| { | |
| "best_metric": 0.9340733266061217, | |
| "best_model_checkpoint": "/mnt/data4_HDD_14TB/yang/voxceleb-checkpoints/etdnn/voxceleb1/pretrain/ce-len3-bs256-lr5e-4/checkpoint-5230", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 5230, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03824091778202677, | |
| "grad_norm": 10.999133110046387, | |
| "learning_rate": 1.9120458891013384e-05, | |
| "loss": 7.1473, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07648183556405354, | |
| "grad_norm": 10.200960159301758, | |
| "learning_rate": 3.824091778202677e-05, | |
| "loss": 7.1283, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1147227533460803, | |
| "grad_norm": 9.468878746032715, | |
| "learning_rate": 5.736137667304015e-05, | |
| "loss": 7.0847, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15296367112810708, | |
| "grad_norm": 8.736529350280762, | |
| "learning_rate": 7.648183556405354e-05, | |
| "loss": 7.0321, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19120458891013384, | |
| "grad_norm": 6.856561183929443, | |
| "learning_rate": 9.560229445506692e-05, | |
| "loss": 6.955, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2294455066921606, | |
| "grad_norm": 5.473419666290283, | |
| "learning_rate": 0.0001147227533460803, | |
| "loss": 6.8474, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2676864244741874, | |
| "grad_norm": 3.7059528827667236, | |
| "learning_rate": 0.0001338432122370937, | |
| "loss": 6.7002, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.30592734225621415, | |
| "grad_norm": 2.9125759601593018, | |
| "learning_rate": 0.00015296367112810707, | |
| "loss": 6.5368, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3441682600382409, | |
| "grad_norm": 2.5546936988830566, | |
| "learning_rate": 0.00017208413001912047, | |
| "loss": 6.3399, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3824091778202677, | |
| "grad_norm": 2.418347120285034, | |
| "learning_rate": 0.00019120458891013384, | |
| "loss": 6.1511, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.42065009560229444, | |
| "grad_norm": 2.2986104488372803, | |
| "learning_rate": 0.0002103250478011472, | |
| "loss": 6.015, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.4588910133843212, | |
| "grad_norm": 2.227661609649658, | |
| "learning_rate": 0.0002294455066921606, | |
| "loss": 5.8224, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.497131931166348, | |
| "grad_norm": 2.2191731929779053, | |
| "learning_rate": 0.00024856596558317403, | |
| "loss": 5.6838, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5353728489483748, | |
| "grad_norm": 2.2345361709594727, | |
| "learning_rate": 0.0002676864244741874, | |
| "loss": 5.5764, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.5736137667304015, | |
| "grad_norm": 2.2029287815093994, | |
| "learning_rate": 0.0002868068833652008, | |
| "loss": 5.4715, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6118546845124283, | |
| "grad_norm": 2.1917331218719482, | |
| "learning_rate": 0.00030592734225621415, | |
| "loss": 5.321, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6500956022944551, | |
| "grad_norm": 2.2598936557769775, | |
| "learning_rate": 0.0003250478011472275, | |
| "loss": 5.2218, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.6883365200764818, | |
| "grad_norm": 2.3335328102111816, | |
| "learning_rate": 0.00034416826003824094, | |
| "loss": 5.1449, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7265774378585086, | |
| "grad_norm": 2.3767151832580566, | |
| "learning_rate": 0.0003632887189292543, | |
| "loss": 5.0062, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.7648183556405354, | |
| "grad_norm": 2.357161521911621, | |
| "learning_rate": 0.0003824091778202677, | |
| "loss": 4.9352, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8030592734225621, | |
| "grad_norm": 2.271972179412842, | |
| "learning_rate": 0.00040152963671128105, | |
| "loss": 4.8959, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.8413001912045889, | |
| "grad_norm": 2.4624767303466797, | |
| "learning_rate": 0.0004206500956022944, | |
| "loss": 4.7951, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.8795411089866156, | |
| "grad_norm": 2.1839234828948975, | |
| "learning_rate": 0.00043977055449330785, | |
| "loss": 4.6734, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.9177820267686424, | |
| "grad_norm": 2.3169806003570557, | |
| "learning_rate": 0.0004588910133843212, | |
| "loss": 4.6458, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.9560229445506692, | |
| "grad_norm": 2.3231122493743896, | |
| "learning_rate": 0.0004780114722753346, | |
| "loss": 4.5657, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.994263862332696, | |
| "grad_norm": 2.34975266456604, | |
| "learning_rate": 0.0004971319311663481, | |
| "loss": 4.489, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.1722166162125799, | |
| "eval_loss": 4.208949089050293, | |
| "eval_runtime": 1092.4973, | |
| "eval_samples_per_second": 13.606, | |
| "eval_steps_per_second": 13.606, | |
| "step": 523 | |
| }, | |
| { | |
| "epoch": 1.0325047801147227, | |
| "grad_norm": 2.3869619369506836, | |
| "learning_rate": 0.0004981941788825154, | |
| "loss": 4.3496, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.0707456978967496, | |
| "grad_norm": 2.2834343910217285, | |
| "learning_rate": 0.0004960696834501806, | |
| "loss": 4.3108, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.1089866156787762, | |
| "grad_norm": 2.192180633544922, | |
| "learning_rate": 0.0004939451880178457, | |
| "loss": 4.229, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.147227533460803, | |
| "grad_norm": 2.161285161972046, | |
| "learning_rate": 0.0004918206925855109, | |
| "loss": 4.1804, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.1854684512428297, | |
| "grad_norm": 2.2634170055389404, | |
| "learning_rate": 0.0004896961971531761, | |
| "loss": 4.118, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.2237093690248566, | |
| "grad_norm": 2.1758434772491455, | |
| "learning_rate": 0.00048757170172084133, | |
| "loss": 4.0295, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.2619502868068833, | |
| "grad_norm": 2.25357985496521, | |
| "learning_rate": 0.0004854472062885065, | |
| "loss": 3.9559, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.3001912045889101, | |
| "grad_norm": 2.2077853679656982, | |
| "learning_rate": 0.0004833227108561717, | |
| "loss": 3.9295, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.338432122370937, | |
| "grad_norm": 2.28712797164917, | |
| "learning_rate": 0.0004811982154238369, | |
| "loss": 3.8221, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.3766730401529637, | |
| "grad_norm": 2.2988157272338867, | |
| "learning_rate": 0.000479073719991502, | |
| "loss": 3.7764, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.4149139579349903, | |
| "grad_norm": 2.391023874282837, | |
| "learning_rate": 0.0004769492245591672, | |
| "loss": 3.7518, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.4531548757170172, | |
| "grad_norm": 2.065288782119751, | |
| "learning_rate": 0.00047482472912683236, | |
| "loss": 3.7074, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.491395793499044, | |
| "grad_norm": 2.288686990737915, | |
| "learning_rate": 0.00047270023369449756, | |
| "loss": 3.6558, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.5296367112810707, | |
| "grad_norm": 2.228173017501831, | |
| "learning_rate": 0.00047057573826216276, | |
| "loss": 3.6038, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.5678776290630974, | |
| "grad_norm": 2.1894149780273438, | |
| "learning_rate": 0.0004684512428298279, | |
| "loss": 3.5405, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.6061185468451242, | |
| "grad_norm": 2.234192371368408, | |
| "learning_rate": 0.0004663267473974931, | |
| "loss": 3.4855, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.644359464627151, | |
| "grad_norm": 2.309387445449829, | |
| "learning_rate": 0.0004642022519651583, | |
| "loss": 3.4771, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.682600382409178, | |
| "grad_norm": 2.183579206466675, | |
| "learning_rate": 0.00046207775653282345, | |
| "loss": 3.3984, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.7208413001912046, | |
| "grad_norm": 2.286858558654785, | |
| "learning_rate": 0.00045995326110048865, | |
| "loss": 3.3318, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.7590822179732313, | |
| "grad_norm": 2.2087490558624268, | |
| "learning_rate": 0.00045782876566815385, | |
| "loss": 3.3256, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.7973231357552581, | |
| "grad_norm": 2.2571709156036377, | |
| "learning_rate": 0.000455704270235819, | |
| "loss": 3.2644, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.835564053537285, | |
| "grad_norm": 2.346696138381958, | |
| "learning_rate": 0.0004535797748034842, | |
| "loss": 3.2687, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.8738049713193117, | |
| "grad_norm": 2.309258460998535, | |
| "learning_rate": 0.0004514552793711494, | |
| "loss": 3.2346, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.9120458891013383, | |
| "grad_norm": 2.200920820236206, | |
| "learning_rate": 0.00044933078393881453, | |
| "loss": 3.1507, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.9502868068833652, | |
| "grad_norm": 2.1627931594848633, | |
| "learning_rate": 0.00044720628850647973, | |
| "loss": 3.0884, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.988527724665392, | |
| "grad_norm": 2.350935220718384, | |
| "learning_rate": 0.00044508179307414493, | |
| "loss": 3.0685, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.41096535486041036, | |
| "eval_loss": 2.762111186981201, | |
| "eval_runtime": 1228.094, | |
| "eval_samples_per_second": 12.104, | |
| "eval_steps_per_second": 12.104, | |
| "step": 1046 | |
| }, | |
| { | |
| "epoch": 2.026768642447419, | |
| "grad_norm": 2.137504816055298, | |
| "learning_rate": 0.0004429572976418101, | |
| "loss": 3.0334, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.0650095602294454, | |
| "grad_norm": 2.2636399269104004, | |
| "learning_rate": 0.0004408328022094752, | |
| "loss": 2.8843, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.1032504780114722, | |
| "grad_norm": 2.141150712966919, | |
| "learning_rate": 0.0004387083067771405, | |
| "loss": 2.909, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.141491395793499, | |
| "grad_norm": 2.243589401245117, | |
| "learning_rate": 0.0004365838113448056, | |
| "loss": 2.8407, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.179732313575526, | |
| "grad_norm": 2.3046114444732666, | |
| "learning_rate": 0.00043445931591247077, | |
| "loss": 2.8377, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.2179732313575524, | |
| "grad_norm": 2.375234603881836, | |
| "learning_rate": 0.000432334820480136, | |
| "loss": 2.7642, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.2562141491395793, | |
| "grad_norm": 2.36006236076355, | |
| "learning_rate": 0.00043021032504780116, | |
| "loss": 2.8011, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.294455066921606, | |
| "grad_norm": 2.246645450592041, | |
| "learning_rate": 0.0004280858296154663, | |
| "loss": 2.7502, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.332695984703633, | |
| "grad_norm": 2.420032501220703, | |
| "learning_rate": 0.0004259613341831315, | |
| "loss": 2.6646, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.3709369024856595, | |
| "grad_norm": 2.3049144744873047, | |
| "learning_rate": 0.0004238368387507967, | |
| "loss": 2.7294, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.4091778202676863, | |
| "grad_norm": 2.3154027462005615, | |
| "learning_rate": 0.00042171234331846185, | |
| "loss": 2.7323, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.447418738049713, | |
| "grad_norm": 2.3349056243896484, | |
| "learning_rate": 0.00041958784788612705, | |
| "loss": 2.6435, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.48565965583174, | |
| "grad_norm": 2.3369665145874023, | |
| "learning_rate": 0.00041746335245379225, | |
| "loss": 2.5938, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.5239005736137665, | |
| "grad_norm": 2.115408182144165, | |
| "learning_rate": 0.0004153388570214574, | |
| "loss": 2.5557, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.5621414913957934, | |
| "grad_norm": 2.2280750274658203, | |
| "learning_rate": 0.00041321436158912254, | |
| "loss": 2.5262, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.6003824091778203, | |
| "grad_norm": 2.3640174865722656, | |
| "learning_rate": 0.0004110898661567878, | |
| "loss": 2.5302, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.638623326959847, | |
| "grad_norm": 2.220531940460205, | |
| "learning_rate": 0.00040896537072445294, | |
| "loss": 2.4982, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 2.676864244741874, | |
| "grad_norm": 2.19985294342041, | |
| "learning_rate": 0.0004068408752921181, | |
| "loss": 2.4884, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.7151051625239004, | |
| "grad_norm": 2.083179235458374, | |
| "learning_rate": 0.00040471637985978334, | |
| "loss": 2.4422, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 2.7533460803059273, | |
| "grad_norm": 2.171346664428711, | |
| "learning_rate": 0.0004025918844274485, | |
| "loss": 2.4187, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.791586998087954, | |
| "grad_norm": 2.330702304840088, | |
| "learning_rate": 0.0004004673889951136, | |
| "loss": 2.3988, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 2.8298279158699806, | |
| "grad_norm": 2.3132684230804443, | |
| "learning_rate": 0.0003983428935627789, | |
| "loss": 2.3606, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 2.8680688336520075, | |
| "grad_norm": 2.188427686691284, | |
| "learning_rate": 0.000396218398130444, | |
| "loss": 2.3411, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.9063097514340344, | |
| "grad_norm": 2.1979410648345947, | |
| "learning_rate": 0.00039409390269810917, | |
| "loss": 2.3314, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 2.9445506692160612, | |
| "grad_norm": 2.246650218963623, | |
| "learning_rate": 0.0003919694072657744, | |
| "loss": 2.3221, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 2.982791586998088, | |
| "grad_norm": 2.2953240871429443, | |
| "learning_rate": 0.00038984491183343957, | |
| "loss": 2.2892, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6542885973763874, | |
| "eval_loss": 1.6626524925231934, | |
| "eval_runtime": 1164.8432, | |
| "eval_samples_per_second": 12.761, | |
| "eval_steps_per_second": 12.761, | |
| "step": 1569 | |
| }, | |
| { | |
| "epoch": 3.0210325047801145, | |
| "grad_norm": 2.2067298889160156, | |
| "learning_rate": 0.0003877204164011047, | |
| "loss": 2.2228, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 3.0592734225621414, | |
| "grad_norm": 2.2736258506774902, | |
| "learning_rate": 0.00038559592096876997, | |
| "loss": 2.1788, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.0975143403441683, | |
| "grad_norm": 2.324817419052124, | |
| "learning_rate": 0.0003834714255364351, | |
| "loss": 2.1743, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 3.135755258126195, | |
| "grad_norm": 2.188101291656494, | |
| "learning_rate": 0.00038134693010410026, | |
| "loss": 2.1267, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 3.173996175908222, | |
| "grad_norm": 2.2637600898742676, | |
| "learning_rate": 0.00037922243467176546, | |
| "loss": 2.0835, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 3.2122370936902485, | |
| "grad_norm": 2.2315452098846436, | |
| "learning_rate": 0.00037709793923943065, | |
| "loss": 2.0944, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 3.2504780114722753, | |
| "grad_norm": 2.236682176589966, | |
| "learning_rate": 0.0003749734438070958, | |
| "loss": 2.0867, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.288718929254302, | |
| "grad_norm": 2.1970863342285156, | |
| "learning_rate": 0.000372848948374761, | |
| "loss": 2.0138, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 3.3269598470363286, | |
| "grad_norm": 2.2857913970947266, | |
| "learning_rate": 0.0003707244529424262, | |
| "loss": 2.0081, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 3.3652007648183555, | |
| "grad_norm": 2.256211280822754, | |
| "learning_rate": 0.00036859995751009134, | |
| "loss": 1.9887, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 3.4034416826003824, | |
| "grad_norm": 2.3634960651397705, | |
| "learning_rate": 0.00036647546207775654, | |
| "loss": 1.9965, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 3.4416826003824093, | |
| "grad_norm": 2.2254064083099365, | |
| "learning_rate": 0.00036435096664542174, | |
| "loss": 2.0024, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.479923518164436, | |
| "grad_norm": 2.271703004837036, | |
| "learning_rate": 0.0003622264712130869, | |
| "loss": 2.0024, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 3.5181644359464626, | |
| "grad_norm": 2.220524549484253, | |
| "learning_rate": 0.0003601019757807521, | |
| "loss": 1.9561, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 3.5564053537284894, | |
| "grad_norm": 2.3010692596435547, | |
| "learning_rate": 0.0003579774803484173, | |
| "loss": 1.9641, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 3.5946462715105163, | |
| "grad_norm": 2.2860300540924072, | |
| "learning_rate": 0.00035585298491608243, | |
| "loss": 1.9032, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 3.632887189292543, | |
| "grad_norm": 2.2572293281555176, | |
| "learning_rate": 0.00035372848948374763, | |
| "loss": 1.9115, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.67112810707457, | |
| "grad_norm": 2.441771984100342, | |
| "learning_rate": 0.0003516039940514128, | |
| "loss": 1.8957, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 3.7093690248565965, | |
| "grad_norm": 2.3750736713409424, | |
| "learning_rate": 0.00034947949861907797, | |
| "loss": 1.8638, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 3.7476099426386233, | |
| "grad_norm": 2.114288806915283, | |
| "learning_rate": 0.00034735500318674317, | |
| "loss": 1.8105, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 3.78585086042065, | |
| "grad_norm": 2.3129258155822754, | |
| "learning_rate": 0.0003452305077544083, | |
| "loss": 1.8908, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 3.8240917782026767, | |
| "grad_norm": 2.217604398727417, | |
| "learning_rate": 0.0003431060123220735, | |
| "loss": 1.8116, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.8623326959847035, | |
| "grad_norm": 2.2592735290527344, | |
| "learning_rate": 0.0003409815168897387, | |
| "loss": 1.8325, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 3.9005736137667304, | |
| "grad_norm": 2.321389675140381, | |
| "learning_rate": 0.00033885702145740386, | |
| "loss": 1.7757, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 3.9388145315487573, | |
| "grad_norm": 2.0709521770477295, | |
| "learning_rate": 0.00033673252602506906, | |
| "loss": 1.7801, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 3.977055449330784, | |
| "grad_norm": 2.3316352367401123, | |
| "learning_rate": 0.00033460803059273426, | |
| "loss": 1.7576, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.758627648839556, | |
| "eval_loss": 1.1761366128921509, | |
| "eval_runtime": 910.1664, | |
| "eval_samples_per_second": 16.332, | |
| "eval_steps_per_second": 16.332, | |
| "step": 2092 | |
| }, | |
| { | |
| "epoch": 4.015296367112811, | |
| "grad_norm": 2.2446036338806152, | |
| "learning_rate": 0.0003324835351603994, | |
| "loss": 1.7157, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.053537284894838, | |
| "grad_norm": 2.1550064086914062, | |
| "learning_rate": 0.0003303590397280646, | |
| "loss": 1.6126, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 4.091778202676864, | |
| "grad_norm": 2.3542799949645996, | |
| "learning_rate": 0.0003282345442957298, | |
| "loss": 1.6417, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 4.130019120458891, | |
| "grad_norm": 2.3036603927612305, | |
| "learning_rate": 0.00032611004886339494, | |
| "loss": 1.6108, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 4.168260038240918, | |
| "grad_norm": 2.168532133102417, | |
| "learning_rate": 0.00032398555343106014, | |
| "loss": 1.5939, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 4.2065009560229445, | |
| "grad_norm": 2.257270097732544, | |
| "learning_rate": 0.0003218610579987253, | |
| "loss": 1.6157, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.244741873804971, | |
| "grad_norm": 2.2476625442504883, | |
| "learning_rate": 0.0003197365625663905, | |
| "loss": 1.6477, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 4.282982791586998, | |
| "grad_norm": 2.162182092666626, | |
| "learning_rate": 0.00031761206713405563, | |
| "loss": 1.5589, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 4.321223709369025, | |
| "grad_norm": 2.084498167037964, | |
| "learning_rate": 0.00031548757170172083, | |
| "loss": 1.5683, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 4.359464627151052, | |
| "grad_norm": 2.1769893169403076, | |
| "learning_rate": 0.00031336307626938603, | |
| "loss": 1.5324, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 4.397705544933078, | |
| "grad_norm": 2.1961722373962402, | |
| "learning_rate": 0.0003112385808370512, | |
| "loss": 1.5551, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.435946462715105, | |
| "grad_norm": 2.2195024490356445, | |
| "learning_rate": 0.0003091140854047164, | |
| "loss": 1.5724, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 4.474187380497132, | |
| "grad_norm": 2.18676495552063, | |
| "learning_rate": 0.0003069895899723816, | |
| "loss": 1.5322, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 4.512428298279159, | |
| "grad_norm": 2.2305195331573486, | |
| "learning_rate": 0.0003048650945400467, | |
| "loss": 1.5207, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 4.550669216061186, | |
| "grad_norm": 2.297010898590088, | |
| "learning_rate": 0.0003027405991077119, | |
| "loss": 1.5712, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 4.588910133843212, | |
| "grad_norm": 2.261610746383667, | |
| "learning_rate": 0.0003006161036753771, | |
| "loss": 1.513, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.627151051625239, | |
| "grad_norm": 2.1144425868988037, | |
| "learning_rate": 0.00029849160824304226, | |
| "loss": 1.4854, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 4.665391969407266, | |
| "grad_norm": 2.344909191131592, | |
| "learning_rate": 0.00029636711281070746, | |
| "loss": 1.5025, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 4.7036328871892925, | |
| "grad_norm": 2.1943275928497314, | |
| "learning_rate": 0.00029424261737837266, | |
| "loss": 1.4917, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 4.741873804971319, | |
| "grad_norm": 2.191070079803467, | |
| "learning_rate": 0.0002921181219460378, | |
| "loss": 1.4539, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 4.780114722753346, | |
| "grad_norm": 2.142045259475708, | |
| "learning_rate": 0.000289993626513703, | |
| "loss": 1.4688, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.818355640535373, | |
| "grad_norm": 2.203152656555176, | |
| "learning_rate": 0.0002878691310813682, | |
| "loss": 1.4236, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 4.8565965583174, | |
| "grad_norm": 2.264139175415039, | |
| "learning_rate": 0.00028574463564903335, | |
| "loss": 1.4211, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 4.894837476099426, | |
| "grad_norm": 2.173497438430786, | |
| "learning_rate": 0.0002836201402166985, | |
| "loss": 1.4236, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 4.933078393881453, | |
| "grad_norm": 2.069350481033325, | |
| "learning_rate": 0.00028149564478436375, | |
| "loss": 1.3804, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 4.97131931166348, | |
| "grad_norm": 2.1907660961151123, | |
| "learning_rate": 0.0002793711493520289, | |
| "loss": 1.3706, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8203834510595358, | |
| "eval_loss": 0.8903235793113708, | |
| "eval_runtime": 813.2495, | |
| "eval_samples_per_second": 18.279, | |
| "eval_steps_per_second": 18.279, | |
| "step": 2615 | |
| }, | |
| { | |
| "epoch": 5.009560229445507, | |
| "grad_norm": 2.2395236492156982, | |
| "learning_rate": 0.00027724665391969404, | |
| "loss": 1.3611, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 5.047801147227533, | |
| "grad_norm": 2.164022922515869, | |
| "learning_rate": 0.0002751221584873593, | |
| "loss": 1.2885, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 5.08604206500956, | |
| "grad_norm": 2.335524559020996, | |
| "learning_rate": 0.00027299766305502443, | |
| "loss": 1.3024, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 5.124282982791587, | |
| "grad_norm": 2.3510003089904785, | |
| "learning_rate": 0.0002708731676226896, | |
| "loss": 1.2819, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 5.162523900573614, | |
| "grad_norm": 2.2182960510253906, | |
| "learning_rate": 0.00026874867219035483, | |
| "loss": 1.2927, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.2007648183556405, | |
| "grad_norm": 2.164422035217285, | |
| "learning_rate": 0.00026662417675802, | |
| "loss": 1.2905, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 5.239005736137667, | |
| "grad_norm": 2.2719788551330566, | |
| "learning_rate": 0.0002644996813256851, | |
| "loss": 1.3067, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 5.277246653919694, | |
| "grad_norm": 2.1733806133270264, | |
| "learning_rate": 0.0002623751858933504, | |
| "loss": 1.2778, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 5.315487571701721, | |
| "grad_norm": 2.0940115451812744, | |
| "learning_rate": 0.0002602506904610155, | |
| "loss": 1.2553, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 5.353728489483748, | |
| "grad_norm": 2.049928665161133, | |
| "learning_rate": 0.00025812619502868067, | |
| "loss": 1.2572, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.3919694072657744, | |
| "grad_norm": 2.039212703704834, | |
| "learning_rate": 0.0002560016995963459, | |
| "loss": 1.2139, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 5.430210325047801, | |
| "grad_norm": 2.2491774559020996, | |
| "learning_rate": 0.00025387720416401106, | |
| "loss": 1.2591, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 5.468451242829828, | |
| "grad_norm": 2.1145877838134766, | |
| "learning_rate": 0.0002517527087316762, | |
| "loss": 1.2458, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 5.506692160611855, | |
| "grad_norm": 2.1274683475494385, | |
| "learning_rate": 0.0002496282132993414, | |
| "loss": 1.227, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 5.544933078393882, | |
| "grad_norm": 2.0846710205078125, | |
| "learning_rate": 0.0002475037178670066, | |
| "loss": 1.1897, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.583173996175908, | |
| "grad_norm": 2.143688678741455, | |
| "learning_rate": 0.0002453792224346718, | |
| "loss": 1.2133, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 5.621414913957935, | |
| "grad_norm": 2.2568676471710205, | |
| "learning_rate": 0.00024325472700233695, | |
| "loss": 1.2194, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 5.659655831739962, | |
| "grad_norm": 2.266052484512329, | |
| "learning_rate": 0.00024113023157000212, | |
| "loss": 1.1895, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 5.6978967495219885, | |
| "grad_norm": 2.352609395980835, | |
| "learning_rate": 0.0002390057361376673, | |
| "loss": 1.1849, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 5.736137667304015, | |
| "grad_norm": 2.1269915103912354, | |
| "learning_rate": 0.0002368812407053325, | |
| "loss": 1.205, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.774378585086042, | |
| "grad_norm": 2.20017147064209, | |
| "learning_rate": 0.00023475674527299767, | |
| "loss": 1.1366, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 5.812619502868069, | |
| "grad_norm": 2.0107390880584717, | |
| "learning_rate": 0.00023263224984066284, | |
| "loss": 1.1268, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 5.850860420650095, | |
| "grad_norm": 2.091771364212036, | |
| "learning_rate": 0.00023050775440832804, | |
| "loss": 1.1619, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 5.8891013384321225, | |
| "grad_norm": 2.261592388153076, | |
| "learning_rate": 0.0002283832589759932, | |
| "loss": 1.1789, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 5.927342256214149, | |
| "grad_norm": 2.126786947250366, | |
| "learning_rate": 0.00022625876354365838, | |
| "loss": 1.1492, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.965583173996176, | |
| "grad_norm": 2.123417615890503, | |
| "learning_rate": 0.00022413426811132355, | |
| "loss": 1.1258, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8432559704002691, | |
| "eval_loss": 0.7555378079414368, | |
| "eval_runtime": 1118.6028, | |
| "eval_samples_per_second": 13.289, | |
| "eval_steps_per_second": 13.289, | |
| "step": 3138 | |
| }, | |
| { | |
| "epoch": 6.003824091778203, | |
| "grad_norm": 2.0020689964294434, | |
| "learning_rate": 0.00022200977267898875, | |
| "loss": 1.1086, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 6.042065009560229, | |
| "grad_norm": 2.0867371559143066, | |
| "learning_rate": 0.00021988527724665392, | |
| "loss": 1.0815, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 6.080305927342256, | |
| "grad_norm": 1.9533103704452515, | |
| "learning_rate": 0.0002177607818143191, | |
| "loss": 1.0412, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 6.118546845124283, | |
| "grad_norm": 2.2522337436676025, | |
| "learning_rate": 0.0002156362863819843, | |
| "loss": 1.0312, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.15678776290631, | |
| "grad_norm": 2.058983087539673, | |
| "learning_rate": 0.00021351179094964944, | |
| "loss": 1.0449, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 6.195028680688337, | |
| "grad_norm": 2.1774373054504395, | |
| "learning_rate": 0.00021138729551731464, | |
| "loss": 1.0532, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 6.233269598470363, | |
| "grad_norm": 2.0248055458068848, | |
| "learning_rate": 0.00020926280008497984, | |
| "loss": 1.0461, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 6.27151051625239, | |
| "grad_norm": 2.066737651824951, | |
| "learning_rate": 0.00020713830465264498, | |
| "loss": 1.0243, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 6.309751434034417, | |
| "grad_norm": 2.100133180618286, | |
| "learning_rate": 0.00020501380922031018, | |
| "loss": 1.0309, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.347992351816444, | |
| "grad_norm": 2.192960739135742, | |
| "learning_rate": 0.00020288931378797538, | |
| "loss": 1.0147, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 6.3862332695984705, | |
| "grad_norm": 2.123382329940796, | |
| "learning_rate": 0.00020076481835564053, | |
| "loss": 1.0229, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 6.424474187380497, | |
| "grad_norm": 2.162189245223999, | |
| "learning_rate": 0.00019864032292330573, | |
| "loss": 1.011, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 6.462715105162524, | |
| "grad_norm": 2.254284143447876, | |
| "learning_rate": 0.0001965158274909709, | |
| "loss": 0.9929, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 6.500956022944551, | |
| "grad_norm": 2.2183146476745605, | |
| "learning_rate": 0.00019439133205863607, | |
| "loss": 0.9852, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.539196940726577, | |
| "grad_norm": 2.1704065799713135, | |
| "learning_rate": 0.00019226683662630127, | |
| "loss": 1.0129, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 6.577437858508604, | |
| "grad_norm": 2.0828094482421875, | |
| "learning_rate": 0.00019014234119396644, | |
| "loss": 0.9667, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 6.615678776290631, | |
| "grad_norm": 2.3150341510772705, | |
| "learning_rate": 0.0001880178457616316, | |
| "loss": 0.9769, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 6.653919694072657, | |
| "grad_norm": 2.0636775493621826, | |
| "learning_rate": 0.0001858933503292968, | |
| "loss": 0.979, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 6.692160611854685, | |
| "grad_norm": 2.171602487564087, | |
| "learning_rate": 0.00018376885489696196, | |
| "loss": 0.9667, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.730401529636711, | |
| "grad_norm": 2.1150522232055664, | |
| "learning_rate": 0.00018164435946462716, | |
| "loss": 0.9593, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 6.768642447418738, | |
| "grad_norm": 2.096452236175537, | |
| "learning_rate": 0.00017951986403229233, | |
| "loss": 0.9828, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 6.806883365200765, | |
| "grad_norm": 2.2779390811920166, | |
| "learning_rate": 0.0001773953685999575, | |
| "loss": 0.9648, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 6.845124282982791, | |
| "grad_norm": 2.2793240547180176, | |
| "learning_rate": 0.0001752708731676227, | |
| "loss": 0.9639, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 6.8833652007648185, | |
| "grad_norm": 2.0822837352752686, | |
| "learning_rate": 0.00017314637773528787, | |
| "loss": 0.9421, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.921606118546845, | |
| "grad_norm": 2.1541502475738525, | |
| "learning_rate": 0.00017102188230295304, | |
| "loss": 0.9277, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 6.959847036328872, | |
| "grad_norm": 2.1241962909698486, | |
| "learning_rate": 0.00016889738687061824, | |
| "loss": 0.9335, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 6.998087954110899, | |
| "grad_norm": 2.2354745864868164, | |
| "learning_rate": 0.00016677289143828341, | |
| "loss": 0.9379, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.889673730238816, | |
| "eval_loss": 0.5586521029472351, | |
| "eval_runtime": 904.3879, | |
| "eval_samples_per_second": 16.437, | |
| "eval_steps_per_second": 16.437, | |
| "step": 3661 | |
| }, | |
| { | |
| "epoch": 7.036328871892925, | |
| "grad_norm": 2.109501600265503, | |
| "learning_rate": 0.0001646483960059486, | |
| "loss": 0.8485, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 7.074569789674952, | |
| "grad_norm": 2.2034125328063965, | |
| "learning_rate": 0.00016252390057361376, | |
| "loss": 0.897, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 7.112810707456979, | |
| "grad_norm": 2.053767204284668, | |
| "learning_rate": 0.00016039940514127896, | |
| "loss": 0.859, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 7.151051625239006, | |
| "grad_norm": 1.977229118347168, | |
| "learning_rate": 0.00015827490970894413, | |
| "loss": 0.8643, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 7.189292543021033, | |
| "grad_norm": 2.0782172679901123, | |
| "learning_rate": 0.0001561504142766093, | |
| "loss": 0.8665, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 7.227533460803059, | |
| "grad_norm": 1.9473881721496582, | |
| "learning_rate": 0.0001540259188442745, | |
| "loss": 0.8748, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 7.265774378585086, | |
| "grad_norm": 2.0109825134277344, | |
| "learning_rate": 0.00015190142341193967, | |
| "loss": 0.8589, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.304015296367113, | |
| "grad_norm": 2.1534509658813477, | |
| "learning_rate": 0.00014977692797960484, | |
| "loss": 0.8373, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 7.342256214149139, | |
| "grad_norm": 2.1457502841949463, | |
| "learning_rate": 0.00014765243254727004, | |
| "loss": 0.8487, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 7.3804971319311665, | |
| "grad_norm": 2.200446844100952, | |
| "learning_rate": 0.0001455279371149352, | |
| "loss": 0.841, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 7.418738049713193, | |
| "grad_norm": 2.0335378646850586, | |
| "learning_rate": 0.0001434034416826004, | |
| "loss": 0.846, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 7.45697896749522, | |
| "grad_norm": 2.1471550464630127, | |
| "learning_rate": 0.0001412789462502656, | |
| "loss": 0.8442, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.495219885277247, | |
| "grad_norm": 2.2554261684417725, | |
| "learning_rate": 0.00013915445081793073, | |
| "loss": 0.8651, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 7.533460803059273, | |
| "grad_norm": 2.1498897075653076, | |
| "learning_rate": 0.00013702995538559593, | |
| "loss": 0.839, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 7.5717017208413, | |
| "grad_norm": 2.1460678577423096, | |
| "learning_rate": 0.00013490545995326113, | |
| "loss": 0.8449, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 7.609942638623327, | |
| "grad_norm": 2.194730281829834, | |
| "learning_rate": 0.00013278096452092628, | |
| "loss": 0.8075, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 7.648183556405353, | |
| "grad_norm": 2.291327714920044, | |
| "learning_rate": 0.00013065646908859147, | |
| "loss": 0.8163, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.686424474187381, | |
| "grad_norm": 2.049647808074951, | |
| "learning_rate": 0.00012853197365625662, | |
| "loss": 0.7999, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 7.724665391969407, | |
| "grad_norm": 2.026001214981079, | |
| "learning_rate": 0.00012640747822392182, | |
| "loss": 0.7721, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 7.762906309751434, | |
| "grad_norm": 2.1242027282714844, | |
| "learning_rate": 0.00012428298279158702, | |
| "loss": 0.7957, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 7.801147227533461, | |
| "grad_norm": 2.078123092651367, | |
| "learning_rate": 0.0001221584873592522, | |
| "loss": 0.8244, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 7.839388145315487, | |
| "grad_norm": 2.028822183609009, | |
| "learning_rate": 0.00012003399192691736, | |
| "loss": 0.8066, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.8776290630975145, | |
| "grad_norm": 2.30352520942688, | |
| "learning_rate": 0.00011790949649458253, | |
| "loss": 0.7892, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 7.915869980879541, | |
| "grad_norm": 2.100221872329712, | |
| "learning_rate": 0.00011578500106224772, | |
| "loss": 0.7703, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 7.954110898661568, | |
| "grad_norm": 2.1163010597229004, | |
| "learning_rate": 0.00011366050562991289, | |
| "loss": 0.7958, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 7.992351816443595, | |
| "grad_norm": 2.194089412689209, | |
| "learning_rate": 0.00011153601019757808, | |
| "loss": 0.7925, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9117389841910528, | |
| "eval_loss": 0.4518139362335205, | |
| "eval_runtime": 1278.572, | |
| "eval_samples_per_second": 11.626, | |
| "eval_steps_per_second": 11.626, | |
| "step": 4184 | |
| }, | |
| { | |
| "epoch": 8.030592734225621, | |
| "grad_norm": 2.024029016494751, | |
| "learning_rate": 0.00010941151476524326, | |
| "loss": 0.7768, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 8.068833652007648, | |
| "grad_norm": 2.101900577545166, | |
| "learning_rate": 0.00010728701933290843, | |
| "loss": 0.7174, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 8.107074569789676, | |
| "grad_norm": 1.8774019479751587, | |
| "learning_rate": 0.0001051625239005736, | |
| "loss": 0.7331, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 8.145315487571702, | |
| "grad_norm": 2.0548768043518066, | |
| "learning_rate": 0.0001030380284682388, | |
| "loss": 0.6988, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 8.183556405353729, | |
| "grad_norm": 2.09981107711792, | |
| "learning_rate": 0.00010091353303590398, | |
| "loss": 0.7329, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 8.221797323135755, | |
| "grad_norm": 1.9958146810531616, | |
| "learning_rate": 9.878903760356915e-05, | |
| "loss": 0.7085, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 8.260038240917781, | |
| "grad_norm": 1.9749237298965454, | |
| "learning_rate": 9.666454217123433e-05, | |
| "loss": 0.715, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 8.29827915869981, | |
| "grad_norm": 1.9691994190216064, | |
| "learning_rate": 9.454004673889952e-05, | |
| "loss": 0.7151, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 8.336520076481836, | |
| "grad_norm": 2.1549344062805176, | |
| "learning_rate": 9.241555130656469e-05, | |
| "loss": 0.7249, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 8.374760994263863, | |
| "grad_norm": 2.2257604598999023, | |
| "learning_rate": 9.029105587422986e-05, | |
| "loss": 0.7207, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 8.413001912045889, | |
| "grad_norm": 2.0424046516418457, | |
| "learning_rate": 8.816656044189505e-05, | |
| "loss": 0.6981, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.451242829827915, | |
| "grad_norm": 1.8724883794784546, | |
| "learning_rate": 8.604206500956024e-05, | |
| "loss": 0.7153, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 8.489483747609942, | |
| "grad_norm": 2.0893712043762207, | |
| "learning_rate": 8.391756957722541e-05, | |
| "loss": 0.7254, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 8.52772466539197, | |
| "grad_norm": 2.0398759841918945, | |
| "learning_rate": 8.179307414489059e-05, | |
| "loss": 0.7184, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 8.565965583173996, | |
| "grad_norm": 2.1322920322418213, | |
| "learning_rate": 7.966857871255577e-05, | |
| "loss": 0.7189, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 8.604206500956023, | |
| "grad_norm": 2.0477068424224854, | |
| "learning_rate": 7.754408328022095e-05, | |
| "loss": 0.705, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.64244741873805, | |
| "grad_norm": 2.0737295150756836, | |
| "learning_rate": 7.541958784788614e-05, | |
| "loss": 0.7285, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 8.680688336520076, | |
| "grad_norm": 1.9542038440704346, | |
| "learning_rate": 7.329509241555131e-05, | |
| "loss": 0.7028, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 8.718929254302104, | |
| "grad_norm": 1.9409737586975098, | |
| "learning_rate": 7.117059698321648e-05, | |
| "loss": 0.7031, | |
| "step": 4560 | |
| }, | |
| { | |
| "epoch": 8.75717017208413, | |
| "grad_norm": 1.982315182685852, | |
| "learning_rate": 6.904610155088168e-05, | |
| "loss": 0.6909, | |
| "step": 4580 | |
| }, | |
| { | |
| "epoch": 8.795411089866157, | |
| "grad_norm": 2.2722830772399902, | |
| "learning_rate": 6.692160611854685e-05, | |
| "loss": 0.7083, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.833652007648183, | |
| "grad_norm": 1.9158116579055786, | |
| "learning_rate": 6.479711068621202e-05, | |
| "loss": 0.6885, | |
| "step": 4620 | |
| }, | |
| { | |
| "epoch": 8.87189292543021, | |
| "grad_norm": 2.058122396469116, | |
| "learning_rate": 6.26726152538772e-05, | |
| "loss": 0.6912, | |
| "step": 4640 | |
| }, | |
| { | |
| "epoch": 8.910133843212238, | |
| "grad_norm": 2.013585090637207, | |
| "learning_rate": 6.054811982154239e-05, | |
| "loss": 0.6937, | |
| "step": 4660 | |
| }, | |
| { | |
| "epoch": 8.948374760994264, | |
| "grad_norm": 2.1662707328796387, | |
| "learning_rate": 5.8423624389207567e-05, | |
| "loss": 0.7003, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 8.98661567877629, | |
| "grad_norm": 2.06309175491333, | |
| "learning_rate": 5.6299128956872745e-05, | |
| "loss": 0.6733, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.929297006390851, | |
| "eval_loss": 0.38889044523239136, | |
| "eval_runtime": 1336.4014, | |
| "eval_samples_per_second": 11.123, | |
| "eval_steps_per_second": 11.123, | |
| "step": 4707 | |
| }, | |
| { | |
| "epoch": 9.024856596558317, | |
| "grad_norm": 1.8986388444900513, | |
| "learning_rate": 5.4174633524537924e-05, | |
| "loss": 0.6593, | |
| "step": 4720 | |
| }, | |
| { | |
| "epoch": 9.063097514340344, | |
| "grad_norm": 1.9107776880264282, | |
| "learning_rate": 5.20501380922031e-05, | |
| "loss": 0.6354, | |
| "step": 4740 | |
| }, | |
| { | |
| "epoch": 9.101338432122372, | |
| "grad_norm": 2.0714404582977295, | |
| "learning_rate": 4.992564265986829e-05, | |
| "loss": 0.6531, | |
| "step": 4760 | |
| }, | |
| { | |
| "epoch": 9.139579349904398, | |
| "grad_norm": 1.9571270942687988, | |
| "learning_rate": 4.780114722753346e-05, | |
| "loss": 0.6429, | |
| "step": 4780 | |
| }, | |
| { | |
| "epoch": 9.177820267686425, | |
| "grad_norm": 1.9968469142913818, | |
| "learning_rate": 4.5676651795198646e-05, | |
| "loss": 0.6498, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 9.216061185468451, | |
| "grad_norm": 1.982832670211792, | |
| "learning_rate": 4.355215636286382e-05, | |
| "loss": 0.6141, | |
| "step": 4820 | |
| }, | |
| { | |
| "epoch": 9.254302103250478, | |
| "grad_norm": 2.035123825073242, | |
| "learning_rate": 4.1427660930529e-05, | |
| "loss": 0.6269, | |
| "step": 4840 | |
| }, | |
| { | |
| "epoch": 9.292543021032504, | |
| "grad_norm": 1.8670984506607056, | |
| "learning_rate": 3.930316549819418e-05, | |
| "loss": 0.6143, | |
| "step": 4860 | |
| }, | |
| { | |
| "epoch": 9.330783938814532, | |
| "grad_norm": 1.9521456956863403, | |
| "learning_rate": 3.7178670065859354e-05, | |
| "loss": 0.6352, | |
| "step": 4880 | |
| }, | |
| { | |
| "epoch": 9.369024856596559, | |
| "grad_norm": 1.9804260730743408, | |
| "learning_rate": 3.505417463352454e-05, | |
| "loss": 0.6214, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 9.407265774378585, | |
| "grad_norm": 2.149338483810425, | |
| "learning_rate": 3.292967920118971e-05, | |
| "loss": 0.655, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 9.445506692160611, | |
| "grad_norm": 1.8145477771759033, | |
| "learning_rate": 3.08051837688549e-05, | |
| "loss": 0.6095, | |
| "step": 4940 | |
| }, | |
| { | |
| "epoch": 9.483747609942638, | |
| "grad_norm": 2.009587287902832, | |
| "learning_rate": 2.8680688336520076e-05, | |
| "loss": 0.6314, | |
| "step": 4960 | |
| }, | |
| { | |
| "epoch": 9.521988527724666, | |
| "grad_norm": 1.946964979171753, | |
| "learning_rate": 2.655619290418526e-05, | |
| "loss": 0.6272, | |
| "step": 4980 | |
| }, | |
| { | |
| "epoch": 9.560229445506693, | |
| "grad_norm": 2.2486774921417236, | |
| "learning_rate": 2.4431697471850437e-05, | |
| "loss": 0.6133, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.598470363288719, | |
| "grad_norm": 1.9570778608322144, | |
| "learning_rate": 2.2307202039515616e-05, | |
| "loss": 0.6185, | |
| "step": 5020 | |
| }, | |
| { | |
| "epoch": 9.636711281070745, | |
| "grad_norm": 1.8994139432907104, | |
| "learning_rate": 2.0182706607180795e-05, | |
| "loss": 0.6366, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 9.674952198852772, | |
| "grad_norm": 2.102374792098999, | |
| "learning_rate": 1.8058211174845977e-05, | |
| "loss": 0.6414, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 9.7131931166348, | |
| "grad_norm": 2.073791027069092, | |
| "learning_rate": 1.5933715742511156e-05, | |
| "loss": 0.6053, | |
| "step": 5080 | |
| }, | |
| { | |
| "epoch": 9.751434034416826, | |
| "grad_norm": 1.9205766916275024, | |
| "learning_rate": 1.3809220310176333e-05, | |
| "loss": 0.6032, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 9.789674952198853, | |
| "grad_norm": 2.04294490814209, | |
| "learning_rate": 1.1684724877841512e-05, | |
| "loss": 0.6202, | |
| "step": 5120 | |
| }, | |
| { | |
| "epoch": 9.82791586998088, | |
| "grad_norm": 1.9936119318008423, | |
| "learning_rate": 9.560229445506692e-06, | |
| "loss": 0.5882, | |
| "step": 5140 | |
| }, | |
| { | |
| "epoch": 9.866156787762906, | |
| "grad_norm": 1.8728615045547485, | |
| "learning_rate": 7.435734013171872e-06, | |
| "loss": 0.5955, | |
| "step": 5160 | |
| }, | |
| { | |
| "epoch": 9.904397705544934, | |
| "grad_norm": 2.069910764694214, | |
| "learning_rate": 5.311238580837051e-06, | |
| "loss": 0.6049, | |
| "step": 5180 | |
| }, | |
| { | |
| "epoch": 9.94263862332696, | |
| "grad_norm": 1.9873836040496826, | |
| "learning_rate": 3.1867431485022306e-06, | |
| "loss": 0.6251, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 9.980879541108987, | |
| "grad_norm": 1.932947039604187, | |
| "learning_rate": 1.0622477161674102e-06, | |
| "loss": 0.6187, | |
| "step": 5220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9340733266061217, | |
| "eval_loss": 0.35941872000694275, | |
| "eval_runtime": 1417.2186, | |
| "eval_samples_per_second": 10.489, | |
| "eval_steps_per_second": 10.489, | |
| "step": 5230 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 5230, | |
| "total_flos": 2.7283641066057605e+18, | |
| "train_loss": 1.9864544938672797, | |
| "train_runtime": 94798.7749, | |
| "train_samples_per_second": 14.112, | |
| "train_steps_per_second": 0.055 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 5230, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.7283641066057605e+18, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |