{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9951106765045782,
  "eval_steps": 500,
  "global_step": 11194,
  "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": 0.4152100682258606,
      "learning_rate": 2e-05,
      "loss": 0.8754,
      "step": 5090
    },
    {
      "epoch": 0.45337363321184104,
      "grad_norm": 0.44853463768959045,
      "learning_rate": 2e-05,
      "loss": 0.8542,
      "step": 5100
    },
    {
      "epoch": 0.4542626011200996,
      "grad_norm": 0.4312889873981476,
      "learning_rate": 2e-05,
      "loss": 0.8323,
      "step": 5110
    },
    {
      "epoch": 0.45515156902835807,
      "grad_norm": 0.4685893654823303,
      "learning_rate": 2e-05,
      "loss": 0.8421,
      "step": 5120
    },
    {
      "epoch": 0.4560405369366166,
      "grad_norm": 0.41682958602905273,
      "learning_rate": 2e-05,
      "loss": 0.8592,
      "step": 5130
    },
    {
      "epoch": 0.4569295048448751,
      "grad_norm": 0.453452467918396,
      "learning_rate": 2e-05,
      "loss": 0.8836,
      "step": 5140
    },
    {
      "epoch": 0.4578184727531336,
      "grad_norm": 0.43062254786491394,
      "learning_rate": 2e-05,
      "loss": 0.8674,
      "step": 5150
    },
    {
      "epoch": 0.45870744066139213,
      "grad_norm": 0.44043225049972534,
      "learning_rate": 2e-05,
      "loss": 0.8481,
      "step": 5160
    },
    {
      "epoch": 0.4595964085696506,
      "grad_norm": 0.40919843316078186,
      "learning_rate": 2e-05,
      "loss": 0.8572,
      "step": 5170
    },
    {
      "epoch": 0.46048537647790916,
      "grad_norm": 0.41787055134773254,
      "learning_rate": 2e-05,
      "loss": 0.8465,
      "step": 5180
    },
    {
      "epoch": 0.46137434438616765,
      "grad_norm": 0.47128838300704956,
      "learning_rate": 2e-05,
      "loss": 0.8899,
      "step": 5190
    },
    {
      "epoch": 0.4622633122944262,
      "grad_norm": 0.41438183188438416,
      "learning_rate": 2e-05,
      "loss": 0.8608,
      "step": 5200
    },
    {
      "epoch": 0.4631522802026847,
      "grad_norm": 0.45180341601371765,
      "learning_rate": 2e-05,
      "loss": 0.8725,
      "step": 5210
    },
    {
      "epoch": 0.4640412481109432,
      "grad_norm": 0.49082064628601074,
      "learning_rate": 2e-05,
      "loss": 0.8682,
      "step": 5220
    },
    {
      "epoch": 0.4649302160192017,
      "grad_norm": 0.4286251366138458,
      "learning_rate": 2e-05,
      "loss": 0.8574,
      "step": 5230
    },
    {
      "epoch": 0.46581918392746025,
      "grad_norm": 0.43935835361480713,
      "learning_rate": 2e-05,
      "loss": 0.8711,
      "step": 5240
    },
    {
      "epoch": 0.46670815183571873,
      "grad_norm": 0.4557168781757355,
      "learning_rate": 2e-05,
      "loss": 0.8637,
      "step": 5250
    },
    {
      "epoch": 0.4675971197439772,
      "grad_norm": 0.4608123302459717,
      "learning_rate": 2e-05,
      "loss": 0.8545,
      "step": 5260
    },
    {
      "epoch": 0.46848608765223576,
      "grad_norm": 0.46600890159606934,
      "learning_rate": 2e-05,
      "loss": 0.8688,
      "step": 5270
    },
    {
      "epoch": 0.46937505556049425,
      "grad_norm": 0.4447679817676544,
      "learning_rate": 2e-05,
      "loss": 0.8514,
      "step": 5280
    },
    {
      "epoch": 0.4702640234687528,
      "grad_norm": 0.47428056597709656,
      "learning_rate": 2e-05,
      "loss": 0.8524,
      "step": 5290
    },
    {
      "epoch": 0.4711529913770113,
      "grad_norm": 0.4861878752708435,
      "learning_rate": 2e-05,
      "loss": 0.8506,
      "step": 5300
    },
    {
      "epoch": 0.4720419592852698,
      "grad_norm": 0.45437613129615784,
      "learning_rate": 2e-05,
      "loss": 0.8855,
      "step": 5310
    },
    {
      "epoch": 0.4729309271935283,
      "grad_norm": 0.4574771523475647,
      "learning_rate": 2e-05,
      "loss": 0.8555,
      "step": 5320
    },
    {
      "epoch": 0.47381989510178685,
      "grad_norm": 0.4115237295627594,
      "learning_rate": 2e-05,
      "loss": 0.8722,
      "step": 5330
    },
    {
      "epoch": 0.47470886301004533,
      "grad_norm": 0.43305709958076477,
      "learning_rate": 2e-05,
      "loss": 0.8485,
      "step": 5340
    },
    {
      "epoch": 0.4755978309183039,
      "grad_norm": 0.4115385115146637,
      "learning_rate": 2e-05,
      "loss": 0.8578,
      "step": 5350
    },
    {
      "epoch": 0.47648679882656236,
      "grad_norm": 0.40595391392707825,
      "learning_rate": 2e-05,
      "loss": 0.8564,
      "step": 5360
    },
    {
      "epoch": 0.47737576673482085,
      "grad_norm": 0.41054949164390564,
      "learning_rate": 2e-05,
      "loss": 0.8598,
      "step": 5370
    },
    {
      "epoch": 0.4782647346430794,
      "grad_norm": 0.43531540036201477,
      "learning_rate": 2e-05,
      "loss": 0.8516,
      "step": 5380
    },
    {
      "epoch": 0.4791537025513379,
      "grad_norm": 0.45188960433006287,
      "learning_rate": 2e-05,
      "loss": 0.8818,
      "step": 5390
    },
    {
      "epoch": 0.4800426704595964,
      "grad_norm": 0.4307803213596344,
      "learning_rate": 2e-05,
      "loss": 0.8552,
      "step": 5400
    },
    {
      "epoch": 0.4809316383678549,
      "grad_norm": 0.4749660789966583,
      "learning_rate": 2e-05,
      "loss": 0.8381,
      "step": 5410
    },
    {
      "epoch": 0.48182060627611345,
      "grad_norm": 0.4446052610874176,
      "learning_rate": 2e-05,
      "loss": 0.8513,
      "step": 5420
    },
    {
      "epoch": 0.48270957418437194,
      "grad_norm": 0.43527573347091675,
      "learning_rate": 2e-05,
      "loss": 0.8401,
      "step": 5430
    },
    {
      "epoch": 0.4835985420926305,
      "grad_norm": 0.45180532336235046,
      "learning_rate": 2e-05,
      "loss": 0.8726,
      "step": 5440
    },
    {
      "epoch": 0.48448751000088897,
      "grad_norm": 0.4339911639690399,
      "learning_rate": 2e-05,
      "loss": 0.8318,
      "step": 5450
    },
    {
      "epoch": 0.48537647790914745,
      "grad_norm": 0.43470919132232666,
      "learning_rate": 2e-05,
      "loss": 0.8427,
      "step": 5460
    },
    {
      "epoch": 0.486265445817406,
      "grad_norm": 0.424111008644104,
      "learning_rate": 2e-05,
      "loss": 0.8587,
      "step": 5470
    },
    {
      "epoch": 0.4871544137256645,
      "grad_norm": 0.44621583819389343,
      "learning_rate": 2e-05,
      "loss": 0.8509,
      "step": 5480
    },
    {
      "epoch": 0.488043381633923,
      "grad_norm": 0.4414016604423523,
      "learning_rate": 2e-05,
      "loss": 0.8644,
      "step": 5490
    },
    {
      "epoch": 0.4889323495421815,
      "grad_norm": 0.43371519446372986,
      "learning_rate": 2e-05,
      "loss": 0.8594,
      "step": 5500
    },
    {
      "epoch": 0.48982131745044005,
      "grad_norm": 0.45571044087409973,
      "learning_rate": 2e-05,
      "loss": 0.8554,
      "step": 5510
    },
    {
      "epoch": 0.49071028535869854,
      "grad_norm": 0.3886682391166687,
      "learning_rate": 2e-05,
      "loss": 0.8309,
      "step": 5520
    },
    {
      "epoch": 0.4915992532669571,
      "grad_norm": 0.40275195240974426,
      "learning_rate": 2e-05,
      "loss": 0.8507,
      "step": 5530
    },
    {
      "epoch": 0.49248822117521557,
      "grad_norm": 0.4461636543273926,
      "learning_rate": 2e-05,
      "loss": 0.8755,
      "step": 5540
    },
    {
      "epoch": 0.4933771890834741,
      "grad_norm": 0.4367828965187073,
      "learning_rate": 2e-05,
      "loss": 0.8594,
      "step": 5550
    },
    {
      "epoch": 0.4942661569917326,
      "grad_norm": 0.4549225866794586,
      "learning_rate": 2e-05,
      "loss": 0.8506,
      "step": 5560
    },
    {
      "epoch": 0.4951551248999911,
      "grad_norm": 0.44123366475105286,
      "learning_rate": 2e-05,
      "loss": 0.8295,
      "step": 5570
    },
    {
      "epoch": 0.4960440928082496,
      "grad_norm": 0.4385070502758026,
      "learning_rate": 2e-05,
      "loss": 0.8682,
      "step": 5580
    },
    {
      "epoch": 0.4969330607165081,
      "grad_norm": 0.47192367911338806,
      "learning_rate": 2e-05,
      "loss": 0.8188,
      "step": 5590
    },
    {
      "epoch": 0.49782202862476665,
      "grad_norm": 0.43115681409835815,
      "learning_rate": 2e-05,
      "loss": 0.8302,
      "step": 5600
    },
    {
      "epoch": 0.49871099653302514,
      "grad_norm": 0.41383543610572815,
      "learning_rate": 2e-05,
      "loss": 0.835,
      "step": 5610
    },
    {
      "epoch": 0.4995999644412837,
      "grad_norm": 0.42274683713912964,
      "learning_rate": 2e-05,
      "loss": 0.853,
      "step": 5620
    },
    {
      "epoch": 0.5004889323495422,
      "grad_norm": 0.42314282059669495,
      "learning_rate": 2e-05,
      "loss": 0.8373,
      "step": 5630
    },
    {
      "epoch": 0.5013779002578007,
      "grad_norm": 1.1123954057693481,
      "learning_rate": 2e-05,
      "loss": 0.8351,
      "step": 5640
    },
    {
      "epoch": 0.5022668681660593,
      "grad_norm": 0.4355487823486328,
      "learning_rate": 2e-05,
      "loss": 0.8471,
      "step": 5650
    },
    {
      "epoch": 0.5031558360743177,
      "grad_norm": 0.4739534854888916,
      "learning_rate": 2e-05,
      "loss": 0.8535,
      "step": 5660
    },
    {
      "epoch": 0.5040448039825762,
      "grad_norm": 0.41240936517715454,
      "learning_rate": 2e-05,
      "loss": 0.8691,
      "step": 5670
    },
    {
      "epoch": 0.5049337718908348,
      "grad_norm": 0.45636141300201416,
      "learning_rate": 2e-05,
      "loss": 0.8374,
      "step": 5680
    },
    {
      "epoch": 0.5058227397990932,
      "grad_norm": 0.46528834104537964,
      "learning_rate": 2e-05,
      "loss": 0.8447,
      "step": 5690
    },
    {
      "epoch": 0.5067117077073517,
      "grad_norm": 0.450166255235672,
      "learning_rate": 2e-05,
      "loss": 0.8625,
      "step": 5700
    },
    {
      "epoch": 0.5076006756156103,
      "grad_norm": 0.43522635102272034,
      "learning_rate": 2e-05,
      "loss": 0.8559,
      "step": 5710
    },
    {
      "epoch": 0.5084896435238688,
      "grad_norm": 0.4199116826057434,
      "learning_rate": 2e-05,
      "loss": 0.8477,
      "step": 5720
    },
    {
      "epoch": 0.5093786114321273,
      "grad_norm": 0.43494316935539246,
      "learning_rate": 2e-05,
      "loss": 0.8785,
      "step": 5730
    },
    {
      "epoch": 0.5102675793403858,
      "grad_norm": 0.5219797492027283,
      "learning_rate": 2e-05,
      "loss": 0.8699,
      "step": 5740
    },
    {
      "epoch": 0.5111565472486443,
      "grad_norm": 0.4437577426433563,
      "learning_rate": 2e-05,
      "loss": 0.8369,
      "step": 5750
    },
    {
      "epoch": 0.5120455151569029,
      "grad_norm": 0.4573316276073456,
      "learning_rate": 2e-05,
      "loss": 0.8428,
      "step": 5760
    },
    {
      "epoch": 0.5129344830651613,
      "grad_norm": 0.5878850221633911,
      "learning_rate": 2e-05,
      "loss": 0.865,
      "step": 5770
    },
    {
      "epoch": 0.5138234509734199,
      "grad_norm": 0.41801321506500244,
      "learning_rate": 2e-05,
      "loss": 0.8601,
      "step": 5780
    },
    {
      "epoch": 0.5147124188816784,
      "grad_norm": 0.5907896161079407,
      "learning_rate": 2e-05,
      "loss": 0.8512,
      "step": 5790
    },
    {
      "epoch": 0.5156013867899368,
      "grad_norm": 0.43416842818260193,
      "learning_rate": 2e-05,
      "loss": 0.8522,
      "step": 5800
    },
    {
      "epoch": 0.5164903546981954,
      "grad_norm": 0.47870421409606934,
      "learning_rate": 2e-05,
      "loss": 0.8803,
      "step": 5810
    },
    {
      "epoch": 0.5173793226064539,
      "grad_norm": 0.46265655755996704,
      "learning_rate": 2e-05,
      "loss": 0.8519,
      "step": 5820
    },
    {
      "epoch": 0.5182682905147125,
      "grad_norm": 0.4341539442539215,
      "learning_rate": 2e-05,
      "loss": 0.8359,
      "step": 5830
    },
    {
      "epoch": 0.5191572584229709,
      "grad_norm": 0.4009212851524353,
      "learning_rate": 2e-05,
      "loss": 0.8175,
      "step": 5840
    },
    {
      "epoch": 0.5200462263312294,
      "grad_norm": 0.4222092926502228,
      "learning_rate": 2e-05,
      "loss": 0.851,
      "step": 5850
    },
    {
      "epoch": 0.520935194239488,
      "grad_norm": 0.45038577914237976,
      "learning_rate": 2e-05,
      "loss": 0.8625,
      "step": 5860
    },
    {
      "epoch": 0.5218241621477465,
      "grad_norm": 0.4213423728942871,
      "learning_rate": 2e-05,
      "loss": 0.8598,
      "step": 5870
    },
    {
      "epoch": 0.522713130056005,
      "grad_norm": 0.43569427728652954,
      "learning_rate": 2e-05,
      "loss": 0.8642,
      "step": 5880
    },
    {
      "epoch": 0.5236020979642635,
      "grad_norm": 0.4321537911891937,
      "learning_rate": 2e-05,
      "loss": 0.8443,
      "step": 5890
    },
    {
      "epoch": 0.524491065872522,
      "grad_norm": 0.4431305527687073,
      "learning_rate": 2e-05,
      "loss": 0.8586,
      "step": 5900
    },
    {
      "epoch": 0.5253800337807805,
      "grad_norm": 0.42940574884414673,
      "learning_rate": 2e-05,
      "loss": 0.8183,
      "step": 5910
    },
    {
      "epoch": 0.526269001689039,
      "grad_norm": 0.4347728192806244,
      "learning_rate": 2e-05,
      "loss": 0.8447,
      "step": 5920
    },
    {
      "epoch": 0.5271579695972975,
      "grad_norm": 0.4522504508495331,
      "learning_rate": 2e-05,
      "loss": 0.8469,
      "step": 5930
    },
    {
      "epoch": 0.5280469375055561,
      "grad_norm": 0.4529230296611786,
      "learning_rate": 2e-05,
      "loss": 0.8403,
      "step": 5940
    },
    {
      "epoch": 0.5289359054138145,
      "grad_norm": 0.4483349025249481,
      "learning_rate": 2e-05,
      "loss": 0.8386,
      "step": 5950
    },
    {
      "epoch": 0.5298248733220731,
      "grad_norm": 0.39993539452552795,
      "learning_rate": 2e-05,
      "loss": 0.8366,
      "step": 5960
    },
    {
      "epoch": 0.5307138412303316,
      "grad_norm": 0.4244546890258789,
      "learning_rate": 2e-05,
      "loss": 0.8401,
      "step": 5970
    },
    {
      "epoch": 0.5316028091385901,
      "grad_norm": 0.47948145866394043,
      "learning_rate": 2e-05,
      "loss": 0.8615,
      "step": 5980
    },
    {
      "epoch": 0.5324917770468486,
      "grad_norm": 0.4141213297843933,
      "learning_rate": 2e-05,
      "loss": 0.843,
      "step": 5990
    },
    {
      "epoch": 0.5333807449551071,
      "grad_norm": 0.4163646697998047,
      "learning_rate": 2e-05,
      "loss": 0.9009,
      "step": 6000
    },
    {
      "epoch": 0.5342697128633657,
      "grad_norm": 0.4395085573196411,
      "learning_rate": 2e-05,
      "loss": 0.8638,
      "step": 6010
    },
    {
      "epoch": 0.5351586807716241,
      "grad_norm": 0.4790187478065491,
      "learning_rate": 2e-05,
      "loss": 0.8554,
      "step": 6020
    },
    {
      "epoch": 0.5360476486798826,
      "grad_norm": 0.427715539932251,
      "learning_rate": 2e-05,
      "loss": 0.8668,
      "step": 6030
    },
    {
      "epoch": 0.5369366165881412,
      "grad_norm": 0.452620267868042,
      "learning_rate": 2e-05,
      "loss": 0.8451,
      "step": 6040
    },
    {
      "epoch": 0.5378255844963997,
      "grad_norm": 0.4643652141094208,
      "learning_rate": 2e-05,
      "loss": 0.8927,
      "step": 6050
    },
    {
      "epoch": 0.5387145524046582,
      "grad_norm": 0.4801163077354431,
      "learning_rate": 2e-05,
      "loss": 0.8485,
      "step": 6060
    },
    {
      "epoch": 0.5396035203129167,
      "grad_norm": 0.43468740582466125,
      "learning_rate": 2e-05,
      "loss": 0.8568,
      "step": 6070
    },
    {
      "epoch": 0.5404924882211752,
      "grad_norm": 0.44157174229621887,
      "learning_rate": 2e-05,
      "loss": 0.8473,
      "step": 6080
    },
    {
      "epoch": 0.5413814561294338,
      "grad_norm": 0.4338991641998291,
      "learning_rate": 2e-05,
      "loss": 0.8391,
      "step": 6090
    },
    {
      "epoch": 0.5422704240376922,
      "grad_norm": 0.6790364384651184,
      "learning_rate": 2e-05,
      "loss": 0.8818,
      "step": 6100
    },
    {
      "epoch": 0.5431593919459508,
      "grad_norm": 0.43302908539772034,
      "learning_rate": 2e-05,
      "loss": 0.8184,
      "step": 6110
    },
    {
      "epoch": 0.5440483598542093,
      "grad_norm": 0.4233933389186859,
      "learning_rate": 2e-05,
      "loss": 0.8269,
      "step": 6120
    },
    {
      "epoch": 0.5449373277624677,
      "grad_norm": 0.415390282869339,
      "learning_rate": 2e-05,
      "loss": 0.844,
      "step": 6130
    },
    {
      "epoch": 0.5458262956707263,
      "grad_norm": 0.4095494747161865,
      "learning_rate": 2e-05,
      "loss": 0.865,
      "step": 6140
    },
    {
      "epoch": 0.5467152635789848,
      "grad_norm": 0.4126962125301361,
      "learning_rate": 2e-05,
      "loss": 0.8371,
      "step": 6150
    },
    {
      "epoch": 0.5476042314872434,
      "grad_norm": 0.4291003942489624,
      "learning_rate": 2e-05,
      "loss": 0.8309,
      "step": 6160
    },
    {
      "epoch": 0.5484931993955018,
      "grad_norm": 0.44764161109924316,
      "learning_rate": 2e-05,
      "loss": 0.8567,
      "step": 6170
    },
    {
      "epoch": 0.5493821673037603,
      "grad_norm": 0.42271336913108826,
      "learning_rate": 2e-05,
      "loss": 0.8253,
      "step": 6180
    },
    {
      "epoch": 0.5502711352120189,
      "grad_norm": 0.45998215675354004,
      "learning_rate": 2e-05,
      "loss": 0.8502,
      "step": 6190
    },
    {
      "epoch": 0.5511601031202774,
      "grad_norm": 0.41794055700302124,
      "learning_rate": 2e-05,
      "loss": 0.8647,
      "step": 6200
    },
    {
      "epoch": 0.5520490710285358,
      "grad_norm": 0.4010458290576935,
      "learning_rate": 2e-05,
      "loss": 0.8405,
      "step": 6210
    },
    {
      "epoch": 0.5529380389367944,
      "grad_norm": 0.43693119287490845,
      "learning_rate": 2e-05,
      "loss": 0.8603,
      "step": 6220
    },
    {
      "epoch": 0.5538270068450529,
      "grad_norm": 0.44117116928100586,
      "learning_rate": 2e-05,
      "loss": 0.8543,
      "step": 6230
    },
    {
      "epoch": 0.5547159747533114,
      "grad_norm": 0.43402010202407837,
      "learning_rate": 2e-05,
      "loss": 0.8452,
      "step": 6240
    },
    {
      "epoch": 0.5556049426615699,
      "grad_norm": 0.48749595880508423,
      "learning_rate": 2e-05,
      "loss": 0.8626,
      "step": 6250
    },
    {
      "epoch": 0.5564939105698284,
      "grad_norm": 0.42768076062202454,
      "learning_rate": 2e-05,
      "loss": 0.8328,
      "step": 6260
    },
    {
      "epoch": 0.557382878478087,
      "grad_norm": 0.5466820597648621,
      "learning_rate": 2e-05,
      "loss": 0.8247,
      "step": 6270
    },
    {
      "epoch": 0.5582718463863454,
      "grad_norm": 0.4047822952270508,
      "learning_rate": 2e-05,
      "loss": 0.8645,
      "step": 6280
    },
    {
      "epoch": 0.559160814294604,
      "grad_norm": 0.4790191054344177,
      "learning_rate": 2e-05,
      "loss": 0.8545,
      "step": 6290
    },
    {
      "epoch": 0.5600497822028625,
      "grad_norm": 0.4446786344051361,
      "learning_rate": 2e-05,
      "loss": 0.8251,
      "step": 6300
    },
    {
      "epoch": 0.560938750111121,
      "grad_norm": 0.4722145199775696,
      "learning_rate": 2e-05,
      "loss": 0.8236,
      "step": 6310
    },
    {
      "epoch": 0.5618277180193795,
      "grad_norm": 0.7340241074562073,
      "learning_rate": 2e-05,
      "loss": 0.8334,
      "step": 6320
    },
    {
      "epoch": 0.562716685927638,
      "grad_norm": 0.5257857441902161,
      "learning_rate": 2e-05,
      "loss": 0.8332,
      "step": 6330
    },
    {
      "epoch": 0.5636056538358966,
      "grad_norm": 0.4421565532684326,
      "learning_rate": 2e-05,
      "loss": 0.8628,
      "step": 6340
    },
    {
      "epoch": 0.564494621744155,
      "grad_norm": 0.41452983021736145,
      "learning_rate": 2e-05,
      "loss": 0.8289,
      "step": 6350
    },
    {
      "epoch": 0.5653835896524135,
      "grad_norm": 0.45369890332221985,
      "learning_rate": 2e-05,
      "loss": 0.8208,
      "step": 6360
    },
    {
      "epoch": 0.5662725575606721,
      "grad_norm": 0.41900870203971863,
      "learning_rate": 2e-05,
      "loss": 0.8272,
      "step": 6370
    },
    {
      "epoch": 0.5671615254689306,
      "grad_norm": 0.4446555972099304,
      "learning_rate": 2e-05,
      "loss": 0.8487,
      "step": 6380
    },
    {
      "epoch": 0.568050493377189,
      "grad_norm": 0.47371572256088257,
      "learning_rate": 2e-05,
      "loss": 0.8482,
      "step": 6390
    },
    {
      "epoch": 0.5689394612854476,
      "grad_norm": 0.42145735025405884,
      "learning_rate": 2e-05,
      "loss": 0.7945,
      "step": 6400
    },
    {
      "epoch": 0.5698284291937061,
      "grad_norm": 0.46672433614730835,
      "learning_rate": 2e-05,
      "loss": 0.8419,
      "step": 6410
    },
    {
      "epoch": 0.5707173971019647,
      "grad_norm": 0.42447903752326965,
      "learning_rate": 2e-05,
      "loss": 0.8248,
      "step": 6420
    },
    {
      "epoch": 0.5716063650102231,
      "grad_norm": 0.4254499077796936,
      "learning_rate": 2e-05,
      "loss": 0.8621,
      "step": 6430
    },
    {
      "epoch": 0.5724953329184816,
      "grad_norm": 0.43189167976379395,
      "learning_rate": 2e-05,
      "loss": 0.8572,
      "step": 6440
    },
    {
      "epoch": 0.5733843008267402,
      "grad_norm": 0.4194925129413605,
      "learning_rate": 2e-05,
      "loss": 0.8445,
      "step": 6450
    },
    {
      "epoch": 0.5742732687349986,
      "grad_norm": 0.4310603439807892,
      "learning_rate": 2e-05,
      "loss": 0.8052,
      "step": 6460
    },
    {
      "epoch": 0.5751622366432572,
      "grad_norm": 0.5101267099380493,
      "learning_rate": 2e-05,
      "loss": 0.8428,
      "step": 6470
    },
    {
      "epoch": 0.5760512045515157,
      "grad_norm": 0.4320722222328186,
      "learning_rate": 2e-05,
      "loss": 0.8087,
      "step": 6480
    },
    {
      "epoch": 0.5769401724597742,
      "grad_norm": 0.45333653688430786,
      "learning_rate": 2e-05,
      "loss": 0.8485,
      "step": 6490
    },
    {
      "epoch": 0.5778291403680327,
      "grad_norm": 0.4470157027244568,
      "learning_rate": 2e-05,
      "loss": 0.8733,
      "step": 6500
    },
    {
      "epoch": 0.5787181082762912,
      "grad_norm": 0.44196566939353943,
      "learning_rate": 2e-05,
      "loss": 0.817,
      "step": 6510
    },
    {
      "epoch": 0.5796070761845498,
      "grad_norm": 0.9676843881607056,
      "learning_rate": 2e-05,
      "loss": 0.8475,
      "step": 6520
    },
    {
      "epoch": 0.5804960440928082,
      "grad_norm": 0.42976272106170654,
      "learning_rate": 2e-05,
      "loss": 0.8622,
      "step": 6530
    },
    {
      "epoch": 0.5813850120010667,
      "grad_norm": 0.4123668074607849,
      "learning_rate": 2e-05,
      "loss": 0.8496,
      "step": 6540
    },
    {
      "epoch": 0.5822739799093253,
      "grad_norm": 0.41329994797706604,
      "learning_rate": 2e-05,
      "loss": 0.8329,
      "step": 6550
    },
    {
      "epoch": 0.5831629478175838,
      "grad_norm": 0.43061667680740356,
      "learning_rate": 2e-05,
      "loss": 0.8644,
      "step": 6560
    },
    {
      "epoch": 0.5840519157258423,
      "grad_norm": 0.41453754901885986,
      "learning_rate": 2e-05,
      "loss": 0.8764,
      "step": 6570
    },
    {
      "epoch": 0.5849408836341008,
      "grad_norm": 0.45204082131385803,
      "learning_rate": 2e-05,
      "loss": 0.8303,
      "step": 6580
    },
    {
      "epoch": 0.5858298515423593,
      "grad_norm": 0.40258970856666565,
      "learning_rate": 2e-05,
      "loss": 0.8238,
      "step": 6590
    },
    {
      "epoch": 0.5867188194506179,
      "grad_norm": 0.46025893092155457,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 6600
    },
    {
      "epoch": 0.5876077873588763,
      "grad_norm": 0.43104374408721924,
      "learning_rate": 2e-05,
      "loss": 0.8394,
      "step": 6610
    },
    {
      "epoch": 0.5884967552671349,
      "grad_norm": 0.41658109426498413,
      "learning_rate": 2e-05,
      "loss": 0.8311,
      "step": 6620
    },
    {
      "epoch": 0.5893857231753934,
      "grad_norm": 0.4144572615623474,
      "learning_rate": 2e-05,
      "loss": 0.7991,
      "step": 6630
    },
    {
      "epoch": 0.5902746910836518,
      "grad_norm": 0.44897204637527466,
      "learning_rate": 2e-05,
      "loss": 0.8375,
      "step": 6640
    },
    {
      "epoch": 0.5911636589919104,
      "grad_norm": 0.4406551718711853,
      "learning_rate": 2e-05,
      "loss": 0.8177,
      "step": 6650
    },
    {
      "epoch": 0.5920526269001689,
      "grad_norm": 0.5023402571678162,
      "learning_rate": 2e-05,
      "loss": 0.8431,
      "step": 6660
    },
    {
      "epoch": 0.5929415948084275,
      "grad_norm": 0.43675655126571655,
      "learning_rate": 2e-05,
      "loss": 0.8407,
      "step": 6670
    },
    {
      "epoch": 0.5938305627166859,
      "grad_norm": 0.4514894187450409,
      "learning_rate": 2e-05,
      "loss": 0.8489,
      "step": 6680
    },
    {
      "epoch": 0.5947195306249444,
      "grad_norm": 0.4401957392692566,
      "learning_rate": 2e-05,
      "loss": 0.8793,
      "step": 6690
    },
    {
      "epoch": 0.595608498533203,
      "grad_norm": 0.4395398497581482,
      "learning_rate": 2e-05,
      "loss": 0.8531,
      "step": 6700
    },
    {
      "epoch": 0.5964974664414615,
      "grad_norm": 0.44599610567092896,
      "learning_rate": 2e-05,
      "loss": 0.8462,
      "step": 6710
    },
    {
      "epoch": 0.5973864343497199,
      "grad_norm": 0.43483826518058777,
      "learning_rate": 2e-05,
      "loss": 0.8431,
      "step": 6720
    },
    {
      "epoch": 0.5982754022579785,
      "grad_norm": 0.43601951003074646,
      "learning_rate": 2e-05,
      "loss": 0.8158,
      "step": 6730
    },
    {
      "epoch": 0.599164370166237,
      "grad_norm": 0.425688236951828,
      "learning_rate": 2e-05,
      "loss": 0.8485,
      "step": 6740
    },
    {
      "epoch": 0.6000533380744955,
      "grad_norm": 0.43710899353027344,
      "learning_rate": 2e-05,
      "loss": 0.8329,
      "step": 6750
    },
    {
      "epoch": 0.600942305982754,
      "grad_norm": 0.4152549207210541,
      "learning_rate": 2e-05,
      "loss": 0.8384,
      "step": 6760
    },
    {
      "epoch": 0.6018312738910125,
      "grad_norm": 0.4004783034324646,
      "learning_rate": 2e-05,
      "loss": 0.85,
      "step": 6770
    },
    {
      "epoch": 0.6027202417992711,
      "grad_norm": 0.4235054850578308,
      "learning_rate": 2e-05,
      "loss": 0.8608,
      "step": 6780
    },
    {
      "epoch": 0.6036092097075295,
      "grad_norm": 0.4164123833179474,
      "learning_rate": 2e-05,
      "loss": 0.8488,
      "step": 6790
    },
    {
      "epoch": 0.6044981776157881,
      "grad_norm": 0.42861080169677734,
      "learning_rate": 2e-05,
      "loss": 0.864,
      "step": 6800
    },
    {
      "epoch": 0.6053871455240466,
      "grad_norm": 0.43809929490089417,
      "learning_rate": 2e-05,
      "loss": 0.8423,
      "step": 6810
    },
    {
      "epoch": 0.6062761134323051,
      "grad_norm": 0.4452234208583832,
      "learning_rate": 2e-05,
      "loss": 0.8017,
      "step": 6820
    },
    {
      "epoch": 0.6071650813405636,
      "grad_norm": 0.4168837368488312,
      "learning_rate": 2e-05,
      "loss": 0.8423,
      "step": 6830
    },
    {
      "epoch": 0.6080540492488221,
      "grad_norm": 0.40204375982284546,
      "learning_rate": 2e-05,
      "loss": 0.8038,
      "step": 6840
    },
    {
      "epoch": 0.6089430171570807,
      "grad_norm": 0.4045926332473755,
      "learning_rate": 2e-05,
      "loss": 0.8372,
      "step": 6850
    },
    {
      "epoch": 0.6098319850653391,
      "grad_norm": 0.44481104612350464,
      "learning_rate": 2e-05,
      "loss": 0.8407,
      "step": 6860
    },
    {
      "epoch": 0.6107209529735976,
      "grad_norm": 0.4702940583229065,
      "learning_rate": 2e-05,
      "loss": 0.8701,
      "step": 6870
    },
    {
      "epoch": 0.6116099208818562,
      "grad_norm": 0.43351835012435913,
      "learning_rate": 2e-05,
      "loss": 0.855,
      "step": 6880
    },
    {
      "epoch": 0.6124988887901147,
      "grad_norm": 0.4448101222515106,
      "learning_rate": 2e-05,
      "loss": 0.8191,
      "step": 6890
    },
    {
      "epoch": 0.6133878566983731,
      "grad_norm": 0.46052518486976624,
      "learning_rate": 2e-05,
      "loss": 0.8362,
      "step": 6900
    },
    {
      "epoch": 0.6142768246066317,
      "grad_norm": 0.4140656888484955,
      "learning_rate": 2e-05,
      "loss": 0.8488,
      "step": 6910
    },
    {
      "epoch": 0.6151657925148902,
      "grad_norm": 0.4561617970466614,
      "learning_rate": 2e-05,
      "loss": 0.8276,
      "step": 6920
    },
    {
      "epoch": 0.6160547604231488,
      "grad_norm": 0.4391039311885834,
      "learning_rate": 2e-05,
      "loss": 0.864,
      "step": 6930
    },
    {
      "epoch": 0.6169437283314072,
      "grad_norm": 0.44055435061454773,
      "learning_rate": 2e-05,
      "loss": 0.8656,
      "step": 6940
    },
    {
      "epoch": 0.6178326962396657,
      "grad_norm": 0.42682865262031555,
      "learning_rate": 2e-05,
      "loss": 0.8179,
      "step": 6950
    },
    {
      "epoch": 0.6187216641479243,
      "grad_norm": 0.40694037079811096,
      "learning_rate": 2e-05,
      "loss": 0.8072,
      "step": 6960
    },
    {
      "epoch": 0.6196106320561827,
      "grad_norm": 0.432834655046463,
      "learning_rate": 2e-05,
      "loss": 0.8193,
      "step": 6970
    },
    {
      "epoch": 0.6204995999644413,
      "grad_norm": 0.4349261224269867,
      "learning_rate": 2e-05,
      "loss": 0.8662,
      "step": 6980
    },
    {
      "epoch": 0.6213885678726998,
      "grad_norm": 0.4014347195625305,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 6990
    },
    {
      "epoch": 0.6222775357809583,
      "grad_norm": 0.4711790382862091,
      "learning_rate": 2e-05,
      "loss": 0.8411,
      "step": 7000
    },
    {
      "epoch": 0.6231665036892168,
      "grad_norm": 0.4515135884284973,
      "learning_rate": 2e-05,
      "loss": 0.8332,
      "step": 7010
    },
    {
      "epoch": 0.6240554715974753,
      "grad_norm": 0.43790239095687866,
      "learning_rate": 2e-05,
      "loss": 0.8149,
      "step": 7020
    },
    {
      "epoch": 0.6249444395057339,
      "grad_norm": 0.42519545555114746,
      "learning_rate": 2e-05,
      "loss": 0.8397,
      "step": 7030
    },
    {
      "epoch": 0.6258334074139924,
      "grad_norm": 0.42393261194229126,
      "learning_rate": 2e-05,
      "loss": 0.8262,
      "step": 7040
    },
    {
      "epoch": 0.6267223753222508,
      "grad_norm": 0.42167550325393677,
      "learning_rate": 2e-05,
      "loss": 0.8345,
      "step": 7050
    },
    {
      "epoch": 0.6276113432305094,
      "grad_norm": 0.43001675605773926,
      "learning_rate": 2e-05,
      "loss": 0.8622,
      "step": 7060
    },
    {
      "epoch": 0.6285003111387679,
      "grad_norm": 0.4084045886993408,
      "learning_rate": 2e-05,
      "loss": 0.8251,
      "step": 7070
    },
    {
      "epoch": 0.6293892790470264,
      "grad_norm": 0.42174044251441956,
      "learning_rate": 2e-05,
      "loss": 0.8393,
      "step": 7080
    },
    {
      "epoch": 0.6302782469552849,
      "grad_norm": 0.42673105001449585,
      "learning_rate": 2e-05,
      "loss": 0.8292,
      "step": 7090
    },
    {
      "epoch": 0.6311672148635434,
      "grad_norm": 0.4536789357662201,
      "learning_rate": 2e-05,
      "loss": 0.849,
      "step": 7100
    },
    {
      "epoch": 0.632056182771802,
      "grad_norm": 0.4451805353164673,
      "learning_rate": 2e-05,
      "loss": 0.8267,
      "step": 7110
    },
    {
      "epoch": 0.6329451506800604,
      "grad_norm": 0.42965081334114075,
      "learning_rate": 2e-05,
      "loss": 0.8188,
      "step": 7120
    },
    {
      "epoch": 0.633834118588319,
      "grad_norm": 0.40687575936317444,
      "learning_rate": 2e-05,
      "loss": 0.8476,
      "step": 7130
    },
    {
      "epoch": 0.6347230864965775,
      "grad_norm": 0.4426710605621338,
      "learning_rate": 2e-05,
      "loss": 0.8342,
      "step": 7140
    },
    {
      "epoch": 0.635612054404836,
      "grad_norm": 0.5465063452720642,
      "learning_rate": 2e-05,
      "loss": 0.8234,
      "step": 7150
    },
    {
      "epoch": 0.6365010223130945,
      "grad_norm": 0.4803895652294159,
      "learning_rate": 2e-05,
      "loss": 0.8056,
      "step": 7160
    },
    {
      "epoch": 0.637389990221353,
      "grad_norm": 0.4149646461009979,
      "learning_rate": 2e-05,
      "loss": 0.8555,
      "step": 7170
    },
    {
      "epoch": 0.6382789581296116,
      "grad_norm": 0.4877215325832367,
      "learning_rate": 2e-05,
      "loss": 0.8026,
      "step": 7180
    },
    {
      "epoch": 0.63916792603787,
      "grad_norm": 0.4329800307750702,
      "learning_rate": 2e-05,
      "loss": 0.8179,
      "step": 7190
    },
    {
      "epoch": 0.6400568939461285,
      "grad_norm": 0.5567352175712585,
      "learning_rate": 2e-05,
      "loss": 0.8089,
      "step": 7200
    },
    {
      "epoch": 0.6409458618543871,
      "grad_norm": 0.49261757731437683,
      "learning_rate": 2e-05,
      "loss": 0.8517,
      "step": 7210
    },
    {
      "epoch": 0.6418348297626456,
      "grad_norm": 0.4212228059768677,
      "learning_rate": 2e-05,
      "loss": 0.8279,
      "step": 7220
    },
    {
      "epoch": 0.642723797670904,
      "grad_norm": 0.4334050118923187,
      "learning_rate": 2e-05,
      "loss": 0.8324,
      "step": 7230
    },
    {
      "epoch": 0.6436127655791626,
      "grad_norm": 0.4356308579444885,
      "learning_rate": 2e-05,
      "loss": 0.8152,
      "step": 7240
    },
    {
      "epoch": 0.6445017334874211,
      "grad_norm": 0.6391989588737488,
      "learning_rate": 2e-05,
      "loss": 0.8409,
      "step": 7250
    },
    {
      "epoch": 0.6453907013956797,
      "grad_norm": 0.4149230420589447,
      "learning_rate": 2e-05,
      "loss": 0.835,
      "step": 7260
    },
    {
      "epoch": 0.6462796693039381,
      "grad_norm": 0.47466468811035156,
      "learning_rate": 2e-05,
      "loss": 0.7811,
      "step": 7270
    },
    {
      "epoch": 0.6471686372121966,
      "grad_norm": 0.4377541244029999,
      "learning_rate": 2e-05,
      "loss": 0.8324,
      "step": 7280
    },
    {
      "epoch": 0.6480576051204552,
      "grad_norm": 0.41136160492897034,
      "learning_rate": 2e-05,
      "loss": 0.8476,
      "step": 7290
    },
    {
      "epoch": 0.6489465730287136,
      "grad_norm": 0.43524637818336487,
      "learning_rate": 2e-05,
      "loss": 0.8381,
      "step": 7300
    },
    {
      "epoch": 0.6498355409369722,
      "grad_norm": 0.43635448813438416,
      "learning_rate": 2e-05,
      "loss": 0.8266,
      "step": 7310
    },
    {
      "epoch": 0.6507245088452307,
      "grad_norm": 0.4506995379924774,
      "learning_rate": 2e-05,
      "loss": 0.8508,
      "step": 7320
    },
    {
      "epoch": 0.6516134767534892,
      "grad_norm": 0.4616965055465698,
      "learning_rate": 2e-05,
      "loss": 0.7919,
      "step": 7330
    },
    {
      "epoch": 0.6525024446617477,
      "grad_norm": 0.44731423258781433,
      "learning_rate": 2e-05,
      "loss": 0.844,
      "step": 7340
    },
    {
      "epoch": 0.6533914125700062,
      "grad_norm": 0.40850594639778137,
      "learning_rate": 2e-05,
      "loss": 0.8115,
      "step": 7350
    },
    {
      "epoch": 0.6542803804782648,
      "grad_norm": 0.4451155364513397,
      "learning_rate": 2e-05,
      "loss": 0.8225,
      "step": 7360
    },
    {
      "epoch": 0.6551693483865233,
      "grad_norm": 0.43127143383026123,
      "learning_rate": 2e-05,
      "loss": 0.8661,
      "step": 7370
    },
    {
      "epoch": 0.6560583162947817,
      "grad_norm": 0.40748530626296997,
      "learning_rate": 2e-05,
      "loss": 0.8157,
      "step": 7380
    },
    {
      "epoch": 0.6569472842030403,
      "grad_norm": 0.6737169027328491,
      "learning_rate": 2e-05,
      "loss": 0.8243,
      "step": 7390
    },
    {
      "epoch": 0.6578362521112988,
      "grad_norm": 0.4323747456073761,
      "learning_rate": 2e-05,
      "loss": 0.831,
      "step": 7400
    },
    {
      "epoch": 0.6587252200195572,
      "grad_norm": 0.4288110136985779,
      "learning_rate": 2e-05,
      "loss": 0.8212,
      "step": 7410
    },
    {
      "epoch": 0.6596141879278158,
      "grad_norm": 0.527877688407898,
      "learning_rate": 2e-05,
      "loss": 0.8517,
      "step": 7420
    },
    {
      "epoch": 0.6605031558360743,
      "grad_norm": 0.441108763217926,
      "learning_rate": 2e-05,
      "loss": 0.808,
      "step": 7430
    },
    {
      "epoch": 0.6613921237443329,
      "grad_norm": 0.46566042304039,
      "learning_rate": 2e-05,
      "loss": 0.8278,
      "step": 7440
    },
    {
      "epoch": 0.6622810916525913,
      "grad_norm": 0.4497864246368408,
      "learning_rate": 2e-05,
      "loss": 0.832,
      "step": 7450
    },
    {
      "epoch": 0.6631700595608498,
      "grad_norm": 0.42945578694343567,
      "learning_rate": 2e-05,
      "loss": 0.803,
      "step": 7460
    },
    {
      "epoch": 0.6640590274691084,
      "grad_norm": 0.5866878032684326,
      "learning_rate": 2e-05,
      "loss": 0.8445,
      "step": 7470
    },
    {
      "epoch": 0.6649479953773669,
      "grad_norm": 0.4364074170589447,
      "learning_rate": 2e-05,
      "loss": 0.8382,
      "step": 7480
    },
    {
      "epoch": 0.6658369632856254,
      "grad_norm": 0.4456654489040375,
      "learning_rate": 2e-05,
      "loss": 0.8067,
      "step": 7490
    },
    {
      "epoch": 0.6667259311938839,
      "grad_norm": 0.5686672329902649,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 7500
    },
    {
      "epoch": 0.6676148991021424,
      "grad_norm": 0.4275640845298767,
      "learning_rate": 2e-05,
      "loss": 0.8414,
      "step": 7510
    },
    {
      "epoch": 0.6685038670104009,
      "grad_norm": 0.433285117149353,
      "learning_rate": 2e-05,
      "loss": 0.8156,
      "step": 7520
    },
    {
      "epoch": 0.6693928349186594,
      "grad_norm": 0.44586434960365295,
      "learning_rate": 2e-05,
      "loss": 0.8542,
      "step": 7530
    },
    {
      "epoch": 0.670281802826918,
      "grad_norm": 0.45324957370758057,
      "learning_rate": 2e-05,
      "loss": 0.831,
      "step": 7540
    },
    {
      "epoch": 0.6711707707351765,
      "grad_norm": 0.4702342748641968,
      "learning_rate": 2e-05,
      "loss": 0.7996,
      "step": 7550
    },
    {
      "epoch": 0.6720597386434349,
      "grad_norm": 0.4410693347454071,
      "learning_rate": 2e-05,
      "loss": 0.8124,
      "step": 7560
    },
    {
      "epoch": 0.6729487065516935,
      "grad_norm": 0.48080357909202576,
      "learning_rate": 2e-05,
      "loss": 0.8039,
      "step": 7570
    },
    {
      "epoch": 0.673837674459952,
      "grad_norm": 0.47827407717704773,
      "learning_rate": 2e-05,
      "loss": 0.8277,
      "step": 7580
    },
    {
      "epoch": 0.6747266423682106,
      "grad_norm": 0.4933921694755554,
      "learning_rate": 2e-05,
      "loss": 0.818,
      "step": 7590
    },
    {
      "epoch": 0.675615610276469,
      "grad_norm": 0.45800861716270447,
      "learning_rate": 2e-05,
      "loss": 0.7975,
      "step": 7600
    },
    {
      "epoch": 0.6765045781847275,
      "grad_norm": 0.4388594329357147,
      "learning_rate": 2e-05,
      "loss": 0.8431,
      "step": 7610
    },
    {
      "epoch": 0.6773935460929861,
      "grad_norm": 0.4385128319263458,
      "learning_rate": 2e-05,
      "loss": 0.8251,
      "step": 7620
    },
    {
      "epoch": 0.6782825140012445,
      "grad_norm": 0.4399843215942383,
      "learning_rate": 2e-05,
      "loss": 0.823,
      "step": 7630
    },
    {
      "epoch": 0.679171481909503,
      "grad_norm": 0.43579229712486267,
      "learning_rate": 2e-05,
      "loss": 0.8113,
      "step": 7640
    },
    {
      "epoch": 0.6800604498177616,
      "grad_norm": 0.4776645600795746,
      "learning_rate": 2e-05,
      "loss": 0.8088,
      "step": 7650
    },
    {
      "epoch": 0.6809494177260201,
      "grad_norm": 0.4208528995513916,
      "learning_rate": 2e-05,
      "loss": 0.8313,
      "step": 7660
    },
    {
      "epoch": 0.6818383856342786,
      "grad_norm": 0.4396716058254242,
      "learning_rate": 2e-05,
      "loss": 0.7973,
      "step": 7670
    },
    {
      "epoch": 0.6827273535425371,
      "grad_norm": 0.445252388715744,
      "learning_rate": 2e-05,
      "loss": 0.8346,
      "step": 7680
    },
    {
      "epoch": 0.6836163214507956,
      "grad_norm": 0.48601850867271423,
      "learning_rate": 2e-05,
      "loss": 0.7863,
      "step": 7690
    },
    {
      "epoch": 0.6845052893590542,
      "grad_norm": 1.053794503211975,
      "learning_rate": 2e-05,
      "loss": 0.8122,
      "step": 7700
    },
    {
      "epoch": 0.6853942572673126,
      "grad_norm": 0.42846912145614624,
      "learning_rate": 2e-05,
      "loss": 0.8388,
      "step": 7710
    },
    {
      "epoch": 0.6862832251755712,
      "grad_norm": 0.4107976257801056,
      "learning_rate": 2e-05,
      "loss": 0.7834,
      "step": 7720
    },
    {
      "epoch": 0.6871721930838297,
      "grad_norm": 0.42571184039115906,
      "learning_rate": 2e-05,
      "loss": 0.8379,
      "step": 7730
    },
    {
      "epoch": 0.6880611609920881,
      "grad_norm": 0.4173625111579895,
      "learning_rate": 2e-05,
      "loss": 0.8302,
      "step": 7740
    },
    {
      "epoch": 0.6889501289003467,
      "grad_norm": 0.41880202293395996,
      "learning_rate": 2e-05,
      "loss": 0.8498,
      "step": 7750
    },
    {
      "epoch": 0.6898390968086052,
      "grad_norm": 0.45978063344955444,
      "learning_rate": 2e-05,
      "loss": 0.834,
      "step": 7760
    },
    {
      "epoch": 0.6907280647168638,
      "grad_norm": 0.4542718231678009,
      "learning_rate": 2e-05,
      "loss": 0.7975,
      "step": 7770
    },
    {
      "epoch": 0.6916170326251222,
      "grad_norm": 0.4782005548477173,
      "learning_rate": 2e-05,
      "loss": 0.8061,
      "step": 7780
    },
    {
      "epoch": 0.6925060005333807,
      "grad_norm": 0.4693487882614136,
      "learning_rate": 2e-05,
      "loss": 0.8129,
      "step": 7790
    },
    {
      "epoch": 0.6933949684416393,
      "grad_norm": 0.4502740502357483,
      "learning_rate": 2e-05,
      "loss": 0.8081,
      "step": 7800
    },
    {
      "epoch": 0.6942839363498978,
      "grad_norm": 0.4064286947250366,
      "learning_rate": 2e-05,
      "loss": 0.8394,
      "step": 7810
    },
    {
      "epoch": 0.6951729042581563,
      "grad_norm": 0.4019455015659332,
      "learning_rate": 2e-05,
      "loss": 0.8553,
      "step": 7820
    },
    {
      "epoch": 0.6960618721664148,
      "grad_norm": 0.40903815627098083,
      "learning_rate": 2e-05,
      "loss": 0.8414,
      "step": 7830
    },
    {
      "epoch": 0.6969508400746733,
      "grad_norm": 0.40911242365837097,
      "learning_rate": 2e-05,
      "loss": 0.8286,
      "step": 7840
    },
    {
      "epoch": 0.6978398079829318,
      "grad_norm": 0.4351511299610138,
      "learning_rate": 2e-05,
      "loss": 0.8201,
      "step": 7850
    },
    {
      "epoch": 0.6987287758911903,
      "grad_norm": 0.4626946449279785,
      "learning_rate": 2e-05,
      "loss": 0.8575,
      "step": 7860
    },
    {
      "epoch": 0.6996177437994489,
      "grad_norm": 0.4467368423938751,
      "learning_rate": 2e-05,
      "loss": 0.8143,
      "step": 7870
    },
    {
      "epoch": 0.7005067117077074,
      "grad_norm": 0.7355163097381592,
      "learning_rate": 2e-05,
      "loss": 0.8379,
      "step": 7880
    },
    {
      "epoch": 0.7013956796159658,
      "grad_norm": 0.4317020773887634,
      "learning_rate": 2e-05,
      "loss": 0.826,
      "step": 7890
    },
    {
      "epoch": 0.7022846475242244,
      "grad_norm": 0.44636988639831543,
      "learning_rate": 2e-05,
      "loss": 0.7858,
      "step": 7900
    },
    {
      "epoch": 0.7031736154324829,
      "grad_norm": 0.42319321632385254,
      "learning_rate": 2e-05,
      "loss": 0.8103,
      "step": 7910
    },
    {
      "epoch": 0.7040625833407415,
      "grad_norm": 0.4444860816001892,
      "learning_rate": 2e-05,
      "loss": 0.8165,
      "step": 7920
    },
    {
      "epoch": 0.7049515512489999,
      "grad_norm": 0.427090048789978,
      "learning_rate": 2e-05,
      "loss": 0.7987,
      "step": 7930
    },
    {
      "epoch": 0.7058405191572584,
      "grad_norm": 0.4884425103664398,
      "learning_rate": 2e-05,
      "loss": 0.8164,
      "step": 7940
    },
    {
      "epoch": 0.706729487065517,
      "grad_norm": 0.4308551251888275,
      "learning_rate": 2e-05,
      "loss": 0.8359,
      "step": 7950
    },
    {
      "epoch": 0.7076184549737754,
      "grad_norm": 0.42863738536834717,
      "learning_rate": 2e-05,
      "loss": 0.8457,
      "step": 7960
    },
    {
      "epoch": 0.7085074228820339,
      "grad_norm": 0.42633193731307983,
      "learning_rate": 2e-05,
      "loss": 0.7978,
      "step": 7970
    },
    {
      "epoch": 0.7093963907902925,
      "grad_norm": 0.4049590826034546,
      "learning_rate": 2e-05,
      "loss": 0.8136,
      "step": 7980
    },
    {
      "epoch": 0.710285358698551,
      "grad_norm": 0.4387713372707367,
      "learning_rate": 2e-05,
      "loss": 0.8133,
      "step": 7990
    },
    {
      "epoch": 0.7111743266068095,
      "grad_norm": 0.4449150860309601,
      "learning_rate": 2e-05,
      "loss": 0.8326,
      "step": 8000
    },
    {
      "epoch": 0.712063294515068,
      "grad_norm": 0.43443921208381653,
      "learning_rate": 2e-05,
      "loss": 0.8095,
      "step": 8010
    },
    {
      "epoch": 0.7129522624233265,
      "grad_norm": 0.4353339970111847,
      "learning_rate": 2e-05,
      "loss": 0.8257,
      "step": 8020
    },
    {
      "epoch": 0.713841230331585,
      "grad_norm": 0.44458910822868347,
      "learning_rate": 2e-05,
      "loss": 0.8171,
      "step": 8030
    },
    {
      "epoch": 0.7147301982398435,
      "grad_norm": 0.46637725830078125,
      "learning_rate": 2e-05,
      "loss": 0.8079,
      "step": 8040
    },
    {
      "epoch": 0.7156191661481021,
      "grad_norm": 0.43437546491622925,
      "learning_rate": 2e-05,
      "loss": 0.8099,
      "step": 8050
    },
    {
      "epoch": 0.7165081340563606,
      "grad_norm": 0.42220667004585266,
      "learning_rate": 2e-05,
      "loss": 0.8091,
      "step": 8060
    },
    {
      "epoch": 0.717397101964619,
      "grad_norm": 0.44379591941833496,
      "learning_rate": 2e-05,
      "loss": 0.8048,
      "step": 8070
    },
    {
      "epoch": 0.7182860698728776,
      "grad_norm": 0.44161292910575867,
      "learning_rate": 2e-05,
      "loss": 0.8007,
      "step": 8080
    },
    {
      "epoch": 0.7191750377811361,
      "grad_norm": 0.432823121547699,
      "learning_rate": 2e-05,
      "loss": 0.8381,
      "step": 8090
    },
    {
      "epoch": 0.7200640056893947,
      "grad_norm": 0.6839956641197205,
      "learning_rate": 2e-05,
      "loss": 0.7968,
      "step": 8100
    },
    {
      "epoch": 0.7209529735976531,
      "grad_norm": 0.46361783146858215,
      "learning_rate": 2e-05,
      "loss": 0.8134,
      "step": 8110
    },
    {
      "epoch": 0.7218419415059116,
      "grad_norm": 0.42806437611579895,
      "learning_rate": 2e-05,
      "loss": 0.8073,
      "step": 8120
    },
    {
      "epoch": 0.7227309094141702,
      "grad_norm": 0.4221607744693756,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 8130
    },
    {
      "epoch": 0.7236198773224286,
      "grad_norm": 0.4217561185359955,
      "learning_rate": 2e-05,
      "loss": 0.7635,
      "step": 8140
    },
    {
      "epoch": 0.7245088452306871,
      "grad_norm": 0.5766949653625488,
      "learning_rate": 2e-05,
      "loss": 0.805,
      "step": 8150
    },
    {
      "epoch": 0.7253978131389457,
      "grad_norm": 0.41743677854537964,
      "learning_rate": 2e-05,
      "loss": 0.8071,
      "step": 8160
    },
    {
      "epoch": 0.7262867810472042,
      "grad_norm": 0.4412586987018585,
      "learning_rate": 2e-05,
      "loss": 0.8461,
      "step": 8170
    },
    {
      "epoch": 0.7271757489554627,
      "grad_norm": 0.4252506196498871,
      "learning_rate": 2e-05,
      "loss": 0.811,
      "step": 8180
    },
    {
      "epoch": 0.7280647168637212,
      "grad_norm": 1.1769479513168335,
      "learning_rate": 2e-05,
      "loss": 0.8244,
      "step": 8190
    },
    {
      "epoch": 0.7289536847719797,
      "grad_norm": 0.6084337830543518,
      "learning_rate": 2e-05,
      "loss": 0.8225,
      "step": 8200
    },
    {
      "epoch": 0.7298426526802383,
      "grad_norm": 0.4107128381729126,
      "learning_rate": 2e-05,
      "loss": 0.8373,
      "step": 8210
    },
    {
      "epoch": 0.7307316205884967,
      "grad_norm": 0.45130422711372375,
      "learning_rate": 2e-05,
      "loss": 0.8108,
      "step": 8220
    },
    {
      "epoch": 0.7316205884967553,
      "grad_norm": 0.4697266221046448,
      "learning_rate": 2e-05,
      "loss": 0.8052,
      "step": 8230
    },
    {
      "epoch": 0.7325095564050138,
      "grad_norm": 0.47850725054740906,
      "learning_rate": 2e-05,
      "loss": 0.83,
      "step": 8240
    },
    {
      "epoch": 0.7333985243132722,
      "grad_norm": 0.4215230941772461,
      "learning_rate": 2e-05,
      "loss": 0.8197,
      "step": 8250
    },
    {
      "epoch": 0.7342874922215308,
      "grad_norm": 0.43214040994644165,
      "learning_rate": 2e-05,
      "loss": 0.8165,
      "step": 8260
    },
    {
      "epoch": 0.7351764601297893,
      "grad_norm": 0.4207126498222351,
      "learning_rate": 2e-05,
      "loss": 0.8173,
      "step": 8270
    },
    {
      "epoch": 0.7360654280380479,
      "grad_norm": 0.4162978231906891,
      "learning_rate": 2e-05,
      "loss": 0.8009,
      "step": 8280
    },
    {
      "epoch": 0.7369543959463063,
      "grad_norm": 0.4562695026397705,
      "learning_rate": 2e-05,
      "loss": 0.805,
      "step": 8290
    },
    {
      "epoch": 0.7378433638545648,
      "grad_norm": 0.44911906123161316,
      "learning_rate": 2e-05,
      "loss": 0.8066,
      "step": 8300
    },
    {
      "epoch": 0.7387323317628234,
      "grad_norm": 0.44587841629981995,
      "learning_rate": 2e-05,
      "loss": 0.8195,
      "step": 8310
    },
    {
      "epoch": 0.7396212996710819,
      "grad_norm": 0.4086598753929138,
      "learning_rate": 2e-05,
      "loss": 0.8255,
      "step": 8320
    },
    {
      "epoch": 0.7405102675793404,
      "grad_norm": 0.4192577600479126,
      "learning_rate": 2e-05,
      "loss": 0.8212,
      "step": 8330
    },
    {
      "epoch": 0.7413992354875989,
      "grad_norm": 0.4602523744106293,
      "learning_rate": 2e-05,
      "loss": 0.8185,
      "step": 8340
    },
    {
      "epoch": 0.7422882033958574,
      "grad_norm": 0.4165593385696411,
      "learning_rate": 2e-05,
      "loss": 0.8181,
      "step": 8350
    },
    {
      "epoch": 0.7431771713041159,
      "grad_norm": 0.48770880699157715,
      "learning_rate": 2e-05,
      "loss": 0.8352,
      "step": 8360
    },
    {
      "epoch": 0.7440661392123744,
      "grad_norm": 0.46819040179252625,
      "learning_rate": 2e-05,
      "loss": 0.8325,
      "step": 8370
    },
    {
      "epoch": 0.744955107120633,
      "grad_norm": 0.4363252520561218,
      "learning_rate": 2e-05,
      "loss": 0.8181,
      "step": 8380
    },
    {
      "epoch": 0.7458440750288915,
      "grad_norm": 0.43392571806907654,
      "learning_rate": 2e-05,
      "loss": 0.8255,
      "step": 8390
    },
    {
      "epoch": 0.7467330429371499,
      "grad_norm": 0.48683926463127136,
      "learning_rate": 2e-05,
      "loss": 0.8001,
      "step": 8400
    },
    {
      "epoch": 0.7476220108454085,
      "grad_norm": 0.43214860558509827,
      "learning_rate": 2e-05,
      "loss": 0.8097,
      "step": 8410
    },
    {
      "epoch": 0.748510978753667,
      "grad_norm": 0.435784250497818,
      "learning_rate": 2e-05,
      "loss": 0.8155,
      "step": 8420
    },
    {
      "epoch": 0.7493999466619256,
      "grad_norm": 0.4375058710575104,
      "learning_rate": 2e-05,
      "loss": 0.8375,
      "step": 8430
    },
    {
      "epoch": 0.750288914570184,
      "grad_norm": 0.41872087121009827,
      "learning_rate": 2e-05,
      "loss": 0.8014,
      "step": 8440
    },
    {
      "epoch": 0.7511778824784425,
      "grad_norm": 0.4573659300804138,
      "learning_rate": 2e-05,
      "loss": 0.8019,
      "step": 8450
    },
    {
      "epoch": 0.7520668503867011,
      "grad_norm": 0.43358829617500305,
      "learning_rate": 2e-05,
      "loss": 0.8181,
      "step": 8460
    },
    {
      "epoch": 0.7529558182949595,
      "grad_norm": 0.40546298027038574,
      "learning_rate": 2e-05,
      "loss": 0.7988,
      "step": 8470
    },
    {
      "epoch": 0.753844786203218,
      "grad_norm": 0.4194851517677307,
      "learning_rate": 2e-05,
      "loss": 0.8163,
      "step": 8480
    },
    {
      "epoch": 0.7547337541114766,
      "grad_norm": 0.42072299122810364,
      "learning_rate": 2e-05,
      "loss": 0.8357,
      "step": 8490
    },
    {
      "epoch": 0.7556227220197351,
      "grad_norm": 0.41284969449043274,
      "learning_rate": 2e-05,
      "loss": 0.7864,
      "step": 8500
    },
    {
      "epoch": 0.7565116899279936,
      "grad_norm": 0.4464804530143738,
      "learning_rate": 2e-05,
      "loss": 0.7858,
      "step": 8510
    },
    {
      "epoch": 0.7574006578362521,
      "grad_norm": 0.44678646326065063,
      "learning_rate": 2e-05,
      "loss": 0.7895,
      "step": 8520
    },
    {
      "epoch": 0.7582896257445106,
      "grad_norm": 0.4128533899784088,
      "learning_rate": 2e-05,
      "loss": 0.8292,
      "step": 8530
    },
    {
      "epoch": 0.7591785936527692,
      "grad_norm": 0.44808879494667053,
      "learning_rate": 2e-05,
      "loss": 0.786,
      "step": 8540
    },
    {
      "epoch": 0.7600675615610276,
      "grad_norm": 0.4333820044994354,
      "learning_rate": 2e-05,
      "loss": 0.8116,
      "step": 8550
    },
    {
      "epoch": 0.7609565294692862,
      "grad_norm": 0.42960503697395325,
      "learning_rate": 2e-05,
      "loss": 0.818,
      "step": 8560
    },
    {
      "epoch": 0.7618454973775447,
      "grad_norm": 0.4117410480976105,
      "learning_rate": 2e-05,
      "loss": 0.8272,
      "step": 8570
    },
    {
      "epoch": 0.7627344652858031,
      "grad_norm": 0.4132250249385834,
      "learning_rate": 2e-05,
      "loss": 0.7899,
      "step": 8580
    },
    {
      "epoch": 0.7636234331940617,
      "grad_norm": 0.42917436361312866,
      "learning_rate": 2e-05,
      "loss": 0.7992,
      "step": 8590
    },
    {
      "epoch": 0.7645124011023202,
      "grad_norm": 0.45649459958076477,
      "learning_rate": 2e-05,
      "loss": 0.8169,
      "step": 8600
    },
    {
      "epoch": 0.7654013690105788,
      "grad_norm": 0.44682660698890686,
      "learning_rate": 2e-05,
      "loss": 0.7996,
      "step": 8610
    },
    {
      "epoch": 0.7662903369188372,
      "grad_norm": 0.4425507187843323,
      "learning_rate": 2e-05,
      "loss": 0.8045,
      "step": 8620
    },
    {
      "epoch": 0.7671793048270957,
      "grad_norm": 0.45848390460014343,
      "learning_rate": 2e-05,
      "loss": 0.8016,
      "step": 8630
    },
    {
      "epoch": 0.7680682727353543,
      "grad_norm": 0.4272679388523102,
      "learning_rate": 2e-05,
      "loss": 0.8391,
      "step": 8640
    },
    {
      "epoch": 0.7689572406436128,
      "grad_norm": 0.4503556489944458,
      "learning_rate": 2e-05,
      "loss": 0.7885,
      "step": 8650
    },
    {
      "epoch": 0.7698462085518712,
      "grad_norm": 0.44096842408180237,
      "learning_rate": 2e-05,
      "loss": 0.8249,
      "step": 8660
    },
    {
      "epoch": 0.7707351764601298,
      "grad_norm": 0.5391594767570496,
      "learning_rate": 2e-05,
      "loss": 0.8234,
      "step": 8670
    },
    {
      "epoch": 0.7716241443683883,
      "grad_norm": 0.4615217447280884,
      "learning_rate": 2e-05,
      "loss": 0.8256,
      "step": 8680
    },
    {
      "epoch": 0.7725131122766468,
      "grad_norm": 0.43299633264541626,
      "learning_rate": 2e-05,
      "loss": 0.7975,
      "step": 8690
    },
    {
      "epoch": 0.7734020801849053,
      "grad_norm": 0.40458884835243225,
      "learning_rate": 2e-05,
      "loss": 0.7862,
      "step": 8700
    },
    {
      "epoch": 0.7742910480931638,
      "grad_norm": 0.4393770694732666,
      "learning_rate": 2e-05,
      "loss": 0.7991,
      "step": 8710
    },
    {
      "epoch": 0.7751800160014224,
      "grad_norm": 0.4450920522212982,
      "learning_rate": 2e-05,
      "loss": 0.8287,
      "step": 8720
    },
    {
      "epoch": 0.7760689839096808,
      "grad_norm": 0.4866134524345398,
      "learning_rate": 2e-05,
      "loss": 0.7981,
      "step": 8730
    },
    {
      "epoch": 0.7769579518179394,
      "grad_norm": 0.4510369598865509,
      "learning_rate": 2e-05,
      "loss": 0.821,
      "step": 8740
    },
    {
      "epoch": 0.7778469197261979,
      "grad_norm": 0.4555661678314209,
      "learning_rate": 2e-05,
      "loss": 0.8076,
      "step": 8750
    },
    {
      "epoch": 0.7787358876344564,
      "grad_norm": 0.4143234193325043,
      "learning_rate": 2e-05,
      "loss": 0.8041,
      "step": 8760
    },
    {
      "epoch": 0.7796248555427149,
      "grad_norm": 0.41745519638061523,
      "learning_rate": 2e-05,
      "loss": 0.8188,
      "step": 8770
    },
    {
      "epoch": 0.7805138234509734,
      "grad_norm": 0.45255017280578613,
      "learning_rate": 2e-05,
      "loss": 0.8289,
      "step": 8780
    },
    {
      "epoch": 0.781402791359232,
      "grad_norm": 0.45360854268074036,
      "learning_rate": 2e-05,
      "loss": 0.8142,
      "step": 8790
    },
    {
      "epoch": 0.7822917592674904,
      "grad_norm": 0.4983905255794525,
      "learning_rate": 2e-05,
      "loss": 0.8138,
      "step": 8800
    },
    {
      "epoch": 0.7831807271757489,
      "grad_norm": 0.43971651792526245,
      "learning_rate": 2e-05,
      "loss": 0.7981,
      "step": 8810
    },
    {
      "epoch": 0.7840696950840075,
      "grad_norm": 0.47563013434410095,
      "learning_rate": 2e-05,
      "loss": 0.7996,
      "step": 8820
    },
    {
      "epoch": 0.784958662992266,
      "grad_norm": 0.4220598042011261,
      "learning_rate": 2e-05,
      "loss": 0.8106,
      "step": 8830
    },
    {
      "epoch": 0.7858476309005245,
      "grad_norm": 0.41671910881996155,
      "learning_rate": 2e-05,
      "loss": 0.7922,
      "step": 8840
    },
    {
      "epoch": 0.786736598808783,
      "grad_norm": 0.4089275002479553,
      "learning_rate": 2e-05,
      "loss": 0.8012,
      "step": 8850
    },
    {
      "epoch": 0.7876255667170415,
      "grad_norm": 0.6030315160751343,
      "learning_rate": 2e-05,
      "loss": 0.8221,
      "step": 8860
    },
    {
      "epoch": 0.7885145346253001,
      "grad_norm": 0.47993841767311096,
      "learning_rate": 2e-05,
      "loss": 0.8227,
      "step": 8870
    },
    {
      "epoch": 0.7894035025335585,
      "grad_norm": 0.38673415780067444,
      "learning_rate": 2e-05,
      "loss": 0.8124,
      "step": 8880
    },
    {
      "epoch": 0.790292470441817,
      "grad_norm": 0.43819916248321533,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 8890
    },
    {
      "epoch": 0.7911814383500756,
      "grad_norm": 0.46342647075653076,
      "learning_rate": 2e-05,
      "loss": 0.8076,
      "step": 8900
    },
    {
      "epoch": 0.792070406258334,
      "grad_norm": 0.4472184181213379,
      "learning_rate": 2e-05,
      "loss": 0.8261,
      "step": 8910
    },
    {
      "epoch": 0.7929593741665926,
      "grad_norm": 0.43255186080932617,
      "learning_rate": 2e-05,
      "loss": 0.8222,
      "step": 8920
    },
    {
      "epoch": 0.7938483420748511,
      "grad_norm": 0.5088943243026733,
      "learning_rate": 2e-05,
      "loss": 0.7884,
      "step": 8930
    },
    {
      "epoch": 0.7947373099831097,
      "grad_norm": 0.4279080629348755,
      "learning_rate": 2e-05,
      "loss": 0.7665,
      "step": 8940
    },
    {
      "epoch": 0.7956262778913681,
      "grad_norm": 0.42185622453689575,
      "learning_rate": 2e-05,
      "loss": 0.7918,
      "step": 8950
    },
    {
      "epoch": 0.7965152457996266,
      "grad_norm": 0.47230008244514465,
      "learning_rate": 2e-05,
      "loss": 0.7901,
      "step": 8960
    },
    {
      "epoch": 0.7974042137078852,
      "grad_norm": 0.4274393320083618,
      "learning_rate": 2e-05,
      "loss": 0.7741,
      "step": 8970
    },
    {
      "epoch": 0.7982931816161437,
      "grad_norm": 0.4455341100692749,
      "learning_rate": 2e-05,
      "loss": 0.8232,
      "step": 8980
    },
    {
      "epoch": 0.7991821495244021,
      "grad_norm": 0.4812625050544739,
      "learning_rate": 2e-05,
      "loss": 0.7868,
      "step": 8990
    },
    {
      "epoch": 0.8000711174326607,
      "grad_norm": 0.45974430441856384,
      "learning_rate": 2e-05,
      "loss": 0.8049,
      "step": 9000
    },
    {
      "epoch": 0.8009600853409192,
      "grad_norm": 0.44384467601776123,
      "learning_rate": 2e-05,
      "loss": 0.8221,
      "step": 9010
    },
    {
      "epoch": 0.8018490532491777,
      "grad_norm": 0.42463982105255127,
      "learning_rate": 2e-05,
      "loss": 0.8227,
      "step": 9020
    },
    {
      "epoch": 0.8027380211574362,
      "grad_norm": 0.439568430185318,
      "learning_rate": 2e-05,
      "loss": 0.8126,
      "step": 9030
    },
    {
      "epoch": 0.8036269890656947,
      "grad_norm": 0.437845915555954,
      "learning_rate": 2e-05,
      "loss": 0.806,
      "step": 9040
    },
    {
      "epoch": 0.8045159569739533,
      "grad_norm": 0.4260004460811615,
      "learning_rate": 2e-05,
      "loss": 0.7921,
      "step": 9050
    },
    {
      "epoch": 0.8054049248822117,
      "grad_norm": 0.45636993646621704,
      "learning_rate": 2e-05,
      "loss": 0.8273,
      "step": 9060
    },
    {
      "epoch": 0.8062938927904703,
      "grad_norm": 0.4608955681324005,
      "learning_rate": 2e-05,
      "loss": 0.8024,
      "step": 9070
    },
    {
      "epoch": 0.8071828606987288,
      "grad_norm": 0.4401666820049286,
      "learning_rate": 2e-05,
      "loss": 0.7871,
      "step": 9080
    },
    {
      "epoch": 0.8080718286069873,
      "grad_norm": 0.4252643287181854,
      "learning_rate": 2e-05,
      "loss": 0.7894,
      "step": 9090
    },
    {
      "epoch": 0.8089607965152458,
      "grad_norm": 0.3989095985889435,
      "learning_rate": 2e-05,
      "loss": 0.8072,
      "step": 9100
    },
    {
      "epoch": 0.8098497644235043,
      "grad_norm": 0.4120527505874634,
      "learning_rate": 2e-05,
      "loss": 0.7981,
      "step": 9110
    },
    {
      "epoch": 0.8107387323317629,
      "grad_norm": 0.4508918821811676,
      "learning_rate": 2e-05,
      "loss": 0.844,
      "step": 9120
    },
    {
      "epoch": 0.8116277002400213,
      "grad_norm": 0.4133751690387726,
      "learning_rate": 2e-05,
      "loss": 0.8335,
      "step": 9130
    },
    {
      "epoch": 0.8125166681482798,
      "grad_norm": 0.4510253965854645,
      "learning_rate": 2e-05,
      "loss": 0.8305,
      "step": 9140
    },
    {
      "epoch": 0.8134056360565384,
      "grad_norm": 0.4445471167564392,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 9150
    },
    {
      "epoch": 0.8142946039647969,
      "grad_norm": 0.42254552245140076,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 9160
    },
    {
      "epoch": 0.8151835718730553,
      "grad_norm": 0.44365352392196655,
      "learning_rate": 2e-05,
      "loss": 0.7888,
      "step": 9170
    },
    {
      "epoch": 0.8160725397813139,
      "grad_norm": 0.43372127413749695,
      "learning_rate": 2e-05,
      "loss": 0.828,
      "step": 9180
    },
    {
      "epoch": 0.8169615076895724,
      "grad_norm": 0.42076003551483154,
      "learning_rate": 2e-05,
      "loss": 0.7985,
      "step": 9190
    },
    {
      "epoch": 0.817850475597831,
      "grad_norm": 0.4768742620944977,
      "learning_rate": 2e-05,
      "loss": 0.8078,
      "step": 9200
    },
    {
      "epoch": 0.8187394435060894,
      "grad_norm": 0.46173882484436035,
      "learning_rate": 2e-05,
      "loss": 0.7964,
      "step": 9210
    },
    {
      "epoch": 0.819628411414348,
      "grad_norm": 0.4002527594566345,
      "learning_rate": 2e-05,
      "loss": 0.7927,
      "step": 9220
    },
    {
      "epoch": 0.8205173793226065,
      "grad_norm": 0.45619165897369385,
      "learning_rate": 2e-05,
      "loss": 0.8088,
      "step": 9230
    },
    {
      "epoch": 0.8214063472308649,
      "grad_norm": 0.43854233622550964,
      "learning_rate": 2e-05,
      "loss": 0.813,
      "step": 9240
    },
    {
      "epoch": 0.8222953151391235,
      "grad_norm": 0.43604493141174316,
      "learning_rate": 2e-05,
      "loss": 0.8273,
      "step": 9250
    },
    {
      "epoch": 0.823184283047382,
      "grad_norm": 0.4195738732814789,
      "learning_rate": 2e-05,
      "loss": 0.8031,
      "step": 9260
    },
    {
      "epoch": 0.8240732509556405,
      "grad_norm": 0.40929853916168213,
      "learning_rate": 2e-05,
      "loss": 0.7727,
      "step": 9270
    },
    {
      "epoch": 0.824962218863899,
      "grad_norm": 0.41806504130363464,
      "learning_rate": 2e-05,
      "loss": 0.8144,
      "step": 9280
    },
    {
      "epoch": 0.8258511867721575,
      "grad_norm": 0.4030514061450958,
      "learning_rate": 2e-05,
      "loss": 0.7469,
      "step": 9290
    },
    {
      "epoch": 0.8267401546804161,
      "grad_norm": 0.49216851592063904,
      "learning_rate": 2e-05,
      "loss": 0.8381,
      "step": 9300
    },
    {
      "epoch": 0.8276291225886746,
      "grad_norm": 0.4287469685077667,
      "learning_rate": 2e-05,
      "loss": 0.7762,
      "step": 9310
    },
    {
      "epoch": 0.828518090496933,
      "grad_norm": 0.4712383449077606,
      "learning_rate": 2e-05,
      "loss": 0.8199,
      "step": 9320
    },
    {
      "epoch": 0.8294070584051916,
      "grad_norm": 0.48641642928123474,
      "learning_rate": 2e-05,
      "loss": 0.8205,
      "step": 9330
    },
    {
      "epoch": 0.8302960263134501,
      "grad_norm": 0.433573842048645,
      "learning_rate": 2e-05,
      "loss": 0.808,
      "step": 9340
    },
    {
      "epoch": 0.8311849942217086,
      "grad_norm": 0.4109918773174286,
      "learning_rate": 2e-05,
      "loss": 0.8181,
      "step": 9350
    },
    {
      "epoch": 0.8320739621299671,
      "grad_norm": 0.4149409234523773,
      "learning_rate": 2e-05,
      "loss": 0.7921,
      "step": 9360
    },
    {
      "epoch": 0.8329629300382256,
      "grad_norm": 0.4718027114868164,
      "learning_rate": 2e-05,
      "loss": 0.8167,
      "step": 9370
    },
    {
      "epoch": 0.8338518979464842,
      "grad_norm": 0.4404934048652649,
      "learning_rate": 2e-05,
      "loss": 0.7842,
      "step": 9380
    },
    {
      "epoch": 0.8347408658547426,
      "grad_norm": 0.413827121257782,
      "learning_rate": 2e-05,
      "loss": 0.7994,
      "step": 9390
    },
    {
      "epoch": 0.8356298337630012,
      "grad_norm": 0.44615113735198975,
      "learning_rate": 2e-05,
      "loss": 0.7989,
      "step": 9400
    },
    {
      "epoch": 0.8365188016712597,
      "grad_norm": 0.45582330226898193,
      "learning_rate": 2e-05,
      "loss": 0.8113,
      "step": 9410
    },
    {
      "epoch": 0.8374077695795181,
      "grad_norm": 0.4371797740459442,
      "learning_rate": 2e-05,
      "loss": 0.7981,
      "step": 9420
    },
    {
      "epoch": 0.8382967374877767,
      "grad_norm": 0.46183905005455017,
      "learning_rate": 2e-05,
      "loss": 0.8086,
      "step": 9430
    },
    {
      "epoch": 0.8391857053960352,
      "grad_norm": 0.46742501854896545,
      "learning_rate": 2e-05,
      "loss": 0.8161,
      "step": 9440
    },
    {
      "epoch": 0.8400746733042938,
      "grad_norm": 0.41528990864753723,
      "learning_rate": 2e-05,
      "loss": 0.7848,
      "step": 9450
    },
    {
      "epoch": 0.8409636412125522,
      "grad_norm": 0.42323559522628784,
      "learning_rate": 2e-05,
      "loss": 0.7783,
      "step": 9460
    },
    {
      "epoch": 0.8418526091208107,
      "grad_norm": 0.4313512444496155,
      "learning_rate": 2e-05,
      "loss": 0.835,
      "step": 9470
    },
    {
      "epoch": 0.8427415770290693,
      "grad_norm": 0.43949875235557556,
      "learning_rate": 2e-05,
      "loss": 0.8058,
      "step": 9480
    },
    {
      "epoch": 0.8436305449373278,
      "grad_norm": 0.44999250769615173,
      "learning_rate": 2e-05,
      "loss": 0.817,
      "step": 9490
    },
    {
      "epoch": 0.8445195128455862,
      "grad_norm": 0.42804211378097534,
      "learning_rate": 2e-05,
      "loss": 0.7768,
      "step": 9500
    },
    {
      "epoch": 0.8454084807538448,
      "grad_norm": 0.432513952255249,
      "learning_rate": 2e-05,
      "loss": 0.7893,
      "step": 9510
    },
    {
      "epoch": 0.8462974486621033,
      "grad_norm": 0.4346504807472229,
      "learning_rate": 2e-05,
      "loss": 0.8109,
      "step": 9520
    },
    {
      "epoch": 0.8471864165703618,
      "grad_norm": 0.41460680961608887,
      "learning_rate": 2e-05,
      "loss": 0.7948,
      "step": 9530
    },
    {
      "epoch": 0.8480753844786203,
      "grad_norm": 0.4242995083332062,
      "learning_rate": 2e-05,
      "loss": 0.786,
      "step": 9540
    },
    {
      "epoch": 0.8489643523868788,
      "grad_norm": 0.43096038699150085,
      "learning_rate": 2e-05,
      "loss": 0.7931,
      "step": 9550
    },
    {
      "epoch": 0.8498533202951374,
      "grad_norm": 0.5234934091567993,
      "learning_rate": 2e-05,
      "loss": 0.8149,
      "step": 9560
    },
    {
      "epoch": 0.8507422882033958,
      "grad_norm": 0.4435233175754547,
      "learning_rate": 2e-05,
      "loss": 0.782,
      "step": 9570
    },
    {
      "epoch": 0.8516312561116544,
      "grad_norm": 0.4421083331108093,
      "learning_rate": 2e-05,
      "loss": 0.7986,
      "step": 9580
    },
    {
      "epoch": 0.8525202240199129,
      "grad_norm": 0.4610024392604828,
      "learning_rate": 2e-05,
      "loss": 0.8222,
      "step": 9590
    },
    {
      "epoch": 0.8534091919281714,
      "grad_norm": 0.44692614674568176,
      "learning_rate": 2e-05,
      "loss": 0.804,
      "step": 9600
    },
    {
      "epoch": 0.8542981598364299,
      "grad_norm": 0.4296981990337372,
      "learning_rate": 2e-05,
      "loss": 0.7838,
      "step": 9610
    },
    {
      "epoch": 0.8551871277446884,
      "grad_norm": 0.42874860763549805,
      "learning_rate": 2e-05,
      "loss": 0.7932,
      "step": 9620
    },
    {
      "epoch": 0.856076095652947,
      "grad_norm": 0.4548424184322357,
      "learning_rate": 2e-05,
      "loss": 0.8305,
      "step": 9630
    },
    {
      "epoch": 0.8569650635612054,
      "grad_norm": 0.43946072459220886,
      "learning_rate": 2e-05,
      "loss": 0.7708,
      "step": 9640
    },
    {
      "epoch": 0.8578540314694639,
      "grad_norm": 0.42643532156944275,
      "learning_rate": 2e-05,
      "loss": 0.7956,
      "step": 9650
    },
    {
      "epoch": 0.8587429993777225,
      "grad_norm": 0.43128225207328796,
      "learning_rate": 2e-05,
      "loss": 0.8026,
      "step": 9660
    },
    {
      "epoch": 0.859631967285981,
      "grad_norm": 0.5285941362380981,
      "learning_rate": 2e-05,
      "loss": 0.8128,
      "step": 9670
    },
    {
      "epoch": 0.8605209351942394,
      "grad_norm": 0.4363904297351837,
      "learning_rate": 2e-05,
      "loss": 0.8056,
      "step": 9680
    },
    {
      "epoch": 0.861409903102498,
      "grad_norm": 0.4123137593269348,
      "learning_rate": 2e-05,
      "loss": 0.8051,
      "step": 9690
    },
    {
      "epoch": 0.8622988710107565,
      "grad_norm": 0.4067586064338684,
      "learning_rate": 2e-05,
      "loss": 0.7915,
      "step": 9700
    },
    {
      "epoch": 0.8631878389190151,
      "grad_norm": 0.4448852241039276,
      "learning_rate": 2e-05,
      "loss": 0.8171,
      "step": 9710
    },
    {
      "epoch": 0.8640768068272735,
      "grad_norm": 0.4848615229129791,
      "learning_rate": 2e-05,
      "loss": 0.7995,
      "step": 9720
    },
    {
      "epoch": 0.864965774735532,
      "grad_norm": 0.45604127645492554,
      "learning_rate": 2e-05,
      "loss": 0.7898,
      "step": 9730
    },
    {
      "epoch": 0.8658547426437906,
      "grad_norm": 0.4043772518634796,
      "learning_rate": 2e-05,
      "loss": 0.7932,
      "step": 9740
    },
    {
      "epoch": 0.866743710552049,
      "grad_norm": 0.41787466406822205,
      "learning_rate": 2e-05,
      "loss": 0.7949,
      "step": 9750
    },
    {
      "epoch": 0.8676326784603076,
      "grad_norm": 0.4674339294433594,
      "learning_rate": 2e-05,
      "loss": 0.7785,
      "step": 9760
    },
    {
      "epoch": 0.8685216463685661,
      "grad_norm": 0.4248625934123993,
      "learning_rate": 2e-05,
      "loss": 0.8273,
      "step": 9770
    },
    {
      "epoch": 0.8694106142768246,
      "grad_norm": 0.43547874689102173,
      "learning_rate": 2e-05,
      "loss": 0.8196,
      "step": 9780
    },
    {
      "epoch": 0.8702995821850831,
      "grad_norm": 0.40893542766571045,
      "learning_rate": 2e-05,
      "loss": 0.8045,
      "step": 9790
    },
    {
      "epoch": 0.8711885500933416,
      "grad_norm": 0.4327576160430908,
      "learning_rate": 2e-05,
      "loss": 0.8041,
      "step": 9800
    },
    {
      "epoch": 0.8720775180016002,
      "grad_norm": 0.41223272681236267,
      "learning_rate": 2e-05,
      "loss": 0.785,
      "step": 9810
    },
    {
      "epoch": 0.8729664859098587,
      "grad_norm": 0.46906736493110657,
      "learning_rate": 2e-05,
      "loss": 0.8042,
      "step": 9820
    },
    {
      "epoch": 0.8738554538181171,
      "grad_norm": 0.4456269443035126,
      "learning_rate": 2e-05,
      "loss": 0.7989,
      "step": 9830
    },
    {
      "epoch": 0.8747444217263757,
      "grad_norm": 0.4735063314437866,
      "learning_rate": 2e-05,
      "loss": 0.7988,
      "step": 9840
    },
    {
      "epoch": 0.8756333896346342,
      "grad_norm": 0.4465423822402954,
      "learning_rate": 2e-05,
      "loss": 0.8011,
      "step": 9850
    },
    {
      "epoch": 0.8765223575428926,
      "grad_norm": 0.5290051102638245,
      "learning_rate": 2e-05,
      "loss": 0.7486,
      "step": 9860
    },
    {
      "epoch": 0.8774113254511512,
      "grad_norm": 0.4081931710243225,
      "learning_rate": 2e-05,
      "loss": 0.7782,
      "step": 9870
    },
    {
      "epoch": 0.8783002933594097,
      "grad_norm": 0.44354891777038574,
      "learning_rate": 2e-05,
      "loss": 0.7828,
      "step": 9880
    },
    {
      "epoch": 0.8791892612676683,
      "grad_norm": 0.4221998453140259,
      "learning_rate": 2e-05,
      "loss": 0.766,
      "step": 9890
    },
    {
      "epoch": 0.8800782291759267,
      "grad_norm": 0.4387543201446533,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 9900
    },
    {
      "epoch": 0.8809671970841852,
      "grad_norm": 0.44046086072921753,
      "learning_rate": 2e-05,
      "loss": 0.7893,
      "step": 9910
    },
    {
      "epoch": 0.8818561649924438,
      "grad_norm": 0.428305983543396,
      "learning_rate": 2e-05,
      "loss": 0.7772,
      "step": 9920
    },
    {
      "epoch": 0.8827451329007023,
      "grad_norm": 0.46368592977523804,
      "learning_rate": 2e-05,
      "loss": 0.8216,
      "step": 9930
    },
    {
      "epoch": 0.8836341008089608,
      "grad_norm": 0.4238978326320648,
      "learning_rate": 2e-05,
      "loss": 0.7601,
      "step": 9940
    },
    {
      "epoch": 0.8845230687172193,
      "grad_norm": 0.4312882721424103,
      "learning_rate": 2e-05,
      "loss": 0.8272,
      "step": 9950
    },
    {
      "epoch": 0.8854120366254778,
      "grad_norm": 0.46035677194595337,
      "learning_rate": 2e-05,
      "loss": 0.8367,
      "step": 9960
    },
    {
      "epoch": 0.8863010045337363,
      "grad_norm": 0.4385230541229248,
      "learning_rate": 2e-05,
      "loss": 0.7805,
      "step": 9970
    },
    {
      "epoch": 0.8871899724419948,
      "grad_norm": 0.5122193098068237,
      "learning_rate": 2e-05,
      "loss": 0.7895,
      "step": 9980
    },
    {
      "epoch": 0.8880789403502534,
      "grad_norm": 0.4304662048816681,
      "learning_rate": 2e-05,
      "loss": 0.8068,
      "step": 9990
    },
    {
      "epoch": 0.8889679082585119,
      "grad_norm": 0.4421577751636505,
      "learning_rate": 2e-05,
      "loss": 0.8058,
      "step": 10000
    },
    {
      "epoch": 0.8898568761667703,
      "grad_norm": 0.42811697721481323,
      "learning_rate": 2e-05,
      "loss": 0.796,
      "step": 10010
    },
    {
      "epoch": 0.8907458440750289,
      "grad_norm": 0.4280967116355896,
      "learning_rate": 2e-05,
      "loss": 0.803,
      "step": 10020
    },
    {
      "epoch": 0.8916348119832874,
      "grad_norm": 0.44291746616363525,
      "learning_rate": 2e-05,
      "loss": 0.789,
      "step": 10030
    },
    {
      "epoch": 0.892523779891546,
      "grad_norm": 0.4556031823158264,
      "learning_rate": 2e-05,
      "loss": 0.8057,
      "step": 10040
    },
    {
      "epoch": 0.8934127477998044,
      "grad_norm": 0.43084263801574707,
      "learning_rate": 2e-05,
      "loss": 0.8135,
      "step": 10050
    },
    {
      "epoch": 0.8943017157080629,
      "grad_norm": 0.4144885540008545,
      "learning_rate": 2e-05,
      "loss": 0.8065,
      "step": 10060
    },
    {
      "epoch": 0.8951906836163215,
      "grad_norm": 0.4528968632221222,
      "learning_rate": 2e-05,
      "loss": 0.8104,
      "step": 10070
    },
    {
      "epoch": 0.8960796515245799,
      "grad_norm": 0.407578706741333,
      "learning_rate": 2e-05,
      "loss": 0.8126,
      "step": 10080
    },
    {
      "epoch": 0.8969686194328385,
      "grad_norm": 0.4567622244358063,
      "learning_rate": 2e-05,
      "loss": 0.7901,
      "step": 10090
    },
    {
      "epoch": 0.897857587341097,
      "grad_norm": 0.4707995653152466,
      "learning_rate": 2e-05,
      "loss": 0.8211,
      "step": 10100
    },
    {
      "epoch": 0.8987465552493555,
      "grad_norm": 0.4658237397670746,
      "learning_rate": 2e-05,
      "loss": 0.7928,
      "step": 10110
    },
    {
      "epoch": 0.899635523157614,
      "grad_norm": 0.4281655251979828,
      "learning_rate": 2e-05,
      "loss": 0.7983,
      "step": 10120
    },
    {
      "epoch": 0.9005244910658725,
      "grad_norm": 0.46391355991363525,
      "learning_rate": 2e-05,
      "loss": 0.7665,
      "step": 10130
    },
    {
      "epoch": 0.901413458974131,
      "grad_norm": 0.47400742769241333,
      "learning_rate": 2e-05,
      "loss": 0.7922,
      "step": 10140
    },
    {
      "epoch": 0.9023024268823896,
      "grad_norm": 0.45130470395088196,
      "learning_rate": 2e-05,
      "loss": 0.7977,
      "step": 10150
    },
    {
      "epoch": 0.903191394790648,
      "grad_norm": 0.48439857363700867,
      "learning_rate": 2e-05,
      "loss": 0.7656,
      "step": 10160
    },
    {
      "epoch": 0.9040803626989066,
      "grad_norm": 0.4267183244228363,
      "learning_rate": 2e-05,
      "loss": 0.7915,
      "step": 10170
    },
    {
      "epoch": 0.9049693306071651,
      "grad_norm": 0.4312112033367157,
      "learning_rate": 1.999957358086795e-05,
      "loss": 0.8046,
      "step": 10180
    },
    {
      "epoch": 0.9058582985154235,
      "grad_norm": 0.4299607574939728,
      "learning_rate": 1.9991993813020483e-05,
      "loss": 0.7887,
      "step": 10190
    },
    {
      "epoch": 0.9067472664236821,
      "grad_norm": 0.42948296666145325,
      "learning_rate": 1.9974946373028628e-05,
      "loss": 0.801,
      "step": 10200
    },
    {
      "epoch": 0.9076362343319406,
      "grad_norm": 0.6642808318138123,
      "learning_rate": 1.994844749501764e-05,
      "loss": 0.802,
      "step": 10210
    },
    {
      "epoch": 0.9085252022401992,
      "grad_norm": 0.47548291087150574,
      "learning_rate": 1.9912522413632988e-05,
      "loss": 0.8125,
      "step": 10220
    },
    {
      "epoch": 0.9094141701484576,
      "grad_norm": 0.4876030683517456,
      "learning_rate": 1.9867205340009656e-05,
      "loss": 0.7914,
      "step": 10230
    },
    {
      "epoch": 0.9103031380567161,
      "grad_norm": 0.7801253795623779,
      "learning_rate": 1.9812539429193164e-05,
      "loss": 0.7729,
      "step": 10240
    },
    {
      "epoch": 0.9111921059649747,
      "grad_norm": 0.5149628520011902,
      "learning_rate": 1.9748576739043423e-05,
      "loss": 0.7974,
      "step": 10250
    },
    {
      "epoch": 0.9120810738732332,
      "grad_norm": 0.444723904132843,
      "learning_rate": 1.967537818066047e-05,
      "loss": 0.7739,
      "step": 10260
    },
    {
      "epoch": 0.9129700417814917,
      "grad_norm": 0.41241803765296936,
      "learning_rate": 1.959301346037941e-05,
      "loss": 0.7859,
      "step": 10270
    },
    {
      "epoch": 0.9138590096897502,
      "grad_norm": 0.41188496351242065,
      "learning_rate": 1.9501561013389646e-05,
      "loss": 0.772,
      "step": 10280
    },
    {
      "epoch": 0.9147479775980087,
      "grad_norm": 0.42549169063568115,
      "learning_rate": 1.9401107929041795e-05,
      "loss": 0.7921,
      "step": 10290
    },
    {
      "epoch": 0.9156369455062672,
      "grad_norm": 0.4924401640892029,
      "learning_rate": 1.9291749867913225e-05,
      "loss": 0.8059,
      "step": 10300
    },
    {
      "epoch": 0.9165259134145257,
      "grad_norm": 0.4294026494026184,
      "learning_rate": 1.91735909707114e-05,
      "loss": 0.8186,
      "step": 10310
    },
    {
      "epoch": 0.9174148813227843,
      "grad_norm": 0.44960930943489075,
      "learning_rate": 1.9046743759101545e-05,
      "loss": 0.8035,
      "step": 10320
    },
    {
      "epoch": 0.9183038492310428,
      "grad_norm": 0.44129130244255066,
      "learning_rate": 1.8911329028553324e-05,
      "loss": 0.7911,
      "step": 10330
    },
    {
      "epoch": 0.9191928171393012,
      "grad_norm": 0.44071710109710693,
      "learning_rate": 1.876747573330835e-05,
      "loss": 0.801,
      "step": 10340
    },
    {
      "epoch": 0.9200817850475598,
      "grad_norm": 0.43820035457611084,
      "learning_rate": 1.8615320863578244e-05,
      "loss": 0.7783,
      "step": 10350
    },
    {
      "epoch": 0.9209707529558183,
      "grad_norm": 0.419459730386734,
      "learning_rate": 1.8455009315090036e-05,
      "loss": 0.777,
      "step": 10360
    },
    {
      "epoch": 0.9218597208640769,
      "grad_norm": 0.4235645830631256,
      "learning_rate": 1.8286693751103294e-05,
      "loss": 0.7851,
      "step": 10370
    },
    {
      "epoch": 0.9227486887723353,
      "grad_norm": 0.41444143652915955,
      "learning_rate": 1.811053445703024e-05,
      "loss": 0.7733,
      "step": 10380
    },
    {
      "epoch": 0.9236376566805938,
      "grad_norm": 0.43107080459594727,
      "learning_rate": 1.7926699187797375e-05,
      "loss": 0.7935,
      "step": 10390
    },
    {
      "epoch": 0.9245266245888524,
      "grad_norm": 0.4274972081184387,
      "learning_rate": 1.7735363008093995e-05,
      "loss": 0.7575,
      "step": 10400
    },
    {
      "epoch": 0.9254155924971108,
      "grad_norm": 0.42977020144462585,
      "learning_rate": 1.7536708125659577e-05,
      "loss": 0.7997,
      "step": 10410
    },
    {
      "epoch": 0.9263045604053693,
      "grad_norm": 0.45263445377349854,
      "learning_rate": 1.733092371776905e-05,
      "loss": 0.7953,
      "step": 10420
    },
    {
      "epoch": 0.9271935283136279,
      "grad_norm": 0.4227323532104492,
      "learning_rate": 1.7118205751080915e-05,
      "loss": 0.7796,
      "step": 10430
    },
    {
      "epoch": 0.9280824962218864,
      "grad_norm": 0.44623422622680664,
      "learning_rate": 1.6898756795019935e-05,
      "loss": 0.8023,
      "step": 10440
    },
    {
      "epoch": 0.9289714641301449,
      "grad_norm": 0.4756172001361847,
      "learning_rate": 1.667278582887211e-05,
      "loss": 0.8071,
      "step": 10450
    },
    {
      "epoch": 0.9298604320384034,
      "grad_norm": 0.41845107078552246,
      "learning_rate": 1.6440508042775545e-05,
      "loss": 0.784,
      "step": 10460
    },
    {
      "epoch": 0.930749399946662,
      "grad_norm": 0.4344635605812073,
      "learning_rate": 1.6202144632796796e-05,
      "loss": 0.8015,
      "step": 10470
    },
    {
      "epoch": 0.9316383678549205,
      "grad_norm": 0.41996198892593384,
      "learning_rate": 1.595792259028783e-05,
      "loss": 0.8201,
      "step": 10480
    },
    {
      "epoch": 0.9325273357631789,
      "grad_norm": 0.4551287591457367,
      "learning_rate": 1.570807448572419e-05,
      "loss": 0.8005,
      "step": 10490
    },
    {
      "epoch": 0.9334163036714375,
      "grad_norm": 0.41586896777153015,
      "learning_rate": 1.5452838247230197e-05,
      "loss": 0.7779,
      "step": 10500
    },
    {
      "epoch": 0.934305271579696,
      "grad_norm": 0.4236511290073395,
      "learning_rate": 1.5192456934002106e-05,
      "loss": 0.8041,
      "step": 10510
    },
    {
      "epoch": 0.9351942394879544,
      "grad_norm": 0.42870306968688965,
      "learning_rate": 1.492717850484502e-05,
      "loss": 0.8108,
      "step": 10520
    },
    {
      "epoch": 0.936083207396213,
      "grad_norm": 0.43775588274002075,
      "learning_rate": 1.4657255582043904e-05,
      "loss": 0.7892,
      "step": 10530
    },
    {
      "epoch": 0.9369721753044715,
      "grad_norm": 0.4598296880722046,
      "learning_rate": 1.4382945210793663e-05,
      "loss": 0.7943,
      "step": 10540
    },
    {
      "epoch": 0.9378611432127301,
      "grad_norm": 0.46645382046699524,
      "learning_rate": 1.4104508614417267e-05,
      "loss": 0.7537,
      "step": 10550
    },
    {
      "epoch": 0.9387501111209885,
      "grad_norm": 0.4148613512516022,
      "learning_rate": 1.382221094560512e-05,
      "loss": 0.7649,
      "step": 10560
    },
    {
      "epoch": 0.939639079029247,
      "grad_norm": 0.41749846935272217,
      "learning_rate": 1.3536321033912506e-05,
      "loss": 0.7957,
      "step": 10570
    },
    {
      "epoch": 0.9405280469375056,
      "grad_norm": 0.47564345598220825,
      "learning_rate": 1.3247111129755609e-05,
      "loss": 0.8025,
      "step": 10580
    },
    {
      "epoch": 0.9414170148457641,
      "grad_norm": 0.44455015659332275,
      "learning_rate": 1.2954856645149868e-05,
      "loss": 0.7939,
      "step": 10590
    },
    {
      "epoch": 0.9423059827540226,
      "grad_norm": 0.5462555885314941,
      "learning_rate": 1.2659835891437553e-05,
      "loss": 0.7517,
      "step": 10600
    },
    {
      "epoch": 0.9431949506622811,
      "grad_norm": 0.4248961806297302,
      "learning_rate": 1.236232981425436e-05,
      "loss": 0.8019,
      "step": 10610
    },
    {
      "epoch": 0.9440839185705396,
      "grad_norm": 0.4119073450565338,
      "learning_rate": 1.2062621725987402e-05,
      "loss": 0.7664,
      "step": 10620
    },
    {
      "epoch": 0.9449728864787981,
      "grad_norm": 0.43481484055519104,
      "learning_rate": 1.1760997035979318e-05,
      "loss": 0.7793,
      "step": 10630
    },
    {
      "epoch": 0.9458618543870566,
      "grad_norm": 0.42372000217437744,
      "learning_rate": 1.1457742978735514e-05,
      "loss": 0.7593,
      "step": 10640
    },
    {
      "epoch": 0.9467508222953152,
      "grad_norm": 0.3973965346813202,
      "learning_rate": 1.1153148340393298e-05,
      "loss": 0.7814,
      "step": 10650
    },
    {
      "epoch": 0.9476397902035737,
      "grad_norm": 0.4345109462738037,
      "learning_rate": 1.0847503183713424e-05,
      "loss": 0.7837,
      "step": 10660
    },
    {
      "epoch": 0.9485287581118321,
      "grad_norm": 0.4273507297039032,
      "learning_rate": 1.0541098571855896e-05,
      "loss": 0.783,
      "step": 10670
    },
    {
      "epoch": 0.9494177260200907,
      "grad_norm": 0.4440676271915436,
      "learning_rate": 1.0234226291203123e-05,
      "loss": 0.769,
      "step": 10680
    },
    {
      "epoch": 0.9503066939283492,
      "grad_norm": 0.439797580242157,
      "learning_rate": 9.927178573494333e-06,
      "loss": 0.7862,
      "step": 10690
    },
    {
      "epoch": 0.9511956618366078,
      "grad_norm": 0.4213857352733612,
      "learning_rate": 9.620247817535902e-06,
      "loss": 0.8061,
      "step": 10700
    },
    {
      "epoch": 0.9520846297448662,
      "grad_norm": 0.4127386510372162,
      "learning_rate": 9.313726310752581e-06,
      "loss": 0.7708,
      "step": 10710
    },
    {
      "epoch": 0.9529735976531247,
      "grad_norm": 0.38233572244644165,
      "learning_rate": 9.00790595084477e-06,
      "loss": 0.7735,
      "step": 10720
    },
    {
      "epoch": 0.9538625655613833,
      "grad_norm": 0.4532533586025238,
      "learning_rate": 8.703077967816944e-06,
      "loss": 0.8157,
      "step": 10730
    },
    {
      "epoch": 0.9547515334696417,
      "grad_norm": 0.45771583914756775,
      "learning_rate": 8.39953264664192e-06,
      "loss": 0.7701,
      "step": 10740
    },
    {
      "epoch": 0.9556405013779002,
      "grad_norm": 0.435138463973999,
      "learning_rate": 8.097559050825058e-06,
      "loss": 0.7993,
      "step": 10750
    },
    {
      "epoch": 0.9565294692861588,
      "grad_norm": 0.4369286000728607,
      "learning_rate": 7.79744474713166e-06,
      "loss": 0.7572,
      "step": 10760
    },
    {
      "epoch": 0.9574184371944173,
      "grad_norm": 0.4499463737010956,
      "learning_rate": 7.499475531739696e-06,
      "loss": 0.7551,
      "step": 10770
    },
    {
      "epoch": 0.9583074051026758,
      "grad_norm": 0.41811731457710266,
      "learning_rate": 7.203935158078662e-06,
      "loss": 0.7756,
      "step": 10780
    },
    {
      "epoch": 0.9591963730109343,
      "grad_norm": 0.42284777760505676,
      "learning_rate": 6.911105066613709e-06,
      "loss": 0.7588,
      "step": 10790
    },
    {
      "epoch": 0.9600853409191928,
      "grad_norm": 0.417892724275589,
      "learning_rate": 6.621264116832394e-06,
      "loss": 0.7776,
      "step": 10800
    },
    {
      "epoch": 0.9609743088274514,
      "grad_norm": 0.42348670959472656,
      "learning_rate": 6.334688321689274e-06,
      "loss": 0.7635,
      "step": 10810
    },
    {
      "epoch": 0.9618632767357098,
      "grad_norm": 0.4163253903388977,
      "learning_rate": 6.051650584761237e-06,
      "loss": 0.7818,
      "step": 10820
    },
    {
      "epoch": 0.9627522446439684,
      "grad_norm": 0.43692997097969055,
      "learning_rate": 5.772420440363841e-06,
      "loss": 0.763,
      "step": 10830
    },
    {
      "epoch": 0.9636412125522269,
      "grad_norm": 0.4110613465309143,
      "learning_rate": 5.497263796876207e-06,
      "loss": 0.7619,
      "step": 10840
    },
    {
      "epoch": 0.9645301804604853,
      "grad_norm": 0.5100053548812866,
      "learning_rate": 5.226442683518813e-06,
      "loss": 0.7982,
      "step": 10850
    },
    {
      "epoch": 0.9654191483687439,
      "grad_norm": 0.4234011471271515,
      "learning_rate": 4.960215000825432e-06,
      "loss": 0.7722,
      "step": 10860
    },
    {
      "epoch": 0.9663081162770024,
      "grad_norm": 0.45340582728385925,
      "learning_rate": 4.698834275046711e-06,
      "loss": 0.757,
      "step": 10870
    },
    {
      "epoch": 0.967197084185261,
      "grad_norm": 0.40960660576820374,
      "learning_rate": 4.442549416719423e-06,
      "loss": 0.7873,
      "step": 10880
    },
    {
      "epoch": 0.9680860520935194,
      "grad_norm": 0.4064622223377228,
      "learning_rate": 4.1916044836311565e-06,
      "loss": 0.7369,
      "step": 10890
    },
    {
      "epoch": 0.9689750200017779,
      "grad_norm": 0.4462379217147827,
      "learning_rate": 3.94623844840626e-06,
      "loss": 0.7636,
      "step": 10900
    },
    {
      "epoch": 0.9698639879100365,
      "grad_norm": 0.4036785960197449,
      "learning_rate": 3.7066849709343467e-06,
      "loss": 0.7681,
      "step": 10910
    },
    {
      "epoch": 0.9707529558182949,
      "grad_norm": 0.4320697486400604,
      "learning_rate": 3.473172175858059e-06,
      "loss": 0.7724,
      "step": 10920
    },
    {
      "epoch": 0.9716419237265534,
      "grad_norm": 0.4174574315547943,
      "learning_rate": 3.2459224353320046e-06,
      "loss": 0.7511,
      "step": 10930
    },
    {
      "epoch": 0.972530891634812,
      "grad_norm": 0.4044733941555023,
      "learning_rate": 3.025152157259735e-06,
      "loss": 0.7948,
      "step": 10940
    },
    {
      "epoch": 0.9734198595430705,
      "grad_norm": 0.4187937378883362,
      "learning_rate": 2.811071579210425e-06,
      "loss": 0.7893,
      "step": 10950
    },
    {
      "epoch": 0.974308827451329,
      "grad_norm": 0.414851576089859,
      "learning_rate": 2.603884568211513e-06,
      "loss": 0.7463,
      "step": 10960
    },
    {
      "epoch": 0.9751977953595875,
      "grad_norm": 0.4280945360660553,
      "learning_rate": 2.4037884266079434e-06,
      "loss": 0.7618,
      "step": 10970
    },
    {
      "epoch": 0.976086763267846,
      "grad_norm": 0.3985748291015625,
      "learning_rate": 2.2109737041728955e-06,
      "loss": 0.7719,
      "step": 10980
    },
    {
      "epoch": 0.9769757311761046,
      "grad_norm": 0.4270165264606476,
      "learning_rate": 2.0256240166489454e-06,
      "loss": 0.725,
      "step": 10990
    },
    {
      "epoch": 0.977864699084363,
      "grad_norm": 0.4121732711791992,
      "learning_rate": 1.847915870892413e-06,
      "loss": 0.7459,
      "step": 11000
    },
    {
      "epoch": 0.9787536669926216,
      "grad_norm": 0.41830912232398987,
      "learning_rate": 1.6780184967874715e-06,
      "loss": 0.7685,
      "step": 11010
    },
    {
      "epoch": 0.9796426349008801,
      "grad_norm": 0.4337053596973419,
      "learning_rate": 1.5160936860900294e-06,
      "loss": 0.7762,
      "step": 11020
    },
    {
      "epoch": 0.9805316028091385,
      "grad_norm": 0.38978487253189087,
      "learning_rate": 1.3622956383548762e-06,
      "loss": 0.7697,
      "step": 11030
    },
    {
      "epoch": 0.9814205707173971,
      "grad_norm": 0.41118234395980835,
      "learning_rate": 1.216770814092834e-06,
      "loss": 0.7687,
      "step": 11040
    },
    {
      "epoch": 0.9823095386256556,
      "grad_norm": 0.4283043444156647,
      "learning_rate": 1.0796577952977013e-06,
      "loss": 0.7768,
      "step": 11050
    },
    {
      "epoch": 0.9831985065339142,
      "grad_norm": 0.4294658899307251,
      "learning_rate": 9.510871534758705e-07,
      "loss": 0.7801,
      "step": 11060
    },
    {
      "epoch": 0.9840874744421726,
      "grad_norm": 0.3994626998901367,
      "learning_rate": 8.311813253042334e-07,
      "loss": 0.7368,
      "step": 11070
    },
    {
      "epoch": 0.9849764423504311,
      "grad_norm": 0.4273434579372406,
      "learning_rate": 7.200544960348308e-07,
      "loss": 0.7947,
      "step": 11080
    },
    {
      "epoch": 0.9858654102586897,
      "grad_norm": 0.41497233510017395,
      "learning_rate": 6.178124907572545e-07,
      "loss": 0.7804,
      "step": 11090
    },
    {
      "epoch": 0.9867543781669482,
      "grad_norm": 0.42127880454063416,
      "learning_rate": 5.245526736223528e-07,
      "loss": 0.7624,
      "step": 11100
    },
    {
      "epoch": 0.9876433460752067,
      "grad_norm": 0.4057467579841614,
      "learning_rate": 4.403638551232205e-07,
      "loss": 0.7529,
      "step": 11110
    },
    {
      "epoch": 0.9885323139834652,
      "grad_norm": 0.39926406741142273,
      "learning_rate": 3.6532620752175616e-07,
      "loss": 0.8013,
      "step": 11120
    },
    {
      "epoch": 0.9894212818917237,
      "grad_norm": 0.40959155559539795,
      "learning_rate": 2.9951118850133247e-07,
      "loss": 0.7634,
      "step": 11130
    },
    {
      "epoch": 0.9903102497999822,
      "grad_norm": 0.438797265291214,
      "learning_rate": 2.429814731182774e-07,
      "loss": 0.7839,
      "step": 11140
    },
    {
      "epoch": 0.9911992177082407,
      "grad_norm": 0.39777156710624695,
      "learning_rate": 1.957908941169813e-07,
      "loss": 0.7875,
      "step": 11150
    },
    {
      "epoch": 0.9920881856164993,
      "grad_norm": 0.40033936500549316,
      "learning_rate": 1.5798439066545254e-07,
      "loss": 0.7679,
      "step": 11160
    },
    {
      "epoch": 0.9929771535247578,
      "grad_norm": 0.41727349162101746,
      "learning_rate": 1.2959796556014471e-07,
      "loss": 0.7654,
      "step": 11170
    },
    {
      "epoch": 0.9938661214330162,
      "grad_norm": 0.47736215591430664,
      "learning_rate": 1.1065865094081884e-07,
      "loss": 0.7627,
      "step": 11180
    },
    {
      "epoch": 0.9947550893412748,
      "grad_norm": 0.4145434498786926,
      "learning_rate": 1.011844825480734e-07,
      "loss": 0.7898,
      "step": 11190
    },
    {
      "epoch": 0.9951106765045782,
      "step": 11194,
      "total_flos": 4.547514943786923e+19,
      "train_loss": 0.00443073098140251,
      "train_runtime": 609.4553,
      "train_samples_per_second": 220.407,
      "train_steps_per_second": 18.367
    }
  ],
  "logging_steps": 10,
  "max_steps": 11194,
  "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": 4.547514943786923e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}