{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6747069241798094,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 0.7081,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 0.6774,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 0.6781,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 0.6772,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 0.6802,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 0.6749,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 0.6745,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 0.6729,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 0.6586,
      "step": 90
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 0.636,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 0.6237,
      "step": 110
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 0.6098,
      "step": 120
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 0.5978,
      "step": 130
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 0.602,
      "step": 140
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 0.5959,
      "step": 150
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 0.5906,
      "step": 160
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 0.5778,
      "step": 170
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 0.5795,
      "step": 180
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 0.5883,
      "step": 190
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 0.604,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 0.5882,
      "step": 210
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 0.5747,
      "step": 220
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 0.5745,
      "step": 230
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 0.5846,
      "step": 240
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 0.5823,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 0.5722,
      "step": 260
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 0.5599,
      "step": 270
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 0.5716,
      "step": 280
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 0.5832,
      "step": 290
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 0.5745,
      "step": 300
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 0.5691,
      "step": 310
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 0.5711,
      "step": 320
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 0.5821,
      "step": 330
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 0.5557,
      "step": 340
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 0.5581,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 0.5761,
      "step": 360
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 0.5643,
      "step": 370
    },
    {
      "epoch": 0.13,
      "learning_rate": 5e-05,
      "loss": 0.5745,
      "step": 380
    },
    {
      "epoch": 0.13,
      "learning_rate": 5e-05,
      "loss": 0.5702,
      "step": 390
    },
    {
      "epoch": 0.13,
      "learning_rate": 5e-05,
      "loss": 0.5648,
      "step": 400
    },
    {
      "epoch": 0.14,
      "learning_rate": 5e-05,
      "loss": 0.5691,
      "step": 410
    },
    {
      "epoch": 0.14,
      "learning_rate": 5e-05,
      "loss": 0.556,
      "step": 420
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 0.5578,
      "step": 430
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 0.564,
      "step": 440
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 0.5562,
      "step": 450
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 0.559,
      "step": 460
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 0.565,
      "step": 470
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 0.5508,
      "step": 480
    },
    {
      "epoch": 0.17,
      "learning_rate": 5e-05,
      "loss": 0.5588,
      "step": 490
    },
    {
      "epoch": 0.17,
      "learning_rate": 5e-05,
      "loss": 0.5584,
      "step": 500
    },
    {
      "epoch": 0.17,
      "learning_rate": 5e-05,
      "loss": 0.5555,
      "step": 510
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 0.5597,
      "step": 520
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 0.5483,
      "step": 530
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 0.5621,
      "step": 540
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 0.5568,
      "step": 550
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 0.5527,
      "step": 560
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 0.5524,
      "step": 570
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 0.5458,
      "step": 580
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 0.5426,
      "step": 590
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 0.5524,
      "step": 600
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 0.5553,
      "step": 610
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 0.5582,
      "step": 620
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 0.5518,
      "step": 630
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 0.5485,
      "step": 640
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 0.5511,
      "step": 650
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 0.5444,
      "step": 660
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 0.5522,
      "step": 670
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 0.5417,
      "step": 680
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 0.5543,
      "step": 690
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 0.56,
      "step": 700
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 0.5394,
      "step": 710
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 0.5475,
      "step": 720
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 0.5437,
      "step": 730
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 0.5438,
      "step": 740
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 0.5454,
      "step": 750
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 0.5481,
      "step": 760
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 0.5434,
      "step": 770
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 0.5422,
      "step": 780
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 0.5465,
      "step": 790
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 0.5447,
      "step": 800
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 0.5396,
      "step": 810
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 0.543,
      "step": 820
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 0.5461,
      "step": 830
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 0.5482,
      "step": 840
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 0.5426,
      "step": 850
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 0.5453,
      "step": 860
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 0.5493,
      "step": 870
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 0.5477,
      "step": 880
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 0.542,
      "step": 890
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 0.5393,
      "step": 900
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 0.5365,
      "step": 910
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 0.5433,
      "step": 920
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 0.542,
      "step": 930
    },
    {
      "epoch": 0.32,
      "learning_rate": 5e-05,
      "loss": 0.5368,
      "step": 940
    },
    {
      "epoch": 0.32,
      "learning_rate": 5e-05,
      "loss": 0.542,
      "step": 950
    },
    {
      "epoch": 0.32,
      "learning_rate": 5e-05,
      "loss": 0.5467,
      "step": 960
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 0.5382,
      "step": 970
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 0.5366,
      "step": 980
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 0.5354,
      "step": 990
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 0.5322,
      "step": 1000
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 0.5379,
      "step": 1010
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 0.5398,
      "step": 1020
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 0.5337,
      "step": 1030
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 0.5384,
      "step": 1040
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 0.5256,
      "step": 1050
    },
    {
      "epoch": 0.36,
      "learning_rate": 5e-05,
      "loss": 0.5353,
      "step": 1060
    },
    {
      "epoch": 0.36,
      "learning_rate": 5e-05,
      "loss": 0.5233,
      "step": 1070
    },
    {
      "epoch": 0.36,
      "learning_rate": 5e-05,
      "loss": 0.5483,
      "step": 1080
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 0.5362,
      "step": 1090
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 0.5426,
      "step": 1100
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 0.5347,
      "step": 1110
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 0.5377,
      "step": 1120
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 0.5374,
      "step": 1130
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 0.532,
      "step": 1140
    },
    {
      "epoch": 0.39,
      "learning_rate": 5e-05,
      "loss": 0.5315,
      "step": 1150
    },
    {
      "epoch": 0.39,
      "learning_rate": 5e-05,
      "loss": 0.5274,
      "step": 1160
    },
    {
      "epoch": 0.39,
      "learning_rate": 5e-05,
      "loss": 0.537,
      "step": 1170
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 0.5427,
      "step": 1180
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 0.5288,
      "step": 1190
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 0.5259,
      "step": 1200
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 0.531,
      "step": 1210
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 0.5263,
      "step": 1220
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 0.5305,
      "step": 1230
    },
    {
      "epoch": 0.42,
      "learning_rate": 5e-05,
      "loss": 0.5245,
      "step": 1240
    },
    {
      "epoch": 0.42,
      "learning_rate": 5e-05,
      "loss": 0.5306,
      "step": 1250
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 0.526,
      "step": 1260
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 0.5308,
      "step": 1270
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 0.5226,
      "step": 1280
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 0.5295,
      "step": 1290
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 0.5297,
      "step": 1300
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 0.5293,
      "step": 1310
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 0.5387,
      "step": 1320
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 0.5252,
      "step": 1330
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 0.524,
      "step": 1340
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 0.5231,
      "step": 1350
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 0.5248,
      "step": 1360
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 0.5332,
      "step": 1370
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-05,
      "loss": 0.5366,
      "step": 1380
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-05,
      "loss": 0.5231,
      "step": 1390
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-05,
      "loss": 0.5206,
      "step": 1400
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 0.5237,
      "step": 1410
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 0.5201,
      "step": 1420
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 0.5366,
      "step": 1430
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 0.5336,
      "step": 1440
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 0.5282,
      "step": 1450
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 0.5269,
      "step": 1460
    },
    {
      "epoch": 0.5,
      "learning_rate": 5e-05,
      "loss": 0.5245,
      "step": 1470
    },
    {
      "epoch": 0.5,
      "learning_rate": 5e-05,
      "loss": 0.5203,
      "step": 1480
    },
    {
      "epoch": 0.5,
      "learning_rate": 5e-05,
      "loss": 0.5225,
      "step": 1490
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 0.5311,
      "step": 1500
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 0.5229,
      "step": 1510
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 0.5266,
      "step": 1520
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 0.5283,
      "step": 1530
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 0.5279,
      "step": 1540
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 0.5299,
      "step": 1550
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 0.5267,
      "step": 1560
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 0.5186,
      "step": 1570
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 0.5238,
      "step": 1580
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 0.5249,
      "step": 1590
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 0.5225,
      "step": 1600
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 0.5226,
      "step": 1610
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 0.5188,
      "step": 1620
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 0.525,
      "step": 1630
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 0.5249,
      "step": 1640
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 0.5225,
      "step": 1650
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 0.5281,
      "step": 1660
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 0.5204,
      "step": 1670
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 0.5193,
      "step": 1680
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 0.5278,
      "step": 1690
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 0.5224,
      "step": 1700
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 0.5218,
      "step": 1710
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 0.5171,
      "step": 1720
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 0.5239,
      "step": 1730
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 0.5231,
      "step": 1740
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 0.5164,
      "step": 1750
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 0.5181,
      "step": 1760
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 0.5166,
      "step": 1770
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 0.5179,
      "step": 1780
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 0.5169,
      "step": 1790
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 0.5275,
      "step": 1800
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 0.5179,
      "step": 1810
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 0.5216,
      "step": 1820
    },
    {
      "epoch": 0.62,
      "learning_rate": 5e-05,
      "loss": 0.5236,
      "step": 1830
    },
    {
      "epoch": 0.62,
      "learning_rate": 5e-05,
      "loss": 0.5125,
      "step": 1840
    },
    {
      "epoch": 0.62,
      "learning_rate": 5e-05,
      "loss": 0.527,
      "step": 1850
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 0.5224,
      "step": 1860
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 0.5304,
      "step": 1870
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 0.5242,
      "step": 1880
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 0.5351,
      "step": 1890
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 0.5262,
      "step": 1900
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 0.5178,
      "step": 1910
    },
    {
      "epoch": 0.65,
      "learning_rate": 5e-05,
      "loss": 0.5104,
      "step": 1920
    },
    {
      "epoch": 0.65,
      "learning_rate": 5e-05,
      "loss": 0.5172,
      "step": 1930
    },
    {
      "epoch": 0.65,
      "learning_rate": 5e-05,
      "loss": 0.517,
      "step": 1940
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 0.5154,
      "step": 1950
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 0.5285,
      "step": 1960
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 0.5163,
      "step": 1970
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 0.5104,
      "step": 1980
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 0.5228,
      "step": 1990
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 0.5131,
      "step": 2000
    }
  ],
  "logging_steps": 10,
  "max_steps": 8892,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}