{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4975553382522891,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008889679082585119,
      "grad_norm": 4.238861560821533,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 1.7261,
      "step": 10
    },
    {
      "epoch": 0.0017779358165170237,
      "grad_norm": 2.0652520656585693,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 1.6328,
      "step": 20
    },
    {
      "epoch": 0.0026669037247755356,
      "grad_norm": 1.8617759943008423,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.3995,
      "step": 30
    },
    {
      "epoch": 0.0035558716330340474,
      "grad_norm": 1.4548546075820923,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.3222,
      "step": 40
    },
    {
      "epoch": 0.004444839541292559,
      "grad_norm": 0.696451723575592,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 1.2108,
      "step": 50
    },
    {
      "epoch": 0.005333807449551071,
      "grad_norm": 0.679622232913971,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 1.1247,
      "step": 60
    },
    {
      "epoch": 0.006222775357809583,
      "grad_norm": 0.6772049069404602,
      "learning_rate": 7.814044943820226e-06,
      "loss": 1.0773,
      "step": 70
    },
    {
      "epoch": 0.007111743266068095,
      "grad_norm": 0.5850681662559509,
      "learning_rate": 8.932022471910112e-06,
      "loss": 1.0619,
      "step": 80
    },
    {
      "epoch": 0.008000711174326608,
      "grad_norm": 0.595498263835907,
      "learning_rate": 1.005e-05,
      "loss": 1.0314,
      "step": 90
    },
    {
      "epoch": 0.008889679082585119,
      "grad_norm": 0.533685028553009,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 1.0384,
      "step": 100
    },
    {
      "epoch": 0.009778646990843631,
      "grad_norm": 0.5636975765228271,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 1.055,
      "step": 110
    },
    {
      "epoch": 0.010667614899102142,
      "grad_norm": 0.523317277431488,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 1.051,
      "step": 120
    },
    {
      "epoch": 0.011556582807360655,
      "grad_norm": 0.548605740070343,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 1.0005,
      "step": 130
    },
    {
      "epoch": 0.012445550715619166,
      "grad_norm": 0.556210458278656,
      "learning_rate": 1.563988764044944e-05,
      "loss": 1.0724,
      "step": 140
    },
    {
      "epoch": 0.013334518623877679,
      "grad_norm": 0.6226356029510498,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 1.0316,
      "step": 150
    },
    {
      "epoch": 0.01422348653213619,
      "grad_norm": 0.5927873849868774,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 1.0391,
      "step": 160
    },
    {
      "epoch": 0.015112454440394702,
      "grad_norm": 0.6042047739028931,
      "learning_rate": 1.89938202247191e-05,
      "loss": 1.02,
      "step": 170
    },
    {
      "epoch": 0.016001422348653215,
      "grad_norm": 0.5578976273536682,
      "learning_rate": 2e-05,
      "loss": 1.0348,
      "step": 180
    },
    {
      "epoch": 0.016890390256911726,
      "grad_norm": 0.6130852699279785,
      "learning_rate": 2e-05,
      "loss": 1.0207,
      "step": 190
    },
    {
      "epoch": 0.017779358165170237,
      "grad_norm": 0.5213446617126465,
      "learning_rate": 2e-05,
      "loss": 1.06,
      "step": 200
    },
    {
      "epoch": 0.018668326073428748,
      "grad_norm": 0.5071802139282227,
      "learning_rate": 2e-05,
      "loss": 1.0331,
      "step": 210
    },
    {
      "epoch": 0.019557293981687263,
      "grad_norm": 0.49541351199150085,
      "learning_rate": 2e-05,
      "loss": 1.0726,
      "step": 220
    },
    {
      "epoch": 0.020446261889945774,
      "grad_norm": 0.4953780472278595,
      "learning_rate": 2e-05,
      "loss": 1.0192,
      "step": 230
    },
    {
      "epoch": 0.021335229798204285,
      "grad_norm": 0.5415428280830383,
      "learning_rate": 2e-05,
      "loss": 1.0232,
      "step": 240
    },
    {
      "epoch": 0.022224197706462796,
      "grad_norm": 0.529830813407898,
      "learning_rate": 2e-05,
      "loss": 1.0408,
      "step": 250
    },
    {
      "epoch": 0.02311316561472131,
      "grad_norm": 0.5176880955696106,
      "learning_rate": 2e-05,
      "loss": 1.0416,
      "step": 260
    },
    {
      "epoch": 0.02400213352297982,
      "grad_norm": 0.5344821810722351,
      "learning_rate": 2e-05,
      "loss": 0.9997,
      "step": 270
    },
    {
      "epoch": 0.024891101431238332,
      "grad_norm": 0.5449062585830688,
      "learning_rate": 2e-05,
      "loss": 1.0481,
      "step": 280
    },
    {
      "epoch": 0.025780069339496843,
      "grad_norm": 0.5149458050727844,
      "learning_rate": 2e-05,
      "loss": 1.0454,
      "step": 290
    },
    {
      "epoch": 0.026669037247755358,
      "grad_norm": 0.49063146114349365,
      "learning_rate": 2e-05,
      "loss": 1.0381,
      "step": 300
    },
    {
      "epoch": 0.02755800515601387,
      "grad_norm": 0.48867329955101013,
      "learning_rate": 2e-05,
      "loss": 1.0128,
      "step": 310
    },
    {
      "epoch": 0.02844697306427238,
      "grad_norm": 0.5222336649894714,
      "learning_rate": 2e-05,
      "loss": 1.0318,
      "step": 320
    },
    {
      "epoch": 0.02933594097253089,
      "grad_norm": 0.589815080165863,
      "learning_rate": 2e-05,
      "loss": 0.9987,
      "step": 330
    },
    {
      "epoch": 0.030224908880789405,
      "grad_norm": 0.522380530834198,
      "learning_rate": 2e-05,
      "loss": 1.0427,
      "step": 340
    },
    {
      "epoch": 0.031113876789047916,
      "grad_norm": 0.5397540330886841,
      "learning_rate": 2e-05,
      "loss": 1.0103,
      "step": 350
    },
    {
      "epoch": 0.03200284469730643,
      "grad_norm": 1.9739325046539307,
      "learning_rate": 2e-05,
      "loss": 1.0189,
      "step": 360
    },
    {
      "epoch": 0.03289181260556494,
      "grad_norm": 0.5190985202789307,
      "learning_rate": 2e-05,
      "loss": 1.0137,
      "step": 370
    },
    {
      "epoch": 0.03378078051382345,
      "grad_norm": 0.49212539196014404,
      "learning_rate": 2e-05,
      "loss": 1.0271,
      "step": 380
    },
    {
      "epoch": 0.03466974842208196,
      "grad_norm": 0.48474904894828796,
      "learning_rate": 2e-05,
      "loss": 1.0107,
      "step": 390
    },
    {
      "epoch": 0.035558716330340474,
      "grad_norm": 0.63145512342453,
      "learning_rate": 2e-05,
      "loss": 0.9979,
      "step": 400
    },
    {
      "epoch": 0.03644768423859899,
      "grad_norm": 0.489841103553772,
      "learning_rate": 2e-05,
      "loss": 1.0007,
      "step": 410
    },
    {
      "epoch": 0.037336652146857496,
      "grad_norm": 0.5093907117843628,
      "learning_rate": 2e-05,
      "loss": 0.9841,
      "step": 420
    },
    {
      "epoch": 0.03822562005511601,
      "grad_norm": 0.5208895206451416,
      "learning_rate": 2e-05,
      "loss": 1.0045,
      "step": 430
    },
    {
      "epoch": 0.039114587963374525,
      "grad_norm": 0.4912340044975281,
      "learning_rate": 2e-05,
      "loss": 1.0158,
      "step": 440
    },
    {
      "epoch": 0.04000355587163303,
      "grad_norm": 0.5119054317474365,
      "learning_rate": 2e-05,
      "loss": 0.9946,
      "step": 450
    },
    {
      "epoch": 0.04089252377989155,
      "grad_norm": 0.49770587682724,
      "learning_rate": 2e-05,
      "loss": 0.9851,
      "step": 460
    },
    {
      "epoch": 0.041781491688150055,
      "grad_norm": 0.4990149140357971,
      "learning_rate": 2e-05,
      "loss": 1.0499,
      "step": 470
    },
    {
      "epoch": 0.04267045959640857,
      "grad_norm": 0.4940296411514282,
      "learning_rate": 2e-05,
      "loss": 0.9977,
      "step": 480
    },
    {
      "epoch": 0.043559427504667084,
      "grad_norm": 0.47203925251960754,
      "learning_rate": 2e-05,
      "loss": 0.9953,
      "step": 490
    },
    {
      "epoch": 0.04444839541292559,
      "grad_norm": 0.5077679753303528,
      "learning_rate": 2e-05,
      "loss": 1.0069,
      "step": 500
    },
    {
      "epoch": 0.045337363321184106,
      "grad_norm": 0.5261061191558838,
      "learning_rate": 2e-05,
      "loss": 1.0212,
      "step": 510
    },
    {
      "epoch": 0.04622633122944262,
      "grad_norm": 0.49914729595184326,
      "learning_rate": 2e-05,
      "loss": 1.0094,
      "step": 520
    },
    {
      "epoch": 0.04711529913770113,
      "grad_norm": 0.49721768498420715,
      "learning_rate": 2e-05,
      "loss": 0.9993,
      "step": 530
    },
    {
      "epoch": 0.04800426704595964,
      "grad_norm": 0.46859484910964966,
      "learning_rate": 2e-05,
      "loss": 0.9901,
      "step": 540
    },
    {
      "epoch": 0.04889323495421815,
      "grad_norm": 0.4489647150039673,
      "learning_rate": 2e-05,
      "loss": 0.99,
      "step": 550
    },
    {
      "epoch": 0.049782202862476664,
      "grad_norm": 0.472504198551178,
      "learning_rate": 2e-05,
      "loss": 1.024,
      "step": 560
    },
    {
      "epoch": 0.05067117077073518,
      "grad_norm": 0.5098609924316406,
      "learning_rate": 2e-05,
      "loss": 1.0048,
      "step": 570
    },
    {
      "epoch": 0.051560138678993686,
      "grad_norm": 0.45842158794403076,
      "learning_rate": 2e-05,
      "loss": 1.0026,
      "step": 580
    },
    {
      "epoch": 0.0524491065872522,
      "grad_norm": 0.4619347155094147,
      "learning_rate": 2e-05,
      "loss": 0.9719,
      "step": 590
    },
    {
      "epoch": 0.053338074495510715,
      "grad_norm": 0.5127968788146973,
      "learning_rate": 2e-05,
      "loss": 1.0013,
      "step": 600
    },
    {
      "epoch": 0.05422704240376922,
      "grad_norm": 0.47510862350463867,
      "learning_rate": 2e-05,
      "loss": 0.9684,
      "step": 610
    },
    {
      "epoch": 0.05511601031202774,
      "grad_norm": 0.48674413561820984,
      "learning_rate": 2e-05,
      "loss": 0.9922,
      "step": 620
    },
    {
      "epoch": 0.056004978220286245,
      "grad_norm": 0.4563288390636444,
      "learning_rate": 2e-05,
      "loss": 0.9954,
      "step": 630
    },
    {
      "epoch": 0.05689394612854476,
      "grad_norm": 0.4836796820163727,
      "learning_rate": 2e-05,
      "loss": 0.9772,
      "step": 640
    },
    {
      "epoch": 0.05778291403680327,
      "grad_norm": 0.5137700438499451,
      "learning_rate": 2e-05,
      "loss": 0.991,
      "step": 650
    },
    {
      "epoch": 0.05867188194506178,
      "grad_norm": 0.48672813177108765,
      "learning_rate": 2e-05,
      "loss": 0.9755,
      "step": 660
    },
    {
      "epoch": 0.059560849853320295,
      "grad_norm": 0.501348078250885,
      "learning_rate": 2e-05,
      "loss": 0.9903,
      "step": 670
    },
    {
      "epoch": 0.06044981776157881,
      "grad_norm": 0.48052978515625,
      "learning_rate": 2e-05,
      "loss": 0.9822,
      "step": 680
    },
    {
      "epoch": 0.06133878566983732,
      "grad_norm": 0.7372881770133972,
      "learning_rate": 2e-05,
      "loss": 0.9958,
      "step": 690
    },
    {
      "epoch": 0.06222775357809583,
      "grad_norm": 0.5022790431976318,
      "learning_rate": 2e-05,
      "loss": 1.0131,
      "step": 700
    },
    {
      "epoch": 0.06311672148635435,
      "grad_norm": 0.497873991727829,
      "learning_rate": 2e-05,
      "loss": 0.985,
      "step": 710
    },
    {
      "epoch": 0.06400568939461286,
      "grad_norm": 1.2517738342285156,
      "learning_rate": 2e-05,
      "loss": 0.9648,
      "step": 720
    },
    {
      "epoch": 0.06489465730287136,
      "grad_norm": 0.4873271882534027,
      "learning_rate": 2e-05,
      "loss": 0.9837,
      "step": 730
    },
    {
      "epoch": 0.06578362521112988,
      "grad_norm": 0.5012655854225159,
      "learning_rate": 2e-05,
      "loss": 0.985,
      "step": 740
    },
    {
      "epoch": 0.06667259311938839,
      "grad_norm": 0.482709676027298,
      "learning_rate": 2e-05,
      "loss": 0.9743,
      "step": 750
    },
    {
      "epoch": 0.0675615610276469,
      "grad_norm": 0.4815094470977783,
      "learning_rate": 2e-05,
      "loss": 0.9818,
      "step": 760
    },
    {
      "epoch": 0.06845052893590542,
      "grad_norm": 0.4626294672489166,
      "learning_rate": 2e-05,
      "loss": 0.9953,
      "step": 770
    },
    {
      "epoch": 0.06933949684416392,
      "grad_norm": 0.4867992103099823,
      "learning_rate": 2e-05,
      "loss": 0.9743,
      "step": 780
    },
    {
      "epoch": 0.07022846475242243,
      "grad_norm": 0.49126219749450684,
      "learning_rate": 2e-05,
      "loss": 0.972,
      "step": 790
    },
    {
      "epoch": 0.07111743266068095,
      "grad_norm": 0.5143160223960876,
      "learning_rate": 2e-05,
      "loss": 0.9534,
      "step": 800
    },
    {
      "epoch": 0.07200640056893946,
      "grad_norm": 0.5352365374565125,
      "learning_rate": 2e-05,
      "loss": 1.0183,
      "step": 810
    },
    {
      "epoch": 0.07289536847719798,
      "grad_norm": 0.5039393901824951,
      "learning_rate": 2e-05,
      "loss": 0.9831,
      "step": 820
    },
    {
      "epoch": 0.07378433638545648,
      "grad_norm": 0.461434006690979,
      "learning_rate": 2e-05,
      "loss": 0.9805,
      "step": 830
    },
    {
      "epoch": 0.07467330429371499,
      "grad_norm": 0.48597556352615356,
      "learning_rate": 2e-05,
      "loss": 0.9592,
      "step": 840
    },
    {
      "epoch": 0.07556227220197351,
      "grad_norm": 0.5177674889564514,
      "learning_rate": 2e-05,
      "loss": 0.9739,
      "step": 850
    },
    {
      "epoch": 0.07645124011023202,
      "grad_norm": 0.48002588748931885,
      "learning_rate": 2e-05,
      "loss": 0.9878,
      "step": 860
    },
    {
      "epoch": 0.07734020801849054,
      "grad_norm": 0.5129883885383606,
      "learning_rate": 2e-05,
      "loss": 0.9756,
      "step": 870
    },
    {
      "epoch": 0.07822917592674905,
      "grad_norm": 0.5066726803779602,
      "learning_rate": 2e-05,
      "loss": 0.9699,
      "step": 880
    },
    {
      "epoch": 0.07911814383500755,
      "grad_norm": 0.4818406403064728,
      "learning_rate": 2e-05,
      "loss": 0.9898,
      "step": 890
    },
    {
      "epoch": 0.08000711174326607,
      "grad_norm": 0.48721879720687866,
      "learning_rate": 2e-05,
      "loss": 0.9503,
      "step": 900
    },
    {
      "epoch": 0.08089607965152458,
      "grad_norm": 0.4314747452735901,
      "learning_rate": 2e-05,
      "loss": 0.9891,
      "step": 910
    },
    {
      "epoch": 0.0817850475597831,
      "grad_norm": 0.4804377853870392,
      "learning_rate": 2e-05,
      "loss": 0.9727,
      "step": 920
    },
    {
      "epoch": 0.08267401546804161,
      "grad_norm": 0.48248469829559326,
      "learning_rate": 2e-05,
      "loss": 0.967,
      "step": 930
    },
    {
      "epoch": 0.08356298337630011,
      "grad_norm": 0.45454731583595276,
      "learning_rate": 2e-05,
      "loss": 0.9854,
      "step": 940
    },
    {
      "epoch": 0.08445195128455862,
      "grad_norm": 0.47664403915405273,
      "learning_rate": 2e-05,
      "loss": 0.987,
      "step": 950
    },
    {
      "epoch": 0.08534091919281714,
      "grad_norm": 0.47163963317871094,
      "learning_rate": 2e-05,
      "loss": 0.9979,
      "step": 960
    },
    {
      "epoch": 0.08622988710107565,
      "grad_norm": 0.5137403011322021,
      "learning_rate": 2e-05,
      "loss": 0.9728,
      "step": 970
    },
    {
      "epoch": 0.08711885500933417,
      "grad_norm": 0.449590802192688,
      "learning_rate": 2e-05,
      "loss": 0.9599,
      "step": 980
    },
    {
      "epoch": 0.08800782291759268,
      "grad_norm": 0.4698745310306549,
      "learning_rate": 2e-05,
      "loss": 0.9475,
      "step": 990
    },
    {
      "epoch": 0.08889679082585118,
      "grad_norm": 0.45183730125427246,
      "learning_rate": 2e-05,
      "loss": 0.9683,
      "step": 1000
    },
    {
      "epoch": 0.0897857587341097,
      "grad_norm": 0.45844438672065735,
      "learning_rate": 2e-05,
      "loss": 0.9835,
      "step": 1010
    },
    {
      "epoch": 0.09067472664236821,
      "grad_norm": 0.4777911305427551,
      "learning_rate": 2e-05,
      "loss": 0.9922,
      "step": 1020
    },
    {
      "epoch": 0.09156369455062673,
      "grad_norm": 0.4591684937477112,
      "learning_rate": 2e-05,
      "loss": 0.9448,
      "step": 1030
    },
    {
      "epoch": 0.09245266245888524,
      "grad_norm": 0.47120004892349243,
      "learning_rate": 2e-05,
      "loss": 0.9668,
      "step": 1040
    },
    {
      "epoch": 0.09334163036714374,
      "grad_norm": 0.467716783285141,
      "learning_rate": 2e-05,
      "loss": 0.9661,
      "step": 1050
    },
    {
      "epoch": 0.09423059827540226,
      "grad_norm": 0.479279488325119,
      "learning_rate": 2e-05,
      "loss": 0.9951,
      "step": 1060
    },
    {
      "epoch": 0.09511956618366077,
      "grad_norm": 0.4522221088409424,
      "learning_rate": 2e-05,
      "loss": 0.9553,
      "step": 1070
    },
    {
      "epoch": 0.09600853409191928,
      "grad_norm": 0.4667496979236603,
      "learning_rate": 2e-05,
      "loss": 0.9659,
      "step": 1080
    },
    {
      "epoch": 0.0968975020001778,
      "grad_norm": 0.47797060012817383,
      "learning_rate": 2e-05,
      "loss": 0.9628,
      "step": 1090
    },
    {
      "epoch": 0.0977864699084363,
      "grad_norm": 0.47188839316368103,
      "learning_rate": 2e-05,
      "loss": 0.9738,
      "step": 1100
    },
    {
      "epoch": 0.09867543781669481,
      "grad_norm": 0.47207027673721313,
      "learning_rate": 2e-05,
      "loss": 0.9473,
      "step": 1110
    },
    {
      "epoch": 0.09956440572495333,
      "grad_norm": 0.4718773066997528,
      "learning_rate": 2e-05,
      "loss": 0.9692,
      "step": 1120
    },
    {
      "epoch": 0.10045337363321184,
      "grad_norm": 0.45084095001220703,
      "learning_rate": 2e-05,
      "loss": 0.9595,
      "step": 1130
    },
    {
      "epoch": 0.10134234154147036,
      "grad_norm": 0.4841742217540741,
      "learning_rate": 2e-05,
      "loss": 0.9701,
      "step": 1140
    },
    {
      "epoch": 0.10223130944972887,
      "grad_norm": 0.5060001015663147,
      "learning_rate": 2e-05,
      "loss": 0.966,
      "step": 1150
    },
    {
      "epoch": 0.10312027735798737,
      "grad_norm": 0.4384157061576843,
      "learning_rate": 2e-05,
      "loss": 0.9809,
      "step": 1160
    },
    {
      "epoch": 0.10400924526624589,
      "grad_norm": 0.4938412606716156,
      "learning_rate": 2e-05,
      "loss": 0.9664,
      "step": 1170
    },
    {
      "epoch": 0.1048982131745044,
      "grad_norm": 0.5100815892219543,
      "learning_rate": 2e-05,
      "loss": 0.964,
      "step": 1180
    },
    {
      "epoch": 0.10578718108276292,
      "grad_norm": 0.4635246992111206,
      "learning_rate": 2e-05,
      "loss": 0.9641,
      "step": 1190
    },
    {
      "epoch": 0.10667614899102143,
      "grad_norm": 0.5389490127563477,
      "learning_rate": 2e-05,
      "loss": 0.9767,
      "step": 1200
    },
    {
      "epoch": 0.10756511689927993,
      "grad_norm": 0.46785902976989746,
      "learning_rate": 2e-05,
      "loss": 0.9496,
      "step": 1210
    },
    {
      "epoch": 0.10845408480753845,
      "grad_norm": 0.4610632359981537,
      "learning_rate": 2e-05,
      "loss": 0.9565,
      "step": 1220
    },
    {
      "epoch": 0.10934305271579696,
      "grad_norm": 0.44586533308029175,
      "learning_rate": 2e-05,
      "loss": 0.9652,
      "step": 1230
    },
    {
      "epoch": 0.11023202062405547,
      "grad_norm": 0.5270296931266785,
      "learning_rate": 2e-05,
      "loss": 0.9595,
      "step": 1240
    },
    {
      "epoch": 0.11112098853231399,
      "grad_norm": 0.4567857086658478,
      "learning_rate": 2e-05,
      "loss": 0.9806,
      "step": 1250
    },
    {
      "epoch": 0.11200995644057249,
      "grad_norm": 0.4863816201686859,
      "learning_rate": 2e-05,
      "loss": 0.9617,
      "step": 1260
    },
    {
      "epoch": 0.112898924348831,
      "grad_norm": 0.4874187409877777,
      "learning_rate": 2e-05,
      "loss": 0.9785,
      "step": 1270
    },
    {
      "epoch": 0.11378789225708952,
      "grad_norm": 0.4692484736442566,
      "learning_rate": 2e-05,
      "loss": 0.9593,
      "step": 1280
    },
    {
      "epoch": 0.11467686016534803,
      "grad_norm": 0.6816130876541138,
      "learning_rate": 2e-05,
      "loss": 0.9237,
      "step": 1290
    },
    {
      "epoch": 0.11556582807360655,
      "grad_norm": 0.49590855836868286,
      "learning_rate": 2e-05,
      "loss": 0.9628,
      "step": 1300
    },
    {
      "epoch": 0.11645479598186506,
      "grad_norm": 0.45354118943214417,
      "learning_rate": 2e-05,
      "loss": 0.9396,
      "step": 1310
    },
    {
      "epoch": 0.11734376389012356,
      "grad_norm": 0.4449705481529236,
      "learning_rate": 2e-05,
      "loss": 0.9473,
      "step": 1320
    },
    {
      "epoch": 0.11823273179838208,
      "grad_norm": 0.44771867990493774,
      "learning_rate": 2e-05,
      "loss": 0.954,
      "step": 1330
    },
    {
      "epoch": 0.11912169970664059,
      "grad_norm": 0.4958712160587311,
      "learning_rate": 2e-05,
      "loss": 0.9637,
      "step": 1340
    },
    {
      "epoch": 0.1200106676148991,
      "grad_norm": 0.4684152603149414,
      "learning_rate": 2e-05,
      "loss": 0.9383,
      "step": 1350
    },
    {
      "epoch": 0.12089963552315762,
      "grad_norm": 0.5252647399902344,
      "learning_rate": 2e-05,
      "loss": 0.9668,
      "step": 1360
    },
    {
      "epoch": 0.12178860343141612,
      "grad_norm": 0.4878622591495514,
      "learning_rate": 2e-05,
      "loss": 0.9357,
      "step": 1370
    },
    {
      "epoch": 0.12267757133967463,
      "grad_norm": 0.46592438220977783,
      "learning_rate": 2e-05,
      "loss": 0.9515,
      "step": 1380
    },
    {
      "epoch": 0.12356653924793315,
      "grad_norm": 0.5377469062805176,
      "learning_rate": 2e-05,
      "loss": 0.9431,
      "step": 1390
    },
    {
      "epoch": 0.12445550715619166,
      "grad_norm": 0.472373902797699,
      "learning_rate": 2e-05,
      "loss": 0.9305,
      "step": 1400
    },
    {
      "epoch": 0.12534447506445018,
      "grad_norm": 0.4335474967956543,
      "learning_rate": 2e-05,
      "loss": 0.9501,
      "step": 1410
    },
    {
      "epoch": 0.1262334429727087,
      "grad_norm": 0.43715789914131165,
      "learning_rate": 2e-05,
      "loss": 0.9565,
      "step": 1420
    },
    {
      "epoch": 0.1271224108809672,
      "grad_norm": 0.473441481590271,
      "learning_rate": 2e-05,
      "loss": 0.9341,
      "step": 1430
    },
    {
      "epoch": 0.12801137878922572,
      "grad_norm": 0.4540274143218994,
      "learning_rate": 2e-05,
      "loss": 0.9584,
      "step": 1440
    },
    {
      "epoch": 0.1289003466974842,
      "grad_norm": 0.45082035660743713,
      "learning_rate": 2e-05,
      "loss": 0.9772,
      "step": 1450
    },
    {
      "epoch": 0.12978931460574272,
      "grad_norm": 0.43487876653671265,
      "learning_rate": 2e-05,
      "loss": 0.9642,
      "step": 1460
    },
    {
      "epoch": 0.13067828251400124,
      "grad_norm": 0.45913758873939514,
      "learning_rate": 2e-05,
      "loss": 0.9387,
      "step": 1470
    },
    {
      "epoch": 0.13156725042225975,
      "grad_norm": 0.458950936794281,
      "learning_rate": 2e-05,
      "loss": 0.9597,
      "step": 1480
    },
    {
      "epoch": 0.13245621833051827,
      "grad_norm": 0.462766170501709,
      "learning_rate": 2e-05,
      "loss": 0.9871,
      "step": 1490
    },
    {
      "epoch": 0.13334518623877678,
      "grad_norm": 0.4476396441459656,
      "learning_rate": 2e-05,
      "loss": 0.9311,
      "step": 1500
    },
    {
      "epoch": 0.1342341541470353,
      "grad_norm": 0.46129149198532104,
      "learning_rate": 2e-05,
      "loss": 0.9348,
      "step": 1510
    },
    {
      "epoch": 0.1351231220552938,
      "grad_norm": 0.46218064427375793,
      "learning_rate": 2e-05,
      "loss": 0.9261,
      "step": 1520
    },
    {
      "epoch": 0.13601208996355232,
      "grad_norm": 0.4620305895805359,
      "learning_rate": 2e-05,
      "loss": 0.9347,
      "step": 1530
    },
    {
      "epoch": 0.13690105787181084,
      "grad_norm": 0.603297770023346,
      "learning_rate": 2e-05,
      "loss": 0.9829,
      "step": 1540
    },
    {
      "epoch": 0.13779002578006935,
      "grad_norm": 0.5085828900337219,
      "learning_rate": 2e-05,
      "loss": 0.9259,
      "step": 1550
    },
    {
      "epoch": 0.13867899368832784,
      "grad_norm": 0.4496013820171356,
      "learning_rate": 2e-05,
      "loss": 0.9561,
      "step": 1560
    },
    {
      "epoch": 0.13956796159658635,
      "grad_norm": 0.468447208404541,
      "learning_rate": 2e-05,
      "loss": 0.9603,
      "step": 1570
    },
    {
      "epoch": 0.14045692950484487,
      "grad_norm": 0.45801347494125366,
      "learning_rate": 2e-05,
      "loss": 0.9326,
      "step": 1580
    },
    {
      "epoch": 0.14134589741310338,
      "grad_norm": 0.46467089653015137,
      "learning_rate": 2e-05,
      "loss": 0.9255,
      "step": 1590
    },
    {
      "epoch": 0.1422348653213619,
      "grad_norm": 0.43190547823905945,
      "learning_rate": 2e-05,
      "loss": 0.9339,
      "step": 1600
    },
    {
      "epoch": 0.1431238332296204,
      "grad_norm": 0.4498319923877716,
      "learning_rate": 2e-05,
      "loss": 0.9288,
      "step": 1610
    },
    {
      "epoch": 0.14401280113787893,
      "grad_norm": 0.4732903838157654,
      "learning_rate": 2e-05,
      "loss": 0.9505,
      "step": 1620
    },
    {
      "epoch": 0.14490176904613744,
      "grad_norm": 0.4331393539905548,
      "learning_rate": 2e-05,
      "loss": 0.9487,
      "step": 1630
    },
    {
      "epoch": 0.14579073695439596,
      "grad_norm": 0.4622068405151367,
      "learning_rate": 2e-05,
      "loss": 0.9322,
      "step": 1640
    },
    {
      "epoch": 0.14667970486265447,
      "grad_norm": 0.44649264216423035,
      "learning_rate": 2e-05,
      "loss": 0.944,
      "step": 1650
    },
    {
      "epoch": 0.14756867277091296,
      "grad_norm": 0.44571954011917114,
      "learning_rate": 2e-05,
      "loss": 0.9305,
      "step": 1660
    },
    {
      "epoch": 0.14845764067917147,
      "grad_norm": 0.47047024965286255,
      "learning_rate": 2e-05,
      "loss": 0.9837,
      "step": 1670
    },
    {
      "epoch": 0.14934660858742999,
      "grad_norm": 0.4326838254928589,
      "learning_rate": 2e-05,
      "loss": 0.9455,
      "step": 1680
    },
    {
      "epoch": 0.1502355764956885,
      "grad_norm": 0.4418770372867584,
      "learning_rate": 2e-05,
      "loss": 0.9395,
      "step": 1690
    },
    {
      "epoch": 0.15112454440394701,
      "grad_norm": 0.4838589131832123,
      "learning_rate": 2e-05,
      "loss": 0.9467,
      "step": 1700
    },
    {
      "epoch": 0.15201351231220553,
      "grad_norm": 0.46900397539138794,
      "learning_rate": 2e-05,
      "loss": 0.9553,
      "step": 1710
    },
    {
      "epoch": 0.15290248022046404,
      "grad_norm": 0.4528874456882477,
      "learning_rate": 2e-05,
      "loss": 0.9536,
      "step": 1720
    },
    {
      "epoch": 0.15379144812872256,
      "grad_norm": 0.43615949153900146,
      "learning_rate": 2e-05,
      "loss": 0.9439,
      "step": 1730
    },
    {
      "epoch": 0.15468041603698107,
      "grad_norm": 0.4381987154483795,
      "learning_rate": 2e-05,
      "loss": 0.9625,
      "step": 1740
    },
    {
      "epoch": 0.1555693839452396,
      "grad_norm": 0.4465603530406952,
      "learning_rate": 2e-05,
      "loss": 0.9219,
      "step": 1750
    },
    {
      "epoch": 0.1564583518534981,
      "grad_norm": 0.4344707727432251,
      "learning_rate": 2e-05,
      "loss": 0.9348,
      "step": 1760
    },
    {
      "epoch": 0.1573473197617566,
      "grad_norm": 0.47260528802871704,
      "learning_rate": 2e-05,
      "loss": 0.9142,
      "step": 1770
    },
    {
      "epoch": 0.1582362876700151,
      "grad_norm": 0.4460822641849518,
      "learning_rate": 2e-05,
      "loss": 0.9283,
      "step": 1780
    },
    {
      "epoch": 0.15912525557827362,
      "grad_norm": 0.46511539816856384,
      "learning_rate": 2e-05,
      "loss": 0.9484,
      "step": 1790
    },
    {
      "epoch": 0.16001422348653213,
      "grad_norm": 0.4567843973636627,
      "learning_rate": 2e-05,
      "loss": 0.9499,
      "step": 1800
    },
    {
      "epoch": 0.16090319139479065,
      "grad_norm": 0.4729498326778412,
      "learning_rate": 2e-05,
      "loss": 0.9633,
      "step": 1810
    },
    {
      "epoch": 0.16179215930304916,
      "grad_norm": 0.44997239112854004,
      "learning_rate": 2e-05,
      "loss": 0.9255,
      "step": 1820
    },
    {
      "epoch": 0.16268112721130767,
      "grad_norm": 0.4489092230796814,
      "learning_rate": 2e-05,
      "loss": 0.947,
      "step": 1830
    },
    {
      "epoch": 0.1635700951195662,
      "grad_norm": 0.5339633226394653,
      "learning_rate": 2e-05,
      "loss": 0.9434,
      "step": 1840
    },
    {
      "epoch": 0.1644590630278247,
      "grad_norm": 0.4252396523952484,
      "learning_rate": 2e-05,
      "loss": 0.9074,
      "step": 1850
    },
    {
      "epoch": 0.16534803093608322,
      "grad_norm": 0.477722704410553,
      "learning_rate": 2e-05,
      "loss": 0.9234,
      "step": 1860
    },
    {
      "epoch": 0.16623699884434173,
      "grad_norm": 0.46278834342956543,
      "learning_rate": 2e-05,
      "loss": 0.8918,
      "step": 1870
    },
    {
      "epoch": 0.16712596675260022,
      "grad_norm": 0.4720710515975952,
      "learning_rate": 2e-05,
      "loss": 0.9356,
      "step": 1880
    },
    {
      "epoch": 0.16801493466085873,
      "grad_norm": 0.4468317925930023,
      "learning_rate": 2e-05,
      "loss": 0.9237,
      "step": 1890
    },
    {
      "epoch": 0.16890390256911725,
      "grad_norm": 0.4632788300514221,
      "learning_rate": 2e-05,
      "loss": 0.9552,
      "step": 1900
    },
    {
      "epoch": 0.16979287047737576,
      "grad_norm": 0.42456817626953125,
      "learning_rate": 2e-05,
      "loss": 0.901,
      "step": 1910
    },
    {
      "epoch": 0.17068183838563428,
      "grad_norm": 0.4745561182498932,
      "learning_rate": 2e-05,
      "loss": 0.9236,
      "step": 1920
    },
    {
      "epoch": 0.1715708062938928,
      "grad_norm": 0.43758249282836914,
      "learning_rate": 2e-05,
      "loss": 0.9205,
      "step": 1930
    },
    {
      "epoch": 0.1724597742021513,
      "grad_norm": 0.45937028527259827,
      "learning_rate": 2e-05,
      "loss": 0.8825,
      "step": 1940
    },
    {
      "epoch": 0.17334874211040982,
      "grad_norm": 0.43388575315475464,
      "learning_rate": 2e-05,
      "loss": 0.9254,
      "step": 1950
    },
    {
      "epoch": 0.17423771001866833,
      "grad_norm": 0.4711926281452179,
      "learning_rate": 2e-05,
      "loss": 0.9437,
      "step": 1960
    },
    {
      "epoch": 0.17512667792692685,
      "grad_norm": 0.44924527406692505,
      "learning_rate": 2e-05,
      "loss": 0.9241,
      "step": 1970
    },
    {
      "epoch": 0.17601564583518536,
      "grad_norm": 0.42580580711364746,
      "learning_rate": 2e-05,
      "loss": 0.9162,
      "step": 1980
    },
    {
      "epoch": 0.17690461374344385,
      "grad_norm": 0.4601577818393707,
      "learning_rate": 2e-05,
      "loss": 0.9473,
      "step": 1990
    },
    {
      "epoch": 0.17779358165170236,
      "grad_norm": 0.4731871485710144,
      "learning_rate": 2e-05,
      "loss": 0.9462,
      "step": 2000
    },
    {
      "epoch": 0.17868254955996088,
      "grad_norm": 0.4369046688079834,
      "learning_rate": 2e-05,
      "loss": 0.9724,
      "step": 2010
    },
    {
      "epoch": 0.1795715174682194,
      "grad_norm": 0.4617767333984375,
      "learning_rate": 2e-05,
      "loss": 0.9402,
      "step": 2020
    },
    {
      "epoch": 0.1804604853764779,
      "grad_norm": 0.45765435695648193,
      "learning_rate": 2e-05,
      "loss": 0.9275,
      "step": 2030
    },
    {
      "epoch": 0.18134945328473642,
      "grad_norm": 0.44978591799736023,
      "learning_rate": 2e-05,
      "loss": 0.9216,
      "step": 2040
    },
    {
      "epoch": 0.18223842119299494,
      "grad_norm": 0.45181888341903687,
      "learning_rate": 2e-05,
      "loss": 0.9389,
      "step": 2050
    },
    {
      "epoch": 0.18312738910125345,
      "grad_norm": 0.6255255341529846,
      "learning_rate": 2e-05,
      "loss": 0.9226,
      "step": 2060
    },
    {
      "epoch": 0.18401635700951197,
      "grad_norm": 0.44182655215263367,
      "learning_rate": 2e-05,
      "loss": 0.9273,
      "step": 2070
    },
    {
      "epoch": 0.18490532491777048,
      "grad_norm": 0.4544832408428192,
      "learning_rate": 2e-05,
      "loss": 0.9089,
      "step": 2080
    },
    {
      "epoch": 0.18579429282602897,
      "grad_norm": 0.44694894552230835,
      "learning_rate": 2e-05,
      "loss": 0.9495,
      "step": 2090
    },
    {
      "epoch": 0.18668326073428748,
      "grad_norm": 0.4754018485546112,
      "learning_rate": 2e-05,
      "loss": 0.9208,
      "step": 2100
    },
    {
      "epoch": 0.187572228642546,
      "grad_norm": 0.43904373049736023,
      "learning_rate": 2e-05,
      "loss": 0.9171,
      "step": 2110
    },
    {
      "epoch": 0.1884611965508045,
      "grad_norm": 0.43585121631622314,
      "learning_rate": 2e-05,
      "loss": 0.9224,
      "step": 2120
    },
    {
      "epoch": 0.18935016445906303,
      "grad_norm": 0.4322623908519745,
      "learning_rate": 2e-05,
      "loss": 0.9096,
      "step": 2130
    },
    {
      "epoch": 0.19023913236732154,
      "grad_norm": 0.45563212037086487,
      "learning_rate": 2e-05,
      "loss": 0.9504,
      "step": 2140
    },
    {
      "epoch": 0.19112810027558005,
      "grad_norm": 0.46097010374069214,
      "learning_rate": 2e-05,
      "loss": 0.9338,
      "step": 2150
    },
    {
      "epoch": 0.19201706818383857,
      "grad_norm": 0.43504899740219116,
      "learning_rate": 2e-05,
      "loss": 0.92,
      "step": 2160
    },
    {
      "epoch": 0.19290603609209708,
      "grad_norm": 0.4069807827472687,
      "learning_rate": 2e-05,
      "loss": 0.9284,
      "step": 2170
    },
    {
      "epoch": 0.1937950040003556,
      "grad_norm": 0.4753881096839905,
      "learning_rate": 2e-05,
      "loss": 0.9163,
      "step": 2180
    },
    {
      "epoch": 0.1946839719086141,
      "grad_norm": 0.42152369022369385,
      "learning_rate": 2e-05,
      "loss": 0.898,
      "step": 2190
    },
    {
      "epoch": 0.1955729398168726,
      "grad_norm": 0.451825350522995,
      "learning_rate": 2e-05,
      "loss": 0.9228,
      "step": 2200
    },
    {
      "epoch": 0.1964619077251311,
      "grad_norm": 0.47456032037734985,
      "learning_rate": 2e-05,
      "loss": 0.9294,
      "step": 2210
    },
    {
      "epoch": 0.19735087563338963,
      "grad_norm": 0.4597964286804199,
      "learning_rate": 2e-05,
      "loss": 0.942,
      "step": 2220
    },
    {
      "epoch": 0.19823984354164814,
      "grad_norm": 0.46146100759506226,
      "learning_rate": 2e-05,
      "loss": 1.0043,
      "step": 2230
    },
    {
      "epoch": 0.19912881144990666,
      "grad_norm": 0.43969210982322693,
      "learning_rate": 2e-05,
      "loss": 0.9286,
      "step": 2240
    },
    {
      "epoch": 0.20001777935816517,
      "grad_norm": 0.4327503740787506,
      "learning_rate": 2e-05,
      "loss": 0.9348,
      "step": 2250
    },
    {
      "epoch": 0.20090674726642369,
      "grad_norm": 0.4259391725063324,
      "learning_rate": 2e-05,
      "loss": 0.9348,
      "step": 2260
    },
    {
      "epoch": 0.2017957151746822,
      "grad_norm": 0.4238135516643524,
      "learning_rate": 2e-05,
      "loss": 0.9743,
      "step": 2270
    },
    {
      "epoch": 0.20268468308294071,
      "grad_norm": 0.43770846724510193,
      "learning_rate": 2e-05,
      "loss": 0.9308,
      "step": 2280
    },
    {
      "epoch": 0.20357365099119923,
      "grad_norm": 0.45749926567077637,
      "learning_rate": 2e-05,
      "loss": 0.9328,
      "step": 2290
    },
    {
      "epoch": 0.20446261889945774,
      "grad_norm": 0.4421558976173401,
      "learning_rate": 2e-05,
      "loss": 0.9015,
      "step": 2300
    },
    {
      "epoch": 0.20535158680771623,
      "grad_norm": 0.44268208742141724,
      "learning_rate": 2e-05,
      "loss": 0.9204,
      "step": 2310
    },
    {
      "epoch": 0.20624055471597474,
      "grad_norm": 0.4365476667881012,
      "learning_rate": 2e-05,
      "loss": 0.9408,
      "step": 2320
    },
    {
      "epoch": 0.20712952262423326,
      "grad_norm": 0.4565020799636841,
      "learning_rate": 2e-05,
      "loss": 0.9241,
      "step": 2330
    },
    {
      "epoch": 0.20801849053249177,
      "grad_norm": 0.43321648240089417,
      "learning_rate": 2e-05,
      "loss": 0.9455,
      "step": 2340
    },
    {
      "epoch": 0.2089074584407503,
      "grad_norm": 0.44953176379203796,
      "learning_rate": 2e-05,
      "loss": 0.9006,
      "step": 2350
    },
    {
      "epoch": 0.2097964263490088,
      "grad_norm": 0.43321532011032104,
      "learning_rate": 2e-05,
      "loss": 0.9259,
      "step": 2360
    },
    {
      "epoch": 0.21068539425726732,
      "grad_norm": 0.43398517370224,
      "learning_rate": 2e-05,
      "loss": 0.9238,
      "step": 2370
    },
    {
      "epoch": 0.21157436216552583,
      "grad_norm": 0.4406549334526062,
      "learning_rate": 2e-05,
      "loss": 0.9206,
      "step": 2380
    },
    {
      "epoch": 0.21246333007378435,
      "grad_norm": 0.45563289523124695,
      "learning_rate": 2e-05,
      "loss": 0.9107,
      "step": 2390
    },
    {
      "epoch": 0.21335229798204286,
      "grad_norm": 0.4393308758735657,
      "learning_rate": 2e-05,
      "loss": 0.9071,
      "step": 2400
    },
    {
      "epoch": 0.21424126589030135,
      "grad_norm": 0.4431764781475067,
      "learning_rate": 2e-05,
      "loss": 0.9264,
      "step": 2410
    },
    {
      "epoch": 0.21513023379855986,
      "grad_norm": 0.423136830329895,
      "learning_rate": 2e-05,
      "loss": 0.9301,
      "step": 2420
    },
    {
      "epoch": 0.21601920170681838,
      "grad_norm": 0.42495959997177124,
      "learning_rate": 2e-05,
      "loss": 0.9082,
      "step": 2430
    },
    {
      "epoch": 0.2169081696150769,
      "grad_norm": 0.4279535710811615,
      "learning_rate": 2e-05,
      "loss": 0.8823,
      "step": 2440
    },
    {
      "epoch": 0.2177971375233354,
      "grad_norm": 0.4287704527378082,
      "learning_rate": 2e-05,
      "loss": 0.9135,
      "step": 2450
    },
    {
      "epoch": 0.21868610543159392,
      "grad_norm": 0.43721771240234375,
      "learning_rate": 2e-05,
      "loss": 0.9327,
      "step": 2460
    },
    {
      "epoch": 0.21957507333985243,
      "grad_norm": 0.4303989112377167,
      "learning_rate": 2e-05,
      "loss": 0.9058,
      "step": 2470
    },
    {
      "epoch": 0.22046404124811095,
      "grad_norm": 0.43652382493019104,
      "learning_rate": 2e-05,
      "loss": 0.9157,
      "step": 2480
    },
    {
      "epoch": 0.22135300915636946,
      "grad_norm": 0.4485083520412445,
      "learning_rate": 2e-05,
      "loss": 0.9024,
      "step": 2490
    },
    {
      "epoch": 0.22224197706462798,
      "grad_norm": 0.4460156261920929,
      "learning_rate": 2e-05,
      "loss": 0.9064,
      "step": 2500
    },
    {
      "epoch": 0.2231309449728865,
      "grad_norm": 0.4979501962661743,
      "learning_rate": 2e-05,
      "loss": 0.8991,
      "step": 2510
    },
    {
      "epoch": 0.22401991288114498,
      "grad_norm": 0.4301646053791046,
      "learning_rate": 2e-05,
      "loss": 0.9366,
      "step": 2520
    },
    {
      "epoch": 0.2249088807894035,
      "grad_norm": 0.45495837926864624,
      "learning_rate": 2e-05,
      "loss": 0.9183,
      "step": 2530
    },
    {
      "epoch": 0.225797848697662,
      "grad_norm": 0.5348328351974487,
      "learning_rate": 2e-05,
      "loss": 0.8963,
      "step": 2540
    },
    {
      "epoch": 0.22668681660592052,
      "grad_norm": 0.4004691541194916,
      "learning_rate": 2e-05,
      "loss": 0.9202,
      "step": 2550
    },
    {
      "epoch": 0.22757578451417904,
      "grad_norm": 0.4097733199596405,
      "learning_rate": 2e-05,
      "loss": 0.9039,
      "step": 2560
    },
    {
      "epoch": 0.22846475242243755,
      "grad_norm": 0.43968674540519714,
      "learning_rate": 2e-05,
      "loss": 0.9029,
      "step": 2570
    },
    {
      "epoch": 0.22935372033069606,
      "grad_norm": 0.43621495366096497,
      "learning_rate": 2e-05,
      "loss": 0.9446,
      "step": 2580
    },
    {
      "epoch": 0.23024268823895458,
      "grad_norm": 0.42816436290740967,
      "learning_rate": 2e-05,
      "loss": 0.8994,
      "step": 2590
    },
    {
      "epoch": 0.2311316561472131,
      "grad_norm": 0.4392223656177521,
      "learning_rate": 2e-05,
      "loss": 0.9385,
      "step": 2600
    },
    {
      "epoch": 0.2320206240554716,
      "grad_norm": 0.4421624541282654,
      "learning_rate": 2e-05,
      "loss": 0.9011,
      "step": 2610
    },
    {
      "epoch": 0.23290959196373012,
      "grad_norm": 0.3971703052520752,
      "learning_rate": 2e-05,
      "loss": 0.9167,
      "step": 2620
    },
    {
      "epoch": 0.2337985598719886,
      "grad_norm": 0.43121039867401123,
      "learning_rate": 2e-05,
      "loss": 0.9061,
      "step": 2630
    },
    {
      "epoch": 0.23468752778024712,
      "grad_norm": 0.4663209617137909,
      "learning_rate": 2e-05,
      "loss": 0.9412,
      "step": 2640
    },
    {
      "epoch": 0.23557649568850564,
      "grad_norm": 0.4751022756099701,
      "learning_rate": 2e-05,
      "loss": 0.9339,
      "step": 2650
    },
    {
      "epoch": 0.23646546359676415,
      "grad_norm": 0.4347291886806488,
      "learning_rate": 2e-05,
      "loss": 0.914,
      "step": 2660
    },
    {
      "epoch": 0.23735443150502267,
      "grad_norm": 0.3965953588485718,
      "learning_rate": 2e-05,
      "loss": 0.8902,
      "step": 2670
    },
    {
      "epoch": 0.23824339941328118,
      "grad_norm": 0.43548595905303955,
      "learning_rate": 2e-05,
      "loss": 0.9258,
      "step": 2680
    },
    {
      "epoch": 0.2391323673215397,
      "grad_norm": 0.4344216585159302,
      "learning_rate": 2e-05,
      "loss": 0.9047,
      "step": 2690
    },
    {
      "epoch": 0.2400213352297982,
      "grad_norm": 0.43349316716194153,
      "learning_rate": 2e-05,
      "loss": 0.8821,
      "step": 2700
    },
    {
      "epoch": 0.24091030313805673,
      "grad_norm": 0.42652013897895813,
      "learning_rate": 2e-05,
      "loss": 0.901,
      "step": 2710
    },
    {
      "epoch": 0.24179927104631524,
      "grad_norm": 0.43868887424468994,
      "learning_rate": 2e-05,
      "loss": 0.9002,
      "step": 2720
    },
    {
      "epoch": 0.24268823895457373,
      "grad_norm": 0.45351046323776245,
      "learning_rate": 2e-05,
      "loss": 0.8924,
      "step": 2730
    },
    {
      "epoch": 0.24357720686283224,
      "grad_norm": 0.4470480978488922,
      "learning_rate": 2e-05,
      "loss": 0.927,
      "step": 2740
    },
    {
      "epoch": 0.24446617477109076,
      "grad_norm": 0.45134204626083374,
      "learning_rate": 2e-05,
      "loss": 0.9118,
      "step": 2750
    },
    {
      "epoch": 0.24535514267934927,
      "grad_norm": 0.4969102740287781,
      "learning_rate": 2e-05,
      "loss": 0.9182,
      "step": 2760
    },
    {
      "epoch": 0.24624411058760778,
      "grad_norm": 0.4372037351131439,
      "learning_rate": 2e-05,
      "loss": 0.8988,
      "step": 2770
    },
    {
      "epoch": 0.2471330784958663,
      "grad_norm": 0.5171802639961243,
      "learning_rate": 2e-05,
      "loss": 0.9302,
      "step": 2780
    },
    {
      "epoch": 0.2480220464041248,
      "grad_norm": 0.40998777747154236,
      "learning_rate": 2e-05,
      "loss": 0.9283,
      "step": 2790
    },
    {
      "epoch": 0.24891101431238333,
      "grad_norm": 0.4390648603439331,
      "learning_rate": 2e-05,
      "loss": 0.8991,
      "step": 2800
    },
    {
      "epoch": 0.24979998222064184,
      "grad_norm": 0.4708453416824341,
      "learning_rate": 2e-05,
      "loss": 0.9045,
      "step": 2810
    },
    {
      "epoch": 0.25068895012890036,
      "grad_norm": 0.4334997832775116,
      "learning_rate": 2e-05,
      "loss": 0.9089,
      "step": 2820
    },
    {
      "epoch": 0.25157791803715884,
      "grad_norm": 0.47281280159950256,
      "learning_rate": 2e-05,
      "loss": 0.9176,
      "step": 2830
    },
    {
      "epoch": 0.2524668859454174,
      "grad_norm": 0.4157651960849762,
      "learning_rate": 2e-05,
      "loss": 0.9067,
      "step": 2840
    },
    {
      "epoch": 0.25335585385367587,
      "grad_norm": 0.43792033195495605,
      "learning_rate": 2e-05,
      "loss": 0.8985,
      "step": 2850
    },
    {
      "epoch": 0.2542448217619344,
      "grad_norm": 0.455698162317276,
      "learning_rate": 2e-05,
      "loss": 0.893,
      "step": 2860
    },
    {
      "epoch": 0.2551337896701929,
      "grad_norm": 0.4315062165260315,
      "learning_rate": 2e-05,
      "loss": 0.9184,
      "step": 2870
    },
    {
      "epoch": 0.25602275757845144,
      "grad_norm": 0.4467574954032898,
      "learning_rate": 2e-05,
      "loss": 0.9108,
      "step": 2880
    },
    {
      "epoch": 0.25691172548670993,
      "grad_norm": 0.5476098656654358,
      "learning_rate": 2e-05,
      "loss": 0.9343,
      "step": 2890
    },
    {
      "epoch": 0.2578006933949684,
      "grad_norm": 0.431232750415802,
      "learning_rate": 2e-05,
      "loss": 0.9115,
      "step": 2900
    },
    {
      "epoch": 0.25868966130322696,
      "grad_norm": 0.4723472595214844,
      "learning_rate": 2e-05,
      "loss": 0.9344,
      "step": 2910
    },
    {
      "epoch": 0.25957862921148545,
      "grad_norm": 0.4483966529369354,
      "learning_rate": 2e-05,
      "loss": 0.9202,
      "step": 2920
    },
    {
      "epoch": 0.260467597119744,
      "grad_norm": 0.44388502836227417,
      "learning_rate": 2e-05,
      "loss": 0.9355,
      "step": 2930
    },
    {
      "epoch": 0.2613565650280025,
      "grad_norm": 0.46426472067832947,
      "learning_rate": 2e-05,
      "loss": 0.9057,
      "step": 2940
    },
    {
      "epoch": 0.262245532936261,
      "grad_norm": 0.4535738229751587,
      "learning_rate": 2e-05,
      "loss": 0.9229,
      "step": 2950
    },
    {
      "epoch": 0.2631345008445195,
      "grad_norm": 0.41703757643699646,
      "learning_rate": 2e-05,
      "loss": 0.8998,
      "step": 2960
    },
    {
      "epoch": 0.26402346875277805,
      "grad_norm": 0.4153919816017151,
      "learning_rate": 2e-05,
      "loss": 0.9311,
      "step": 2970
    },
    {
      "epoch": 0.26491243666103653,
      "grad_norm": 0.46921271085739136,
      "learning_rate": 2e-05,
      "loss": 0.8981,
      "step": 2980
    },
    {
      "epoch": 0.2658014045692951,
      "grad_norm": 0.46266472339630127,
      "learning_rate": 2e-05,
      "loss": 0.8888,
      "step": 2990
    },
    {
      "epoch": 0.26669037247755356,
      "grad_norm": 0.44326508045196533,
      "learning_rate": 2e-05,
      "loss": 0.9197,
      "step": 3000
    },
    {
      "epoch": 0.26757934038581205,
      "grad_norm": 0.4367375075817108,
      "learning_rate": 2e-05,
      "loss": 0.8977,
      "step": 3010
    },
    {
      "epoch": 0.2684683082940706,
      "grad_norm": 0.4375791549682617,
      "learning_rate": 2e-05,
      "loss": 0.9079,
      "step": 3020
    },
    {
      "epoch": 0.2693572762023291,
      "grad_norm": 0.42957520484924316,
      "learning_rate": 2e-05,
      "loss": 0.9397,
      "step": 3030
    },
    {
      "epoch": 0.2702462441105876,
      "grad_norm": 0.43384110927581787,
      "learning_rate": 2e-05,
      "loss": 0.9004,
      "step": 3040
    },
    {
      "epoch": 0.2711352120188461,
      "grad_norm": 0.4357619881629944,
      "learning_rate": 2e-05,
      "loss": 0.9178,
      "step": 3050
    },
    {
      "epoch": 0.27202417992710465,
      "grad_norm": 0.45047247409820557,
      "learning_rate": 2e-05,
      "loss": 0.9025,
      "step": 3060
    },
    {
      "epoch": 0.27291314783536313,
      "grad_norm": 0.4550107419490814,
      "learning_rate": 2e-05,
      "loss": 0.9346,
      "step": 3070
    },
    {
      "epoch": 0.2738021157436217,
      "grad_norm": 0.45373380184173584,
      "learning_rate": 2e-05,
      "loss": 0.9069,
      "step": 3080
    },
    {
      "epoch": 0.27469108365188016,
      "grad_norm": 0.43642041087150574,
      "learning_rate": 2e-05,
      "loss": 0.9017,
      "step": 3090
    },
    {
      "epoch": 0.2755800515601387,
      "grad_norm": 0.421867698431015,
      "learning_rate": 2e-05,
      "loss": 0.8947,
      "step": 3100
    },
    {
      "epoch": 0.2764690194683972,
      "grad_norm": 0.4463402330875397,
      "learning_rate": 2e-05,
      "loss": 0.9013,
      "step": 3110
    },
    {
      "epoch": 0.2773579873766557,
      "grad_norm": 0.4239545464515686,
      "learning_rate": 2e-05,
      "loss": 0.9067,
      "step": 3120
    },
    {
      "epoch": 0.2782469552849142,
      "grad_norm": 0.4566056430339813,
      "learning_rate": 2e-05,
      "loss": 0.8866,
      "step": 3130
    },
    {
      "epoch": 0.2791359231931727,
      "grad_norm": 0.4473591148853302,
      "learning_rate": 2e-05,
      "loss": 0.9081,
      "step": 3140
    },
    {
      "epoch": 0.28002489110143125,
      "grad_norm": 0.41697973012924194,
      "learning_rate": 2e-05,
      "loss": 0.8995,
      "step": 3150
    },
    {
      "epoch": 0.28091385900968974,
      "grad_norm": 0.44013944268226624,
      "learning_rate": 2e-05,
      "loss": 0.8726,
      "step": 3160
    },
    {
      "epoch": 0.2818028269179483,
      "grad_norm": 0.4331235885620117,
      "learning_rate": 2e-05,
      "loss": 0.88,
      "step": 3170
    },
    {
      "epoch": 0.28269179482620677,
      "grad_norm": 0.4298301935195923,
      "learning_rate": 2e-05,
      "loss": 0.9093,
      "step": 3180
    },
    {
      "epoch": 0.2835807627344653,
      "grad_norm": 0.4418012797832489,
      "learning_rate": 2e-05,
      "loss": 0.8626,
      "step": 3190
    },
    {
      "epoch": 0.2844697306427238,
      "grad_norm": 0.4697241485118866,
      "learning_rate": 2e-05,
      "loss": 0.9156,
      "step": 3200
    },
    {
      "epoch": 0.28535869855098234,
      "grad_norm": 0.48490825295448303,
      "learning_rate": 2e-05,
      "loss": 0.9152,
      "step": 3210
    },
    {
      "epoch": 0.2862476664592408,
      "grad_norm": 0.49913570284843445,
      "learning_rate": 2e-05,
      "loss": 0.9076,
      "step": 3220
    },
    {
      "epoch": 0.2871366343674993,
      "grad_norm": 0.4422288239002228,
      "learning_rate": 2e-05,
      "loss": 0.9142,
      "step": 3230
    },
    {
      "epoch": 0.28802560227575785,
      "grad_norm": 0.43101540207862854,
      "learning_rate": 2e-05,
      "loss": 0.9156,
      "step": 3240
    },
    {
      "epoch": 0.28891457018401634,
      "grad_norm": 0.42662280797958374,
      "learning_rate": 2e-05,
      "loss": 0.9073,
      "step": 3250
    },
    {
      "epoch": 0.2898035380922749,
      "grad_norm": 0.4074089825153351,
      "learning_rate": 2e-05,
      "loss": 0.8995,
      "step": 3260
    },
    {
      "epoch": 0.29069250600053337,
      "grad_norm": 0.43171247839927673,
      "learning_rate": 2e-05,
      "loss": 0.9058,
      "step": 3270
    },
    {
      "epoch": 0.2915814739087919,
      "grad_norm": 0.42101308703422546,
      "learning_rate": 2e-05,
      "loss": 0.9116,
      "step": 3280
    },
    {
      "epoch": 0.2924704418170504,
      "grad_norm": 0.48281070590019226,
      "learning_rate": 2e-05,
      "loss": 0.9052,
      "step": 3290
    },
    {
      "epoch": 0.29335940972530894,
      "grad_norm": 0.4471592903137207,
      "learning_rate": 2e-05,
      "loss": 0.8821,
      "step": 3300
    },
    {
      "epoch": 0.2942483776335674,
      "grad_norm": 0.4851568341255188,
      "learning_rate": 2e-05,
      "loss": 0.9114,
      "step": 3310
    },
    {
      "epoch": 0.2951373455418259,
      "grad_norm": 0.4939979016780853,
      "learning_rate": 2e-05,
      "loss": 0.9202,
      "step": 3320
    },
    {
      "epoch": 0.29602631345008446,
      "grad_norm": 0.4668613076210022,
      "learning_rate": 2e-05,
      "loss": 0.9131,
      "step": 3330
    },
    {
      "epoch": 0.29691528135834294,
      "grad_norm": 0.44769349694252014,
      "learning_rate": 2e-05,
      "loss": 0.8822,
      "step": 3340
    },
    {
      "epoch": 0.2978042492666015,
      "grad_norm": 0.4628336727619171,
      "learning_rate": 2e-05,
      "loss": 0.9148,
      "step": 3350
    },
    {
      "epoch": 0.29869321717485997,
      "grad_norm": 0.45139649510383606,
      "learning_rate": 2e-05,
      "loss": 0.9101,
      "step": 3360
    },
    {
      "epoch": 0.2995821850831185,
      "grad_norm": 0.40553897619247437,
      "learning_rate": 2e-05,
      "loss": 0.9359,
      "step": 3370
    },
    {
      "epoch": 0.300471152991377,
      "grad_norm": 0.42898911237716675,
      "learning_rate": 2e-05,
      "loss": 0.8839,
      "step": 3380
    },
    {
      "epoch": 0.30136012089963554,
      "grad_norm": 0.45891132950782776,
      "learning_rate": 2e-05,
      "loss": 0.8768,
      "step": 3390
    },
    {
      "epoch": 0.30224908880789403,
      "grad_norm": 0.44235119223594666,
      "learning_rate": 2e-05,
      "loss": 0.8828,
      "step": 3400
    },
    {
      "epoch": 0.30313805671615257,
      "grad_norm": 0.4573551118373871,
      "learning_rate": 2e-05,
      "loss": 0.8964,
      "step": 3410
    },
    {
      "epoch": 0.30402702462441106,
      "grad_norm": 0.41882309317588806,
      "learning_rate": 2e-05,
      "loss": 0.8715,
      "step": 3420
    },
    {
      "epoch": 0.30491599253266954,
      "grad_norm": 0.43964776396751404,
      "learning_rate": 2e-05,
      "loss": 0.8944,
      "step": 3430
    },
    {
      "epoch": 0.3058049604409281,
      "grad_norm": 0.4233776330947876,
      "learning_rate": 2e-05,
      "loss": 0.8891,
      "step": 3440
    },
    {
      "epoch": 0.3066939283491866,
      "grad_norm": 0.4284922182559967,
      "learning_rate": 2e-05,
      "loss": 0.9195,
      "step": 3450
    },
    {
      "epoch": 0.3075828962574451,
      "grad_norm": 0.4250510632991791,
      "learning_rate": 2e-05,
      "loss": 0.8971,
      "step": 3460
    },
    {
      "epoch": 0.3084718641657036,
      "grad_norm": 0.4597267508506775,
      "learning_rate": 2e-05,
      "loss": 0.888,
      "step": 3470
    },
    {
      "epoch": 0.30936083207396214,
      "grad_norm": 0.4276660084724426,
      "learning_rate": 2e-05,
      "loss": 0.8956,
      "step": 3480
    },
    {
      "epoch": 0.31024979998222063,
      "grad_norm": 0.44020989537239075,
      "learning_rate": 2e-05,
      "loss": 0.899,
      "step": 3490
    },
    {
      "epoch": 0.3111387678904792,
      "grad_norm": 0.4431859850883484,
      "learning_rate": 2e-05,
      "loss": 0.8705,
      "step": 3500
    },
    {
      "epoch": 0.31202773579873766,
      "grad_norm": 0.47489258646965027,
      "learning_rate": 2e-05,
      "loss": 0.891,
      "step": 3510
    },
    {
      "epoch": 0.3129167037069962,
      "grad_norm": 0.4463152587413788,
      "learning_rate": 2e-05,
      "loss": 0.9123,
      "step": 3520
    },
    {
      "epoch": 0.3138056716152547,
      "grad_norm": 0.44237059354782104,
      "learning_rate": 2e-05,
      "loss": 0.8817,
      "step": 3530
    },
    {
      "epoch": 0.3146946395235132,
      "grad_norm": 0.6374925374984741,
      "learning_rate": 2e-05,
      "loss": 0.8715,
      "step": 3540
    },
    {
      "epoch": 0.3155836074317717,
      "grad_norm": 0.45814210176467896,
      "learning_rate": 2e-05,
      "loss": 0.8914,
      "step": 3550
    },
    {
      "epoch": 0.3164725753400302,
      "grad_norm": 0.44097962975502014,
      "learning_rate": 2e-05,
      "loss": 0.8942,
      "step": 3560
    },
    {
      "epoch": 0.31736154324828875,
      "grad_norm": 0.43970420956611633,
      "learning_rate": 2e-05,
      "loss": 0.8844,
      "step": 3570
    },
    {
      "epoch": 0.31825051115654723,
      "grad_norm": 0.4468035101890564,
      "learning_rate": 2e-05,
      "loss": 0.9067,
      "step": 3580
    },
    {
      "epoch": 0.3191394790648058,
      "grad_norm": 0.435278981924057,
      "learning_rate": 2e-05,
      "loss": 0.8743,
      "step": 3590
    },
    {
      "epoch": 0.32002844697306426,
      "grad_norm": 0.4380773603916168,
      "learning_rate": 2e-05,
      "loss": 0.9058,
      "step": 3600
    },
    {
      "epoch": 0.3209174148813228,
      "grad_norm": 0.42821770906448364,
      "learning_rate": 2e-05,
      "loss": 0.8924,
      "step": 3610
    },
    {
      "epoch": 0.3218063827895813,
      "grad_norm": 0.4328218698501587,
      "learning_rate": 2e-05,
      "loss": 0.9053,
      "step": 3620
    },
    {
      "epoch": 0.32269535069783983,
      "grad_norm": 0.43773573637008667,
      "learning_rate": 2e-05,
      "loss": 0.892,
      "step": 3630
    },
    {
      "epoch": 0.3235843186060983,
      "grad_norm": 0.44607359170913696,
      "learning_rate": 2e-05,
      "loss": 0.8936,
      "step": 3640
    },
    {
      "epoch": 0.3244732865143568,
      "grad_norm": 0.873892068862915,
      "learning_rate": 2e-05,
      "loss": 0.8776,
      "step": 3650
    },
    {
      "epoch": 0.32536225442261535,
      "grad_norm": 0.42603957653045654,
      "learning_rate": 2e-05,
      "loss": 0.9201,
      "step": 3660
    },
    {
      "epoch": 0.32625122233087384,
      "grad_norm": 0.43324801325798035,
      "learning_rate": 2e-05,
      "loss": 0.8921,
      "step": 3670
    },
    {
      "epoch": 0.3271401902391324,
      "grad_norm": 0.4143255650997162,
      "learning_rate": 2e-05,
      "loss": 0.9045,
      "step": 3680
    },
    {
      "epoch": 0.32802915814739086,
      "grad_norm": 0.41029876470565796,
      "learning_rate": 2e-05,
      "loss": 0.8793,
      "step": 3690
    },
    {
      "epoch": 0.3289181260556494,
      "grad_norm": 0.46699365973472595,
      "learning_rate": 2e-05,
      "loss": 0.8891,
      "step": 3700
    },
    {
      "epoch": 0.3298070939639079,
      "grad_norm": 0.4288537800312042,
      "learning_rate": 2e-05,
      "loss": 0.884,
      "step": 3710
    },
    {
      "epoch": 0.33069606187216644,
      "grad_norm": 0.43615034222602844,
      "learning_rate": 2e-05,
      "loss": 0.8716,
      "step": 3720
    },
    {
      "epoch": 0.3315850297804249,
      "grad_norm": 0.4557756781578064,
      "learning_rate": 2e-05,
      "loss": 0.8909,
      "step": 3730
    },
    {
      "epoch": 0.33247399768868346,
      "grad_norm": 0.48360738158226013,
      "learning_rate": 2e-05,
      "loss": 0.895,
      "step": 3740
    },
    {
      "epoch": 0.33336296559694195,
      "grad_norm": 0.4450119733810425,
      "learning_rate": 2e-05,
      "loss": 0.8558,
      "step": 3750
    },
    {
      "epoch": 0.33425193350520044,
      "grad_norm": 0.43574458360671997,
      "learning_rate": 2e-05,
      "loss": 0.8761,
      "step": 3760
    },
    {
      "epoch": 0.335140901413459,
      "grad_norm": 0.6523852944374084,
      "learning_rate": 2e-05,
      "loss": 0.9014,
      "step": 3770
    },
    {
      "epoch": 0.33602986932171747,
      "grad_norm": 0.4512973129749298,
      "learning_rate": 2e-05,
      "loss": 0.8831,
      "step": 3780
    },
    {
      "epoch": 0.336918837229976,
      "grad_norm": 0.38467535376548767,
      "learning_rate": 2e-05,
      "loss": 0.8922,
      "step": 3790
    },
    {
      "epoch": 0.3378078051382345,
      "grad_norm": 0.42140308022499084,
      "learning_rate": 2e-05,
      "loss": 0.8952,
      "step": 3800
    },
    {
      "epoch": 0.33869677304649304,
      "grad_norm": 0.7853624820709229,
      "learning_rate": 2e-05,
      "loss": 0.8755,
      "step": 3810
    },
    {
      "epoch": 0.3395857409547515,
      "grad_norm": 0.456306517124176,
      "learning_rate": 2e-05,
      "loss": 0.9122,
      "step": 3820
    },
    {
      "epoch": 0.34047470886301007,
      "grad_norm": 0.44224637746810913,
      "learning_rate": 2e-05,
      "loss": 0.879,
      "step": 3830
    },
    {
      "epoch": 0.34136367677126855,
      "grad_norm": 0.4817962050437927,
      "learning_rate": 2e-05,
      "loss": 0.8698,
      "step": 3840
    },
    {
      "epoch": 0.3422526446795271,
      "grad_norm": 0.4291815161705017,
      "learning_rate": 2e-05,
      "loss": 0.8907,
      "step": 3850
    },
    {
      "epoch": 0.3431416125877856,
      "grad_norm": 0.42189449071884155,
      "learning_rate": 2e-05,
      "loss": 0.8964,
      "step": 3860
    },
    {
      "epoch": 0.34403058049604407,
      "grad_norm": 0.4154491722583771,
      "learning_rate": 2e-05,
      "loss": 0.8631,
      "step": 3870
    },
    {
      "epoch": 0.3449195484043026,
      "grad_norm": 0.4724118709564209,
      "learning_rate": 2e-05,
      "loss": 0.888,
      "step": 3880
    },
    {
      "epoch": 0.3458085163125611,
      "grad_norm": 0.44710826873779297,
      "learning_rate": 2e-05,
      "loss": 0.8508,
      "step": 3890
    },
    {
      "epoch": 0.34669748422081964,
      "grad_norm": 0.42160674929618835,
      "learning_rate": 2e-05,
      "loss": 0.8947,
      "step": 3900
    },
    {
      "epoch": 0.3475864521290781,
      "grad_norm": 0.4110996425151825,
      "learning_rate": 2e-05,
      "loss": 0.855,
      "step": 3910
    },
    {
      "epoch": 0.34847542003733667,
      "grad_norm": 0.41057881712913513,
      "learning_rate": 2e-05,
      "loss": 0.8874,
      "step": 3920
    },
    {
      "epoch": 0.34936438794559516,
      "grad_norm": 0.44590920209884644,
      "learning_rate": 2e-05,
      "loss": 0.8577,
      "step": 3930
    },
    {
      "epoch": 0.3502533558538537,
      "grad_norm": 0.48844611644744873,
      "learning_rate": 2e-05,
      "loss": 0.8835,
      "step": 3940
    },
    {
      "epoch": 0.3511423237621122,
      "grad_norm": 0.41316306591033936,
      "learning_rate": 2e-05,
      "loss": 0.9094,
      "step": 3950
    },
    {
      "epoch": 0.3520312916703707,
      "grad_norm": 0.4613005518913269,
      "learning_rate": 2e-05,
      "loss": 0.8772,
      "step": 3960
    },
    {
      "epoch": 0.3529202595786292,
      "grad_norm": 0.4151155948638916,
      "learning_rate": 2e-05,
      "loss": 0.8842,
      "step": 3970
    },
    {
      "epoch": 0.3538092274868877,
      "grad_norm": 0.4251476228237152,
      "learning_rate": 2e-05,
      "loss": 0.8809,
      "step": 3980
    },
    {
      "epoch": 0.35469819539514624,
      "grad_norm": 0.4338812828063965,
      "learning_rate": 2e-05,
      "loss": 0.8785,
      "step": 3990
    },
    {
      "epoch": 0.35558716330340473,
      "grad_norm": 0.3898870646953583,
      "learning_rate": 2e-05,
      "loss": 0.8839,
      "step": 4000
    },
    {
      "epoch": 0.35647613121166327,
      "grad_norm": 0.4402524530887604,
      "learning_rate": 2e-05,
      "loss": 0.9019,
      "step": 4010
    },
    {
      "epoch": 0.35736509911992176,
      "grad_norm": 0.4277661442756653,
      "learning_rate": 2e-05,
      "loss": 0.8633,
      "step": 4020
    },
    {
      "epoch": 0.3582540670281803,
      "grad_norm": 0.43866926431655884,
      "learning_rate": 2e-05,
      "loss": 0.8769,
      "step": 4030
    },
    {
      "epoch": 0.3591430349364388,
      "grad_norm": 0.4379994869232178,
      "learning_rate": 2e-05,
      "loss": 0.8669,
      "step": 4040
    },
    {
      "epoch": 0.36003200284469733,
      "grad_norm": 0.4321524500846863,
      "learning_rate": 2e-05,
      "loss": 0.8585,
      "step": 4050
    },
    {
      "epoch": 0.3609209707529558,
      "grad_norm": 0.4352610111236572,
      "learning_rate": 2e-05,
      "loss": 0.908,
      "step": 4060
    },
    {
      "epoch": 0.3618099386612143,
      "grad_norm": 0.3967607021331787,
      "learning_rate": 2e-05,
      "loss": 0.8623,
      "step": 4070
    },
    {
      "epoch": 0.36269890656947285,
      "grad_norm": 0.4171207845211029,
      "learning_rate": 2e-05,
      "loss": 0.8628,
      "step": 4080
    },
    {
      "epoch": 0.36358787447773133,
      "grad_norm": 0.43129152059555054,
      "learning_rate": 2e-05,
      "loss": 0.8814,
      "step": 4090
    },
    {
      "epoch": 0.3644768423859899,
      "grad_norm": 0.40105265378952026,
      "learning_rate": 2e-05,
      "loss": 0.8453,
      "step": 4100
    },
    {
      "epoch": 0.36536581029424836,
      "grad_norm": 0.44715967774391174,
      "learning_rate": 2e-05,
      "loss": 0.861,
      "step": 4110
    },
    {
      "epoch": 0.3662547782025069,
      "grad_norm": 0.4838224947452545,
      "learning_rate": 2e-05,
      "loss": 0.8704,
      "step": 4120
    },
    {
      "epoch": 0.3671437461107654,
      "grad_norm": 0.4485843777656555,
      "learning_rate": 2e-05,
      "loss": 0.8699,
      "step": 4130
    },
    {
      "epoch": 0.36803271401902393,
      "grad_norm": 0.44495731592178345,
      "learning_rate": 2e-05,
      "loss": 0.8877,
      "step": 4140
    },
    {
      "epoch": 0.3689216819272824,
      "grad_norm": 0.423117071390152,
      "learning_rate": 2e-05,
      "loss": 0.8935,
      "step": 4150
    },
    {
      "epoch": 0.36981064983554096,
      "grad_norm": 0.44600898027420044,
      "learning_rate": 2e-05,
      "loss": 0.8904,
      "step": 4160
    },
    {
      "epoch": 0.37069961774379945,
      "grad_norm": 0.4504396319389343,
      "learning_rate": 2e-05,
      "loss": 0.8724,
      "step": 4170
    },
    {
      "epoch": 0.37158858565205793,
      "grad_norm": 0.4439404308795929,
      "learning_rate": 2e-05,
      "loss": 0.8791,
      "step": 4180
    },
    {
      "epoch": 0.3724775535603165,
      "grad_norm": 0.4300064742565155,
      "learning_rate": 2e-05,
      "loss": 0.878,
      "step": 4190
    },
    {
      "epoch": 0.37336652146857496,
      "grad_norm": 0.45836833119392395,
      "learning_rate": 2e-05,
      "loss": 0.8602,
      "step": 4200
    },
    {
      "epoch": 0.3742554893768335,
      "grad_norm": 0.46192407608032227,
      "learning_rate": 2e-05,
      "loss": 0.8806,
      "step": 4210
    },
    {
      "epoch": 0.375144457285092,
      "grad_norm": 0.44279104471206665,
      "learning_rate": 2e-05,
      "loss": 0.9066,
      "step": 4220
    },
    {
      "epoch": 0.37603342519335053,
      "grad_norm": 0.43591928482055664,
      "learning_rate": 2e-05,
      "loss": 0.8829,
      "step": 4230
    },
    {
      "epoch": 0.376922393101609,
      "grad_norm": 0.44853675365448,
      "learning_rate": 2e-05,
      "loss": 0.8816,
      "step": 4240
    },
    {
      "epoch": 0.37781136100986756,
      "grad_norm": 0.4316171109676361,
      "learning_rate": 2e-05,
      "loss": 0.8826,
      "step": 4250
    },
    {
      "epoch": 0.37870032891812605,
      "grad_norm": 0.43002820014953613,
      "learning_rate": 2e-05,
      "loss": 0.852,
      "step": 4260
    },
    {
      "epoch": 0.3795892968263846,
      "grad_norm": 0.4232701361179352,
      "learning_rate": 2e-05,
      "loss": 0.8465,
      "step": 4270
    },
    {
      "epoch": 0.3804782647346431,
      "grad_norm": 0.43845826387405396,
      "learning_rate": 2e-05,
      "loss": 0.8897,
      "step": 4280
    },
    {
      "epoch": 0.38136723264290157,
      "grad_norm": 0.42489296197891235,
      "learning_rate": 2e-05,
      "loss": 0.9105,
      "step": 4290
    },
    {
      "epoch": 0.3822562005511601,
      "grad_norm": 0.43635889887809753,
      "learning_rate": 2e-05,
      "loss": 0.8884,
      "step": 4300
    },
    {
      "epoch": 0.3831451684594186,
      "grad_norm": 0.5024684071540833,
      "learning_rate": 2e-05,
      "loss": 0.8887,
      "step": 4310
    },
    {
      "epoch": 0.38403413636767714,
      "grad_norm": 0.4134429097175598,
      "learning_rate": 2e-05,
      "loss": 0.8871,
      "step": 4320
    },
    {
      "epoch": 0.3849231042759356,
      "grad_norm": 0.45940929651260376,
      "learning_rate": 2e-05,
      "loss": 0.8854,
      "step": 4330
    },
    {
      "epoch": 0.38581207218419417,
      "grad_norm": 0.4452202618122101,
      "learning_rate": 2e-05,
      "loss": 0.9098,
      "step": 4340
    },
    {
      "epoch": 0.38670104009245265,
      "grad_norm": 0.44986090064048767,
      "learning_rate": 2e-05,
      "loss": 0.8769,
      "step": 4350
    },
    {
      "epoch": 0.3875900080007112,
      "grad_norm": 0.5838043093681335,
      "learning_rate": 2e-05,
      "loss": 0.8801,
      "step": 4360
    },
    {
      "epoch": 0.3884789759089697,
      "grad_norm": 0.4439884424209595,
      "learning_rate": 2e-05,
      "loss": 0.8663,
      "step": 4370
    },
    {
      "epoch": 0.3893679438172282,
      "grad_norm": 0.4562339782714844,
      "learning_rate": 2e-05,
      "loss": 0.8886,
      "step": 4380
    },
    {
      "epoch": 0.3902569117254867,
      "grad_norm": 4.9865522384643555,
      "learning_rate": 2e-05,
      "loss": 0.9079,
      "step": 4390
    },
    {
      "epoch": 0.3911458796337452,
      "grad_norm": 0.44887351989746094,
      "learning_rate": 2e-05,
      "loss": 0.8622,
      "step": 4400
    },
    {
      "epoch": 0.39203484754200374,
      "grad_norm": 0.4354223310947418,
      "learning_rate": 2e-05,
      "loss": 0.872,
      "step": 4410
    },
    {
      "epoch": 0.3929238154502622,
      "grad_norm": 0.4442330598831177,
      "learning_rate": 2e-05,
      "loss": 0.8887,
      "step": 4420
    },
    {
      "epoch": 0.39381278335852077,
      "grad_norm": 0.4355458617210388,
      "learning_rate": 2e-05,
      "loss": 0.8412,
      "step": 4430
    },
    {
      "epoch": 0.39470175126677925,
      "grad_norm": 0.43551182746887207,
      "learning_rate": 2e-05,
      "loss": 0.8699,
      "step": 4440
    },
    {
      "epoch": 0.3955907191750378,
      "grad_norm": 0.41794008016586304,
      "learning_rate": 2e-05,
      "loss": 0.8588,
      "step": 4450
    },
    {
      "epoch": 0.3964796870832963,
      "grad_norm": 0.41012197732925415,
      "learning_rate": 2e-05,
      "loss": 0.8749,
      "step": 4460
    },
    {
      "epoch": 0.3973686549915548,
      "grad_norm": 0.4389755427837372,
      "learning_rate": 2e-05,
      "loss": 0.8681,
      "step": 4470
    },
    {
      "epoch": 0.3982576228998133,
      "grad_norm": 0.4350631833076477,
      "learning_rate": 2e-05,
      "loss": 0.8909,
      "step": 4480
    },
    {
      "epoch": 0.39914659080807185,
      "grad_norm": 0.4518719017505646,
      "learning_rate": 2e-05,
      "loss": 0.8787,
      "step": 4490
    },
    {
      "epoch": 0.40003555871633034,
      "grad_norm": 0.4485830068588257,
      "learning_rate": 2e-05,
      "loss": 0.8876,
      "step": 4500
    },
    {
      "epoch": 0.40092452662458883,
      "grad_norm": 0.4184865653514862,
      "learning_rate": 2e-05,
      "loss": 0.868,
      "step": 4510
    },
    {
      "epoch": 0.40181349453284737,
      "grad_norm": 0.4456641972064972,
      "learning_rate": 2e-05,
      "loss": 0.8703,
      "step": 4520
    },
    {
      "epoch": 0.40270246244110586,
      "grad_norm": 0.4468088448047638,
      "learning_rate": 2e-05,
      "loss": 0.8584,
      "step": 4530
    },
    {
      "epoch": 0.4035914303493644,
      "grad_norm": 0.5828477144241333,
      "learning_rate": 2e-05,
      "loss": 0.8686,
      "step": 4540
    },
    {
      "epoch": 0.4044803982576229,
      "grad_norm": 0.42055702209472656,
      "learning_rate": 2e-05,
      "loss": 0.8985,
      "step": 4550
    },
    {
      "epoch": 0.40536936616588143,
      "grad_norm": 0.44557300209999084,
      "learning_rate": 2e-05,
      "loss": 0.8734,
      "step": 4560
    },
    {
      "epoch": 0.4062583340741399,
      "grad_norm": 0.4572277367115021,
      "learning_rate": 2e-05,
      "loss": 0.856,
      "step": 4570
    },
    {
      "epoch": 0.40714730198239846,
      "grad_norm": 0.44720762968063354,
      "learning_rate": 2e-05,
      "loss": 0.8514,
      "step": 4580
    },
    {
      "epoch": 0.40803626989065694,
      "grad_norm": 0.4495958983898163,
      "learning_rate": 2e-05,
      "loss": 0.879,
      "step": 4590
    },
    {
      "epoch": 0.4089252377989155,
      "grad_norm": 0.4514221251010895,
      "learning_rate": 2e-05,
      "loss": 0.8821,
      "step": 4600
    },
    {
      "epoch": 0.409814205707174,
      "grad_norm": 0.43793654441833496,
      "learning_rate": 2e-05,
      "loss": 0.8655,
      "step": 4610
    },
    {
      "epoch": 0.41070317361543246,
      "grad_norm": 0.4157227873802185,
      "learning_rate": 2e-05,
      "loss": 0.8716,
      "step": 4620
    },
    {
      "epoch": 0.411592141523691,
      "grad_norm": 0.4308115541934967,
      "learning_rate": 2e-05,
      "loss": 0.8387,
      "step": 4630
    },
    {
      "epoch": 0.4124811094319495,
      "grad_norm": 0.4266626834869385,
      "learning_rate": 2e-05,
      "loss": 0.8682,
      "step": 4640
    },
    {
      "epoch": 0.41337007734020803,
      "grad_norm": 0.4247201085090637,
      "learning_rate": 2e-05,
      "loss": 0.8683,
      "step": 4650
    },
    {
      "epoch": 0.4142590452484665,
      "grad_norm": 0.4056819975376129,
      "learning_rate": 2e-05,
      "loss": 0.8825,
      "step": 4660
    },
    {
      "epoch": 0.41514801315672506,
      "grad_norm": 0.4517332911491394,
      "learning_rate": 2e-05,
      "loss": 0.8686,
      "step": 4670
    },
    {
      "epoch": 0.41603698106498355,
      "grad_norm": 0.4322936534881592,
      "learning_rate": 2e-05,
      "loss": 0.8572,
      "step": 4680
    },
    {
      "epoch": 0.4169259489732421,
      "grad_norm": 0.44361260533332825,
      "learning_rate": 2e-05,
      "loss": 0.8604,
      "step": 4690
    },
    {
      "epoch": 0.4178149168815006,
      "grad_norm": 0.4788258969783783,
      "learning_rate": 2e-05,
      "loss": 0.8605,
      "step": 4700
    },
    {
      "epoch": 0.41870388478975906,
      "grad_norm": 0.6463165879249573,
      "learning_rate": 2e-05,
      "loss": 0.8691,
      "step": 4710
    },
    {
      "epoch": 0.4195928526980176,
      "grad_norm": 0.443398654460907,
      "learning_rate": 2e-05,
      "loss": 0.8503,
      "step": 4720
    },
    {
      "epoch": 0.4204818206062761,
      "grad_norm": 0.4216567873954773,
      "learning_rate": 2e-05,
      "loss": 0.8437,
      "step": 4730
    },
    {
      "epoch": 0.42137078851453463,
      "grad_norm": 0.4536304473876953,
      "learning_rate": 2e-05,
      "loss": 0.8608,
      "step": 4740
    },
    {
      "epoch": 0.4222597564227931,
      "grad_norm": 0.3989715278148651,
      "learning_rate": 2e-05,
      "loss": 0.8607,
      "step": 4750
    },
    {
      "epoch": 0.42314872433105166,
      "grad_norm": 0.4211702346801758,
      "learning_rate": 2e-05,
      "loss": 0.8684,
      "step": 4760
    },
    {
      "epoch": 0.42403769223931015,
      "grad_norm": 0.7876825928688049,
      "learning_rate": 2e-05,
      "loss": 0.8478,
      "step": 4770
    },
    {
      "epoch": 0.4249266601475687,
      "grad_norm": 0.43140843510627747,
      "learning_rate": 2e-05,
      "loss": 0.8726,
      "step": 4780
    },
    {
      "epoch": 0.4258156280558272,
      "grad_norm": 0.4446166157722473,
      "learning_rate": 2e-05,
      "loss": 0.8589,
      "step": 4790
    },
    {
      "epoch": 0.4267045959640857,
      "grad_norm": 0.43559712171554565,
      "learning_rate": 2e-05,
      "loss": 0.8362,
      "step": 4800
    },
    {
      "epoch": 0.4275935638723442,
      "grad_norm": 0.4382726550102234,
      "learning_rate": 2e-05,
      "loss": 0.8756,
      "step": 4810
    },
    {
      "epoch": 0.4284825317806027,
      "grad_norm": 0.41629907488822937,
      "learning_rate": 2e-05,
      "loss": 0.8606,
      "step": 4820
    },
    {
      "epoch": 0.42937149968886124,
      "grad_norm": 0.44978195428848267,
      "learning_rate": 2e-05,
      "loss": 0.8606,
      "step": 4830
    },
    {
      "epoch": 0.4302604675971197,
      "grad_norm": 0.44057467579841614,
      "learning_rate": 2e-05,
      "loss": 0.8597,
      "step": 4840
    },
    {
      "epoch": 0.43114943550537826,
      "grad_norm": 0.45374205708503723,
      "learning_rate": 2e-05,
      "loss": 0.8448,
      "step": 4850
    },
    {
      "epoch": 0.43203840341363675,
      "grad_norm": 0.4728107154369354,
      "learning_rate": 2e-05,
      "loss": 0.8879,
      "step": 4860
    },
    {
      "epoch": 0.4329273713218953,
      "grad_norm": 0.41493260860443115,
      "learning_rate": 2e-05,
      "loss": 0.8836,
      "step": 4870
    },
    {
      "epoch": 0.4338163392301538,
      "grad_norm": 0.42182251811027527,
      "learning_rate": 2e-05,
      "loss": 0.8978,
      "step": 4880
    },
    {
      "epoch": 0.4347053071384123,
      "grad_norm": 0.4477763772010803,
      "learning_rate": 2e-05,
      "loss": 0.8756,
      "step": 4890
    },
    {
      "epoch": 0.4355942750466708,
      "grad_norm": 0.4344884157180786,
      "learning_rate": 2e-05,
      "loss": 0.8681,
      "step": 4900
    },
    {
      "epoch": 0.43648324295492935,
      "grad_norm": 0.41144901514053345,
      "learning_rate": 2e-05,
      "loss": 0.8607,
      "step": 4910
    },
    {
      "epoch": 0.43737221086318784,
      "grad_norm": 0.4463854730129242,
      "learning_rate": 2e-05,
      "loss": 0.8637,
      "step": 4920
    },
    {
      "epoch": 0.4382611787714463,
      "grad_norm": 0.43071019649505615,
      "learning_rate": 2e-05,
      "loss": 0.865,
      "step": 4930
    },
    {
      "epoch": 0.43915014667970487,
      "grad_norm": 0.4892447292804718,
      "learning_rate": 2e-05,
      "loss": 0.8573,
      "step": 4940
    },
    {
      "epoch": 0.44003911458796335,
      "grad_norm": 0.4109754264354706,
      "learning_rate": 2e-05,
      "loss": 0.8639,
      "step": 4950
    },
    {
      "epoch": 0.4409280824962219,
      "grad_norm": 0.44643068313598633,
      "learning_rate": 2e-05,
      "loss": 0.8698,
      "step": 4960
    },
    {
      "epoch": 0.4418170504044804,
      "grad_norm": 0.4940479099750519,
      "learning_rate": 2e-05,
      "loss": 0.8624,
      "step": 4970
    },
    {
      "epoch": 0.4427060183127389,
      "grad_norm": 0.43415969610214233,
      "learning_rate": 2e-05,
      "loss": 0.869,
      "step": 4980
    },
    {
      "epoch": 0.4435949862209974,
      "grad_norm": 0.5503201484680176,
      "learning_rate": 2e-05,
      "loss": 0.8601,
      "step": 4990
    },
    {
      "epoch": 0.44448395412925595,
      "grad_norm": 0.4226771891117096,
      "learning_rate": 2e-05,
      "loss": 0.8824,
      "step": 5000
    },
    {
      "epoch": 0.44537292203751444,
      "grad_norm": 0.4197985529899597,
      "learning_rate": 2e-05,
      "loss": 0.8752,
      "step": 5010
    },
    {
      "epoch": 0.446261889945773,
      "grad_norm": 0.4497617781162262,
      "learning_rate": 2e-05,
      "loss": 0.8828,
      "step": 5020
    },
    {
      "epoch": 0.44715085785403147,
      "grad_norm": 0.43590760231018066,
      "learning_rate": 2e-05,
      "loss": 0.8753,
      "step": 5030
    },
    {
      "epoch": 0.44803982576228996,
      "grad_norm": 0.4442792534828186,
      "learning_rate": 2e-05,
      "loss": 0.8632,
      "step": 5040
    },
    {
      "epoch": 0.4489287936705485,
      "grad_norm": 0.43888193368911743,
      "learning_rate": 2e-05,
      "loss": 0.8846,
      "step": 5050
    },
    {
      "epoch": 0.449817761578807,
      "grad_norm": 0.4566144645214081,
      "learning_rate": 2e-05,
      "loss": 0.8548,
      "step": 5060
    },
    {
      "epoch": 0.4507067294870655,
      "grad_norm": 0.45631855726242065,
      "learning_rate": 2e-05,
      "loss": 0.8472,
      "step": 5070
    },
    {
      "epoch": 0.451595697395324,
      "grad_norm": 0.43353480100631714,
      "learning_rate": 2e-05,
      "loss": 0.8479,
      "step": 5080
    },
    {
      "epoch": 0.45248466530358256,
      "grad_norm": 0.4151061475276947,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.8754,
      "step": 5090
    },
    {
      "epoch": 0.45337363321184104,
      "grad_norm": 0.44305944442749023,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.8542,
      "step": 5100
    },
    {
      "epoch": 0.4542626011200996,
      "grad_norm": 0.42879927158355713,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.8324,
      "step": 5110
    },
    {
      "epoch": 0.45515156902835807,
      "grad_norm": 0.4453643262386322,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.842,
      "step": 5120
    },
    {
      "epoch": 0.4560405369366166,
      "grad_norm": 0.42254212498664856,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.8591,
      "step": 5130
    },
    {
      "epoch": 0.4569295048448751,
      "grad_norm": 0.4523744583129883,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.8836,
      "step": 5140
    },
    {
      "epoch": 0.4578184727531336,
      "grad_norm": 0.428847074508667,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.8673,
      "step": 5150
    },
    {
      "epoch": 0.45870744066139213,
      "grad_norm": 0.8905320167541504,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.8478,
      "step": 5160
    },
    {
      "epoch": 0.4595964085696506,
      "grad_norm": 0.4058988392353058,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.8567,
      "step": 5170
    },
    {
      "epoch": 0.46048537647790916,
      "grad_norm": 0.41792622208595276,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.8457,
      "step": 5180
    },
    {
      "epoch": 0.46137434438616765,
      "grad_norm": 0.44021958112716675,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.8888,
      "step": 5190
    },
    {
      "epoch": 0.4622633122944262,
      "grad_norm": 0.4766295254230499,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.8593,
      "step": 5200
    },
    {
      "epoch": 0.4631522802026847,
      "grad_norm": 0.4465012550354004,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.8708,
      "step": 5210
    },
    {
      "epoch": 0.4640412481109432,
      "grad_norm": 0.4719880521297455,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.8661,
      "step": 5220
    },
    {
      "epoch": 0.4649302160192017,
      "grad_norm": 0.4134722054004669,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.8552,
      "step": 5230
    },
    {
      "epoch": 0.46581918392746025,
      "grad_norm": 0.4404371976852417,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.8686,
      "step": 5240
    },
    {
      "epoch": 0.46670815183571873,
      "grad_norm": 0.4459529221057892,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.8606,
      "step": 5250
    },
    {
      "epoch": 0.4675971197439772,
      "grad_norm": 0.41834092140197754,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.8511,
      "step": 5260
    },
    {
      "epoch": 0.46848608765223576,
      "grad_norm": 0.4472412168979645,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.8651,
      "step": 5270
    },
    {
      "epoch": 0.46937505556049425,
      "grad_norm": 0.43029430508613586,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.8468,
      "step": 5280
    },
    {
      "epoch": 0.4702640234687528,
      "grad_norm": 0.45277583599090576,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.8473,
      "step": 5290
    },
    {
      "epoch": 0.4711529913770113,
      "grad_norm": 0.547594428062439,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.8448,
      "step": 5300
    },
    {
      "epoch": 0.4720419592852698,
      "grad_norm": 0.41689711809158325,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.8791,
      "step": 5310
    },
    {
      "epoch": 0.4729309271935283,
      "grad_norm": 0.4195074737071991,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.8483,
      "step": 5320
    },
    {
      "epoch": 0.47381989510178685,
      "grad_norm": 0.40814030170440674,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.8647,
      "step": 5330
    },
    {
      "epoch": 0.47470886301004533,
      "grad_norm": 0.4236510396003723,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.8405,
      "step": 5340
    },
    {
      "epoch": 0.4755978309183039,
      "grad_norm": 0.4169195890426636,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.8496,
      "step": 5350
    },
    {
      "epoch": 0.47648679882656236,
      "grad_norm": 0.39508602023124695,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.8478,
      "step": 5360
    },
    {
      "epoch": 0.47737576673482085,
      "grad_norm": 0.39964550733566284,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.8507,
      "step": 5370
    },
    {
      "epoch": 0.4782647346430794,
      "grad_norm": 0.41549521684646606,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.8412,
      "step": 5380
    },
    {
      "epoch": 0.4791537025513379,
      "grad_norm": 0.4259960651397705,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.8712,
      "step": 5390
    },
    {
      "epoch": 0.4800426704595964,
      "grad_norm": 0.4117063879966736,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.8447,
      "step": 5400
    },
    {
      "epoch": 0.4809316383678549,
      "grad_norm": 0.4396529793739319,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.8267,
      "step": 5410
    },
    {
      "epoch": 0.48182060627611345,
      "grad_norm": 0.44393667578697205,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.8393,
      "step": 5420
    },
    {
      "epoch": 0.48270957418437194,
      "grad_norm": 0.40403181314468384,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.8272,
      "step": 5430
    },
    {
      "epoch": 0.4835985420926305,
      "grad_norm": 0.4234180450439453,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.8604,
      "step": 5440
    },
    {
      "epoch": 0.48448751000088897,
      "grad_norm": 0.40846946835517883,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.8188,
      "step": 5450
    },
    {
      "epoch": 0.48537647790914745,
      "grad_norm": 0.4168575704097748,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.8307,
      "step": 5460
    },
    {
      "epoch": 0.486265445817406,
      "grad_norm": 0.41752156615257263,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.8455,
      "step": 5470
    },
    {
      "epoch": 0.4871544137256645,
      "grad_norm": 0.41761812567710876,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.8366,
      "step": 5480
    },
    {
      "epoch": 0.488043381633923,
      "grad_norm": 0.43264442682266235,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.8491,
      "step": 5490
    },
    {
      "epoch": 0.4889323495421815,
      "grad_norm": 0.4171850383281708,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.8463,
      "step": 5500
    },
    {
      "epoch": 0.48982131745044005,
      "grad_norm": 0.43236735463142395,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.8422,
      "step": 5510
    },
    {
      "epoch": 0.49071028535869854,
      "grad_norm": 0.3773476481437683,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.8181,
      "step": 5520
    },
    {
      "epoch": 0.4915992532669571,
      "grad_norm": 0.38137203454971313,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.8367,
      "step": 5530
    },
    {
      "epoch": 0.49248822117521557,
      "grad_norm": 0.42881423234939575,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.8605,
      "step": 5540
    },
    {
      "epoch": 0.4933771890834741,
      "grad_norm": 0.4339230954647064,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.8455,
      "step": 5550
    },
    {
      "epoch": 0.4942661569917326,
      "grad_norm": 0.6631712317466736,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.8353,
      "step": 5560
    },
    {
      "epoch": 0.4951551248999911,
      "grad_norm": 0.42285656929016113,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.8174,
      "step": 5570
    },
    {
      "epoch": 0.4960440928082496,
      "grad_norm": 0.42415425181388855,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.8546,
      "step": 5580
    },
    {
      "epoch": 0.4969330607165081,
      "grad_norm": 0.43446242809295654,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.8092,
      "step": 5590
    },
    {
      "epoch": 0.4975553382522891,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.028524470069439513,
      "train_runtime": 1767.174,
      "train_samples_per_second": 38.006,
      "train_steps_per_second": 3.167
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}