{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.025,
      "grad_norm": 198.0,
      "learning_rate": 1.747190540874941e-05,
      "loss": 0.5751,
      "step": 5
    },
    {
      "epoch": 0.05,
      "grad_norm": 186.0,
      "learning_rate": 3.931178716968617e-05,
      "loss": 0.4175,
      "step": 10
    },
    {
      "epoch": 0.075,
      "grad_norm": 270.0,
      "learning_rate": 6.115166893062294e-05,
      "loss": 0.3509,
      "step": 15
    },
    {
      "epoch": 0.1,
      "grad_norm": 110.5,
      "learning_rate": 8.29915506915597e-05,
      "loss": 0.3508,
      "step": 20
    },
    {
      "epoch": 0.125,
      "grad_norm": 111.0,
      "learning_rate": 0.00010483143245249646,
      "loss": 0.4222,
      "step": 25
    },
    {
      "epoch": 0.15,
      "grad_norm": 83.5,
      "learning_rate": 0.00012667131421343323,
      "loss": 0.6498,
      "step": 30
    },
    {
      "epoch": 0.175,
      "grad_norm": 65.0,
      "learning_rate": 0.00014851119597437,
      "loss": 0.8894,
      "step": 35
    },
    {
      "epoch": 0.2,
      "grad_norm": 104.5,
      "learning_rate": 0.0001528649930352754,
      "loss": 0.6775,
      "step": 40
    },
    {
      "epoch": 0.225,
      "grad_norm": 29.5,
      "learning_rate": 0.00015280740168512177,
      "loss": 0.6998,
      "step": 45
    },
    {
      "epoch": 0.25,
      "grad_norm": 17.0,
      "learning_rate": 0.00015270555654763282,
      "loss": 0.6402,
      "step": 50
    },
    {
      "epoch": 0.275,
      "grad_norm": 16.75,
      "learning_rate": 0.00015255953633738878,
      "loss": 0.4757,
      "step": 55
    },
    {
      "epoch": 0.3,
      "grad_norm": 11.75,
      "learning_rate": 0.0001523694539112214,
      "loss": 0.3831,
      "step": 60
    },
    {
      "epoch": 0.325,
      "grad_norm": 6.5625,
      "learning_rate": 0.00015213545618098876,
      "loss": 0.3437,
      "step": 65
    },
    {
      "epoch": 0.35,
      "grad_norm": 6.03125,
      "learning_rate": 0.00015185772400002907,
      "loss": 0.2986,
      "step": 70
    },
    {
      "epoch": 0.375,
      "grad_norm": 4.46875,
      "learning_rate": 0.00015153647202338207,
      "loss": 0.2774,
      "step": 75
    },
    {
      "epoch": 0.4,
      "grad_norm": 7.6875,
      "learning_rate": 0.00015117194854188525,
      "loss": 0.2703,
      "step": 80
    },
    {
      "epoch": 0.425,
      "grad_norm": 5.65625,
      "learning_rate": 0.00015076443529027353,
      "loss": 0.2501,
      "step": 85
    },
    {
      "epoch": 0.45,
      "grad_norm": 26.375,
      "learning_rate": 0.00015031424722943083,
      "loss": 0.3738,
      "step": 90
    },
    {
      "epoch": 0.475,
      "grad_norm": 8.1875,
      "learning_rate": 0.00014982173230296148,
      "loss": 0.2765,
      "step": 95
    },
    {
      "epoch": 0.5,
      "grad_norm": 7.125,
      "learning_rate": 0.00014928727116826976,
      "loss": 0.2929,
      "step": 100
    },
    {
      "epoch": 0.525,
      "grad_norm": 7.8125,
      "learning_rate": 0.00014871127690235564,
      "loss": 0.2699,
      "step": 105
    },
    {
      "epoch": 0.55,
      "grad_norm": 6.125,
      "learning_rate": 0.00014809419468255356,
      "loss": 0.269,
      "step": 110
    },
    {
      "epoch": 0.575,
      "grad_norm": 4.1875,
      "learning_rate": 0.00014743650144246167,
      "loss": 0.2615,
      "step": 115
    },
    {
      "epoch": 0.6,
      "grad_norm": 3.828125,
      "learning_rate": 0.00014673870550332703,
      "loss": 0.2325,
      "step": 120
    },
    {
      "epoch": 0.625,
      "grad_norm": 5.375,
      "learning_rate": 0.00014600134618117166,
      "loss": 0.2295,
      "step": 125
    },
    {
      "epoch": 0.65,
      "grad_norm": 3.953125,
      "learning_rate": 0.0001452249933699633,
      "loss": 0.2407,
      "step": 130
    },
    {
      "epoch": 0.675,
      "grad_norm": 2.09375,
      "learning_rate": 0.0001444102471011529,
      "loss": 0.2192,
      "step": 135
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00014355773707991926,
      "loss": 0.2133,
      "step": 140
    },
    {
      "epoch": 0.725,
      "grad_norm": 2.078125,
      "learning_rate": 0.00014266812219847945,
      "loss": 0.2014,
      "step": 145
    },
    {
      "epoch": 0.75,
      "grad_norm": 2.0625,
      "learning_rate": 0.00014174209002684087,
      "loss": 0.191,
      "step": 150
    },
    {
      "epoch": 0.775,
      "grad_norm": 2.828125,
      "learning_rate": 0.00014078035628138847,
      "loss": 0.2024,
      "step": 155
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.015625,
      "learning_rate": 0.00013978366427171864,
      "loss": 0.1871,
      "step": 160
    },
    {
      "epoch": 0.825,
      "grad_norm": 1.90625,
      "learning_rate": 0.00013875278432614612,
      "loss": 0.1751,
      "step": 165
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00013768851319632887,
      "loss": 0.1679,
      "step": 170
    },
    {
      "epoch": 0.875,
      "grad_norm": 2.015625,
      "learning_rate": 0.00013659167344147067,
      "loss": 0.1818,
      "step": 175
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0001354631127925774,
      "loss": 0.1692,
      "step": 180
    },
    {
      "epoch": 0.925,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001343037034972584,
      "loss": 0.1619,
      "step": 185
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001331143416455796,
      "loss": 0.1617,
      "step": 190
    },
    {
      "epoch": 0.975,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00013189594647748868,
      "loss": 0.1615,
      "step": 195
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.5,
      "learning_rate": 0.00013064945967234835,
      "loss": 0.1689,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.16072207689285278,
      "eval_runtime": 0.5972,
      "eval_samples_per_second": 43.533,
      "eval_steps_per_second": 43.533,
      "step": 200
    },
    {
      "epoch": 1.025,
      "grad_norm": 1.828125,
      "learning_rate": 0.0001293758446211266,
      "loss": 0.1629,
      "step": 205
    },
    {
      "epoch": 1.05,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012807608568180618,
      "loss": 0.1624,
      "step": 210
    },
    {
      "epoch": 1.075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00012675118741858906,
      "loss": 0.1614,
      "step": 215
    },
    {
      "epoch": 1.1,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012540217382548384,
      "loss": 0.1636,
      "step": 220
    },
    {
      "epoch": 1.125,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001240300875348761,
      "loss": 0.1567,
      "step": 225
    },
    {
      "epoch": 1.15,
      "grad_norm": 2.609375,
      "learning_rate": 0.0001226359890116935,
      "loss": 0.1663,
      "step": 230
    },
    {
      "epoch": 1.175,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00012122095573378837,
      "loss": 0.1774,
      "step": 235
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.015625,
      "learning_rate": 0.00011978608135917105,
      "loss": 0.1701,
      "step": 240
    },
    {
      "epoch": 1.225,
      "grad_norm": 2.59375,
      "learning_rate": 0.00011833247488073823,
      "loss": 0.1853,
      "step": 245
    },
    {
      "epoch": 1.25,
      "grad_norm": 2.453125,
      "learning_rate": 0.00011686125976914878,
      "loss": 0.1948,
      "step": 250
    },
    {
      "epoch": 1.275,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011537357310451031,
      "loss": 0.1733,
      "step": 255
    },
    {
      "epoch": 1.3,
      "grad_norm": 1.890625,
      "learning_rate": 0.00011387056469754679,
      "loss": 0.1624,
      "step": 260
    },
    {
      "epoch": 1.325,
      "grad_norm": 2.390625,
      "learning_rate": 0.00011235339620092721,
      "loss": 0.1684,
      "step": 265
    },
    {
      "epoch": 1.35,
      "grad_norm": 2.390625,
      "learning_rate": 0.0001108232402114416,
      "loss": 0.169,
      "step": 270
    },
    {
      "epoch": 1.375,
      "grad_norm": 2.65625,
      "learning_rate": 0.0001092812793637186,
      "loss": 0.1789,
      "step": 275
    },
    {
      "epoch": 1.4,
      "grad_norm": 1.6875,
      "learning_rate": 0.0001077287054161847,
      "loss": 0.1695,
      "step": 280
    },
    {
      "epoch": 1.425,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010616671832997237,
      "loss": 0.1671,
      "step": 285
    },
    {
      "epoch": 1.45,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00010459652534148764,
      "loss": 0.1675,
      "step": 290
    },
    {
      "epoch": 1.475,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00010301934002935564,
      "loss": 0.159,
      "step": 295
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00010143638137646338,
      "loss": 0.1578,
      "step": 300
    },
    {
      "epoch": 1.525,
      "grad_norm": 1.171875,
      "learning_rate": 9.984887282782665e-05,
      "loss": 0.1588,
      "step": 305
    },
    {
      "epoch": 1.55,
      "grad_norm": 1.0546875,
      "learning_rate": 9.825804134500727e-05,
      "loss": 0.1535,
      "step": 310
    },
    {
      "epoch": 1.575,
      "grad_norm": 1.296875,
      "learning_rate": 9.666511645781328e-05,
      "loss": 0.1587,
      "step": 315
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.109375,
      "learning_rate": 9.507132931401333e-05,
      "loss": 0.1514,
      "step": 320
    },
    {
      "epoch": 1.625,
      "grad_norm": 1.21875,
      "learning_rate": 9.347791172780155e-05,
      "loss": 0.1486,
      "step": 325
    },
    {
      "epoch": 1.65,
      "grad_norm": 1.0703125,
      "learning_rate": 9.188609522774628e-05,
      "loss": 0.149,
      "step": 330
    },
    {
      "epoch": 1.675,
      "grad_norm": 0.87109375,
      "learning_rate": 9.029711010496061e-05,
      "loss": 0.1467,
      "step": 335
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.8984375,
      "learning_rate": 8.871218446222844e-05,
      "loss": 0.1456,
      "step": 340
    },
    {
      "epoch": 1.725,
      "grad_norm": 0.7734375,
      "learning_rate": 8.713254326482237e-05,
      "loss": 0.1439,
      "step": 345
    },
    {
      "epoch": 1.75,
      "grad_norm": 1.0625,
      "learning_rate": 8.555940739374653e-05,
      "loss": 0.1426,
      "step": 350
    },
    {
      "epoch": 1.775,
      "grad_norm": 0.9453125,
      "learning_rate": 8.399399270213575e-05,
      "loss": 0.1399,
      "step": 355
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.953125,
      "learning_rate": 8.243750907554097e-05,
      "loss": 0.1436,
      "step": 360
    },
    {
      "epoch": 1.825,
      "grad_norm": 0.7421875,
      "learning_rate": 8.089115949682696e-05,
      "loss": 0.1375,
      "step": 365
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.6640625,
      "learning_rate": 7.935613911640464e-05,
      "loss": 0.1384,
      "step": 370
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.828125,
      "learning_rate": 7.783363432851746e-05,
      "loss": 0.1366,
      "step": 375
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.859375,
      "learning_rate": 7.632482185429501e-05,
      "loss": 0.1374,
      "step": 380
    },
    {
      "epoch": 1.925,
      "grad_norm": 0.87890625,
      "learning_rate": 7.483086783228284e-05,
      "loss": 0.1413,
      "step": 385
    },
    {
      "epoch": 1.95,
      "grad_norm": 1.1015625,
      "learning_rate": 7.335292691715154e-05,
      "loss": 0.1371,
      "step": 390
    },
    {
      "epoch": 1.975,
      "grad_norm": 0.8359375,
      "learning_rate": 7.189214138728142e-05,
      "loss": 0.1413,
      "step": 395
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8359375,
      "learning_rate": 7.044964026191261e-05,
      "loss": 0.1378,
      "step": 400
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.13336719572544098,
      "eval_runtime": 0.5924,
      "eval_samples_per_second": 43.887,
      "eval_steps_per_second": 43.887,
      "step": 400
    },
    {
      "epoch": 2.025,
      "grad_norm": 0.8125,
      "learning_rate": 6.902653842854314e-05,
      "loss": 0.1382,
      "step": 405
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.6640625,
      "learning_rate": 6.762393578124894e-05,
      "loss": 0.1273,
      "step": 410
    },
    {
      "epoch": 2.075,
      "grad_norm": 0.6640625,
      "learning_rate": 6.624291637059237e-05,
      "loss": 0.1319,
      "step": 415
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.6796875,
      "learning_rate": 6.48845475657757e-05,
      "loss": 0.1296,
      "step": 420
    },
    {
      "epoch": 2.125,
      "grad_norm": 0.68359375,
      "learning_rate": 6.354987922968741e-05,
      "loss": 0.136,
      "step": 425
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.8203125,
      "learning_rate": 6.223994290747898e-05,
      "loss": 0.1332,
      "step": 430
    },
    {
      "epoch": 2.175,
      "grad_norm": 0.79296875,
      "learning_rate": 6.095575102929907e-05,
      "loss": 0.1337,
      "step": 435
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.7421875,
      "learning_rate": 5.969829612780141e-05,
      "loss": 0.1295,
      "step": 440
    },
    {
      "epoch": 2.225,
      "grad_norm": 0.85546875,
      "learning_rate": 5.8468550071031296e-05,
      "loss": 0.1291,
      "step": 445
    },
    {
      "epoch": 2.25,
      "grad_norm": 1.0390625,
      "learning_rate": 5.726746331128316e-05,
      "loss": 0.1364,
      "step": 450
    },
    {
      "epoch": 2.275,
      "grad_norm": 1.0,
      "learning_rate": 5.609596415051039e-05,
      "loss": 0.1365,
      "step": 455
    },
    {
      "epoch": 2.3,
      "grad_norm": 1.0390625,
      "learning_rate": 5.495495802285465e-05,
      "loss": 0.1317,
      "step": 460
    },
    {
      "epoch": 2.325,
      "grad_norm": 0.7890625,
      "learning_rate": 5.384532679484933e-05,
      "loss": 0.1296,
      "step": 465
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.66796875,
      "learning_rate": 5.276792808383817e-05,
      "loss": 0.1355,
      "step": 470
    },
    {
      "epoch": 2.375,
      "grad_norm": 0.875,
      "learning_rate": 5.1723594595135666e-05,
      "loss": 0.1324,
      "step": 475
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.77734375,
      "learning_rate": 5.07131334784416e-05,
      "loss": 0.1315,
      "step": 480
    },
    {
      "epoch": 2.425,
      "grad_norm": 0.8359375,
      "learning_rate": 4.973732570400718e-05,
      "loss": 0.1284,
      "step": 485
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.79296875,
      "learning_rate": 4.879692545903476e-05,
      "loss": 0.1345,
      "step": 490
    },
    {
      "epoch": 2.475,
      "grad_norm": 0.875,
      "learning_rate": 4.789265956477791e-05,
      "loss": 0.126,
      "step": 495
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.953125,
      "learning_rate": 4.702522691479217e-05,
      "loss": 0.1345,
      "step": 500
    },
    {
      "epoch": 2.5,
      "eval_loss": 0.12545974552631378,
      "eval_runtime": 0.5917,
      "eval_samples_per_second": 43.943,
      "eval_steps_per_second": 43.943,
      "step": 500
    },
    {
      "epoch": 2.525,
      "grad_norm": 0.8125,
      "learning_rate": 4.619529793477068e-05,
      "loss": 0.1291,
      "step": 505
    },
    {
      "epoch": 2.55,
      "grad_norm": 1.078125,
      "learning_rate": 4.540351406438219e-05,
      "loss": 0.1301,
      "step": 510
    },
    {
      "epoch": 2.575,
      "grad_norm": 0.921875,
      "learning_rate": 4.465048726151201e-05,
      "loss": 0.128,
      "step": 515
    },
    {
      "epoch": 2.6,
      "grad_norm": 1.1875,
      "learning_rate": 4.393679952928885e-05,
      "loss": 0.1288,
      "step": 520
    },
    {
      "epoch": 2.625,
      "grad_norm": 1.203125,
      "learning_rate": 4.3263002466263436e-05,
      "loss": 0.1281,
      "step": 525
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.9609375,
      "learning_rate": 4.262961684008613e-05,
      "loss": 0.1305,
      "step": 530
    },
    {
      "epoch": 2.675,
      "grad_norm": 0.88671875,
      "learning_rate": 4.203713218501353e-05,
      "loss": 0.1258,
      "step": 535
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.9921875,
      "learning_rate": 4.1486006423554745e-05,
      "loss": 0.1284,
      "step": 540
    },
    {
      "epoch": 2.725,
      "grad_norm": 0.84765625,
      "learning_rate": 4.097666551254989e-05,
      "loss": 0.1269,
      "step": 545
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.92578125,
      "learning_rate": 4.0509503113954545e-05,
      "loss": 0.128,
      "step": 550
    },
    {
      "epoch": 2.775,
      "grad_norm": 0.90234375,
      "learning_rate": 4.008488029058422e-05,
      "loss": 0.1293,
      "step": 555
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.85546875,
      "learning_rate": 3.9703125227054376e-05,
      "loss": 0.132,
      "step": 560
    },
    {
      "epoch": 2.825,
      "grad_norm": 0.859375,
      "learning_rate": 3.9364532976131475e-05,
      "loss": 0.1247,
      "step": 565
    },
    {
      "epoch": 2.85,
      "grad_norm": 1.6796875,
      "learning_rate": 3.906936523069101e-05,
      "loss": 0.1246,
      "step": 570
    },
    {
      "epoch": 2.875,
      "grad_norm": 1.0390625,
      "learning_rate": 3.8817850121459174e-05,
      "loss": 0.1246,
      "step": 575
    },
    {
      "epoch": 2.9,
      "grad_norm": 1.03125,
      "learning_rate": 3.861018204069391e-05,
      "loss": 0.1322,
      "step": 580
    },
    {
      "epoch": 2.925,
      "grad_norm": 0.859375,
      "learning_rate": 3.8446521491942034e-05,
      "loss": 0.1277,
      "step": 585
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.984375,
      "learning_rate": 3.832699496598859e-05,
      "loss": 0.1264,
      "step": 590
    },
    {
      "epoch": 2.975,
      "grad_norm": 0.81640625,
      "learning_rate": 3.8251694843093894e-05,
      "loss": 0.1277,
      "step": 595
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.78515625,
      "learning_rate": 3.8220679321594226e-05,
      "loss": 0.1288,
      "step": 600
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.1242036521434784,
      "eval_runtime": 0.5964,
      "eval_samples_per_second": 43.593,
      "eval_steps_per_second": 43.593,
      "step": 600
    }
  ],
  "logging_steps": 5,
  "max_steps": 600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.19491360014336e+16,
  "train_batch_size": 46,
  "trial_name": null,
  "trial_params": null
}