{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 11.849372384937238,
  "eval_steps": 500,
  "global_step": 708,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "grad_norm": 5.769251346588135,
      "learning_rate": 4.9975392245612254e-05,
      "loss": 2.1374,
      "step": 10
    },
    {
      "epoch": 0.33,
      "grad_norm": 5.128970623016357,
      "learning_rate": 4.9901617425775067e-05,
      "loss": 0.4576,
      "step": 20
    },
    {
      "epoch": 0.5,
      "grad_norm": 4.582453727722168,
      "learning_rate": 4.9778820775100185e-05,
      "loss": 0.2497,
      "step": 30
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.4966882467269897,
      "learning_rate": 4.9607244033573156e-05,
      "loss": 0.1604,
      "step": 40
    },
    {
      "epoch": 0.84,
      "grad_norm": 3.9401819705963135,
      "learning_rate": 4.93872249706591e-05,
      "loss": 0.1206,
      "step": 50
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.5695924758911133,
      "learning_rate": 4.91191967203629e-05,
      "loss": 0.1109,
      "step": 60
    },
    {
      "epoch": 1.17,
      "grad_norm": 1.77476167678833,
      "learning_rate": 4.8803686928552736e-05,
      "loss": 0.057,
      "step": 70
    },
    {
      "epoch": 1.34,
      "grad_norm": 2.8595950603485107,
      "learning_rate": 4.84413167142257e-05,
      "loss": 0.0569,
      "step": 80
    },
    {
      "epoch": 1.51,
      "grad_norm": 3.9191925525665283,
      "learning_rate": 4.803279944676032e-05,
      "loss": 0.0634,
      "step": 90
    },
    {
      "epoch": 1.67,
      "grad_norm": 2.3463213443756104,
      "learning_rate": 4.7578939341563095e-05,
      "loss": 0.0634,
      "step": 100
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.22609496116638184,
      "learning_rate": 4.70806298768736e-05,
      "loss": 0.0437,
      "step": 110
    },
    {
      "epoch": 2.01,
      "grad_norm": 2.9491143226623535,
      "learning_rate": 4.653885203484515e-05,
      "loss": 0.0627,
      "step": 120
    },
    {
      "epoch": 2.18,
      "grad_norm": 1.0097993612289429,
      "learning_rate": 4.595467237036329e-05,
      "loss": 0.0361,
      "step": 130
    },
    {
      "epoch": 2.34,
      "grad_norm": 2.7911064624786377,
      "learning_rate": 4.532924091140417e-05,
      "loss": 0.0267,
      "step": 140
    },
    {
      "epoch": 2.51,
      "grad_norm": 1.8213822841644287,
      "learning_rate": 4.466378889506607e-05,
      "loss": 0.0362,
      "step": 150
    },
    {
      "epoch": 2.68,
      "grad_norm": 0.4782440960407257,
      "learning_rate": 4.395962634373097e-05,
      "loss": 0.0403,
      "step": 160
    },
    {
      "epoch": 2.85,
      "grad_norm": 1.6825015544891357,
      "learning_rate": 4.3218139486127854e-05,
      "loss": 0.0302,
      "step": 170
    },
    {
      "epoch": 3.01,
      "grad_norm": 0.773872971534729,
      "learning_rate": 4.2440788028374624e-05,
      "loss": 0.048,
      "step": 180
    },
    {
      "epoch": 3.18,
      "grad_norm": 0.590453565120697,
      "learning_rate": 4.1629102280370904e-05,
      "loss": 0.0219,
      "step": 190
    },
    {
      "epoch": 3.35,
      "grad_norm": 0.30445072054862976,
      "learning_rate": 4.0784680143198836e-05,
      "loss": 0.0215,
      "step": 200
    },
    {
      "epoch": 3.51,
      "grad_norm": 0.2872461676597595,
      "learning_rate": 3.990918396346254e-05,
      "loss": 0.0264,
      "step": 210
    },
    {
      "epoch": 3.68,
      "grad_norm": 0.9185941219329834,
      "learning_rate": 3.900433726075865e-05,
      "loss": 0.029,
      "step": 220
    },
    {
      "epoch": 3.85,
      "grad_norm": 1.8721823692321777,
      "learning_rate": 3.8071921334720696e-05,
      "loss": 0.031,
      "step": 230
    },
    {
      "epoch": 4.02,
      "grad_norm": 0.34600716829299927,
      "learning_rate": 3.711377175831626e-05,
      "loss": 0.0205,
      "step": 240
    },
    {
      "epoch": 4.18,
      "grad_norm": 0.4261312484741211,
      "learning_rate": 3.613177476430079e-05,
      "loss": 0.0172,
      "step": 250
    },
    {
      "epoch": 4.35,
      "grad_norm": 1.4647105932235718,
      "learning_rate": 3.512786353194134e-05,
      "loss": 0.0162,
      "step": 260
    },
    {
      "epoch": 4.52,
      "grad_norm": 0.12821082770824432,
      "learning_rate": 3.410401438132056e-05,
      "loss": 0.0136,
      "step": 270
    },
    {
      "epoch": 4.69,
      "grad_norm": 0.6188161373138428,
      "learning_rate": 3.3062242882712724e-05,
      "loss": 0.0186,
      "step": 280
    },
    {
      "epoch": 4.85,
      "grad_norm": 0.3787616789340973,
      "learning_rate": 3.200459988869111e-05,
      "loss": 0.0147,
      "step": 290
    },
    {
      "epoch": 5.02,
      "grad_norm": 0.27198734879493713,
      "learning_rate": 3.093316749677788e-05,
      "loss": 0.0117,
      "step": 300
    },
    {
      "epoch": 5.19,
      "grad_norm": 0.38122686743736267,
      "learning_rate": 2.985005495058446e-05,
      "loss": 0.0099,
      "step": 310
    },
    {
      "epoch": 5.36,
      "grad_norm": 0.4904286861419678,
      "learning_rate": 2.875739448751176e-05,
      "loss": 0.013,
      "step": 320
    },
    {
      "epoch": 5.52,
      "grad_norm": 5.21450662612915,
      "learning_rate": 2.7657337141184138e-05,
      "loss": 0.0178,
      "step": 330
    },
    {
      "epoch": 5.69,
      "grad_norm": 0.09181027114391327,
      "learning_rate": 2.655204850688085e-05,
      "loss": 0.006,
      "step": 340
    },
    {
      "epoch": 5.86,
      "grad_norm": 0.7764429450035095,
      "learning_rate": 2.5443704478301154e-05,
      "loss": 0.0132,
      "step": 350
    },
    {
      "epoch": 6.03,
      "grad_norm": 0.3942791819572449,
      "learning_rate": 2.433448696405563e-05,
      "loss": 0.0086,
      "step": 360
    },
    {
      "epoch": 6.19,
      "grad_norm": 0.32931485772132874,
      "learning_rate": 2.3226579592316538e-05,
      "loss": 0.0073,
      "step": 370
    },
    {
      "epoch": 6.36,
      "grad_norm": 0.0025393522810190916,
      "learning_rate": 2.2122163412082927e-05,
      "loss": 0.0068,
      "step": 380
    },
    {
      "epoch": 6.53,
      "grad_norm": 0.17522938549518585,
      "learning_rate": 2.1023412599523204e-05,
      "loss": 0.0055,
      "step": 390
    },
    {
      "epoch": 6.69,
      "grad_norm": 0.07497023046016693,
      "learning_rate": 1.993249017784766e-05,
      "loss": 0.0104,
      "step": 400
    },
    {
      "epoch": 6.86,
      "grad_norm": 0.34479033946990967,
      "learning_rate": 1.8851543759137007e-05,
      "loss": 0.0086,
      "step": 410
    },
    {
      "epoch": 7.03,
      "grad_norm": 0.4811187982559204,
      "learning_rate": 1.778270131650948e-05,
      "loss": 0.0077,
      "step": 420
    },
    {
      "epoch": 7.2,
      "grad_norm": 0.33907610177993774,
      "learning_rate": 1.672806699494966e-05,
      "loss": 0.0028,
      "step": 430
    },
    {
      "epoch": 7.36,
      "grad_norm": 0.12760841846466064,
      "learning_rate": 1.5689716969045848e-05,
      "loss": 0.0049,
      "step": 440
    },
    {
      "epoch": 7.53,
      "grad_norm": 0.10119038820266724,
      "learning_rate": 1.4669695355790552e-05,
      "loss": 0.0041,
      "step": 450
    },
    {
      "epoch": 7.7,
      "grad_norm": 0.1975838541984558,
      "learning_rate": 1.3670010190490073e-05,
      "loss": 0.0051,
      "step": 460
    },
    {
      "epoch": 7.87,
      "grad_norm": 0.0027128455694764853,
      "learning_rate": 1.2692629473705453e-05,
      "loss": 0.0062,
      "step": 470
    },
    {
      "epoch": 8.03,
      "grad_norm": 0.45006152987480164,
      "learning_rate": 1.173947729700644e-05,
      "loss": 0.0036,
      "step": 480
    },
    {
      "epoch": 8.2,
      "grad_norm": 0.0031273181084543467,
      "learning_rate": 1.081243005516571e-05,
      "loss": 0.0027,
      "step": 490
    },
    {
      "epoch": 8.37,
      "grad_norm": 0.11013814806938171,
      "learning_rate": 9.913312752249903e-06,
      "loss": 0.0025,
      "step": 500
    },
    {
      "epoch": 8.54,
      "grad_norm": 0.16879281401634216,
      "learning_rate": 9.043895408879505e-06,
      "loss": 0.0029,
      "step": 510
    },
    {
      "epoch": 8.7,
      "grad_norm": 0.4711211323738098,
      "learning_rate": 8.20588957773018e-06,
      "loss": 0.0026,
      "step": 520
    },
    {
      "epoch": 8.87,
      "grad_norm": 0.4484475553035736,
      "learning_rate": 7.400944974135427e-06,
      "loss": 0.0029,
      "step": 530
    },
    {
      "epoch": 9.04,
      "grad_norm": 0.0020081661641597748,
      "learning_rate": 6.6306462284233234e-06,
      "loss": 0.0022,
      "step": 540
    },
    {
      "epoch": 9.21,
      "grad_norm": 0.051797155290842056,
      "learning_rate": 5.896509766381028e-06,
      "loss": 0.0011,
      "step": 550
    },
    {
      "epoch": 9.37,
      "grad_norm": 0.05627438426017761,
      "learning_rate": 5.199980823988157e-06,
      "loss": 0.0016,
      "step": 560
    },
    {
      "epoch": 9.54,
      "grad_norm": 0.054639093577861786,
      "learning_rate": 4.542430602295774e-06,
      "loss": 0.0012,
      "step": 570
    },
    {
      "epoch": 9.71,
      "grad_norm": 0.09268685430288315,
      "learning_rate": 3.925153568052123e-06,
      "loss": 0.0011,
      "step": 580
    },
    {
      "epoch": 9.87,
      "grad_norm": 0.3072189688682556,
      "learning_rate": 3.3493649053890326e-06,
      "loss": 0.0018,
      "step": 590
    },
    {
      "epoch": 10.04,
      "grad_norm": 0.002750764600932598,
      "learning_rate": 2.8161981235857143e-06,
      "loss": 0.001,
      "step": 600
    },
    {
      "epoch": 10.21,
      "grad_norm": 0.14409799873828888,
      "learning_rate": 2.3267028256193036e-06,
      "loss": 0.0011,
      "step": 610
    },
    {
      "epoch": 10.38,
      "grad_norm": 0.0033888304606080055,
      "learning_rate": 1.881842641895104e-06,
      "loss": 0.0014,
      "step": 620
    },
    {
      "epoch": 10.54,
      "grad_norm": 0.05172213539481163,
      "learning_rate": 1.4824933332241692e-06,
      "loss": 0.0004,
      "step": 630
    },
    {
      "epoch": 10.71,
      "grad_norm": 0.26695069670677185,
      "learning_rate": 1.129441066782702e-06,
      "loss": 0.001,
      "step": 640
    },
    {
      "epoch": 10.88,
      "grad_norm": 0.06269415467977524,
      "learning_rate": 8.233808684473959e-07,
      "loss": 0.0008,
      "step": 650
    },
    {
      "epoch": 11.05,
      "grad_norm": 0.1800147444009781,
      "learning_rate": 5.649152545533332e-07,
      "loss": 0.0013,
      "step": 660
    },
    {
      "epoch": 11.21,
      "grad_norm": 0.033635422587394714,
      "learning_rate": 3.5455304576806347e-07,
      "loss": 0.0008,
      "step": 670
    },
    {
      "epoch": 11.38,
      "grad_norm": 0.011946323327720165,
      "learning_rate": 1.927083654168854e-07,
      "loss": 0.0004,
      "step": 680
    },
    {
      "epoch": 11.55,
      "grad_norm": 0.1268051713705063,
      "learning_rate": 7.969982423124689e-08,
      "loss": 0.0007,
      "step": 690
    },
    {
      "epoch": 11.72,
      "grad_norm": 0.20600733160972595,
      "learning_rate": 1.5749893125160954e-08,
      "loss": 0.0011,
      "step": 700
    },
    {
      "epoch": 11.85,
      "step": 708,
      "total_flos": 2.9001892500996096e+16,
      "train_loss": 0.05958240834198955,
      "train_runtime": 2189.2208,
      "train_samples_per_second": 2.62,
      "train_steps_per_second": 0.323
    }
  ],
  "logging_steps": 10,
  "max_steps": 708,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 12,
  "save_steps": 1000,
  "total_flos": 2.9001892500996096e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}