{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.954337899543379,
  "eval_steps": 500,
  "global_step": 1090,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0091324200913242,
      "grad_norm": 382.0,
      "learning_rate": 1.8348623853211011e-06,
      "loss": 46.9033,
      "step": 1
    },
    {
      "epoch": 0.045662100456621,
      "grad_norm": 318.0,
      "learning_rate": 9.174311926605506e-06,
      "loss": 46.3618,
      "step": 5
    },
    {
      "epoch": 0.091324200913242,
      "grad_norm": 139.0,
      "learning_rate": 1.834862385321101e-05,
      "loss": 39.3883,
      "step": 10
    },
    {
      "epoch": 0.136986301369863,
      "grad_norm": 52.25,
      "learning_rate": 2.7522935779816515e-05,
      "loss": 29.4216,
      "step": 15
    },
    {
      "epoch": 0.182648401826484,
      "grad_norm": 20.25,
      "learning_rate": 3.669724770642202e-05,
      "loss": 24.7169,
      "step": 20
    },
    {
      "epoch": 0.228310502283105,
      "grad_norm": 10.0,
      "learning_rate": 4.587155963302753e-05,
      "loss": 21.4244,
      "step": 25
    },
    {
      "epoch": 0.273972602739726,
      "grad_norm": 5.0,
      "learning_rate": 5.504587155963303e-05,
      "loss": 19.7804,
      "step": 30
    },
    {
      "epoch": 0.319634703196347,
      "grad_norm": 4.6875,
      "learning_rate": 6.422018348623854e-05,
      "loss": 19.1075,
      "step": 35
    },
    {
      "epoch": 0.365296803652968,
      "grad_norm": 8.5,
      "learning_rate": 7.339449541284404e-05,
      "loss": 18.037,
      "step": 40
    },
    {
      "epoch": 0.410958904109589,
      "grad_norm": 18.75,
      "learning_rate": 8.256880733944955e-05,
      "loss": 17.2652,
      "step": 45
    },
    {
      "epoch": 0.45662100456621,
      "grad_norm": 29.875,
      "learning_rate": 9.174311926605506e-05,
      "loss": 14.4775,
      "step": 50
    },
    {
      "epoch": 0.502283105022831,
      "grad_norm": 37.75,
      "learning_rate": 0.00010091743119266055,
      "loss": 9.6302,
      "step": 55
    },
    {
      "epoch": 0.547945205479452,
      "grad_norm": 8.75,
      "learning_rate": 0.00011009174311926606,
      "loss": 4.0499,
      "step": 60
    },
    {
      "epoch": 0.593607305936073,
      "grad_norm": 7.03125,
      "learning_rate": 0.00011926605504587157,
      "loss": 2.4784,
      "step": 65
    },
    {
      "epoch": 0.639269406392694,
      "grad_norm": 2.546875,
      "learning_rate": 0.00012844036697247707,
      "loss": 2.0967,
      "step": 70
    },
    {
      "epoch": 0.684931506849315,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00013761467889908258,
      "loss": 1.8211,
      "step": 75
    },
    {
      "epoch": 0.730593607305936,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001467889908256881,
      "loss": 1.6757,
      "step": 80
    },
    {
      "epoch": 0.776255707762557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001559633027522936,
      "loss": 1.5691,
      "step": 85
    },
    {
      "epoch": 0.821917808219178,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0001651376146788991,
      "loss": 1.4872,
      "step": 90
    },
    {
      "epoch": 0.867579908675799,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00017431192660550458,
      "loss": 1.4228,
      "step": 95
    },
    {
      "epoch": 0.91324200913242,
      "grad_norm": 2.546875,
      "learning_rate": 0.00018348623853211012,
      "loss": 1.3778,
      "step": 100
    },
    {
      "epoch": 0.958904109589041,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001926605504587156,
      "loss": 1.3554,
      "step": 105
    },
    {
      "epoch": 0.9954337899543378,
      "eval_loss": 2.645094871520996,
      "eval_runtime": 0.2786,
      "eval_samples_per_second": 35.888,
      "eval_steps_per_second": 3.589,
      "step": 109
    },
    {
      "epoch": 1.004566210045662,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00019999948721966259,
      "loss": 1.326,
      "step": 110
    },
    {
      "epoch": 1.0502283105022832,
      "grad_norm": 1.015625,
      "learning_rate": 0.00019998154046002822,
      "loss": 1.291,
      "step": 115
    },
    {
      "epoch": 1.095890410958904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001999379599421534,
      "loss": 1.2721,
      "step": 120
    },
    {
      "epoch": 1.1415525114155252,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019986875683942535,
      "loss": 1.2479,
      "step": 125
    },
    {
      "epoch": 1.187214611872146,
      "grad_norm": 3.375,
      "learning_rate": 0.00019977394889447524,
      "loss": 1.2491,
      "step": 130
    },
    {
      "epoch": 1.2328767123287672,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019965356041462955,
      "loss": 1.2212,
      "step": 135
    },
    {
      "epoch": 1.278538812785388,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019950762226567781,
      "loss": 1.2246,
      "step": 140
    },
    {
      "epoch": 1.3242009132420092,
      "grad_norm": 1.984375,
      "learning_rate": 0.00019933617186395917,
      "loss": 1.2387,
      "step": 145
    },
    {
      "epoch": 1.36986301369863,
      "grad_norm": 2.46875,
      "learning_rate": 0.00019913925316676945,
      "loss": 1.203,
      "step": 150
    },
    {
      "epoch": 1.4155251141552512,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00019891691666109113,
      "loss": 1.1869,
      "step": 155
    },
    {
      "epoch": 1.461187214611872,
      "grad_norm": 2.28125,
      "learning_rate": 0.00019866921935064906,
      "loss": 1.1858,
      "step": 160
    },
    {
      "epoch": 1.5068493150684932,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019839622474129596,
      "loss": 1.1696,
      "step": 165
    },
    {
      "epoch": 1.5525114155251143,
      "grad_norm": 0.81640625,
      "learning_rate": 0.00019809800282473013,
      "loss": 1.1624,
      "step": 170
    },
    {
      "epoch": 1.5981735159817352,
      "grad_norm": 1.734375,
      "learning_rate": 0.0001977746300605507,
      "loss": 1.1494,
      "step": 175
    },
    {
      "epoch": 1.643835616438356,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019742618935665476,
      "loss": 1.1314,
      "step": 180
    },
    {
      "epoch": 1.6894977168949772,
      "grad_norm": 2.078125,
      "learning_rate": 0.00019705277004798073,
      "loss": 1.1407,
      "step": 185
    },
    {
      "epoch": 1.7351598173515983,
      "grad_norm": 2.0625,
      "learning_rate": 0.0001966544678736044,
      "loss": 1.1287,
      "step": 190
    },
    {
      "epoch": 1.7808219178082192,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00019623138495219292,
      "loss": 1.1407,
      "step": 195
    },
    {
      "epoch": 1.82648401826484,
      "grad_norm": 12.0,
      "learning_rate": 0.00019578362975582292,
      "loss": 1.1151,
      "step": 200
    },
    {
      "epoch": 1.8721461187214612,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019531131708217005,
      "loss": 1.1221,
      "step": 205
    },
    {
      "epoch": 1.9178082191780823,
      "grad_norm": 0.90625,
      "learning_rate": 0.0001948145680250766,
      "loss": 1.0982,
      "step": 210
    },
    {
      "epoch": 1.9634703196347032,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00019429350994350483,
      "loss": 1.0898,
      "step": 215
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.508340358734131,
      "eval_runtime": 0.2456,
      "eval_samples_per_second": 40.72,
      "eval_steps_per_second": 4.072,
      "step": 219
    },
    {
      "epoch": 2.009132420091324,
      "grad_norm": 2.59375,
      "learning_rate": 0.00019374827642888398,
      "loss": 1.1068,
      "step": 220
    },
    {
      "epoch": 2.0547945205479454,
      "grad_norm": 4.09375,
      "learning_rate": 0.0001931790072708596,
      "loss": 1.0932,
      "step": 225
    },
    {
      "epoch": 2.1004566210045663,
      "grad_norm": 1.5,
      "learning_rate": 0.00019258584842145343,
      "loss": 1.1001,
      "step": 230
    },
    {
      "epoch": 2.146118721461187,
      "grad_norm": 11.25,
      "learning_rate": 0.00019196895195764362,
      "loss": 1.1001,
      "step": 235
    },
    {
      "epoch": 2.191780821917808,
      "grad_norm": 1.9375,
      "learning_rate": 0.0001913284760423745,
      "loss": 1.1046,
      "step": 240
    },
    {
      "epoch": 2.237442922374429,
      "grad_norm": 2.828125,
      "learning_rate": 0.00019066458488400584,
      "loss": 1.0795,
      "step": 245
    },
    {
      "epoch": 2.2831050228310503,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018997744869421246,
      "loss": 1.0767,
      "step": 250
    },
    {
      "epoch": 2.328767123287671,
      "grad_norm": 4.875,
      "learning_rate": 0.00018926724364434446,
      "loss": 1.059,
      "step": 255
    },
    {
      "epoch": 2.374429223744292,
      "grad_norm": 3.171875,
      "learning_rate": 0.0001885341518202595,
      "loss": 1.0695,
      "step": 260
    },
    {
      "epoch": 2.4200913242009134,
      "grad_norm": 0.734375,
      "learning_rate": 0.00018777836117563892,
      "loss": 1.0709,
      "step": 265
    },
    {
      "epoch": 2.4657534246575343,
      "grad_norm": 7.59375,
      "learning_rate": 0.00018700006548379898,
      "loss": 1.0677,
      "step": 270
    },
    {
      "epoch": 2.5114155251141552,
      "grad_norm": 0.984375,
      "learning_rate": 0.0001861994642880105,
      "loss": 1.0693,
      "step": 275
    },
    {
      "epoch": 2.557077625570776,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00018537676285033887,
      "loss": 1.0508,
      "step": 280
    },
    {
      "epoch": 2.602739726027397,
      "grad_norm": 0.578125,
      "learning_rate": 0.0001845321720990181,
      "loss": 1.0449,
      "step": 285
    },
    {
      "epoch": 2.6484018264840183,
      "grad_norm": 1.015625,
      "learning_rate": 0.00018366590857437184,
      "loss": 1.0562,
      "step": 290
    },
    {
      "epoch": 2.6940639269406392,
      "grad_norm": 1.734375,
      "learning_rate": 0.00018277819437329576,
      "loss": 1.0428,
      "step": 295
    },
    {
      "epoch": 2.73972602739726,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018186925709231532,
      "loss": 1.0321,
      "step": 300
    },
    {
      "epoch": 2.7853881278538815,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001809393297692334,
      "loss": 1.0253,
      "step": 305
    },
    {
      "epoch": 2.8310502283105023,
      "grad_norm": 3.1875,
      "learning_rate": 0.0001799886508233829,
      "loss": 1.0377,
      "step": 310
    },
    {
      "epoch": 2.8767123287671232,
      "grad_norm": 3.484375,
      "learning_rate": 0.0001790174639944997,
      "loss": 1.0359,
      "step": 315
    },
    {
      "epoch": 2.922374429223744,
      "grad_norm": 6.09375,
      "learning_rate": 0.00017802601828023138,
      "loss": 1.0428,
      "step": 320
    },
    {
      "epoch": 2.968036529680365,
      "grad_norm": 3.03125,
      "learning_rate": 0.00017701456787229804,
      "loss": 1.0434,
      "step": 325
    },
    {
      "epoch": 2.9954337899543377,
      "eval_loss": 2.480058193206787,
      "eval_runtime": 0.2581,
      "eval_samples_per_second": 38.741,
      "eval_steps_per_second": 3.874,
      "step": 328
    },
    {
      "epoch": 3.0136986301369864,
      "grad_norm": 7.625,
      "learning_rate": 0.0001759833720913214,
      "loss": 1.0302,
      "step": 330
    },
    {
      "epoch": 3.0593607305936072,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00017493269532033883,
      "loss": 1.0273,
      "step": 335
    },
    {
      "epoch": 3.105022831050228,
      "grad_norm": 1.71875,
      "learning_rate": 0.0001738628069370195,
      "loss": 1.0212,
      "step": 340
    },
    {
      "epoch": 3.1506849315068495,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017277398124460023,
      "loss": 1.013,
      "step": 345
    },
    {
      "epoch": 3.1963470319634704,
      "grad_norm": 2.390625,
      "learning_rate": 0.000171666497401558,
      "loss": 1.0077,
      "step": 350
    },
    {
      "epoch": 3.2420091324200913,
      "grad_norm": 0.921875,
      "learning_rate": 0.0001705406393500381,
      "loss": 1.0111,
      "step": 355
    },
    {
      "epoch": 3.287671232876712,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016939669574305566,
      "loss": 1.0047,
      "step": 360
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0001682349598704892,
      "loss": 0.9977,
      "step": 365
    },
    {
      "epoch": 3.3789954337899544,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016705572958388576,
      "loss": 0.9914,
      "step": 370
    },
    {
      "epoch": 3.4246575342465753,
      "grad_norm": 1.046875,
      "learning_rate": 0.00016585930722009601,
      "loss": 1.0012,
      "step": 375
    },
    {
      "epoch": 3.470319634703196,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00016464599952375998,
      "loss": 0.9888,
      "step": 380
    },
    {
      "epoch": 3.5159817351598175,
      "grad_norm": 0.73828125,
      "learning_rate": 0.000163416117568662,
      "loss": 1.0036,
      "step": 385
    },
    {
      "epoch": 3.5616438356164384,
      "grad_norm": 2.515625,
      "learning_rate": 0.0001621699766779763,
      "loss": 0.9963,
      "step": 390
    },
    {
      "epoch": 3.6073059360730593,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016090789634342278,
      "loss": 0.9955,
      "step": 395
    },
    {
      "epoch": 3.65296803652968,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015963020014335438,
      "loss": 0.9953,
      "step": 400
    },
    {
      "epoch": 3.6986301369863015,
      "grad_norm": 2.015625,
      "learning_rate": 0.0001583372156597961,
      "loss": 0.9959,
      "step": 405
    },
    {
      "epoch": 3.7442922374429224,
      "grad_norm": 1.75,
      "learning_rate": 0.00015702927439445826,
      "loss": 0.9906,
      "step": 410
    },
    {
      "epoch": 3.7899543378995433,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015570671168374438,
      "loss": 0.9849,
      "step": 415
    },
    {
      "epoch": 3.8356164383561646,
      "grad_norm": 2.234375,
      "learning_rate": 0.00015436986661277577,
      "loss": 0.9697,
      "step": 420
    },
    {
      "epoch": 3.8812785388127855,
      "grad_norm": 1.5,
      "learning_rate": 0.0001530190819284555,
      "loss": 0.979,
      "step": 425
    },
    {
      "epoch": 3.9269406392694064,
      "grad_norm": 10.75,
      "learning_rate": 0.00015165470395159313,
      "loss": 0.9715,
      "step": 430
    },
    {
      "epoch": 3.9726027397260273,
      "grad_norm": 2.71875,
      "learning_rate": 0.0001502770824881133,
      "loss": 0.9864,
      "step": 435
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.474334239959717,
      "eval_runtime": 0.2363,
      "eval_samples_per_second": 42.318,
      "eval_steps_per_second": 4.232,
      "step": 438
    },
    {
      "epoch": 4.018264840182648,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014888657073937076,
      "loss": 0.9764,
      "step": 440
    },
    {
      "epoch": 4.063926940639269,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00014748352521159493,
      "loss": 0.9564,
      "step": 445
    },
    {
      "epoch": 4.109589041095891,
      "grad_norm": 0.7265625,
      "learning_rate": 0.0001460683056244869,
      "loss": 0.9573,
      "step": 450
    },
    {
      "epoch": 4.155251141552512,
      "grad_norm": 11.5625,
      "learning_rate": 0.00014464127481899312,
      "loss": 0.957,
      "step": 455
    },
    {
      "epoch": 4.200913242009133,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00014320279866427796,
      "loss": 0.9596,
      "step": 460
    },
    {
      "epoch": 4.2465753424657535,
      "grad_norm": 2.03125,
      "learning_rate": 0.00014175324596392075,
      "loss": 0.9647,
      "step": 465
    },
    {
      "epoch": 4.292237442922374,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014029298836135988,
      "loss": 0.9632,
      "step": 470
    },
    {
      "epoch": 4.337899543378995,
      "grad_norm": 5.09375,
      "learning_rate": 0.00013882240024460927,
      "loss": 0.9664,
      "step": 475
    },
    {
      "epoch": 4.383561643835616,
      "grad_norm": 2.96875,
      "learning_rate": 0.0001373418586502706,
      "loss": 0.964,
      "step": 480
    },
    {
      "epoch": 4.429223744292237,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001358517431668672,
      "loss": 0.9531,
      "step": 485
    },
    {
      "epoch": 4.474885844748858,
      "grad_norm": 1.046875,
      "learning_rate": 0.00013435243583752294,
      "loss": 0.958,
      "step": 490
    },
    {
      "epoch": 4.52054794520548,
      "grad_norm": 0.6953125,
      "learning_rate": 0.00013284432106201233,
      "loss": 0.9514,
      "step": 495
    },
    {
      "epoch": 4.566210045662101,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00013132778549820618,
      "loss": 0.9588,
      "step": 500
    },
    {
      "epoch": 4.6118721461187215,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00012980321796293836,
      "loss": 0.9494,
      "step": 505
    },
    {
      "epoch": 4.657534246575342,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00012827100933231905,
      "loss": 0.9508,
      "step": 510
    },
    {
      "epoch": 4.703196347031963,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00012673155244151985,
      "loss": 0.9557,
      "step": 515
    },
    {
      "epoch": 4.748858447488584,
      "grad_norm": 0.75,
      "learning_rate": 0.000125185241984057,
      "loss": 0.9508,
      "step": 520
    },
    {
      "epoch": 4.794520547945205,
      "grad_norm": 0.76953125,
      "learning_rate": 0.00012363247441059776,
      "loss": 0.9562,
      "step": 525
    },
    {
      "epoch": 4.840182648401827,
      "grad_norm": 1.078125,
      "learning_rate": 0.00012207364782731655,
      "loss": 0.9542,
      "step": 530
    },
    {
      "epoch": 4.885844748858448,
      "grad_norm": 2.1875,
      "learning_rate": 0.00012050916189382646,
      "loss": 0.9606,
      "step": 535
    },
    {
      "epoch": 4.931506849315069,
      "grad_norm": 1.84375,
      "learning_rate": 0.00011893941772071249,
      "loss": 0.9424,
      "step": 540
    },
    {
      "epoch": 4.9771689497716896,
      "grad_norm": 1.9375,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.9371,
      "step": 545
    },
    {
      "epoch": 4.995433789954338,
      "eval_loss": 2.485384464263916,
      "eval_runtime": 0.2567,
      "eval_samples_per_second": 38.96,
      "eval_steps_per_second": 3.896,
      "step": 547
    },
    {
      "epoch": 5.0228310502283104,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0001157857657354354,
      "loss": 0.9249,
      "step": 550
    },
    {
      "epoch": 5.068493150684931,
      "grad_norm": 2.53125,
      "learning_rate": 0.00011420266647205231,
      "loss": 0.9271,
      "step": 555
    },
    {
      "epoch": 5.114155251141552,
      "grad_norm": 12.0,
      "learning_rate": 0.00011261592585930576,
      "loss": 0.9329,
      "step": 560
    },
    {
      "epoch": 5.159817351598173,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00011102595071354472,
      "loss": 0.9238,
      "step": 565
    },
    {
      "epoch": 5.205479452054795,
      "grad_norm": 6.3125,
      "learning_rate": 0.00010943314868040364,
      "loss": 0.9134,
      "step": 570
    },
    {
      "epoch": 5.251141552511416,
      "grad_norm": 0.69921875,
      "learning_rate": 0.00010783792813028827,
      "loss": 0.91,
      "step": 575
    },
    {
      "epoch": 5.296803652968037,
      "grad_norm": 4.375,
      "learning_rate": 0.00010624069805367559,
      "loss": 0.9193,
      "step": 580
    },
    {
      "epoch": 5.342465753424658,
      "grad_norm": 0.984375,
      "learning_rate": 0.00010464186795625482,
      "loss": 0.9101,
      "step": 585
    },
    {
      "epoch": 5.3881278538812785,
      "grad_norm": 3.1875,
      "learning_rate": 0.00010304184775393642,
      "loss": 0.9122,
      "step": 590
    },
    {
      "epoch": 5.433789954337899,
      "grad_norm": 0.90625,
      "learning_rate": 0.00010144104766775572,
      "loss": 0.9126,
      "step": 595
    },
    {
      "epoch": 5.47945205479452,
      "grad_norm": 1.171875,
      "learning_rate": 9.983987811869862e-05,
      "loss": 0.9177,
      "step": 600
    },
    {
      "epoch": 5.525114155251142,
      "grad_norm": 0.53515625,
      "learning_rate": 9.823874962247564e-05,
      "loss": 0.9089,
      "step": 605
    },
    {
      "epoch": 5.570776255707763,
      "grad_norm": 1.0078125,
      "learning_rate": 9.663807268427198e-05,
      "loss": 0.9112,
      "step": 610
    },
    {
      "epoch": 5.616438356164384,
      "grad_norm": 0.69140625,
      "learning_rate": 9.503825769350017e-05,
      "loss": 0.9142,
      "step": 615
    },
    {
      "epoch": 5.662100456621005,
      "grad_norm": 0.8359375,
      "learning_rate": 9.343971481858246e-05,
      "loss": 0.9068,
      "step": 620
    },
    {
      "epoch": 5.707762557077626,
      "grad_norm": 0.92578125,
      "learning_rate": 9.184285390178978e-05,
      "loss": 0.9134,
      "step": 625
    },
    {
      "epoch": 5.7534246575342465,
      "grad_norm": 1.890625,
      "learning_rate": 9.024808435416434e-05,
      "loss": 0.9106,
      "step": 630
    },
    {
      "epoch": 5.799086757990867,
      "grad_norm": 0.90234375,
      "learning_rate": 8.865581505055291e-05,
      "loss": 0.9108,
      "step": 635
    },
    {
      "epoch": 5.844748858447488,
      "grad_norm": 1.6015625,
      "learning_rate": 8.706645422477739e-05,
      "loss": 0.9027,
      "step": 640
    },
    {
      "epoch": 5.890410958904109,
      "grad_norm": 4.90625,
      "learning_rate": 8.548040936496989e-05,
      "loss": 0.9217,
      "step": 645
    },
    {
      "epoch": 5.936073059360731,
      "grad_norm": 7.09375,
      "learning_rate": 8.389808710909881e-05,
      "loss": 0.9227,
      "step": 650
    },
    {
      "epoch": 5.981735159817352,
      "grad_norm": 6.625,
      "learning_rate": 8.231989314071317e-05,
      "loss": 0.9157,
      "step": 655
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.46421480178833,
      "eval_runtime": 0.2356,
      "eval_samples_per_second": 42.441,
      "eval_steps_per_second": 4.244,
      "step": 657
    },
    {
      "epoch": 6.027397260273973,
      "grad_norm": 0.7109375,
      "learning_rate": 8.07462320849313e-05,
      "loss": 0.902,
      "step": 660
    },
    {
      "epoch": 6.073059360730594,
      "grad_norm": 1.0703125,
      "learning_rate": 7.917750740470117e-05,
      "loss": 0.8855,
      "step": 665
    },
    {
      "epoch": 6.1187214611872145,
      "grad_norm": 1.25,
      "learning_rate": 7.761412129735852e-05,
      "loss": 0.9014,
      "step": 670
    },
    {
      "epoch": 6.164383561643835,
      "grad_norm": 0.984375,
      "learning_rate": 7.605647459150961e-05,
      "loss": 0.8863,
      "step": 675
    },
    {
      "epoch": 6.210045662100456,
      "grad_norm": 1.296875,
      "learning_rate": 7.450496664426477e-05,
      "loss": 0.8804,
      "step": 680
    },
    {
      "epoch": 6.255707762557078,
      "grad_norm": 1.1171875,
      "learning_rate": 7.295999523884921e-05,
      "loss": 0.8795,
      "step": 685
    },
    {
      "epoch": 6.301369863013699,
      "grad_norm": 0.890625,
      "learning_rate": 7.142195648261747e-05,
      "loss": 0.8855,
      "step": 690
    },
    {
      "epoch": 6.34703196347032,
      "grad_norm": 0.671875,
      "learning_rate": 6.989124470549745e-05,
      "loss": 0.8799,
      "step": 695
    },
    {
      "epoch": 6.392694063926941,
      "grad_norm": 0.796875,
      "learning_rate": 6.83682523588902e-05,
      "loss": 0.8731,
      "step": 700
    },
    {
      "epoch": 6.438356164383562,
      "grad_norm": 2.015625,
      "learning_rate": 6.685336991505122e-05,
      "loss": 0.8818,
      "step": 705
    },
    {
      "epoch": 6.4840182648401825,
      "grad_norm": 1.4140625,
      "learning_rate": 6.534698576697939e-05,
      "loss": 0.8792,
      "step": 710
    },
    {
      "epoch": 6.529680365296803,
      "grad_norm": 0.8125,
      "learning_rate": 6.384948612883873e-05,
      "loss": 0.8713,
      "step": 715
    },
    {
      "epoch": 6.575342465753424,
      "grad_norm": 0.53515625,
      "learning_rate": 6.2361254936939e-05,
      "loss": 0.8762,
      "step": 720
    },
    {
      "epoch": 6.621004566210045,
      "grad_norm": 0.515625,
      "learning_rate": 6.088267375130023e-05,
      "loss": 0.8708,
      "step": 725
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.70703125,
      "learning_rate": 5.941412165782645e-05,
      "loss": 0.8797,
      "step": 730
    },
    {
      "epoch": 6.712328767123288,
      "grad_norm": 1.21875,
      "learning_rate": 5.79559751711138e-05,
      "loss": 0.8634,
      "step": 735
    },
    {
      "epoch": 6.757990867579909,
      "grad_norm": 0.81640625,
      "learning_rate": 5.650860813791785e-05,
      "loss": 0.872,
      "step": 740
    },
    {
      "epoch": 6.80365296803653,
      "grad_norm": 0.625,
      "learning_rate": 5.507239164130501e-05,
      "loss": 0.8661,
      "step": 745
    },
    {
      "epoch": 6.8493150684931505,
      "grad_norm": 0.828125,
      "learning_rate": 5.364769390551225e-05,
      "loss": 0.8744,
      "step": 750
    },
    {
      "epoch": 6.894977168949771,
      "grad_norm": 1.5234375,
      "learning_rate": 5.2234880201540284e-05,
      "loss": 0.8662,
      "step": 755
    },
    {
      "epoch": 6.940639269406392,
      "grad_norm": 0.94140625,
      "learning_rate": 5.0834312753503124e-05,
      "loss": 0.8764,
      "step": 760
    },
    {
      "epoch": 6.986301369863014,
      "grad_norm": 0.4609375,
      "learning_rate": 4.9446350645759885e-05,
      "loss": 0.8657,
      "step": 765
    },
    {
      "epoch": 6.995433789954338,
      "eval_loss": 2.5075883865356445,
      "eval_runtime": 0.2561,
      "eval_samples_per_second": 39.047,
      "eval_steps_per_second": 3.905,
      "step": 766
    },
    {
      "epoch": 7.031963470319635,
      "grad_norm": 0.408203125,
      "learning_rate": 4.807134973085036e-05,
      "loss": 0.8614,
      "step": 770
    },
    {
      "epoch": 7.077625570776256,
      "grad_norm": 0.56640625,
      "learning_rate": 4.6709662538260267e-05,
      "loss": 0.8457,
      "step": 775
    },
    {
      "epoch": 7.123287671232877,
      "grad_norm": 0.447265625,
      "learning_rate": 4.53616381840377e-05,
      "loss": 0.8502,
      "step": 780
    },
    {
      "epoch": 7.168949771689498,
      "grad_norm": 0.64453125,
      "learning_rate": 4.402762228128531e-05,
      "loss": 0.8536,
      "step": 785
    },
    {
      "epoch": 7.2146118721461185,
      "grad_norm": 0.78515625,
      "learning_rate": 4.2707956851550016e-05,
      "loss": 0.8531,
      "step": 790
    },
    {
      "epoch": 7.260273972602739,
      "grad_norm": 0.68359375,
      "learning_rate": 4.140298023713416e-05,
      "loss": 0.8609,
      "step": 795
    },
    {
      "epoch": 7.30593607305936,
      "grad_norm": 0.671875,
      "learning_rate": 4.011302701434937e-05,
      "loss": 0.8529,
      "step": 800
    },
    {
      "epoch": 7.351598173515982,
      "grad_norm": 0.96875,
      "learning_rate": 3.8838427907736476e-05,
      "loss": 0.8566,
      "step": 805
    },
    {
      "epoch": 7.397260273972603,
      "grad_norm": 0.6171875,
      "learning_rate": 3.757950970527249e-05,
      "loss": 0.8508,
      "step": 810
    },
    {
      "epoch": 7.442922374429224,
      "grad_norm": 0.435546875,
      "learning_rate": 3.633659517458736e-05,
      "loss": 0.8513,
      "step": 815
    },
    {
      "epoch": 7.488584474885845,
      "grad_norm": 0.466796875,
      "learning_rate": 3.5110002980210975e-05,
      "loss": 0.8499,
      "step": 820
    },
    {
      "epoch": 7.534246575342466,
      "grad_norm": 0.43359375,
      "learning_rate": 3.3900047601872596e-05,
      "loss": 0.8493,
      "step": 825
    },
    {
      "epoch": 7.579908675799087,
      "grad_norm": 0.6171875,
      "learning_rate": 3.270703925387279e-05,
      "loss": 0.851,
      "step": 830
    },
    {
      "epoch": 7.6255707762557075,
      "grad_norm": 0.5234375,
      "learning_rate": 3.153128380554941e-05,
      "loss": 0.8452,
      "step": 835
    },
    {
      "epoch": 7.671232876712329,
      "grad_norm": 0.43359375,
      "learning_rate": 3.037308270285709e-05,
      "loss": 0.862,
      "step": 840
    },
    {
      "epoch": 7.71689497716895,
      "grad_norm": 0.625,
      "learning_rate": 2.923273289108115e-05,
      "loss": 0.8487,
      "step": 845
    },
    {
      "epoch": 7.762557077625571,
      "grad_norm": 0.39453125,
      "learning_rate": 2.8110526738705344e-05,
      "loss": 0.8516,
      "step": 850
    },
    {
      "epoch": 7.808219178082192,
      "grad_norm": 0.546875,
      "learning_rate": 2.7006751962452882e-05,
      "loss": 0.8541,
      "step": 855
    },
    {
      "epoch": 7.853881278538813,
      "grad_norm": 0.419921875,
      "learning_rate": 2.592169155352031e-05,
      "loss": 0.8486,
      "step": 860
    },
    {
      "epoch": 7.899543378995434,
      "grad_norm": 0.86328125,
      "learning_rate": 2.485562370502279e-05,
      "loss": 0.8402,
      "step": 865
    },
    {
      "epoch": 7.945205479452055,
      "grad_norm": 0.412109375,
      "learning_rate": 2.3808821740669606e-05,
      "loss": 0.8474,
      "step": 870
    },
    {
      "epoch": 7.9908675799086755,
      "grad_norm": 0.40625,
      "learning_rate": 2.2781554044688015e-05,
      "loss": 0.8393,
      "step": 875
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.515906810760498,
      "eval_runtime": 0.2401,
      "eval_samples_per_second": 41.645,
      "eval_steps_per_second": 4.164,
      "step": 876
    },
    {
      "epoch": 8.036529680365296,
      "grad_norm": 0.5390625,
      "learning_rate": 2.1774083993013718e-05,
      "loss": 0.8404,
      "step": 880
    },
    {
      "epoch": 8.082191780821917,
      "grad_norm": 0.5234375,
      "learning_rate": 2.078666988576504e-05,
      "loss": 0.837,
      "step": 885
    },
    {
      "epoch": 8.127853881278538,
      "grad_norm": 0.58203125,
      "learning_rate": 1.9819564881018983e-05,
      "loss": 0.8372,
      "step": 890
    },
    {
      "epoch": 8.173515981735159,
      "grad_norm": 0.486328125,
      "learning_rate": 1.887301692990494e-05,
      "loss": 0.846,
      "step": 895
    },
    {
      "epoch": 8.219178082191782,
      "grad_norm": 0.439453125,
      "learning_rate": 1.7947268713034127e-05,
      "loss": 0.8461,
      "step": 900
    },
    {
      "epoch": 8.264840182648403,
      "grad_norm": 0.4140625,
      "learning_rate": 1.7042557578279626e-05,
      "loss": 0.8373,
      "step": 905
    },
    {
      "epoch": 8.310502283105023,
      "grad_norm": 0.46484375,
      "learning_rate": 1.6159115479924257e-05,
      "loss": 0.8422,
      "step": 910
    },
    {
      "epoch": 8.356164383561644,
      "grad_norm": 0.4453125,
      "learning_rate": 1.529716891919074e-05,
      "loss": 0.8403,
      "step": 915
    },
    {
      "epoch": 8.401826484018265,
      "grad_norm": 0.419921875,
      "learning_rate": 1.4456938886170412e-05,
      "loss": 0.8343,
      "step": 920
    },
    {
      "epoch": 8.447488584474886,
      "grad_norm": 0.43359375,
      "learning_rate": 1.3638640803164516e-05,
      "loss": 0.8355,
      "step": 925
    },
    {
      "epoch": 8.493150684931507,
      "grad_norm": 0.41796875,
      "learning_rate": 1.2842484469453365e-05,
      "loss": 0.841,
      "step": 930
    },
    {
      "epoch": 8.538812785388128,
      "grad_norm": 0.455078125,
      "learning_rate": 1.2068674007506786e-05,
      "loss": 0.8396,
      "step": 935
    },
    {
      "epoch": 8.584474885844749,
      "grad_norm": 0.40625,
      "learning_rate": 1.1317407810650372e-05,
      "loss": 0.8377,
      "step": 940
    },
    {
      "epoch": 8.63013698630137,
      "grad_norm": 0.392578125,
      "learning_rate": 1.058887849220026e-05,
      "loss": 0.8348,
      "step": 945
    },
    {
      "epoch": 8.67579908675799,
      "grad_norm": 0.396484375,
      "learning_rate": 9.883272836080116e-06,
      "loss": 0.8388,
      "step": 950
    },
    {
      "epoch": 8.721461187214611,
      "grad_norm": 0.396484375,
      "learning_rate": 9.200771748932513e-06,
      "loss": 0.8366,
      "step": 955
    },
    {
      "epoch": 8.767123287671232,
      "grad_norm": 0.416015625,
      "learning_rate": 8.541550213737171e-06,
      "loss": 0.8436,
      "step": 960
    },
    {
      "epoch": 8.812785388127853,
      "grad_norm": 0.455078125,
      "learning_rate": 7.905777244947954e-06,
      "loss": 0.8409,
      "step": 965
    },
    {
      "epoch": 8.858447488584474,
      "grad_norm": 0.4140625,
      "learning_rate": 7.293615845160196e-06,
      "loss": 0.8377,
      "step": 970
    },
    {
      "epoch": 8.904109589041095,
      "grad_norm": 0.408203125,
      "learning_rate": 6.705222963319191e-06,
      "loss": 0.8425,
      "step": 975
    },
    {
      "epoch": 8.949771689497716,
      "grad_norm": 0.419921875,
      "learning_rate": 6.140749454480932e-06,
      "loss": 0.8371,
      "step": 980
    },
    {
      "epoch": 8.995433789954339,
      "grad_norm": 0.408203125,
      "learning_rate": 5.6003400411351325e-06,
      "loss": 0.8462,
      "step": 985
    },
    {
      "epoch": 8.995433789954339,
      "eval_loss": 2.518533706665039,
      "eval_runtime": 0.2546,
      "eval_samples_per_second": 39.282,
      "eval_steps_per_second": 3.928,
      "step": 985
    },
    {
      "epoch": 9.04109589041096,
      "grad_norm": 0.392578125,
      "learning_rate": 5.0841332761005e-06,
      "loss": 0.8404,
      "step": 990
    },
    {
      "epoch": 9.08675799086758,
      "grad_norm": 0.416015625,
      "learning_rate": 4.592261507001993e-06,
      "loss": 0.8303,
      "step": 995
    },
    {
      "epoch": 9.132420091324201,
      "grad_norm": 0.42578125,
      "learning_rate": 4.124850842338779e-06,
      "loss": 0.8325,
      "step": 1000
    },
    {
      "epoch": 9.178082191780822,
      "grad_norm": 0.4140625,
      "learning_rate": 3.6820211191520125e-06,
      "loss": 0.8353,
      "step": 1005
    },
    {
      "epoch": 9.223744292237443,
      "grad_norm": 0.419921875,
      "learning_rate": 3.263885872300343e-06,
      "loss": 0.8347,
      "step": 1010
    },
    {
      "epoch": 9.269406392694064,
      "grad_norm": 0.3984375,
      "learning_rate": 2.8705523053513816e-06,
      "loss": 0.8329,
      "step": 1015
    },
    {
      "epoch": 9.315068493150685,
      "grad_norm": 0.421875,
      "learning_rate": 2.502121263096224e-06,
      "loss": 0.8369,
      "step": 1020
    },
    {
      "epoch": 9.360730593607306,
      "grad_norm": 0.419921875,
      "learning_rate": 2.1586872056944428e-06,
      "loss": 0.8324,
      "step": 1025
    },
    {
      "epoch": 9.406392694063927,
      "grad_norm": 0.412109375,
      "learning_rate": 1.840338184455881e-06,
      "loss": 0.8383,
      "step": 1030
    },
    {
      "epoch": 9.452054794520548,
      "grad_norm": 0.400390625,
      "learning_rate": 1.5471558192656777e-06,
      "loss": 0.8315,
      "step": 1035
    },
    {
      "epoch": 9.497716894977168,
      "grad_norm": 0.435546875,
      "learning_rate": 1.2792152776580968e-06,
      "loss": 0.8437,
      "step": 1040
    },
    {
      "epoch": 9.54337899543379,
      "grad_norm": 0.419921875,
      "learning_rate": 1.036585255544764e-06,
      "loss": 0.8418,
      "step": 1045
    },
    {
      "epoch": 9.58904109589041,
      "grad_norm": 0.400390625,
      "learning_rate": 8.193279596020121e-07,
      "loss": 0.8346,
      "step": 1050
    },
    {
      "epoch": 9.634703196347033,
      "grad_norm": 0.40234375,
      "learning_rate": 6.274990913221035e-07,
      "loss": 0.8415,
      "step": 1055
    },
    {
      "epoch": 9.680365296803654,
      "grad_norm": 0.416015625,
      "learning_rate": 4.6114783273213393e-07,
      "loss": 0.8339,
      "step": 1060
    },
    {
      "epoch": 9.726027397260275,
      "grad_norm": 0.439453125,
      "learning_rate": 3.203168337845508e-07,
      "loss": 0.8331,
      "step": 1065
    },
    {
      "epoch": 9.771689497716896,
      "grad_norm": 0.400390625,
      "learning_rate": 2.05042201422323e-07,
      "loss": 0.8458,
      "step": 1070
    },
    {
      "epoch": 9.817351598173516,
      "grad_norm": 0.41015625,
      "learning_rate": 1.1535349032167908e-07,
      "loss": 0.8444,
      "step": 1075
    },
    {
      "epoch": 9.863013698630137,
      "grad_norm": 0.400390625,
      "learning_rate": 5.127369531473525e-08,
      "loss": 0.8486,
      "step": 1080
    },
    {
      "epoch": 9.908675799086758,
      "grad_norm": 0.39453125,
      "learning_rate": 1.2819245493955744e-08,
      "loss": 0.8473,
      "step": 1085
    },
    {
      "epoch": 9.954337899543379,
      "grad_norm": 0.439453125,
      "learning_rate": 0.0,
      "loss": 0.8359,
      "step": 1090
    },
    {
      "epoch": 9.954337899543379,
      "eval_loss": 2.515676259994507,
      "eval_runtime": 0.2343,
      "eval_samples_per_second": 42.683,
      "eval_steps_per_second": 4.268,
      "step": 1090
    },
    {
      "epoch": 9.954337899543379,
      "step": 1090,
      "total_flos": 3.327732991202951e+18,
      "train_loss": 2.136770288659892,
      "train_runtime": 2636.8816,
      "train_samples_per_second": 26.554,
      "train_steps_per_second": 0.413
    }
  ],
  "logging_steps": 5,
  "max_steps": 1090,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 3.327732991202951e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}