape": [ 4096 ], @@ -809,7 +927,7 @@ "byteOffset": 20447232 }, { - "name": "param_61", + "name": "param_70", "shape": [ 4096 ], @@ -819,7 +937,7 @@ "byteOffset": 20455424 }, { - "name": "param_62", + "name": "param_71", "shape": [ 6144, 512 @@ -832,12 +950,12 @@ ] }, { - "dataPath": "params_shard_21.bin", + "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_66", + "name": "param_75", "shape": [ 28672, 512 @@ -850,12 +968,12 @@ ] }, { - "dataPath": "params_shard_22.bin", + "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_68", + "name": "param_77", "shape": [ 4096, 1792 @@ -868,12 +986,12 @@ ] }, { - "dataPath": "params_shard_23.bin", + "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 22036480, "records": [ { - "name": "param_63", + "name": "param_72", "shape": [ 6144, 128 @@ -884,7 +1002,7 @@ "byteOffset": 0 }, { - "name": "param_64", + "name": "param_73", "shape": [ 4096, 512 @@ -895,7 +1013,7 @@ "byteOffset": 1572864 }, { - "name": "param_65", + "name": "param_74", "shape": [ 4096, 128 @@ -906,7 +1024,7 @@ "byteOffset": 9961472 }, { - "name": "param_67", + "name": "param_76", "shape": [ 28672, 128 @@ -917,7 +1035,7 @@ "byteOffset": 11010048 }, { - "name": "param_69", + "name": "param_78", "shape": [ 4096, 448 @@ -928,7 +1046,7 @@ "byteOffset": 18350080 }, { - "name": "param_70", + "name": "param_79", "shape": [ 4096 ], @@ -938,7 +1056,7 @@ "byteOffset": 22020096 }, { - "name": "param_71", + "name": "param_80", "shape": [ 4096 ], @@ -950,12 +1068,12 @@ ] }, { - "dataPath": "params_shard_24.bin", + "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_76", + "name": "param_85", "shape": [ 28672, 512 @@ -968,12 +1086,12 @@ ] }, { - "dataPath": "params_shard_25.bin", + "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_78", + "name": "param_87", "shape": [ 4096, 1792 @@ -986,12 +1104,12 @@ ] }, { - "dataPath": "params_shard_26.bin", + "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 30932992, "records": [ { - "name": "param_72", + "name": "param_81", "shape": [ 6144, 512 @@ -1002,7 +1120,7 @@ "byteOffset": 0 }, { - "name": "param_73", + "name": "param_82", "shape": [ 6144, 128 @@ -1013,7 +1131,7 @@ "byteOffset": 12582912 }, { - "name": "param_74", + "name": "param_83", "shape": [ 4096, 512 @@ -1024,7 +1142,7 @@ "byteOffset": 14155776 }, { - "name": "param_75", + "name": "param_84", "shape": [ 4096, 128 @@ -1035,7 +1153,7 @@ "byteOffset": 22544384 }, { - "name": "param_77", + "name": "param_86", "shape": [ 28672, 128 @@ -1048,12 +1166,12 @@ ] }, { - "dataPath": "params_shard_27.bin", + "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_86", + "name": "param_95", "shape": [ 28672, 512 @@ -1066,12 +1184,12 @@ ] }, { - "dataPath": "params_shard_28.bin", + "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 27279360, "records": [ { - "name": "param_79", + "name": "param_88", "shape": [ 4096, 448 @@ -1082,7 +1200,7 @@ "byteOffset": 0 }, { - "name": "param_80", + "name": "param_89", "shape": [ 4096 ], @@ -1092,7 +1210,7 @@ "byteOffset": 3670016 }, { - "name": "param_81", + "name": "param_90", "shape": [ 4096 ], @@ -1102,7 +1220,7 @@ "byteOffset": 3678208 }, { - "name": "param_82", + "name": "param_91", "shape": [ 6144, 512 @@ -1113,7 +1231,7 @@ "byteOffset": 3686400 }, { - "name": "param_83", + "name": "param_92", "shape": [ 6144, 128 @@ -1124,7 +1242,7 @@ "byteOffset": 16269312 }, { - "name": "param_84", + "name": "param_93", "shape": [ 4096, 512 @@ -1135,7 +1253,7 @@ "byteOffset": 17842176 }, { - "name": "param_85", + "name": "param_94", "shape": [ 4096, 128 @@ -1148,12 +1266,12 @@ ] }, { - "dataPath": "params_shard_29.bin", + "dataPath": "params_shard_32.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_88", + "name": "param_97", "shape": [ 4096, 1792 @@ -1166,12 +1284,12 @@ ] }, { - "dataPath": "params_shard_30.bin", + "dataPath": "params_shard_33.bin", "format": "raw-shard", "nbytes": 25182208, "records": [ { - "name": "param_87", + "name": "param_96", "shape": [ 28672, 128 @@ -1182,7 +1300,7 @@ "byteOffset": 0 }, { - "name": "param_89", + "name": "param_98", "shape": [ 4096, 448 @@ -1193,7 +1311,7 @@ "byteOffset": 7340032 }, { - "name": "param_90", + "name": "param_99", "shape": [ 4096 ], @@ -1203,7 +1321,7 @@ "byteOffset": 11010048 }, { - "name": "param_91", + "name": "param_100", "shape": [ 4096 ], @@ -1213,7 +1331,7 @@ "byteOffset": 11018240 }, { - "name": "param_92", + "name": "param_101", "shape": [ 6144, 512 @@ -1224,7 +1342,7 @@ "byteOffset": 11026432 }, { - "name": "param_93", + "name": "param_102", "shape": [ 6144, 128 @@ -1237,12 +1355,12 @@ ] }, { - "dataPath": "params_shard_31.bin", + "dataPath": "params_shard_34.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_96", + "name": "param_105", "shape": [ 28672, 512 @@ -1255,12 +1373,12 @@ ] }, { - "dataPath": "params_shard_32.bin", + "dataPath": "params_shard_35.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_98", + "name": "param_107", "shape": [ 4096, 1792 @@ -1273,12 +1391,12 @@ ] }, { - "dataPath": "params_shard_33.bin", + "dataPath": "params_shard_36.bin", "format": "raw-shard", "nbytes": 33046528, "records": [ { - "name": "param_94", + "name": "param_103", "shape": [ 4096, 512 @@ -1289,7 +1407,7 @@ "byteOffset": 0 }, { - "name": "param_95", + "name": "param_104", "shape": [ 4096, 128 @@ -1300,7 +1418,7 @@ "byteOffset": 8388608 }, { - "name": "param_97", + "name": "param_106", "shape": [ 28672, 128 @@ -1311,7 +1429,7 @@ "byteOffset": 9437184 }, { - "name": "param_99", + "name": "param_108", "shape": [ 4096, 448 @@ -1322,7 +1440,7 @@ "byteOffset": 16777216 }, { - "name": "param_100", + "name": "param_109", "shape": [ 4096 ], @@ -1332,7 +1450,7 @@ "byteOffset": 20447232 }, { - "name": "param_101", + "name": "param_110", "shape": [ 4096 ], @@ -1342,7 +1460,7 @@ "byteOffset": 20455424 }, { - "name": "param_102", + "name": "param_111", "shape": [ 6144, 512 @@ -1355,12 +1473,12 @@ ] }, { - "dataPath": "params_shard_34.bin", + "dataPath": "params_shard_37.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_106", + "name": "param_115", "shape": [ 28672, 512 @@ -1373,12 +1491,12 @@ ] }, { - "dataPath": "params_shard_35.bin", + "dataPath": "params_shard_38.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_108", + "name": "param_117", "shape": [ 4096, 1792 @@ -1391,12 +1509,12 @@ ] }, { - "dataPath": "params_shard_36.bin", + "dataPath": "params_shard_39.bin", "format": "raw-shard", "nbytes": 22036480, "records": [ { - "name": "param_103", + "name": "param_112", "shape": [ 6144, 128 @@ -1407,7 +1525,7 @@ "byteOffset": 0 }, { - "name": "param_104", + "name": "param_113", "shape": [ 4096, 512 @@ -1418,7 +1536,7 @@ "byteOffset": 1572864 }, { - "name": "param_105", + "name": "param_114", "shape": [ 4096, 128 @@ -1429,7 +1547,7 @@ "byteOffset": 9961472 }, { - "name": "param_107", + "name": "param_116", "shape": [ 28672, 128 @@ -1440,7 +1558,7 @@ "byteOffset": 11010048 }, { - "name": "param_109", + "name": "param_118", "shape": [ 4096, 448 @@ -1451,7 +1569,7 @@ "byteOffset": 18350080 }, { - "name": "param_110", + "name": "param_119", "shape": [ 4096 ], @@ -1461,7 +1579,7 @@ "byteOffset": 22020096 }, { - "name": "param_111", + "name": "param_120", "shape": [ 4096 ], @@ -1473,12 +1591,12 @@ ] }, { - "dataPath": "params_shard_37.bin", + "dataPath": "params_shard_40.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_116", + "name": "param_125", "shape": [ 28672, 512 @@ -1491,12 +1609,12 @@ ] }, { - "dataPath": "params_shard_38.bin", + "dataPath": "params_shard_41.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_118", + "name": "param_127", "shape": [ 4096, 1792 @@ -1509,12 +1627,12 @@ ] }, { - "dataPath": "params_shard_39.bin", + "dataPath": "params_shard_42.bin", "format": "raw-shard", "nbytes": 30932992, "records": [ { - "name": "param_112", + "name": "param_121", "shape": [ 6144, 512 @@ -1525,7 +1643,7 @@ "byteOffset": 0 }, { - "name": "param_113", + "name": "param_122", "shape": [ 6144, 128 @@ -1536,7 +1654,7 @@ "byteOffset": 12582912 }, { - "name": "param_114", + "name": "param_123", "shape": [ 4096, 512 @@ -1547,7 +1665,7 @@ "byteOffset": 14155776 }, { - "name": "param_115", + "name": "param_124", "shape": [ 4096, 128 @@ -1558,7 +1676,7 @@ "byteOffset": 22544384 }, { - "name": "param_117", + "name": "param_126", "shape": [ 28672, 128 @@ -1571,12 +1689,12 @@ ] }, { - "dataPath": "params_shard_40.bin", + "dataPath": "params_shard_43.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_126", + "name": "param_135", "shape": [ 28672, 512 @@ -1589,12 +1707,12 @@ ] }, { - "dataPath": "params_shard_41.bin", + "dataPath": "params_shard_44.bin", "format": "raw-shard", "nbytes": 27279360, "records": [ { - "name": "param_119", + "name": "param_128", "shape": [ 4096, 448 @@ -1605,7 +1723,7 @@ "byteOffset": 0 }, { - "name": "param_120", + "name": "param_129", "shape": [ 4096 ], @@ -1615,7 +1733,7 @@ "byteOffset": 3670016 }, { - "name": "param_121", + "name": "param_130", "shape": [ 4096 ], @@ -1625,7 +1743,7 @@ "byteOffset": 3678208 }, { - "name": "param_122", + "name": "param_131", "shape": [ 6144, 512 @@ -1636,7 +1754,7 @@ "byteOffset": 3686400 }, { - "name": "param_123", + "name": "param_132", "shape": [ 6144, 128 @@ -1647,7 +1765,7 @@ "byteOffset": 16269312 }, { - "name": "param_124", + "name": "param_133", "shape": [ 4096, 512 @@ -1658,7 +1776,7 @@ "byteOffset": 17842176 }, { - "name": "param_125", + "name": "param_134", "shape": [ 4096, 128 @@ -1671,12 +1789,12 @@ ] }, { - "dataPath": "params_shard_42.bin", + "dataPath": "params_shard_45.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_128", + "name": "param_137", "shape": [ 4096, 1792 @@ -1689,12 +1807,12 @@ ] }, { - "dataPath": "params_shard_43.bin", + "dataPath": "params_shard_46.bin", "format": "raw-shard", "nbytes": 25182208, "records": [ { - "name": "param_127", + "name": "param_136", "shape": [ 28672, 128 @@ -1705,7 +1823,7 @@ "byteOffset": 0 }, { - "name": "param_129", + "name": "param_138", "shape": [ 4096, 448 @@ -1716,7 +1834,7 @@ "byteOffset": 7340032 }, { - "name": "param_130", + "name": "param_139", "shape": [ 4096 ], @@ -1726,7 +1844,7 @@ "byteOffset": 11010048 }, { - "name": "param_131", + "name": "param_140", "shape": [ 4096 ], @@ -1736,7 +1854,7 @@ "byteOffset": 11018240 }, { - "name": "param_132", + "name": "param_141", "shape": [ 6144, 512 @@ -1747,7 +1865,7 @@ "byteOffset": 11026432 }, { - "name": "param_133", + "name": "param_142", "shape": [ 6144, 128 @@ -1760,12 +1878,12 @@ ] }, { - "dataPath": "params_shard_44.bin", + "dataPath": "params_shard_47.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_136", + "name": "param_145", "shape": [ 28672, 512 @@ -1778,12 +1896,12 @@ ] }, { - "dataPath": "params_shard_45.bin", + "dataPath": "params_shard_48.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_138", + "name": "param_147", "shape": [ 4096, 1792 @@ -1796,12 +1914,12 @@ ] }, { - "dataPath": "params_shard_46.bin", + "dataPath": "params_shard_49.bin", "format": "raw-shard", "nbytes": 33046528, "records": [ { - "name": "param_134", + "name": "param_143", "shape": [ 4096, 512 @@ -1812,7 +1930,7 @@ "byteOffset": 0 }, { - "name": "param_135", + "name": "param_144", "shape": [ 4096, 128 @@ -1823,7 +1941,7 @@ "byteOffset": 8388608 }, { - "name": "param_137", + "name": "param_146", "shape": [ 28672, 128 @@ -1834,7 +1952,7 @@ "byteOffset": 9437184 }, { - "name": "param_139", + "name": "param_148", "shape": [ 4096, 448 @@ -1845,7 +1963,7 @@ "byteOffset": 16777216 }, { - "name": "param_140", + "name": "param_149", "shape": [ 4096 ], @@ -1855,7 +1973,7 @@ "byteOffset": 20447232 }, { - "name": "param_141", + "name": "param_150", "shape": [ 4096 ], @@ -1865,7 +1983,7 @@ "byteOffset": 20455424 }, { - "name": "param_142", + "name": "param_151", "shape": [ 6144, 512 @@ -1878,12 +1996,12 @@ ] }, { - "dataPath": "params_shard_47.bin", + "dataPath": "params_shard_50.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_146", + "name": "param_155", "shape": [ 28672, 512 @@ -1896,12 +2014,12 @@ ] }, { - "dataPath": "params_shard_48.bin", + "dataPath": "params_shard_51.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_148", + "name": "param_157", "shape": [ 4096, 1792 @@ -1914,12 +2032,12 @@ ] }, { - "dataPath": "params_shard_49.bin", + "dataPath": "params_shard_52.bin", "format": "raw-shard", "nbytes": 22036480, "records": [ { - "name": "param_143", + "name": "param_152", "shape": [ 6144, 128 @@ -1930,7 +2048,7 @@ "byteOffset": 0 }, { - "name": "param_144", + "name": "param_153", "shape": [ 4096, 512 @@ -1941,7 +2059,7 @@ "byteOffset": 1572864 }, { - "name": "param_145", + "name": "param_154", "shape": [ 4096, 128 @@ -1952,7 +2070,7 @@ "byteOffset": 9961472 }, { - "name": "param_147", + "name": "param_156", "shape": [ 28672, 128 @@ -1963,7 +2081,7 @@ "byteOffset": 11010048 }, { - "name": "param_149", + "name": "param_158", "shape": [ 4096, 448 @@ -1974,7 +2092,7 @@ "byteOffset": 18350080 }, { - "name": "param_150", + "name": "param_159", "shape": [ 4096 ], @@ -1984,7 +2102,7 @@ "byteOffset": 22020096 }, { - "name": "param_151", + "name": "param_160", "shape": [ 4096 ], @@ -1996,12 +2114,12 @@ ] }, { - "dataPath": "params_shard_50.bin", + "dataPath": "params_shard_53.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_156", + "name": "param_165", "shape": [ 28672, 512 @@ -2014,12 +2132,12 @@ ] }, { - "dataPath": "params_shard_51.bin", + "dataPath": "params_shard_54.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_158", + "name": "param_167", "shape": [ 4096, 1792 @@ -2032,12 +2150,12 @@ ] }, { - "dataPath": "params_shard_52.bin", + "dataPath": "params_shard_55.bin", "format": "raw-shard", "nbytes": 30932992, "records": [ { - "name": "param_152", + "name": "param_161", "shape": [ 6144, 512 @@ -2048,7 +2166,7 @@ "byteOffset": 0 }, { - "name": "param_153", + "name": "param_162", "shape": [ 6144, 128 @@ -2059,7 +2177,7 @@ "byteOffset": 12582912 }, { - "name": "param_154", + "name": "param_163", "shape": [ 4096, 512 @@ -2070,7 +2188,7 @@ "byteOffset": 14155776 }, { - "name": "param_155", + "name": "param_164", "shape": [ 4096, 128 @@ -2081,7 +2199,7 @@ "byteOffset": 22544384 }, { - "name": "param_157", + "name": "param_166", "shape": [ 28672, 128 @@ -2094,12 +2212,12 @@ ] }, { - "dataPath": "params_shard_53.bin", + "dataPath": "params_shard_56.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_166", + "name": "param_175", "shape": [ 28672, 512 @@ -2112,12 +2230,12 @@ ] }, { - "dataPath": "params_shard_54.bin", + "dataPath": "params_shard_57.bin", "format": "raw-shard", "nbytes": 27279360, "records": [ { - "name": "param_159", + "name": "param_168", "shape": [ 4096, 448 @@ -2128,7 +2246,7 @@ "byteOffset": 0 }, { - "name": "param_160", + "name": "param_169", "shape": [ 4096 ], @@ -2138,7 +2256,7 @@ "byteOffset": 3670016 }, { - "name": "param_161", + "name": "param_170", "shape": [ 4096 ], @@ -2148,7 +2266,7 @@ "byteOffset": 3678208 }, { - "name": "param_162", + "name": "param_171", "shape": [ 6144, 512 @@ -2159,7 +2277,7 @@ "byteOffset": 3686400 }, { - "name": "param_163", + "name": "param_172", "shape": [ 6144, 128 @@ -2170,7 +2288,7 @@ "byteOffset": 16269312 }, { - "name": "param_164", + "name": "param_173", "shape": [ 4096, 512 @@ -2181,7 +2299,7 @@ "byteOffset": 17842176 }, { - "name": "param_165", + "name": "param_174", "shape": [ 4096, 128 @@ -2194,12 +2312,12 @@ ] }, { - "dataPath": "params_shard_55.bin", + "dataPath": "params_shard_58.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_168", + "name": "param_177", "shape": [ 4096, 1792 @@ -2212,12 +2330,12 @@ ] }, { - "dataPath": "params_shard_56.bin", + "dataPath": "params_shard_59.bin", "format": "raw-shard", "nbytes": 25182208, "records": [ { - "name": "param_167", + "name": "param_176", "shape": [ 28672, 128 @@ -2228,7 +2346,7 @@ "byteOffset": 0 }, { - "name": "param_169", + "name": "param_178", "shape": [ 4096, 448 @@ -2239,7 +2357,7 @@ "byteOffset": 7340032 }, { - "name": "param_170", + "name": "param_179", "shape": [ 4096 ], @@ -2249,7 +2367,7 @@ "byteOffset": 11010048 }, { - "name": "param_171", + "name": "param_180", "shape": [ 4096 ], @@ -2259,7 +2377,7 @@ "byteOffset": 11018240 }, { - "name": "param_172", + "name": "param_181", "shape": [ 6144, 512 @@ -2270,7 +2388,7 @@ "byteOffset": 11026432 }, { - "name": "param_173", + "name": "param_182", "shape": [ 6144, 128 @@ -2283,12 +2401,12 @@ ] }, { - "dataPath": "params_shard_57.bin", + "dataPath": "params_shard_60.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_176", + "name": "param_185", "shape": [ 28672, 512 @@ -2301,12 +2419,12 @@ ] }, { - "dataPath": "params_shard_58.bin", + "dataPath": "params_shard_61.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_178", + "name": "param_187", "shape": [ 4096, 1792 @@ -2319,12 +2437,12 @@ ] }, { - "dataPath": "params_shard_59.bin", + "dataPath": "params_shard_62.bin", "format": "raw-shard", "nbytes": 33046528, "records": [ { - "name": "param_174", + "name": "param_183", "shape": [ 4096, 512 @@ -2335,7 +2453,7 @@ "byteOffset": 0 }, { - "name": "param_175", + "name": "param_184", "shape": [ 4096, 128 @@ -2346,7 +2464,7 @@ "byteOffset": 8388608 }, { - "name": "param_177", + "name": "param_186", "shape": [ 28672, 128 @@ -2357,7 +2475,7 @@ "byteOffset": 9437184 }, { - "name": "param_179", + "name": "param_188", "shape": [ 4096, 448 @@ -2368,7 +2486,7 @@ "byteOffset": 16777216 }, { - "name": "param_180", + "name": "param_189", "shape": [ 4096 ], @@ -2378,7 +2496,7 @@ "byteOffset": 20447232 }, { - "name": "param_181", + "name": "param_190", "shape": [ 4096 ], @@ -2388,7 +2506,7 @@ "byteOffset": 20455424 }, { - "name": "param_182", + "name": "param_191", "shape": [ 6144, 512 @@ -2401,12 +2519,12 @@ ] }, { - "dataPath": "params_shard_60.bin", + "dataPath": "params_shard_63.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_186", + "name": "param_195", "shape": [ 28672, 512 @@ -2419,12 +2537,12 @@ ] }, { - "dataPath": "params_shard_61.bin", + "dataPath": "params_shard_64.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_188", + "name": "param_197", "shape": [ 4096, 1792 @@ -2437,12 +2555,12 @@ ] }, { - "dataPath": "params_shard_62.bin", + "dataPath": "params_shard_65.bin", "format": "raw-shard", "nbytes": 22036480, "records": [ { - "name": "param_183", + "name": "param_192", "shape": [ 6144, 128 @@ -2453,7 +2571,7 @@ "byteOffset": 0 }, { - "name": "param_184", + "name": "param_193", "shape": [ 4096, 512 @@ -2464,7 +2582,7 @@ "byteOffset": 1572864 }, { - "name": "param_185", + "name": "param_194", "shape": [ 4096, 128 @@ -2475,7 +2593,7 @@ "byteOffset": 9961472 }, { - "name": "param_187", + "name": "param_196", "shape": [ 28672, 128 @@ -2486,7 +2604,7 @@ "byteOffset": 11010048 }, { - "name": "param_189", + "name": "param_198", "shape": [ 4096, 448 @@ -2497,7 +2615,7 @@ "byteOffset": 18350080 }, { - "name": "param_190", + "name": "param_199", "shape": [ 4096 ], @@ -2507,7 +2625,7 @@ "byteOffset": 22020096 }, { - "name": "param_191", + "name": "param_200", "shape": [ 4096 ], @@ -2519,12 +2637,12 @@ ] }, { - "dataPath": "params_shard_63.bin", + "dataPath": "params_shard_66.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_196", + "name": "param_205", "shape": [ 28672, 512 @@ -2537,12 +2655,12 @@ ] }, { - "dataPath": "params_shard_64.bin", + "dataPath": "params_shard_67.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_198", + "name": "param_207", "shape": [ 4096, 1792 @@ -2555,12 +2673,12 @@ ] }, { - "dataPath": "params_shard_65.bin", + "dataPath": "params_shard_68.bin", "format": "raw-shard", "nbytes": 30932992, "records": [ { - "name": "param_192", + "name": "param_201", "shape": [ 6144, 512 @@ -2571,7 +2689,7 @@ "byteOffset": 0 }, { - "name": "param_193", + "name": "param_202", "shape": [ 6144, 128 @@ -2582,7 +2700,7 @@ "byteOffset": 12582912 }, { - "name": "param_194", + "name": "param_203", "shape": [ 4096, 512 @@ -2593,7 +2711,7 @@ "byteOffset": 14155776 }, { - "name": "param_195", + "name": "param_204", "shape": [ 4096, 128 @@ -2604,7 +2722,7 @@ "byteOffset": 22544384 }, { - "name": "param_197", + "name": "param_206", "shape": [ 28672, 128 @@ -2617,12 +2735,12 @@ ] }, { - "dataPath": "params_shard_66.bin", + "dataPath": "params_shard_69.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_206", + "name": "param_215", "shape": [ 28672, 512 @@ -2635,12 +2753,12 @@ ] }, { - "dataPath": "params_shard_67.bin", + "dataPath": "params_shard_70.bin", "format": "raw-shard", "nbytes": 27279360, "records": [ { - "name": "param_199", + "name": "param_208", "shape": [ 4096, 448 @@ -2651,7 +2769,7 @@ "byteOffset": 0 }, { - "name": "param_200", + "name": "param_209", "shape": [ 4096 ], @@ -2661,7 +2779,7 @@ "byteOffset": 3670016 }, { - "name": "param_201", + "name": "param_210", "shape": [ 4096 ], @@ -2671,7 +2789,7 @@ "byteOffset": 3678208 }, { - "name": "param_202", + "name": "param_211", "shape": [ 6144, 512 @@ -2682,7 +2800,7 @@ "byteOffset": 3686400 }, { - "name": "param_203", + "name": "param_212", "shape": [ 6144, 128 @@ -2693,7 +2811,7 @@ "byteOffset": 16269312 }, { - "name": "param_204", + "name": "param_213", "shape": [ 4096, 512 @@ -2704,7 +2822,7 @@ "byteOffset": 17842176 }, { - "name": "param_205", + "name": "param_214", "shape": [ 4096, 128 @@ -2717,12 +2835,12 @@ ] }, { - "dataPath": "params_shard_68.bin", + "dataPath": "params_shard_71.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_208", + "name": "param_217", "shape": [ 4096, 1792 @@ -2735,12 +2853,12 @@ ] }, { - "dataPath": "params_shard_69.bin", + "dataPath": "params_shard_72.bin", "format": "raw-shard", "nbytes": 25182208, "records": [ { - "name": "param_207", + "name": "param_216", "shape": [ 28672, 128 @@ -2751,7 +2869,7 @@ "byteOffset": 0 }, { - "name": "param_209", + "name": "param_218", "shape": [ 4096, 448 @@ -2762,7 +2880,7 @@ "byteOffset": 7340032 }, { - "name": "param_210", + "name": "param_219", "shape": [ 4096 ], @@ -2772,7 +2890,7 @@ "byteOffset": 11010048 }, { - "name": "param_211", + "name": "param_220", "shape": [ 4096 ], @@ -2782,7 +2900,7 @@ "byteOffset": 11018240 }, { - "name": "param_212", + "name": "param_221", "shape": [ 6144, 512 @@ -2793,7 +2911,7 @@ "byteOffset": 11026432 }, { - "name": "param_213", + "name": "param_222", "shape": [ 6144, 128 @@ -2806,12 +2924,12 @@ ] }, { - "dataPath": "params_shard_70.bin", + "dataPath": "params_shard_73.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_216", + "name": "param_225", "shape": [ 28672, 512 @@ -2824,12 +2942,12 @@ ] }, { - "dataPath": "params_shard_71.bin", + "dataPath": "params_shard_74.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_218", + "name": "param_227", "shape": [ 4096, 1792 @@ -2842,12 +2960,12 @@ ] }, { - "dataPath": "params_shard_72.bin", + "dataPath": "params_shard_75.bin", "format": "raw-shard", "nbytes": 33046528, "records": [ { - "name": "param_214", + "name": "param_223", "shape": [ 4096, 512 @@ -2858,7 +2976,7 @@ "byteOffset": 0 }, { - "name": "param_215", + "name": "param_224", "shape": [ 4096, 128 @@ -2869,7 +2987,7 @@ "byteOffset": 8388608 }, { - "name": "param_217", + "name": "param_226", "shape": [ 28672, 128 @@ -2880,7 +2998,7 @@ "byteOffset": 9437184 }, { - "name": "param_219", + "name": "param_228", "shape": [ 4096, 448 @@ -2891,7 +3009,7 @@ "byteOffset": 16777216 }, { - "name": "param_220", + "name": "param_229", "shape": [ 4096 ], @@ -2901,7 +3019,7 @@ "byteOffset": 20447232 }, { - "name": "param_221", + "name": "param_230", "shape": [ 4096 ], @@ -2911,7 +3029,7 @@ "byteOffset": 20455424 }, { - "name": "param_222", + "name": "param_231", "shape": [ 6144, 512 @@ -2924,12 +3042,12 @@ ] }, { - "dataPath": "params_shard_73.bin", + "dataPath": "params_shard_76.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_226", + "name": "param_235", "shape": [ 28672, 512 @@ -2942,12 +3060,12 @@ ] }, { - "dataPath": "params_shard_74.bin", + "dataPath": "params_shard_77.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_228", + "name": "param_237", "shape": [ 4096, 1792 @@ -2960,12 +3078,12 @@ ] }, { - "dataPath": "params_shard_75.bin", + "dataPath": "params_shard_78.bin", "format": "raw-shard", "nbytes": 22036480, "records": [ { - "name": "param_223", + "name": "param_232", "shape": [ 6144, 128 @@ -2976,7 +3094,7 @@ "byteOffset": 0 }, { - "name": "param_224", + "name": "param_233", "shape": [ 4096, 512 @@ -2987,7 +3105,7 @@ "byteOffset": 1572864 }, { - "name": "param_225", + "name": "param_234", "shape": [ 4096, 128 @@ -2998,7 +3116,7 @@ "byteOffset": 9961472 }, { - "name": "param_227", + "name": "param_236", "shape": [ 28672, 128 @@ -3009,7 +3127,7 @@ "byteOffset": 11010048 }, { - "name": "param_229", + "name": "param_238", "shape": [ 4096, 448 @@ -3020,7 +3138,7 @@ "byteOffset": 18350080 }, { - "name": "param_230", + "name": "param_239", "shape": [ 4096 ], @@ -3030,7 +3148,7 @@ "byteOffset": 22020096 }, { - "name": "param_231", + "name": "param_240", "shape": [ 4096 ], @@ -3042,12 +3160,12 @@ ] }, { - "dataPath": "params_shard_76.bin", + "dataPath": "params_shard_79.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_236", + "name": "param_245", "shape": [ 28672, 512 @@ -3060,12 +3178,12 @@ ] }, { - "dataPath": "params_shard_77.bin", + "dataPath": "params_shard_80.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_238", + "name": "param_247", "shape": [ 4096, 1792 @@ -3078,12 +3196,12 @@ ] }, { - "dataPath": "params_shard_78.bin", + "dataPath": "params_shard_81.bin", "format": "raw-shard", "nbytes": 30932992, "records": [ { - "name": "param_232", + "name": "param_241", "shape": [ 6144, 512 @@ -3094,7 +3212,7 @@ "byteOffset": 0 }, { - "name": "param_233", + "name": "param_242", "shape": [ 6144, 128 @@ -3105,7 +3223,7 @@ "byteOffset": 12582912 }, { - "name": "param_234", + "name": "param_243", "shape": [ 4096, 512 @@ -3116,7 +3234,7 @@ "byteOffset": 14155776 }, { - "name": "param_235", + "name": "param_244", "shape": [ 4096, 128 @@ -3127,7 +3245,7 @@ "byteOffset": 22544384 }, { - "name": "param_237", + "name": "param_246", "shape": [ 28672, 128 @@ -3140,12 +3258,12 @@ ] }, { - "dataPath": "params_shard_79.bin", + "dataPath": "params_shard_82.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_246", + "name": "param_255", "shape": [ 28672, 512 @@ -3158,12 +3276,12 @@ ] }, { - "dataPath": "params_shard_80.bin", + "dataPath": "params_shard_83.bin", "format": "raw-shard", "nbytes": 27279360, "records": [ { - "name": "param_239", + "name": "param_248", "shape": [ 4096, 448 @@ -3174,7 +3292,7 @@ "byteOffset": 0 }, { - "name": "param_240", + "name": "param_249", "shape": [ 4096 ], @@ -3184,7 +3302,7 @@ "byteOffset": 3670016 }, { - "name": "param_241", + "name": "param_250", "shape": [ 4096 ], @@ -3194,7 +3312,7 @@ "byteOffset": 3678208 }, { - "name": "param_242", + "name": "param_251", "shape": [ 6144, 512 @@ -3205,7 +3323,7 @@ "byteOffset": 3686400 }, { - "name": "param_243", + "name": "param_252", "shape": [ 6144, 128 @@ -3216,7 +3334,7 @@ "byteOffset": 16269312 }, { - "name": "param_244", + "name": "param_253", "shape": [ 4096, 512 @@ -3227,7 +3345,7 @@ "byteOffset": 17842176 }, { - "name": "param_245", + "name": "param_254", "shape": [ 4096, 128 @@ -3240,12 +3358,12 @@ ] }, { - "dataPath": "params_shard_81.bin", + "dataPath": "params_shard_84.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_248", + "name": "param_257", "shape": [ 4096, 1792 @@ -3258,12 +3376,12 @@ ] }, { - "dataPath": "params_shard_82.bin", + "dataPath": "params_shard_85.bin", "format": "raw-shard", "nbytes": 25182208, "records": [ { - "name": "param_247", + "name": "param_256", "shape": [ 28672, 128 @@ -3274,7 +3392,7 @@ "byteOffset": 0 }, { - "name": "param_249", + "name": "param_258", "shape": [ 4096, 448 @@ -3285,7 +3403,7 @@ "byteOffset": 7340032 }, { - "name": "param_250", + "name": "param_259", "shape": [ 4096 ], @@ -3295,7 +3413,7 @@ "byteOffset": 11010048 }, { - "name": "param_251", + "name": "param_260", "shape": [ 4096 ], @@ -3305,7 +3423,7 @@ "byteOffset": 11018240 }, { - "name": "param_252", + "name": "param_261", "shape": [ 6144, 512 @@ -3316,7 +3434,7 @@ "byteOffset": 11026432 }, { - "name": "param_253", + "name": "param_262", "shape": [ 6144, 128 @@ -3329,12 +3447,12 @@ ] }, { - "dataPath": "params_shard_83.bin", + "dataPath": "params_shard_86.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_256", + "name": "param_265", "shape": [ 28672, 512 @@ -3347,12 +3465,12 @@ ] }, { - "dataPath": "params_shard_84.bin", + "dataPath": "params_shard_87.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_258", + "name": "param_267", "shape": [ 4096, 1792 @@ -3365,12 +3483,12 @@ ] }, { - "dataPath": "params_shard_85.bin", + "dataPath": "params_shard_88.bin", "format": "raw-shard", "nbytes": 33046528, "records": [ { - "name": "param_254", + "name": "param_263", "shape": [ 4096, 512 @@ -3381,7 +3499,7 @@ "byteOffset": 0 }, { - "name": "param_255", + "name": "param_264", "shape": [ 4096, 128 @@ -3392,7 +3510,7 @@ "byteOffset": 8388608 }, { - "name": "param_257", + "name": "param_266", "shape": [ 28672, 128 @@ -3403,7 +3521,7 @@ "byteOffset": 9437184 }, { - "name": "param_259", + "name": "param_268", "shape": [ 4096, 448 @@ -3414,7 +3532,7 @@ "byteOffset": 16777216 }, { - "name": "param_260", + "name": "param_269", "shape": [ 4096 ], @@ -3424,7 +3542,7 @@ "byteOffset": 20447232 }, { - "name": "param_261", + "name": "param_270", "shape": [ 4096 ], @@ -3434,7 +3552,7 @@ "byteOffset": 20455424 }, { - "name": "param_262", + "name": "param_271", "shape": [ 6144, 512 @@ -3447,12 +3565,12 @@ ] }, { - "dataPath": "params_shard_86.bin", + "dataPath": "params_shard_89.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_266", + "name": "param_275", "shape": [ 28672, 512 @@ -3465,12 +3583,12 @@ ] }, { - "dataPath": "params_shard_87.bin", + "dataPath": "params_shard_90.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_268", + "name": "param_277", "shape": [ 4096, 1792 @@ -3483,12 +3601,12 @@ ] }, { - "dataPath": "params_shard_88.bin", + "dataPath": "params_shard_91.bin", "format": "raw-shard", "nbytes": 22036480, "records": [ { - "name": "param_263", + "name": "param_272", "shape": [ 6144, 128 @@ -3499,7 +3617,7 @@ "byteOffset": 0 }, { - "name": "param_264", + "name": "param_273", "shape": [ 4096, 512 @@ -3510,7 +3628,7 @@ "byteOffset": 1572864 }, { - "name": "param_265", + "name": "param_274", "shape": [ 4096, 128 @@ -3521,7 +3639,7 @@ "byteOffset": 9961472 }, { - "name": "param_267", + "name": "param_276", "shape": [ 28672, 128 @@ -3532,7 +3650,7 @@ "byteOffset": 11010048 }, { - "name": "param_269", + "name": "param_278", "shape": [ 4096, 448 @@ -3543,7 +3661,7 @@ "byteOffset": 18350080 }, { - "name": "param_270", + "name": "param_279", "shape": [ 4096 ], @@ -3553,7 +3671,7 @@ "byteOffset": 22020096 }, { - "name": "param_271", + "name": "param_280", "shape": [ 4096 ], @@ -3565,12 +3683,12 @@ ] }, { - "dataPath": "params_shard_89.bin", + "dataPath": "params_shard_92.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_276", + "name": "param_285", "shape": [ 28672, 512 @@ -3583,12 +3701,12 @@ ] }, { - "dataPath": "params_shard_90.bin", + "dataPath": "params_shard_93.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_278", + "name": "param_287", "shape": [ 4096, 1792 @@ -3601,12 +3719,12 @@ ] }, { - "dataPath": "params_shard_91.bin", + "dataPath": "params_shard_94.bin", "format": "raw-shard", "nbytes": 30932992, "records": [ { - "name": "param_272", + "name": "param_281", "shape": [ 6144, 512 @@ -3617,7 +3735,7 @@ "byteOffset": 0 }, { - "name": "param_273", + "name": "param_282", "shape": [ 6144, 128 @@ -3628,7 +3746,7 @@ "byteOffset": 12582912 }, { - "name": "param_274", + "name": "param_283", "shape": [ 4096, 512 @@ -3639,7 +3757,7 @@ "byteOffset": 14155776 }, { - "name": "param_275", + "name": "param_284", "shape": [ 4096, 128 @@ -3650,7 +3768,7 @@ "byteOffset": 22544384 }, { - "name": "param_277", + "name": "param_286", "shape": [ 28672, 128 @@ -3663,12 +3781,12 @@ ] }, { - "dataPath": "params_shard_92.bin", + "dataPath": "params_shard_95.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_286", + "name": "param_295", "shape": [ 28672, 512 @@ -3681,12 +3799,12 @@ ] }, { - "dataPath": "params_shard_93.bin", + "dataPath": "params_shard_96.bin", "format": "raw-shard", "nbytes": 27279360, "records": [ { - "name": "param_279", + "name": "param_288", "shape": [ 4096, 448 @@ -3697,7 +3815,7 @@ "byteOffset": 0 }, { - "name": "param_280", + "name": "param_289", "shape": [ 4096 ], @@ -3707,7 +3825,7 @@ "byteOffset": 3670016 }, { - "name": "param_281", + "name": "param_290", "shape": [ 4096 ], @@ -3717,7 +3835,7 @@ "byteOffset": 3678208 }, { - "name": "param_282", + "name": "param_291", "shape": [ 6144, 512 @@ -3728,7 +3846,7 @@ "byteOffset": 3686400 }, { - "name": "param_283", + "name": "param_292", "shape": [ 6144, 128 @@ -3739,7 +3857,7 @@ "byteOffset": 16269312 }, { - "name": "param_284", + "name": "param_293", "shape": [ 4096, 512 @@ -3750,7 +3868,7 @@ "byteOffset": 17842176 }, { - "name": "param_285", + "name": "param_294", "shape": [ 4096, 128 @@ -3763,12 +3881,12 @@ ] }, { - "dataPath": "params_shard_94.bin", + "dataPath": "params_shard_97.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_288", + "name": "param_297", "shape": [ 4096, 1792 @@ -3781,12 +3899,12 @@ ] }, { - "dataPath": "params_shard_95.bin", + "dataPath": "params_shard_98.bin", "format": "raw-shard", "nbytes": 25182208, "records": [ { - "name": "param_287", + "name": "param_296", "shape": [ 28672, 128 @@ -3797,7 +3915,7 @@ "byteOffset": 0 }, { - "name": "param_289", + "name": "param_298", "shape": [ 4096, 448 @@ -3808,7 +3926,7 @@ "byteOffset": 7340032 }, { - "name": "param_290", + "name": "param_299", "shape": [ 4096 ], @@ -3818,7 +3936,7 @@ "byteOffset": 11010048 }, { - "name": "param_291", + "name": "param_300", "shape": [ 4096 ], @@ -3828,7 +3946,7 @@ "byteOffset": 11018240 }, { - "name": "param_292", + "name": "param_301", "shape": [ 6144, 512 @@ -3839,7 +3957,7 @@ "byteOffset": 11026432 }, { - "name": "param_293", + "name": "param_302", "shape": [ 6144, 128 @@ -3852,12 +3970,12 @@ ] }, { - "dataPath": "params_shard_96.bin", + "dataPath": "params_shard_99.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_296", + "name": "param_305", "shape": [ 28672, 512 @@ -3870,12 +3988,12 @@ ] }, { - "dataPath": "params_shard_97.bin", + "dataPath": "params_shard_100.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_298", + "name": "param_307", "shape": [ 4096, 1792 @@ -3888,12 +4006,12 @@ ] }, { - "dataPath": "params_shard_98.bin", + "dataPath": "params_shard_101.bin", "format": "raw-shard", "nbytes": 33046528, "records": [ { - "name": "param_294", + "name": "param_303", "shape": [ 4096, 512 @@ -3904,7 +4022,7 @@ "byteOffset": 0 }, { - "name": "param_295", + "name": "param_304", "shape": [ 4096, 128 @@ -3915,7 +4033,7 @@ "byteOffset": 8388608 }, { - "name": "param_297", + "name": "param_306", "shape": [ 28672, 128 @@ -3926,7 +4044,7 @@ "byteOffset": 9437184 }, { - "name": "param_299", + "name": "param_308", "shape": [ 4096, 448 @@ -3937,7 +4055,7 @@ "byteOffset": 16777216 }, { - "name": "param_300", + "name": "param_309", "shape": [ 4096 ], @@ -3947,7 +4065,7 @@ "byteOffset": 20447232 }, { - "name": "param_301", + "name": "param_310", "shape": [ 4096 ], @@ -3957,7 +4075,7 @@ "byteOffset": 20455424 }, { - "name": "param_302", + "name": "param_311", "shape": [ 6144, 512 @@ -3970,12 +4088,12 @@ ] }, { - "dataPath": "params_shard_99.bin", + "dataPath": "params_shard_102.bin", "format": "raw-shard", "nbytes": 58720256, "records": [ { - "name": "param_306", + "name": "param_315", "shape": [ 28672, 512 @@ -3988,12 +4106,12 @@ ] }, { - "dataPath": "params_shard_100.bin", + "dataPath": "params_shard_103.bin", "format": "raw-shard", "nbytes": 29360128, "records": [ { - "name": "param_308", + "name": "param_317", "shape": [ 4096, 1792 @@ -4006,141 +4124,30 @@ ] }, { - "dataPath": "params_shard_101.bin", + "dataPath": "params_shard_104.bin", "format": "raw-shard", - "nbytes": 22036480, + "nbytes": 262144000, "records": [ { - "name": "param_303", - "shape": [ - 6144, - 128 - ], - "dtype": "float16", - "format": "raw", - "nbytes": 1572864, - "byteOffset": 0 - }, - { - "name": "param_304", - "shape": [ - 4096, - 512 - ], - "dtype": "uint32", - "format": "raw", - "nbytes": 8388608, - "byteOffset": 1572864 - }, - { - "name": "param_305", - "shape": [ - 4096, - 128 - ], - "dtype": "float16", - "format": "raw", - "nbytes": 1048576, - "byteOffset": 9961472 - }, - { - "name": "param_307", - "shape": [ - 28672, - 128 - ], - "dtype": "float16", - "format": "raw", - "nbytes": 7340032, - "byteOffset": 11010048 - }, - { - "name": "param_309", - "shape": [ - 4096, - 448 - ], - "dtype": "float16", - "format": "raw", - "nbytes": 3670016, - "byteOffset": 18350080 - }, - { - "name": "param_310", - "shape": [ - 4096 - ], - "dtype": "float16", - "format": "raw", - "nbytes": 8192, - "byteOffset": 22020096 - }, - { - "name": "param_311", + "name": "param_322", "shape": [ + 32000, 4096 ], "dtype": "float16", "format": "raw", - "nbytes": 8192, - "byteOffset": 22028288 - } - ] - }, - { - "dataPath": "params_shard_102.bin", - "format": "raw-shard", - "nbytes": 58720256, - "records": [ - { - "name": "param_316", - "shape": [ - 28672, - 512 - ], - "dtype": "uint32", - "format": "raw", - "nbytes": 58720256, - "byteOffset": 0 - } - ] - }, - { - "dataPath": "params_shard_103.bin", - "format": "raw-shard", - "nbytes": 29360128, - "records": [ - { - "name": "param_318", - "shape": [ - 4096, - 1792 - ], - "dtype": "uint32", - "format": "raw", - "nbytes": 29360128, + "nbytes": 262144000, "byteOffset": 0 } ] }, { - "dataPath": "params_shard_104.bin", + "dataPath": "params_shard_105.bin", "format": "raw-shard", - "nbytes": 30932992, + "nbytes": 23093248, "records": [ { "name": "param_312", - "shape": [ - 6144, - 512 - ], - "dtype": "uint32", - "format": "raw", - "nbytes": 12582912, - "byteOffset": 0 - }, - { - "name": "param_313", "shape": [ 6144, 128 @@ -4148,10 +4155,10 @@ "dtype": "float16", "format": "raw", "nbytes": 1572864, - "byteOffset": 12582912 + "byteOffset": 0 }, { - "name": "param_314", + "name": "param_313", "shape": [ 4096, 512 @@ -4159,10 +4166,10 @@ "dtype": "uint32", "format": "raw", "nbytes": 8388608, - "byteOffset": 14155776 + "byteOffset": 1572864 }, { - "name": "param_315", + "name": "param_314", "shape": [ 4096, 128 @@ -4170,10 +4177,10 @@ "dtype": "float16", "format": "raw", "nbytes": 1048576, - "byteOffset": 22544384 + "byteOffset": 9961472 }, { - "name": "param_317", + "name": "param_316", "shape": [ 28672, 128 @@ -4181,35 +4188,10 @@ "dtype": "float16", "format": "raw", "nbytes": 7340032, - "byteOffset": 23592960 - } - ] - }, - { - "dataPath": "params_shard_105.bin", - "format": "raw-shard", - "nbytes": 65536000, - "records": [ - { - "name": "param_323", - "shape": [ - 32000, - 512 - ], - "dtype": "uint32", - "format": "raw", - "nbytes": 65536000, - "byteOffset": 0 - } - ] - }, - { - "dataPath": "params_shard_106.bin", - "format": "raw-shard", - "nbytes": 12935168, - "records": [ + "byteOffset": 11010048 + }, { - "name": "param_319", + "name": "param_318", "shape": [ 4096, 448 @@ -4217,51 +4199,40 @@ "dtype": "float16", "format": "raw", "nbytes": 3670016, - "byteOffset": 0 + "byteOffset": 18350080 }, { - "name": "param_320", + "name": "param_319", "shape": [ 4096 ], "dtype": "float16", "format": "raw", "nbytes": 8192, - "byteOffset": 3670016 + "byteOffset": 22020096 }, { - "name": "param_321", + "name": "param_320", "shape": [ 4096 ], "dtype": "float16", "format": "raw", "nbytes": 8192, - "byteOffset": 3678208 + "byteOffset": 22028288 }, { - "name": "param_322", + "name": "param_321", "shape": [ 4096 ], "dtype": "float16", "format": "raw", "nbytes": 8192, - "byteOffset": 3686400 + "byteOffset": 22036480 }, { - "name": "param_324", - "shape": [ - 32000, - 128 - ], - "dtype": "float16", - "format": "raw", - "nbytes": 8192000, - "byteOffset": 3694592 - }, - { - "name": "param_325", + "name": "param_323", "shape": [ 2048, 128 @@ -4269,10 +4240,10 @@ "dtype": "float16", "format": "raw", "nbytes": 524288, - "byteOffset": 11886592 + "byteOffset": 22044672 }, { - "name": "param_326", + "name": "param_324", "shape": [ 2048, 128 @@ -4280,7 +4251,7 @@ "dtype": "float16", "format": "raw", "nbytes": 524288, - "byteOffset": 12410880 + "byteOffset": 22568960 } ] }