{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 58.67970660146699,
  "eval_steps": 2000,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 8e-05,
      "loss": 0.8756,
      "step": 10
    },
    {
      "epoch": 0.39,
      "learning_rate": 8e-05,
      "loss": 0.8401,
      "step": 20
    },
    {
      "epoch": 0.59,
      "learning_rate": 8e-05,
      "loss": 0.8574,
      "step": 30
    },
    {
      "epoch": 0.78,
      "learning_rate": 8e-05,
      "loss": 0.8282,
      "step": 40
    },
    {
      "epoch": 0.98,
      "learning_rate": 8e-05,
      "loss": 0.7797,
      "step": 50
    },
    {
      "epoch": 1.17,
      "learning_rate": 8e-05,
      "loss": 0.7198,
      "step": 60
    },
    {
      "epoch": 1.37,
      "learning_rate": 8e-05,
      "loss": 0.6546,
      "step": 70
    },
    {
      "epoch": 1.56,
      "learning_rate": 8e-05,
      "loss": 0.6477,
      "step": 80
    },
    {
      "epoch": 1.76,
      "learning_rate": 8e-05,
      "loss": 0.6173,
      "step": 90
    },
    {
      "epoch": 1.96,
      "learning_rate": 8e-05,
      "loss": 0.6119,
      "step": 100
    },
    {
      "epoch": 2.15,
      "learning_rate": 8e-05,
      "loss": 0.5535,
      "step": 110
    },
    {
      "epoch": 2.35,
      "learning_rate": 8e-05,
      "loss": 0.4959,
      "step": 120
    },
    {
      "epoch": 2.54,
      "learning_rate": 8e-05,
      "loss": 0.5077,
      "step": 130
    },
    {
      "epoch": 2.74,
      "learning_rate": 8e-05,
      "loss": 0.5021,
      "step": 140
    },
    {
      "epoch": 2.93,
      "learning_rate": 8e-05,
      "loss": 0.4717,
      "step": 150
    },
    {
      "epoch": 3.13,
      "learning_rate": 8e-05,
      "loss": 0.427,
      "step": 160
    },
    {
      "epoch": 3.33,
      "learning_rate": 8e-05,
      "loss": 0.361,
      "step": 170
    },
    {
      "epoch": 3.52,
      "learning_rate": 8e-05,
      "loss": 0.3622,
      "step": 180
    },
    {
      "epoch": 3.72,
      "learning_rate": 8e-05,
      "loss": 0.3425,
      "step": 190
    },
    {
      "epoch": 3.91,
      "learning_rate": 8e-05,
      "loss": 0.3531,
      "step": 200
    },
    {
      "epoch": 4.11,
      "learning_rate": 8e-05,
      "loss": 0.3376,
      "step": 210
    },
    {
      "epoch": 4.3,
      "learning_rate": 8e-05,
      "loss": 0.2356,
      "step": 220
    },
    {
      "epoch": 4.5,
      "learning_rate": 8e-05,
      "loss": 0.2514,
      "step": 230
    },
    {
      "epoch": 4.69,
      "learning_rate": 8e-05,
      "loss": 0.2369,
      "step": 240
    },
    {
      "epoch": 4.89,
      "learning_rate": 8e-05,
      "loss": 0.2485,
      "step": 250
    },
    {
      "epoch": 5.09,
      "learning_rate": 8e-05,
      "loss": 0.2164,
      "step": 260
    },
    {
      "epoch": 5.28,
      "learning_rate": 8e-05,
      "loss": 0.1498,
      "step": 270
    },
    {
      "epoch": 5.48,
      "learning_rate": 8e-05,
      "loss": 0.1495,
      "step": 280
    },
    {
      "epoch": 5.67,
      "learning_rate": 8e-05,
      "loss": 0.166,
      "step": 290
    },
    {
      "epoch": 5.87,
      "learning_rate": 8e-05,
      "loss": 0.2012,
      "step": 300
    },
    {
      "epoch": 6.06,
      "learning_rate": 8e-05,
      "loss": 0.1569,
      "step": 310
    },
    {
      "epoch": 6.26,
      "learning_rate": 8e-05,
      "loss": 0.0878,
      "step": 320
    },
    {
      "epoch": 6.45,
      "learning_rate": 8e-05,
      "loss": 0.0969,
      "step": 330
    },
    {
      "epoch": 6.65,
      "learning_rate": 8e-05,
      "loss": 0.1229,
      "step": 340
    },
    {
      "epoch": 6.85,
      "learning_rate": 8e-05,
      "loss": 0.1199,
      "step": 350
    },
    {
      "epoch": 7.04,
      "learning_rate": 8e-05,
      "loss": 0.0996,
      "step": 360
    },
    {
      "epoch": 7.24,
      "learning_rate": 8e-05,
      "loss": 0.0568,
      "step": 370
    },
    {
      "epoch": 7.43,
      "learning_rate": 8e-05,
      "loss": 0.0792,
      "step": 380
    },
    {
      "epoch": 7.63,
      "learning_rate": 8e-05,
      "loss": 0.0781,
      "step": 390
    },
    {
      "epoch": 7.82,
      "learning_rate": 8e-05,
      "loss": 0.0852,
      "step": 400
    },
    {
      "epoch": 8.02,
      "learning_rate": 8e-05,
      "loss": 0.0611,
      "step": 410
    },
    {
      "epoch": 8.22,
      "learning_rate": 8e-05,
      "loss": 0.0399,
      "step": 420
    },
    {
      "epoch": 8.41,
      "learning_rate": 8e-05,
      "loss": 0.0597,
      "step": 430
    },
    {
      "epoch": 8.61,
      "learning_rate": 8e-05,
      "loss": 0.0502,
      "step": 440
    },
    {
      "epoch": 8.8,
      "learning_rate": 8e-05,
      "loss": 0.0535,
      "step": 450
    },
    {
      "epoch": 9.0,
      "learning_rate": 8e-05,
      "loss": 0.0347,
      "step": 460
    },
    {
      "epoch": 9.19,
      "learning_rate": 8e-05,
      "loss": 0.0323,
      "step": 470
    },
    {
      "epoch": 9.39,
      "learning_rate": 8e-05,
      "loss": 0.039,
      "step": 480
    },
    {
      "epoch": 9.58,
      "learning_rate": 8e-05,
      "loss": 0.0335,
      "step": 490
    },
    {
      "epoch": 9.78,
      "learning_rate": 8e-05,
      "loss": 0.0364,
      "step": 500
    },
    {
      "epoch": 9.98,
      "learning_rate": 8e-05,
      "loss": 0.0268,
      "step": 510
    },
    {
      "epoch": 10.17,
      "learning_rate": 8e-05,
      "loss": 0.0245,
      "step": 520
    },
    {
      "epoch": 10.37,
      "learning_rate": 8e-05,
      "loss": 0.0229,
      "step": 530
    },
    {
      "epoch": 10.56,
      "learning_rate": 8e-05,
      "loss": 0.0256,
      "step": 540
    },
    {
      "epoch": 10.76,
      "learning_rate": 8e-05,
      "loss": 0.0217,
      "step": 550
    },
    {
      "epoch": 10.95,
      "learning_rate": 8e-05,
      "loss": 0.0204,
      "step": 560
    },
    {
      "epoch": 11.15,
      "learning_rate": 8e-05,
      "loss": 0.0173,
      "step": 570
    },
    {
      "epoch": 11.34,
      "learning_rate": 8e-05,
      "loss": 0.0173,
      "step": 580
    },
    {
      "epoch": 11.54,
      "learning_rate": 8e-05,
      "loss": 0.0171,
      "step": 590
    },
    {
      "epoch": 11.74,
      "learning_rate": 8e-05,
      "loss": 0.0123,
      "step": 600
    },
    {
      "epoch": 11.93,
      "learning_rate": 8e-05,
      "loss": 0.0146,
      "step": 610
    },
    {
      "epoch": 12.13,
      "learning_rate": 8e-05,
      "loss": 0.0145,
      "step": 620
    },
    {
      "epoch": 12.32,
      "learning_rate": 8e-05,
      "loss": 0.0156,
      "step": 630
    },
    {
      "epoch": 12.52,
      "learning_rate": 8e-05,
      "loss": 0.0096,
      "step": 640
    },
    {
      "epoch": 12.71,
      "learning_rate": 8e-05,
      "loss": 0.0103,
      "step": 650
    },
    {
      "epoch": 12.91,
      "learning_rate": 8e-05,
      "loss": 0.0104,
      "step": 660
    },
    {
      "epoch": 13.11,
      "learning_rate": 8e-05,
      "loss": 0.0096,
      "step": 670
    },
    {
      "epoch": 13.3,
      "learning_rate": 8e-05,
      "loss": 0.0074,
      "step": 680
    },
    {
      "epoch": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.0095,
      "step": 690
    },
    {
      "epoch": 13.69,
      "learning_rate": 8e-05,
      "loss": 0.0087,
      "step": 700
    },
    {
      "epoch": 13.89,
      "learning_rate": 8e-05,
      "loss": 0.0084,
      "step": 710
    },
    {
      "epoch": 14.08,
      "learning_rate": 8e-05,
      "loss": 0.0081,
      "step": 720
    },
    {
      "epoch": 14.28,
      "learning_rate": 8e-05,
      "loss": 0.0048,
      "step": 730
    },
    {
      "epoch": 14.47,
      "learning_rate": 8e-05,
      "loss": 0.0044,
      "step": 740
    },
    {
      "epoch": 14.67,
      "learning_rate": 8e-05,
      "loss": 0.008,
      "step": 750
    },
    {
      "epoch": 14.87,
      "learning_rate": 8e-05,
      "loss": 0.0067,
      "step": 760
    },
    {
      "epoch": 15.06,
      "learning_rate": 8e-05,
      "loss": 0.0065,
      "step": 770
    },
    {
      "epoch": 15.26,
      "learning_rate": 8e-05,
      "loss": 0.0061,
      "step": 780
    },
    {
      "epoch": 15.45,
      "learning_rate": 8e-05,
      "loss": 0.0074,
      "step": 790
    },
    {
      "epoch": 15.65,
      "learning_rate": 8e-05,
      "loss": 0.0062,
      "step": 800
    },
    {
      "epoch": 15.84,
      "learning_rate": 8e-05,
      "loss": 0.008,
      "step": 810
    },
    {
      "epoch": 16.04,
      "learning_rate": 8e-05,
      "loss": 0.0071,
      "step": 820
    },
    {
      "epoch": 16.23,
      "learning_rate": 8e-05,
      "loss": 0.0055,
      "step": 830
    },
    {
      "epoch": 16.43,
      "learning_rate": 8e-05,
      "loss": 0.0057,
      "step": 840
    },
    {
      "epoch": 16.63,
      "learning_rate": 8e-05,
      "loss": 0.0073,
      "step": 850
    },
    {
      "epoch": 16.82,
      "learning_rate": 8e-05,
      "loss": 0.0066,
      "step": 860
    },
    {
      "epoch": 17.02,
      "learning_rate": 8e-05,
      "loss": 0.0067,
      "step": 870
    },
    {
      "epoch": 17.21,
      "learning_rate": 8e-05,
      "loss": 0.0039,
      "step": 880
    },
    {
      "epoch": 17.41,
      "learning_rate": 8e-05,
      "loss": 0.0062,
      "step": 890
    },
    {
      "epoch": 17.6,
      "learning_rate": 8e-05,
      "loss": 0.0062,
      "step": 900
    },
    {
      "epoch": 17.8,
      "learning_rate": 8e-05,
      "loss": 0.0065,
      "step": 910
    },
    {
      "epoch": 18.0,
      "learning_rate": 8e-05,
      "loss": 0.0047,
      "step": 920
    },
    {
      "epoch": 18.19,
      "learning_rate": 8e-05,
      "loss": 0.0051,
      "step": 930
    },
    {
      "epoch": 18.39,
      "learning_rate": 8e-05,
      "loss": 0.0044,
      "step": 940
    },
    {
      "epoch": 18.58,
      "learning_rate": 8e-05,
      "loss": 0.0041,
      "step": 950
    },
    {
      "epoch": 18.78,
      "learning_rate": 8e-05,
      "loss": 0.0039,
      "step": 960
    },
    {
      "epoch": 18.97,
      "learning_rate": 8e-05,
      "loss": 0.0027,
      "step": 970
    },
    {
      "epoch": 19.17,
      "learning_rate": 8e-05,
      "loss": 0.0046,
      "step": 980
    },
    {
      "epoch": 19.36,
      "learning_rate": 8e-05,
      "loss": 0.0051,
      "step": 990
    },
    {
      "epoch": 19.56,
      "learning_rate": 8e-05,
      "loss": 0.0071,
      "step": 1000
    },
    {
      "epoch": 19.76,
      "learning_rate": 8e-05,
      "loss": 0.0064,
      "step": 1010
    },
    {
      "epoch": 19.95,
      "learning_rate": 8e-05,
      "loss": 0.0065,
      "step": 1020
    },
    {
      "epoch": 20.15,
      "learning_rate": 8e-05,
      "loss": 0.0056,
      "step": 1030
    },
    {
      "epoch": 20.34,
      "learning_rate": 8e-05,
      "loss": 0.0073,
      "step": 1040
    },
    {
      "epoch": 20.54,
      "learning_rate": 8e-05,
      "loss": 0.0078,
      "step": 1050
    },
    {
      "epoch": 20.73,
      "learning_rate": 8e-05,
      "loss": 0.0062,
      "step": 1060
    },
    {
      "epoch": 20.93,
      "learning_rate": 8e-05,
      "loss": 0.0045,
      "step": 1070
    },
    {
      "epoch": 21.12,
      "learning_rate": 8e-05,
      "loss": 0.0048,
      "step": 1080
    },
    {
      "epoch": 21.32,
      "learning_rate": 8e-05,
      "loss": 0.0036,
      "step": 1090
    },
    {
      "epoch": 21.52,
      "learning_rate": 8e-05,
      "loss": 0.005,
      "step": 1100
    },
    {
      "epoch": 21.71,
      "learning_rate": 8e-05,
      "loss": 0.0048,
      "step": 1110
    },
    {
      "epoch": 21.91,
      "learning_rate": 8e-05,
      "loss": 0.007,
      "step": 1120
    },
    {
      "epoch": 22.1,
      "learning_rate": 8e-05,
      "loss": 0.0061,
      "step": 1130
    },
    {
      "epoch": 22.3,
      "learning_rate": 8e-05,
      "loss": 0.0086,
      "step": 1140
    },
    {
      "epoch": 22.49,
      "learning_rate": 8e-05,
      "loss": 0.0049,
      "step": 1150
    },
    {
      "epoch": 22.69,
      "learning_rate": 8e-05,
      "loss": 0.008,
      "step": 1160
    },
    {
      "epoch": 22.89,
      "learning_rate": 8e-05,
      "loss": 0.0058,
      "step": 1170
    },
    {
      "epoch": 23.08,
      "learning_rate": 8e-05,
      "loss": 0.0078,
      "step": 1180
    },
    {
      "epoch": 23.28,
      "learning_rate": 8e-05,
      "loss": 0.0051,
      "step": 1190
    },
    {
      "epoch": 23.47,
      "learning_rate": 8e-05,
      "loss": 0.0051,
      "step": 1200
    },
    {
      "epoch": 23.67,
      "learning_rate": 8e-05,
      "loss": 0.0061,
      "step": 1210
    },
    {
      "epoch": 23.86,
      "learning_rate": 8e-05,
      "loss": 0.0072,
      "step": 1220
    },
    {
      "epoch": 24.06,
      "learning_rate": 8e-05,
      "loss": 0.0048,
      "step": 1230
    },
    {
      "epoch": 24.25,
      "learning_rate": 8e-05,
      "loss": 0.0041,
      "step": 1240
    },
    {
      "epoch": 24.45,
      "learning_rate": 8e-05,
      "loss": 0.0036,
      "step": 1250
    },
    {
      "epoch": 24.65,
      "learning_rate": 8e-05,
      "loss": 0.0043,
      "step": 1260
    },
    {
      "epoch": 24.84,
      "learning_rate": 8e-05,
      "loss": 0.0046,
      "step": 1270
    },
    {
      "epoch": 25.04,
      "learning_rate": 8e-05,
      "loss": 0.0055,
      "step": 1280
    },
    {
      "epoch": 25.23,
      "learning_rate": 8e-05,
      "loss": 0.0026,
      "step": 1290
    },
    {
      "epoch": 25.43,
      "learning_rate": 8e-05,
      "loss": 0.0043,
      "step": 1300
    },
    {
      "epoch": 25.62,
      "learning_rate": 8e-05,
      "loss": 0.0044,
      "step": 1310
    },
    {
      "epoch": 25.82,
      "learning_rate": 8e-05,
      "loss": 0.0041,
      "step": 1320
    },
    {
      "epoch": 26.01,
      "learning_rate": 8e-05,
      "loss": 0.0041,
      "step": 1330
    },
    {
      "epoch": 26.21,
      "learning_rate": 8e-05,
      "loss": 0.003,
      "step": 1340
    },
    {
      "epoch": 26.41,
      "learning_rate": 8e-05,
      "loss": 0.0021,
      "step": 1350
    },
    {
      "epoch": 26.6,
      "learning_rate": 8e-05,
      "loss": 0.0028,
      "step": 1360
    },
    {
      "epoch": 26.8,
      "learning_rate": 8e-05,
      "loss": 0.0036,
      "step": 1370
    },
    {
      "epoch": 26.99,
      "learning_rate": 8e-05,
      "loss": 0.0036,
      "step": 1380
    },
    {
      "epoch": 27.19,
      "learning_rate": 8e-05,
      "loss": 0.0024,
      "step": 1390
    },
    {
      "epoch": 27.38,
      "learning_rate": 8e-05,
      "loss": 0.0031,
      "step": 1400
    },
    {
      "epoch": 27.58,
      "learning_rate": 8e-05,
      "loss": 0.0021,
      "step": 1410
    },
    {
      "epoch": 27.78,
      "learning_rate": 8e-05,
      "loss": 0.0032,
      "step": 1420
    },
    {
      "epoch": 27.97,
      "learning_rate": 8e-05,
      "loss": 0.0024,
      "step": 1430
    },
    {
      "epoch": 28.17,
      "learning_rate": 8e-05,
      "loss": 0.0027,
      "step": 1440
    },
    {
      "epoch": 28.36,
      "learning_rate": 8e-05,
      "loss": 0.0014,
      "step": 1450
    },
    {
      "epoch": 28.56,
      "learning_rate": 8e-05,
      "loss": 0.0022,
      "step": 1460
    },
    {
      "epoch": 28.75,
      "learning_rate": 8e-05,
      "loss": 0.0025,
      "step": 1470
    },
    {
      "epoch": 28.95,
      "learning_rate": 8e-05,
      "loss": 0.0021,
      "step": 1480
    },
    {
      "epoch": 29.14,
      "learning_rate": 8e-05,
      "loss": 0.0024,
      "step": 1490
    },
    {
      "epoch": 29.34,
      "learning_rate": 8e-05,
      "loss": 0.0024,
      "step": 1500
    },
    {
      "epoch": 29.54,
      "learning_rate": 8e-05,
      "loss": 0.0023,
      "step": 1510
    },
    {
      "epoch": 29.73,
      "learning_rate": 8e-05,
      "loss": 0.0015,
      "step": 1520
    },
    {
      "epoch": 29.93,
      "learning_rate": 8e-05,
      "loss": 0.0013,
      "step": 1530
    },
    {
      "epoch": 30.12,
      "learning_rate": 8e-05,
      "loss": 0.0018,
      "step": 1540
    },
    {
      "epoch": 30.32,
      "learning_rate": 8e-05,
      "loss": 0.001,
      "step": 1550
    },
    {
      "epoch": 30.51,
      "learning_rate": 8e-05,
      "loss": 0.0019,
      "step": 1560
    },
    {
      "epoch": 30.71,
      "learning_rate": 8e-05,
      "loss": 0.0014,
      "step": 1570
    },
    {
      "epoch": 30.9,
      "learning_rate": 8e-05,
      "loss": 0.0019,
      "step": 1580
    },
    {
      "epoch": 31.1,
      "learning_rate": 8e-05,
      "loss": 0.0031,
      "step": 1590
    },
    {
      "epoch": 31.3,
      "learning_rate": 8e-05,
      "loss": 0.0014,
      "step": 1600
    },
    {
      "epoch": 31.49,
      "learning_rate": 8e-05,
      "loss": 0.0007,
      "step": 1610
    },
    {
      "epoch": 31.69,
      "learning_rate": 8e-05,
      "loss": 0.0016,
      "step": 1620
    },
    {
      "epoch": 31.88,
      "learning_rate": 8e-05,
      "loss": 0.0018,
      "step": 1630
    },
    {
      "epoch": 32.08,
      "learning_rate": 8e-05,
      "loss": 0.0017,
      "step": 1640
    },
    {
      "epoch": 32.27,
      "learning_rate": 8e-05,
      "loss": 0.0011,
      "step": 1650
    },
    {
      "epoch": 32.47,
      "learning_rate": 8e-05,
      "loss": 0.0021,
      "step": 1660
    },
    {
      "epoch": 32.67,
      "learning_rate": 8e-05,
      "loss": 0.0023,
      "step": 1670
    },
    {
      "epoch": 32.86,
      "learning_rate": 8e-05,
      "loss": 0.0013,
      "step": 1680
    },
    {
      "epoch": 33.06,
      "learning_rate": 8e-05,
      "loss": 0.002,
      "step": 1690
    },
    {
      "epoch": 33.25,
      "learning_rate": 8e-05,
      "loss": 0.0017,
      "step": 1700
    },
    {
      "epoch": 33.45,
      "learning_rate": 8e-05,
      "loss": 0.0034,
      "step": 1710
    },
    {
      "epoch": 33.64,
      "learning_rate": 8e-05,
      "loss": 0.0024,
      "step": 1720
    },
    {
      "epoch": 33.84,
      "learning_rate": 8e-05,
      "loss": 0.003,
      "step": 1730
    },
    {
      "epoch": 34.03,
      "learning_rate": 8e-05,
      "loss": 0.0036,
      "step": 1740
    },
    {
      "epoch": 34.23,
      "learning_rate": 8e-05,
      "loss": 0.0029,
      "step": 1750
    },
    {
      "epoch": 34.43,
      "learning_rate": 8e-05,
      "loss": 0.0035,
      "step": 1760
    },
    {
      "epoch": 34.62,
      "learning_rate": 8e-05,
      "loss": 0.004,
      "step": 1770
    },
    {
      "epoch": 34.82,
      "learning_rate": 8e-05,
      "loss": 0.0035,
      "step": 1780
    },
    {
      "epoch": 35.01,
      "learning_rate": 8e-05,
      "loss": 0.0051,
      "step": 1790
    },
    {
      "epoch": 35.21,
      "learning_rate": 8e-05,
      "loss": 0.0038,
      "step": 1800
    },
    {
      "epoch": 35.4,
      "learning_rate": 8e-05,
      "loss": 0.0035,
      "step": 1810
    },
    {
      "epoch": 35.6,
      "learning_rate": 8e-05,
      "loss": 0.004,
      "step": 1820
    },
    {
      "epoch": 35.79,
      "learning_rate": 8e-05,
      "loss": 0.0037,
      "step": 1830
    },
    {
      "epoch": 35.99,
      "learning_rate": 8e-05,
      "loss": 0.0055,
      "step": 1840
    },
    {
      "epoch": 36.19,
      "learning_rate": 8e-05,
      "loss": 0.0112,
      "step": 1850
    },
    {
      "epoch": 36.38,
      "learning_rate": 8e-05,
      "loss": 0.0031,
      "step": 1860
    },
    {
      "epoch": 36.58,
      "learning_rate": 8e-05,
      "loss": 0.0044,
      "step": 1870
    },
    {
      "epoch": 36.77,
      "learning_rate": 8e-05,
      "loss": 0.0057,
      "step": 1880
    },
    {
      "epoch": 36.97,
      "learning_rate": 8e-05,
      "loss": 0.0083,
      "step": 1890
    },
    {
      "epoch": 37.16,
      "learning_rate": 8e-05,
      "loss": 0.0043,
      "step": 1900
    },
    {
      "epoch": 37.36,
      "learning_rate": 8e-05,
      "loss": 0.0047,
      "step": 1910
    },
    {
      "epoch": 37.56,
      "learning_rate": 8e-05,
      "loss": 0.0048,
      "step": 1920
    },
    {
      "epoch": 37.75,
      "learning_rate": 8e-05,
      "loss": 0.0068,
      "step": 1930
    },
    {
      "epoch": 37.95,
      "learning_rate": 8e-05,
      "loss": 0.0068,
      "step": 1940
    },
    {
      "epoch": 38.14,
      "learning_rate": 8e-05,
      "loss": 0.006,
      "step": 1950
    },
    {
      "epoch": 38.34,
      "learning_rate": 8e-05,
      "loss": 0.0069,
      "step": 1960
    },
    {
      "epoch": 38.53,
      "learning_rate": 8e-05,
      "loss": 0.0069,
      "step": 1970
    },
    {
      "epoch": 38.73,
      "learning_rate": 8e-05,
      "loss": 0.005,
      "step": 1980
    },
    {
      "epoch": 38.92,
      "learning_rate": 8e-05,
      "loss": 0.0084,
      "step": 1990
    },
    {
      "epoch": 39.12,
      "learning_rate": 8e-05,
      "loss": 0.0067,
      "step": 2000
    },
    {
      "epoch": 39.12,
      "eval_loss": 0.8936046361923218,
      "eval_runtime": 0.3385,
      "eval_samples_per_second": 2.954,
      "eval_steps_per_second": 2.954,
      "step": 2000
    },
    {
      "epoch": 39.32,
      "learning_rate": 8e-05,
      "loss": 0.0051,
      "step": 2010
    },
    {
      "epoch": 39.51,
      "learning_rate": 8e-05,
      "loss": 0.0059,
      "step": 2020
    },
    {
      "epoch": 39.71,
      "learning_rate": 8e-05,
      "loss": 0.0069,
      "step": 2030
    },
    {
      "epoch": 39.9,
      "learning_rate": 8e-05,
      "loss": 0.0076,
      "step": 2040
    },
    {
      "epoch": 40.1,
      "learning_rate": 8e-05,
      "loss": 0.0061,
      "step": 2050
    },
    {
      "epoch": 40.29,
      "learning_rate": 8e-05,
      "loss": 0.0075,
      "step": 2060
    },
    {
      "epoch": 40.49,
      "learning_rate": 8e-05,
      "loss": 0.0066,
      "step": 2070
    },
    {
      "epoch": 40.68,
      "learning_rate": 8e-05,
      "loss": 0.0081,
      "step": 2080
    },
    {
      "epoch": 40.88,
      "learning_rate": 8e-05,
      "loss": 0.0062,
      "step": 2090
    },
    {
      "epoch": 41.08,
      "learning_rate": 8e-05,
      "loss": 0.0059,
      "step": 2100
    },
    {
      "epoch": 41.27,
      "learning_rate": 8e-05,
      "loss": 0.0035,
      "step": 2110
    },
    {
      "epoch": 41.47,
      "learning_rate": 8e-05,
      "loss": 0.0032,
      "step": 2120
    },
    {
      "epoch": 41.66,
      "learning_rate": 8e-05,
      "loss": 0.0061,
      "step": 2130
    },
    {
      "epoch": 41.86,
      "learning_rate": 8e-05,
      "loss": 0.0048,
      "step": 2140
    },
    {
      "epoch": 42.05,
      "learning_rate": 8e-05,
      "loss": 0.0079,
      "step": 2150
    },
    {
      "epoch": 42.25,
      "learning_rate": 8e-05,
      "loss": 0.0055,
      "step": 2160
    },
    {
      "epoch": 42.44,
      "learning_rate": 8e-05,
      "loss": 0.0038,
      "step": 2170
    },
    {
      "epoch": 42.64,
      "learning_rate": 8e-05,
      "loss": 0.0047,
      "step": 2180
    },
    {
      "epoch": 42.84,
      "learning_rate": 8e-05,
      "loss": 0.0032,
      "step": 2190
    },
    {
      "epoch": 43.03,
      "learning_rate": 8e-05,
      "loss": 0.0073,
      "step": 2200
    },
    {
      "epoch": 43.23,
      "learning_rate": 8e-05,
      "loss": 0.004,
      "step": 2210
    },
    {
      "epoch": 43.42,
      "learning_rate": 8e-05,
      "loss": 0.0058,
      "step": 2220
    },
    {
      "epoch": 43.62,
      "learning_rate": 8e-05,
      "loss": 0.0042,
      "step": 2230
    },
    {
      "epoch": 43.81,
      "learning_rate": 8e-05,
      "loss": 0.0045,
      "step": 2240
    },
    {
      "epoch": 44.01,
      "learning_rate": 8e-05,
      "loss": 0.0033,
      "step": 2250
    },
    {
      "epoch": 44.21,
      "learning_rate": 8e-05,
      "loss": 0.0025,
      "step": 2260
    },
    {
      "epoch": 44.4,
      "learning_rate": 8e-05,
      "loss": 0.0051,
      "step": 2270
    },
    {
      "epoch": 44.6,
      "learning_rate": 8e-05,
      "loss": 0.0049,
      "step": 2280
    },
    {
      "epoch": 44.79,
      "learning_rate": 8e-05,
      "loss": 0.004,
      "step": 2290
    },
    {
      "epoch": 44.99,
      "learning_rate": 8e-05,
      "loss": 0.0035,
      "step": 2300
    },
    {
      "epoch": 45.18,
      "learning_rate": 8e-05,
      "loss": 0.0023,
      "step": 2310
    },
    {
      "epoch": 45.38,
      "learning_rate": 8e-05,
      "loss": 0.0036,
      "step": 2320
    },
    {
      "epoch": 45.57,
      "learning_rate": 8e-05,
      "loss": 0.0035,
      "step": 2330
    },
    {
      "epoch": 45.77,
      "learning_rate": 8e-05,
      "loss": 0.0043,
      "step": 2340
    },
    {
      "epoch": 45.97,
      "learning_rate": 8e-05,
      "loss": 0.0041,
      "step": 2350
    },
    {
      "epoch": 46.16,
      "learning_rate": 8e-05,
      "loss": 0.0032,
      "step": 2360
    },
    {
      "epoch": 46.36,
      "learning_rate": 8e-05,
      "loss": 0.0026,
      "step": 2370
    },
    {
      "epoch": 46.55,
      "learning_rate": 8e-05,
      "loss": 0.0055,
      "step": 2380
    },
    {
      "epoch": 46.75,
      "learning_rate": 8e-05,
      "loss": 0.0038,
      "step": 2390
    },
    {
      "epoch": 46.94,
      "learning_rate": 8e-05,
      "loss": 0.0038,
      "step": 2400
    },
    {
      "epoch": 47.14,
      "learning_rate": 8e-05,
      "loss": 0.0017,
      "step": 2410
    },
    {
      "epoch": 47.33,
      "learning_rate": 8e-05,
      "loss": 0.002,
      "step": 2420
    },
    {
      "epoch": 47.53,
      "learning_rate": 8e-05,
      "loss": 0.0021,
      "step": 2430
    },
    {
      "epoch": 47.73,
      "learning_rate": 8e-05,
      "loss": 0.0018,
      "step": 2440
    },
    {
      "epoch": 47.92,
      "learning_rate": 8e-05,
      "loss": 0.002,
      "step": 2450
    },
    {
      "epoch": 48.12,
      "learning_rate": 8e-05,
      "loss": 0.003,
      "step": 2460
    },
    {
      "epoch": 48.31,
      "learning_rate": 8e-05,
      "loss": 0.0023,
      "step": 2470
    },
    {
      "epoch": 48.51,
      "learning_rate": 8e-05,
      "loss": 0.0019,
      "step": 2480
    },
    {
      "epoch": 48.7,
      "learning_rate": 8e-05,
      "loss": 0.0025,
      "step": 2490
    },
    {
      "epoch": 48.9,
      "learning_rate": 8e-05,
      "loss": 0.0017,
      "step": 2500
    },
    {
      "epoch": 49.1,
      "learning_rate": 8e-05,
      "loss": 0.0014,
      "step": 2510
    },
    {
      "epoch": 49.29,
      "learning_rate": 8e-05,
      "loss": 0.0015,
      "step": 2520
    },
    {
      "epoch": 49.49,
      "learning_rate": 8e-05,
      "loss": 0.0012,
      "step": 2530
    },
    {
      "epoch": 49.68,
      "learning_rate": 8e-05,
      "loss": 0.0024,
      "step": 2540
    },
    {
      "epoch": 49.88,
      "learning_rate": 8e-05,
      "loss": 0.0019,
      "step": 2550
    },
    {
      "epoch": 50.07,
      "learning_rate": 8e-05,
      "loss": 0.0013,
      "step": 2560
    },
    {
      "epoch": 50.27,
      "learning_rate": 8e-05,
      "loss": 0.0023,
      "step": 2570
    },
    {
      "epoch": 50.46,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2580
    },
    {
      "epoch": 50.66,
      "learning_rate": 8e-05,
      "loss": 0.0013,
      "step": 2590
    },
    {
      "epoch": 50.86,
      "learning_rate": 8e-05,
      "loss": 0.0012,
      "step": 2600
    },
    {
      "epoch": 51.05,
      "learning_rate": 8e-05,
      "loss": 0.0025,
      "step": 2610
    },
    {
      "epoch": 51.25,
      "learning_rate": 8e-05,
      "loss": 0.002,
      "step": 2620
    },
    {
      "epoch": 51.44,
      "learning_rate": 8e-05,
      "loss": 0.0013,
      "step": 2630
    },
    {
      "epoch": 51.64,
      "learning_rate": 8e-05,
      "loss": 0.0018,
      "step": 2640
    },
    {
      "epoch": 51.83,
      "learning_rate": 8e-05,
      "loss": 0.0015,
      "step": 2650
    },
    {
      "epoch": 52.03,
      "learning_rate": 8e-05,
      "loss": 0.0011,
      "step": 2660
    },
    {
      "epoch": 52.22,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2670
    },
    {
      "epoch": 52.42,
      "learning_rate": 8e-05,
      "loss": 0.0014,
      "step": 2680
    },
    {
      "epoch": 52.62,
      "learning_rate": 8e-05,
      "loss": 0.0012,
      "step": 2690
    },
    {
      "epoch": 52.81,
      "learning_rate": 8e-05,
      "loss": 0.0011,
      "step": 2700
    },
    {
      "epoch": 53.01,
      "learning_rate": 8e-05,
      "loss": 0.0019,
      "step": 2710
    },
    {
      "epoch": 53.2,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2720
    },
    {
      "epoch": 53.4,
      "learning_rate": 8e-05,
      "loss": 0.0013,
      "step": 2730
    },
    {
      "epoch": 53.59,
      "learning_rate": 8e-05,
      "loss": 0.0009,
      "step": 2740
    },
    {
      "epoch": 53.79,
      "learning_rate": 8e-05,
      "loss": 0.001,
      "step": 2750
    },
    {
      "epoch": 53.99,
      "learning_rate": 8e-05,
      "loss": 0.001,
      "step": 2760
    },
    {
      "epoch": 54.18,
      "learning_rate": 8e-05,
      "loss": 0.0009,
      "step": 2770
    },
    {
      "epoch": 54.38,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2780
    },
    {
      "epoch": 54.57,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2790
    },
    {
      "epoch": 54.77,
      "learning_rate": 8e-05,
      "loss": 0.0004,
      "step": 2800
    },
    {
      "epoch": 54.96,
      "learning_rate": 8e-05,
      "loss": 0.0009,
      "step": 2810
    },
    {
      "epoch": 55.16,
      "learning_rate": 8e-05,
      "loss": 0.0004,
      "step": 2820
    },
    {
      "epoch": 55.35,
      "learning_rate": 8e-05,
      "loss": 0.0005,
      "step": 2830
    },
    {
      "epoch": 55.55,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2840
    },
    {
      "epoch": 55.75,
      "learning_rate": 8e-05,
      "loss": 0.0009,
      "step": 2850
    },
    {
      "epoch": 55.94,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2860
    },
    {
      "epoch": 56.14,
      "learning_rate": 8e-05,
      "loss": 0.0006,
      "step": 2870
    },
    {
      "epoch": 56.33,
      "learning_rate": 8e-05,
      "loss": 0.0006,
      "step": 2880
    },
    {
      "epoch": 56.53,
      "learning_rate": 8e-05,
      "loss": 0.0005,
      "step": 2890
    },
    {
      "epoch": 56.72,
      "learning_rate": 8e-05,
      "loss": 0.0003,
      "step": 2900
    },
    {
      "epoch": 56.92,
      "learning_rate": 8e-05,
      "loss": 0.0026,
      "step": 2910
    },
    {
      "epoch": 57.11,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2920
    },
    {
      "epoch": 57.31,
      "learning_rate": 8e-05,
      "loss": 0.0008,
      "step": 2930
    },
    {
      "epoch": 57.51,
      "learning_rate": 8e-05,
      "loss": 0.0016,
      "step": 2940
    },
    {
      "epoch": 57.7,
      "learning_rate": 8e-05,
      "loss": 0.0004,
      "step": 2950
    },
    {
      "epoch": 57.9,
      "learning_rate": 8e-05,
      "loss": 0.0003,
      "step": 2960
    },
    {
      "epoch": 58.09,
      "learning_rate": 8e-05,
      "loss": 0.0007,
      "step": 2970
    },
    {
      "epoch": 58.29,
      "learning_rate": 8e-05,
      "loss": 0.0004,
      "step": 2980
    },
    {
      "epoch": 58.48,
      "learning_rate": 8e-05,
      "loss": 0.0002,
      "step": 2990
    },
    {
      "epoch": 58.68,
      "learning_rate": 8e-05,
      "loss": 0.0007,
      "step": 3000
    }
  ],
  "logging_steps": 10,
  "max_steps": 100000,
  "num_train_epochs": 1961,
  "save_steps": 500,
  "total_flos": 2.34588386134997e+18,
  "trial_name": null,
  "trial_params": null
}