{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.025,
      "grad_norm": 6.393901263952026,
      "learning_rate": 5e-06,
      "loss": 1.1278,
      "step": 10
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.705956397184389,
      "learning_rate": 5e-06,
      "loss": 1.0093,
      "step": 20
    },
    {
      "epoch": 0.075,
      "grad_norm": 1.68971140105189,
      "learning_rate": 5e-06,
      "loss": 0.9776,
      "step": 30
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.2896479687833042,
      "learning_rate": 5e-06,
      "loss": 0.9461,
      "step": 40
    },
    {
      "epoch": 0.125,
      "grad_norm": 2.337387810419128,
      "learning_rate": 5e-06,
      "loss": 0.9341,
      "step": 50
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.882007339408848,
      "learning_rate": 5e-06,
      "loss": 0.9133,
      "step": 60
    },
    {
      "epoch": 0.175,
      "grad_norm": 1.0405536021297026,
      "learning_rate": 5e-06,
      "loss": 0.9034,
      "step": 70
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.1273905617480575,
      "learning_rate": 5e-06,
      "loss": 0.8946,
      "step": 80
    },
    {
      "epoch": 0.225,
      "grad_norm": 0.8899956341459144,
      "learning_rate": 5e-06,
      "loss": 0.8897,
      "step": 90
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7569552721791574,
      "learning_rate": 5e-06,
      "loss": 0.8838,
      "step": 100
    },
    {
      "epoch": 0.275,
      "grad_norm": 0.775187782225504,
      "learning_rate": 5e-06,
      "loss": 0.8804,
      "step": 110
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.7058567462838982,
      "learning_rate": 5e-06,
      "loss": 0.8742,
      "step": 120
    },
    {
      "epoch": 0.325,
      "grad_norm": 1.2857109201084875,
      "learning_rate": 5e-06,
      "loss": 0.8689,
      "step": 130
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.833965423282999,
      "learning_rate": 5e-06,
      "loss": 0.8664,
      "step": 140
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.8452787015074426,
      "learning_rate": 5e-06,
      "loss": 0.8647,
      "step": 150
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.6905100077591924,
      "learning_rate": 5e-06,
      "loss": 0.8614,
      "step": 160
    },
    {
      "epoch": 0.425,
      "grad_norm": 0.6534167059133549,
      "learning_rate": 5e-06,
      "loss": 0.8591,
      "step": 170
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.7163070342801734,
      "learning_rate": 5e-06,
      "loss": 0.8538,
      "step": 180
    },
    {
      "epoch": 0.475,
      "grad_norm": 0.8379014374017074,
      "learning_rate": 5e-06,
      "loss": 0.8519,
      "step": 190
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.7058002653208569,
      "learning_rate": 5e-06,
      "loss": 0.8519,
      "step": 200
    },
    {
      "epoch": 0.525,
      "grad_norm": 0.7433143437252816,
      "learning_rate": 5e-06,
      "loss": 0.8486,
      "step": 210
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.7600103941580408,
      "learning_rate": 5e-06,
      "loss": 0.844,
      "step": 220
    },
    {
      "epoch": 0.575,
      "grad_norm": 0.7419761211751795,
      "learning_rate": 5e-06,
      "loss": 0.8447,
      "step": 230
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.6020153457959969,
      "learning_rate": 5e-06,
      "loss": 0.8448,
      "step": 240
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.5834136183612872,
      "learning_rate": 5e-06,
      "loss": 0.8415,
      "step": 250
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.6861809869702314,
      "learning_rate": 5e-06,
      "loss": 0.8412,
      "step": 260
    },
    {
      "epoch": 0.675,
      "grad_norm": 0.6424298466812792,
      "learning_rate": 5e-06,
      "loss": 0.8446,
      "step": 270
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.7062264581938785,
      "learning_rate": 5e-06,
      "loss": 0.8371,
      "step": 280
    },
    {
      "epoch": 0.725,
      "grad_norm": 0.7413013325508084,
      "learning_rate": 5e-06,
      "loss": 0.8368,
      "step": 290
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6755467656899127,
      "learning_rate": 5e-06,
      "loss": 0.8384,
      "step": 300
    },
    {
      "epoch": 0.775,
      "grad_norm": 0.7186164830975353,
      "learning_rate": 5e-06,
      "loss": 0.8346,
      "step": 310
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5707506171843091,
      "learning_rate": 5e-06,
      "loss": 0.8331,
      "step": 320
    },
    {
      "epoch": 0.825,
      "grad_norm": 0.9112862077459049,
      "learning_rate": 5e-06,
      "loss": 0.8324,
      "step": 330
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.6828679795547313,
      "learning_rate": 5e-06,
      "loss": 0.8354,
      "step": 340
    },
    {
      "epoch": 0.875,
      "grad_norm": 0.6927842206009195,
      "learning_rate": 5e-06,
      "loss": 0.829,
      "step": 350
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.6008408676809543,
      "learning_rate": 5e-06,
      "loss": 0.8307,
      "step": 360
    },
    {
      "epoch": 0.925,
      "grad_norm": 0.6206381764947855,
      "learning_rate": 5e-06,
      "loss": 0.8267,
      "step": 370
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5454612469710348,
      "learning_rate": 5e-06,
      "loss": 0.8247,
      "step": 380
    },
    {
      "epoch": 0.975,
      "grad_norm": 0.5220893940655968,
      "learning_rate": 5e-06,
      "loss": 0.8256,
      "step": 390
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5672238225320995,
      "learning_rate": 5e-06,
      "loss": 0.8272,
      "step": 400
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8262304067611694,
      "eval_runtime": 283.5953,
      "eval_samples_per_second": 38.001,
      "eval_steps_per_second": 0.596,
      "step": 400
    },
    {
      "epoch": 1.025,
      "grad_norm": 1.1102286013620262,
      "learning_rate": 5e-06,
      "loss": 0.7842,
      "step": 410
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.7607811738450544,
      "learning_rate": 5e-06,
      "loss": 0.7849,
      "step": 420
    },
    {
      "epoch": 1.075,
      "grad_norm": 0.7653482131592415,
      "learning_rate": 5e-06,
      "loss": 0.7849,
      "step": 430
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.705031839749241,
      "learning_rate": 5e-06,
      "loss": 0.779,
      "step": 440
    },
    {
      "epoch": 1.125,
      "grad_norm": 0.5370550513294627,
      "learning_rate": 5e-06,
      "loss": 0.7831,
      "step": 450
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.5298114725731297,
      "learning_rate": 5e-06,
      "loss": 0.7818,
      "step": 460
    },
    {
      "epoch": 1.175,
      "grad_norm": 0.6734194289163318,
      "learning_rate": 5e-06,
      "loss": 0.7772,
      "step": 470
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5897636893907213,
      "learning_rate": 5e-06,
      "loss": 0.7813,
      "step": 480
    },
    {
      "epoch": 1.225,
      "grad_norm": 0.604266980297563,
      "learning_rate": 5e-06,
      "loss": 0.7836,
      "step": 490
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.6714899816376678,
      "learning_rate": 5e-06,
      "loss": 0.7801,
      "step": 500
    },
    {
      "epoch": 1.275,
      "grad_norm": 0.8168625249318959,
      "learning_rate": 5e-06,
      "loss": 0.7806,
      "step": 510
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.7932681977775804,
      "learning_rate": 5e-06,
      "loss": 0.7776,
      "step": 520
    },
    {
      "epoch": 1.325,
      "grad_norm": 0.8862643278939233,
      "learning_rate": 5e-06,
      "loss": 0.7796,
      "step": 530
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.7421035460384174,
      "learning_rate": 5e-06,
      "loss": 0.7823,
      "step": 540
    },
    {
      "epoch": 1.375,
      "grad_norm": 0.5962649233174171,
      "learning_rate": 5e-06,
      "loss": 0.7783,
      "step": 550
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.9067010653184415,
      "learning_rate": 5e-06,
      "loss": 0.7755,
      "step": 560
    },
    {
      "epoch": 1.425,
      "grad_norm": 0.9272486884932043,
      "learning_rate": 5e-06,
      "loss": 0.7811,
      "step": 570
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.634573401992294,
      "learning_rate": 5e-06,
      "loss": 0.7783,
      "step": 580
    },
    {
      "epoch": 1.475,
      "grad_norm": 0.7647615867543145,
      "learning_rate": 5e-06,
      "loss": 0.7789,
      "step": 590
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.7186976301425001,
      "learning_rate": 5e-06,
      "loss": 0.7732,
      "step": 600
    },
    {
      "epoch": 1.525,
      "grad_norm": 0.6741640245568876,
      "learning_rate": 5e-06,
      "loss": 0.7803,
      "step": 610
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.7673936498272227,
      "learning_rate": 5e-06,
      "loss": 0.7771,
      "step": 620
    },
    {
      "epoch": 1.575,
      "grad_norm": 0.6762446070662301,
      "learning_rate": 5e-06,
      "loss": 0.7781,
      "step": 630
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.694582981661087,
      "learning_rate": 5e-06,
      "loss": 0.7803,
      "step": 640
    },
    {
      "epoch": 1.625,
      "grad_norm": 0.6911825951192573,
      "learning_rate": 5e-06,
      "loss": 0.7755,
      "step": 650
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.7214035778878081,
      "learning_rate": 5e-06,
      "loss": 0.7784,
      "step": 660
    },
    {
      "epoch": 1.675,
      "grad_norm": 0.565532898038658,
      "learning_rate": 5e-06,
      "loss": 0.7798,
      "step": 670
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.7144730054018653,
      "learning_rate": 5e-06,
      "loss": 0.7779,
      "step": 680
    },
    {
      "epoch": 1.725,
      "grad_norm": 0.68929751507137,
      "learning_rate": 5e-06,
      "loss": 0.7743,
      "step": 690
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.6232088446530882,
      "learning_rate": 5e-06,
      "loss": 0.7764,
      "step": 700
    },
    {
      "epoch": 1.775,
      "grad_norm": 0.5516916387958084,
      "learning_rate": 5e-06,
      "loss": 0.7743,
      "step": 710
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.6556060585205762,
      "learning_rate": 5e-06,
      "loss": 0.7795,
      "step": 720
    },
    {
      "epoch": 1.825,
      "grad_norm": 0.6395664223230667,
      "learning_rate": 5e-06,
      "loss": 0.7727,
      "step": 730
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.7264470243374237,
      "learning_rate": 5e-06,
      "loss": 0.7716,
      "step": 740
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.7348089904512866,
      "learning_rate": 5e-06,
      "loss": 0.7759,
      "step": 750
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.5362704236504767,
      "learning_rate": 5e-06,
      "loss": 0.7725,
      "step": 760
    },
    {
      "epoch": 1.925,
      "grad_norm": 0.6132209869820257,
      "learning_rate": 5e-06,
      "loss": 0.7746,
      "step": 770
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.5674492579544563,
      "learning_rate": 5e-06,
      "loss": 0.7762,
      "step": 780
    },
    {
      "epoch": 1.975,
      "grad_norm": 0.8144733224752085,
      "learning_rate": 5e-06,
      "loss": 0.771,
      "step": 790
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6438437017098873,
      "learning_rate": 5e-06,
      "loss": 0.7753,
      "step": 800
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8092445731163025,
      "eval_runtime": 279.3948,
      "eval_samples_per_second": 38.573,
      "eval_steps_per_second": 0.605,
      "step": 800
    },
    {
      "epoch": 2.025,
      "grad_norm": 0.741980442805714,
      "learning_rate": 5e-06,
      "loss": 0.731,
      "step": 810
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.6348046724923234,
      "learning_rate": 5e-06,
      "loss": 0.7254,
      "step": 820
    },
    {
      "epoch": 2.075,
      "grad_norm": 0.6998662875286445,
      "learning_rate": 5e-06,
      "loss": 0.722,
      "step": 830
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.6662824597359785,
      "learning_rate": 5e-06,
      "loss": 0.7251,
      "step": 840
    },
    {
      "epoch": 2.125,
      "grad_norm": 0.5887524561855076,
      "learning_rate": 5e-06,
      "loss": 0.7329,
      "step": 850
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.773871523528161,
      "learning_rate": 5e-06,
      "loss": 0.7282,
      "step": 860
    },
    {
      "epoch": 2.175,
      "grad_norm": 0.6520605524681873,
      "learning_rate": 5e-06,
      "loss": 0.7283,
      "step": 870
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.6481774807052064,
      "learning_rate": 5e-06,
      "loss": 0.7333,
      "step": 880
    },
    {
      "epoch": 2.225,
      "grad_norm": 0.5980371010517554,
      "learning_rate": 5e-06,
      "loss": 0.7307,
      "step": 890
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.6698106025945666,
      "learning_rate": 5e-06,
      "loss": 0.7276,
      "step": 900
    },
    {
      "epoch": 2.275,
      "grad_norm": 0.6064989242913533,
      "learning_rate": 5e-06,
      "loss": 0.7284,
      "step": 910
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.5694949139445875,
      "learning_rate": 5e-06,
      "loss": 0.7287,
      "step": 920
    },
    {
      "epoch": 2.325,
      "grad_norm": 0.6518761067013964,
      "learning_rate": 5e-06,
      "loss": 0.7272,
      "step": 930
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.6644802771746268,
      "learning_rate": 5e-06,
      "loss": 0.7301,
      "step": 940
    },
    {
      "epoch": 2.375,
      "grad_norm": 0.564358869683258,
      "learning_rate": 5e-06,
      "loss": 0.7333,
      "step": 950
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.6052088075734858,
      "learning_rate": 5e-06,
      "loss": 0.7279,
      "step": 960
    },
    {
      "epoch": 2.425,
      "grad_norm": 0.5588110088845489,
      "learning_rate": 5e-06,
      "loss": 0.7307,
      "step": 970
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.7921721794918313,
      "learning_rate": 5e-06,
      "loss": 0.727,
      "step": 980
    },
    {
      "epoch": 2.475,
      "grad_norm": 0.633397399719071,
      "learning_rate": 5e-06,
      "loss": 0.7315,
      "step": 990
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.5866962766860978,
      "learning_rate": 5e-06,
      "loss": 0.732,
      "step": 1000
    },
    {
      "epoch": 2.525,
      "grad_norm": 0.7798403321881853,
      "learning_rate": 5e-06,
      "loss": 0.7301,
      "step": 1010
    },
    {
      "epoch": 2.55,
      "grad_norm": 0.6522313154237275,
      "learning_rate": 5e-06,
      "loss": 0.727,
      "step": 1020
    },
    {
      "epoch": 2.575,
      "grad_norm": 0.5918752307210033,
      "learning_rate": 5e-06,
      "loss": 0.732,
      "step": 1030
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.7013642379103808,
      "learning_rate": 5e-06,
      "loss": 0.7304,
      "step": 1040
    },
    {
      "epoch": 2.625,
      "grad_norm": 0.6019984161152461,
      "learning_rate": 5e-06,
      "loss": 0.7301,
      "step": 1050
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.6025521914638767,
      "learning_rate": 5e-06,
      "loss": 0.7303,
      "step": 1060
    },
    {
      "epoch": 2.675,
      "grad_norm": 0.8383133859288988,
      "learning_rate": 5e-06,
      "loss": 0.7352,
      "step": 1070
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.7176072744526589,
      "learning_rate": 5e-06,
      "loss": 0.7325,
      "step": 1080
    },
    {
      "epoch": 2.725,
      "grad_norm": 0.6330508627766337,
      "learning_rate": 5e-06,
      "loss": 0.7297,
      "step": 1090
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.708854127664677,
      "learning_rate": 5e-06,
      "loss": 0.7302,
      "step": 1100
    },
    {
      "epoch": 2.775,
      "grad_norm": 0.7080026819402018,
      "learning_rate": 5e-06,
      "loss": 0.7338,
      "step": 1110
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.6486046388056755,
      "learning_rate": 5e-06,
      "loss": 0.7333,
      "step": 1120
    },
    {
      "epoch": 2.825,
      "grad_norm": 0.6098247605462774,
      "learning_rate": 5e-06,
      "loss": 0.7368,
      "step": 1130
    },
    {
      "epoch": 2.85,
      "grad_norm": 0.707595465934189,
      "learning_rate": 5e-06,
      "loss": 0.7315,
      "step": 1140
    },
    {
      "epoch": 2.875,
      "grad_norm": 0.6598068743344158,
      "learning_rate": 5e-06,
      "loss": 0.7328,
      "step": 1150
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.6256793408527994,
      "learning_rate": 5e-06,
      "loss": 0.7335,
      "step": 1160
    },
    {
      "epoch": 2.925,
      "grad_norm": 0.6477061017951039,
      "learning_rate": 5e-06,
      "loss": 0.7296,
      "step": 1170
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.6150717312512685,
      "learning_rate": 5e-06,
      "loss": 0.7336,
      "step": 1180
    },
    {
      "epoch": 2.975,
      "grad_norm": 0.5375035486490041,
      "learning_rate": 5e-06,
      "loss": 0.7316,
      "step": 1190
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6000906442308059,
      "learning_rate": 5e-06,
      "loss": 0.7331,
      "step": 1200
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.8070117235183716,
      "eval_runtime": 271.2171,
      "eval_samples_per_second": 39.736,
      "eval_steps_per_second": 0.623,
      "step": 1200
    },
    {
      "epoch": 3.0,
      "step": 1200,
      "total_flos": 2009835314872320.0,
      "train_loss": 0.7931587568918864,
      "train_runtime": 40308.7538,
      "train_samples_per_second": 15.238,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 10,
  "max_steps": 1200,
  "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": 2009835314872320.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}