{
  "best_metric": 0.049902621656656265,
  "best_model_checkpoint": "/content/drive/MyDrive/Hkth_data/WNS Hackathon/model/checkpoint-1400",
  "epoch": 3.953488372093023,
  "eval_steps": 100,
  "global_step": 1700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.00019895348837209301,
      "loss": 0.3296,
      "step": 10
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019779069767441863,
      "loss": 0.3506,
      "step": 20
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001966279069767442,
      "loss": 0.3365,
      "step": 30
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001954651162790698,
      "loss": 0.1444,
      "step": 40
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019430232558139535,
      "loss": 0.2242,
      "step": 50
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019313953488372094,
      "loss": 0.1912,
      "step": 60
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019197674418604653,
      "loss": 0.1604,
      "step": 70
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001908139534883721,
      "loss": 0.2392,
      "step": 80
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001896511627906977,
      "loss": 0.1543,
      "step": 90
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00018848837209302325,
      "loss": 0.3238,
      "step": 100
    },
    {
      "epoch": 0.23,
      "eval_accuracy": 0.9471947194719472,
      "eval_loss": 0.1719779074192047,
      "eval_runtime": 418.2527,
      "eval_samples_per_second": 2.898,
      "eval_steps_per_second": 0.363,
      "step": 100
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00018732558139534884,
      "loss": 0.2097,
      "step": 110
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00018616279069767443,
      "loss": 0.1673,
      "step": 120
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018500000000000002,
      "loss": 0.2386,
      "step": 130
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001838372093023256,
      "loss": 0.3442,
      "step": 140
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018267441860465118,
      "loss": 0.2316,
      "step": 150
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018151162790697674,
      "loss": 0.1904,
      "step": 160
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00018034883720930233,
      "loss": 0.1186,
      "step": 170
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017918604651162793,
      "loss": 0.145,
      "step": 180
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001780232558139535,
      "loss": 0.1418,
      "step": 190
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017686046511627908,
      "loss": 0.1078,
      "step": 200
    },
    {
      "epoch": 0.47,
      "eval_accuracy": 0.9562706270627063,
      "eval_loss": 0.14491133391857147,
      "eval_runtime": 43.7761,
      "eval_samples_per_second": 27.686,
      "eval_steps_per_second": 3.472,
      "step": 200
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017569767441860464,
      "loss": 0.0595,
      "step": 210
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017453488372093023,
      "loss": 0.0677,
      "step": 220
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017337209302325583,
      "loss": 0.1667,
      "step": 230
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00017220930232558142,
      "loss": 0.199,
      "step": 240
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017104651162790698,
      "loss": 0.2125,
      "step": 250
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00016988372093023257,
      "loss": 0.2511,
      "step": 260
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00016872093023255816,
      "loss": 0.1656,
      "step": 270
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016755813953488373,
      "loss": 0.1384,
      "step": 280
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00016639534883720932,
      "loss": 0.0628,
      "step": 290
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00016534883720930235,
      "loss": 0.2622,
      "step": 300
    },
    {
      "epoch": 0.7,
      "eval_accuracy": 0.9249174917491749,
      "eval_loss": 0.18467029929161072,
      "eval_runtime": 43.6859,
      "eval_samples_per_second": 27.743,
      "eval_steps_per_second": 3.479,
      "step": 300
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001641860465116279,
      "loss": 0.1592,
      "step": 310
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001630232558139535,
      "loss": 0.1944,
      "step": 320
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00016186046511627907,
      "loss": 0.0672,
      "step": 330
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016069767441860466,
      "loss": 0.0579,
      "step": 340
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015953488372093025,
      "loss": 0.0437,
      "step": 350
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001583720930232558,
      "loss": 0.308,
      "step": 360
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001572093023255814,
      "loss": 0.2108,
      "step": 370
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00015604651162790697,
      "loss": 0.1127,
      "step": 380
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00015488372093023256,
      "loss": 0.1465,
      "step": 390
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015372093023255815,
      "loss": 0.0433,
      "step": 400
    },
    {
      "epoch": 0.93,
      "eval_accuracy": 0.9702970297029703,
      "eval_loss": 0.110055111348629,
      "eval_runtime": 44.1236,
      "eval_samples_per_second": 27.468,
      "eval_steps_per_second": 3.445,
      "step": 400
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015255813953488374,
      "loss": 0.0939,
      "step": 410
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001513953488372093,
      "loss": 0.2253,
      "step": 420
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001502325581395349,
      "loss": 0.0982,
      "step": 430
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001490697674418605,
      "loss": 0.0136,
      "step": 440
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014790697674418605,
      "loss": 0.0822,
      "step": 450
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00014674418604651164,
      "loss": 0.0479,
      "step": 460
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001455813953488372,
      "loss": 0.1399,
      "step": 470
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001444186046511628,
      "loss": 0.0975,
      "step": 480
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00014325581395348836,
      "loss": 0.0308,
      "step": 490
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014209302325581395,
      "loss": 0.0962,
      "step": 500
    },
    {
      "epoch": 1.16,
      "eval_accuracy": 0.9702970297029703,
      "eval_loss": 0.09571434557437897,
      "eval_runtime": 44.8921,
      "eval_samples_per_second": 26.998,
      "eval_steps_per_second": 3.386,
      "step": 500
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014093023255813954,
      "loss": 0.12,
      "step": 510
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00013976744186046513,
      "loss": 0.0626,
      "step": 520
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013860465116279072,
      "loss": 0.0878,
      "step": 530
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001374418604651163,
      "loss": 0.1058,
      "step": 540
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00013627906976744188,
      "loss": 0.0938,
      "step": 550
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00013511627906976744,
      "loss": 0.0142,
      "step": 560
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00013395348837209303,
      "loss": 0.0696,
      "step": 570
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001327906976744186,
      "loss": 0.0015,
      "step": 580
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001316279069767442,
      "loss": 0.0736,
      "step": 590
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00013046511627906975,
      "loss": 0.0479,
      "step": 600
    },
    {
      "epoch": 1.4,
      "eval_accuracy": 0.9678217821782178,
      "eval_loss": 0.1245083436369896,
      "eval_runtime": 44.2549,
      "eval_samples_per_second": 27.387,
      "eval_steps_per_second": 3.435,
      "step": 600
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012930232558139534,
      "loss": 0.0414,
      "step": 610
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00012813953488372093,
      "loss": 0.0234,
      "step": 620
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00012697674418604653,
      "loss": 0.0148,
      "step": 630
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012581395348837212,
      "loss": 0.0723,
      "step": 640
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012465116279069768,
      "loss": 0.1016,
      "step": 650
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012348837209302327,
      "loss": 0.0057,
      "step": 660
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012232558139534883,
      "loss": 0.0512,
      "step": 670
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012116279069767443,
      "loss": 0.0263,
      "step": 680
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00012,
      "loss": 0.0562,
      "step": 690
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00011883720930232558,
      "loss": 0.0594,
      "step": 700
    },
    {
      "epoch": 1.63,
      "eval_accuracy": 0.985973597359736,
      "eval_loss": 0.06257853657007217,
      "eval_runtime": 44.5792,
      "eval_samples_per_second": 27.188,
      "eval_steps_per_second": 3.41,
      "step": 700
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011767441860465116,
      "loss": 0.0027,
      "step": 710
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00011651162790697674,
      "loss": 0.0164,
      "step": 720
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00011534883720930234,
      "loss": 0.0406,
      "step": 730
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011418604651162792,
      "loss": 0.0317,
      "step": 740
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0001130232558139535,
      "loss": 0.0539,
      "step": 750
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011186046511627907,
      "loss": 0.0158,
      "step": 760
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011069767441860466,
      "loss": 0.1054,
      "step": 770
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010953488372093024,
      "loss": 0.0112,
      "step": 780
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010837209302325582,
      "loss": 0.019,
      "step": 790
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001072093023255814,
      "loss": 0.027,
      "step": 800
    },
    {
      "epoch": 1.86,
      "eval_accuracy": 0.985973597359736,
      "eval_loss": 0.056669652462005615,
      "eval_runtime": 44.7355,
      "eval_samples_per_second": 27.093,
      "eval_steps_per_second": 3.398,
      "step": 800
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010604651162790697,
      "loss": 0.0206,
      "step": 810
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010488372093023255,
      "loss": 0.0296,
      "step": 820
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010372093023255815,
      "loss": 0.0775,
      "step": 830
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010255813953488373,
      "loss": 0.0278,
      "step": 840
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010139534883720931,
      "loss": 0.0091,
      "step": 850
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0001002325581395349,
      "loss": 0.0019,
      "step": 860
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.906976744186048e-05,
      "loss": 0.0014,
      "step": 870
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.790697674418605e-05,
      "loss": 0.0044,
      "step": 880
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.674418604651163e-05,
      "loss": 0.0002,
      "step": 890
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.558139534883721e-05,
      "loss": 0.0003,
      "step": 900
    },
    {
      "epoch": 2.09,
      "eval_accuracy": 0.985973597359736,
      "eval_loss": 0.0684327706694603,
      "eval_runtime": 44.3211,
      "eval_samples_per_second": 27.346,
      "eval_steps_per_second": 3.43,
      "step": 900
    },
    {
      "epoch": 2.12,
      "learning_rate": 9.441860465116279e-05,
      "loss": 0.0243,
      "step": 910
    },
    {
      "epoch": 2.14,
      "learning_rate": 9.325581395348838e-05,
      "loss": 0.0383,
      "step": 920
    },
    {
      "epoch": 2.16,
      "learning_rate": 9.209302325581396e-05,
      "loss": 0.0046,
      "step": 930
    },
    {
      "epoch": 2.19,
      "learning_rate": 9.093023255813953e-05,
      "loss": 0.0955,
      "step": 940
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.976744186046512e-05,
      "loss": 0.0122,
      "step": 950
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.86046511627907e-05,
      "loss": 0.0197,
      "step": 960
    },
    {
      "epoch": 2.26,
      "learning_rate": 8.744186046511629e-05,
      "loss": 0.0197,
      "step": 970
    },
    {
      "epoch": 2.28,
      "learning_rate": 8.627906976744187e-05,
      "loss": 0.0025,
      "step": 980
    },
    {
      "epoch": 2.3,
      "learning_rate": 8.511627906976745e-05,
      "loss": 0.0001,
      "step": 990
    },
    {
      "epoch": 2.33,
      "learning_rate": 8.395348837209302e-05,
      "loss": 0.0272,
      "step": 1000
    },
    {
      "epoch": 2.33,
      "eval_accuracy": 0.9851485148514851,
      "eval_loss": 0.07581602782011032,
      "eval_runtime": 44.565,
      "eval_samples_per_second": 27.196,
      "eval_steps_per_second": 3.411,
      "step": 1000
    },
    {
      "epoch": 2.35,
      "learning_rate": 8.27906976744186e-05,
      "loss": 0.0516,
      "step": 1010
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.162790697674419e-05,
      "loss": 0.001,
      "step": 1020
    },
    {
      "epoch": 2.4,
      "learning_rate": 8.046511627906977e-05,
      "loss": 0.0027,
      "step": 1030
    },
    {
      "epoch": 2.42,
      "learning_rate": 7.930232558139535e-05,
      "loss": 0.0002,
      "step": 1040
    },
    {
      "epoch": 2.44,
      "learning_rate": 7.813953488372094e-05,
      "loss": 0.0683,
      "step": 1050
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.697674418604652e-05,
      "loss": 0.0016,
      "step": 1060
    },
    {
      "epoch": 2.49,
      "learning_rate": 7.58139534883721e-05,
      "loss": 0.0195,
      "step": 1070
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.465116279069768e-05,
      "loss": 0.0818,
      "step": 1080
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.348837209302326e-05,
      "loss": 0.0016,
      "step": 1090
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.232558139534884e-05,
      "loss": 0.0275,
      "step": 1100
    },
    {
      "epoch": 2.56,
      "eval_accuracy": 0.9876237623762376,
      "eval_loss": 0.06026723235845566,
      "eval_runtime": 44.5699,
      "eval_samples_per_second": 27.193,
      "eval_steps_per_second": 3.41,
      "step": 1100
    },
    {
      "epoch": 2.58,
      "learning_rate": 7.116279069767442e-05,
      "loss": 0.0005,
      "step": 1110
    },
    {
      "epoch": 2.6,
      "learning_rate": 7e-05,
      "loss": 0.0017,
      "step": 1120
    },
    {
      "epoch": 2.63,
      "learning_rate": 6.883720930232558e-05,
      "loss": 0.0004,
      "step": 1130
    },
    {
      "epoch": 2.65,
      "learning_rate": 6.767441860465116e-05,
      "loss": 0.0023,
      "step": 1140
    },
    {
      "epoch": 2.67,
      "learning_rate": 6.651162790697675e-05,
      "loss": 0.0004,
      "step": 1150
    },
    {
      "epoch": 2.7,
      "learning_rate": 6.534883720930233e-05,
      "loss": 0.0017,
      "step": 1160
    },
    {
      "epoch": 2.72,
      "learning_rate": 6.418604651162791e-05,
      "loss": 0.0005,
      "step": 1170
    },
    {
      "epoch": 2.74,
      "learning_rate": 6.30232558139535e-05,
      "loss": 0.0001,
      "step": 1180
    },
    {
      "epoch": 2.77,
      "learning_rate": 6.186046511627908e-05,
      "loss": 0.0013,
      "step": 1190
    },
    {
      "epoch": 2.79,
      "learning_rate": 6.0697674418604654e-05,
      "loss": 0.0007,
      "step": 1200
    },
    {
      "epoch": 2.79,
      "eval_accuracy": 0.9900990099009901,
      "eval_loss": 0.05105312541127205,
      "eval_runtime": 44.7068,
      "eval_samples_per_second": 27.11,
      "eval_steps_per_second": 3.4,
      "step": 1200
    },
    {
      "epoch": 2.81,
      "learning_rate": 5.953488372093024e-05,
      "loss": 0.0002,
      "step": 1210
    },
    {
      "epoch": 2.84,
      "learning_rate": 5.8372093023255815e-05,
      "loss": 0.0002,
      "step": 1220
    },
    {
      "epoch": 2.86,
      "learning_rate": 5.720930232558139e-05,
      "loss": 0.0024,
      "step": 1230
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.6046511627906984e-05,
      "loss": 0.0009,
      "step": 1240
    },
    {
      "epoch": 2.91,
      "learning_rate": 5.488372093023256e-05,
      "loss": 0.0002,
      "step": 1250
    },
    {
      "epoch": 2.93,
      "learning_rate": 5.3720930232558145e-05,
      "loss": 0.0497,
      "step": 1260
    },
    {
      "epoch": 2.95,
      "learning_rate": 5.255813953488372e-05,
      "loss": 0.0006,
      "step": 1270
    },
    {
      "epoch": 2.98,
      "learning_rate": 5.13953488372093e-05,
      "loss": 0.0004,
      "step": 1280
    },
    {
      "epoch": 3.0,
      "learning_rate": 5.023255813953489e-05,
      "loss": 0.0001,
      "step": 1290
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.906976744186046e-05,
      "loss": 0.0001,
      "step": 1300
    },
    {
      "epoch": 3.02,
      "eval_accuracy": 0.9900990099009901,
      "eval_loss": 0.061022818088531494,
      "eval_runtime": 44.3547,
      "eval_samples_per_second": 27.325,
      "eval_steps_per_second": 3.427,
      "step": 1300
    },
    {
      "epoch": 3.05,
      "learning_rate": 4.790697674418605e-05,
      "loss": 0.0241,
      "step": 1310
    },
    {
      "epoch": 3.07,
      "learning_rate": 4.674418604651163e-05,
      "loss": 0.0005,
      "step": 1320
    },
    {
      "epoch": 3.09,
      "learning_rate": 4.5581395348837214e-05,
      "loss": 0.0003,
      "step": 1330
    },
    {
      "epoch": 3.12,
      "learning_rate": 4.441860465116279e-05,
      "loss": 0.0002,
      "step": 1340
    },
    {
      "epoch": 3.14,
      "learning_rate": 4.325581395348837e-05,
      "loss": 0.0006,
      "step": 1350
    },
    {
      "epoch": 3.16,
      "learning_rate": 4.209302325581396e-05,
      "loss": 0.0001,
      "step": 1360
    },
    {
      "epoch": 3.19,
      "learning_rate": 4.093023255813954e-05,
      "loss": 0.0001,
      "step": 1370
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.9767441860465115e-05,
      "loss": 0.0005,
      "step": 1380
    },
    {
      "epoch": 3.23,
      "learning_rate": 3.86046511627907e-05,
      "loss": 0.0001,
      "step": 1390
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.7441860465116276e-05,
      "loss": 0.0001,
      "step": 1400
    },
    {
      "epoch": 3.26,
      "eval_accuracy": 0.990924092409241,
      "eval_loss": 0.049902621656656265,
      "eval_runtime": 44.1915,
      "eval_samples_per_second": 27.426,
      "eval_steps_per_second": 3.44,
      "step": 1400
    },
    {
      "epoch": 3.28,
      "learning_rate": 3.627906976744187e-05,
      "loss": 0.0002,
      "step": 1410
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.5116279069767445e-05,
      "loss": 0.0001,
      "step": 1420
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.395348837209302e-05,
      "loss": 0.0001,
      "step": 1430
    },
    {
      "epoch": 3.35,
      "learning_rate": 3.2790697674418606e-05,
      "loss": 0.0001,
      "step": 1440
    },
    {
      "epoch": 3.37,
      "learning_rate": 3.162790697674419e-05,
      "loss": 0.0001,
      "step": 1450
    },
    {
      "epoch": 3.4,
      "learning_rate": 3.0465116279069768e-05,
      "loss": 0.0001,
      "step": 1460
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.9302325581395352e-05,
      "loss": 0.0001,
      "step": 1470
    },
    {
      "epoch": 3.44,
      "learning_rate": 2.813953488372093e-05,
      "loss": 0.0001,
      "step": 1480
    },
    {
      "epoch": 3.47,
      "learning_rate": 2.6976744186046517e-05,
      "loss": 0.0002,
      "step": 1490
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.5813953488372094e-05,
      "loss": 0.0001,
      "step": 1500
    },
    {
      "epoch": 3.49,
      "eval_accuracy": 0.990924092409241,
      "eval_loss": 0.05676552653312683,
      "eval_runtime": 44.3782,
      "eval_samples_per_second": 27.311,
      "eval_steps_per_second": 3.425,
      "step": 1500
    },
    {
      "epoch": 3.51,
      "learning_rate": 2.4651162790697675e-05,
      "loss": 0.0001,
      "step": 1510
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.3488372093023256e-05,
      "loss": 0.0,
      "step": 1520
    },
    {
      "epoch": 3.56,
      "learning_rate": 2.2325581395348837e-05,
      "loss": 0.0,
      "step": 1530
    },
    {
      "epoch": 3.58,
      "learning_rate": 2.116279069767442e-05,
      "loss": 0.0001,
      "step": 1540
    },
    {
      "epoch": 3.6,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1550
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.8837209302325582e-05,
      "loss": 0.0001,
      "step": 1560
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.7674418604651163e-05,
      "loss": 0.0,
      "step": 1570
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.6511627906976744e-05,
      "loss": 0.0,
      "step": 1580
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.5348837209302328e-05,
      "loss": 0.0004,
      "step": 1590
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.4186046511627907e-05,
      "loss": 0.0001,
      "step": 1600
    },
    {
      "epoch": 3.72,
      "eval_accuracy": 0.9917491749174917,
      "eval_loss": 0.05982881784439087,
      "eval_runtime": 44.2442,
      "eval_samples_per_second": 27.393,
      "eval_steps_per_second": 3.435,
      "step": 1600
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.3023255813953488e-05,
      "loss": 0.0001,
      "step": 1610
    },
    {
      "epoch": 3.77,
      "learning_rate": 1.186046511627907e-05,
      "loss": 0.0055,
      "step": 1620
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.0697674418604651e-05,
      "loss": 0.0001,
      "step": 1630
    },
    {
      "epoch": 3.81,
      "learning_rate": 9.534883720930234e-06,
      "loss": 0.0,
      "step": 1640
    },
    {
      "epoch": 3.84,
      "learning_rate": 8.372093023255815e-06,
      "loss": 0.0001,
      "step": 1650
    },
    {
      "epoch": 3.86,
      "learning_rate": 7.209302325581396e-06,
      "loss": 0.0001,
      "step": 1660
    },
    {
      "epoch": 3.88,
      "learning_rate": 6.046511627906977e-06,
      "loss": 0.0,
      "step": 1670
    },
    {
      "epoch": 3.91,
      "learning_rate": 4.883720930232559e-06,
      "loss": 0.0,
      "step": 1680
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.72093023255814e-06,
      "loss": 0.0,
      "step": 1690
    },
    {
      "epoch": 3.95,
      "learning_rate": 2.558139534883721e-06,
      "loss": 0.0,
      "step": 1700
    },
    {
      "epoch": 3.95,
      "eval_accuracy": 0.990924092409241,
      "eval_loss": 0.05301262065768242,
      "eval_runtime": 44.0827,
      "eval_samples_per_second": 27.494,
      "eval_steps_per_second": 3.448,
      "step": 1700
    }
  ],
  "logging_steps": 10,
  "max_steps": 1720,
  "num_train_epochs": 4,
  "save_steps": 100,
  "total_flos": 6.205770439364837e+18,
  "trial_name": null,
  "trial_params": null
}