{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 3448,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 4.4935,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.576,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9832,
      "step": 30
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8691,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7926,
      "step": 50
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.795,
      "step": 60
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7327,
      "step": 70
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7195,
      "step": 80
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7307,
      "step": 90
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6724,
      "step": 100
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.6968,
      "step": 110
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.7335,
      "step": 120
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.7235,
      "step": 130
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.6764,
      "step": 140
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.6414,
      "step": 150
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.6739,
      "step": 160
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.6421,
      "step": 170
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.6436,
      "step": 180
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.635,
      "step": 190
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.6113,
      "step": 200
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.6381,
      "step": 210
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.6109,
      "step": 220
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.614,
      "step": 230
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.6144,
      "step": 240
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.5815,
      "step": 250
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.6123,
      "step": 260
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.5971,
      "step": 270
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.5952,
      "step": 280
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.6136,
      "step": 290
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.5892,
      "step": 300
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.6023,
      "step": 310
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.5737,
      "step": 320
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.578,
      "step": 330
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.5737,
      "step": 340
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.5561,
      "step": 350
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.5794,
      "step": 360
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 0.5677,
      "step": 370
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.5603,
      "step": 380
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.5622,
      "step": 390
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.544,
      "step": 400
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.5675,
      "step": 410
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.5656,
      "step": 420
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.5494,
      "step": 430
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.5525,
      "step": 440
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002,
      "loss": 0.5325,
      "step": 450
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.5571,
      "step": 460
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 0.5422,
      "step": 470
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.5415,
      "step": 480
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.5473,
      "step": 490
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.532,
      "step": 500
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.5466,
      "step": 510
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.5431,
      "step": 520
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.5358,
      "step": 530
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.5304,
      "step": 540
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002,
      "loss": 0.5142,
      "step": 550
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.5426,
      "step": 560
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.5303,
      "step": 570
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.5285,
      "step": 580
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.5239,
      "step": 590
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.5144,
      "step": 600
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.5326,
      "step": 610
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002,
      "loss": 0.5147,
      "step": 620
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.5238,
      "step": 630
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002,
      "loss": 0.5328,
      "step": 640
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.5057,
      "step": 650
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.5276,
      "step": 660
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.5204,
      "step": 670
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002,
      "loss": 0.52,
      "step": 680
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.5197,
      "step": 690
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.5032,
      "step": 700
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.5275,
      "step": 710
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002,
      "loss": 0.5047,
      "step": 720
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.5055,
      "step": 730
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.5216,
      "step": 740
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.5019,
      "step": 750
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.5134,
      "step": 760
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 0.4997,
      "step": 770
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 0.4985,
      "step": 780
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.5033,
      "step": 790
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.4919,
      "step": 800
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.5104,
      "step": 810
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.4918,
      "step": 820
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002,
      "loss": 0.4877,
      "step": 830
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.5057,
      "step": 840
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.4728,
      "step": 850
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.5035,
      "step": 860
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.4923032224178314,
      "eval_runtime": 5.6257,
      "eval_samples_per_second": 177.756,
      "eval_steps_per_second": 2.844,
      "step": 862
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002,
      "loss": 0.4254,
      "step": 870
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.4024,
      "step": 880
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002,
      "loss": 0.4108,
      "step": 890
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.3998,
      "step": 900
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 0.4134,
      "step": 910
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.418,
      "step": 920
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.4121,
      "step": 930
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.4197,
      "step": 940
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.4064,
      "step": 950
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.4144,
      "step": 960
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.4218,
      "step": 970
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 0.4118,
      "step": 980
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.4216,
      "step": 990
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002,
      "loss": 0.4056,
      "step": 1000
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.4157,
      "step": 1010
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.4177,
      "step": 1020
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.4059,
      "step": 1030
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.4206,
      "step": 1040
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.4175,
      "step": 1050
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.4223,
      "step": 1060
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.4179,
      "step": 1070
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.4149,
      "step": 1080
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.4166,
      "step": 1090
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002,
      "loss": 0.4097,
      "step": 1100
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.4266,
      "step": 1110
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002,
      "loss": 0.429,
      "step": 1120
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.4159,
      "step": 1130
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.4239,
      "step": 1140
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.4074,
      "step": 1150
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.4212,
      "step": 1160
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.4208,
      "step": 1170
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.4156,
      "step": 1180
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.4275,
      "step": 1190
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.462,
      "step": 1200
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.4299,
      "step": 1210
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002,
      "loss": 0.4216,
      "step": 1220
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.4112,
      "step": 1230
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.4219,
      "step": 1240
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002,
      "loss": 0.4081,
      "step": 1250
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.4218,
      "step": 1260
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.4182,
      "step": 1270
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.4104,
      "step": 1280
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002,
      "loss": 0.4216,
      "step": 1290
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.4098,
      "step": 1300
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002,
      "loss": 0.4132,
      "step": 1310
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.4118,
      "step": 1320
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.4194,
      "step": 1330
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.4206,
      "step": 1340
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002,
      "loss": 0.4073,
      "step": 1350
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.4167,
      "step": 1360
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.406,
      "step": 1370
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.4093,
      "step": 1380
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.418,
      "step": 1390
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.4035,
      "step": 1400
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002,
      "loss": 0.4139,
      "step": 1410
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.4058,
      "step": 1420
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.4063,
      "step": 1430
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.4155,
      "step": 1440
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.4307,
      "step": 1450
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.4259,
      "step": 1460
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0002,
      "loss": 0.4049,
      "step": 1470
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.4101,
      "step": 1480
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.4553,
      "step": 1490
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0002,
      "loss": 0.4012,
      "step": 1500
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.4094,
      "step": 1510
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.4003,
      "step": 1520
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.4019,
      "step": 1530
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002,
      "loss": 0.4107,
      "step": 1540
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.4054,
      "step": 1550
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0002,
      "loss": 0.4154,
      "step": 1560
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.407,
      "step": 1570
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.4067,
      "step": 1580
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.4109,
      "step": 1590
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0002,
      "loss": 0.4056,
      "step": 1600
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.4159,
      "step": 1610
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.3987,
      "step": 1620
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.3993,
      "step": 1630
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.4081,
      "step": 1640
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0002,
      "loss": 0.3993,
      "step": 1650
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.4112,
      "step": 1660
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.4052,
      "step": 1670
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.4028,
      "step": 1680
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0002,
      "loss": 0.4099,
      "step": 1690
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 0.4026,
      "step": 1700
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0002,
      "loss": 0.4015,
      "step": 1710
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0002,
      "loss": 0.4046,
      "step": 1720
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4530680179595947,
      "eval_runtime": 5.6216,
      "eval_samples_per_second": 177.885,
      "eval_steps_per_second": 2.846,
      "step": 1724
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0002,
      "loss": 0.3456,
      "step": 1730
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0002,
      "loss": 0.3203,
      "step": 1740
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0002,
      "loss": 0.3087,
      "step": 1750
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 0.3051,
      "step": 1760
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0002,
      "loss": 0.3183,
      "step": 1770
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0002,
      "loss": 0.3098,
      "step": 1780
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0002,
      "loss": 0.3256,
      "step": 1790
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0002,
      "loss": 0.3189,
      "step": 1800
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0002,
      "loss": 0.3162,
      "step": 1810
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 0.326,
      "step": 1820
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0002,
      "loss": 0.3127,
      "step": 1830
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0002,
      "loss": 0.3306,
      "step": 1840
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0002,
      "loss": 0.3255,
      "step": 1850
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0002,
      "loss": 0.3199,
      "step": 1860
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0002,
      "loss": 0.3265,
      "step": 1870
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0002,
      "loss": 0.3185,
      "step": 1880
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0002,
      "loss": 0.3306,
      "step": 1890
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0002,
      "loss": 0.3238,
      "step": 1900
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0002,
      "loss": 0.3297,
      "step": 1910
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0002,
      "loss": 0.3348,
      "step": 1920
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0002,
      "loss": 0.329,
      "step": 1930
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0002,
      "loss": 0.3422,
      "step": 1940
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002,
      "loss": 0.3325,
      "step": 1950
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0002,
      "loss": 0.3327,
      "step": 1960
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0002,
      "loss": 0.3398,
      "step": 1970
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0002,
      "loss": 0.3279,
      "step": 1980
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0002,
      "loss": 0.3463,
      "step": 1990
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0002,
      "loss": 0.3295,
      "step": 2000
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 0.3354,
      "step": 2010
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0002,
      "loss": 0.3411,
      "step": 2020
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0002,
      "loss": 0.3263,
      "step": 2030
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0002,
      "loss": 0.3379,
      "step": 2040
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0002,
      "loss": 0.3305,
      "step": 2050
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0002,
      "loss": 0.3296,
      "step": 2060
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 0.3489,
      "step": 2070
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0002,
      "loss": 0.3325,
      "step": 2080
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0002,
      "loss": 0.34,
      "step": 2090
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0002,
      "loss": 0.34,
      "step": 2100
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0002,
      "loss": 0.3424,
      "step": 2110
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0002,
      "loss": 0.3421,
      "step": 2120
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0002,
      "loss": 0.336,
      "step": 2130
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002,
      "loss": 0.3481,
      "step": 2140
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0002,
      "loss": 0.3426,
      "step": 2150
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0002,
      "loss": 0.3425,
      "step": 2160
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0002,
      "loss": 0.3442,
      "step": 2170
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0002,
      "loss": 0.3351,
      "step": 2180
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0002,
      "loss": 0.3582,
      "step": 2190
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0002,
      "loss": 0.3429,
      "step": 2200
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0002,
      "loss": 0.3421,
      "step": 2210
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0002,
      "loss": 0.3444,
      "step": 2220
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0002,
      "loss": 0.3345,
      "step": 2230
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0002,
      "loss": 0.3484,
      "step": 2240
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0002,
      "loss": 0.3451,
      "step": 2250
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 0.341,
      "step": 2260
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0002,
      "loss": 0.3447,
      "step": 2270
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0002,
      "loss": 0.3354,
      "step": 2280
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0002,
      "loss": 0.3476,
      "step": 2290
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0002,
      "loss": 0.3377,
      "step": 2300
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0002,
      "loss": 0.3349,
      "step": 2310
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0002,
      "loss": 0.3432,
      "step": 2320
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0002,
      "loss": 0.3321,
      "step": 2330
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0002,
      "loss": 0.3475,
      "step": 2340
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0002,
      "loss": 0.3367,
      "step": 2350
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0002,
      "loss": 0.3408,
      "step": 2360
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0002,
      "loss": 0.3465,
      "step": 2370
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0002,
      "loss": 0.3331,
      "step": 2380
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 0.3532,
      "step": 2390
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0002,
      "loss": 0.3425,
      "step": 2400
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0002,
      "loss": 0.3412,
      "step": 2410
    },
    {
      "epoch": 2.81,
      "learning_rate": 0.0002,
      "loss": 0.344,
      "step": 2420
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.0002,
      "loss": 0.3409,
      "step": 2430
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.0002,
      "loss": 0.3455,
      "step": 2440
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0002,
      "loss": 0.344,
      "step": 2450
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.0002,
      "loss": 0.3386,
      "step": 2460
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.0002,
      "loss": 0.3443,
      "step": 2470
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0002,
      "loss": 0.3364,
      "step": 2480
    },
    {
      "epoch": 2.89,
      "learning_rate": 0.0002,
      "loss": 0.3584,
      "step": 2490
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0002,
      "loss": 0.3469,
      "step": 2500
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0002,
      "loss": 0.3476,
      "step": 2510
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0002,
      "loss": 0.3544,
      "step": 2520
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0002,
      "loss": 0.3384,
      "step": 2530
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.0002,
      "loss": 0.351,
      "step": 2540
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.0002,
      "loss": 0.3383,
      "step": 2550
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.0002,
      "loss": 0.3435,
      "step": 2560
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0002,
      "loss": 0.3542,
      "step": 2570
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0002,
      "loss": 0.3395,
      "step": 2580
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.43056488037109375,
      "eval_runtime": 5.6316,
      "eval_samples_per_second": 177.57,
      "eval_steps_per_second": 2.841,
      "step": 2586
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0002,
      "loss": 0.3188,
      "step": 2590
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0002,
      "loss": 0.2607,
      "step": 2600
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0002,
      "loss": 0.2477,
      "step": 2610
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0002,
      "loss": 0.2591,
      "step": 2620
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0002,
      "loss": 0.2514,
      "step": 2630
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0002,
      "loss": 0.2539,
      "step": 2640
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0002,
      "loss": 0.2631,
      "step": 2650
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0002,
      "loss": 0.2571,
      "step": 2660
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0002,
      "loss": 0.2645,
      "step": 2670
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0002,
      "loss": 0.2571,
      "step": 2680
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0002,
      "loss": 0.2715,
      "step": 2690
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0002,
      "loss": 0.2692,
      "step": 2700
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0002,
      "loss": 0.2582,
      "step": 2710
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0002,
      "loss": 0.2683,
      "step": 2720
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0002,
      "loss": 0.2612,
      "step": 2730
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0002,
      "loss": 0.2736,
      "step": 2740
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0002,
      "loss": 0.2678,
      "step": 2750
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0002,
      "loss": 0.2623,
      "step": 2760
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0002,
      "loss": 0.2768,
      "step": 2770
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0002,
      "loss": 0.2651,
      "step": 2780
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0002,
      "loss": 0.2702,
      "step": 2790
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0002,
      "loss": 0.2744,
      "step": 2800
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0002,
      "loss": 0.2686,
      "step": 2810
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0002,
      "loss": 0.278,
      "step": 2820
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0002,
      "loss": 0.2724,
      "step": 2830
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0002,
      "loss": 0.284,
      "step": 2840
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0002,
      "loss": 0.2831,
      "step": 2850
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0002,
      "loss": 0.2736,
      "step": 2860
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0002,
      "loss": 0.288,
      "step": 2870
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0002,
      "loss": 0.2779,
      "step": 2880
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0002,
      "loss": 0.2841,
      "step": 2890
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0002,
      "loss": 0.2822,
      "step": 2900
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0002,
      "loss": 0.2751,
      "step": 2910
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0002,
      "loss": 0.2851,
      "step": 2920
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0002,
      "loss": 0.2723,
      "step": 2930
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0002,
      "loss": 0.2852,
      "step": 2940
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0002,
      "loss": 0.2871,
      "step": 2950
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0002,
      "loss": 0.2774,
      "step": 2960
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0002,
      "loss": 0.2857,
      "step": 2970
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0002,
      "loss": 0.2768,
      "step": 2980
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0002,
      "loss": 0.2888,
      "step": 2990
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0002,
      "loss": 0.2879,
      "step": 3000
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0002,
      "loss": 0.2798,
      "step": 3010
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0002,
      "loss": 0.2908,
      "step": 3020
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0002,
      "loss": 0.2756,
      "step": 3030
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0002,
      "loss": 0.2941,
      "step": 3040
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0002,
      "loss": 0.2893,
      "step": 3050
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0002,
      "loss": 0.2803,
      "step": 3060
    },
    {
      "epoch": 3.56,
      "learning_rate": 0.0002,
      "loss": 0.2899,
      "step": 3070
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.0002,
      "loss": 0.2833,
      "step": 3080
    },
    {
      "epoch": 3.58,
      "learning_rate": 0.0002,
      "loss": 0.2908,
      "step": 3090
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0002,
      "loss": 0.2919,
      "step": 3100
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.0002,
      "loss": 0.2865,
      "step": 3110
    },
    {
      "epoch": 3.62,
      "learning_rate": 0.0002,
      "loss": 0.3033,
      "step": 3120
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.0002,
      "loss": 0.2802,
      "step": 3130
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0002,
      "loss": 0.2918,
      "step": 3140
    },
    {
      "epoch": 3.65,
      "learning_rate": 0.0002,
      "loss": 0.2911,
      "step": 3150
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.0002,
      "loss": 0.2882,
      "step": 3160
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.0002,
      "loss": 0.2873,
      "step": 3170
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0002,
      "loss": 0.286,
      "step": 3180
    },
    {
      "epoch": 3.7,
      "learning_rate": 0.0002,
      "loss": 0.294,
      "step": 3190
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.0002,
      "loss": 0.2903,
      "step": 3200
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.0002,
      "loss": 0.2818,
      "step": 3210
    },
    {
      "epoch": 3.74,
      "learning_rate": 0.0002,
      "loss": 0.2925,
      "step": 3220
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.0002,
      "loss": 0.2834,
      "step": 3230
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0002,
      "loss": 0.2932,
      "step": 3240
    },
    {
      "epoch": 3.77,
      "learning_rate": 0.0002,
      "loss": 0.3,
      "step": 3250
    },
    {
      "epoch": 3.78,
      "learning_rate": 0.0002,
      "loss": 0.2875,
      "step": 3260
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.0002,
      "loss": 0.2922,
      "step": 3270
    },
    {
      "epoch": 3.81,
      "learning_rate": 0.0002,
      "loss": 0.2855,
      "step": 3280
    },
    {
      "epoch": 3.82,
      "learning_rate": 0.0002,
      "loss": 0.2977,
      "step": 3290
    },
    {
      "epoch": 3.83,
      "learning_rate": 0.0002,
      "loss": 0.2987,
      "step": 3300
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0002,
      "loss": 0.2929,
      "step": 3310
    },
    {
      "epoch": 3.85,
      "learning_rate": 0.0002,
      "loss": 0.2979,
      "step": 3320
    },
    {
      "epoch": 3.86,
      "learning_rate": 0.0002,
      "loss": 0.2904,
      "step": 3330
    },
    {
      "epoch": 3.87,
      "learning_rate": 0.0002,
      "loss": 0.2977,
      "step": 3340
    },
    {
      "epoch": 3.89,
      "learning_rate": 0.0002,
      "loss": 0.2936,
      "step": 3350
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.0002,
      "loss": 0.2924,
      "step": 3360
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.0002,
      "loss": 0.2982,
      "step": 3370
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.0002,
      "loss": 0.2897,
      "step": 3380
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.0002,
      "loss": 0.3037,
      "step": 3390
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.0002,
      "loss": 0.3005,
      "step": 3400
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0002,
      "loss": 0.294,
      "step": 3410
    },
    {
      "epoch": 3.97,
      "learning_rate": 0.0002,
      "loss": 0.3017,
      "step": 3420
    },
    {
      "epoch": 3.98,
      "learning_rate": 0.0002,
      "loss": 0.2917,
      "step": 3430
    },
    {
      "epoch": 3.99,
      "learning_rate": 0.0002,
      "loss": 0.3012,
      "step": 3440
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.4269254803657532,
      "eval_runtime": 5.6372,
      "eval_samples_per_second": 177.394,
      "eval_steps_per_second": 2.838,
      "step": 3448
    }
  ],
  "max_steps": 10000,
  "num_train_epochs": 12,
  "total_flos": 1.6741838366462444e+18,
  "trial_name": null,
  "trial_params": null
}