{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.9890764647467725,
  "global_step": 2510,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.0,
      "loss": 0.0956,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0,
      "loss": 0.1428,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0,
      "loss": 0.0953,
      "step": 30
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0,
      "loss": 0.0613,
      "step": 40
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0,
      "loss": 0.0052,
      "step": 50
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0,
      "loss": 0.0532,
      "step": 60
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0,
      "loss": 0.0105,
      "step": 70
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0,
      "loss": 0.0314,
      "step": 80
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0,
      "loss": 0.0145,
      "step": 90
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0,
      "loss": 0.0738,
      "step": 100
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0,
      "loss": 0.0689,
      "step": 110
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0,
      "loss": 0.0173,
      "step": 120
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0,
      "loss": 0.047,
      "step": 130
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0,
      "loss": 0.1202,
      "step": 140
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0,
      "loss": 0.1458,
      "step": 150
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0,
      "loss": 0.0388,
      "step": 160
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0,
      "loss": 0.0518,
      "step": 170
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0,
      "loss": 0.1362,
      "step": 180
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0,
      "loss": 0.0068,
      "step": 190
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0,
      "loss": 0.0802,
      "step": 200
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0,
      "loss": 0.0025,
      "step": 210
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0,
      "loss": 0.002,
      "step": 220
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0,
      "loss": 0.03,
      "step": 230
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0,
      "loss": 0.0193,
      "step": 240
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0,
      "loss": 0.051,
      "step": 250
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0,
      "loss": 0.1223,
      "step": 260
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0,
      "loss": 0.1329,
      "step": 270
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0,
      "loss": 0.0536,
      "step": 280
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0,
      "loss": 0.1244,
      "step": 290
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0,
      "loss": 0.0772,
      "step": 300
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0,
      "loss": 0.0393,
      "step": 310
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0,
      "loss": 0.0082,
      "step": 320
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0,
      "loss": 0.0683,
      "step": 330
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0,
      "loss": 0.0961,
      "step": 340
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0,
      "loss": 0.0054,
      "step": 350
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0,
      "loss": 0.0486,
      "step": 360
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0,
      "loss": 0.1173,
      "step": 370
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0,
      "loss": 0.0658,
      "step": 380
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0,
      "loss": 0.1945,
      "step": 390
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0,
      "loss": 0.0169,
      "step": 400
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0,
      "loss": 0.1126,
      "step": 410
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0,
      "loss": 0.0566,
      "step": 420
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0,
      "loss": 0.1206,
      "step": 430
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0,
      "loss": 0.0652,
      "step": 440
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0,
      "loss": 0.0336,
      "step": 450
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0,
      "loss": 0.1087,
      "step": 460
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0,
      "loss": 0.0813,
      "step": 470
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0,
      "loss": 0.0551,
      "step": 480
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0,
      "loss": 0.1438,
      "step": 490
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0,
      "loss": 0.0044,
      "step": 500
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.9484127163887024,
      "eval_loss": 0.2529471218585968,
      "eval_runtime": 19.6111,
      "eval_samples_per_second": 25.7,
      "eval_steps_per_second": 6.425,
      "step": 500
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0,
      "loss": 0.0058,
      "step": 510
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0,
      "loss": 0.0095,
      "step": 520
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0,
      "loss": 0.0162,
      "step": 530
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0,
      "loss": 0.0024,
      "step": 540
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0,
      "loss": 0.1816,
      "step": 550
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0,
      "loss": 0.0086,
      "step": 560
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0,
      "loss": 0.0044,
      "step": 570
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0,
      "loss": 0.1037,
      "step": 580
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0,
      "loss": 0.0139,
      "step": 590
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0,
      "loss": 0.0781,
      "step": 600
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0,
      "loss": 0.0274,
      "step": 610
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0,
      "loss": 0.248,
      "step": 620
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0,
      "loss": 0.094,
      "step": 630
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0,
      "loss": 0.0492,
      "step": 640
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0,
      "loss": 0.2059,
      "step": 650
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0,
      "loss": 0.05,
      "step": 660
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0,
      "loss": 0.1069,
      "step": 670
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0,
      "loss": 0.0823,
      "step": 680
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0,
      "loss": 0.0839,
      "step": 690
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0,
      "loss": 0.1075,
      "step": 700
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0,
      "loss": 0.036,
      "step": 710
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0,
      "loss": 0.0371,
      "step": 720
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0,
      "loss": 0.1644,
      "step": 730
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0,
      "loss": 0.0058,
      "step": 740
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0,
      "loss": 0.1455,
      "step": 750
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0,
      "loss": 0.0995,
      "step": 760
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0,
      "loss": 0.0143,
      "step": 770
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0,
      "loss": 0.0722,
      "step": 780
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0,
      "loss": 0.0564,
      "step": 790
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0,
      "loss": 0.0579,
      "step": 800
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0,
      "loss": 0.0293,
      "step": 810
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0,
      "loss": 0.0041,
      "step": 820
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0,
      "loss": 0.0738,
      "step": 830
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0,
      "loss": 0.0723,
      "step": 840
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0,
      "loss": 0.0691,
      "step": 850
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0,
      "loss": 0.1105,
      "step": 860
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0,
      "loss": 0.0431,
      "step": 870
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0,
      "loss": 0.0099,
      "step": 880
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0,
      "loss": 0.0372,
      "step": 890
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0,
      "loss": 0.0031,
      "step": 900
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0,
      "loss": 0.1861,
      "step": 910
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0,
      "loss": 0.0132,
      "step": 920
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0,
      "loss": 0.1948,
      "step": 930
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0,
      "loss": 0.0089,
      "step": 940
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0,
      "loss": 0.003,
      "step": 950
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0,
      "loss": 0.0022,
      "step": 960
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0,
      "loss": 0.0136,
      "step": 970
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0,
      "loss": 0.0829,
      "step": 980
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0,
      "loss": 0.0076,
      "step": 990
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0,
      "loss": 0.0082,
      "step": 1000
    },
    {
      "epoch": 1.99,
      "eval_accuracy": 0.9484127163887024,
      "eval_loss": 0.2529471218585968,
      "eval_runtime": 23.0993,
      "eval_samples_per_second": 21.819,
      "eval_steps_per_second": 5.455,
      "step": 1000
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0,
      "loss": 0.0855,
      "step": 1010
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0,
      "loss": 0.0352,
      "step": 1020
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0,
      "loss": 0.1112,
      "step": 1030
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0,
      "loss": 0.0089,
      "step": 1040
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0,
      "loss": 0.0055,
      "step": 1050
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0,
      "loss": 0.1579,
      "step": 1060
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0,
      "loss": 0.0573,
      "step": 1070
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0,
      "loss": 0.1156,
      "step": 1080
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0,
      "loss": 0.0173,
      "step": 1090
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0,
      "loss": 0.1601,
      "step": 1100
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0,
      "loss": 0.0059,
      "step": 1110
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0,
      "loss": 0.1865,
      "step": 1120
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0,
      "loss": 0.0519,
      "step": 1130
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0,
      "loss": 0.0657,
      "step": 1140
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0,
      "loss": 0.0577,
      "step": 1150
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0,
      "loss": 0.0885,
      "step": 1160
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0,
      "loss": 0.0032,
      "step": 1170
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0,
      "loss": 0.0808,
      "step": 1180
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0,
      "loss": 0.048,
      "step": 1190
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0,
      "loss": 0.0184,
      "step": 1200
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0,
      "loss": 0.0039,
      "step": 1210
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0,
      "loss": 0.0313,
      "step": 1220
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0,
      "loss": 0.069,
      "step": 1230
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0,
      "loss": 0.1047,
      "step": 1240
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0,
      "loss": 0.0696,
      "step": 1250
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0,
      "loss": 0.0971,
      "step": 1260
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0,
      "loss": 0.1439,
      "step": 1270
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0,
      "loss": 0.0776,
      "step": 1280
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0,
      "loss": 0.0376,
      "step": 1290
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0,
      "loss": 0.0631,
      "step": 1300
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0,
      "loss": 0.0936,
      "step": 1310
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0,
      "loss": 0.0363,
      "step": 1320
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0,
      "loss": 0.0061,
      "step": 1330
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0,
      "loss": 0.0015,
      "step": 1340
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0,
      "loss": 0.073,
      "step": 1350
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0,
      "loss": 0.1017,
      "step": 1360
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0,
      "loss": 0.1227,
      "step": 1370
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0,
      "loss": 0.0862,
      "step": 1380
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0,
      "loss": 0.0406,
      "step": 1390
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0,
      "loss": 0.0647,
      "step": 1400
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0,
      "loss": 0.0014,
      "step": 1410
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.0,
      "loss": 0.0527,
      "step": 1420
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0,
      "loss": 0.062,
      "step": 1430
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.0,
      "loss": 0.0039,
      "step": 1440
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0,
      "loss": 0.002,
      "step": 1450
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0,
      "loss": 0.2507,
      "step": 1460
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0,
      "loss": 0.0513,
      "step": 1470
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0,
      "loss": 0.005,
      "step": 1480
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.0,
      "loss": 0.0126,
      "step": 1490
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0,
      "loss": 0.0041,
      "step": 1500
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.9484127163887024,
      "eval_loss": 0.2529471218585968,
      "eval_runtime": 18.8541,
      "eval_samples_per_second": 26.732,
      "eval_steps_per_second": 6.683,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0,
      "loss": 0.093,
      "step": 1510
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0,
      "loss": 0.0606,
      "step": 1520
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0,
      "loss": 0.0034,
      "step": 1530
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0,
      "loss": 0.039,
      "step": 1540
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0,
      "loss": 0.004,
      "step": 1550
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0,
      "loss": 0.0664,
      "step": 1560
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0,
      "loss": 0.0503,
      "step": 1570
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0,
      "loss": 0.0861,
      "step": 1580
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0,
      "loss": 0.0054,
      "step": 1590
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0,
      "loss": 0.0855,
      "step": 1600
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0,
      "loss": 0.1265,
      "step": 1610
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0,
      "loss": 0.0166,
      "step": 1620
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0,
      "loss": 0.0298,
      "step": 1630
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0,
      "loss": 0.0567,
      "step": 1640
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0,
      "loss": 0.1123,
      "step": 1650
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0,
      "loss": 0.1002,
      "step": 1660
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0,
      "loss": 0.0601,
      "step": 1670
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0,
      "loss": 0.0047,
      "step": 1680
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0,
      "loss": 0.0385,
      "step": 1690
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0,
      "loss": 0.0204,
      "step": 1700
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0,
      "loss": 0.0921,
      "step": 1710
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0,
      "loss": 0.0043,
      "step": 1720
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0,
      "loss": 0.1291,
      "step": 1730
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0,
      "loss": 0.0044,
      "step": 1740
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0,
      "loss": 0.0833,
      "step": 1750
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0,
      "loss": 0.0035,
      "step": 1760
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0,
      "loss": 0.0119,
      "step": 1770
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0,
      "loss": 0.0121,
      "step": 1780
    },
    {
      "epoch": 3.56,
      "learning_rate": 0.0,
      "loss": 0.0127,
      "step": 1790
    },
    {
      "epoch": 3.58,
      "learning_rate": 0.0,
      "loss": 0.0987,
      "step": 1800
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0,
      "loss": 0.0101,
      "step": 1810
    },
    {
      "epoch": 3.62,
      "learning_rate": 0.0,
      "loss": 0.0751,
      "step": 1820
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0,
      "loss": 0.0624,
      "step": 1830
    },
    {
      "epoch": 3.66,
      "learning_rate": 0.0,
      "loss": 0.0354,
      "step": 1840
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.0,
      "loss": 0.0017,
      "step": 1850
    },
    {
      "epoch": 3.7,
      "learning_rate": 0.0,
      "loss": 0.0373,
      "step": 1860
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.0,
      "loss": 0.0541,
      "step": 1870
    },
    {
      "epoch": 3.74,
      "learning_rate": 0.0,
      "loss": 0.0523,
      "step": 1880
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0,
      "loss": 0.035,
      "step": 1890
    },
    {
      "epoch": 3.78,
      "learning_rate": 0.0,
      "loss": 0.0153,
      "step": 1900
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0,
      "loss": 0.1175,
      "step": 1910
    },
    {
      "epoch": 3.82,
      "learning_rate": 0.0,
      "loss": 0.0691,
      "step": 1920
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0,
      "loss": 0.0708,
      "step": 1930
    },
    {
      "epoch": 3.86,
      "learning_rate": 0.0,
      "loss": 0.0024,
      "step": 1940
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.0,
      "loss": 0.1188,
      "step": 1950
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.0,
      "loss": 0.0656,
      "step": 1960
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.0,
      "loss": 0.1651,
      "step": 1970
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.0,
      "loss": 0.086,
      "step": 1980
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0,
      "loss": 0.1358,
      "step": 1990
    },
    {
      "epoch": 3.98,
      "learning_rate": 0.0,
      "loss": 0.0437,
      "step": 2000
    },
    {
      "epoch": 3.98,
      "eval_accuracy": 0.9484127163887024,
      "eval_loss": 0.2529471218585968,
      "eval_runtime": 20.8751,
      "eval_samples_per_second": 24.144,
      "eval_steps_per_second": 6.036,
      "step": 2000
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0,
      "loss": 0.0859,
      "step": 2010
    },
    {
      "epoch": 4.02,
      "learning_rate": 0.0,
      "loss": 0.1498,
      "step": 2020
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.0,
      "loss": 0.0834,
      "step": 2030
    },
    {
      "epoch": 4.06,
      "learning_rate": 0.0,
      "loss": 0.0372,
      "step": 2040
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.0,
      "loss": 0.17,
      "step": 2050
    },
    {
      "epoch": 4.1,
      "learning_rate": 0.0,
      "loss": 0.0045,
      "step": 2060
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.0,
      "loss": 0.0043,
      "step": 2070
    },
    {
      "epoch": 4.14,
      "learning_rate": 0.0,
      "loss": 0.0555,
      "step": 2080
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.0,
      "loss": 0.0567,
      "step": 2090
    },
    {
      "epoch": 4.17,
      "learning_rate": 0.0,
      "loss": 0.0428,
      "step": 2100
    },
    {
      "epoch": 4.19,
      "learning_rate": 0.0,
      "loss": 0.246,
      "step": 2110
    },
    {
      "epoch": 4.21,
      "learning_rate": 0.0,
      "loss": 0.0879,
      "step": 2120
    },
    {
      "epoch": 4.23,
      "learning_rate": 0.0,
      "loss": 0.0723,
      "step": 2130
    },
    {
      "epoch": 4.25,
      "learning_rate": 0.0,
      "loss": 0.0668,
      "step": 2140
    },
    {
      "epoch": 4.27,
      "learning_rate": 0.0,
      "loss": 0.0104,
      "step": 2150
    },
    {
      "epoch": 4.29,
      "learning_rate": 0.0,
      "loss": 0.0354,
      "step": 2160
    },
    {
      "epoch": 4.31,
      "learning_rate": 0.0,
      "loss": 0.1192,
      "step": 2170
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.0,
      "loss": 0.0557,
      "step": 2180
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.0,
      "loss": 0.233,
      "step": 2190
    },
    {
      "epoch": 4.37,
      "learning_rate": 0.0,
      "loss": 0.0158,
      "step": 2200
    },
    {
      "epoch": 4.39,
      "learning_rate": 0.0,
      "loss": 0.009,
      "step": 2210
    },
    {
      "epoch": 4.41,
      "learning_rate": 0.0,
      "loss": 0.0515,
      "step": 2220
    },
    {
      "epoch": 4.43,
      "learning_rate": 0.0,
      "loss": 0.0693,
      "step": 2230
    },
    {
      "epoch": 4.45,
      "learning_rate": 0.0,
      "loss": 0.0554,
      "step": 2240
    },
    {
      "epoch": 4.47,
      "learning_rate": 0.0,
      "loss": 0.0558,
      "step": 2250
    },
    {
      "epoch": 4.49,
      "learning_rate": 0.0,
      "loss": 0.0594,
      "step": 2260
    },
    {
      "epoch": 4.51,
      "learning_rate": 0.0,
      "loss": 0.1239,
      "step": 2270
    },
    {
      "epoch": 4.53,
      "learning_rate": 0.0,
      "loss": 0.0655,
      "step": 2280
    },
    {
      "epoch": 4.55,
      "learning_rate": 0.0,
      "loss": 0.0444,
      "step": 2290
    },
    {
      "epoch": 4.57,
      "learning_rate": 0.0,
      "loss": 0.0049,
      "step": 2300
    },
    {
      "epoch": 4.59,
      "learning_rate": 0.0,
      "loss": 0.1547,
      "step": 2310
    },
    {
      "epoch": 4.61,
      "learning_rate": 0.0,
      "loss": 0.0394,
      "step": 2320
    },
    {
      "epoch": 4.63,
      "learning_rate": 0.0,
      "loss": 0.0607,
      "step": 2330
    },
    {
      "epoch": 4.65,
      "learning_rate": 0.0,
      "loss": 0.0062,
      "step": 2340
    },
    {
      "epoch": 4.67,
      "learning_rate": 0.0,
      "loss": 0.2009,
      "step": 2350
    },
    {
      "epoch": 4.69,
      "learning_rate": 0.0,
      "loss": 0.0279,
      "step": 2360
    },
    {
      "epoch": 4.71,
      "learning_rate": 0.0,
      "loss": 0.1167,
      "step": 2370
    },
    {
      "epoch": 4.73,
      "learning_rate": 0.0,
      "loss": 0.1043,
      "step": 2380
    },
    {
      "epoch": 4.75,
      "learning_rate": 0.0,
      "loss": 0.1177,
      "step": 2390
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.0,
      "loss": 0.0038,
      "step": 2400
    },
    {
      "epoch": 4.79,
      "learning_rate": 0.0,
      "loss": 0.0638,
      "step": 2410
    },
    {
      "epoch": 4.81,
      "learning_rate": 0.0,
      "loss": 0.0515,
      "step": 2420
    },
    {
      "epoch": 4.83,
      "learning_rate": 0.0,
      "loss": 0.0376,
      "step": 2430
    },
    {
      "epoch": 4.85,
      "learning_rate": 0.0,
      "loss": 0.0064,
      "step": 2440
    },
    {
      "epoch": 4.87,
      "learning_rate": 0.0,
      "loss": 0.0028,
      "step": 2450
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.0,
      "loss": 0.0049,
      "step": 2460
    },
    {
      "epoch": 4.91,
      "learning_rate": 0.0,
      "loss": 0.0267,
      "step": 2470
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.0,
      "loss": 0.1005,
      "step": 2480
    },
    {
      "epoch": 4.95,
      "learning_rate": 0.0,
      "loss": 0.0818,
      "step": 2490
    },
    {
      "epoch": 4.97,
      "learning_rate": 0.0,
      "loss": 0.0371,
      "step": 2500
    },
    {
      "epoch": 4.97,
      "eval_accuracy": 0.9484127163887024,
      "eval_loss": 0.2529471218585968,
      "eval_runtime": 19.8825,
      "eval_samples_per_second": 25.349,
      "eval_steps_per_second": 6.337,
      "step": 2500
    },
    {
      "epoch": 4.99,
      "learning_rate": 0.0,
      "loss": 0.1018,
      "step": 2510
    }
  ],
  "max_steps": 2515,
  "num_train_epochs": 5,
  "total_flos": 7.171139575642413e+17,
  "trial_name": null,
  "trial_params": null
}