{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 50,
  "global_step": 537,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009324009324009324,
      "grad_norm": 269.7618103027344,
      "learning_rate": 1.8604651162790698e-07,
      "loss": 13.883856201171875,
      "step": 5
    },
    {
      "epoch": 0.018648018648018648,
      "grad_norm": 256.5934143066406,
      "learning_rate": 4.186046511627907e-07,
      "loss": 13.883564758300782,
      "step": 10
    },
    {
      "epoch": 0.027972027972027972,
      "grad_norm": 247.4810028076172,
      "learning_rate": 6.511627906976745e-07,
      "loss": 13.845947265625,
      "step": 15
    },
    {
      "epoch": 0.037296037296037296,
      "grad_norm": 243.70272827148438,
      "learning_rate": 8.837209302325582e-07,
      "loss": 13.705046081542969,
      "step": 20
    },
    {
      "epoch": 0.046620046620046623,
      "grad_norm": 238.2322235107422,
      "learning_rate": 1.116279069767442e-06,
      "loss": 13.556683349609376,
      "step": 25
    },
    {
      "epoch": 0.055944055944055944,
      "grad_norm": 241.71214294433594,
      "learning_rate": 1.3488372093023258e-06,
      "loss": 13.285650634765625,
      "step": 30
    },
    {
      "epoch": 0.06526806526806526,
      "grad_norm": 221.19960021972656,
      "learning_rate": 1.5813953488372093e-06,
      "loss": 13.05435791015625,
      "step": 35
    },
    {
      "epoch": 0.07459207459207459,
      "grad_norm": 205.08326721191406,
      "learning_rate": 1.8139534883720933e-06,
      "loss": 12.770874786376954,
      "step": 40
    },
    {
      "epoch": 0.08391608391608392,
      "grad_norm": 170.5838623046875,
      "learning_rate": 2.0465116279069768e-06,
      "loss": 12.280790710449219,
      "step": 45
    },
    {
      "epoch": 0.09324009324009325,
      "grad_norm": 96.90670776367188,
      "learning_rate": 2.2790697674418607e-06,
      "loss": 11.730733489990234,
      "step": 50
    },
    {
      "epoch": 0.09324009324009325,
      "eval_com_accuracy": 0.2,
      "eval_rew_accuracy": 0.276,
      "step": 50
    },
    {
      "epoch": 0.10256410256410256,
      "grad_norm": 61.14415740966797,
      "learning_rate": 2.5116279069767446e-06,
      "loss": 11.25308609008789,
      "step": 55
    },
    {
      "epoch": 0.11188811188811189,
      "grad_norm": 56.027095794677734,
      "learning_rate": 2.744186046511628e-06,
      "loss": 11.197045135498048,
      "step": 60
    },
    {
      "epoch": 0.12121212121212122,
      "grad_norm": 46.54409408569336,
      "learning_rate": 2.9767441860465116e-06,
      "loss": 10.874276733398437,
      "step": 65
    },
    {
      "epoch": 0.13053613053613053,
      "grad_norm": 40.62468719482422,
      "learning_rate": 3.2093023255813956e-06,
      "loss": 10.69852294921875,
      "step": 70
    },
    {
      "epoch": 0.13986013986013987,
      "grad_norm": 37.44975662231445,
      "learning_rate": 3.4418604651162795e-06,
      "loss": 10.560739135742187,
      "step": 75
    },
    {
      "epoch": 0.14918414918414918,
      "grad_norm": 32.19042205810547,
      "learning_rate": 3.674418604651163e-06,
      "loss": 10.408053588867187,
      "step": 80
    },
    {
      "epoch": 0.1585081585081585,
      "grad_norm": 27.208942413330078,
      "learning_rate": 3.906976744186047e-06,
      "loss": 10.108213806152344,
      "step": 85
    },
    {
      "epoch": 0.16783216783216784,
      "grad_norm": 24.76529312133789,
      "learning_rate": 4.1395348837209304e-06,
      "loss": 9.929393005371093,
      "step": 90
    },
    {
      "epoch": 0.17715617715617715,
      "grad_norm": 23.917495727539062,
      "learning_rate": 4.372093023255815e-06,
      "loss": 9.677659606933593,
      "step": 95
    },
    {
      "epoch": 0.1864801864801865,
      "grad_norm": 22.98892593383789,
      "learning_rate": 4.604651162790698e-06,
      "loss": 9.481817626953125,
      "step": 100
    },
    {
      "epoch": 0.1864801864801865,
      "eval_com_accuracy": 0.176,
      "eval_rew_accuracy": 0.264,
      "step": 100
    },
    {
      "epoch": 0.1958041958041958,
      "grad_norm": 25.177200317382812,
      "learning_rate": 4.837209302325582e-06,
      "loss": 9.42150650024414,
      "step": 105
    },
    {
      "epoch": 0.20512820512820512,
      "grad_norm": 18.95603370666504,
      "learning_rate": 5.069767441860466e-06,
      "loss": 9.267645263671875,
      "step": 110
    },
    {
      "epoch": 0.21445221445221446,
      "grad_norm": 22.155452728271484,
      "learning_rate": 5.302325581395349e-06,
      "loss": 9.06854248046875,
      "step": 115
    },
    {
      "epoch": 0.22377622377622378,
      "grad_norm": 20.26909065246582,
      "learning_rate": 5.534883720930233e-06,
      "loss": 9.045303344726562,
      "step": 120
    },
    {
      "epoch": 0.2331002331002331,
      "grad_norm": 26.49457359313965,
      "learning_rate": 5.7674418604651175e-06,
      "loss": 8.91327362060547,
      "step": 125
    },
    {
      "epoch": 0.24242424242424243,
      "grad_norm": 19.886178970336914,
      "learning_rate": 6e-06,
      "loss": 8.950289154052735,
      "step": 130
    },
    {
      "epoch": 0.2517482517482518,
      "grad_norm": 18.22088050842285,
      "learning_rate": 6.2325581395348845e-06,
      "loss": 8.710862731933593,
      "step": 135
    },
    {
      "epoch": 0.26107226107226106,
      "grad_norm": 21.09984016418457,
      "learning_rate": 6.465116279069767e-06,
      "loss": 8.762718200683594,
      "step": 140
    },
    {
      "epoch": 0.2703962703962704,
      "grad_norm": 17.96813201904297,
      "learning_rate": 6.6976744186046515e-06,
      "loss": 8.533232116699219,
      "step": 145
    },
    {
      "epoch": 0.27972027972027974,
      "grad_norm": 20.52909278869629,
      "learning_rate": 6.930232558139536e-06,
      "loss": 8.279258728027344,
      "step": 150
    },
    {
      "epoch": 0.27972027972027974,
      "eval_com_accuracy": 0.152,
      "eval_rew_accuracy": 0.252,
      "step": 150
    },
    {
      "epoch": 0.289044289044289,
      "grad_norm": 19.475040435791016,
      "learning_rate": 7.1627906976744185e-06,
      "loss": 8.369435882568359,
      "step": 155
    },
    {
      "epoch": 0.29836829836829837,
      "grad_norm": 16.741973876953125,
      "learning_rate": 7.395348837209303e-06,
      "loss": 8.280664825439453,
      "step": 160
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 19.79235076904297,
      "learning_rate": 7.627906976744187e-06,
      "loss": 8.200827026367188,
      "step": 165
    },
    {
      "epoch": 0.317016317016317,
      "grad_norm": 32.991554260253906,
      "learning_rate": 7.86046511627907e-06,
      "loss": 7.973581695556641,
      "step": 170
    },
    {
      "epoch": 0.32634032634032634,
      "grad_norm": 17.949445724487305,
      "learning_rate": 8.093023255813955e-06,
      "loss": 8.141377258300782,
      "step": 175
    },
    {
      "epoch": 0.3356643356643357,
      "grad_norm": 16.19559097290039,
      "learning_rate": 8.325581395348837e-06,
      "loss": 7.837094116210937,
      "step": 180
    },
    {
      "epoch": 0.34498834498834496,
      "grad_norm": 23.260622024536133,
      "learning_rate": 8.558139534883722e-06,
      "loss": 7.763467407226562,
      "step": 185
    },
    {
      "epoch": 0.3543123543123543,
      "grad_norm": 24.858335494995117,
      "learning_rate": 8.790697674418606e-06,
      "loss": 7.886874389648438,
      "step": 190
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 18.59360122680664,
      "learning_rate": 9.023255813953489e-06,
      "loss": 7.7624969482421875,
      "step": 195
    },
    {
      "epoch": 0.372960372960373,
      "grad_norm": 18.056495666503906,
      "learning_rate": 9.255813953488373e-06,
      "loss": 7.4621437072753904,
      "step": 200
    },
    {
      "epoch": 0.372960372960373,
      "eval_com_accuracy": 0.144,
      "eval_rew_accuracy": 0.332,
      "step": 200
    },
    {
      "epoch": 0.3822843822843823,
      "grad_norm": 16.496261596679688,
      "learning_rate": 9.488372093023258e-06,
      "loss": 7.526261901855468,
      "step": 205
    },
    {
      "epoch": 0.3916083916083916,
      "grad_norm": 16.77888298034668,
      "learning_rate": 9.72093023255814e-06,
      "loss": 7.624967193603515,
      "step": 210
    },
    {
      "epoch": 0.40093240093240096,
      "grad_norm": 17.096778869628906,
      "learning_rate": 9.953488372093025e-06,
      "loss": 7.337849426269531,
      "step": 215
    },
    {
      "epoch": 0.41025641025641024,
      "grad_norm": 18.3544979095459,
      "learning_rate": 9.999894343918674e-06,
      "loss": 7.354197692871094,
      "step": 220
    },
    {
      "epoch": 0.4195804195804196,
      "grad_norm": 15.287137985229492,
      "learning_rate": 9.999465123741172e-06,
      "loss": 7.085916137695312,
      "step": 225
    },
    {
      "epoch": 0.4289044289044289,
      "grad_norm": 17.03432273864746,
      "learning_rate": 9.998705764284132e-06,
      "loss": 7.059596252441406,
      "step": 230
    },
    {
      "epoch": 0.4382284382284382,
      "grad_norm": 15.470928192138672,
      "learning_rate": 9.99761631569179e-06,
      "loss": 7.118296813964844,
      "step": 235
    },
    {
      "epoch": 0.44755244755244755,
      "grad_norm": 15.732117652893066,
      "learning_rate": 9.9961968499058e-06,
      "loss": 6.981436157226563,
      "step": 240
    },
    {
      "epoch": 0.4568764568764569,
      "grad_norm": 20.92630958557129,
      "learning_rate": 9.994447460660473e-06,
      "loss": 6.96953125,
      "step": 245
    },
    {
      "epoch": 0.4662004662004662,
      "grad_norm": 14.848095893859863,
      "learning_rate": 9.992368263476585e-06,
      "loss": 6.831211090087891,
      "step": 250
    },
    {
      "epoch": 0.4662004662004662,
      "eval_com_accuracy": 0.132,
      "eval_rew_accuracy": 0.312,
      "step": 250
    },
    {
      "epoch": 0.4755244755244755,
      "grad_norm": 18.59755516052246,
      "learning_rate": 9.989959395653756e-06,
      "loss": 6.873024749755859,
      "step": 255
    },
    {
      "epoch": 0.48484848484848486,
      "grad_norm": 15.705364227294922,
      "learning_rate": 9.98722101626138e-06,
      "loss": 6.633563232421875,
      "step": 260
    },
    {
      "epoch": 0.49417249417249415,
      "grad_norm": 39.2294807434082,
      "learning_rate": 9.984153306128124e-06,
      "loss": 6.63948974609375,
      "step": 265
    },
    {
      "epoch": 0.5034965034965035,
      "grad_norm": 13.209358215332031,
      "learning_rate": 9.980756467829977e-06,
      "loss": 6.688297271728516,
      "step": 270
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 14.645393371582031,
      "learning_rate": 9.977030725676887e-06,
      "loss": 6.4885505676269535,
      "step": 275
    },
    {
      "epoch": 0.5221445221445221,
      "grad_norm": 17.008819580078125,
      "learning_rate": 9.972976325697938e-06,
      "loss": 6.78935546875,
      "step": 280
    },
    {
      "epoch": 0.5314685314685315,
      "grad_norm": 16.029518127441406,
      "learning_rate": 9.96859353562511e-06,
      "loss": 6.681609344482422,
      "step": 285
    },
    {
      "epoch": 0.5407925407925408,
      "grad_norm": 13.878348350524902,
      "learning_rate": 9.963882644875594e-06,
      "loss": 6.528868103027344,
      "step": 290
    },
    {
      "epoch": 0.5501165501165501,
      "grad_norm": 13.996464729309082,
      "learning_rate": 9.958843964532683e-06,
      "loss": 6.431344604492187,
      "step": 295
    },
    {
      "epoch": 0.5594405594405595,
      "grad_norm": 15.72003173828125,
      "learning_rate": 9.953477827325229e-06,
      "loss": 6.5927978515625,
      "step": 300
    },
    {
      "epoch": 0.5594405594405595,
      "eval_com_accuracy": 0.148,
      "eval_rew_accuracy": 0.368,
      "step": 300
    },
    {
      "epoch": 0.5687645687645687,
      "grad_norm": 16.416725158691406,
      "learning_rate": 9.947784587605678e-06,
      "loss": 6.411885833740234,
      "step": 305
    },
    {
      "epoch": 0.578088578088578,
      "grad_norm": 14.880790710449219,
      "learning_rate": 9.941764621326655e-06,
      "loss": 6.5298927307128904,
      "step": 310
    },
    {
      "epoch": 0.5874125874125874,
      "grad_norm": 17.828296661376953,
      "learning_rate": 9.935418326016153e-06,
      "loss": 6.045618057250977,
      "step": 315
    },
    {
      "epoch": 0.5967365967365967,
      "grad_norm": 13.30245590209961,
      "learning_rate": 9.928746120751275e-06,
      "loss": 6.1650840759277346,
      "step": 320
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 13.527432441711426,
      "learning_rate": 9.921748446130564e-06,
      "loss": 6.349945831298828,
      "step": 325
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 14.491817474365234,
      "learning_rate": 9.9144257642449e-06,
      "loss": 6.232943725585938,
      "step": 330
    },
    {
      "epoch": 0.6247086247086248,
      "grad_norm": 15.7533597946167,
      "learning_rate": 9.906778558647e-06,
      "loss": 6.564260864257813,
      "step": 335
    },
    {
      "epoch": 0.634032634032634,
      "grad_norm": 16.538602828979492,
      "learning_rate": 9.898807334319471e-06,
      "loss": 6.122506332397461,
      "step": 340
    },
    {
      "epoch": 0.6433566433566433,
      "grad_norm": 16.459688186645508,
      "learning_rate": 9.890512617641474e-06,
      "loss": 6.332431030273438,
      "step": 345
    },
    {
      "epoch": 0.6526806526806527,
      "grad_norm": 14.879058837890625,
      "learning_rate": 9.881894956353963e-06,
      "loss": 6.118003082275391,
      "step": 350
    },
    {
      "epoch": 0.6526806526806527,
      "eval_com_accuracy": 0.164,
      "eval_rew_accuracy": 0.364,
      "step": 350
    },
    {
      "epoch": 0.662004662004662,
      "grad_norm": 16.02516746520996,
      "learning_rate": 9.87295491952351e-06,
      "loss": 6.241584014892578,
      "step": 355
    },
    {
      "epoch": 0.6713286713286714,
      "grad_norm": 14.9837646484375,
      "learning_rate": 9.863693097504733e-06,
      "loss": 6.083467864990235,
      "step": 360
    },
    {
      "epoch": 0.6806526806526807,
      "grad_norm": 14.138517379760742,
      "learning_rate": 9.854110101901308e-06,
      "loss": 6.143560791015625,
      "step": 365
    },
    {
      "epoch": 0.6899766899766899,
      "grad_norm": 12.696243286132812,
      "learning_rate": 9.844206565525585e-06,
      "loss": 6.061418914794922,
      "step": 370
    },
    {
      "epoch": 0.6993006993006993,
      "grad_norm": 15.22846508026123,
      "learning_rate": 9.833983142356792e-06,
      "loss": 6.193332672119141,
      "step": 375
    },
    {
      "epoch": 0.7086247086247086,
      "grad_norm": 15.284857749938965,
      "learning_rate": 9.823440507497863e-06,
      "loss": 6.0868377685546875,
      "step": 380
    },
    {
      "epoch": 0.717948717948718,
      "grad_norm": 14.117541313171387,
      "learning_rate": 9.812579357130848e-06,
      "loss": 6.289948272705078,
      "step": 385
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 12.648367881774902,
      "learning_rate": 9.801400408470943e-06,
      "loss": 6.194795227050781,
      "step": 390
    },
    {
      "epoch": 0.7365967365967366,
      "grad_norm": 12.394021034240723,
      "learning_rate": 9.789904399719124e-06,
      "loss": 6.031880950927734,
      "step": 395
    },
    {
      "epoch": 0.745920745920746,
      "grad_norm": 15.566425323486328,
      "learning_rate": 9.778092090013416e-06,
      "loss": 6.0966644287109375,
      "step": 400
    },
    {
      "epoch": 0.745920745920746,
      "eval_com_accuracy": 0.184,
      "eval_rew_accuracy": 0.364,
      "step": 400
    },
    {
      "epoch": 0.7552447552447552,
      "grad_norm": 14.885455131530762,
      "learning_rate": 9.76596425937874e-06,
      "loss": 6.476510620117187,
      "step": 405
    },
    {
      "epoch": 0.7645687645687645,
      "grad_norm": 16.627117156982422,
      "learning_rate": 9.753521708675426e-06,
      "loss": 5.9605262756347654,
      "step": 410
    },
    {
      "epoch": 0.7738927738927739,
      "grad_norm": 14.517923355102539,
      "learning_rate": 9.740765259546312e-06,
      "loss": 6.1104694366455075,
      "step": 415
    },
    {
      "epoch": 0.7832167832167832,
      "grad_norm": 12.007637977600098,
      "learning_rate": 9.727695754362498e-06,
      "loss": 5.964785766601563,
      "step": 420
    },
    {
      "epoch": 0.7925407925407926,
      "grad_norm": 16.17314338684082,
      "learning_rate": 9.714314056167711e-06,
      "loss": 5.976337432861328,
      "step": 425
    },
    {
      "epoch": 0.8018648018648019,
      "grad_norm": 13.246196746826172,
      "learning_rate": 9.700621048621322e-06,
      "loss": 5.936331939697266,
      "step": 430
    },
    {
      "epoch": 0.8111888111888111,
      "grad_norm": 15.030324935913086,
      "learning_rate": 9.686617635939988e-06,
      "loss": 6.209032821655273,
      "step": 435
    },
    {
      "epoch": 0.8205128205128205,
      "grad_norm": 13.853086471557617,
      "learning_rate": 9.672304742837945e-06,
      "loss": 5.985638427734375,
      "step": 440
    },
    {
      "epoch": 0.8298368298368298,
      "grad_norm": 13.534188270568848,
      "learning_rate": 9.657683314465948e-06,
      "loss": 5.7971649169921875,
      "step": 445
    },
    {
      "epoch": 0.8391608391608392,
      "grad_norm": 26.1879825592041,
      "learning_rate": 9.642754316348846e-06,
      "loss": 5.949985504150391,
      "step": 450
    },
    {
      "epoch": 0.8391608391608392,
      "eval_com_accuracy": 0.176,
      "eval_rew_accuracy": 0.34,
      "step": 450
    },
    {
      "epoch": 0.8484848484848485,
      "grad_norm": 16.848834991455078,
      "learning_rate": 9.627518734321837e-06,
      "loss": 6.1193115234375,
      "step": 455
    },
    {
      "epoch": 0.8578088578088578,
      "grad_norm": 30.765796661376953,
      "learning_rate": 9.61197757446536e-06,
      "loss": 5.74780158996582,
      "step": 460
    },
    {
      "epoch": 0.8671328671328671,
      "grad_norm": 13.27775764465332,
      "learning_rate": 9.596131863038664e-06,
      "loss": 5.994970703125,
      "step": 465
    },
    {
      "epoch": 0.8764568764568764,
      "grad_norm": 14.923792839050293,
      "learning_rate": 9.579982646412039e-06,
      "loss": 5.810473251342773,
      "step": 470
    },
    {
      "epoch": 0.8857808857808858,
      "grad_norm": 14.351722717285156,
      "learning_rate": 9.563530990997707e-06,
      "loss": 6.09692268371582,
      "step": 475
    },
    {
      "epoch": 0.8951048951048951,
      "grad_norm": 11.83583927154541,
      "learning_rate": 9.546777983179421e-06,
      "loss": 5.867227935791016,
      "step": 480
    },
    {
      "epoch": 0.9044289044289044,
      "grad_norm": 19.415884017944336,
      "learning_rate": 9.529724729240712e-06,
      "loss": 5.834615707397461,
      "step": 485
    },
    {
      "epoch": 0.9137529137529138,
      "grad_norm": 12.147261619567871,
      "learning_rate": 9.512372355291838e-06,
      "loss": 6.039286041259766,
      "step": 490
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 12.706372261047363,
      "learning_rate": 9.494722007195427e-06,
      "loss": 6.06304702758789,
      "step": 495
    },
    {
      "epoch": 0.9324009324009324,
      "grad_norm": 12.642041206359863,
      "learning_rate": 9.476774850490803e-06,
      "loss": 5.9729866027832035,
      "step": 500
    },
    {
      "epoch": 0.9324009324009324,
      "eval_com_accuracy": 0.204,
      "eval_rew_accuracy": 0.328,
      "step": 500
    },
    {
      "epoch": 0.9417249417249417,
      "grad_norm": 12.66169548034668,
      "learning_rate": 9.458532070317021e-06,
      "loss": 5.8281097412109375,
      "step": 505
    },
    {
      "epoch": 0.951048951048951,
      "grad_norm": 13.191854476928711,
      "learning_rate": 9.439994871334614e-06,
      "loss": 6.183546447753907,
      "step": 510
    },
    {
      "epoch": 0.9603729603729604,
      "grad_norm": 12.52676010131836,
      "learning_rate": 9.421164477646031e-06,
      "loss": 5.785601043701172,
      "step": 515
    },
    {
      "epoch": 0.9696969696969697,
      "grad_norm": 13.225418090820312,
      "learning_rate": 9.402042132714817e-06,
      "loss": 5.784825134277344,
      "step": 520
    },
    {
      "epoch": 0.9790209790209791,
      "grad_norm": 12.035611152648926,
      "learning_rate": 9.382629099283486e-06,
      "loss": 5.9122272491455075,
      "step": 525
    },
    {
      "epoch": 0.9883449883449883,
      "grad_norm": 12.166351318359375,
      "learning_rate": 9.362926659290149e-06,
      "loss": 5.790053939819336,
      "step": 530
    },
    {
      "epoch": 0.9976689976689976,
      "grad_norm": 13.22154426574707,
      "learning_rate": 9.342936113783855e-06,
      "loss": 5.745006561279297,
      "step": 535
    }
  ],
  "logging_steps": 5,
  "max_steps": 2148,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}