{
  "best_metric": 1.4924039840698242,
  "best_model_checkpoint": "./output/checkpoint-3900",
  "epoch": 0.34580599397056216,
  "eval_steps": 150,
  "global_step": 3900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008866820358219542,
      "grad_norm": 12.221611976623535,
      "learning_rate": 1.0000000000000004e-05,
      "loss": 1.5409,
      "step": 10
    },
    {
      "epoch": 0.0017733640716439084,
      "grad_norm": 10.172511100769043,
      "learning_rate": 2.000000000000001e-05,
      "loss": 1.5034,
      "step": 20
    },
    {
      "epoch": 0.0026600461074658627,
      "grad_norm": 7.491011619567871,
      "learning_rate": 3.0000000000000008e-05,
      "loss": 1.5275,
      "step": 30
    },
    {
      "epoch": 0.003546728143287817,
      "grad_norm": 8.695981979370117,
      "learning_rate": 4.000000000000002e-05,
      "loss": 1.6105,
      "step": 40
    },
    {
      "epoch": 0.004433410179109771,
      "grad_norm": 9.496545791625977,
      "learning_rate": 5.0000000000000016e-05,
      "loss": 1.5488,
      "step": 50
    },
    {
      "epoch": 0.005320092214931725,
      "grad_norm": 8.968175888061523,
      "learning_rate": 6.0000000000000015e-05,
      "loss": 1.5128,
      "step": 60
    },
    {
      "epoch": 0.00620677425075368,
      "grad_norm": 9.554357528686523,
      "learning_rate": 7.000000000000002e-05,
      "loss": 1.5392,
      "step": 70
    },
    {
      "epoch": 0.007093456286575634,
      "grad_norm": 10.577646255493164,
      "learning_rate": 8.000000000000003e-05,
      "loss": 1.7024,
      "step": 80
    },
    {
      "epoch": 0.007980138322397589,
      "grad_norm": 7.967463493347168,
      "learning_rate": 9.000000000000003e-05,
      "loss": 1.6036,
      "step": 90
    },
    {
      "epoch": 0.008866820358219542,
      "grad_norm": 10.403460502624512,
      "learning_rate": 0.00010000000000000003,
      "loss": 1.7425,
      "step": 100
    },
    {
      "epoch": 0.009753502394041496,
      "grad_norm": 7.898177623748779,
      "learning_rate": 9.999897234791833e-05,
      "loss": 1.6833,
      "step": 110
    },
    {
      "epoch": 0.01064018442986345,
      "grad_norm": 9.69663143157959,
      "learning_rate": 9.9995889433916e-05,
      "loss": 1.5979,
      "step": 120
    },
    {
      "epoch": 0.011526866465685405,
      "grad_norm": 10.984465599060059,
      "learning_rate": 9.999075138471954e-05,
      "loss": 1.8234,
      "step": 130
    },
    {
      "epoch": 0.01241354850150736,
      "grad_norm": 7.519535541534424,
      "learning_rate": 9.998355841153402e-05,
      "loss": 1.5598,
      "step": 140
    },
    {
      "epoch": 0.013300230537329314,
      "grad_norm": 10.765931129455566,
      "learning_rate": 9.997431081003442e-05,
      "loss": 1.6027,
      "step": 150
    },
    {
      "epoch": 0.013300230537329314,
      "eval_loss": 1.7432395219802856,
      "eval_runtime": 59.4356,
      "eval_samples_per_second": 8.412,
      "eval_steps_per_second": 8.412,
      "step": 150
    },
    {
      "epoch": 0.014186912573151267,
      "grad_norm": 7.04581356048584,
      "learning_rate": 9.996300896035342e-05,
      "loss": 1.625,
      "step": 160
    },
    {
      "epoch": 0.015073594608973222,
      "grad_norm": 10.549965858459473,
      "learning_rate": 9.994965332706576e-05,
      "loss": 1.7696,
      "step": 170
    },
    {
      "epoch": 0.015960276644795178,
      "grad_norm": 17.5423641204834,
      "learning_rate": 9.993424445916925e-05,
      "loss": 1.6977,
      "step": 180
    },
    {
      "epoch": 0.01684695868061713,
      "grad_norm": 15.295858383178711,
      "learning_rate": 9.991678299006208e-05,
      "loss": 1.7496,
      "step": 190
    },
    {
      "epoch": 0.017733640716439084,
      "grad_norm": 14.050594329833984,
      "learning_rate": 9.989726963751685e-05,
      "loss": 1.6698,
      "step": 200
    },
    {
      "epoch": 0.01862032275226104,
      "grad_norm": 9.636832237243652,
      "learning_rate": 9.987570520365106e-05,
      "loss": 1.7278,
      "step": 210
    },
    {
      "epoch": 0.019507004788082993,
      "grad_norm": 10.37330150604248,
      "learning_rate": 9.985209057489412e-05,
      "loss": 1.6985,
      "step": 220
    },
    {
      "epoch": 0.02039368682390495,
      "grad_norm": 8.965282440185547,
      "learning_rate": 9.982642672195095e-05,
      "loss": 1.652,
      "step": 230
    },
    {
      "epoch": 0.0212803688597269,
      "grad_norm": 11.881516456604004,
      "learning_rate": 9.979871469976199e-05,
      "loss": 1.6963,
      "step": 240
    },
    {
      "epoch": 0.022167050895548854,
      "grad_norm": 8.806026458740234,
      "learning_rate": 9.976895564745994e-05,
      "loss": 1.6385,
      "step": 250
    },
    {
      "epoch": 0.02305373293137081,
      "grad_norm": 8.204315185546875,
      "learning_rate": 9.97371507883229e-05,
      "loss": 1.8237,
      "step": 260
    },
    {
      "epoch": 0.023940414967192764,
      "grad_norm": 7.4131879806518555,
      "learning_rate": 9.970330142972404e-05,
      "loss": 1.6853,
      "step": 270
    },
    {
      "epoch": 0.02482709700301472,
      "grad_norm": 8.07067584991455,
      "learning_rate": 9.966740896307794e-05,
      "loss": 1.6961,
      "step": 280
    },
    {
      "epoch": 0.025713779038836673,
      "grad_norm": 10.194974899291992,
      "learning_rate": 9.962947486378328e-05,
      "loss": 1.6454,
      "step": 290
    },
    {
      "epoch": 0.02660046107465863,
      "grad_norm": 9.990214347839355,
      "learning_rate": 9.958950069116233e-05,
      "loss": 1.7213,
      "step": 300
    },
    {
      "epoch": 0.02660046107465863,
      "eval_loss": 1.710451364517212,
      "eval_runtime": 59.6513,
      "eval_samples_per_second": 8.382,
      "eval_steps_per_second": 8.382,
      "step": 300
    },
    {
      "epoch": 0.02748714311048058,
      "grad_norm": 9.65303897857666,
      "learning_rate": 9.954748808839677e-05,
      "loss": 1.6927,
      "step": 310
    },
    {
      "epoch": 0.028373825146302534,
      "grad_norm": 11.48330307006836,
      "learning_rate": 9.950343878246013e-05,
      "loss": 1.6804,
      "step": 320
    },
    {
      "epoch": 0.02926050718212449,
      "grad_norm": 8.232380867004395,
      "learning_rate": 9.945735458404684e-05,
      "loss": 1.6682,
      "step": 330
    },
    {
      "epoch": 0.030147189217946443,
      "grad_norm": 8.22367000579834,
      "learning_rate": 9.940923738749781e-05,
      "loss": 1.705,
      "step": 340
    },
    {
      "epoch": 0.0310338712537684,
      "grad_norm": 8.59586238861084,
      "learning_rate": 9.935908917072255e-05,
      "loss": 1.7646,
      "step": 350
    },
    {
      "epoch": 0.031920553289590356,
      "grad_norm": 11.29736328125,
      "learning_rate": 9.930691199511777e-05,
      "loss": 1.5999,
      "step": 360
    },
    {
      "epoch": 0.032807235325412305,
      "grad_norm": 7.492193222045898,
      "learning_rate": 9.925270800548287e-05,
      "loss": 1.7898,
      "step": 370
    },
    {
      "epoch": 0.03369391736123426,
      "grad_norm": 9.845335006713867,
      "learning_rate": 9.919647942993151e-05,
      "loss": 1.7107,
      "step": 380
    },
    {
      "epoch": 0.03458059939705622,
      "grad_norm": 6.399631500244141,
      "learning_rate": 9.913822857980023e-05,
      "loss": 1.8382,
      "step": 390
    },
    {
      "epoch": 0.03546728143287817,
      "grad_norm": 14.003453254699707,
      "learning_rate": 9.90779578495533e-05,
      "loss": 1.5307,
      "step": 400
    },
    {
      "epoch": 0.03635396346870012,
      "grad_norm": 7.186993598937988,
      "learning_rate": 9.90156697166844e-05,
      "loss": 1.6269,
      "step": 410
    },
    {
      "epoch": 0.03724064550452208,
      "grad_norm": 10.279706001281738,
      "learning_rate": 9.895136674161468e-05,
      "loss": 1.7234,
      "step": 420
    },
    {
      "epoch": 0.038127327540344036,
      "grad_norm": 15.368351936340332,
      "learning_rate": 9.888505156758762e-05,
      "loss": 1.6987,
      "step": 430
    },
    {
      "epoch": 0.039014009576165985,
      "grad_norm": 12.989547729492188,
      "learning_rate": 9.881672692056024e-05,
      "loss": 1.7605,
      "step": 440
    },
    {
      "epoch": 0.03990069161198794,
      "grad_norm": 10.625384330749512,
      "learning_rate": 9.87463956090912e-05,
      "loss": 1.7047,
      "step": 450
    },
    {
      "epoch": 0.03990069161198794,
      "eval_loss": 1.7071540355682373,
      "eval_runtime": 59.4581,
      "eval_samples_per_second": 8.409,
      "eval_steps_per_second": 8.409,
      "step": 450
    },
    {
      "epoch": 0.0407873736478099,
      "grad_norm": 13.621452331542969,
      "learning_rate": 9.867406052422526e-05,
      "loss": 1.6615,
      "step": 460
    },
    {
      "epoch": 0.04167405568363185,
      "grad_norm": 11.978480339050293,
      "learning_rate": 9.859972463937443e-05,
      "loss": 1.712,
      "step": 470
    },
    {
      "epoch": 0.0425607377194538,
      "grad_norm": 8.805001258850098,
      "learning_rate": 9.852339101019577e-05,
      "loss": 1.6285,
      "step": 480
    },
    {
      "epoch": 0.04344741975527576,
      "grad_norm": 6.610208034515381,
      "learning_rate": 9.84450627744658e-05,
      "loss": 1.6863,
      "step": 490
    },
    {
      "epoch": 0.04433410179109771,
      "grad_norm": 8.630401611328125,
      "learning_rate": 9.83647431519515e-05,
      "loss": 1.7646,
      "step": 500
    },
    {
      "epoch": 0.045220783826919665,
      "grad_norm": 9.726890563964844,
      "learning_rate": 9.828243544427798e-05,
      "loss": 1.7782,
      "step": 510
    },
    {
      "epoch": 0.04610746586274162,
      "grad_norm": 9.516313552856445,
      "learning_rate": 9.81981430347927e-05,
      "loss": 1.7227,
      "step": 520
    },
    {
      "epoch": 0.04699414789856358,
      "grad_norm": 8.171401023864746,
      "learning_rate": 9.811186938842648e-05,
      "loss": 1.6726,
      "step": 530
    },
    {
      "epoch": 0.04788082993438553,
      "grad_norm": 9.201889991760254,
      "learning_rate": 9.8023618051551e-05,
      "loss": 1.4651,
      "step": 540
    },
    {
      "epoch": 0.04876751197020748,
      "grad_norm": 8.409120559692383,
      "learning_rate": 9.793339265183306e-05,
      "loss": 1.5729,
      "step": 550
    },
    {
      "epoch": 0.04965419400602944,
      "grad_norm": 8.624395370483398,
      "learning_rate": 9.784119689808547e-05,
      "loss": 1.7279,
      "step": 560
    },
    {
      "epoch": 0.05054087604185139,
      "grad_norm": 9.849170684814453,
      "learning_rate": 9.774703458011455e-05,
      "loss": 1.7016,
      "step": 570
    },
    {
      "epoch": 0.051427558077673345,
      "grad_norm": 10.26559829711914,
      "learning_rate": 9.765090956856439e-05,
      "loss": 1.6263,
      "step": 580
    },
    {
      "epoch": 0.0523142401134953,
      "grad_norm": 8.199737548828125,
      "learning_rate": 9.755282581475771e-05,
      "loss": 1.6802,
      "step": 590
    },
    {
      "epoch": 0.05320092214931726,
      "grad_norm": 8.513672828674316,
      "learning_rate": 9.745278735053346e-05,
      "loss": 1.8307,
      "step": 600
    },
    {
      "epoch": 0.05320092214931726,
      "eval_loss": 1.7056752443313599,
      "eval_runtime": 59.3846,
      "eval_samples_per_second": 8.42,
      "eval_steps_per_second": 8.42,
      "step": 600
    },
    {
      "epoch": 0.05408760418513921,
      "grad_norm": 6.43255615234375,
      "learning_rate": 9.73507982880811e-05,
      "loss": 1.6386,
      "step": 610
    },
    {
      "epoch": 0.05497428622096116,
      "grad_norm": 7.917320728302002,
      "learning_rate": 9.724686281977149e-05,
      "loss": 1.4825,
      "step": 620
    },
    {
      "epoch": 0.05586096825678312,
      "grad_norm": 7.494490146636963,
      "learning_rate": 9.714098521798468e-05,
      "loss": 1.6621,
      "step": 630
    },
    {
      "epoch": 0.05674765029260507,
      "grad_norm": 7.454612731933594,
      "learning_rate": 9.703316983493417e-05,
      "loss": 1.6406,
      "step": 640
    },
    {
      "epoch": 0.057634332328427025,
      "grad_norm": 8.165127754211426,
      "learning_rate": 9.692342110248805e-05,
      "loss": 1.7403,
      "step": 650
    },
    {
      "epoch": 0.05852101436424898,
      "grad_norm": 7.547806739807129,
      "learning_rate": 9.68117435319869e-05,
      "loss": 1.8081,
      "step": 660
    },
    {
      "epoch": 0.05940769640007094,
      "grad_norm": 6.669929504394531,
      "learning_rate": 9.669814171405819e-05,
      "loss": 1.5977,
      "step": 670
    },
    {
      "epoch": 0.06029437843589289,
      "grad_norm": 10.013628005981445,
      "learning_rate": 9.658262031842773e-05,
      "loss": 1.7354,
      "step": 680
    },
    {
      "epoch": 0.06118106047171484,
      "grad_norm": 7.262237071990967,
      "learning_rate": 9.646518409372763e-05,
      "loss": 1.73,
      "step": 690
    },
    {
      "epoch": 0.0620677425075368,
      "grad_norm": 10.55190658569336,
      "learning_rate": 9.634583786730112e-05,
      "loss": 1.5776,
      "step": 700
    },
    {
      "epoch": 0.06295442454335876,
      "grad_norm": 6.680520534515381,
      "learning_rate": 9.62245865450041e-05,
      "loss": 1.6498,
      "step": 710
    },
    {
      "epoch": 0.06384110657918071,
      "grad_norm": 8.451497077941895,
      "learning_rate": 9.610143511100356e-05,
      "loss": 1.5667,
      "step": 720
    },
    {
      "epoch": 0.06472778861500265,
      "grad_norm": 6.46457576751709,
      "learning_rate": 9.597638862757256e-05,
      "loss": 1.6396,
      "step": 730
    },
    {
      "epoch": 0.06561447065082461,
      "grad_norm": 7.483574867248535,
      "learning_rate": 9.58494522348823e-05,
      "loss": 1.7103,
      "step": 740
    },
    {
      "epoch": 0.06650115268664657,
      "grad_norm": 7.17427921295166,
      "learning_rate": 9.572063115079066e-05,
      "loss": 1.7465,
      "step": 750
    },
    {
      "epoch": 0.06650115268664657,
      "eval_loss": 1.7024177312850952,
      "eval_runtime": 59.3886,
      "eval_samples_per_second": 8.419,
      "eval_steps_per_second": 8.419,
      "step": 750
    },
    {
      "epoch": 0.06738783472246852,
      "grad_norm": 10.202959060668945,
      "learning_rate": 9.558993067062788e-05,
      "loss": 1.5595,
      "step": 760
    },
    {
      "epoch": 0.06827451675829048,
      "grad_norm": 8.259611129760742,
      "learning_rate": 9.545735616697878e-05,
      "loss": 1.6273,
      "step": 770
    },
    {
      "epoch": 0.06916119879411244,
      "grad_norm": 10.105010032653809,
      "learning_rate": 9.532291308946193e-05,
      "loss": 1.6441,
      "step": 780
    },
    {
      "epoch": 0.07004788082993439,
      "grad_norm": 11.567327499389648,
      "learning_rate": 9.518660696450571e-05,
      "loss": 1.7721,
      "step": 790
    },
    {
      "epoch": 0.07093456286575633,
      "grad_norm": 7.6176838874816895,
      "learning_rate": 9.504844339512098e-05,
      "loss": 1.5159,
      "step": 800
    },
    {
      "epoch": 0.07182124490157829,
      "grad_norm": 8.693132400512695,
      "learning_rate": 9.490842806067098e-05,
      "loss": 1.7085,
      "step": 810
    },
    {
      "epoch": 0.07270792693740025,
      "grad_norm": 8.85315227508545,
      "learning_rate": 9.476656671663768e-05,
      "loss": 1.6169,
      "step": 820
    },
    {
      "epoch": 0.0735946089732222,
      "grad_norm": 8.727082252502441,
      "learning_rate": 9.462286519438532e-05,
      "loss": 1.5426,
      "step": 830
    },
    {
      "epoch": 0.07448129100904416,
      "grad_norm": 9.754756927490234,
      "learning_rate": 9.447732940092063e-05,
      "loss": 1.5564,
      "step": 840
    },
    {
      "epoch": 0.07536797304486612,
      "grad_norm": 9.421085357666016,
      "learning_rate": 9.432996531865004e-05,
      "loss": 1.6749,
      "step": 850
    },
    {
      "epoch": 0.07625465508068807,
      "grad_norm": 8.565560340881348,
      "learning_rate": 9.418077900513379e-05,
      "loss": 1.5869,
      "step": 860
    },
    {
      "epoch": 0.07714133711651001,
      "grad_norm": 6.376341342926025,
      "learning_rate": 9.402977659283693e-05,
      "loss": 1.6504,
      "step": 870
    },
    {
      "epoch": 0.07802801915233197,
      "grad_norm": 8.475419044494629,
      "learning_rate": 9.387696428887718e-05,
      "loss": 1.6349,
      "step": 880
    },
    {
      "epoch": 0.07891470118815393,
      "grad_norm": 8.961620330810547,
      "learning_rate": 9.372234837476981e-05,
      "loss": 1.6964,
      "step": 890
    },
    {
      "epoch": 0.07980138322397588,
      "grad_norm": 10.981645584106445,
      "learning_rate": 9.35659352061695e-05,
      "loss": 1.8223,
      "step": 900
    },
    {
      "epoch": 0.07980138322397588,
      "eval_loss": 1.7124860286712646,
      "eval_runtime": 59.2388,
      "eval_samples_per_second": 8.44,
      "eval_steps_per_second": 8.44,
      "step": 900
    },
    {
      "epoch": 0.08068806525979784,
      "grad_norm": 7.960820198059082,
      "learning_rate": 9.340773121260896e-05,
      "loss": 1.7361,
      "step": 910
    },
    {
      "epoch": 0.0815747472956198,
      "grad_norm": 7.7724785804748535,
      "learning_rate": 9.32477428972347e-05,
      "loss": 1.7118,
      "step": 920
    },
    {
      "epoch": 0.08246142933144175,
      "grad_norm": 12.896247863769531,
      "learning_rate": 9.308597683653978e-05,
      "loss": 1.6849,
      "step": 930
    },
    {
      "epoch": 0.0833481113672637,
      "grad_norm": 10.572025299072266,
      "learning_rate": 9.292243968009333e-05,
      "loss": 1.7373,
      "step": 940
    },
    {
      "epoch": 0.08423479340308565,
      "grad_norm": 15.746474266052246,
      "learning_rate": 9.275713815026734e-05,
      "loss": 1.5829,
      "step": 950
    },
    {
      "epoch": 0.0851214754389076,
      "grad_norm": 8.268464088439941,
      "learning_rate": 9.259007904196024e-05,
      "loss": 1.659,
      "step": 960
    },
    {
      "epoch": 0.08600815747472956,
      "grad_norm": 6.80485200881958,
      "learning_rate": 9.242126922231766e-05,
      "loss": 1.5132,
      "step": 970
    },
    {
      "epoch": 0.08689483951055152,
      "grad_norm": 9.059598922729492,
      "learning_rate": 9.225071563045009e-05,
      "loss": 1.5896,
      "step": 980
    },
    {
      "epoch": 0.08778152154637348,
      "grad_norm": 6.6695170402526855,
      "learning_rate": 9.207842527714768e-05,
      "loss": 1.7439,
      "step": 990
    },
    {
      "epoch": 0.08866820358219542,
      "grad_norm": 8.449771881103516,
      "learning_rate": 9.190440524459205e-05,
      "loss": 1.9,
      "step": 1000
    },
    {
      "epoch": 0.08955488561801737,
      "grad_norm": 45.434688568115234,
      "learning_rate": 9.172866268606516e-05,
      "loss": 1.6689,
      "step": 1010
    },
    {
      "epoch": 0.09044156765383933,
      "grad_norm": 7.402377605438232,
      "learning_rate": 9.155120482565522e-05,
      "loss": 1.7466,
      "step": 1020
    },
    {
      "epoch": 0.09132824968966129,
      "grad_norm": 8.143401145935059,
      "learning_rate": 9.137203895795986e-05,
      "loss": 1.5923,
      "step": 1030
    },
    {
      "epoch": 0.09221493172548324,
      "grad_norm": 7.547423839569092,
      "learning_rate": 9.11911724477861e-05,
      "loss": 1.6487,
      "step": 1040
    },
    {
      "epoch": 0.0931016137613052,
      "grad_norm": 6.291411399841309,
      "learning_rate": 9.100861272984782e-05,
      "loss": 1.7453,
      "step": 1050
    },
    {
      "epoch": 0.0931016137613052,
      "eval_loss": 1.6903132200241089,
      "eval_runtime": 59.3722,
      "eval_samples_per_second": 8.421,
      "eval_steps_per_second": 8.421,
      "step": 1050
    },
    {
      "epoch": 0.09398829579712716,
      "grad_norm": 6.51155948638916,
      "learning_rate": 9.082436730845996e-05,
      "loss": 1.5121,
      "step": 1060
    },
    {
      "epoch": 0.0948749778329491,
      "grad_norm": 7.315041542053223,
      "learning_rate": 9.063844375723016e-05,
      "loss": 1.45,
      "step": 1070
    },
    {
      "epoch": 0.09576165986877105,
      "grad_norm": 9.287749290466309,
      "learning_rate": 9.045084971874741e-05,
      "loss": 1.6892,
      "step": 1080
    },
    {
      "epoch": 0.09664834190459301,
      "grad_norm": 7.6157097816467285,
      "learning_rate": 9.026159290426783e-05,
      "loss": 1.832,
      "step": 1090
    },
    {
      "epoch": 0.09753502394041497,
      "grad_norm": 6.081124782562256,
      "learning_rate": 9.007068109339786e-05,
      "loss": 1.6911,
      "step": 1100
    },
    {
      "epoch": 0.09842170597623692,
      "grad_norm": 7.2468671798706055,
      "learning_rate": 8.987812213377425e-05,
      "loss": 1.6959,
      "step": 1110
    },
    {
      "epoch": 0.09930838801205888,
      "grad_norm": 7.454516887664795,
      "learning_rate": 8.968392394074165e-05,
      "loss": 1.5169,
      "step": 1120
    },
    {
      "epoch": 0.10019507004788084,
      "grad_norm": 10.253645896911621,
      "learning_rate": 8.948809449702714e-05,
      "loss": 1.6779,
      "step": 1130
    },
    {
      "epoch": 0.10108175208370278,
      "grad_norm": 8.075345993041992,
      "learning_rate": 8.929064185241216e-05,
      "loss": 1.6622,
      "step": 1140
    },
    {
      "epoch": 0.10196843411952473,
      "grad_norm": 11.007535934448242,
      "learning_rate": 8.909157412340152e-05,
      "loss": 1.7568,
      "step": 1150
    },
    {
      "epoch": 0.10285511615534669,
      "grad_norm": 8.019722938537598,
      "learning_rate": 8.889089949288989e-05,
      "loss": 1.6177,
      "step": 1160
    },
    {
      "epoch": 0.10374179819116865,
      "grad_norm": 8.618474960327148,
      "learning_rate": 8.868862620982537e-05,
      "loss": 1.5605,
      "step": 1170
    },
    {
      "epoch": 0.1046284802269906,
      "grad_norm": 8.008125305175781,
      "learning_rate": 8.848476258887034e-05,
      "loss": 1.5995,
      "step": 1180
    },
    {
      "epoch": 0.10551516226281256,
      "grad_norm": 11.63944149017334,
      "learning_rate": 8.827931701005976e-05,
      "loss": 1.5778,
      "step": 1190
    },
    {
      "epoch": 0.10640184429863452,
      "grad_norm": 9.485556602478027,
      "learning_rate": 8.807229791845674e-05,
      "loss": 1.547,
      "step": 1200
    },
    {
      "epoch": 0.10640184429863452,
      "eval_loss": 1.6847599744796753,
      "eval_runtime": 59.4403,
      "eval_samples_per_second": 8.412,
      "eval_steps_per_second": 8.412,
      "step": 1200
    },
    {
      "epoch": 0.10728852633445646,
      "grad_norm": 8.38836669921875,
      "learning_rate": 8.786371382380529e-05,
      "loss": 1.6206,
      "step": 1210
    },
    {
      "epoch": 0.10817520837027841,
      "grad_norm": 6.360514163970947,
      "learning_rate": 8.765357330018059e-05,
      "loss": 1.5409,
      "step": 1220
    },
    {
      "epoch": 0.10906189040610037,
      "grad_norm": 8.692633628845215,
      "learning_rate": 8.744188498563644e-05,
      "loss": 1.549,
      "step": 1230
    },
    {
      "epoch": 0.10994857244192233,
      "grad_norm": 6.637635707855225,
      "learning_rate": 8.722865758185038e-05,
      "loss": 1.6354,
      "step": 1240
    },
    {
      "epoch": 0.11083525447774428,
      "grad_norm": 9.428290367126465,
      "learning_rate": 8.70138998537658e-05,
      "loss": 1.84,
      "step": 1250
    },
    {
      "epoch": 0.11172193651356624,
      "grad_norm": 7.926419734954834,
      "learning_rate": 8.679762062923178e-05,
      "loss": 1.6915,
      "step": 1260
    },
    {
      "epoch": 0.1126086185493882,
      "grad_norm": 9.914402961730957,
      "learning_rate": 8.65798287986401e-05,
      "loss": 1.6808,
      "step": 1270
    },
    {
      "epoch": 0.11349530058521014,
      "grad_norm": 11.937222480773926,
      "learning_rate": 8.636053331455989e-05,
      "loss": 1.6807,
      "step": 1280
    },
    {
      "epoch": 0.1143819826210321,
      "grad_norm": 9.676164627075195,
      "learning_rate": 8.613974319136961e-05,
      "loss": 1.6505,
      "step": 1290
    },
    {
      "epoch": 0.11526866465685405,
      "grad_norm": 8.247169494628906,
      "learning_rate": 8.59174675048864e-05,
      "loss": 1.6287,
      "step": 1300
    },
    {
      "epoch": 0.116155346692676,
      "grad_norm": 7.142345905303955,
      "learning_rate": 8.569371539199318e-05,
      "loss": 1.6104,
      "step": 1310
    },
    {
      "epoch": 0.11704202872849796,
      "grad_norm": 7.208193778991699,
      "learning_rate": 8.546849605026292e-05,
      "loss": 1.5853,
      "step": 1320
    },
    {
      "epoch": 0.11792871076431992,
      "grad_norm": 7.444331645965576,
      "learning_rate": 8.524181873758061e-05,
      "loss": 1.5583,
      "step": 1330
    },
    {
      "epoch": 0.11881539280014188,
      "grad_norm": 9.676839828491211,
      "learning_rate": 8.501369277176277e-05,
      "loss": 1.7519,
      "step": 1340
    },
    {
      "epoch": 0.11970207483596382,
      "grad_norm": 8.323569297790527,
      "learning_rate": 8.478412753017435e-05,
      "loss": 1.6618,
      "step": 1350
    },
    {
      "epoch": 0.11970207483596382,
      "eval_loss": 1.6926313638687134,
      "eval_runtime": 59.2541,
      "eval_samples_per_second": 8.438,
      "eval_steps_per_second": 8.438,
      "step": 1350
    },
    {
      "epoch": 0.12058875687178577,
      "grad_norm": 9.658554077148438,
      "learning_rate": 8.455313244934327e-05,
      "loss": 1.8027,
      "step": 1360
    },
    {
      "epoch": 0.12147543890760773,
      "grad_norm": 10.467687606811523,
      "learning_rate": 8.432071702457255e-05,
      "loss": 1.6109,
      "step": 1370
    },
    {
      "epoch": 0.12236212094342969,
      "grad_norm": 8.080743789672852,
      "learning_rate": 8.408689080955001e-05,
      "loss": 1.5724,
      "step": 1380
    },
    {
      "epoch": 0.12324880297925164,
      "grad_norm": 7.568329334259033,
      "learning_rate": 8.38516634159555e-05,
      "loss": 1.5831,
      "step": 1390
    },
    {
      "epoch": 0.1241354850150736,
      "grad_norm": 6.801577091217041,
      "learning_rate": 8.361504451306586e-05,
      "loss": 1.6606,
      "step": 1400
    },
    {
      "epoch": 0.12502216705089556,
      "grad_norm": 8.583120346069336,
      "learning_rate": 8.337704382735743e-05,
      "loss": 1.8597,
      "step": 1410
    },
    {
      "epoch": 0.1259088490867175,
      "grad_norm": 9.193146705627441,
      "learning_rate": 8.313767114210618e-05,
      "loss": 1.6648,
      "step": 1420
    },
    {
      "epoch": 0.12679553112253947,
      "grad_norm": 7.931524753570557,
      "learning_rate": 8.289693629698566e-05,
      "loss": 1.6386,
      "step": 1430
    },
    {
      "epoch": 0.12768221315836142,
      "grad_norm": 8.280115127563477,
      "learning_rate": 8.265484918766245e-05,
      "loss": 1.5829,
      "step": 1440
    },
    {
      "epoch": 0.12856889519418335,
      "grad_norm": 7.822704315185547,
      "learning_rate": 8.241141976538945e-05,
      "loss": 1.7449,
      "step": 1450
    },
    {
      "epoch": 0.1294555772300053,
      "grad_norm": 6.792067050933838,
      "learning_rate": 8.216665803659673e-05,
      "loss": 1.8019,
      "step": 1460
    },
    {
      "epoch": 0.13034225926582726,
      "grad_norm": 6.622004508972168,
      "learning_rate": 8.19205740624803e-05,
      "loss": 1.6184,
      "step": 1470
    },
    {
      "epoch": 0.13122894130164922,
      "grad_norm": 8.733943939208984,
      "learning_rate": 8.167317795858853e-05,
      "loss": 1.7708,
      "step": 1480
    },
    {
      "epoch": 0.13211562333747118,
      "grad_norm": 12.745617866516113,
      "learning_rate": 8.142447989440621e-05,
      "loss": 1.4564,
      "step": 1490
    },
    {
      "epoch": 0.13300230537329313,
      "grad_norm": 16.666255950927734,
      "learning_rate": 8.117449009293671e-05,
      "loss": 1.7387,
      "step": 1500
    },
    {
      "epoch": 0.13300230537329313,
      "eval_loss": 1.6664111614227295,
      "eval_runtime": 59.3291,
      "eval_samples_per_second": 8.428,
      "eval_steps_per_second": 8.428,
      "step": 1500
    },
    {
      "epoch": 0.1338889874091151,
      "grad_norm": 9.83285140991211,
      "learning_rate": 8.09232188302816e-05,
      "loss": 1.545,
      "step": 1510
    },
    {
      "epoch": 0.13477566944493705,
      "grad_norm": 12.203434944152832,
      "learning_rate": 8.067067643521836e-05,
      "loss": 1.8037,
      "step": 1520
    },
    {
      "epoch": 0.135662351480759,
      "grad_norm": 6.172882556915283,
      "learning_rate": 8.041687328877568e-05,
      "loss": 1.5307,
      "step": 1530
    },
    {
      "epoch": 0.13654903351658096,
      "grad_norm": 8.186355590820312,
      "learning_rate": 8.016181982380684e-05,
      "loss": 1.4897,
      "step": 1540
    },
    {
      "epoch": 0.13743571555240292,
      "grad_norm": 6.983214378356934,
      "learning_rate": 7.990552652456082e-05,
      "loss": 1.5217,
      "step": 1550
    },
    {
      "epoch": 0.13832239758822487,
      "grad_norm": 7.637870788574219,
      "learning_rate": 7.964800392625131e-05,
      "loss": 1.4999,
      "step": 1560
    },
    {
      "epoch": 0.13920907962404683,
      "grad_norm": 6.830051422119141,
      "learning_rate": 7.938926261462369e-05,
      "loss": 1.6,
      "step": 1570
    },
    {
      "epoch": 0.14009576165986878,
      "grad_norm": 8.118239402770996,
      "learning_rate": 7.912931322551983e-05,
      "loss": 1.5677,
      "step": 1580
    },
    {
      "epoch": 0.1409824436956907,
      "grad_norm": 6.605192184448242,
      "learning_rate": 7.8868166444441e-05,
      "loss": 1.701,
      "step": 1590
    },
    {
      "epoch": 0.14186912573151267,
      "grad_norm": 8.868054389953613,
      "learning_rate": 7.86058330061085e-05,
      "loss": 1.5491,
      "step": 1600
    },
    {
      "epoch": 0.14275580776733462,
      "grad_norm": 9.033772468566895,
      "learning_rate": 7.834232369402252e-05,
      "loss": 1.6596,
      "step": 1610
    },
    {
      "epoch": 0.14364248980315658,
      "grad_norm": 10.658461570739746,
      "learning_rate": 7.807764934001877e-05,
      "loss": 1.4879,
      "step": 1620
    },
    {
      "epoch": 0.14452917183897854,
      "grad_norm": 8.768953323364258,
      "learning_rate": 7.781182082382326e-05,
      "loss": 1.6105,
      "step": 1630
    },
    {
      "epoch": 0.1454158538748005,
      "grad_norm": 6.057528018951416,
      "learning_rate": 7.754484907260515e-05,
      "loss": 1.3959,
      "step": 1640
    },
    {
      "epoch": 0.14630253591062245,
      "grad_norm": 8.636338233947754,
      "learning_rate": 7.727674506052746e-05,
      "loss": 1.4808,
      "step": 1650
    },
    {
      "epoch": 0.14630253591062245,
      "eval_loss": 1.6922165155410767,
      "eval_runtime": 59.3594,
      "eval_samples_per_second": 8.423,
      "eval_steps_per_second": 8.423,
      "step": 1650
    },
    {
      "epoch": 0.1471892179464444,
      "grad_norm": 8.238250732421875,
      "learning_rate": 7.700751980829603e-05,
      "loss": 1.6605,
      "step": 1660
    },
    {
      "epoch": 0.14807589998226636,
      "grad_norm": 7.141147136688232,
      "learning_rate": 7.67371843827065e-05,
      "loss": 1.7022,
      "step": 1670
    },
    {
      "epoch": 0.14896258201808832,
      "grad_norm": 7.410867691040039,
      "learning_rate": 7.64657498961894e-05,
      "loss": 1.676,
      "step": 1680
    },
    {
      "epoch": 0.14984926405391027,
      "grad_norm": 7.417080402374268,
      "learning_rate": 7.61932275063533e-05,
      "loss": 1.6914,
      "step": 1690
    },
    {
      "epoch": 0.15073594608973223,
      "grad_norm": 29.780086517333984,
      "learning_rate": 7.591962841552628e-05,
      "loss": 1.7629,
      "step": 1700
    },
    {
      "epoch": 0.1516226281255542,
      "grad_norm": 7.910691738128662,
      "learning_rate": 7.564496387029534e-05,
      "loss": 1.4673,
      "step": 1710
    },
    {
      "epoch": 0.15250931016137614,
      "grad_norm": 8.581075668334961,
      "learning_rate": 7.536924516104414e-05,
      "loss": 1.588,
      "step": 1720
    },
    {
      "epoch": 0.15339599219719807,
      "grad_norm": 9.336210250854492,
      "learning_rate": 7.50924836214889e-05,
      "loss": 1.7385,
      "step": 1730
    },
    {
      "epoch": 0.15428267423302003,
      "grad_norm": 7.002266883850098,
      "learning_rate": 7.481469062821253e-05,
      "loss": 1.6644,
      "step": 1740
    },
    {
      "epoch": 0.15516935626884198,
      "grad_norm": 13.515497207641602,
      "learning_rate": 7.453587760019692e-05,
      "loss": 1.7763,
      "step": 1750
    },
    {
      "epoch": 0.15605603830466394,
      "grad_norm": 8.133156776428223,
      "learning_rate": 7.425605599835362e-05,
      "loss": 1.5641,
      "step": 1760
    },
    {
      "epoch": 0.1569427203404859,
      "grad_norm": 6.456133842468262,
      "learning_rate": 7.397523732505273e-05,
      "loss": 1.7309,
      "step": 1770
    },
    {
      "epoch": 0.15782940237630785,
      "grad_norm": 14.745148658752441,
      "learning_rate": 7.369343312364996e-05,
      "loss": 1.6363,
      "step": 1780
    },
    {
      "epoch": 0.1587160844121298,
      "grad_norm": 7.444967746734619,
      "learning_rate": 7.341065497801231e-05,
      "loss": 1.581,
      "step": 1790
    },
    {
      "epoch": 0.15960276644795177,
      "grad_norm": 7.5540595054626465,
      "learning_rate": 7.31269145120418e-05,
      "loss": 1.7818,
      "step": 1800
    },
    {
      "epoch": 0.15960276644795177,
      "eval_loss": 1.6818472146987915,
      "eval_runtime": 59.3071,
      "eval_samples_per_second": 8.431,
      "eval_steps_per_second": 8.431,
      "step": 1800
    },
    {
      "epoch": 0.16048944848377372,
      "grad_norm": 9.014618873596191,
      "learning_rate": 7.284222338919761e-05,
      "loss": 1.7272,
      "step": 1810
    },
    {
      "epoch": 0.16137613051959568,
      "grad_norm": 7.285683631896973,
      "learning_rate": 7.255659331201675e-05,
      "loss": 1.7136,
      "step": 1820
    },
    {
      "epoch": 0.16226281255541763,
      "grad_norm": 8.174327850341797,
      "learning_rate": 7.227003602163298e-05,
      "loss": 1.6277,
      "step": 1830
    },
    {
      "epoch": 0.1631494945912396,
      "grad_norm": 8.313567161560059,
      "learning_rate": 7.198256329729413e-05,
      "loss": 1.5496,
      "step": 1840
    },
    {
      "epoch": 0.16403617662706155,
      "grad_norm": 6.585691452026367,
      "learning_rate": 7.169418695587792e-05,
      "loss": 1.4631,
      "step": 1850
    },
    {
      "epoch": 0.1649228586628835,
      "grad_norm": 6.994291305541992,
      "learning_rate": 7.140491885140631e-05,
      "loss": 1.6103,
      "step": 1860
    },
    {
      "epoch": 0.16580954069870543,
      "grad_norm": 10.888928413391113,
      "learning_rate": 7.111477087455802e-05,
      "loss": 1.5637,
      "step": 1870
    },
    {
      "epoch": 0.1666962227345274,
      "grad_norm": 9.979802131652832,
      "learning_rate": 7.082375495217998e-05,
      "loss": 1.7041,
      "step": 1880
    },
    {
      "epoch": 0.16758290477034934,
      "grad_norm": 7.943565368652344,
      "learning_rate": 7.053188304679693e-05,
      "loss": 1.5222,
      "step": 1890
    },
    {
      "epoch": 0.1684695868061713,
      "grad_norm": 7.496694087982178,
      "learning_rate": 7.02391671561197e-05,
      "loss": 1.4436,
      "step": 1900
    },
    {
      "epoch": 0.16935626884199326,
      "grad_norm": 10.703848838806152,
      "learning_rate": 6.994561931255211e-05,
      "loss": 1.6628,
      "step": 1910
    },
    {
      "epoch": 0.1702429508778152,
      "grad_norm": 9.519083976745605,
      "learning_rate": 6.96512515826962e-05,
      "loss": 1.5567,
      "step": 1920
    },
    {
      "epoch": 0.17112963291363717,
      "grad_norm": 9.946447372436523,
      "learning_rate": 6.935607606685643e-05,
      "loss": 1.5579,
      "step": 1930
    },
    {
      "epoch": 0.17201631494945913,
      "grad_norm": 10.239102363586426,
      "learning_rate": 6.90601048985421e-05,
      "loss": 1.8462,
      "step": 1940
    },
    {
      "epoch": 0.17290299698528108,
      "grad_norm": 7.735941410064697,
      "learning_rate": 6.876335024396873e-05,
      "loss": 1.7393,
      "step": 1950
    },
    {
      "epoch": 0.17290299698528108,
      "eval_loss": 1.6459492444992065,
      "eval_runtime": 59.3323,
      "eval_samples_per_second": 8.427,
      "eval_steps_per_second": 8.427,
      "step": 1950
    },
    {
      "epoch": 0.17378967902110304,
      "grad_norm": 7.9266228675842285,
      "learning_rate": 6.846582430155784e-05,
      "loss": 1.6871,
      "step": 1960
    },
    {
      "epoch": 0.174676361056925,
      "grad_norm": 7.106112003326416,
      "learning_rate": 6.81675393014356e-05,
      "loss": 1.4537,
      "step": 1970
    },
    {
      "epoch": 0.17556304309274695,
      "grad_norm": 7.142345428466797,
      "learning_rate": 6.786850750493007e-05,
      "loss": 1.7198,
      "step": 1980
    },
    {
      "epoch": 0.1764497251285689,
      "grad_norm": 9.443406105041504,
      "learning_rate": 6.756874120406716e-05,
      "loss": 1.8101,
      "step": 1990
    },
    {
      "epoch": 0.17733640716439084,
      "grad_norm": 9.909912109375,
      "learning_rate": 6.72682527210654e-05,
      "loss": 1.5669,
      "step": 2000
    },
    {
      "epoch": 0.1782230892002128,
      "grad_norm": 7.394705772399902,
      "learning_rate": 6.696705440782941e-05,
      "loss": 1.7875,
      "step": 2010
    },
    {
      "epoch": 0.17910977123603475,
      "grad_norm": 6.871555805206299,
      "learning_rate": 6.66651586454421e-05,
      "loss": 1.6649,
      "step": 2020
    },
    {
      "epoch": 0.1799964532718567,
      "grad_norm": 8.4517183303833,
      "learning_rate": 6.636257784365587e-05,
      "loss": 1.5413,
      "step": 2030
    },
    {
      "epoch": 0.18088313530767866,
      "grad_norm": 7.050330638885498,
      "learning_rate": 6.60593244403823e-05,
      "loss": 1.591,
      "step": 2040
    },
    {
      "epoch": 0.18176981734350062,
      "grad_norm": 8.818882942199707,
      "learning_rate": 6.575541090118106e-05,
      "loss": 1.5542,
      "step": 2050
    },
    {
      "epoch": 0.18265649937932257,
      "grad_norm": 9.987334251403809,
      "learning_rate": 6.54508497187474e-05,
      "loss": 1.6636,
      "step": 2060
    },
    {
      "epoch": 0.18354318141514453,
      "grad_norm": 9.73635196685791,
      "learning_rate": 6.514565341239862e-05,
      "loss": 1.8135,
      "step": 2070
    },
    {
      "epoch": 0.18442986345096649,
      "grad_norm": 6.916482925415039,
      "learning_rate": 6.483983452755954e-05,
      "loss": 1.7043,
      "step": 2080
    },
    {
      "epoch": 0.18531654548678844,
      "grad_norm": 13.21588134765625,
      "learning_rate": 6.45334056352467e-05,
      "loss": 1.5767,
      "step": 2090
    },
    {
      "epoch": 0.1862032275226104,
      "grad_norm": 7.8419904708862305,
      "learning_rate": 6.422637933155163e-05,
      "loss": 1.5591,
      "step": 2100
    },
    {
      "epoch": 0.1862032275226104,
      "eval_loss": 1.6301392316818237,
      "eval_runtime": 59.4431,
      "eval_samples_per_second": 8.411,
      "eval_steps_per_second": 8.411,
      "step": 2100
    },
    {
      "epoch": 0.18708990955843235,
      "grad_norm": 8.962486267089844,
      "learning_rate": 6.391876823712319e-05,
      "loss": 1.4843,
      "step": 2110
    },
    {
      "epoch": 0.1879765915942543,
      "grad_norm": 10.67493724822998,
      "learning_rate": 6.361058499664857e-05,
      "loss": 1.6638,
      "step": 2120
    },
    {
      "epoch": 0.18886327363007627,
      "grad_norm": 8.06369686126709,
      "learning_rate": 6.330184227833377e-05,
      "loss": 1.6439,
      "step": 2130
    },
    {
      "epoch": 0.1897499556658982,
      "grad_norm": 9.005534172058105,
      "learning_rate": 6.299255277338267e-05,
      "loss": 1.5289,
      "step": 2140
    },
    {
      "epoch": 0.19063663770172015,
      "grad_norm": 9.255204200744629,
      "learning_rate": 6.268272919547539e-05,
      "loss": 1.46,
      "step": 2150
    },
    {
      "epoch": 0.1915233197375421,
      "grad_norm": 7.344980239868164,
      "learning_rate": 6.237238428024573e-05,
      "loss": 1.4932,
      "step": 2160
    },
    {
      "epoch": 0.19241000177336406,
      "grad_norm": 8.692234992980957,
      "learning_rate": 6.206153078475765e-05,
      "loss": 1.6582,
      "step": 2170
    },
    {
      "epoch": 0.19329668380918602,
      "grad_norm": 7.381601333618164,
      "learning_rate": 6.175018148698078e-05,
      "loss": 1.5007,
      "step": 2180
    },
    {
      "epoch": 0.19418336584500798,
      "grad_norm": 7.794239044189453,
      "learning_rate": 6.143834918526529e-05,
      "loss": 1.6501,
      "step": 2190
    },
    {
      "epoch": 0.19507004788082993,
      "grad_norm": 8.13096809387207,
      "learning_rate": 6.112604669781574e-05,
      "loss": 1.6862,
      "step": 2200
    },
    {
      "epoch": 0.1959567299166519,
      "grad_norm": 6.846219539642334,
      "learning_rate": 6.081328686216419e-05,
      "loss": 1.5702,
      "step": 2210
    },
    {
      "epoch": 0.19684341195247385,
      "grad_norm": 8.771533966064453,
      "learning_rate": 6.0500082534642485e-05,
      "loss": 1.6259,
      "step": 2220
    },
    {
      "epoch": 0.1977300939882958,
      "grad_norm": 6.50418758392334,
      "learning_rate": 6.01864465898538e-05,
      "loss": 1.6948,
      "step": 2230
    },
    {
      "epoch": 0.19861677602411776,
      "grad_norm": 8.83719539642334,
      "learning_rate": 5.987239192014337e-05,
      "loss": 1.643,
      "step": 2240
    },
    {
      "epoch": 0.19950345805993971,
      "grad_norm": 7.24541711807251,
      "learning_rate": 5.955793143506864e-05,
      "loss": 1.624,
      "step": 2250
    },
    {
      "epoch": 0.19950345805993971,
      "eval_loss": 1.5997846126556396,
      "eval_runtime": 59.4561,
      "eval_samples_per_second": 8.41,
      "eval_steps_per_second": 8.41,
      "step": 2250
    },
    {
      "epoch": 0.20039014009576167,
      "grad_norm": 13.114813804626465,
      "learning_rate": 5.9243078060868454e-05,
      "loss": 1.5787,
      "step": 2260
    },
    {
      "epoch": 0.20127682213158363,
      "grad_norm": 6.7087321281433105,
      "learning_rate": 5.8927844739931854e-05,
      "loss": 1.3785,
      "step": 2270
    },
    {
      "epoch": 0.20216350416740556,
      "grad_norm": 6.644030570983887,
      "learning_rate": 5.8612244430265966e-05,
      "loss": 1.5126,
      "step": 2280
    },
    {
      "epoch": 0.2030501862032275,
      "grad_norm": 10.291509628295898,
      "learning_rate": 5.829629010496342e-05,
      "loss": 1.4863,
      "step": 2290
    },
    {
      "epoch": 0.20393686823904947,
      "grad_norm": 6.426754951477051,
      "learning_rate": 5.797999475166898e-05,
      "loss": 1.5586,
      "step": 2300
    },
    {
      "epoch": 0.20482355027487142,
      "grad_norm": 9.044095039367676,
      "learning_rate": 5.766337137204581e-05,
      "loss": 1.5063,
      "step": 2310
    },
    {
      "epoch": 0.20571023231069338,
      "grad_norm": 8.852991104125977,
      "learning_rate": 5.734643298124092e-05,
      "loss": 1.7211,
      "step": 2320
    },
    {
      "epoch": 0.20659691434651534,
      "grad_norm": 73.65837860107422,
      "learning_rate": 5.702919260735016e-05,
      "loss": 1.5191,
      "step": 2330
    },
    {
      "epoch": 0.2074835963823373,
      "grad_norm": 8.413342475891113,
      "learning_rate": 5.671166329088279e-05,
      "loss": 1.5013,
      "step": 2340
    },
    {
      "epoch": 0.20837027841815925,
      "grad_norm": 6.938820838928223,
      "learning_rate": 5.639385808422532e-05,
      "loss": 1.5099,
      "step": 2350
    },
    {
      "epoch": 0.2092569604539812,
      "grad_norm": 7.757599353790283,
      "learning_rate": 5.6075790051105044e-05,
      "loss": 1.5848,
      "step": 2360
    },
    {
      "epoch": 0.21014364248980316,
      "grad_norm": 7.502821445465088,
      "learning_rate": 5.5757472266052994e-05,
      "loss": 1.7166,
      "step": 2370
    },
    {
      "epoch": 0.21103032452562512,
      "grad_norm": 11.332352638244629,
      "learning_rate": 5.543891781386657e-05,
      "loss": 1.671,
      "step": 2380
    },
    {
      "epoch": 0.21191700656144707,
      "grad_norm": 7.515905380249023,
      "learning_rate": 5.512013978907158e-05,
      "loss": 1.6298,
      "step": 2390
    },
    {
      "epoch": 0.21280368859726903,
      "grad_norm": 6.094747543334961,
      "learning_rate": 5.4801151295384105e-05,
      "loss": 1.5135,
      "step": 2400
    },
    {
      "epoch": 0.21280368859726903,
      "eval_loss": 1.5888803005218506,
      "eval_runtime": 59.4453,
      "eval_samples_per_second": 8.411,
      "eval_steps_per_second": 8.411,
      "step": 2400
    },
    {
      "epoch": 0.213690370633091,
      "grad_norm": 7.49708366394043,
      "learning_rate": 5.448196544517169e-05,
      "loss": 1.5031,
      "step": 2410
    },
    {
      "epoch": 0.21457705266891292,
      "grad_norm": 8.41457748413086,
      "learning_rate": 5.4162595358914485e-05,
      "loss": 1.5116,
      "step": 2420
    },
    {
      "epoch": 0.21546373470473487,
      "grad_norm": 7.308359146118164,
      "learning_rate": 5.3843054164665855e-05,
      "loss": 1.4185,
      "step": 2430
    },
    {
      "epoch": 0.21635041674055683,
      "grad_norm": 13.086946487426758,
      "learning_rate": 5.352335499751271e-05,
      "loss": 1.6723,
      "step": 2440
    },
    {
      "epoch": 0.21723709877637878,
      "grad_norm": 7.7518157958984375,
      "learning_rate": 5.3203510999035666e-05,
      "loss": 1.4357,
      "step": 2450
    },
    {
      "epoch": 0.21812378081220074,
      "grad_norm": 7.657406806945801,
      "learning_rate": 5.2883535316768745e-05,
      "loss": 1.5464,
      "step": 2460
    },
    {
      "epoch": 0.2190104628480227,
      "grad_norm": 6.197967529296875,
      "learning_rate": 5.2563441103658975e-05,
      "loss": 1.3296,
      "step": 2470
    },
    {
      "epoch": 0.21989714488384465,
      "grad_norm": 10.529012680053711,
      "learning_rate": 5.224324151752577e-05,
      "loss": 1.6508,
      "step": 2480
    },
    {
      "epoch": 0.2207838269196666,
      "grad_norm": 7.516609191894531,
      "learning_rate": 5.1922949720519935e-05,
      "loss": 1.5441,
      "step": 2490
    },
    {
      "epoch": 0.22167050895548857,
      "grad_norm": 5.759303569793701,
      "learning_rate": 5.160257887858279e-05,
      "loss": 1.5121,
      "step": 2500
    },
    {
      "epoch": 0.22255719099131052,
      "grad_norm": 9.412184715270996,
      "learning_rate": 5.1282142160904794e-05,
      "loss": 1.5882,
      "step": 2510
    },
    {
      "epoch": 0.22344387302713248,
      "grad_norm": 6.849535942077637,
      "learning_rate": 5.096165273938437e-05,
      "loss": 1.4593,
      "step": 2520
    },
    {
      "epoch": 0.22433055506295443,
      "grad_norm": 6.8701300621032715,
      "learning_rate": 5.064112378808638e-05,
      "loss": 1.6232,
      "step": 2530
    },
    {
      "epoch": 0.2252172370987764,
      "grad_norm": 10.939626693725586,
      "learning_rate": 5.032056848270057e-05,
      "loss": 1.605,
      "step": 2540
    },
    {
      "epoch": 0.22610391913459835,
      "grad_norm": 9.180508613586426,
      "learning_rate": 5.0000000000000016e-05,
      "loss": 1.572,
      "step": 2550
    },
    {
      "epoch": 0.22610391913459835,
      "eval_loss": 1.5821589231491089,
      "eval_runtime": 59.3605,
      "eval_samples_per_second": 8.423,
      "eval_steps_per_second": 8.423,
      "step": 2550
    },
    {
      "epoch": 0.22699060117042028,
      "grad_norm": 7.859452247619629,
      "learning_rate": 4.967943151729946e-05,
      "loss": 1.4249,
      "step": 2560
    },
    {
      "epoch": 0.22787728320624223,
      "grad_norm": 8.19686222076416,
      "learning_rate": 4.935887621191365e-05,
      "loss": 1.5235,
      "step": 2570
    },
    {
      "epoch": 0.2287639652420642,
      "grad_norm": 6.515259265899658,
      "learning_rate": 4.903834726061566e-05,
      "loss": 1.6435,
      "step": 2580
    },
    {
      "epoch": 0.22965064727788614,
      "grad_norm": 7.381898403167725,
      "learning_rate": 4.8717857839095245e-05,
      "loss": 1.5029,
      "step": 2590
    },
    {
      "epoch": 0.2305373293137081,
      "grad_norm": 7.736898422241211,
      "learning_rate": 4.8397421121417256e-05,
      "loss": 1.4735,
      "step": 2600
    },
    {
      "epoch": 0.23142401134953006,
      "grad_norm": 5.957932949066162,
      "learning_rate": 4.807705027948009e-05,
      "loss": 1.4774,
      "step": 2610
    },
    {
      "epoch": 0.232310693385352,
      "grad_norm": 6.916577339172363,
      "learning_rate": 4.7756758482474285e-05,
      "loss": 1.5927,
      "step": 2620
    },
    {
      "epoch": 0.23319737542117397,
      "grad_norm": 11.942724227905273,
      "learning_rate": 4.7436558896341064e-05,
      "loss": 1.4889,
      "step": 2630
    },
    {
      "epoch": 0.23408405745699593,
      "grad_norm": 6.2363057136535645,
      "learning_rate": 4.71164646832313e-05,
      "loss": 1.5574,
      "step": 2640
    },
    {
      "epoch": 0.23497073949281788,
      "grad_norm": 7.307931423187256,
      "learning_rate": 4.679648900096437e-05,
      "loss": 1.4904,
      "step": 2650
    },
    {
      "epoch": 0.23585742152863984,
      "grad_norm": 6.648471355438232,
      "learning_rate": 4.6476645002487314e-05,
      "loss": 1.5208,
      "step": 2660
    },
    {
      "epoch": 0.2367441035644618,
      "grad_norm": 7.0004167556762695,
      "learning_rate": 4.615694583533419e-05,
      "loss": 1.5019,
      "step": 2670
    },
    {
      "epoch": 0.23763078560028375,
      "grad_norm": 7.872572422027588,
      "learning_rate": 4.583740464108555e-05,
      "loss": 1.5381,
      "step": 2680
    },
    {
      "epoch": 0.23851746763610568,
      "grad_norm": 8.477245330810547,
      "learning_rate": 4.5518034554828346e-05,
      "loss": 1.5339,
      "step": 2690
    },
    {
      "epoch": 0.23940414967192764,
      "grad_norm": 9.522911071777344,
      "learning_rate": 4.519884870461593e-05,
      "loss": 1.5553,
      "step": 2700
    },
    {
      "epoch": 0.23940414967192764,
      "eval_loss": 1.5684587955474854,
      "eval_runtime": 59.3936,
      "eval_samples_per_second": 8.418,
      "eval_steps_per_second": 8.418,
      "step": 2700
    },
    {
      "epoch": 0.2402908317077496,
      "grad_norm": 6.57583475112915,
      "learning_rate": 4.487986021092845e-05,
      "loss": 1.584,
      "step": 2710
    },
    {
      "epoch": 0.24117751374357155,
      "grad_norm": 10.243136405944824,
      "learning_rate": 4.456108218613348e-05,
      "loss": 1.5983,
      "step": 2720
    },
    {
      "epoch": 0.2420641957793935,
      "grad_norm": 7.099559307098389,
      "learning_rate": 4.424252773394705e-05,
      "loss": 1.5817,
      "step": 2730
    },
    {
      "epoch": 0.24295087781521546,
      "grad_norm": 7.89496374130249,
      "learning_rate": 4.3924209948894995e-05,
      "loss": 1.6713,
      "step": 2740
    },
    {
      "epoch": 0.24383755985103742,
      "grad_norm": 7.425642490386963,
      "learning_rate": 4.360614191577471e-05,
      "loss": 1.4357,
      "step": 2750
    },
    {
      "epoch": 0.24472424188685937,
      "grad_norm": 7.661828994750977,
      "learning_rate": 4.3288336709117256e-05,
      "loss": 1.4964,
      "step": 2760
    },
    {
      "epoch": 0.24561092392268133,
      "grad_norm": 9.707741737365723,
      "learning_rate": 4.297080739264988e-05,
      "loss": 1.3392,
      "step": 2770
    },
    {
      "epoch": 0.24649760595850329,
      "grad_norm": 6.905391693115234,
      "learning_rate": 4.2653567018759114e-05,
      "loss": 1.4212,
      "step": 2780
    },
    {
      "epoch": 0.24738428799432524,
      "grad_norm": 7.846536636352539,
      "learning_rate": 4.233662862795421e-05,
      "loss": 1.6355,
      "step": 2790
    },
    {
      "epoch": 0.2482709700301472,
      "grad_norm": 6.390925884246826,
      "learning_rate": 4.202000524833106e-05,
      "loss": 1.4986,
      "step": 2800
    },
    {
      "epoch": 0.24915765206596915,
      "grad_norm": 7.581704616546631,
      "learning_rate": 4.170370989503664e-05,
      "loss": 1.6026,
      "step": 2810
    },
    {
      "epoch": 0.2500443341017911,
      "grad_norm": 4.918117523193359,
      "learning_rate": 4.138775556973407e-05,
      "loss": 1.4725,
      "step": 2820
    },
    {
      "epoch": 0.25093101613761304,
      "grad_norm": 12.692071914672852,
      "learning_rate": 4.1072155260068185e-05,
      "loss": 1.6113,
      "step": 2830
    },
    {
      "epoch": 0.251817698173435,
      "grad_norm": 6.600620746612549,
      "learning_rate": 4.075692193913158e-05,
      "loss": 1.5967,
      "step": 2840
    },
    {
      "epoch": 0.25270438020925695,
      "grad_norm": 6.879825115203857,
      "learning_rate": 4.0442068564931405e-05,
      "loss": 1.4693,
      "step": 2850
    },
    {
      "epoch": 0.25270438020925695,
      "eval_loss": 1.5608752965927124,
      "eval_runtime": 59.3283,
      "eval_samples_per_second": 8.428,
      "eval_steps_per_second": 8.428,
      "step": 2850
    },
    {
      "epoch": 0.25359106224507894,
      "grad_norm": 7.351998329162598,
      "learning_rate": 4.012760807985666e-05,
      "loss": 1.5118,
      "step": 2860
    },
    {
      "epoch": 0.25447774428090086,
      "grad_norm": 9.371225357055664,
      "learning_rate": 3.9813553410146234e-05,
      "loss": 1.5299,
      "step": 2870
    },
    {
      "epoch": 0.25536442631672285,
      "grad_norm": 7.500007152557373,
      "learning_rate": 3.949991746535754e-05,
      "loss": 1.5855,
      "step": 2880
    },
    {
      "epoch": 0.2562511083525448,
      "grad_norm": 7.6006903648376465,
      "learning_rate": 3.918671313783584e-05,
      "loss": 1.5459,
      "step": 2890
    },
    {
      "epoch": 0.2571377903883667,
      "grad_norm": 6.81592321395874,
      "learning_rate": 3.8873953302184295e-05,
      "loss": 1.361,
      "step": 2900
    },
    {
      "epoch": 0.2580244724241887,
      "grad_norm": 6.851174831390381,
      "learning_rate": 3.856165081473475e-05,
      "loss": 1.2751,
      "step": 2910
    },
    {
      "epoch": 0.2589111544600106,
      "grad_norm": 8.746306419372559,
      "learning_rate": 3.824981851301925e-05,
      "loss": 1.3964,
      "step": 2920
    },
    {
      "epoch": 0.2597978364958326,
      "grad_norm": 8.92397689819336,
      "learning_rate": 3.7938469215242386e-05,
      "loss": 1.5833,
      "step": 2930
    },
    {
      "epoch": 0.26068451853165453,
      "grad_norm": 12.532337188720703,
      "learning_rate": 3.762761571975431e-05,
      "loss": 1.754,
      "step": 2940
    },
    {
      "epoch": 0.2615712005674765,
      "grad_norm": 7.304866313934326,
      "learning_rate": 3.731727080452465e-05,
      "loss": 1.5328,
      "step": 2950
    },
    {
      "epoch": 0.26245788260329844,
      "grad_norm": 7.864557266235352,
      "learning_rate": 3.700744722661737e-05,
      "loss": 1.5286,
      "step": 2960
    },
    {
      "epoch": 0.2633445646391204,
      "grad_norm": 6.201906204223633,
      "learning_rate": 3.669815772166626e-05,
      "loss": 1.5775,
      "step": 2970
    },
    {
      "epoch": 0.26423124667494235,
      "grad_norm": 8.181777954101562,
      "learning_rate": 3.6389415003351454e-05,
      "loss": 1.6203,
      "step": 2980
    },
    {
      "epoch": 0.26511792871076434,
      "grad_norm": 8.13985824584961,
      "learning_rate": 3.608123176287686e-05,
      "loss": 1.4212,
      "step": 2990
    },
    {
      "epoch": 0.26600461074658627,
      "grad_norm": 7.873915672302246,
      "learning_rate": 3.577362066844839e-05,
      "loss": 1.4327,
      "step": 3000
    },
    {
      "epoch": 0.26600461074658627,
      "eval_loss": 1.5497733354568481,
      "eval_runtime": 59.3515,
      "eval_samples_per_second": 8.424,
      "eval_steps_per_second": 8.424,
      "step": 3000
    },
    {
      "epoch": 0.26689129278240825,
      "grad_norm": 9.156728744506836,
      "learning_rate": 3.546659436475333e-05,
      "loss": 1.4905,
      "step": 3010
    },
    {
      "epoch": 0.2677779748182302,
      "grad_norm": 9.707823753356934,
      "learning_rate": 3.516016547244048e-05,
      "loss": 1.537,
      "step": 3020
    },
    {
      "epoch": 0.26866465685405216,
      "grad_norm": 7.352388858795166,
      "learning_rate": 3.485434658760141e-05,
      "loss": 1.4339,
      "step": 3030
    },
    {
      "epoch": 0.2695513388898741,
      "grad_norm": 6.827575206756592,
      "learning_rate": 3.454915028125264e-05,
      "loss": 1.4186,
      "step": 3040
    },
    {
      "epoch": 0.270438020925696,
      "grad_norm": 10.17138671875,
      "learning_rate": 3.424458909881898e-05,
      "loss": 1.7247,
      "step": 3050
    },
    {
      "epoch": 0.271324702961518,
      "grad_norm": 6.4703850746154785,
      "learning_rate": 3.394067555961773e-05,
      "loss": 1.5231,
      "step": 3060
    },
    {
      "epoch": 0.27221138499733993,
      "grad_norm": 10.700502395629883,
      "learning_rate": 3.363742215634417e-05,
      "loss": 1.6463,
      "step": 3070
    },
    {
      "epoch": 0.2730980670331619,
      "grad_norm": 7.374842643737793,
      "learning_rate": 3.333484135455793e-05,
      "loss": 1.4507,
      "step": 3080
    },
    {
      "epoch": 0.27398474906898385,
      "grad_norm": 20.006195068359375,
      "learning_rate": 3.303294559217064e-05,
      "loss": 1.5481,
      "step": 3090
    },
    {
      "epoch": 0.27487143110480583,
      "grad_norm": 9.058706283569336,
      "learning_rate": 3.273174727893464e-05,
      "loss": 1.4567,
      "step": 3100
    },
    {
      "epoch": 0.27575811314062776,
      "grad_norm": 6.159706115722656,
      "learning_rate": 3.243125879593287e-05,
      "loss": 1.6317,
      "step": 3110
    },
    {
      "epoch": 0.27664479517644974,
      "grad_norm": 7.272984981536865,
      "learning_rate": 3.213149249506998e-05,
      "loss": 1.4127,
      "step": 3120
    },
    {
      "epoch": 0.27753147721227167,
      "grad_norm": 7.003290176391602,
      "learning_rate": 3.183246069856444e-05,
      "loss": 1.4704,
      "step": 3130
    },
    {
      "epoch": 0.27841815924809366,
      "grad_norm": 9.395560264587402,
      "learning_rate": 3.15341756984422e-05,
      "loss": 1.4897,
      "step": 3140
    },
    {
      "epoch": 0.2793048412839156,
      "grad_norm": 9.366192817687988,
      "learning_rate": 3.123664975603131e-05,
      "loss": 1.4859,
      "step": 3150
    },
    {
      "epoch": 0.2793048412839156,
      "eval_loss": 1.5420976877212524,
      "eval_runtime": 59.3868,
      "eval_samples_per_second": 8.419,
      "eval_steps_per_second": 8.419,
      "step": 3150
    },
    {
      "epoch": 0.28019152331973757,
      "grad_norm": 9.653646469116211,
      "learning_rate": 3.093989510145792e-05,
      "loss": 1.5305,
      "step": 3160
    },
    {
      "epoch": 0.2810782053555595,
      "grad_norm": 7.124954700469971,
      "learning_rate": 3.0643923933143614e-05,
      "loss": 1.5319,
      "step": 3170
    },
    {
      "epoch": 0.2819648873913814,
      "grad_norm": 7.334105968475342,
      "learning_rate": 3.0348748417303834e-05,
      "loss": 1.4719,
      "step": 3180
    },
    {
      "epoch": 0.2828515694272034,
      "grad_norm": 8.093483924865723,
      "learning_rate": 3.005438068744793e-05,
      "loss": 1.6516,
      "step": 3190
    },
    {
      "epoch": 0.28373825146302534,
      "grad_norm": 7.165219306945801,
      "learning_rate": 2.9760832843880317e-05,
      "loss": 1.5383,
      "step": 3200
    },
    {
      "epoch": 0.2846249334988473,
      "grad_norm": 8.381077766418457,
      "learning_rate": 2.9468116953203113e-05,
      "loss": 1.5084,
      "step": 3210
    },
    {
      "epoch": 0.28551161553466925,
      "grad_norm": 8.025052070617676,
      "learning_rate": 2.917624504782007e-05,
      "loss": 1.4437,
      "step": 3220
    },
    {
      "epoch": 0.28639829757049123,
      "grad_norm": 6.2781453132629395,
      "learning_rate": 2.8885229125442027e-05,
      "loss": 1.6153,
      "step": 3230
    },
    {
      "epoch": 0.28728497960631316,
      "grad_norm": 7.848852157592773,
      "learning_rate": 2.8595081148593748e-05,
      "loss": 1.4542,
      "step": 3240
    },
    {
      "epoch": 0.28817166164213515,
      "grad_norm": 9.784353256225586,
      "learning_rate": 2.8305813044122107e-05,
      "loss": 1.4358,
      "step": 3250
    },
    {
      "epoch": 0.2890583436779571,
      "grad_norm": 8.469407081604004,
      "learning_rate": 2.8017436702705908e-05,
      "loss": 1.4224,
      "step": 3260
    },
    {
      "epoch": 0.28994502571377906,
      "grad_norm": 7.443441390991211,
      "learning_rate": 2.7729963978367048e-05,
      "loss": 1.5289,
      "step": 3270
    },
    {
      "epoch": 0.290831707749601,
      "grad_norm": 9.007468223571777,
      "learning_rate": 2.7443406687983272e-05,
      "loss": 1.4351,
      "step": 3280
    },
    {
      "epoch": 0.29171838978542297,
      "grad_norm": 9.351861953735352,
      "learning_rate": 2.715777661080242e-05,
      "loss": 1.5687,
      "step": 3290
    },
    {
      "epoch": 0.2926050718212449,
      "grad_norm": 7.077907562255859,
      "learning_rate": 2.6873085487958257e-05,
      "loss": 1.7491,
      "step": 3300
    },
    {
      "epoch": 0.2926050718212449,
      "eval_loss": 1.5311921834945679,
      "eval_runtime": 59.4179,
      "eval_samples_per_second": 8.415,
      "eval_steps_per_second": 8.415,
      "step": 3300
    },
    {
      "epoch": 0.29349175385706683,
      "grad_norm": 5.730030059814453,
      "learning_rate": 2.6589345021987728e-05,
      "loss": 1.5217,
      "step": 3310
    },
    {
      "epoch": 0.2943784358928888,
      "grad_norm": 6.85732889175415,
      "learning_rate": 2.6306566876350076e-05,
      "loss": 1.4184,
      "step": 3320
    },
    {
      "epoch": 0.29526511792871074,
      "grad_norm": 9.774615287780762,
      "learning_rate": 2.602476267494732e-05,
      "loss": 1.4287,
      "step": 3330
    },
    {
      "epoch": 0.2961517999645327,
      "grad_norm": 6.502627372741699,
      "learning_rate": 2.5743944001646398e-05,
      "loss": 1.6562,
      "step": 3340
    },
    {
      "epoch": 0.29703848200035465,
      "grad_norm": 10.487425804138184,
      "learning_rate": 2.546412239980313e-05,
      "loss": 1.5361,
      "step": 3350
    },
    {
      "epoch": 0.29792516403617664,
      "grad_norm": 6.752458095550537,
      "learning_rate": 2.518530937178752e-05,
      "loss": 1.473,
      "step": 3360
    },
    {
      "epoch": 0.29881184607199857,
      "grad_norm": 9.102508544921875,
      "learning_rate": 2.4907516378511142e-05,
      "loss": 1.626,
      "step": 3370
    },
    {
      "epoch": 0.29969852810782055,
      "grad_norm": 11.789603233337402,
      "learning_rate": 2.4630754838955908e-05,
      "loss": 1.5676,
      "step": 3380
    },
    {
      "epoch": 0.3005852101436425,
      "grad_norm": 4.690525531768799,
      "learning_rate": 2.4355036129704707e-05,
      "loss": 1.3247,
      "step": 3390
    },
    {
      "epoch": 0.30147189217946446,
      "grad_norm": 8.3900785446167,
      "learning_rate": 2.4080371584473755e-05,
      "loss": 1.5134,
      "step": 3400
    },
    {
      "epoch": 0.3023585742152864,
      "grad_norm": 6.29799747467041,
      "learning_rate": 2.380677249364673e-05,
      "loss": 1.4641,
      "step": 3410
    },
    {
      "epoch": 0.3032452562511084,
      "grad_norm": 6.937891960144043,
      "learning_rate": 2.3534250103810636e-05,
      "loss": 1.5212,
      "step": 3420
    },
    {
      "epoch": 0.3041319382869303,
      "grad_norm": 8.222491264343262,
      "learning_rate": 2.326281561729352e-05,
      "loss": 1.452,
      "step": 3430
    },
    {
      "epoch": 0.3050186203227523,
      "grad_norm": 7.146228313446045,
      "learning_rate": 2.299248019170401e-05,
      "loss": 1.4393,
      "step": 3440
    },
    {
      "epoch": 0.3059053023585742,
      "grad_norm": 8.970130920410156,
      "learning_rate": 2.2723254939472577e-05,
      "loss": 1.5834,
      "step": 3450
    },
    {
      "epoch": 0.3059053023585742,
      "eval_loss": 1.5162127017974854,
      "eval_runtime": 59.4395,
      "eval_samples_per_second": 8.412,
      "eval_steps_per_second": 8.412,
      "step": 3450
    },
    {
      "epoch": 0.30679198439439614,
      "grad_norm": 8.676182746887207,
      "learning_rate": 2.2455150927394888e-05,
      "loss": 1.4935,
      "step": 3460
    },
    {
      "epoch": 0.30767866643021813,
      "grad_norm": 7.3444719314575195,
      "learning_rate": 2.2188179176176773e-05,
      "loss": 1.4348,
      "step": 3470
    },
    {
      "epoch": 0.30856534846604006,
      "grad_norm": 8.14013957977295,
      "learning_rate": 2.1922350659981268e-05,
      "loss": 1.486,
      "step": 3480
    },
    {
      "epoch": 0.30945203050186204,
      "grad_norm": 8.495216369628906,
      "learning_rate": 2.1657676305977525e-05,
      "loss": 1.4254,
      "step": 3490
    },
    {
      "epoch": 0.31033871253768397,
      "grad_norm": 6.765787601470947,
      "learning_rate": 2.1394166993891536e-05,
      "loss": 1.4532,
      "step": 3500
    },
    {
      "epoch": 0.31122539457350595,
      "grad_norm": 7.284378528594971,
      "learning_rate": 2.1131833555559044e-05,
      "loss": 1.4648,
      "step": 3510
    },
    {
      "epoch": 0.3121120766093279,
      "grad_norm": 5.678081512451172,
      "learning_rate": 2.0870686774480203e-05,
      "loss": 1.4927,
      "step": 3520
    },
    {
      "epoch": 0.31299875864514987,
      "grad_norm": 7.111896514892578,
      "learning_rate": 2.0610737385376356e-05,
      "loss": 1.3698,
      "step": 3530
    },
    {
      "epoch": 0.3138854406809718,
      "grad_norm": 8.32049560546875,
      "learning_rate": 2.035199607374872e-05,
      "loss": 1.3701,
      "step": 3540
    },
    {
      "epoch": 0.3147721227167938,
      "grad_norm": 5.531152248382568,
      "learning_rate": 2.009447347543921e-05,
      "loss": 1.4623,
      "step": 3550
    },
    {
      "epoch": 0.3156588047526157,
      "grad_norm": 11.240205764770508,
      "learning_rate": 1.983818017619318e-05,
      "loss": 1.4952,
      "step": 3560
    },
    {
      "epoch": 0.3165454867884377,
      "grad_norm": 9.017026901245117,
      "learning_rate": 1.9583126711224347e-05,
      "loss": 1.3575,
      "step": 3570
    },
    {
      "epoch": 0.3174321688242596,
      "grad_norm": 7.459463596343994,
      "learning_rate": 1.9329323564781685e-05,
      "loss": 1.5883,
      "step": 3580
    },
    {
      "epoch": 0.31831885086008155,
      "grad_norm": 8.614239692687988,
      "learning_rate": 1.907678116971843e-05,
      "loss": 1.3795,
      "step": 3590
    },
    {
      "epoch": 0.31920553289590353,
      "grad_norm": 7.115685939788818,
      "learning_rate": 1.882550990706333e-05,
      "loss": 1.4851,
      "step": 3600
    },
    {
      "epoch": 0.31920553289590353,
      "eval_loss": 1.507421612739563,
      "eval_runtime": 59.4282,
      "eval_samples_per_second": 8.414,
      "eval_steps_per_second": 8.414,
      "step": 3600
    },
    {
      "epoch": 0.32009221493172546,
      "grad_norm": 7.0822882652282715,
      "learning_rate": 1.8575520105593824e-05,
      "loss": 1.4626,
      "step": 3610
    },
    {
      "epoch": 0.32097889696754744,
      "grad_norm": 8.020169258117676,
      "learning_rate": 1.8326822041411528e-05,
      "loss": 1.4378,
      "step": 3620
    },
    {
      "epoch": 0.3218655790033694,
      "grad_norm": 7.411886692047119,
      "learning_rate": 1.8079425937519732e-05,
      "loss": 1.4582,
      "step": 3630
    },
    {
      "epoch": 0.32275226103919136,
      "grad_norm": 8.050488471984863,
      "learning_rate": 1.7833341963403314e-05,
      "loss": 1.5404,
      "step": 3640
    },
    {
      "epoch": 0.3236389430750133,
      "grad_norm": 7.69305944442749,
      "learning_rate": 1.7588580234610594e-05,
      "loss": 1.3396,
      "step": 3650
    },
    {
      "epoch": 0.32452562511083527,
      "grad_norm": 10.572087287902832,
      "learning_rate": 1.7345150812337567e-05,
      "loss": 1.5023,
      "step": 3660
    },
    {
      "epoch": 0.3254123071466572,
      "grad_norm": 9.625089645385742,
      "learning_rate": 1.7103063703014376e-05,
      "loss": 1.3099,
      "step": 3670
    },
    {
      "epoch": 0.3262989891824792,
      "grad_norm": 7.519534587860107,
      "learning_rate": 1.686232885789386e-05,
      "loss": 1.4512,
      "step": 3680
    },
    {
      "epoch": 0.3271856712183011,
      "grad_norm": 8.976761817932129,
      "learning_rate": 1.6622956172642604e-05,
      "loss": 1.5594,
      "step": 3690
    },
    {
      "epoch": 0.3280723532541231,
      "grad_norm": 6.362197399139404,
      "learning_rate": 1.638495548693416e-05,
      "loss": 1.3491,
      "step": 3700
    },
    {
      "epoch": 0.328959035289945,
      "grad_norm": 7.0223469734191895,
      "learning_rate": 1.6148336584044543e-05,
      "loss": 1.5594,
      "step": 3710
    },
    {
      "epoch": 0.329845717325767,
      "grad_norm": 7.8357625007629395,
      "learning_rate": 1.5913109190450035e-05,
      "loss": 1.6264,
      "step": 3720
    },
    {
      "epoch": 0.33073239936158894,
      "grad_norm": 7.040388107299805,
      "learning_rate": 1.5679282975427494e-05,
      "loss": 1.4702,
      "step": 3730
    },
    {
      "epoch": 0.33161908139741086,
      "grad_norm": 8.075628280639648,
      "learning_rate": 1.5446867550656772e-05,
      "loss": 1.5274,
      "step": 3740
    },
    {
      "epoch": 0.33250576343323285,
      "grad_norm": 10.086247444152832,
      "learning_rate": 1.5215872469825685e-05,
      "loss": 1.5678,
      "step": 3750
    },
    {
      "epoch": 0.33250576343323285,
      "eval_loss": 1.499360203742981,
      "eval_runtime": 59.3955,
      "eval_samples_per_second": 8.418,
      "eval_steps_per_second": 8.418,
      "step": 3750
    },
    {
      "epoch": 0.3333924454690548,
      "grad_norm": 6.615363597869873,
      "learning_rate": 1.4986307228237271e-05,
      "loss": 1.4275,
      "step": 3760
    },
    {
      "epoch": 0.33427912750487676,
      "grad_norm": 10.221020698547363,
      "learning_rate": 1.4758181262419428e-05,
      "loss": 1.5383,
      "step": 3770
    },
    {
      "epoch": 0.3351658095406987,
      "grad_norm": 7.772680759429932,
      "learning_rate": 1.4531503949737111e-05,
      "loss": 1.4759,
      "step": 3780
    },
    {
      "epoch": 0.3360524915765207,
      "grad_norm": 12.315176963806152,
      "learning_rate": 1.4306284608006841e-05,
      "loss": 1.6371,
      "step": 3790
    },
    {
      "epoch": 0.3369391736123426,
      "grad_norm": 5.77493953704834,
      "learning_rate": 1.408253249511363e-05,
      "loss": 1.5273,
      "step": 3800
    },
    {
      "epoch": 0.3378258556481646,
      "grad_norm": 7.274715423583984,
      "learning_rate": 1.3860256808630431e-05,
      "loss": 1.5488,
      "step": 3810
    },
    {
      "epoch": 0.3387125376839865,
      "grad_norm": 5.848362922668457,
      "learning_rate": 1.3639466685440138e-05,
      "loss": 1.3308,
      "step": 3820
    },
    {
      "epoch": 0.3395992197198085,
      "grad_norm": 7.360718727111816,
      "learning_rate": 1.3420171201359936e-05,
      "loss": 1.3528,
      "step": 3830
    },
    {
      "epoch": 0.3404859017556304,
      "grad_norm": 7.868961334228516,
      "learning_rate": 1.3202379370768256e-05,
      "loss": 1.4614,
      "step": 3840
    },
    {
      "epoch": 0.3413725837914524,
      "grad_norm": 9.383559226989746,
      "learning_rate": 1.2986100146234235e-05,
      "loss": 1.4874,
      "step": 3850
    },
    {
      "epoch": 0.34225926582727434,
      "grad_norm": 11.13818645477295,
      "learning_rate": 1.277134241814966e-05,
      "loss": 1.3849,
      "step": 3860
    },
    {
      "epoch": 0.34314594786309627,
      "grad_norm": 10.342170715332031,
      "learning_rate": 1.2558115014363595e-05,
      "loss": 1.3115,
      "step": 3870
    },
    {
      "epoch": 0.34403262989891825,
      "grad_norm": 6.494439125061035,
      "learning_rate": 1.2346426699819462e-05,
      "loss": 1.4533,
      "step": 3880
    },
    {
      "epoch": 0.3449193119347402,
      "grad_norm": 9.364208221435547,
      "learning_rate": 1.2136286176194748e-05,
      "loss": 1.4229,
      "step": 3890
    },
    {
      "epoch": 0.34580599397056216,
      "grad_norm": 6.627419471740723,
      "learning_rate": 1.1927702081543282e-05,
      "loss": 1.2769,
      "step": 3900
    },
    {
      "epoch": 0.34580599397056216,
      "eval_loss": 1.4924039840698242,
      "eval_runtime": 59.4072,
      "eval_samples_per_second": 8.416,
      "eval_steps_per_second": 8.416,
      "step": 3900
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 150,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3131363193066783e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}