{
  "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": 0.5854,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 0.5466,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 0.5426,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 0.5393,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 0.55,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 0.5355,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 0.5387,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 0.5299,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 0.5183,
      "step": 90
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 0.5188,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 0.5186,
      "step": 110
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 0.5135,
      "step": 120
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 0.5221,
      "step": 130
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 0.5163,
      "step": 140
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 0.5063,
      "step": 150
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 0.5134,
      "step": 160
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 0.5058,
      "step": 170
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 0.5059,
      "step": 180
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 0.5077,
      "step": 190
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 0.5089,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 0.5075,
      "step": 210
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 0.5017,
      "step": 220
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 0.5171,
      "step": 230
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 0.5022,
      "step": 240
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 0.5005,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 0.5025,
      "step": 260
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 0.5046,
      "step": 270
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 0.5174,
      "step": 280
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 0.5037,
      "step": 290
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 0.4952,
      "step": 300
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 0.4914,
      "step": 310
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 0.4867,
      "step": 320
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 0.4998,
      "step": 330
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 0.482,
      "step": 340
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 0.498,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 0.4982,
      "step": 360
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 0.4931,
      "step": 370
    },
    {
      "epoch": 0.13,
      "learning_rate": 5e-05,
      "loss": 0.4938,
      "step": 380
    },
    {
      "epoch": 0.13,
      "learning_rate": 5e-05,
      "loss": 0.4859,
      "step": 390
    },
    {
      "epoch": 0.13,
      "learning_rate": 5e-05,
      "loss": 0.4852,
      "step": 400
    },
    {
      "epoch": 0.14,
      "learning_rate": 5e-05,
      "loss": 0.4942,
      "step": 410
    },
    {
      "epoch": 0.14,
      "learning_rate": 5e-05,
      "loss": 0.5014,
      "step": 420
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 0.5076,
      "step": 430
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 0.4962,
      "step": 440
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 0.4802,
      "step": 450
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 0.5002,
      "step": 460
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 0.4979,
      "step": 470
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 0.4999,
      "step": 480
    },
    {
      "epoch": 0.17,
      "learning_rate": 5e-05,
      "loss": 0.4908,
      "step": 490
    },
    {
      "epoch": 0.17,
      "learning_rate": 5e-05,
      "loss": 0.4884,
      "step": 500
    },
    {
      "epoch": 0.17,
      "learning_rate": 5e-05,
      "loss": 0.4801,
      "step": 510
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 0.478,
      "step": 520
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 0.4838,
      "step": 530
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 0.4879,
      "step": 540
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 0.4761,
      "step": 550
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 0.4853,
      "step": 560
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 0.4938,
      "step": 570
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 0.477,
      "step": 580
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 0.491,
      "step": 590
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 0.4678,
      "step": 600
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 0.4814,
      "step": 610
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 0.4822,
      "step": 620
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 0.4792,
      "step": 630
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 0.4791,
      "step": 640
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 0.4839,
      "step": 650
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 0.4846,
      "step": 660
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 0.4833,
      "step": 670
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 0.4821,
      "step": 680
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 0.4778,
      "step": 690
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 0.4783,
      "step": 700
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 0.4644,
      "step": 710
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 0.4771,
      "step": 720
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 0.4747,
      "step": 730
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 0.4802,
      "step": 740
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 0.4738,
      "step": 750
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 0.4768,
      "step": 760
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 0.4692,
      "step": 770
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 0.4684,
      "step": 780
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 0.4775,
      "step": 790
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 0.4714,
      "step": 800
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 0.4744,
      "step": 810
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 0.4752,
      "step": 820
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 0.4727,
      "step": 830
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 0.4693,
      "step": 840
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 0.4746,
      "step": 850
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 0.4699,
      "step": 860
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 0.4694,
      "step": 870
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 0.4745,
      "step": 880
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 0.4674,
      "step": 890
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 0.4754,
      "step": 900
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 0.4796,
      "step": 910
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 0.4708,
      "step": 920
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 0.4746,
      "step": 930
    },
    {
      "epoch": 0.32,
      "learning_rate": 5e-05,
      "loss": 0.4768,
      "step": 940
    },
    {
      "epoch": 0.32,
      "learning_rate": 5e-05,
      "loss": 0.4709,
      "step": 950
    },
    {
      "epoch": 0.32,
      "learning_rate": 5e-05,
      "loss": 0.466,
      "step": 960
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 0.4706,
      "step": 970
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 0.4756,
      "step": 980
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 0.4687,
      "step": 990
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 0.4667,
      "step": 1000
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 0.4702,
      "step": 1010
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 0.4739,
      "step": 1020
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 0.472,
      "step": 1030
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 0.4632,
      "step": 1040
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 0.458,
      "step": 1050
    },
    {
      "epoch": 0.36,
      "learning_rate": 5e-05,
      "loss": 0.463,
      "step": 1060
    },
    {
      "epoch": 0.36,
      "learning_rate": 5e-05,
      "loss": 0.4655,
      "step": 1070
    },
    {
      "epoch": 0.36,
      "learning_rate": 5e-05,
      "loss": 0.4611,
      "step": 1080
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 0.4673,
      "step": 1090
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 0.4778,
      "step": 1100
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 0.4746,
      "step": 1110
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 0.4738,
      "step": 1120
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 0.4682,
      "step": 1130
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 0.4627,
      "step": 1140
    },
    {
      "epoch": 0.39,
      "learning_rate": 5e-05,
      "loss": 0.4608,
      "step": 1150
    },
    {
      "epoch": 0.39,
      "learning_rate": 5e-05,
      "loss": 0.4674,
      "step": 1160
    },
    {
      "epoch": 0.39,
      "learning_rate": 5e-05,
      "loss": 0.4718,
      "step": 1170
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 0.4629,
      "step": 1180
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 0.4645,
      "step": 1190
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 0.4612,
      "step": 1200
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 0.4584,
      "step": 1210
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 0.4711,
      "step": 1220
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 0.4645,
      "step": 1230
    },
    {
      "epoch": 0.42,
      "learning_rate": 5e-05,
      "loss": 0.4667,
      "step": 1240
    },
    {
      "epoch": 0.42,
      "learning_rate": 5e-05,
      "loss": 0.4615,
      "step": 1250
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 0.4576,
      "step": 1260
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 0.4634,
      "step": 1270
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 0.4621,
      "step": 1280
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 0.4575,
      "step": 1290
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 0.4602,
      "step": 1300
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 0.4598,
      "step": 1310
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 0.4672,
      "step": 1320
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 0.4613,
      "step": 1330
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 0.4674,
      "step": 1340
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 0.4673,
      "step": 1350
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 0.4524,
      "step": 1360
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 0.4685,
      "step": 1370
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-05,
      "loss": 0.4539,
      "step": 1380
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-05,
      "loss": 0.4643,
      "step": 1390
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-05,
      "loss": 0.4637,
      "step": 1400
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 0.4531,
      "step": 1410
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 0.4595,
      "step": 1420
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 0.4584,
      "step": 1430
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 0.4644,
      "step": 1440
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 0.4667,
      "step": 1450
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 0.4593,
      "step": 1460
    },
    {
      "epoch": 0.5,
      "learning_rate": 5e-05,
      "loss": 0.4585,
      "step": 1470
    },
    {
      "epoch": 0.5,
      "learning_rate": 5e-05,
      "loss": 0.4634,
      "step": 1480
    },
    {
      "epoch": 0.5,
      "learning_rate": 5e-05,
      "loss": 0.4538,
      "step": 1490
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 0.4561,
      "step": 1500
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 0.4559,
      "step": 1510
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 0.4535,
      "step": 1520
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 0.4581,
      "step": 1530
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 0.4623,
      "step": 1540
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 0.4605,
      "step": 1550
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 0.4589,
      "step": 1560
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 0.4539,
      "step": 1570
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 0.4569,
      "step": 1580
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 0.4611,
      "step": 1590
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 0.4649,
      "step": 1600
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 0.4574,
      "step": 1610
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 0.4572,
      "step": 1620
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 0.46,
      "step": 1630
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 0.4669,
      "step": 1640
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 0.4514,
      "step": 1650
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 0.4598,
      "step": 1660
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 0.4578,
      "step": 1670
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 0.4617,
      "step": 1680
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 0.4509,
      "step": 1690
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 0.4646,
      "step": 1700
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 0.457,
      "step": 1710
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 0.4523,
      "step": 1720
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 0.4629,
      "step": 1730
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 0.46,
      "step": 1740
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 0.4598,
      "step": 1750
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 0.4564,
      "step": 1760
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 0.455,
      "step": 1770
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 0.4475,
      "step": 1780
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 0.4591,
      "step": 1790
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 0.4607,
      "step": 1800
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 0.4539,
      "step": 1810
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 0.4543,
      "step": 1820
    },
    {
      "epoch": 0.62,
      "learning_rate": 5e-05,
      "loss": 0.449,
      "step": 1830
    },
    {
      "epoch": 0.62,
      "learning_rate": 5e-05,
      "loss": 0.4536,
      "step": 1840
    },
    {
      "epoch": 0.62,
      "learning_rate": 5e-05,
      "loss": 0.4511,
      "step": 1850
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 0.4547,
      "step": 1860
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 0.4669,
      "step": 1870
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 0.4574,
      "step": 1880
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 0.451,
      "step": 1890
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 0.4531,
      "step": 1900
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 0.4534,
      "step": 1910
    },
    {
      "epoch": 0.65,
      "learning_rate": 5e-05,
      "loss": 0.4537,
      "step": 1920
    },
    {
      "epoch": 0.65,
      "learning_rate": 5e-05,
      "loss": 0.4485,
      "step": 1930
    },
    {
      "epoch": 0.65,
      "learning_rate": 5e-05,
      "loss": 0.4576,
      "step": 1940
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 0.4518,
      "step": 1950
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 0.4601,
      "step": 1960
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 0.4561,
      "step": 1970
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 0.4479,
      "step": 1980
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 0.4519,
      "step": 1990
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 0.4562,
      "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
}