{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9964796996010326,
  "eval_steps": 100,
  "global_step": 1596,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018774935461159353,
      "grad_norm": 0.3424220085144043,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 3.1063,
      "step": 10
    },
    {
      "epoch": 0.03754987092231871,
      "grad_norm": 0.3918570578098297,
      "learning_rate": 4.000000000000001e-06,
      "loss": 3.0935,
      "step": 20
    },
    {
      "epoch": 0.05632480638347806,
      "grad_norm": 0.42540064454078674,
      "learning_rate": 6e-06,
      "loss": 3.0475,
      "step": 30
    },
    {
      "epoch": 0.07509974184463741,
      "grad_norm": 0.47563326358795166,
      "learning_rate": 8.000000000000001e-06,
      "loss": 3.0609,
      "step": 40
    },
    {
      "epoch": 0.09387467730579677,
      "grad_norm": 0.5615227818489075,
      "learning_rate": 1e-05,
      "loss": 3.026,
      "step": 50
    },
    {
      "epoch": 0.11264961276695612,
      "grad_norm": 0.5656551122665405,
      "learning_rate": 1.2e-05,
      "loss": 3.0134,
      "step": 60
    },
    {
      "epoch": 0.13142454822811547,
      "grad_norm": 0.5487608313560486,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 2.9549,
      "step": 70
    },
    {
      "epoch": 0.15019948368927483,
      "grad_norm": 0.48524507880210876,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.9301,
      "step": 80
    },
    {
      "epoch": 0.16897441915043418,
      "grad_norm": 0.5230005383491516,
      "learning_rate": 1.8e-05,
      "loss": 2.9109,
      "step": 90
    },
    {
      "epoch": 0.18774935461159353,
      "grad_norm": 0.38439810276031494,
      "learning_rate": 2e-05,
      "loss": 2.8142,
      "step": 100
    },
    {
      "epoch": 0.18774935461159353,
      "eval_loss": 2.856565237045288,
      "eval_runtime": 115.5533,
      "eval_samples_per_second": 8.195,
      "eval_steps_per_second": 4.102,
      "step": 100
    },
    {
      "epoch": 0.2065242900727529,
      "grad_norm": 0.4284297227859497,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 2.8967,
      "step": 110
    },
    {
      "epoch": 0.22529922553391224,
      "grad_norm": 0.41896769404411316,
      "learning_rate": 2.4e-05,
      "loss": 2.8629,
      "step": 120
    },
    {
      "epoch": 0.2440741609950716,
      "grad_norm": 0.41066381335258484,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 2.8142,
      "step": 130
    },
    {
      "epoch": 0.26284909645623095,
      "grad_norm": 0.44229111075401306,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 2.815,
      "step": 140
    },
    {
      "epoch": 0.28162403191739027,
      "grad_norm": 0.43809670209884644,
      "learning_rate": 3e-05,
      "loss": 2.7881,
      "step": 150
    },
    {
      "epoch": 0.30039896737854965,
      "grad_norm": 0.48944246768951416,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 2.7937,
      "step": 160
    },
    {
      "epoch": 0.319173902839709,
      "grad_norm": 0.5136508941650391,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 2.7524,
      "step": 170
    },
    {
      "epoch": 0.33794883830086836,
      "grad_norm": 0.48745179176330566,
      "learning_rate": 3.6e-05,
      "loss": 2.7675,
      "step": 180
    },
    {
      "epoch": 0.3567237737620277,
      "grad_norm": 0.4948646128177643,
      "learning_rate": 3.8e-05,
      "loss": 2.731,
      "step": 190
    },
    {
      "epoch": 0.37549870922318707,
      "grad_norm": 0.5109882354736328,
      "learning_rate": 4e-05,
      "loss": 2.7235,
      "step": 200
    },
    {
      "epoch": 0.37549870922318707,
      "eval_loss": 2.6776111125946045,
      "eval_runtime": 115.5701,
      "eval_samples_per_second": 8.194,
      "eval_steps_per_second": 4.101,
      "step": 200
    },
    {
      "epoch": 0.3942736446843464,
      "grad_norm": 0.5552621483802795,
      "learning_rate": 4.2e-05,
      "loss": 2.7244,
      "step": 210
    },
    {
      "epoch": 0.4130485801455058,
      "grad_norm": 0.5443555116653442,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 2.6612,
      "step": 220
    },
    {
      "epoch": 0.4318235156066651,
      "grad_norm": 0.6089721918106079,
      "learning_rate": 4.600000000000001e-05,
      "loss": 2.6635,
      "step": 230
    },
    {
      "epoch": 0.4505984510678245,
      "grad_norm": 0.6026090383529663,
      "learning_rate": 4.8e-05,
      "loss": 2.6335,
      "step": 240
    },
    {
      "epoch": 0.4693733865289838,
      "grad_norm": 0.6256532073020935,
      "learning_rate": 5e-05,
      "loss": 2.6262,
      "step": 250
    },
    {
      "epoch": 0.4881483219901432,
      "grad_norm": 0.6782171726226807,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 2.665,
      "step": 260
    },
    {
      "epoch": 0.5069232574513025,
      "grad_norm": 0.7474572658538818,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 2.7222,
      "step": 270
    },
    {
      "epoch": 0.5256981929124619,
      "grad_norm": 0.6566171646118164,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 2.6626,
      "step": 280
    },
    {
      "epoch": 0.5444731283736212,
      "grad_norm": 0.7244133949279785,
      "learning_rate": 5.8e-05,
      "loss": 2.5429,
      "step": 290
    },
    {
      "epoch": 0.5632480638347805,
      "grad_norm": 0.7084689140319824,
      "learning_rate": 6e-05,
      "loss": 2.5809,
      "step": 300
    },
    {
      "epoch": 0.5632480638347805,
      "eval_loss": 2.5663962364196777,
      "eval_runtime": 115.4961,
      "eval_samples_per_second": 8.199,
      "eval_steps_per_second": 4.104,
      "step": 300
    },
    {
      "epoch": 0.5820229992959399,
      "grad_norm": 0.7440884709358215,
      "learning_rate": 6.2e-05,
      "loss": 2.6225,
      "step": 310
    },
    {
      "epoch": 0.6007979347570993,
      "grad_norm": 1.0468900203704834,
      "learning_rate": 6.400000000000001e-05,
      "loss": 2.579,
      "step": 320
    },
    {
      "epoch": 0.6195728702182586,
      "grad_norm": 1.0507537126541138,
      "learning_rate": 6.6e-05,
      "loss": 2.6367,
      "step": 330
    },
    {
      "epoch": 0.638347805679418,
      "grad_norm": 1.0535833835601807,
      "learning_rate": 6.800000000000001e-05,
      "loss": 2.556,
      "step": 340
    },
    {
      "epoch": 0.6571227411405773,
      "grad_norm": 1.5066871643066406,
      "learning_rate": 7e-05,
      "loss": 2.5231,
      "step": 350
    },
    {
      "epoch": 0.6758976766017367,
      "grad_norm": 1.1159995794296265,
      "learning_rate": 7.2e-05,
      "loss": 2.4532,
      "step": 360
    },
    {
      "epoch": 0.694672612062896,
      "grad_norm": 1.4439431428909302,
      "learning_rate": 7.4e-05,
      "loss": 2.4436,
      "step": 370
    },
    {
      "epoch": 0.7134475475240554,
      "grad_norm": 1.1311273574829102,
      "learning_rate": 7.58e-05,
      "loss": 2.4088,
      "step": 380
    },
    {
      "epoch": 0.7322224829852148,
      "grad_norm": 0.9161147475242615,
      "learning_rate": 7.780000000000001e-05,
      "loss": 2.5574,
      "step": 390
    },
    {
      "epoch": 0.7509974184463741,
      "grad_norm": 1.2713381052017212,
      "learning_rate": 7.98e-05,
      "loss": 2.3971,
      "step": 400
    },
    {
      "epoch": 0.7509974184463741,
      "eval_loss": 2.4457809925079346,
      "eval_runtime": 115.5729,
      "eval_samples_per_second": 8.194,
      "eval_steps_per_second": 4.101,
      "step": 400
    },
    {
      "epoch": 0.7697723539075334,
      "grad_norm": 1.4076571464538574,
      "learning_rate": 8.18e-05,
      "loss": 2.5646,
      "step": 410
    },
    {
      "epoch": 0.7885472893686928,
      "grad_norm": 1.0278912782669067,
      "learning_rate": 8.38e-05,
      "loss": 2.5097,
      "step": 420
    },
    {
      "epoch": 0.8073222248298522,
      "grad_norm": 0.9688978791236877,
      "learning_rate": 8.58e-05,
      "loss": 2.4623,
      "step": 430
    },
    {
      "epoch": 0.8260971602910115,
      "grad_norm": 1.3491131067276,
      "learning_rate": 8.78e-05,
      "loss": 2.508,
      "step": 440
    },
    {
      "epoch": 0.8448720957521708,
      "grad_norm": 0.8669747710227966,
      "learning_rate": 8.98e-05,
      "loss": 2.518,
      "step": 450
    },
    {
      "epoch": 0.8636470312133302,
      "grad_norm": 3.474540948867798,
      "learning_rate": 9.180000000000001e-05,
      "loss": 2.4558,
      "step": 460
    },
    {
      "epoch": 0.8824219666744896,
      "grad_norm": 0.9799988865852356,
      "learning_rate": 9.38e-05,
      "loss": 2.3909,
      "step": 470
    },
    {
      "epoch": 0.901196902135649,
      "grad_norm": 1.605360507965088,
      "learning_rate": 9.58e-05,
      "loss": 2.3931,
      "step": 480
    },
    {
      "epoch": 0.9199718375968082,
      "grad_norm": 1.0437480211257935,
      "learning_rate": 9.78e-05,
      "loss": 2.3893,
      "step": 490
    },
    {
      "epoch": 0.9387467730579676,
      "grad_norm": 0.7387896180152893,
      "learning_rate": 9.98e-05,
      "loss": 2.4147,
      "step": 500
    },
    {
      "epoch": 0.9387467730579676,
      "eval_loss": 2.381179094314575,
      "eval_runtime": 115.6216,
      "eval_samples_per_second": 8.191,
      "eval_steps_per_second": 4.1,
      "step": 500
    },
    {
      "epoch": 0.957521708519127,
      "grad_norm": 1.0030367374420166,
      "learning_rate": 9.998336282524578e-05,
      "loss": 2.4192,
      "step": 510
    },
    {
      "epoch": 0.9762966439802864,
      "grad_norm": 2.564517021179199,
      "learning_rate": 9.992586581827853e-05,
      "loss": 2.4279,
      "step": 520
    },
    {
      "epoch": 0.9950715794414456,
      "grad_norm": 0.8353458642959595,
      "learning_rate": 9.983904607709364e-05,
      "loss": 2.4838,
      "step": 530
    },
    {
      "epoch": 1.013846514902605,
      "grad_norm": 0.7976542115211487,
      "learning_rate": 9.970368310819e-05,
      "loss": 2.4366,
      "step": 540
    },
    {
      "epoch": 1.0326214503637643,
      "grad_norm": 1.2872307300567627,
      "learning_rate": 9.952748467735829e-05,
      "loss": 2.3284,
      "step": 550
    },
    {
      "epoch": 1.0513963858249238,
      "grad_norm": 1.1133558750152588,
      "learning_rate": 9.931059554538611e-05,
      "loss": 2.3192,
      "step": 560
    },
    {
      "epoch": 1.070171321286083,
      "grad_norm": 1.0255626440048218,
      "learning_rate": 9.905319390365365e-05,
      "loss": 2.3409,
      "step": 570
    },
    {
      "epoch": 1.0889462567472425,
      "grad_norm": 1.0734648704528809,
      "learning_rate": 9.875549122773535e-05,
      "loss": 2.4633,
      "step": 580
    },
    {
      "epoch": 1.1077211922084018,
      "grad_norm": 0.9561296105384827,
      "learning_rate": 9.841773210365646e-05,
      "loss": 2.3144,
      "step": 590
    },
    {
      "epoch": 1.126496127669561,
      "grad_norm": 0.8402953743934631,
      "learning_rate": 9.804019402694626e-05,
      "loss": 2.3987,
      "step": 600
    },
    {
      "epoch": 1.126496127669561,
      "eval_loss": 2.343639850616455,
      "eval_runtime": 115.6739,
      "eval_samples_per_second": 8.187,
      "eval_steps_per_second": 4.098,
      "step": 600
    },
    {
      "epoch": 1.1452710631307206,
      "grad_norm": 0.9632971882820129,
      "learning_rate": 9.762318717465388e-05,
      "loss": 2.4487,
      "step": 610
    },
    {
      "epoch": 1.1640459985918798,
      "grad_norm": 0.996286153793335,
      "learning_rate": 9.716705415051361e-05,
      "loss": 2.5009,
      "step": 620
    },
    {
      "epoch": 1.1828209340530391,
      "grad_norm": 0.9334232211112976,
      "learning_rate": 9.667216970346914e-05,
      "loss": 2.2991,
      "step": 630
    },
    {
      "epoch": 1.2015958695141986,
      "grad_norm": 1.014267086982727,
      "learning_rate": 9.613894041978795e-05,
      "loss": 2.4963,
      "step": 640
    },
    {
      "epoch": 1.2203708049753579,
      "grad_norm": 1.3219252824783325,
      "learning_rate": 9.556780438901898e-05,
      "loss": 2.4339,
      "step": 650
    },
    {
      "epoch": 1.2391457404365172,
      "grad_norm": 0.7801995873451233,
      "learning_rate": 9.495923084406772e-05,
      "loss": 2.3548,
      "step": 660
    },
    {
      "epoch": 1.2579206758976766,
      "grad_norm": 1.7439610958099365,
      "learning_rate": 9.431371977568483e-05,
      "loss": 2.4373,
      "step": 670
    },
    {
      "epoch": 1.276695611358836,
      "grad_norm": 1.080696702003479,
      "learning_rate": 9.363180152168447e-05,
      "loss": 2.1908,
      "step": 680
    },
    {
      "epoch": 1.2954705468199954,
      "grad_norm": 1.151071548461914,
      "learning_rate": 9.291403633123046e-05,
      "loss": 2.5268,
      "step": 690
    },
    {
      "epoch": 1.3142454822811547,
      "grad_norm": 0.8707749843597412,
      "learning_rate": 9.21610139045477e-05,
      "loss": 2.3,
      "step": 700
    },
    {
      "epoch": 1.3142454822811547,
      "eval_loss": 2.3193163871765137,
      "eval_runtime": 115.797,
      "eval_samples_per_second": 8.178,
      "eval_steps_per_second": 4.093,
      "step": 700
    },
    {
      "epoch": 1.333020417742314,
      "grad_norm": 1.0154194831848145,
      "learning_rate": 9.13733529084374e-05,
      "loss": 2.3169,
      "step": 710
    },
    {
      "epoch": 1.3517953532034734,
      "grad_norm": 1.2447744607925415,
      "learning_rate": 9.055170046799385e-05,
      "loss": 2.2876,
      "step": 720
    },
    {
      "epoch": 1.3705702886646327,
      "grad_norm": 1.545040249824524,
      "learning_rate": 8.969673163494064e-05,
      "loss": 2.3832,
      "step": 730
    },
    {
      "epoch": 1.3893452241257922,
      "grad_norm": 0.9809775352478027,
      "learning_rate": 8.880914883302278e-05,
      "loss": 2.4271,
      "step": 740
    },
    {
      "epoch": 1.4081201595869515,
      "grad_norm": 0.8140939474105835,
      "learning_rate": 8.788968128091083e-05,
      "loss": 2.426,
      "step": 750
    },
    {
      "epoch": 1.4268950950481107,
      "grad_norm": 0.9135342240333557,
      "learning_rate": 8.69390843930906e-05,
      "loss": 2.3396,
      "step": 760
    },
    {
      "epoch": 1.44567003050927,
      "grad_norm": 0.7756408452987671,
      "learning_rate": 8.595813915923113e-05,
      "loss": 2.3238,
      "step": 770
    },
    {
      "epoch": 1.4644449659704295,
      "grad_norm": 0.8543408513069153,
      "learning_rate": 8.494765150254062e-05,
      "loss": 2.509,
      "step": 780
    },
    {
      "epoch": 1.4832199014315888,
      "grad_norm": 0.8993504643440247,
      "learning_rate": 8.390845161763756e-05,
      "loss": 2.1994,
      "step": 790
    },
    {
      "epoch": 1.5019948368927483,
      "grad_norm": 1.2262345552444458,
      "learning_rate": 8.284139328848082e-05,
      "loss": 2.3219,
      "step": 800
    },
    {
      "epoch": 1.5019948368927483,
      "eval_loss": 2.295060873031616,
      "eval_runtime": 115.6726,
      "eval_samples_per_second": 8.187,
      "eval_steps_per_second": 4.098,
      "step": 800
    },
    {
      "epoch": 1.5207697723539075,
      "grad_norm": 0.8056873679161072,
      "learning_rate": 8.174735318691945e-05,
      "loss": 2.4357,
      "step": 810
    },
    {
      "epoch": 1.5395447078150668,
      "grad_norm": 0.9234392642974854,
      "learning_rate": 8.062723015243822e-05,
      "loss": 2.172,
      "step": 820
    },
    {
      "epoch": 1.558319643276226,
      "grad_norm": 1.4100919961929321,
      "learning_rate": 7.948194445369064e-05,
      "loss": 2.4166,
      "step": 830
    },
    {
      "epoch": 1.5770945787373856,
      "grad_norm": 0.9078715443611145,
      "learning_rate": 7.831243703242637e-05,
      "loss": 2.3765,
      "step": 840
    },
    {
      "epoch": 1.595869514198545,
      "grad_norm": 1.0631110668182373,
      "learning_rate": 7.711966873043398e-05,
      "loss": 2.3603,
      "step": 850
    },
    {
      "epoch": 1.6146444496597043,
      "grad_norm": 0.8904982805252075,
      "learning_rate": 7.590461950013424e-05,
      "loss": 2.2791,
      "step": 860
    },
    {
      "epoch": 1.6334193851208636,
      "grad_norm": 1.1240772008895874,
      "learning_rate": 7.46682875994727e-05,
      "loss": 2.4014,
      "step": 870
    },
    {
      "epoch": 1.6521943205820229,
      "grad_norm": 0.9138346314430237,
      "learning_rate": 7.341168877177267e-05,
      "loss": 2.3897,
      "step": 880
    },
    {
      "epoch": 1.6709692560431824,
      "grad_norm": 0.8400964140892029,
      "learning_rate": 7.213585541122261e-05,
      "loss": 2.3004,
      "step": 890
    },
    {
      "epoch": 1.6897441915043419,
      "grad_norm": 1.1160262823104858,
      "learning_rate": 7.084183571468368e-05,
      "loss": 2.377,
      "step": 900
    },
    {
      "epoch": 1.6897441915043419,
      "eval_loss": 2.2762601375579834,
      "eval_runtime": 115.7128,
      "eval_samples_per_second": 8.184,
      "eval_steps_per_second": 4.096,
      "step": 900
    },
    {
      "epoch": 1.7085191269655011,
      "grad_norm": 0.8896321654319763,
      "learning_rate": 6.953069282051396e-05,
      "loss": 2.1619,
      "step": 910
    },
    {
      "epoch": 1.7272940624266604,
      "grad_norm": 0.848534107208252,
      "learning_rate": 6.820350393511732e-05,
      "loss": 2.447,
      "step": 920
    },
    {
      "epoch": 1.7460689978878197,
      "grad_norm": 0.9254674911499023,
      "learning_rate": 6.686135944793395e-05,
      "loss": 2.2199,
      "step": 930
    },
    {
      "epoch": 1.7648439333489792,
      "grad_norm": 0.8377211093902588,
      "learning_rate": 6.550536203560029e-05,
      "loss": 2.3059,
      "step": 940
    },
    {
      "epoch": 1.7836188688101384,
      "grad_norm": 1.01736319065094,
      "learning_rate": 6.413662575601391e-05,
      "loss": 2.1205,
      "step": 950
    },
    {
      "epoch": 1.802393804271298,
      "grad_norm": 0.7986969947814941,
      "learning_rate": 6.27562751330479e-05,
      "loss": 2.2619,
      "step": 960
    },
    {
      "epoch": 1.8211687397324572,
      "grad_norm": 1.054882287979126,
      "learning_rate": 6.136544423266651e-05,
      "loss": 2.1804,
      "step": 970
    },
    {
      "epoch": 1.8399436751936165,
      "grad_norm": 0.8551494479179382,
      "learning_rate": 5.9965275731201366e-05,
      "loss": 2.3119,
      "step": 980
    },
    {
      "epoch": 1.8587186106547757,
      "grad_norm": 1.810271143913269,
      "learning_rate": 5.8556919976553406e-05,
      "loss": 2.4647,
      "step": 990
    },
    {
      "epoch": 1.8774935461159352,
      "grad_norm": 0.7847923636436462,
      "learning_rate": 5.714153404309228e-05,
      "loss": 2.2977,
      "step": 1000
    },
    {
      "epoch": 1.8774935461159352,
      "eval_loss": 2.262303590774536,
      "eval_runtime": 115.6317,
      "eval_samples_per_second": 8.19,
      "eval_steps_per_second": 4.099,
      "step": 1000
    },
    {
      "epoch": 1.8962684815770947,
      "grad_norm": 0.8150405883789062,
      "learning_rate": 5.5720280781029166e-05,
      "loss": 2.2873,
      "step": 1010
    },
    {
      "epoch": 1.915043417038254,
      "grad_norm": 0.7990459203720093,
      "learning_rate": 5.429432786104446e-05,
      "loss": 2.3096,
      "step": 1020
    },
    {
      "epoch": 1.9338183524994133,
      "grad_norm": 0.9583683013916016,
      "learning_rate": 5.286484681495499e-05,
      "loss": 2.3133,
      "step": 1030
    },
    {
      "epoch": 1.9525932879605725,
      "grad_norm": 0.8717370629310608,
      "learning_rate": 5.1433012073209085e-05,
      "loss": 2.3946,
      "step": 1040
    },
    {
      "epoch": 1.971368223421732,
      "grad_norm": 0.8073247671127319,
      "learning_rate": 5e-05,
      "loss": 2.1941,
      "step": 1050
    },
    {
      "epoch": 1.9901431588828915,
      "grad_norm": 1.0495169162750244,
      "learning_rate": 4.856698792679094e-05,
      "loss": 2.2465,
      "step": 1060
    },
    {
      "epoch": 2.0089180943440508,
      "grad_norm": 0.8290483355522156,
      "learning_rate": 4.7135153185045014e-05,
      "loss": 2.2935,
      "step": 1070
    },
    {
      "epoch": 2.02769302980521,
      "grad_norm": 0.8408583998680115,
      "learning_rate": 4.570567213895555e-05,
      "loss": 2.2642,
      "step": 1080
    },
    {
      "epoch": 2.0464679652663693,
      "grad_norm": 1.1089783906936646,
      "learning_rate": 4.427971921897085e-05,
      "loss": 2.3274,
      "step": 1090
    },
    {
      "epoch": 2.0652429007275286,
      "grad_norm": 1.12042236328125,
      "learning_rate": 4.2858465956907724e-05,
      "loss": 2.269,
      "step": 1100
    },
    {
      "epoch": 2.0652429007275286,
      "eval_loss": 2.252450704574585,
      "eval_runtime": 115.7991,
      "eval_samples_per_second": 8.178,
      "eval_steps_per_second": 4.093,
      "step": 1100
    },
    {
      "epoch": 2.0840178361886883,
      "grad_norm": 0.9814222455024719,
      "learning_rate": 4.14430800234466e-05,
      "loss": 2.1596,
      "step": 1110
    },
    {
      "epoch": 2.1027927716498476,
      "grad_norm": 0.8501948118209839,
      "learning_rate": 4.003472426879866e-05,
      "loss": 2.3033,
      "step": 1120
    },
    {
      "epoch": 2.121567707111007,
      "grad_norm": 0.8352840542793274,
      "learning_rate": 3.863455576733349e-05,
      "loss": 2.2893,
      "step": 1130
    },
    {
      "epoch": 2.140342642572166,
      "grad_norm": 0.8014605045318604,
      "learning_rate": 3.724372486695211e-05,
      "loss": 2.2274,
      "step": 1140
    },
    {
      "epoch": 2.1591175780333254,
      "grad_norm": 0.9311611652374268,
      "learning_rate": 3.5863374243986094e-05,
      "loss": 2.0715,
      "step": 1150
    },
    {
      "epoch": 2.177892513494485,
      "grad_norm": 0.8880860209465027,
      "learning_rate": 3.4494637964399724e-05,
      "loss": 2.2517,
      "step": 1160
    },
    {
      "epoch": 2.1966674489556444,
      "grad_norm": 0.9553248882293701,
      "learning_rate": 3.313864055206607e-05,
      "loss": 2.3358,
      "step": 1170
    },
    {
      "epoch": 2.2154423844168036,
      "grad_norm": 1.015997052192688,
      "learning_rate": 3.179649606488267e-05,
      "loss": 2.2711,
      "step": 1180
    },
    {
      "epoch": 2.234217319877963,
      "grad_norm": 0.8522908091545105,
      "learning_rate": 3.046930717948604e-05,
      "loss": 2.2483,
      "step": 1190
    },
    {
      "epoch": 2.252992255339122,
      "grad_norm": 0.844487190246582,
      "learning_rate": 2.9158164285316353e-05,
      "loss": 2.2305,
      "step": 1200
    },
    {
      "epoch": 2.252992255339122,
      "eval_loss": 2.2441632747650146,
      "eval_runtime": 115.7498,
      "eval_samples_per_second": 8.181,
      "eval_steps_per_second": 4.095,
      "step": 1200
    },
    {
      "epoch": 2.2717671908002814,
      "grad_norm": 0.87571120262146,
      "learning_rate": 2.7864144588777403e-05,
      "loss": 2.2648,
      "step": 1210
    },
    {
      "epoch": 2.290542126261441,
      "grad_norm": 0.8913039565086365,
      "learning_rate": 2.6588311228227347e-05,
      "loss": 2.1812,
      "step": 1220
    },
    {
      "epoch": 2.3093170617226004,
      "grad_norm": 1.0479601621627808,
      "learning_rate": 2.5331712400527298e-05,
      "loss": 2.2898,
      "step": 1230
    },
    {
      "epoch": 2.3280919971837597,
      "grad_norm": 0.9554978013038635,
      "learning_rate": 2.4095380499865762e-05,
      "loss": 2.3384,
      "step": 1240
    },
    {
      "epoch": 2.346866932644919,
      "grad_norm": 0.8619790077209473,
      "learning_rate": 2.288033126956604e-05,
      "loss": 2.2454,
      "step": 1250
    },
    {
      "epoch": 2.3656418681060782,
      "grad_norm": 0.980107843875885,
      "learning_rate": 2.1687562967573645e-05,
      "loss": 2.193,
      "step": 1260
    },
    {
      "epoch": 2.3844168035672375,
      "grad_norm": 0.935689389705658,
      "learning_rate": 2.0518055546309362e-05,
      "loss": 2.3337,
      "step": 1270
    },
    {
      "epoch": 2.4031917390283972,
      "grad_norm": 0.8614078164100647,
      "learning_rate": 1.9372769847561788e-05,
      "loss": 2.1658,
      "step": 1280
    },
    {
      "epoch": 2.4219666744895565,
      "grad_norm": 0.9031429886817932,
      "learning_rate": 1.8252646813080565e-05,
      "loss": 2.1985,
      "step": 1290
    },
    {
      "epoch": 2.4407416099507158,
      "grad_norm": 0.9556667804718018,
      "learning_rate": 1.7158606711519194e-05,
      "loss": 2.3866,
      "step": 1300
    },
    {
      "epoch": 2.4407416099507158,
      "eval_loss": 2.2396435737609863,
      "eval_runtime": 115.863,
      "eval_samples_per_second": 8.173,
      "eval_steps_per_second": 4.091,
      "step": 1300
    },
    {
      "epoch": 2.459516545411875,
      "grad_norm": 0.9352513551712036,
      "learning_rate": 1.6091548382362458e-05,
      "loss": 2.2399,
      "step": 1310
    },
    {
      "epoch": 2.4782914808730343,
      "grad_norm": 1.494532585144043,
      "learning_rate": 1.50523484974594e-05,
      "loss": 2.3273,
      "step": 1320
    },
    {
      "epoch": 2.497066416334194,
      "grad_norm": 0.9380200505256653,
      "learning_rate": 1.40418608407689e-05,
      "loss": 2.2071,
      "step": 1330
    },
    {
      "epoch": 2.5158413517953533,
      "grad_norm": 0.8295271396636963,
      "learning_rate": 1.3060915606909413e-05,
      "loss": 2.274,
      "step": 1340
    },
    {
      "epoch": 2.5346162872565126,
      "grad_norm": 0.8793519735336304,
      "learning_rate": 1.2110318719089158e-05,
      "loss": 2.3791,
      "step": 1350
    },
    {
      "epoch": 2.553391222717672,
      "grad_norm": 0.9659551382064819,
      "learning_rate": 1.1190851166977217e-05,
      "loss": 2.2612,
      "step": 1360
    },
    {
      "epoch": 2.572166158178831,
      "grad_norm": 0.9426782727241516,
      "learning_rate": 1.0303268365059382e-05,
      "loss": 2.2907,
      "step": 1370
    },
    {
      "epoch": 2.590941093639991,
      "grad_norm": 1.0491048097610474,
      "learning_rate": 9.448299532006149e-06,
      "loss": 2.3124,
      "step": 1380
    },
    {
      "epoch": 2.60971602910115,
      "grad_norm": 0.9064353704452515,
      "learning_rate": 8.626647091562612e-06,
      "loss": 2.2214,
      "step": 1390
    },
    {
      "epoch": 2.6284909645623094,
      "grad_norm": 0.7819197773933411,
      "learning_rate": 7.838986095452311e-06,
      "loss": 2.3217,
      "step": 1400
    },
    {
      "epoch": 2.6284909645623094,
      "eval_loss": 2.2368621826171875,
      "eval_runtime": 115.8127,
      "eval_samples_per_second": 8.177,
      "eval_steps_per_second": 4.093,
      "step": 1400
    },
    {
      "epoch": 2.6472659000234686,
      "grad_norm": 0.8173615336418152,
      "learning_rate": 7.085963668769552e-06,
      "loss": 2.2193,
      "step": 1410
    },
    {
      "epoch": 2.666040835484628,
      "grad_norm": 0.7969537377357483,
      "learning_rate": 6.36819847831554e-06,
      "loss": 2.3388,
      "step": 1420
    },
    {
      "epoch": 2.6848157709457876,
      "grad_norm": 0.8578535318374634,
      "learning_rate": 5.686280224315188e-06,
      "loss": 2.3577,
      "step": 1430
    },
    {
      "epoch": 2.703590706406947,
      "grad_norm": 0.8416107296943665,
      "learning_rate": 5.040769155932284e-06,
      "loss": 2.2829,
      "step": 1440
    },
    {
      "epoch": 2.722365641868106,
      "grad_norm": 0.8079660534858704,
      "learning_rate": 4.432195610981032e-06,
      "loss": 2.2675,
      "step": 1450
    },
    {
      "epoch": 2.7411405773292654,
      "grad_norm": 0.8975071907043457,
      "learning_rate": 3.861059580212056e-06,
      "loss": 2.3367,
      "step": 1460
    },
    {
      "epoch": 2.7599155127904247,
      "grad_norm": 1.0292245149612427,
      "learning_rate": 3.3278302965308596e-06,
      "loss": 2.3209,
      "step": 1470
    },
    {
      "epoch": 2.7786904482515844,
      "grad_norm": 0.7877647280693054,
      "learning_rate": 2.8329458494863847e-06,
      "loss": 2.2889,
      "step": 1480
    },
    {
      "epoch": 2.7974653837127432,
      "grad_norm": 0.8482506275177002,
      "learning_rate": 2.3768128253461253e-06,
      "loss": 2.3301,
      "step": 1490
    },
    {
      "epoch": 2.816240319173903,
      "grad_norm": 0.7603471279144287,
      "learning_rate": 1.9598059730537466e-06,
      "loss": 2.2007,
      "step": 1500
    },
    {
      "epoch": 2.816240319173903,
      "eval_loss": 2.2355196475982666,
      "eval_runtime": 115.7546,
      "eval_samples_per_second": 8.181,
      "eval_steps_per_second": 4.095,
      "step": 1500
    },
    {
      "epoch": 2.835015254635062,
      "grad_norm": 0.8267006874084473,
      "learning_rate": 1.5822678963435478e-06,
      "loss": 2.392,
      "step": 1510
    },
    {
      "epoch": 2.8537901900962215,
      "grad_norm": 0.8282234072685242,
      "learning_rate": 1.2445087722646575e-06,
      "loss": 2.2968,
      "step": 1520
    },
    {
      "epoch": 2.8725651255573807,
      "grad_norm": 0.8622906804084778,
      "learning_rate": 9.468060963463755e-07,
      "loss": 2.2902,
      "step": 1530
    },
    {
      "epoch": 2.89134006101854,
      "grad_norm": 0.8282898664474487,
      "learning_rate": 6.894044546138845e-07,
      "loss": 2.2151,
      "step": 1540
    },
    {
      "epoch": 2.9101149964796997,
      "grad_norm": 0.8747497797012329,
      "learning_rate": 4.7251532264170895e-07,
      "loss": 2.188,
      "step": 1550
    },
    {
      "epoch": 2.928889931940859,
      "grad_norm": 0.96438068151474,
      "learning_rate": 2.9631689180999457e-07,
      "loss": 2.4261,
      "step": 1560
    },
    {
      "epoch": 2.9476648674020183,
      "grad_norm": 0.8510390520095825,
      "learning_rate": 1.6095392290635393e-07,
      "loss": 2.272,
      "step": 1570
    },
    {
      "epoch": 2.9664398028631775,
      "grad_norm": 0.7817071080207825,
      "learning_rate": 6.653762719355805e-08,
      "loss": 2.3397,
      "step": 1580
    },
    {
      "epoch": 2.985214738324337,
      "grad_norm": 0.8862946033477783,
      "learning_rate": 1.3145575040801606e-08,
      "loss": 2.269,
      "step": 1590
    },
    {
      "epoch": 2.9964796996010326,
      "step": 1596,
      "total_flos": 1.1780062420402176e+18,
      "train_loss": 2.428264606566656,
      "train_runtime": 11538.2228,
      "train_samples_per_second": 2.216,
      "train_steps_per_second": 0.138
    }
  ],
  "logging_steps": 10,
  "max_steps": 1596,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1.1780062420402176e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}