{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2002,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 3e-05,
      "loss": 0.6862,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 3e-05,
      "loss": 0.6634,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 3e-05,
      "loss": 0.6382,
      "step": 15
    },
    {
      "epoch": 0.01,
      "learning_rate": 3e-05,
      "loss": 0.6214,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 3e-05,
      "loss": 0.5948,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 3e-05,
      "loss": 0.5716,
      "step": 30
    },
    {
      "epoch": 0.02,
      "learning_rate": 3e-05,
      "loss": 0.567,
      "step": 35
    },
    {
      "epoch": 0.02,
      "learning_rate": 3e-05,
      "loss": 0.5596,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 3e-05,
      "loss": 0.553,
      "step": 45
    },
    {
      "epoch": 0.02,
      "learning_rate": 3e-05,
      "loss": 0.5403,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 3e-05,
      "loss": 0.5372,
      "step": 55
    },
    {
      "epoch": 0.03,
      "learning_rate": 3e-05,
      "loss": 0.5121,
      "step": 60
    },
    {
      "epoch": 0.03,
      "learning_rate": 3e-05,
      "loss": 0.522,
      "step": 65
    },
    {
      "epoch": 0.03,
      "learning_rate": 3e-05,
      "loss": 0.5278,
      "step": 70
    },
    {
      "epoch": 0.04,
      "learning_rate": 3e-05,
      "loss": 0.5316,
      "step": 75
    },
    {
      "epoch": 0.04,
      "learning_rate": 3e-05,
      "loss": 0.5005,
      "step": 80
    },
    {
      "epoch": 0.04,
      "learning_rate": 3e-05,
      "loss": 0.5078,
      "step": 85
    },
    {
      "epoch": 0.04,
      "learning_rate": 3e-05,
      "loss": 0.503,
      "step": 90
    },
    {
      "epoch": 0.05,
      "learning_rate": 3e-05,
      "loss": 0.5166,
      "step": 95
    },
    {
      "epoch": 0.05,
      "learning_rate": 3e-05,
      "loss": 0.5025,
      "step": 100
    },
    {
      "epoch": 0.05,
      "learning_rate": 3e-05,
      "loss": 0.5147,
      "step": 105
    },
    {
      "epoch": 0.05,
      "learning_rate": 3e-05,
      "loss": 0.5096,
      "step": 110
    },
    {
      "epoch": 0.06,
      "learning_rate": 3e-05,
      "loss": 0.4717,
      "step": 115
    },
    {
      "epoch": 0.06,
      "learning_rate": 3e-05,
      "loss": 0.4894,
      "step": 120
    },
    {
      "epoch": 0.06,
      "learning_rate": 3e-05,
      "loss": 0.5086,
      "step": 125
    },
    {
      "epoch": 0.06,
      "learning_rate": 3e-05,
      "loss": 0.5157,
      "step": 130
    },
    {
      "epoch": 0.07,
      "learning_rate": 3e-05,
      "loss": 0.5278,
      "step": 135
    },
    {
      "epoch": 0.07,
      "learning_rate": 3e-05,
      "loss": 0.5059,
      "step": 140
    },
    {
      "epoch": 0.07,
      "learning_rate": 3e-05,
      "loss": 0.5418,
      "step": 145
    },
    {
      "epoch": 0.07,
      "learning_rate": 3e-05,
      "loss": 0.5156,
      "step": 150
    },
    {
      "epoch": 0.08,
      "learning_rate": 3e-05,
      "loss": 0.505,
      "step": 155
    },
    {
      "epoch": 0.08,
      "learning_rate": 3e-05,
      "loss": 0.4993,
      "step": 160
    },
    {
      "epoch": 0.08,
      "learning_rate": 3e-05,
      "loss": 0.5003,
      "step": 165
    },
    {
      "epoch": 0.08,
      "learning_rate": 3e-05,
      "loss": 0.5121,
      "step": 170
    },
    {
      "epoch": 0.09,
      "learning_rate": 3e-05,
      "loss": 0.4802,
      "step": 175
    },
    {
      "epoch": 0.09,
      "learning_rate": 3e-05,
      "loss": 0.4893,
      "step": 180
    },
    {
      "epoch": 0.09,
      "learning_rate": 3e-05,
      "loss": 0.4967,
      "step": 185
    },
    {
      "epoch": 0.09,
      "learning_rate": 3e-05,
      "loss": 0.5016,
      "step": 190
    },
    {
      "epoch": 0.1,
      "learning_rate": 3e-05,
      "loss": 0.4854,
      "step": 195
    },
    {
      "epoch": 0.1,
      "learning_rate": 3e-05,
      "loss": 0.5074,
      "step": 200
    },
    {
      "epoch": 0.1,
      "learning_rate": 3e-05,
      "loss": 0.5108,
      "step": 205
    },
    {
      "epoch": 0.1,
      "learning_rate": 3e-05,
      "loss": 0.484,
      "step": 210
    },
    {
      "epoch": 0.11,
      "learning_rate": 3e-05,
      "loss": 0.5124,
      "step": 215
    },
    {
      "epoch": 0.11,
      "learning_rate": 3e-05,
      "loss": 0.5182,
      "step": 220
    },
    {
      "epoch": 0.11,
      "learning_rate": 3e-05,
      "loss": 0.5016,
      "step": 225
    },
    {
      "epoch": 0.11,
      "learning_rate": 3e-05,
      "loss": 0.4906,
      "step": 230
    },
    {
      "epoch": 0.12,
      "learning_rate": 3e-05,
      "loss": 0.4816,
      "step": 235
    },
    {
      "epoch": 0.12,
      "learning_rate": 3e-05,
      "loss": 0.5268,
      "step": 240
    },
    {
      "epoch": 0.12,
      "learning_rate": 3e-05,
      "loss": 0.5135,
      "step": 245
    },
    {
      "epoch": 0.12,
      "learning_rate": 3e-05,
      "loss": 0.4752,
      "step": 250
    },
    {
      "epoch": 0.13,
      "learning_rate": 3e-05,
      "loss": 0.4995,
      "step": 255
    },
    {
      "epoch": 0.13,
      "learning_rate": 3e-05,
      "loss": 0.4729,
      "step": 260
    },
    {
      "epoch": 0.13,
      "learning_rate": 3e-05,
      "loss": 0.5037,
      "step": 265
    },
    {
      "epoch": 0.13,
      "learning_rate": 3e-05,
      "loss": 0.4936,
      "step": 270
    },
    {
      "epoch": 0.14,
      "learning_rate": 3e-05,
      "loss": 0.5037,
      "step": 275
    },
    {
      "epoch": 0.14,
      "learning_rate": 3e-05,
      "loss": 0.518,
      "step": 280
    },
    {
      "epoch": 0.14,
      "learning_rate": 3e-05,
      "loss": 0.4979,
      "step": 285
    },
    {
      "epoch": 0.14,
      "learning_rate": 3e-05,
      "loss": 0.5089,
      "step": 290
    },
    {
      "epoch": 0.15,
      "learning_rate": 3e-05,
      "loss": 0.4839,
      "step": 295
    },
    {
      "epoch": 0.15,
      "learning_rate": 3e-05,
      "loss": 0.468,
      "step": 300
    },
    {
      "epoch": 0.15,
      "learning_rate": 3e-05,
      "loss": 0.4933,
      "step": 305
    },
    {
      "epoch": 0.15,
      "learning_rate": 3e-05,
      "loss": 0.4932,
      "step": 310
    },
    {
      "epoch": 0.16,
      "learning_rate": 3e-05,
      "loss": 0.5113,
      "step": 315
    },
    {
      "epoch": 0.16,
      "learning_rate": 3e-05,
      "loss": 0.5042,
      "step": 320
    },
    {
      "epoch": 0.16,
      "learning_rate": 3e-05,
      "loss": 0.4828,
      "step": 325
    },
    {
      "epoch": 0.16,
      "learning_rate": 3e-05,
      "loss": 0.4963,
      "step": 330
    },
    {
      "epoch": 0.17,
      "learning_rate": 3e-05,
      "loss": 0.4928,
      "step": 335
    },
    {
      "epoch": 0.17,
      "learning_rate": 3e-05,
      "loss": 0.4961,
      "step": 340
    },
    {
      "epoch": 0.17,
      "learning_rate": 3e-05,
      "loss": 0.4805,
      "step": 345
    },
    {
      "epoch": 0.17,
      "learning_rate": 3e-05,
      "loss": 0.4867,
      "step": 350
    },
    {
      "epoch": 0.18,
      "learning_rate": 3e-05,
      "loss": 0.5097,
      "step": 355
    },
    {
      "epoch": 0.18,
      "learning_rate": 3e-05,
      "loss": 0.4972,
      "step": 360
    },
    {
      "epoch": 0.18,
      "learning_rate": 3e-05,
      "loss": 0.4881,
      "step": 365
    },
    {
      "epoch": 0.18,
      "learning_rate": 3e-05,
      "loss": 0.4917,
      "step": 370
    },
    {
      "epoch": 0.19,
      "learning_rate": 3e-05,
      "loss": 0.5091,
      "step": 375
    },
    {
      "epoch": 0.19,
      "learning_rate": 3e-05,
      "loss": 0.4943,
      "step": 380
    },
    {
      "epoch": 0.19,
      "learning_rate": 3e-05,
      "loss": 0.488,
      "step": 385
    },
    {
      "epoch": 0.19,
      "learning_rate": 3e-05,
      "loss": 0.482,
      "step": 390
    },
    {
      "epoch": 0.2,
      "learning_rate": 3e-05,
      "loss": 0.4975,
      "step": 395
    },
    {
      "epoch": 0.2,
      "learning_rate": 3e-05,
      "loss": 0.4749,
      "step": 400
    },
    {
      "epoch": 0.2,
      "learning_rate": 3e-05,
      "loss": 0.483,
      "step": 405
    },
    {
      "epoch": 0.2,
      "learning_rate": 3e-05,
      "loss": 0.5165,
      "step": 410
    },
    {
      "epoch": 0.21,
      "learning_rate": 3e-05,
      "loss": 0.4933,
      "step": 415
    },
    {
      "epoch": 0.21,
      "learning_rate": 3e-05,
      "loss": 0.4923,
      "step": 420
    },
    {
      "epoch": 0.21,
      "learning_rate": 3e-05,
      "loss": 0.5069,
      "step": 425
    },
    {
      "epoch": 0.21,
      "learning_rate": 3e-05,
      "loss": 0.5154,
      "step": 430
    },
    {
      "epoch": 0.22,
      "learning_rate": 3e-05,
      "loss": 0.4937,
      "step": 435
    },
    {
      "epoch": 0.22,
      "learning_rate": 3e-05,
      "loss": 0.5119,
      "step": 440
    },
    {
      "epoch": 0.22,
      "learning_rate": 3e-05,
      "loss": 0.4616,
      "step": 445
    },
    {
      "epoch": 0.22,
      "learning_rate": 3e-05,
      "loss": 0.4819,
      "step": 450
    },
    {
      "epoch": 0.23,
      "learning_rate": 3e-05,
      "loss": 0.472,
      "step": 455
    },
    {
      "epoch": 0.23,
      "learning_rate": 3e-05,
      "loss": 0.4761,
      "step": 460
    },
    {
      "epoch": 0.23,
      "learning_rate": 3e-05,
      "loss": 0.4957,
      "step": 465
    },
    {
      "epoch": 0.23,
      "learning_rate": 3e-05,
      "loss": 0.5135,
      "step": 470
    },
    {
      "epoch": 0.24,
      "learning_rate": 3e-05,
      "loss": 0.5026,
      "step": 475
    },
    {
      "epoch": 0.24,
      "learning_rate": 3e-05,
      "loss": 0.4959,
      "step": 480
    },
    {
      "epoch": 0.24,
      "learning_rate": 3e-05,
      "loss": 0.5016,
      "step": 485
    },
    {
      "epoch": 0.24,
      "learning_rate": 3e-05,
      "loss": 0.5108,
      "step": 490
    },
    {
      "epoch": 0.25,
      "learning_rate": 3e-05,
      "loss": 0.5108,
      "step": 495
    },
    {
      "epoch": 0.25,
      "learning_rate": 3e-05,
      "loss": 0.5137,
      "step": 500
    },
    {
      "epoch": 0.25,
      "learning_rate": 3e-05,
      "loss": 0.4952,
      "step": 505
    },
    {
      "epoch": 0.25,
      "learning_rate": 3e-05,
      "loss": 0.5081,
      "step": 510
    },
    {
      "epoch": 0.26,
      "learning_rate": 3e-05,
      "loss": 0.5127,
      "step": 515
    },
    {
      "epoch": 0.26,
      "learning_rate": 3e-05,
      "loss": 0.502,
      "step": 520
    },
    {
      "epoch": 0.26,
      "learning_rate": 3e-05,
      "loss": 0.505,
      "step": 525
    },
    {
      "epoch": 0.26,
      "learning_rate": 3e-05,
      "loss": 0.4954,
      "step": 530
    },
    {
      "epoch": 0.27,
      "learning_rate": 3e-05,
      "loss": 0.4866,
      "step": 535
    },
    {
      "epoch": 0.27,
      "learning_rate": 3e-05,
      "loss": 0.4988,
      "step": 540
    },
    {
      "epoch": 0.27,
      "learning_rate": 3e-05,
      "loss": 0.5086,
      "step": 545
    },
    {
      "epoch": 0.27,
      "learning_rate": 3e-05,
      "loss": 0.4809,
      "step": 550
    },
    {
      "epoch": 0.28,
      "learning_rate": 3e-05,
      "loss": 0.4864,
      "step": 555
    },
    {
      "epoch": 0.28,
      "learning_rate": 3e-05,
      "loss": 0.4824,
      "step": 560
    },
    {
      "epoch": 0.28,
      "learning_rate": 3e-05,
      "loss": 0.5376,
      "step": 565
    },
    {
      "epoch": 0.28,
      "learning_rate": 3e-05,
      "loss": 0.4987,
      "step": 570
    },
    {
      "epoch": 0.29,
      "learning_rate": 3e-05,
      "loss": 0.4944,
      "step": 575
    },
    {
      "epoch": 0.29,
      "learning_rate": 3e-05,
      "loss": 0.5306,
      "step": 580
    },
    {
      "epoch": 0.29,
      "learning_rate": 3e-05,
      "loss": 0.4995,
      "step": 585
    },
    {
      "epoch": 0.29,
      "learning_rate": 3e-05,
      "loss": 0.4892,
      "step": 590
    },
    {
      "epoch": 0.3,
      "learning_rate": 3e-05,
      "loss": 0.5356,
      "step": 595
    },
    {
      "epoch": 0.3,
      "learning_rate": 3e-05,
      "loss": 0.4931,
      "step": 600
    },
    {
      "epoch": 0.3,
      "learning_rate": 3e-05,
      "loss": 0.5032,
      "step": 605
    },
    {
      "epoch": 0.3,
      "learning_rate": 3e-05,
      "loss": 0.4911,
      "step": 610
    },
    {
      "epoch": 0.31,
      "learning_rate": 3e-05,
      "loss": 0.4989,
      "step": 615
    },
    {
      "epoch": 0.31,
      "learning_rate": 3e-05,
      "loss": 0.4995,
      "step": 620
    },
    {
      "epoch": 0.31,
      "learning_rate": 3e-05,
      "loss": 0.4907,
      "step": 625
    },
    {
      "epoch": 0.31,
      "learning_rate": 3e-05,
      "loss": 0.4982,
      "step": 630
    },
    {
      "epoch": 0.32,
      "learning_rate": 3e-05,
      "loss": 0.4569,
      "step": 635
    },
    {
      "epoch": 0.32,
      "learning_rate": 3e-05,
      "loss": 0.4932,
      "step": 640
    },
    {
      "epoch": 0.32,
      "learning_rate": 3e-05,
      "loss": 0.4795,
      "step": 645
    },
    {
      "epoch": 0.32,
      "learning_rate": 3e-05,
      "loss": 0.4873,
      "step": 650
    },
    {
      "epoch": 0.33,
      "learning_rate": 3e-05,
      "loss": 0.4743,
      "step": 655
    },
    {
      "epoch": 0.33,
      "learning_rate": 3e-05,
      "loss": 0.5099,
      "step": 660
    },
    {
      "epoch": 0.33,
      "learning_rate": 3e-05,
      "loss": 0.5015,
      "step": 665
    },
    {
      "epoch": 0.33,
      "learning_rate": 3e-05,
      "loss": 0.49,
      "step": 670
    },
    {
      "epoch": 0.34,
      "learning_rate": 3e-05,
      "loss": 0.4901,
      "step": 675
    },
    {
      "epoch": 0.34,
      "learning_rate": 3e-05,
      "loss": 0.5034,
      "step": 680
    },
    {
      "epoch": 0.34,
      "learning_rate": 3e-05,
      "loss": 0.4973,
      "step": 685
    },
    {
      "epoch": 0.34,
      "learning_rate": 3e-05,
      "loss": 0.4769,
      "step": 690
    },
    {
      "epoch": 0.35,
      "learning_rate": 3e-05,
      "loss": 0.4844,
      "step": 695
    },
    {
      "epoch": 0.35,
      "learning_rate": 3e-05,
      "loss": 0.5038,
      "step": 700
    },
    {
      "epoch": 0.35,
      "learning_rate": 3e-05,
      "loss": 0.4849,
      "step": 705
    },
    {
      "epoch": 0.35,
      "learning_rate": 3e-05,
      "loss": 0.4698,
      "step": 710
    },
    {
      "epoch": 0.36,
      "learning_rate": 3e-05,
      "loss": 0.4789,
      "step": 715
    },
    {
      "epoch": 0.36,
      "learning_rate": 3e-05,
      "loss": 0.5255,
      "step": 720
    },
    {
      "epoch": 0.36,
      "learning_rate": 3e-05,
      "loss": 0.5074,
      "step": 725
    },
    {
      "epoch": 0.36,
      "learning_rate": 3e-05,
      "loss": 0.5016,
      "step": 730
    },
    {
      "epoch": 0.37,
      "learning_rate": 3e-05,
      "loss": 0.5089,
      "step": 735
    },
    {
      "epoch": 0.37,
      "learning_rate": 3e-05,
      "loss": 0.5036,
      "step": 740
    },
    {
      "epoch": 0.37,
      "learning_rate": 3e-05,
      "loss": 0.4816,
      "step": 745
    },
    {
      "epoch": 0.37,
      "learning_rate": 3e-05,
      "loss": 0.4777,
      "step": 750
    },
    {
      "epoch": 0.38,
      "learning_rate": 3e-05,
      "loss": 0.4713,
      "step": 755
    },
    {
      "epoch": 0.38,
      "learning_rate": 3e-05,
      "loss": 0.4929,
      "step": 760
    },
    {
      "epoch": 0.38,
      "learning_rate": 3e-05,
      "loss": 0.5038,
      "step": 765
    },
    {
      "epoch": 0.38,
      "learning_rate": 3e-05,
      "loss": 0.4729,
      "step": 770
    },
    {
      "epoch": 0.39,
      "learning_rate": 3e-05,
      "loss": 0.5002,
      "step": 775
    },
    {
      "epoch": 0.39,
      "learning_rate": 3e-05,
      "loss": 0.5151,
      "step": 780
    },
    {
      "epoch": 0.39,
      "learning_rate": 3e-05,
      "loss": 0.5039,
      "step": 785
    },
    {
      "epoch": 0.39,
      "learning_rate": 3e-05,
      "loss": 0.4866,
      "step": 790
    },
    {
      "epoch": 0.4,
      "learning_rate": 3e-05,
      "loss": 0.484,
      "step": 795
    },
    {
      "epoch": 0.4,
      "learning_rate": 3e-05,
      "loss": 0.5007,
      "step": 800
    },
    {
      "epoch": 0.4,
      "learning_rate": 3e-05,
      "loss": 0.5001,
      "step": 805
    },
    {
      "epoch": 0.4,
      "learning_rate": 3e-05,
      "loss": 0.508,
      "step": 810
    },
    {
      "epoch": 0.41,
      "learning_rate": 3e-05,
      "loss": 0.4831,
      "step": 815
    },
    {
      "epoch": 0.41,
      "learning_rate": 3e-05,
      "loss": 0.5185,
      "step": 820
    },
    {
      "epoch": 0.41,
      "learning_rate": 3e-05,
      "loss": 0.5131,
      "step": 825
    },
    {
      "epoch": 0.41,
      "learning_rate": 3e-05,
      "loss": 0.4893,
      "step": 830
    },
    {
      "epoch": 0.42,
      "learning_rate": 3e-05,
      "loss": 0.4797,
      "step": 835
    },
    {
      "epoch": 0.42,
      "learning_rate": 3e-05,
      "loss": 0.4935,
      "step": 840
    },
    {
      "epoch": 0.42,
      "learning_rate": 3e-05,
      "loss": 0.4896,
      "step": 845
    },
    {
      "epoch": 0.42,
      "learning_rate": 3e-05,
      "loss": 0.4897,
      "step": 850
    },
    {
      "epoch": 0.43,
      "learning_rate": 3e-05,
      "loss": 0.4805,
      "step": 855
    },
    {
      "epoch": 0.43,
      "learning_rate": 3e-05,
      "loss": 0.4933,
      "step": 860
    },
    {
      "epoch": 0.43,
      "learning_rate": 3e-05,
      "loss": 0.4978,
      "step": 865
    },
    {
      "epoch": 0.43,
      "learning_rate": 3e-05,
      "loss": 0.4755,
      "step": 870
    },
    {
      "epoch": 0.44,
      "learning_rate": 3e-05,
      "loss": 0.5159,
      "step": 875
    },
    {
      "epoch": 0.44,
      "learning_rate": 3e-05,
      "loss": 0.5054,
      "step": 880
    },
    {
      "epoch": 0.44,
      "learning_rate": 3e-05,
      "loss": 0.495,
      "step": 885
    },
    {
      "epoch": 0.44,
      "learning_rate": 3e-05,
      "loss": 0.4791,
      "step": 890
    },
    {
      "epoch": 0.45,
      "learning_rate": 3e-05,
      "loss": 0.474,
      "step": 895
    },
    {
      "epoch": 0.45,
      "learning_rate": 3e-05,
      "loss": 0.4795,
      "step": 900
    },
    {
      "epoch": 0.45,
      "learning_rate": 3e-05,
      "loss": 0.4826,
      "step": 905
    },
    {
      "epoch": 0.45,
      "learning_rate": 3e-05,
      "loss": 0.5009,
      "step": 910
    },
    {
      "epoch": 0.46,
      "learning_rate": 3e-05,
      "loss": 0.4823,
      "step": 915
    },
    {
      "epoch": 0.46,
      "learning_rate": 3e-05,
      "loss": 0.5183,
      "step": 920
    },
    {
      "epoch": 0.46,
      "learning_rate": 3e-05,
      "loss": 0.4707,
      "step": 925
    },
    {
      "epoch": 0.46,
      "learning_rate": 3e-05,
      "loss": 0.5133,
      "step": 930
    },
    {
      "epoch": 0.47,
      "learning_rate": 3e-05,
      "loss": 0.4967,
      "step": 935
    },
    {
      "epoch": 0.47,
      "learning_rate": 3e-05,
      "loss": 0.4843,
      "step": 940
    },
    {
      "epoch": 0.47,
      "learning_rate": 3e-05,
      "loss": 0.4693,
      "step": 945
    },
    {
      "epoch": 0.47,
      "learning_rate": 3e-05,
      "loss": 0.4671,
      "step": 950
    },
    {
      "epoch": 0.48,
      "learning_rate": 3e-05,
      "loss": 0.4949,
      "step": 955
    },
    {
      "epoch": 0.48,
      "learning_rate": 3e-05,
      "loss": 0.5138,
      "step": 960
    },
    {
      "epoch": 0.48,
      "learning_rate": 3e-05,
      "loss": 0.4853,
      "step": 965
    },
    {
      "epoch": 0.48,
      "learning_rate": 3e-05,
      "loss": 0.4916,
      "step": 970
    },
    {
      "epoch": 0.49,
      "learning_rate": 3e-05,
      "loss": 0.5141,
      "step": 975
    },
    {
      "epoch": 0.49,
      "learning_rate": 3e-05,
      "loss": 0.4842,
      "step": 980
    },
    {
      "epoch": 0.49,
      "learning_rate": 3e-05,
      "loss": 0.4933,
      "step": 985
    },
    {
      "epoch": 0.49,
      "learning_rate": 3e-05,
      "loss": 0.5022,
      "step": 990
    },
    {
      "epoch": 0.5,
      "learning_rate": 3e-05,
      "loss": 0.4879,
      "step": 995
    },
    {
      "epoch": 0.5,
      "learning_rate": 3e-05,
      "loss": 0.5193,
      "step": 1000
    },
    {
      "epoch": 0.5,
      "learning_rate": 3e-05,
      "loss": 0.473,
      "step": 1005
    },
    {
      "epoch": 0.5,
      "learning_rate": 3e-05,
      "loss": 0.4785,
      "step": 1010
    },
    {
      "epoch": 0.51,
      "learning_rate": 3e-05,
      "loss": 0.485,
      "step": 1015
    },
    {
      "epoch": 0.51,
      "learning_rate": 3e-05,
      "loss": 0.4803,
      "step": 1020
    },
    {
      "epoch": 0.51,
      "learning_rate": 3e-05,
      "loss": 0.4773,
      "step": 1025
    },
    {
      "epoch": 0.51,
      "learning_rate": 3e-05,
      "loss": 0.4771,
      "step": 1030
    },
    {
      "epoch": 0.52,
      "learning_rate": 3e-05,
      "loss": 0.5051,
      "step": 1035
    },
    {
      "epoch": 0.52,
      "learning_rate": 3e-05,
      "loss": 0.4433,
      "step": 1040
    },
    {
      "epoch": 0.52,
      "learning_rate": 3e-05,
      "loss": 0.4708,
      "step": 1045
    },
    {
      "epoch": 0.52,
      "learning_rate": 3e-05,
      "loss": 0.5028,
      "step": 1050
    },
    {
      "epoch": 0.53,
      "learning_rate": 3e-05,
      "loss": 0.5053,
      "step": 1055
    },
    {
      "epoch": 0.53,
      "learning_rate": 3e-05,
      "loss": 0.45,
      "step": 1060
    },
    {
      "epoch": 0.53,
      "learning_rate": 3e-05,
      "loss": 0.5022,
      "step": 1065
    },
    {
      "epoch": 0.53,
      "learning_rate": 3e-05,
      "loss": 0.4668,
      "step": 1070
    },
    {
      "epoch": 0.54,
      "learning_rate": 3e-05,
      "loss": 0.5334,
      "step": 1075
    },
    {
      "epoch": 0.54,
      "learning_rate": 3e-05,
      "loss": 0.4898,
      "step": 1080
    },
    {
      "epoch": 0.54,
      "learning_rate": 3e-05,
      "loss": 0.4664,
      "step": 1085
    },
    {
      "epoch": 0.54,
      "learning_rate": 3e-05,
      "loss": 0.4657,
      "step": 1090
    },
    {
      "epoch": 0.55,
      "learning_rate": 3e-05,
      "loss": 0.4598,
      "step": 1095
    },
    {
      "epoch": 0.55,
      "learning_rate": 3e-05,
      "loss": 0.48,
      "step": 1100
    },
    {
      "epoch": 0.55,
      "learning_rate": 3e-05,
      "loss": 0.4838,
      "step": 1105
    },
    {
      "epoch": 0.55,
      "learning_rate": 3e-05,
      "loss": 0.5101,
      "step": 1110
    },
    {
      "epoch": 0.56,
      "learning_rate": 3e-05,
      "loss": 0.5332,
      "step": 1115
    },
    {
      "epoch": 0.56,
      "learning_rate": 3e-05,
      "loss": 0.4935,
      "step": 1120
    },
    {
      "epoch": 0.56,
      "learning_rate": 3e-05,
      "loss": 0.4861,
      "step": 1125
    },
    {
      "epoch": 0.56,
      "learning_rate": 3e-05,
      "loss": 0.4654,
      "step": 1130
    },
    {
      "epoch": 0.57,
      "learning_rate": 3e-05,
      "loss": 0.4708,
      "step": 1135
    },
    {
      "epoch": 0.57,
      "learning_rate": 3e-05,
      "loss": 0.49,
      "step": 1140
    },
    {
      "epoch": 0.57,
      "learning_rate": 3e-05,
      "loss": 0.4992,
      "step": 1145
    },
    {
      "epoch": 0.57,
      "learning_rate": 3e-05,
      "loss": 0.501,
      "step": 1150
    },
    {
      "epoch": 0.58,
      "learning_rate": 3e-05,
      "loss": 0.507,
      "step": 1155
    },
    {
      "epoch": 0.58,
      "learning_rate": 3e-05,
      "loss": 0.4858,
      "step": 1160
    },
    {
      "epoch": 0.58,
      "learning_rate": 3e-05,
      "loss": 0.5118,
      "step": 1165
    },
    {
      "epoch": 0.58,
      "learning_rate": 3e-05,
      "loss": 0.4985,
      "step": 1170
    },
    {
      "epoch": 0.59,
      "learning_rate": 3e-05,
      "loss": 0.4821,
      "step": 1175
    },
    {
      "epoch": 0.59,
      "learning_rate": 3e-05,
      "loss": 0.4957,
      "step": 1180
    },
    {
      "epoch": 0.59,
      "learning_rate": 3e-05,
      "loss": 0.4722,
      "step": 1185
    },
    {
      "epoch": 0.59,
      "learning_rate": 3e-05,
      "loss": 0.5096,
      "step": 1190
    },
    {
      "epoch": 0.6,
      "learning_rate": 3e-05,
      "loss": 0.4748,
      "step": 1195
    },
    {
      "epoch": 0.6,
      "learning_rate": 3e-05,
      "loss": 0.5074,
      "step": 1200
    },
    {
      "epoch": 0.6,
      "learning_rate": 3e-05,
      "loss": 0.4901,
      "step": 1205
    },
    {
      "epoch": 0.6,
      "learning_rate": 3e-05,
      "loss": 0.4931,
      "step": 1210
    },
    {
      "epoch": 0.61,
      "learning_rate": 3e-05,
      "loss": 0.489,
      "step": 1215
    },
    {
      "epoch": 0.61,
      "learning_rate": 3e-05,
      "loss": 0.48,
      "step": 1220
    },
    {
      "epoch": 0.61,
      "learning_rate": 3e-05,
      "loss": 0.5009,
      "step": 1225
    },
    {
      "epoch": 0.61,
      "learning_rate": 3e-05,
      "loss": 0.4795,
      "step": 1230
    },
    {
      "epoch": 0.62,
      "learning_rate": 3e-05,
      "loss": 0.515,
      "step": 1235
    },
    {
      "epoch": 0.62,
      "learning_rate": 3e-05,
      "loss": 0.488,
      "step": 1240
    },
    {
      "epoch": 0.62,
      "learning_rate": 3e-05,
      "loss": 0.484,
      "step": 1245
    },
    {
      "epoch": 0.62,
      "learning_rate": 3e-05,
      "loss": 0.4764,
      "step": 1250
    },
    {
      "epoch": 0.63,
      "learning_rate": 3e-05,
      "loss": 0.4849,
      "step": 1255
    },
    {
      "epoch": 0.63,
      "learning_rate": 3e-05,
      "loss": 0.487,
      "step": 1260
    },
    {
      "epoch": 0.63,
      "learning_rate": 3e-05,
      "loss": 0.4801,
      "step": 1265
    },
    {
      "epoch": 0.63,
      "learning_rate": 3e-05,
      "loss": 0.4574,
      "step": 1270
    },
    {
      "epoch": 0.64,
      "learning_rate": 3e-05,
      "loss": 0.4595,
      "step": 1275
    },
    {
      "epoch": 0.64,
      "learning_rate": 3e-05,
      "loss": 0.5119,
      "step": 1280
    },
    {
      "epoch": 0.64,
      "learning_rate": 3e-05,
      "loss": 0.5091,
      "step": 1285
    },
    {
      "epoch": 0.64,
      "learning_rate": 3e-05,
      "loss": 0.4992,
      "step": 1290
    },
    {
      "epoch": 0.65,
      "learning_rate": 3e-05,
      "loss": 0.4997,
      "step": 1295
    },
    {
      "epoch": 0.65,
      "learning_rate": 3e-05,
      "loss": 0.4899,
      "step": 1300
    },
    {
      "epoch": 0.65,
      "learning_rate": 3e-05,
      "loss": 0.4852,
      "step": 1305
    },
    {
      "epoch": 0.65,
      "learning_rate": 3e-05,
      "loss": 0.5055,
      "step": 1310
    },
    {
      "epoch": 0.66,
      "learning_rate": 3e-05,
      "loss": 0.5298,
      "step": 1315
    },
    {
      "epoch": 0.66,
      "learning_rate": 3e-05,
      "loss": 0.4985,
      "step": 1320
    },
    {
      "epoch": 0.66,
      "learning_rate": 3e-05,
      "loss": 0.5047,
      "step": 1325
    },
    {
      "epoch": 0.66,
      "learning_rate": 3e-05,
      "loss": 0.4877,
      "step": 1330
    },
    {
      "epoch": 0.67,
      "learning_rate": 3e-05,
      "loss": 0.4611,
      "step": 1335
    },
    {
      "epoch": 0.67,
      "learning_rate": 3e-05,
      "loss": 0.505,
      "step": 1340
    },
    {
      "epoch": 0.67,
      "learning_rate": 3e-05,
      "loss": 0.5215,
      "step": 1345
    },
    {
      "epoch": 0.67,
      "learning_rate": 3e-05,
      "loss": 0.4831,
      "step": 1350
    },
    {
      "epoch": 0.68,
      "learning_rate": 3e-05,
      "loss": 0.5026,
      "step": 1355
    },
    {
      "epoch": 0.68,
      "learning_rate": 3e-05,
      "loss": 0.4686,
      "step": 1360
    },
    {
      "epoch": 0.68,
      "learning_rate": 3e-05,
      "loss": 0.4784,
      "step": 1365
    },
    {
      "epoch": 0.68,
      "learning_rate": 3e-05,
      "loss": 0.487,
      "step": 1370
    },
    {
      "epoch": 0.69,
      "learning_rate": 3e-05,
      "loss": 0.4702,
      "step": 1375
    },
    {
      "epoch": 0.69,
      "learning_rate": 3e-05,
      "loss": 0.4865,
      "step": 1380
    },
    {
      "epoch": 0.69,
      "learning_rate": 3e-05,
      "loss": 0.5031,
      "step": 1385
    },
    {
      "epoch": 0.69,
      "learning_rate": 3e-05,
      "loss": 0.5097,
      "step": 1390
    },
    {
      "epoch": 0.7,
      "learning_rate": 3e-05,
      "loss": 0.4828,
      "step": 1395
    },
    {
      "epoch": 0.7,
      "learning_rate": 3e-05,
      "loss": 0.5069,
      "step": 1400
    },
    {
      "epoch": 0.7,
      "learning_rate": 3e-05,
      "loss": 0.4758,
      "step": 1405
    },
    {
      "epoch": 0.7,
      "learning_rate": 3e-05,
      "loss": 0.4799,
      "step": 1410
    },
    {
      "epoch": 0.71,
      "learning_rate": 3e-05,
      "loss": 0.5066,
      "step": 1415
    },
    {
      "epoch": 0.71,
      "learning_rate": 3e-05,
      "loss": 0.4946,
      "step": 1420
    },
    {
      "epoch": 0.71,
      "learning_rate": 3e-05,
      "loss": 0.5126,
      "step": 1425
    },
    {
      "epoch": 0.71,
      "learning_rate": 3e-05,
      "loss": 0.484,
      "step": 1430
    },
    {
      "epoch": 0.72,
      "learning_rate": 3e-05,
      "loss": 0.5277,
      "step": 1435
    },
    {
      "epoch": 0.72,
      "learning_rate": 3e-05,
      "loss": 0.4852,
      "step": 1440
    },
    {
      "epoch": 0.72,
      "learning_rate": 3e-05,
      "loss": 0.4911,
      "step": 1445
    },
    {
      "epoch": 0.72,
      "learning_rate": 3e-05,
      "loss": 0.4712,
      "step": 1450
    },
    {
      "epoch": 0.73,
      "learning_rate": 3e-05,
      "loss": 0.4782,
      "step": 1455
    },
    {
      "epoch": 0.73,
      "learning_rate": 3e-05,
      "loss": 0.4826,
      "step": 1460
    },
    {
      "epoch": 0.73,
      "learning_rate": 3e-05,
      "loss": 0.4877,
      "step": 1465
    },
    {
      "epoch": 0.73,
      "learning_rate": 3e-05,
      "loss": 0.4867,
      "step": 1470
    },
    {
      "epoch": 0.74,
      "learning_rate": 3e-05,
      "loss": 0.4997,
      "step": 1475
    },
    {
      "epoch": 0.74,
      "learning_rate": 3e-05,
      "loss": 0.478,
      "step": 1480
    },
    {
      "epoch": 0.74,
      "learning_rate": 3e-05,
      "loss": 0.4662,
      "step": 1485
    },
    {
      "epoch": 0.74,
      "learning_rate": 3e-05,
      "loss": 0.4942,
      "step": 1490
    },
    {
      "epoch": 0.75,
      "learning_rate": 3e-05,
      "loss": 0.4547,
      "step": 1495
    },
    {
      "epoch": 0.75,
      "learning_rate": 3e-05,
      "loss": 0.4829,
      "step": 1500
    },
    {
      "epoch": 0.75,
      "learning_rate": 3e-05,
      "loss": 0.4711,
      "step": 1505
    },
    {
      "epoch": 0.75,
      "learning_rate": 3e-05,
      "loss": 0.5094,
      "step": 1510
    },
    {
      "epoch": 0.76,
      "learning_rate": 3e-05,
      "loss": 0.4734,
      "step": 1515
    },
    {
      "epoch": 0.76,
      "learning_rate": 3e-05,
      "loss": 0.5037,
      "step": 1520
    },
    {
      "epoch": 0.76,
      "learning_rate": 3e-05,
      "loss": 0.4871,
      "step": 1525
    },
    {
      "epoch": 0.76,
      "learning_rate": 3e-05,
      "loss": 0.5062,
      "step": 1530
    },
    {
      "epoch": 0.77,
      "learning_rate": 3e-05,
      "loss": 0.4767,
      "step": 1535
    },
    {
      "epoch": 0.77,
      "learning_rate": 3e-05,
      "loss": 0.4845,
      "step": 1540
    },
    {
      "epoch": 0.77,
      "learning_rate": 3e-05,
      "loss": 0.4631,
      "step": 1545
    },
    {
      "epoch": 0.77,
      "learning_rate": 3e-05,
      "loss": 0.4955,
      "step": 1550
    },
    {
      "epoch": 0.78,
      "learning_rate": 3e-05,
      "loss": 0.4818,
      "step": 1555
    },
    {
      "epoch": 0.78,
      "learning_rate": 3e-05,
      "loss": 0.471,
      "step": 1560
    },
    {
      "epoch": 0.78,
      "learning_rate": 3e-05,
      "loss": 0.5047,
      "step": 1565
    },
    {
      "epoch": 0.78,
      "learning_rate": 3e-05,
      "loss": 0.4794,
      "step": 1570
    },
    {
      "epoch": 0.79,
      "learning_rate": 3e-05,
      "loss": 0.4887,
      "step": 1575
    },
    {
      "epoch": 0.79,
      "learning_rate": 3e-05,
      "loss": 0.5055,
      "step": 1580
    },
    {
      "epoch": 0.79,
      "learning_rate": 3e-05,
      "loss": 0.4965,
      "step": 1585
    },
    {
      "epoch": 0.79,
      "learning_rate": 3e-05,
      "loss": 0.4712,
      "step": 1590
    },
    {
      "epoch": 0.8,
      "learning_rate": 3e-05,
      "loss": 0.4827,
      "step": 1595
    },
    {
      "epoch": 0.8,
      "learning_rate": 3e-05,
      "loss": 0.4702,
      "step": 1600
    },
    {
      "epoch": 0.8,
      "learning_rate": 3e-05,
      "loss": 0.4725,
      "step": 1605
    },
    {
      "epoch": 0.8,
      "learning_rate": 3e-05,
      "loss": 0.4634,
      "step": 1610
    },
    {
      "epoch": 0.81,
      "learning_rate": 3e-05,
      "loss": 0.4701,
      "step": 1615
    },
    {
      "epoch": 0.81,
      "learning_rate": 3e-05,
      "loss": 0.5085,
      "step": 1620
    },
    {
      "epoch": 0.81,
      "learning_rate": 3e-05,
      "loss": 0.4937,
      "step": 1625
    },
    {
      "epoch": 0.81,
      "learning_rate": 3e-05,
      "loss": 0.4974,
      "step": 1630
    },
    {
      "epoch": 0.82,
      "learning_rate": 3e-05,
      "loss": 0.4722,
      "step": 1635
    },
    {
      "epoch": 0.82,
      "learning_rate": 3e-05,
      "loss": 0.4689,
      "step": 1640
    },
    {
      "epoch": 0.82,
      "learning_rate": 3e-05,
      "loss": 0.5093,
      "step": 1645
    },
    {
      "epoch": 0.82,
      "learning_rate": 3e-05,
      "loss": 0.5012,
      "step": 1650
    },
    {
      "epoch": 0.83,
      "learning_rate": 3e-05,
      "loss": 0.4917,
      "step": 1655
    },
    {
      "epoch": 0.83,
      "learning_rate": 3e-05,
      "loss": 0.4931,
      "step": 1660
    },
    {
      "epoch": 0.83,
      "learning_rate": 3e-05,
      "loss": 0.4939,
      "step": 1665
    },
    {
      "epoch": 0.83,
      "learning_rate": 3e-05,
      "loss": 0.5113,
      "step": 1670
    },
    {
      "epoch": 0.84,
      "learning_rate": 3e-05,
      "loss": 0.4847,
      "step": 1675
    },
    {
      "epoch": 0.84,
      "learning_rate": 3e-05,
      "loss": 0.4815,
      "step": 1680
    },
    {
      "epoch": 0.84,
      "learning_rate": 3e-05,
      "loss": 0.5155,
      "step": 1685
    },
    {
      "epoch": 0.84,
      "learning_rate": 3e-05,
      "loss": 0.4896,
      "step": 1690
    },
    {
      "epoch": 0.85,
      "learning_rate": 3e-05,
      "loss": 0.4801,
      "step": 1695
    },
    {
      "epoch": 0.85,
      "learning_rate": 3e-05,
      "loss": 0.48,
      "step": 1700
    },
    {
      "epoch": 0.85,
      "learning_rate": 3e-05,
      "loss": 0.4849,
      "step": 1705
    },
    {
      "epoch": 0.85,
      "learning_rate": 3e-05,
      "loss": 0.4625,
      "step": 1710
    },
    {
      "epoch": 0.86,
      "learning_rate": 3e-05,
      "loss": 0.4715,
      "step": 1715
    },
    {
      "epoch": 0.86,
      "learning_rate": 3e-05,
      "loss": 0.4905,
      "step": 1720
    },
    {
      "epoch": 0.86,
      "learning_rate": 3e-05,
      "loss": 0.4877,
      "step": 1725
    },
    {
      "epoch": 0.86,
      "learning_rate": 3e-05,
      "loss": 0.4724,
      "step": 1730
    },
    {
      "epoch": 0.87,
      "learning_rate": 3e-05,
      "loss": 0.4941,
      "step": 1735
    },
    {
      "epoch": 0.87,
      "learning_rate": 3e-05,
      "loss": 0.4744,
      "step": 1740
    },
    {
      "epoch": 0.87,
      "learning_rate": 3e-05,
      "loss": 0.4943,
      "step": 1745
    },
    {
      "epoch": 0.87,
      "learning_rate": 3e-05,
      "loss": 0.4974,
      "step": 1750
    },
    {
      "epoch": 0.88,
      "learning_rate": 3e-05,
      "loss": 0.5005,
      "step": 1755
    },
    {
      "epoch": 0.88,
      "learning_rate": 3e-05,
      "loss": 0.5023,
      "step": 1760
    },
    {
      "epoch": 0.88,
      "learning_rate": 3e-05,
      "loss": 0.5012,
      "step": 1765
    },
    {
      "epoch": 0.88,
      "learning_rate": 3e-05,
      "loss": 0.4434,
      "step": 1770
    },
    {
      "epoch": 0.89,
      "learning_rate": 3e-05,
      "loss": 0.501,
      "step": 1775
    },
    {
      "epoch": 0.89,
      "learning_rate": 3e-05,
      "loss": 0.4699,
      "step": 1780
    },
    {
      "epoch": 0.89,
      "learning_rate": 3e-05,
      "loss": 0.4846,
      "step": 1785
    },
    {
      "epoch": 0.89,
      "learning_rate": 3e-05,
      "loss": 0.463,
      "step": 1790
    },
    {
      "epoch": 0.9,
      "learning_rate": 3e-05,
      "loss": 0.4612,
      "step": 1795
    },
    {
      "epoch": 0.9,
      "learning_rate": 3e-05,
      "loss": 0.484,
      "step": 1800
    },
    {
      "epoch": 0.9,
      "learning_rate": 3e-05,
      "loss": 0.4848,
      "step": 1805
    },
    {
      "epoch": 0.9,
      "learning_rate": 3e-05,
      "loss": 0.4762,
      "step": 1810
    },
    {
      "epoch": 0.91,
      "learning_rate": 3e-05,
      "loss": 0.4905,
      "step": 1815
    },
    {
      "epoch": 0.91,
      "learning_rate": 3e-05,
      "loss": 0.462,
      "step": 1820
    },
    {
      "epoch": 0.91,
      "learning_rate": 3e-05,
      "loss": 0.4988,
      "step": 1825
    },
    {
      "epoch": 0.91,
      "learning_rate": 3e-05,
      "loss": 0.4629,
      "step": 1830
    },
    {
      "epoch": 0.92,
      "learning_rate": 3e-05,
      "loss": 0.4971,
      "step": 1835
    },
    {
      "epoch": 0.92,
      "learning_rate": 3e-05,
      "loss": 0.4702,
      "step": 1840
    },
    {
      "epoch": 0.92,
      "learning_rate": 3e-05,
      "loss": 0.5022,
      "step": 1845
    },
    {
      "epoch": 0.92,
      "learning_rate": 3e-05,
      "loss": 0.4857,
      "step": 1850
    },
    {
      "epoch": 0.93,
      "learning_rate": 3e-05,
      "loss": 0.4779,
      "step": 1855
    },
    {
      "epoch": 0.93,
      "learning_rate": 3e-05,
      "loss": 0.5093,
      "step": 1860
    },
    {
      "epoch": 0.93,
      "learning_rate": 3e-05,
      "loss": 0.501,
      "step": 1865
    },
    {
      "epoch": 0.93,
      "learning_rate": 3e-05,
      "loss": 0.4836,
      "step": 1870
    },
    {
      "epoch": 0.94,
      "learning_rate": 3e-05,
      "loss": 0.5167,
      "step": 1875
    },
    {
      "epoch": 0.94,
      "learning_rate": 3e-05,
      "loss": 0.4801,
      "step": 1880
    },
    {
      "epoch": 0.94,
      "learning_rate": 3e-05,
      "loss": 0.4531,
      "step": 1885
    },
    {
      "epoch": 0.94,
      "learning_rate": 3e-05,
      "loss": 0.4929,
      "step": 1890
    },
    {
      "epoch": 0.95,
      "learning_rate": 3e-05,
      "loss": 0.4888,
      "step": 1895
    },
    {
      "epoch": 0.95,
      "learning_rate": 3e-05,
      "loss": 0.4919,
      "step": 1900
    },
    {
      "epoch": 0.95,
      "learning_rate": 3e-05,
      "loss": 0.4596,
      "step": 1905
    },
    {
      "epoch": 0.95,
      "learning_rate": 3e-05,
      "loss": 0.4869,
      "step": 1910
    },
    {
      "epoch": 0.96,
      "learning_rate": 3e-05,
      "loss": 0.4825,
      "step": 1915
    },
    {
      "epoch": 0.96,
      "learning_rate": 3e-05,
      "loss": 0.513,
      "step": 1920
    },
    {
      "epoch": 0.96,
      "learning_rate": 3e-05,
      "loss": 0.4972,
      "step": 1925
    },
    {
      "epoch": 0.96,
      "learning_rate": 3e-05,
      "loss": 0.4663,
      "step": 1930
    },
    {
      "epoch": 0.97,
      "learning_rate": 3e-05,
      "loss": 0.4654,
      "step": 1935
    },
    {
      "epoch": 0.97,
      "learning_rate": 3e-05,
      "loss": 0.4889,
      "step": 1940
    },
    {
      "epoch": 0.97,
      "learning_rate": 3e-05,
      "loss": 0.5205,
      "step": 1945
    },
    {
      "epoch": 0.97,
      "learning_rate": 3e-05,
      "loss": 0.4829,
      "step": 1950
    },
    {
      "epoch": 0.98,
      "learning_rate": 3e-05,
      "loss": 0.4809,
      "step": 1955
    },
    {
      "epoch": 0.98,
      "learning_rate": 3e-05,
      "loss": 0.4788,
      "step": 1960
    },
    {
      "epoch": 0.98,
      "learning_rate": 3e-05,
      "loss": 0.4946,
      "step": 1965
    },
    {
      "epoch": 0.98,
      "learning_rate": 3e-05,
      "loss": 0.4888,
      "step": 1970
    },
    {
      "epoch": 0.99,
      "learning_rate": 3e-05,
      "loss": 0.4633,
      "step": 1975
    },
    {
      "epoch": 0.99,
      "learning_rate": 3e-05,
      "loss": 0.465,
      "step": 1980
    },
    {
      "epoch": 0.99,
      "learning_rate": 3e-05,
      "loss": 0.4588,
      "step": 1985
    },
    {
      "epoch": 0.99,
      "learning_rate": 3e-05,
      "loss": 0.4724,
      "step": 1990
    },
    {
      "epoch": 1.0,
      "learning_rate": 3e-05,
      "loss": 0.5044,
      "step": 1995
    },
    {
      "epoch": 1.0,
      "learning_rate": 3e-05,
      "loss": 0.5002,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "step": 2002,
      "total_flos": 7.452987125170176e+18,
      "train_loss": 0.49472963369333306,
      "train_runtime": 1452.6312,
      "train_samples_per_second": 22.043,
      "train_steps_per_second": 1.378
    }
  ],
  "logging_steps": 5,
  "max_steps": 2002,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 7.452987125170176e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}