{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 28.0,
  "eval_steps": 500,
  "global_step": 4900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05714285714285714,
      "grad_norm": 17.375,
      "learning_rate": 8e-05,
      "loss": 2.6583,
      "step": 10
    },
    {
      "epoch": 0.11428571428571428,
      "grad_norm": 10.1875,
      "learning_rate": 8e-05,
      "loss": 1.4692,
      "step": 20
    },
    {
      "epoch": 0.17142857142857143,
      "grad_norm": 9.1875,
      "learning_rate": 8e-05,
      "loss": 1.3322,
      "step": 30
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 9.0,
      "learning_rate": 8e-05,
      "loss": 1.2474,
      "step": 40
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 8.5,
      "learning_rate": 8e-05,
      "loss": 1.2488,
      "step": 50
    },
    {
      "epoch": 0.34285714285714286,
      "grad_norm": 22.75,
      "learning_rate": 8e-05,
      "loss": 1.0061,
      "step": 60
    },
    {
      "epoch": 0.4,
      "grad_norm": 16.0,
      "learning_rate": 8e-05,
      "loss": 1.1315,
      "step": 70
    },
    {
      "epoch": 0.45714285714285713,
      "grad_norm": 8.375,
      "learning_rate": 8e-05,
      "loss": 0.9975,
      "step": 80
    },
    {
      "epoch": 0.5142857142857142,
      "grad_norm": 7.8125,
      "learning_rate": 8e-05,
      "loss": 1.2164,
      "step": 90
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 15.0,
      "learning_rate": 8e-05,
      "loss": 1.1313,
      "step": 100
    },
    {
      "epoch": 0.6285714285714286,
      "grad_norm": 8.1875,
      "learning_rate": 8e-05,
      "loss": 1.1283,
      "step": 110
    },
    {
      "epoch": 0.6857142857142857,
      "grad_norm": 9.5625,
      "learning_rate": 8e-05,
      "loss": 1.1187,
      "step": 120
    },
    {
      "epoch": 0.7428571428571429,
      "grad_norm": 15.5,
      "learning_rate": 8e-05,
      "loss": 0.997,
      "step": 130
    },
    {
      "epoch": 0.8,
      "grad_norm": 18.75,
      "learning_rate": 8e-05,
      "loss": 0.8798,
      "step": 140
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 7.15625,
      "learning_rate": 8e-05,
      "loss": 0.7601,
      "step": 150
    },
    {
      "epoch": 0.9142857142857143,
      "grad_norm": 8.75,
      "learning_rate": 8e-05,
      "loss": 0.6739,
      "step": 160
    },
    {
      "epoch": 0.9714285714285714,
      "grad_norm": 7.5625,
      "learning_rate": 8e-05,
      "loss": 0.9945,
      "step": 170
    },
    {
      "epoch": 1.0285714285714285,
      "grad_norm": 7.25,
      "learning_rate": 8e-05,
      "loss": 0.7509,
      "step": 180
    },
    {
      "epoch": 1.0857142857142856,
      "grad_norm": 7.4375,
      "learning_rate": 8e-05,
      "loss": 0.731,
      "step": 190
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 7.625,
      "learning_rate": 8e-05,
      "loss": 0.8672,
      "step": 200
    },
    {
      "epoch": 1.2,
      "grad_norm": 7.53125,
      "learning_rate": 8e-05,
      "loss": 0.7409,
      "step": 210
    },
    {
      "epoch": 1.2571428571428571,
      "grad_norm": 8.375,
      "learning_rate": 8e-05,
      "loss": 0.7572,
      "step": 220
    },
    {
      "epoch": 1.3142857142857143,
      "grad_norm": 7.53125,
      "learning_rate": 8e-05,
      "loss": 0.7667,
      "step": 230
    },
    {
      "epoch": 1.3714285714285714,
      "grad_norm": 7.6875,
      "learning_rate": 8e-05,
      "loss": 0.7002,
      "step": 240
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 8.4375,
      "learning_rate": 8e-05,
      "loss": 0.6992,
      "step": 250
    },
    {
      "epoch": 1.4857142857142858,
      "grad_norm": 15.0,
      "learning_rate": 8e-05,
      "loss": 0.7765,
      "step": 260
    },
    {
      "epoch": 1.5428571428571427,
      "grad_norm": 14.6875,
      "learning_rate": 8e-05,
      "loss": 0.5869,
      "step": 270
    },
    {
      "epoch": 1.6,
      "grad_norm": 14.8125,
      "learning_rate": 8e-05,
      "loss": 0.8268,
      "step": 280
    },
    {
      "epoch": 1.657142857142857,
      "grad_norm": 6.84375,
      "learning_rate": 8e-05,
      "loss": 0.7856,
      "step": 290
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 14.75,
      "learning_rate": 8e-05,
      "loss": 0.6341,
      "step": 300
    },
    {
      "epoch": 1.7714285714285714,
      "grad_norm": 7.21875,
      "learning_rate": 8e-05,
      "loss": 0.794,
      "step": 310
    },
    {
      "epoch": 1.8285714285714287,
      "grad_norm": 14.0625,
      "learning_rate": 8e-05,
      "loss": 0.6861,
      "step": 320
    },
    {
      "epoch": 1.8857142857142857,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.5252,
      "step": 330
    },
    {
      "epoch": 1.9428571428571428,
      "grad_norm": 8.1875,
      "learning_rate": 8e-05,
      "loss": 0.7826,
      "step": 340
    },
    {
      "epoch": 2.0,
      "grad_norm": 15.6875,
      "learning_rate": 8e-05,
      "loss": 0.7721,
      "step": 350
    },
    {
      "epoch": 2.057142857142857,
      "grad_norm": 14.4375,
      "learning_rate": 8e-05,
      "loss": 0.6508,
      "step": 360
    },
    {
      "epoch": 2.1142857142857143,
      "grad_norm": 14.625,
      "learning_rate": 8e-05,
      "loss": 0.6328,
      "step": 370
    },
    {
      "epoch": 2.1714285714285713,
      "grad_norm": 14.375,
      "learning_rate": 8e-05,
      "loss": 0.5822,
      "step": 380
    },
    {
      "epoch": 2.2285714285714286,
      "grad_norm": 7.46875,
      "learning_rate": 8e-05,
      "loss": 0.6849,
      "step": 390
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.5679,
      "step": 400
    },
    {
      "epoch": 2.342857142857143,
      "grad_norm": 6.6875,
      "learning_rate": 8e-05,
      "loss": 0.552,
      "step": 410
    },
    {
      "epoch": 2.4,
      "grad_norm": 10.5625,
      "learning_rate": 8e-05,
      "loss": 0.5635,
      "step": 420
    },
    {
      "epoch": 2.4571428571428573,
      "grad_norm": 13.625,
      "learning_rate": 8e-05,
      "loss": 0.6423,
      "step": 430
    },
    {
      "epoch": 2.5142857142857142,
      "grad_norm": 6.90625,
      "learning_rate": 8e-05,
      "loss": 0.6436,
      "step": 440
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 9.3125,
      "learning_rate": 8e-05,
      "loss": 0.5537,
      "step": 450
    },
    {
      "epoch": 2.6285714285714286,
      "grad_norm": 16.0,
      "learning_rate": 8e-05,
      "loss": 0.5902,
      "step": 460
    },
    {
      "epoch": 2.685714285714286,
      "grad_norm": 15.4375,
      "learning_rate": 8e-05,
      "loss": 0.6193,
      "step": 470
    },
    {
      "epoch": 2.742857142857143,
      "grad_norm": 6.5,
      "learning_rate": 8e-05,
      "loss": 0.6679,
      "step": 480
    },
    {
      "epoch": 2.8,
      "grad_norm": 9.375,
      "learning_rate": 8e-05,
      "loss": 0.5469,
      "step": 490
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 7.5,
      "learning_rate": 8e-05,
      "loss": 0.653,
      "step": 500
    },
    {
      "epoch": 2.857142857142857,
      "eval_loss": 6.050149440765381,
      "eval_runtime": 7.8659,
      "eval_samples_per_second": 63.566,
      "eval_steps_per_second": 63.566,
      "step": 500
    },
    {
      "epoch": 2.914285714285714,
      "grad_norm": 7.15625,
      "learning_rate": 8e-05,
      "loss": 0.6111,
      "step": 510
    },
    {
      "epoch": 2.9714285714285715,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.5121,
      "step": 520
    },
    {
      "epoch": 3.0285714285714285,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.5591,
      "step": 530
    },
    {
      "epoch": 3.085714285714286,
      "grad_norm": 10.5,
      "learning_rate": 8e-05,
      "loss": 0.5514,
      "step": 540
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.4903,
      "step": 550
    },
    {
      "epoch": 3.2,
      "grad_norm": 7.625,
      "learning_rate": 8e-05,
      "loss": 0.5356,
      "step": 560
    },
    {
      "epoch": 3.257142857142857,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.5641,
      "step": 570
    },
    {
      "epoch": 3.314285714285714,
      "grad_norm": 6.875,
      "learning_rate": 8e-05,
      "loss": 0.5261,
      "step": 580
    },
    {
      "epoch": 3.3714285714285714,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.46,
      "step": 590
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 13.5625,
      "learning_rate": 8e-05,
      "loss": 0.4915,
      "step": 600
    },
    {
      "epoch": 3.4857142857142858,
      "grad_norm": 11.1875,
      "learning_rate": 8e-05,
      "loss": 0.5197,
      "step": 610
    },
    {
      "epoch": 3.5428571428571427,
      "grad_norm": 14.8125,
      "learning_rate": 8e-05,
      "loss": 0.5629,
      "step": 620
    },
    {
      "epoch": 3.6,
      "grad_norm": 7.46875,
      "learning_rate": 8e-05,
      "loss": 0.5783,
      "step": 630
    },
    {
      "epoch": 3.657142857142857,
      "grad_norm": 6.59375,
      "learning_rate": 8e-05,
      "loss": 0.5087,
      "step": 640
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 7.28125,
      "learning_rate": 8e-05,
      "loss": 0.4895,
      "step": 650
    },
    {
      "epoch": 3.7714285714285714,
      "grad_norm": 6.6875,
      "learning_rate": 8e-05,
      "loss": 0.5505,
      "step": 660
    },
    {
      "epoch": 3.8285714285714287,
      "grad_norm": 12.5625,
      "learning_rate": 8e-05,
      "loss": 0.5888,
      "step": 670
    },
    {
      "epoch": 3.8857142857142857,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.5155,
      "step": 680
    },
    {
      "epoch": 3.942857142857143,
      "grad_norm": 14.5,
      "learning_rate": 8e-05,
      "loss": 0.4999,
      "step": 690
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.8125,
      "learning_rate": 8e-05,
      "loss": 0.5957,
      "step": 700
    },
    {
      "epoch": 4.057142857142857,
      "grad_norm": 6.625,
      "learning_rate": 8e-05,
      "loss": 0.377,
      "step": 710
    },
    {
      "epoch": 4.114285714285714,
      "grad_norm": 6.6875,
      "learning_rate": 8e-05,
      "loss": 0.3669,
      "step": 720
    },
    {
      "epoch": 4.171428571428572,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.4941,
      "step": 730
    },
    {
      "epoch": 4.228571428571429,
      "grad_norm": 14.25,
      "learning_rate": 8e-05,
      "loss": 0.3737,
      "step": 740
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 7.65625,
      "learning_rate": 8e-05,
      "loss": 0.4366,
      "step": 750
    },
    {
      "epoch": 4.3428571428571425,
      "grad_norm": 9.5625,
      "learning_rate": 8e-05,
      "loss": 0.4763,
      "step": 760
    },
    {
      "epoch": 4.4,
      "grad_norm": 14.1875,
      "learning_rate": 8e-05,
      "loss": 0.4481,
      "step": 770
    },
    {
      "epoch": 4.457142857142857,
      "grad_norm": 13.6875,
      "learning_rate": 8e-05,
      "loss": 0.4707,
      "step": 780
    },
    {
      "epoch": 4.514285714285714,
      "grad_norm": 15.875,
      "learning_rate": 8e-05,
      "loss": 0.4706,
      "step": 790
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 5.78125,
      "learning_rate": 8e-05,
      "loss": 0.3984,
      "step": 800
    },
    {
      "epoch": 4.628571428571428,
      "grad_norm": 7.34375,
      "learning_rate": 8e-05,
      "loss": 0.4837,
      "step": 810
    },
    {
      "epoch": 4.685714285714286,
      "grad_norm": 13.8125,
      "learning_rate": 8e-05,
      "loss": 0.4525,
      "step": 820
    },
    {
      "epoch": 4.742857142857143,
      "grad_norm": 13.75,
      "learning_rate": 8e-05,
      "loss": 0.4504,
      "step": 830
    },
    {
      "epoch": 4.8,
      "grad_norm": 6.53125,
      "learning_rate": 8e-05,
      "loss": 0.4679,
      "step": 840
    },
    {
      "epoch": 4.857142857142857,
      "grad_norm": 6.5,
      "learning_rate": 8e-05,
      "loss": 0.4356,
      "step": 850
    },
    {
      "epoch": 4.914285714285715,
      "grad_norm": 13.75,
      "learning_rate": 8e-05,
      "loss": 0.4415,
      "step": 860
    },
    {
      "epoch": 4.9714285714285715,
      "grad_norm": 12.25,
      "learning_rate": 8e-05,
      "loss": 0.4542,
      "step": 870
    },
    {
      "epoch": 5.0285714285714285,
      "grad_norm": 7.0,
      "learning_rate": 8e-05,
      "loss": 0.4882,
      "step": 880
    },
    {
      "epoch": 5.085714285714285,
      "grad_norm": 5.0,
      "learning_rate": 8e-05,
      "loss": 0.3197,
      "step": 890
    },
    {
      "epoch": 5.142857142857143,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.3614,
      "step": 900
    },
    {
      "epoch": 5.2,
      "grad_norm": 5.84375,
      "learning_rate": 8e-05,
      "loss": 0.3744,
      "step": 910
    },
    {
      "epoch": 5.257142857142857,
      "grad_norm": 10.0,
      "learning_rate": 8e-05,
      "loss": 0.3811,
      "step": 920
    },
    {
      "epoch": 5.314285714285714,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.4349,
      "step": 930
    },
    {
      "epoch": 5.371428571428572,
      "grad_norm": 5.5625,
      "learning_rate": 8e-05,
      "loss": 0.3253,
      "step": 940
    },
    {
      "epoch": 5.428571428571429,
      "grad_norm": 5.46875,
      "learning_rate": 8e-05,
      "loss": 0.4326,
      "step": 950
    },
    {
      "epoch": 5.485714285714286,
      "grad_norm": 4.8125,
      "learning_rate": 8e-05,
      "loss": 0.3973,
      "step": 960
    },
    {
      "epoch": 5.542857142857143,
      "grad_norm": 6.65625,
      "learning_rate": 8e-05,
      "loss": 0.3823,
      "step": 970
    },
    {
      "epoch": 5.6,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.4287,
      "step": 980
    },
    {
      "epoch": 5.6571428571428575,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.4264,
      "step": 990
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 6.46875,
      "learning_rate": 8e-05,
      "loss": 0.3398,
      "step": 1000
    },
    {
      "epoch": 5.714285714285714,
      "eval_loss": 6.685291767120361,
      "eval_runtime": 7.9381,
      "eval_samples_per_second": 62.987,
      "eval_steps_per_second": 62.987,
      "step": 1000
    },
    {
      "epoch": 5.771428571428571,
      "grad_norm": 7.03125,
      "learning_rate": 8e-05,
      "loss": 0.3123,
      "step": 1010
    },
    {
      "epoch": 5.828571428571428,
      "grad_norm": 6.59375,
      "learning_rate": 8e-05,
      "loss": 0.2876,
      "step": 1020
    },
    {
      "epoch": 5.885714285714286,
      "grad_norm": 4.84375,
      "learning_rate": 8e-05,
      "loss": 0.3921,
      "step": 1030
    },
    {
      "epoch": 5.942857142857143,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.3555,
      "step": 1040
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.4375,
      "learning_rate": 8e-05,
      "loss": 0.4566,
      "step": 1050
    },
    {
      "epoch": 6.057142857142857,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.3109,
      "step": 1060
    },
    {
      "epoch": 6.114285714285714,
      "grad_norm": 11.375,
      "learning_rate": 8e-05,
      "loss": 0.4008,
      "step": 1070
    },
    {
      "epoch": 6.171428571428572,
      "grad_norm": 5.71875,
      "learning_rate": 8e-05,
      "loss": 0.3019,
      "step": 1080
    },
    {
      "epoch": 6.228571428571429,
      "grad_norm": 3.671875,
      "learning_rate": 8e-05,
      "loss": 0.3714,
      "step": 1090
    },
    {
      "epoch": 6.285714285714286,
      "grad_norm": 4.5,
      "learning_rate": 8e-05,
      "loss": 0.2272,
      "step": 1100
    },
    {
      "epoch": 6.3428571428571425,
      "grad_norm": 13.9375,
      "learning_rate": 8e-05,
      "loss": 0.3953,
      "step": 1110
    },
    {
      "epoch": 6.4,
      "grad_norm": 11.6875,
      "learning_rate": 8e-05,
      "loss": 0.3158,
      "step": 1120
    },
    {
      "epoch": 6.457142857142857,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.3624,
      "step": 1130
    },
    {
      "epoch": 6.514285714285714,
      "grad_norm": 6.5625,
      "learning_rate": 8e-05,
      "loss": 0.3308,
      "step": 1140
    },
    {
      "epoch": 6.571428571428571,
      "grad_norm": 3.6875,
      "learning_rate": 8e-05,
      "loss": 0.2394,
      "step": 1150
    },
    {
      "epoch": 6.628571428571428,
      "grad_norm": 3.1875,
      "learning_rate": 8e-05,
      "loss": 0.306,
      "step": 1160
    },
    {
      "epoch": 6.685714285714286,
      "grad_norm": 9.75,
      "learning_rate": 8e-05,
      "loss": 0.3698,
      "step": 1170
    },
    {
      "epoch": 6.742857142857143,
      "grad_norm": 3.53125,
      "learning_rate": 8e-05,
      "loss": 0.3485,
      "step": 1180
    },
    {
      "epoch": 6.8,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.342,
      "step": 1190
    },
    {
      "epoch": 6.857142857142857,
      "grad_norm": 28.5,
      "learning_rate": 8e-05,
      "loss": 0.3361,
      "step": 1200
    },
    {
      "epoch": 6.914285714285715,
      "grad_norm": 7.84375,
      "learning_rate": 8e-05,
      "loss": 0.3526,
      "step": 1210
    },
    {
      "epoch": 6.9714285714285715,
      "grad_norm": 8.0,
      "learning_rate": 8e-05,
      "loss": 0.3662,
      "step": 1220
    },
    {
      "epoch": 7.0285714285714285,
      "grad_norm": 4.21875,
      "learning_rate": 8e-05,
      "loss": 0.3469,
      "step": 1230
    },
    {
      "epoch": 7.085714285714285,
      "grad_norm": 7.84375,
      "learning_rate": 8e-05,
      "loss": 0.2773,
      "step": 1240
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 11.9375,
      "learning_rate": 8e-05,
      "loss": 0.2524,
      "step": 1250
    },
    {
      "epoch": 7.2,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.389,
      "step": 1260
    },
    {
      "epoch": 7.257142857142857,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.2766,
      "step": 1270
    },
    {
      "epoch": 7.314285714285714,
      "grad_norm": 2.359375,
      "learning_rate": 8e-05,
      "loss": 0.3331,
      "step": 1280
    },
    {
      "epoch": 7.371428571428572,
      "grad_norm": 11.4375,
      "learning_rate": 8e-05,
      "loss": 0.2712,
      "step": 1290
    },
    {
      "epoch": 7.428571428571429,
      "grad_norm": 11.625,
      "learning_rate": 8e-05,
      "loss": 0.4544,
      "step": 1300
    },
    {
      "epoch": 7.485714285714286,
      "grad_norm": 3.6875,
      "learning_rate": 8e-05,
      "loss": 0.3047,
      "step": 1310
    },
    {
      "epoch": 7.542857142857143,
      "grad_norm": 3.609375,
      "learning_rate": 8e-05,
      "loss": 0.2868,
      "step": 1320
    },
    {
      "epoch": 7.6,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.3294,
      "step": 1330
    },
    {
      "epoch": 7.6571428571428575,
      "grad_norm": 2.78125,
      "learning_rate": 8e-05,
      "loss": 0.3252,
      "step": 1340
    },
    {
      "epoch": 7.714285714285714,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.2855,
      "step": 1350
    },
    {
      "epoch": 7.771428571428571,
      "grad_norm": 3.71875,
      "learning_rate": 8e-05,
      "loss": 0.3158,
      "step": 1360
    },
    {
      "epoch": 7.828571428571428,
      "grad_norm": 7.28125,
      "learning_rate": 8e-05,
      "loss": 0.2767,
      "step": 1370
    },
    {
      "epoch": 7.885714285714286,
      "grad_norm": 2.53125,
      "learning_rate": 8e-05,
      "loss": 0.2423,
      "step": 1380
    },
    {
      "epoch": 7.942857142857143,
      "grad_norm": 6.03125,
      "learning_rate": 8e-05,
      "loss": 0.2216,
      "step": 1390
    },
    {
      "epoch": 8.0,
      "grad_norm": 4.96875,
      "learning_rate": 8e-05,
      "loss": 0.3818,
      "step": 1400
    },
    {
      "epoch": 8.057142857142857,
      "grad_norm": 5.1875,
      "learning_rate": 8e-05,
      "loss": 0.2588,
      "step": 1410
    },
    {
      "epoch": 8.114285714285714,
      "grad_norm": 12.3125,
      "learning_rate": 8e-05,
      "loss": 0.3463,
      "step": 1420
    },
    {
      "epoch": 8.17142857142857,
      "grad_norm": 2.8125,
      "learning_rate": 8e-05,
      "loss": 0.2003,
      "step": 1430
    },
    {
      "epoch": 8.228571428571428,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.3551,
      "step": 1440
    },
    {
      "epoch": 8.285714285714286,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.3274,
      "step": 1450
    },
    {
      "epoch": 8.342857142857143,
      "grad_norm": 1.3046875,
      "learning_rate": 8e-05,
      "loss": 0.2906,
      "step": 1460
    },
    {
      "epoch": 8.4,
      "grad_norm": 1.9453125,
      "learning_rate": 8e-05,
      "loss": 0.3483,
      "step": 1470
    },
    {
      "epoch": 8.457142857142857,
      "grad_norm": 12.125,
      "learning_rate": 8e-05,
      "loss": 0.2865,
      "step": 1480
    },
    {
      "epoch": 8.514285714285714,
      "grad_norm": 13.4375,
      "learning_rate": 8e-05,
      "loss": 0.334,
      "step": 1490
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 5.125,
      "learning_rate": 8e-05,
      "loss": 0.2815,
      "step": 1500
    },
    {
      "epoch": 8.571428571428571,
      "eval_loss": 7.375263690948486,
      "eval_runtime": 7.8964,
      "eval_samples_per_second": 63.32,
      "eval_steps_per_second": 63.32,
      "step": 1500
    },
    {
      "epoch": 8.628571428571428,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.1534,
      "step": 1510
    },
    {
      "epoch": 8.685714285714285,
      "grad_norm": 10.6875,
      "learning_rate": 8e-05,
      "loss": 0.2531,
      "step": 1520
    },
    {
      "epoch": 8.742857142857144,
      "grad_norm": 5.28125,
      "learning_rate": 8e-05,
      "loss": 0.3664,
      "step": 1530
    },
    {
      "epoch": 8.8,
      "grad_norm": 3.390625,
      "learning_rate": 8e-05,
      "loss": 0.2378,
      "step": 1540
    },
    {
      "epoch": 8.857142857142858,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.3167,
      "step": 1550
    },
    {
      "epoch": 8.914285714285715,
      "grad_norm": 3.0,
      "learning_rate": 8e-05,
      "loss": 0.2882,
      "step": 1560
    },
    {
      "epoch": 8.971428571428572,
      "grad_norm": 1.984375,
      "learning_rate": 8e-05,
      "loss": 0.2518,
      "step": 1570
    },
    {
      "epoch": 9.028571428571428,
      "grad_norm": 9.5625,
      "learning_rate": 8e-05,
      "loss": 0.1575,
      "step": 1580
    },
    {
      "epoch": 9.085714285714285,
      "grad_norm": 3.96875,
      "learning_rate": 8e-05,
      "loss": 0.1379,
      "step": 1590
    },
    {
      "epoch": 9.142857142857142,
      "grad_norm": 10.4375,
      "learning_rate": 8e-05,
      "loss": 0.3056,
      "step": 1600
    },
    {
      "epoch": 9.2,
      "grad_norm": 3.234375,
      "learning_rate": 8e-05,
      "loss": 0.2755,
      "step": 1610
    },
    {
      "epoch": 9.257142857142856,
      "grad_norm": 1.078125,
      "learning_rate": 8e-05,
      "loss": 0.2795,
      "step": 1620
    },
    {
      "epoch": 9.314285714285715,
      "grad_norm": 12.25,
      "learning_rate": 8e-05,
      "loss": 0.2939,
      "step": 1630
    },
    {
      "epoch": 9.371428571428572,
      "grad_norm": 4.0,
      "learning_rate": 8e-05,
      "loss": 0.1852,
      "step": 1640
    },
    {
      "epoch": 9.428571428571429,
      "grad_norm": 9.1875,
      "learning_rate": 8e-05,
      "loss": 0.2803,
      "step": 1650
    },
    {
      "epoch": 9.485714285714286,
      "grad_norm": 13.625,
      "learning_rate": 8e-05,
      "loss": 0.2908,
      "step": 1660
    },
    {
      "epoch": 9.542857142857143,
      "grad_norm": 13.4375,
      "learning_rate": 8e-05,
      "loss": 0.3115,
      "step": 1670
    },
    {
      "epoch": 9.6,
      "grad_norm": 3.796875,
      "learning_rate": 8e-05,
      "loss": 0.4095,
      "step": 1680
    },
    {
      "epoch": 9.657142857142857,
      "grad_norm": 2.484375,
      "learning_rate": 8e-05,
      "loss": 0.2278,
      "step": 1690
    },
    {
      "epoch": 9.714285714285714,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.3629,
      "step": 1700
    },
    {
      "epoch": 9.771428571428572,
      "grad_norm": 2.8125,
      "learning_rate": 8e-05,
      "loss": 0.2527,
      "step": 1710
    },
    {
      "epoch": 9.82857142857143,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.2514,
      "step": 1720
    },
    {
      "epoch": 9.885714285714286,
      "grad_norm": 6.90625,
      "learning_rate": 8e-05,
      "loss": 0.238,
      "step": 1730
    },
    {
      "epoch": 9.942857142857143,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.2673,
      "step": 1740
    },
    {
      "epoch": 10.0,
      "grad_norm": 11.5,
      "learning_rate": 8e-05,
      "loss": 0.3351,
      "step": 1750
    },
    {
      "epoch": 10.057142857142857,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.255,
      "step": 1760
    },
    {
      "epoch": 10.114285714285714,
      "grad_norm": 3.171875,
      "learning_rate": 8e-05,
      "loss": 0.2478,
      "step": 1770
    },
    {
      "epoch": 10.17142857142857,
      "grad_norm": 1.25,
      "learning_rate": 8e-05,
      "loss": 0.1548,
      "step": 1780
    },
    {
      "epoch": 10.228571428571428,
      "grad_norm": 3.078125,
      "learning_rate": 8e-05,
      "loss": 0.3276,
      "step": 1790
    },
    {
      "epoch": 10.285714285714286,
      "grad_norm": 2.21875,
      "learning_rate": 8e-05,
      "loss": 0.2305,
      "step": 1800
    },
    {
      "epoch": 10.342857142857143,
      "grad_norm": 11.3125,
      "learning_rate": 8e-05,
      "loss": 0.2625,
      "step": 1810
    },
    {
      "epoch": 10.4,
      "grad_norm": 10.375,
      "learning_rate": 8e-05,
      "loss": 0.2509,
      "step": 1820
    },
    {
      "epoch": 10.457142857142857,
      "grad_norm": 2.25,
      "learning_rate": 8e-05,
      "loss": 0.2091,
      "step": 1830
    },
    {
      "epoch": 10.514285714285714,
      "grad_norm": 3.5,
      "learning_rate": 8e-05,
      "loss": 0.2416,
      "step": 1840
    },
    {
      "epoch": 10.571428571428571,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.3821,
      "step": 1850
    },
    {
      "epoch": 10.628571428571428,
      "grad_norm": 11.0625,
      "learning_rate": 8e-05,
      "loss": 0.3066,
      "step": 1860
    },
    {
      "epoch": 10.685714285714285,
      "grad_norm": 1.078125,
      "learning_rate": 8e-05,
      "loss": 0.1716,
      "step": 1870
    },
    {
      "epoch": 10.742857142857144,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.325,
      "step": 1880
    },
    {
      "epoch": 10.8,
      "grad_norm": 2.046875,
      "learning_rate": 8e-05,
      "loss": 0.2183,
      "step": 1890
    },
    {
      "epoch": 10.857142857142858,
      "grad_norm": 3.90625,
      "learning_rate": 8e-05,
      "loss": 0.2748,
      "step": 1900
    },
    {
      "epoch": 10.914285714285715,
      "grad_norm": 11.5,
      "learning_rate": 8e-05,
      "loss": 0.2748,
      "step": 1910
    },
    {
      "epoch": 10.971428571428572,
      "grad_norm": 2.765625,
      "learning_rate": 8e-05,
      "loss": 0.2093,
      "step": 1920
    },
    {
      "epoch": 11.028571428571428,
      "grad_norm": 2.78125,
      "learning_rate": 8e-05,
      "loss": 0.3204,
      "step": 1930
    },
    {
      "epoch": 11.085714285714285,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.2653,
      "step": 1940
    },
    {
      "epoch": 11.142857142857142,
      "grad_norm": 1.4453125,
      "learning_rate": 8e-05,
      "loss": 0.1379,
      "step": 1950
    },
    {
      "epoch": 11.2,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.3495,
      "step": 1960
    },
    {
      "epoch": 11.257142857142856,
      "grad_norm": 0.96875,
      "learning_rate": 8e-05,
      "loss": 0.0931,
      "step": 1970
    },
    {
      "epoch": 11.314285714285715,
      "grad_norm": 2.1875,
      "learning_rate": 8e-05,
      "loss": 0.3159,
      "step": 1980
    },
    {
      "epoch": 11.371428571428572,
      "grad_norm": 1.7734375,
      "learning_rate": 8e-05,
      "loss": 0.2834,
      "step": 1990
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 2.46875,
      "learning_rate": 8e-05,
      "loss": 0.298,
      "step": 2000
    },
    {
      "epoch": 11.428571428571429,
      "eval_loss": 7.439001083374023,
      "eval_runtime": 7.8289,
      "eval_samples_per_second": 63.866,
      "eval_steps_per_second": 63.866,
      "step": 2000
    },
    {
      "epoch": 11.485714285714286,
      "grad_norm": 2.578125,
      "learning_rate": 8e-05,
      "loss": 0.2116,
      "step": 2010
    },
    {
      "epoch": 11.542857142857143,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.3693,
      "step": 2020
    },
    {
      "epoch": 11.6,
      "grad_norm": 1.40625,
      "learning_rate": 8e-05,
      "loss": 0.1701,
      "step": 2030
    },
    {
      "epoch": 11.657142857142857,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.2949,
      "step": 2040
    },
    {
      "epoch": 11.714285714285714,
      "grad_norm": 1.5546875,
      "learning_rate": 8e-05,
      "loss": 0.207,
      "step": 2050
    },
    {
      "epoch": 11.771428571428572,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.372,
      "step": 2060
    },
    {
      "epoch": 11.82857142857143,
      "grad_norm": 2.296875,
      "learning_rate": 8e-05,
      "loss": 0.3137,
      "step": 2070
    },
    {
      "epoch": 11.885714285714286,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.3433,
      "step": 2080
    },
    {
      "epoch": 11.942857142857143,
      "grad_norm": 1.875,
      "learning_rate": 8e-05,
      "loss": 0.2197,
      "step": 2090
    },
    {
      "epoch": 12.0,
      "grad_norm": 14.3125,
      "learning_rate": 8e-05,
      "loss": 0.2616,
      "step": 2100
    },
    {
      "epoch": 12.057142857142857,
      "grad_norm": 13.9375,
      "learning_rate": 8e-05,
      "loss": 0.3095,
      "step": 2110
    },
    {
      "epoch": 12.114285714285714,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.2683,
      "step": 2120
    },
    {
      "epoch": 12.17142857142857,
      "grad_norm": 3.015625,
      "learning_rate": 8e-05,
      "loss": 0.2083,
      "step": 2130
    },
    {
      "epoch": 12.228571428571428,
      "grad_norm": 1.6015625,
      "learning_rate": 8e-05,
      "loss": 0.17,
      "step": 2140
    },
    {
      "epoch": 12.285714285714286,
      "grad_norm": 2.875,
      "learning_rate": 8e-05,
      "loss": 0.1634,
      "step": 2150
    },
    {
      "epoch": 12.342857142857143,
      "grad_norm": 1.5,
      "learning_rate": 8e-05,
      "loss": 0.3009,
      "step": 2160
    },
    {
      "epoch": 12.4,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.1983,
      "step": 2170
    },
    {
      "epoch": 12.457142857142857,
      "grad_norm": 1.765625,
      "learning_rate": 8e-05,
      "loss": 0.244,
      "step": 2180
    },
    {
      "epoch": 12.514285714285714,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.3322,
      "step": 2190
    },
    {
      "epoch": 12.571428571428571,
      "grad_norm": 13.75,
      "learning_rate": 8e-05,
      "loss": 0.2492,
      "step": 2200
    },
    {
      "epoch": 12.628571428571428,
      "grad_norm": 1.625,
      "learning_rate": 8e-05,
      "loss": 0.2741,
      "step": 2210
    },
    {
      "epoch": 12.685714285714285,
      "grad_norm": 3.21875,
      "learning_rate": 8e-05,
      "loss": 0.2587,
      "step": 2220
    },
    {
      "epoch": 12.742857142857144,
      "grad_norm": 2.515625,
      "learning_rate": 8e-05,
      "loss": 0.3302,
      "step": 2230
    },
    {
      "epoch": 12.8,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.2935,
      "step": 2240
    },
    {
      "epoch": 12.857142857142858,
      "grad_norm": 1.265625,
      "learning_rate": 8e-05,
      "loss": 0.1977,
      "step": 2250
    },
    {
      "epoch": 12.914285714285715,
      "grad_norm": 11.9375,
      "learning_rate": 8e-05,
      "loss": 0.3059,
      "step": 2260
    },
    {
      "epoch": 12.971428571428572,
      "grad_norm": 11.625,
      "learning_rate": 8e-05,
      "loss": 0.2862,
      "step": 2270
    },
    {
      "epoch": 13.028571428571428,
      "grad_norm": 2.546875,
      "learning_rate": 8e-05,
      "loss": 0.2819,
      "step": 2280
    },
    {
      "epoch": 13.085714285714285,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.2451,
      "step": 2290
    },
    {
      "epoch": 13.142857142857142,
      "grad_norm": 1.3125,
      "learning_rate": 8e-05,
      "loss": 0.1908,
      "step": 2300
    },
    {
      "epoch": 13.2,
      "grad_norm": 10.125,
      "learning_rate": 8e-05,
      "loss": 0.1633,
      "step": 2310
    },
    {
      "epoch": 13.257142857142856,
      "grad_norm": 2.890625,
      "learning_rate": 8e-05,
      "loss": 0.2328,
      "step": 2320
    },
    {
      "epoch": 13.314285714285715,
      "grad_norm": 1.8203125,
      "learning_rate": 8e-05,
      "loss": 0.2568,
      "step": 2330
    },
    {
      "epoch": 13.371428571428572,
      "grad_norm": 2.0,
      "learning_rate": 8e-05,
      "loss": 0.2639,
      "step": 2340
    },
    {
      "epoch": 13.428571428571429,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.174,
      "step": 2350
    },
    {
      "epoch": 13.485714285714286,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.2579,
      "step": 2360
    },
    {
      "epoch": 13.542857142857143,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.2111,
      "step": 2370
    },
    {
      "epoch": 13.6,
      "grad_norm": 1.7265625,
      "learning_rate": 8e-05,
      "loss": 0.3664,
      "step": 2380
    },
    {
      "epoch": 13.657142857142857,
      "grad_norm": 2.96875,
      "learning_rate": 8e-05,
      "loss": 0.2923,
      "step": 2390
    },
    {
      "epoch": 13.714285714285714,
      "grad_norm": 1.4296875,
      "learning_rate": 8e-05,
      "loss": 0.206,
      "step": 2400
    },
    {
      "epoch": 13.771428571428572,
      "grad_norm": 11.25,
      "learning_rate": 8e-05,
      "loss": 0.3807,
      "step": 2410
    },
    {
      "epoch": 13.82857142857143,
      "grad_norm": 1.6328125,
      "learning_rate": 8e-05,
      "loss": 0.2344,
      "step": 2420
    },
    {
      "epoch": 13.885714285714286,
      "grad_norm": 12.3125,
      "learning_rate": 8e-05,
      "loss": 0.2925,
      "step": 2430
    },
    {
      "epoch": 13.942857142857143,
      "grad_norm": 2.0,
      "learning_rate": 8e-05,
      "loss": 0.1925,
      "step": 2440
    },
    {
      "epoch": 14.0,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.3541,
      "step": 2450
    },
    {
      "epoch": 14.057142857142857,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.2747,
      "step": 2460
    },
    {
      "epoch": 14.114285714285714,
      "grad_norm": 1.4765625,
      "learning_rate": 8e-05,
      "loss": 0.1139,
      "step": 2470
    },
    {
      "epoch": 14.17142857142857,
      "grad_norm": 3.171875,
      "learning_rate": 8e-05,
      "loss": 0.2117,
      "step": 2480
    },
    {
      "epoch": 14.228571428571428,
      "grad_norm": 2.671875,
      "learning_rate": 8e-05,
      "loss": 0.2845,
      "step": 2490
    },
    {
      "epoch": 14.285714285714286,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.2977,
      "step": 2500
    },
    {
      "epoch": 14.285714285714286,
      "eval_loss": 8.100706100463867,
      "eval_runtime": 7.5531,
      "eval_samples_per_second": 66.198,
      "eval_steps_per_second": 66.198,
      "step": 2500
    },
    {
      "epoch": 14.342857142857143,
      "grad_norm": 1.640625,
      "learning_rate": 8e-05,
      "loss": 0.2609,
      "step": 2510
    },
    {
      "epoch": 14.4,
      "grad_norm": 2.046875,
      "learning_rate": 8e-05,
      "loss": 0.2843,
      "step": 2520
    },
    {
      "epoch": 14.457142857142857,
      "grad_norm": 12.125,
      "learning_rate": 8e-05,
      "loss": 0.2175,
      "step": 2530
    },
    {
      "epoch": 14.514285714285714,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.2517,
      "step": 2540
    },
    {
      "epoch": 14.571428571428571,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.4215,
      "step": 2550
    },
    {
      "epoch": 14.628571428571428,
      "grad_norm": 2.921875,
      "learning_rate": 8e-05,
      "loss": 0.2922,
      "step": 2560
    },
    {
      "epoch": 14.685714285714285,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.1465,
      "step": 2570
    },
    {
      "epoch": 14.742857142857144,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.1883,
      "step": 2580
    },
    {
      "epoch": 14.8,
      "grad_norm": 1.703125,
      "learning_rate": 8e-05,
      "loss": 0.2466,
      "step": 2590
    },
    {
      "epoch": 14.857142857142858,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.3114,
      "step": 2600
    },
    {
      "epoch": 14.914285714285715,
      "grad_norm": 1.2421875,
      "learning_rate": 8e-05,
      "loss": 0.3011,
      "step": 2610
    },
    {
      "epoch": 14.971428571428572,
      "grad_norm": 2.46875,
      "learning_rate": 8e-05,
      "loss": 0.1617,
      "step": 2620
    },
    {
      "epoch": 15.028571428571428,
      "grad_norm": 2.125,
      "learning_rate": 8e-05,
      "loss": 0.2399,
      "step": 2630
    },
    {
      "epoch": 15.085714285714285,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.3677,
      "step": 2640
    },
    {
      "epoch": 15.142857142857142,
      "grad_norm": 3.09375,
      "learning_rate": 8e-05,
      "loss": 0.1981,
      "step": 2650
    },
    {
      "epoch": 15.2,
      "grad_norm": 1.9375,
      "learning_rate": 8e-05,
      "loss": 0.1607,
      "step": 2660
    },
    {
      "epoch": 15.257142857142856,
      "grad_norm": 14.1875,
      "learning_rate": 8e-05,
      "loss": 0.2697,
      "step": 2670
    },
    {
      "epoch": 15.314285714285715,
      "grad_norm": 54.5,
      "learning_rate": 8e-05,
      "loss": 0.2507,
      "step": 2680
    },
    {
      "epoch": 15.371428571428572,
      "grad_norm": 9.625,
      "learning_rate": 8e-05,
      "loss": 0.2347,
      "step": 2690
    },
    {
      "epoch": 15.428571428571429,
      "grad_norm": 1.921875,
      "learning_rate": 8e-05,
      "loss": 0.176,
      "step": 2700
    },
    {
      "epoch": 15.485714285714286,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.2868,
      "step": 2710
    },
    {
      "epoch": 15.542857142857143,
      "grad_norm": 1.0859375,
      "learning_rate": 8e-05,
      "loss": 0.2933,
      "step": 2720
    },
    {
      "epoch": 15.6,
      "grad_norm": 1.328125,
      "learning_rate": 8e-05,
      "loss": 0.1255,
      "step": 2730
    },
    {
      "epoch": 15.657142857142857,
      "grad_norm": 0.86328125,
      "learning_rate": 8e-05,
      "loss": 0.0946,
      "step": 2740
    },
    {
      "epoch": 15.714285714285714,
      "grad_norm": 0.87890625,
      "learning_rate": 8e-05,
      "loss": 0.215,
      "step": 2750
    },
    {
      "epoch": 15.771428571428572,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.3107,
      "step": 2760
    },
    {
      "epoch": 15.82857142857143,
      "grad_norm": 1.8671875,
      "learning_rate": 8e-05,
      "loss": 0.3236,
      "step": 2770
    },
    {
      "epoch": 15.885714285714286,
      "grad_norm": 1.0703125,
      "learning_rate": 8e-05,
      "loss": 0.3173,
      "step": 2780
    },
    {
      "epoch": 15.942857142857143,
      "grad_norm": 12.3125,
      "learning_rate": 8e-05,
      "loss": 0.3143,
      "step": 2790
    },
    {
      "epoch": 16.0,
      "grad_norm": 1.7890625,
      "learning_rate": 8e-05,
      "loss": 0.2744,
      "step": 2800
    },
    {
      "epoch": 16.057142857142857,
      "grad_norm": 1.4765625,
      "learning_rate": 8e-05,
      "loss": 0.2547,
      "step": 2810
    },
    {
      "epoch": 16.114285714285714,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.2173,
      "step": 2820
    },
    {
      "epoch": 16.17142857142857,
      "grad_norm": 0.796875,
      "learning_rate": 8e-05,
      "loss": 0.1791,
      "step": 2830
    },
    {
      "epoch": 16.228571428571428,
      "grad_norm": 2.46875,
      "learning_rate": 8e-05,
      "loss": 0.277,
      "step": 2840
    },
    {
      "epoch": 16.285714285714285,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.2995,
      "step": 2850
    },
    {
      "epoch": 16.34285714285714,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.1871,
      "step": 2860
    },
    {
      "epoch": 16.4,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.299,
      "step": 2870
    },
    {
      "epoch": 16.457142857142856,
      "grad_norm": 1.375,
      "learning_rate": 8e-05,
      "loss": 0.2655,
      "step": 2880
    },
    {
      "epoch": 16.514285714285712,
      "grad_norm": 1.0390625,
      "learning_rate": 8e-05,
      "loss": 0.17,
      "step": 2890
    },
    {
      "epoch": 16.571428571428573,
      "grad_norm": 2.578125,
      "learning_rate": 8e-05,
      "loss": 0.2232,
      "step": 2900
    },
    {
      "epoch": 16.62857142857143,
      "grad_norm": 10.4375,
      "learning_rate": 8e-05,
      "loss": 0.2826,
      "step": 2910
    },
    {
      "epoch": 16.685714285714287,
      "grad_norm": 11.8125,
      "learning_rate": 8e-05,
      "loss": 0.3659,
      "step": 2920
    },
    {
      "epoch": 16.742857142857144,
      "grad_norm": 1.71875,
      "learning_rate": 8e-05,
      "loss": 0.26,
      "step": 2930
    },
    {
      "epoch": 16.8,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.2699,
      "step": 2940
    },
    {
      "epoch": 16.857142857142858,
      "grad_norm": 1.515625,
      "learning_rate": 8e-05,
      "loss": 0.238,
      "step": 2950
    },
    {
      "epoch": 16.914285714285715,
      "grad_norm": 3.390625,
      "learning_rate": 8e-05,
      "loss": 0.2692,
      "step": 2960
    },
    {
      "epoch": 16.97142857142857,
      "grad_norm": 1.5625,
      "learning_rate": 8e-05,
      "loss": 0.1514,
      "step": 2970
    },
    {
      "epoch": 17.02857142857143,
      "grad_norm": 1.140625,
      "learning_rate": 8e-05,
      "loss": 0.1644,
      "step": 2980
    },
    {
      "epoch": 17.085714285714285,
      "grad_norm": 12.5625,
      "learning_rate": 8e-05,
      "loss": 0.3619,
      "step": 2990
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 1.1328125,
      "learning_rate": 8e-05,
      "loss": 0.1402,
      "step": 3000
    },
    {
      "epoch": 17.142857142857142,
      "eval_loss": 8.192225456237793,
      "eval_runtime": 7.8572,
      "eval_samples_per_second": 63.636,
      "eval_steps_per_second": 63.636,
      "step": 3000
    },
    {
      "epoch": 17.2,
      "grad_norm": 1.1328125,
      "learning_rate": 8e-05,
      "loss": 0.279,
      "step": 3010
    },
    {
      "epoch": 17.257142857142856,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.394,
      "step": 3020
    },
    {
      "epoch": 17.314285714285713,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.2386,
      "step": 3030
    },
    {
      "epoch": 17.37142857142857,
      "grad_norm": 11.3125,
      "learning_rate": 8e-05,
      "loss": 0.3098,
      "step": 3040
    },
    {
      "epoch": 17.428571428571427,
      "grad_norm": 10.875,
      "learning_rate": 8e-05,
      "loss": 0.1654,
      "step": 3050
    },
    {
      "epoch": 17.485714285714284,
      "grad_norm": 2.453125,
      "learning_rate": 8e-05,
      "loss": 0.2084,
      "step": 3060
    },
    {
      "epoch": 17.542857142857144,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.2444,
      "step": 3070
    },
    {
      "epoch": 17.6,
      "grad_norm": 1.328125,
      "learning_rate": 8e-05,
      "loss": 0.2768,
      "step": 3080
    },
    {
      "epoch": 17.65714285714286,
      "grad_norm": 1.6328125,
      "learning_rate": 8e-05,
      "loss": 0.2225,
      "step": 3090
    },
    {
      "epoch": 17.714285714285715,
      "grad_norm": 0.8984375,
      "learning_rate": 8e-05,
      "loss": 0.2237,
      "step": 3100
    },
    {
      "epoch": 17.771428571428572,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.231,
      "step": 3110
    },
    {
      "epoch": 17.82857142857143,
      "grad_norm": 12.125,
      "learning_rate": 8e-05,
      "loss": 0.1867,
      "step": 3120
    },
    {
      "epoch": 17.885714285714286,
      "grad_norm": 0.96875,
      "learning_rate": 8e-05,
      "loss": 0.2167,
      "step": 3130
    },
    {
      "epoch": 17.942857142857143,
      "grad_norm": 1.5,
      "learning_rate": 8e-05,
      "loss": 0.1154,
      "step": 3140
    },
    {
      "epoch": 18.0,
      "grad_norm": 11.375,
      "learning_rate": 8e-05,
      "loss": 0.3082,
      "step": 3150
    },
    {
      "epoch": 18.057142857142857,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.283,
      "step": 3160
    },
    {
      "epoch": 18.114285714285714,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.2062,
      "step": 3170
    },
    {
      "epoch": 18.17142857142857,
      "grad_norm": 11.0625,
      "learning_rate": 8e-05,
      "loss": 0.3119,
      "step": 3180
    },
    {
      "epoch": 18.228571428571428,
      "grad_norm": 1.6015625,
      "learning_rate": 8e-05,
      "loss": 0.2351,
      "step": 3190
    },
    {
      "epoch": 18.285714285714285,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.2614,
      "step": 3200
    },
    {
      "epoch": 18.34285714285714,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.2687,
      "step": 3210
    },
    {
      "epoch": 18.4,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.3616,
      "step": 3220
    },
    {
      "epoch": 18.457142857142856,
      "grad_norm": 2.203125,
      "learning_rate": 8e-05,
      "loss": 0.2505,
      "step": 3230
    },
    {
      "epoch": 18.514285714285712,
      "grad_norm": 1.7734375,
      "learning_rate": 8e-05,
      "loss": 0.18,
      "step": 3240
    },
    {
      "epoch": 18.571428571428573,
      "grad_norm": 1.421875,
      "learning_rate": 8e-05,
      "loss": 0.2074,
      "step": 3250
    },
    {
      "epoch": 18.62857142857143,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.2854,
      "step": 3260
    },
    {
      "epoch": 18.685714285714287,
      "grad_norm": 2.34375,
      "learning_rate": 8e-05,
      "loss": 0.1504,
      "step": 3270
    },
    {
      "epoch": 18.742857142857144,
      "grad_norm": 1.5625,
      "learning_rate": 8e-05,
      "loss": 0.2151,
      "step": 3280
    },
    {
      "epoch": 18.8,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.2479,
      "step": 3290
    },
    {
      "epoch": 18.857142857142858,
      "grad_norm": 1.65625,
      "learning_rate": 8e-05,
      "loss": 0.2759,
      "step": 3300
    },
    {
      "epoch": 18.914285714285715,
      "grad_norm": 1.78125,
      "learning_rate": 8e-05,
      "loss": 0.3888,
      "step": 3310
    },
    {
      "epoch": 18.97142857142857,
      "grad_norm": 1.46875,
      "learning_rate": 8e-05,
      "loss": 0.1576,
      "step": 3320
    },
    {
      "epoch": 19.02857142857143,
      "grad_norm": 11.375,
      "learning_rate": 8e-05,
      "loss": 0.2192,
      "step": 3330
    },
    {
      "epoch": 19.085714285714285,
      "grad_norm": 14.5625,
      "learning_rate": 8e-05,
      "loss": 0.2596,
      "step": 3340
    },
    {
      "epoch": 19.142857142857142,
      "grad_norm": 0.8828125,
      "learning_rate": 8e-05,
      "loss": 0.1673,
      "step": 3350
    },
    {
      "epoch": 19.2,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.3031,
      "step": 3360
    },
    {
      "epoch": 19.257142857142856,
      "grad_norm": 11.6875,
      "learning_rate": 8e-05,
      "loss": 0.3253,
      "step": 3370
    },
    {
      "epoch": 19.314285714285713,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.257,
      "step": 3380
    },
    {
      "epoch": 19.37142857142857,
      "grad_norm": 10.1875,
      "learning_rate": 8e-05,
      "loss": 0.2625,
      "step": 3390
    },
    {
      "epoch": 19.428571428571427,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.3655,
      "step": 3400
    },
    {
      "epoch": 19.485714285714284,
      "grad_norm": 2.546875,
      "learning_rate": 8e-05,
      "loss": 0.1244,
      "step": 3410
    },
    {
      "epoch": 19.542857142857144,
      "grad_norm": 1.3984375,
      "learning_rate": 8e-05,
      "loss": 0.1875,
      "step": 3420
    },
    {
      "epoch": 19.6,
      "grad_norm": 0.83203125,
      "learning_rate": 8e-05,
      "loss": 0.1918,
      "step": 3430
    },
    {
      "epoch": 19.65714285714286,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.2807,
      "step": 3440
    },
    {
      "epoch": 19.714285714285715,
      "grad_norm": 1.5078125,
      "learning_rate": 8e-05,
      "loss": 0.1853,
      "step": 3450
    },
    {
      "epoch": 19.771428571428572,
      "grad_norm": 1.359375,
      "learning_rate": 8e-05,
      "loss": 0.1957,
      "step": 3460
    },
    {
      "epoch": 19.82857142857143,
      "grad_norm": 2.15625,
      "learning_rate": 8e-05,
      "loss": 0.2659,
      "step": 3470
    },
    {
      "epoch": 19.885714285714286,
      "grad_norm": 11.3125,
      "learning_rate": 8e-05,
      "loss": 0.2069,
      "step": 3480
    },
    {
      "epoch": 19.942857142857143,
      "grad_norm": 0.90625,
      "learning_rate": 8e-05,
      "loss": 0.2495,
      "step": 3490
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.6875,
      "learning_rate": 8e-05,
      "loss": 0.2491,
      "step": 3500
    },
    {
      "epoch": 20.0,
      "eval_loss": 8.880386352539062,
      "eval_runtime": 7.8925,
      "eval_samples_per_second": 63.351,
      "eval_steps_per_second": 63.351,
      "step": 3500
    },
    {
      "epoch": 20.057142857142857,
      "grad_norm": 1.0234375,
      "learning_rate": 8e-05,
      "loss": 0.2605,
      "step": 3510
    },
    {
      "epoch": 20.114285714285714,
      "grad_norm": 13.6875,
      "learning_rate": 8e-05,
      "loss": 0.3452,
      "step": 3520
    },
    {
      "epoch": 20.17142857142857,
      "grad_norm": 1.609375,
      "learning_rate": 8e-05,
      "loss": 0.2802,
      "step": 3530
    },
    {
      "epoch": 20.228571428571428,
      "grad_norm": 1.125,
      "learning_rate": 8e-05,
      "loss": 0.2273,
      "step": 3540
    },
    {
      "epoch": 20.285714285714285,
      "grad_norm": 2.921875,
      "learning_rate": 8e-05,
      "loss": 0.1685,
      "step": 3550
    },
    {
      "epoch": 20.34285714285714,
      "grad_norm": 2.828125,
      "learning_rate": 8e-05,
      "loss": 0.3155,
      "step": 3560
    },
    {
      "epoch": 20.4,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.3041,
      "step": 3570
    },
    {
      "epoch": 20.457142857142856,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.2268,
      "step": 3580
    },
    {
      "epoch": 20.514285714285712,
      "grad_norm": 3.46875,
      "learning_rate": 8e-05,
      "loss": 0.2387,
      "step": 3590
    },
    {
      "epoch": 20.571428571428573,
      "grad_norm": 12.25,
      "learning_rate": 8e-05,
      "loss": 0.2591,
      "step": 3600
    },
    {
      "epoch": 20.62857142857143,
      "grad_norm": 13.8125,
      "learning_rate": 8e-05,
      "loss": 0.2208,
      "step": 3610
    },
    {
      "epoch": 20.685714285714287,
      "grad_norm": 1.921875,
      "learning_rate": 8e-05,
      "loss": 0.1838,
      "step": 3620
    },
    {
      "epoch": 20.742857142857144,
      "grad_norm": 2.734375,
      "learning_rate": 8e-05,
      "loss": 0.2709,
      "step": 3630
    },
    {
      "epoch": 20.8,
      "grad_norm": 13.9375,
      "learning_rate": 8e-05,
      "loss": 0.1863,
      "step": 3640
    },
    {
      "epoch": 20.857142857142858,
      "grad_norm": 0.921875,
      "learning_rate": 8e-05,
      "loss": 0.1883,
      "step": 3650
    },
    {
      "epoch": 20.914285714285715,
      "grad_norm": 1.140625,
      "learning_rate": 8e-05,
      "loss": 0.2428,
      "step": 3660
    },
    {
      "epoch": 20.97142857142857,
      "grad_norm": 1.03125,
      "learning_rate": 8e-05,
      "loss": 0.1156,
      "step": 3670
    },
    {
      "epoch": 21.02857142857143,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.303,
      "step": 3680
    },
    {
      "epoch": 21.085714285714285,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.284,
      "step": 3690
    },
    {
      "epoch": 21.142857142857142,
      "grad_norm": 11.0625,
      "learning_rate": 8e-05,
      "loss": 0.2223,
      "step": 3700
    },
    {
      "epoch": 21.2,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.1959,
      "step": 3710
    },
    {
      "epoch": 21.257142857142856,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.2387,
      "step": 3720
    },
    {
      "epoch": 21.314285714285713,
      "grad_norm": 1.3203125,
      "learning_rate": 8e-05,
      "loss": 0.1537,
      "step": 3730
    },
    {
      "epoch": 21.37142857142857,
      "grad_norm": 1.296875,
      "learning_rate": 8e-05,
      "loss": 0.2123,
      "step": 3740
    },
    {
      "epoch": 21.428571428571427,
      "grad_norm": 1.34375,
      "learning_rate": 8e-05,
      "loss": 0.2465,
      "step": 3750
    },
    {
      "epoch": 21.485714285714284,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.247,
      "step": 3760
    },
    {
      "epoch": 21.542857142857144,
      "grad_norm": 1.0859375,
      "learning_rate": 8e-05,
      "loss": 0.2494,
      "step": 3770
    },
    {
      "epoch": 21.6,
      "grad_norm": 2.296875,
      "learning_rate": 8e-05,
      "loss": 0.1628,
      "step": 3780
    },
    {
      "epoch": 21.65714285714286,
      "grad_norm": 1.0703125,
      "learning_rate": 8e-05,
      "loss": 0.1647,
      "step": 3790
    },
    {
      "epoch": 21.714285714285715,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.2749,
      "step": 3800
    },
    {
      "epoch": 21.771428571428572,
      "grad_norm": 1.265625,
      "learning_rate": 8e-05,
      "loss": 0.2694,
      "step": 3810
    },
    {
      "epoch": 21.82857142857143,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.3419,
      "step": 3820
    },
    {
      "epoch": 21.885714285714286,
      "grad_norm": 1.546875,
      "learning_rate": 8e-05,
      "loss": 0.1667,
      "step": 3830
    },
    {
      "epoch": 21.942857142857143,
      "grad_norm": 4.46875,
      "learning_rate": 8e-05,
      "loss": 0.3001,
      "step": 3840
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.76953125,
      "learning_rate": 8e-05,
      "loss": 0.2372,
      "step": 3850
    },
    {
      "epoch": 22.057142857142857,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.2372,
      "step": 3860
    },
    {
      "epoch": 22.114285714285714,
      "grad_norm": 1.890625,
      "learning_rate": 8e-05,
      "loss": 0.2566,
      "step": 3870
    },
    {
      "epoch": 22.17142857142857,
      "grad_norm": 11.0625,
      "learning_rate": 8e-05,
      "loss": 0.1146,
      "step": 3880
    },
    {
      "epoch": 22.228571428571428,
      "grad_norm": 1.28125,
      "learning_rate": 8e-05,
      "loss": 0.2228,
      "step": 3890
    },
    {
      "epoch": 22.285714285714285,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.2839,
      "step": 3900
    },
    {
      "epoch": 22.34285714285714,
      "grad_norm": 11.125,
      "learning_rate": 8e-05,
      "loss": 0.2664,
      "step": 3910
    },
    {
      "epoch": 22.4,
      "grad_norm": 1.0625,
      "learning_rate": 8e-05,
      "loss": 0.2979,
      "step": 3920
    },
    {
      "epoch": 22.457142857142856,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.2924,
      "step": 3930
    },
    {
      "epoch": 22.514285714285712,
      "grad_norm": 1.59375,
      "learning_rate": 8e-05,
      "loss": 0.201,
      "step": 3940
    },
    {
      "epoch": 22.571428571428573,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.2611,
      "step": 3950
    },
    {
      "epoch": 22.62857142857143,
      "grad_norm": 1.28125,
      "learning_rate": 8e-05,
      "loss": 0.1853,
      "step": 3960
    },
    {
      "epoch": 22.685714285714287,
      "grad_norm": 2.125,
      "learning_rate": 8e-05,
      "loss": 0.1745,
      "step": 3970
    },
    {
      "epoch": 22.742857142857144,
      "grad_norm": 1.0390625,
      "learning_rate": 8e-05,
      "loss": 0.1967,
      "step": 3980
    },
    {
      "epoch": 22.8,
      "grad_norm": 11.1875,
      "learning_rate": 8e-05,
      "loss": 0.2362,
      "step": 3990
    },
    {
      "epoch": 22.857142857142858,
      "grad_norm": 1.46875,
      "learning_rate": 8e-05,
      "loss": 0.1878,
      "step": 4000
    },
    {
      "epoch": 22.857142857142858,
      "eval_loss": 8.872773170471191,
      "eval_runtime": 7.8964,
      "eval_samples_per_second": 63.32,
      "eval_steps_per_second": 63.32,
      "step": 4000
    },
    {
      "epoch": 22.914285714285715,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.3667,
      "step": 4010
    },
    {
      "epoch": 22.97142857142857,
      "grad_norm": 1.0234375,
      "learning_rate": 8e-05,
      "loss": 0.2711,
      "step": 4020
    },
    {
      "epoch": 23.02857142857143,
      "grad_norm": 12.125,
      "learning_rate": 8e-05,
      "loss": 0.2202,
      "step": 4030
    },
    {
      "epoch": 23.085714285714285,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.23,
      "step": 4040
    },
    {
      "epoch": 23.142857142857142,
      "grad_norm": 0.95703125,
      "learning_rate": 8e-05,
      "loss": 0.2866,
      "step": 4050
    },
    {
      "epoch": 23.2,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.2623,
      "step": 4060
    },
    {
      "epoch": 23.257142857142856,
      "grad_norm": 1.28125,
      "learning_rate": 8e-05,
      "loss": 0.1838,
      "step": 4070
    },
    {
      "epoch": 23.314285714285713,
      "grad_norm": 10.125,
      "learning_rate": 8e-05,
      "loss": 0.2214,
      "step": 4080
    },
    {
      "epoch": 23.37142857142857,
      "grad_norm": 1.25,
      "learning_rate": 8e-05,
      "loss": 0.3113,
      "step": 4090
    },
    {
      "epoch": 23.428571428571427,
      "grad_norm": 0.90234375,
      "learning_rate": 8e-05,
      "loss": 0.1955,
      "step": 4100
    },
    {
      "epoch": 23.485714285714284,
      "grad_norm": 1.5,
      "learning_rate": 8e-05,
      "loss": 0.1104,
      "step": 4110
    },
    {
      "epoch": 23.542857142857144,
      "grad_norm": 13.625,
      "learning_rate": 8e-05,
      "loss": 0.2652,
      "step": 4120
    },
    {
      "epoch": 23.6,
      "grad_norm": 0.9765625,
      "learning_rate": 8e-05,
      "loss": 0.3163,
      "step": 4130
    },
    {
      "epoch": 23.65714285714286,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.3631,
      "step": 4140
    },
    {
      "epoch": 23.714285714285715,
      "grad_norm": 1.1171875,
      "learning_rate": 8e-05,
      "loss": 0.1891,
      "step": 4150
    },
    {
      "epoch": 23.771428571428572,
      "grad_norm": 1.0859375,
      "learning_rate": 8e-05,
      "loss": 0.2511,
      "step": 4160
    },
    {
      "epoch": 23.82857142857143,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.2388,
      "step": 4170
    },
    {
      "epoch": 23.885714285714286,
      "grad_norm": 2.015625,
      "learning_rate": 8e-05,
      "loss": 0.1704,
      "step": 4180
    },
    {
      "epoch": 23.942857142857143,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.2243,
      "step": 4190
    },
    {
      "epoch": 24.0,
      "grad_norm": 1.4765625,
      "learning_rate": 8e-05,
      "loss": 0.2253,
      "step": 4200
    },
    {
      "epoch": 24.057142857142857,
      "grad_norm": 11.8125,
      "learning_rate": 8e-05,
      "loss": 0.1691,
      "step": 4210
    },
    {
      "epoch": 24.114285714285714,
      "grad_norm": 11.3125,
      "learning_rate": 8e-05,
      "loss": 0.2168,
      "step": 4220
    },
    {
      "epoch": 24.17142857142857,
      "grad_norm": 0.97265625,
      "learning_rate": 8e-05,
      "loss": 0.2336,
      "step": 4230
    },
    {
      "epoch": 24.228571428571428,
      "grad_norm": 1.8671875,
      "learning_rate": 8e-05,
      "loss": 0.2058,
      "step": 4240
    },
    {
      "epoch": 24.285714285714285,
      "grad_norm": 2.359375,
      "learning_rate": 8e-05,
      "loss": 0.1282,
      "step": 4250
    },
    {
      "epoch": 24.34285714285714,
      "grad_norm": 1.359375,
      "learning_rate": 8e-05,
      "loss": 0.2631,
      "step": 4260
    },
    {
      "epoch": 24.4,
      "grad_norm": 1.0703125,
      "learning_rate": 8e-05,
      "loss": 0.2444,
      "step": 4270
    },
    {
      "epoch": 24.457142857142856,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.2777,
      "step": 4280
    },
    {
      "epoch": 24.514285714285712,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.2395,
      "step": 4290
    },
    {
      "epoch": 24.571428571428573,
      "grad_norm": 1.5625,
      "learning_rate": 8e-05,
      "loss": 0.2291,
      "step": 4300
    },
    {
      "epoch": 24.62857142857143,
      "grad_norm": 2.375,
      "learning_rate": 8e-05,
      "loss": 0.3439,
      "step": 4310
    },
    {
      "epoch": 24.685714285714287,
      "grad_norm": 14.0625,
      "learning_rate": 8e-05,
      "loss": 0.2126,
      "step": 4320
    },
    {
      "epoch": 24.742857142857144,
      "grad_norm": 3.5625,
      "learning_rate": 8e-05,
      "loss": 0.2568,
      "step": 4330
    },
    {
      "epoch": 24.8,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.2513,
      "step": 4340
    },
    {
      "epoch": 24.857142857142858,
      "grad_norm": 0.96875,
      "learning_rate": 8e-05,
      "loss": 0.2694,
      "step": 4350
    },
    {
      "epoch": 24.914285714285715,
      "grad_norm": 1.890625,
      "learning_rate": 8e-05,
      "loss": 0.3128,
      "step": 4360
    },
    {
      "epoch": 24.97142857142857,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.2509,
      "step": 4370
    },
    {
      "epoch": 25.02857142857143,
      "grad_norm": 0.984375,
      "learning_rate": 8e-05,
      "loss": 0.2178,
      "step": 4380
    },
    {
      "epoch": 25.085714285714285,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.2829,
      "step": 4390
    },
    {
      "epoch": 25.142857142857142,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.2785,
      "step": 4400
    },
    {
      "epoch": 25.2,
      "grad_norm": 13.5625,
      "learning_rate": 8e-05,
      "loss": 0.2013,
      "step": 4410
    },
    {
      "epoch": 25.257142857142856,
      "grad_norm": 12.5625,
      "learning_rate": 8e-05,
      "loss": 0.1823,
      "step": 4420
    },
    {
      "epoch": 25.314285714285713,
      "grad_norm": 11.125,
      "learning_rate": 8e-05,
      "loss": 0.199,
      "step": 4430
    },
    {
      "epoch": 25.37142857142857,
      "grad_norm": 0.84375,
      "learning_rate": 8e-05,
      "loss": 0.3073,
      "step": 4440
    },
    {
      "epoch": 25.428571428571427,
      "grad_norm": 3.265625,
      "learning_rate": 8e-05,
      "loss": 0.2571,
      "step": 4450
    },
    {
      "epoch": 25.485714285714284,
      "grad_norm": 1.546875,
      "learning_rate": 8e-05,
      "loss": 0.1958,
      "step": 4460
    },
    {
      "epoch": 25.542857142857144,
      "grad_norm": 1.4765625,
      "learning_rate": 8e-05,
      "loss": 0.2315,
      "step": 4470
    },
    {
      "epoch": 25.6,
      "grad_norm": 1.1171875,
      "learning_rate": 8e-05,
      "loss": 0.2363,
      "step": 4480
    },
    {
      "epoch": 25.65714285714286,
      "grad_norm": 1.0703125,
      "learning_rate": 8e-05,
      "loss": 0.3238,
      "step": 4490
    },
    {
      "epoch": 25.714285714285715,
      "grad_norm": 1.421875,
      "learning_rate": 8e-05,
      "loss": 0.2129,
      "step": 4500
    },
    {
      "epoch": 25.714285714285715,
      "eval_loss": 9.115077018737793,
      "eval_runtime": 7.8421,
      "eval_samples_per_second": 63.759,
      "eval_steps_per_second": 63.759,
      "step": 4500
    },
    {
      "epoch": 25.771428571428572,
      "grad_norm": 0.875,
      "learning_rate": 8e-05,
      "loss": 0.1894,
      "step": 4510
    },
    {
      "epoch": 25.82857142857143,
      "grad_norm": 1.3515625,
      "learning_rate": 8e-05,
      "loss": 0.1831,
      "step": 4520
    },
    {
      "epoch": 25.885714285714286,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.3132,
      "step": 4530
    },
    {
      "epoch": 25.942857142857143,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.3162,
      "step": 4540
    },
    {
      "epoch": 26.0,
      "grad_norm": 1.015625,
      "learning_rate": 8e-05,
      "loss": 0.2133,
      "step": 4550
    },
    {
      "epoch": 26.057142857142857,
      "grad_norm": 2.0625,
      "learning_rate": 8e-05,
      "loss": 0.2219,
      "step": 4560
    },
    {
      "epoch": 26.114285714285714,
      "grad_norm": 1.0703125,
      "learning_rate": 8e-05,
      "loss": 0.2104,
      "step": 4570
    },
    {
      "epoch": 26.17142857142857,
      "grad_norm": 1.671875,
      "learning_rate": 8e-05,
      "loss": 0.2499,
      "step": 4580
    },
    {
      "epoch": 26.228571428571428,
      "grad_norm": 0.88671875,
      "learning_rate": 8e-05,
      "loss": 0.2229,
      "step": 4590
    },
    {
      "epoch": 26.285714285714285,
      "grad_norm": 14.0625,
      "learning_rate": 8e-05,
      "loss": 0.199,
      "step": 4600
    },
    {
      "epoch": 26.34285714285714,
      "grad_norm": 14.25,
      "learning_rate": 8e-05,
      "loss": 0.2306,
      "step": 4610
    },
    {
      "epoch": 26.4,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.2476,
      "step": 4620
    },
    {
      "epoch": 26.457142857142856,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.2139,
      "step": 4630
    },
    {
      "epoch": 26.514285714285712,
      "grad_norm": 1.9375,
      "learning_rate": 8e-05,
      "loss": 0.216,
      "step": 4640
    },
    {
      "epoch": 26.571428571428573,
      "grad_norm": 0.8828125,
      "learning_rate": 8e-05,
      "loss": 0.1741,
      "step": 4650
    },
    {
      "epoch": 26.62857142857143,
      "grad_norm": 1.0078125,
      "learning_rate": 8e-05,
      "loss": 0.2203,
      "step": 4660
    },
    {
      "epoch": 26.685714285714287,
      "grad_norm": 11.5625,
      "learning_rate": 8e-05,
      "loss": 0.3017,
      "step": 4670
    },
    {
      "epoch": 26.742857142857144,
      "grad_norm": 2.15625,
      "learning_rate": 8e-05,
      "loss": 0.2818,
      "step": 4680
    },
    {
      "epoch": 26.8,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.2493,
      "step": 4690
    },
    {
      "epoch": 26.857142857142858,
      "grad_norm": 1.125,
      "learning_rate": 8e-05,
      "loss": 0.2621,
      "step": 4700
    },
    {
      "epoch": 26.914285714285715,
      "grad_norm": 0.87109375,
      "learning_rate": 8e-05,
      "loss": 0.2187,
      "step": 4710
    },
    {
      "epoch": 26.97142857142857,
      "grad_norm": 14.375,
      "learning_rate": 8e-05,
      "loss": 0.3442,
      "step": 4720
    },
    {
      "epoch": 27.02857142857143,
      "grad_norm": 2.5,
      "learning_rate": 8e-05,
      "loss": 0.1336,
      "step": 4730
    },
    {
      "epoch": 27.085714285714285,
      "grad_norm": 0.9765625,
      "learning_rate": 8e-05,
      "loss": 0.2898,
      "step": 4740
    },
    {
      "epoch": 27.142857142857142,
      "grad_norm": 12.3125,
      "learning_rate": 8e-05,
      "loss": 0.3381,
      "step": 4750
    },
    {
      "epoch": 27.2,
      "grad_norm": 0.890625,
      "learning_rate": 8e-05,
      "loss": 0.1815,
      "step": 4760
    },
    {
      "epoch": 27.257142857142856,
      "grad_norm": 10.5625,
      "learning_rate": 8e-05,
      "loss": 0.1338,
      "step": 4770
    },
    {
      "epoch": 27.314285714285713,
      "grad_norm": 1.7734375,
      "learning_rate": 8e-05,
      "loss": 0.2167,
      "step": 4780
    },
    {
      "epoch": 27.37142857142857,
      "grad_norm": 1.28125,
      "learning_rate": 8e-05,
      "loss": 0.253,
      "step": 4790
    },
    {
      "epoch": 27.428571428571427,
      "grad_norm": 10.875,
      "learning_rate": 8e-05,
      "loss": 0.2037,
      "step": 4800
    },
    {
      "epoch": 27.485714285714284,
      "grad_norm": 0.734375,
      "learning_rate": 8e-05,
      "loss": 0.2611,
      "step": 4810
    },
    {
      "epoch": 27.542857142857144,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.1021,
      "step": 4820
    },
    {
      "epoch": 27.6,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.2804,
      "step": 4830
    },
    {
      "epoch": 27.65714285714286,
      "grad_norm": 1.7421875,
      "learning_rate": 8e-05,
      "loss": 0.2014,
      "step": 4840
    },
    {
      "epoch": 27.714285714285715,
      "grad_norm": 9.5625,
      "learning_rate": 8e-05,
      "loss": 0.1588,
      "step": 4850
    },
    {
      "epoch": 27.771428571428572,
      "grad_norm": 2.4375,
      "learning_rate": 8e-05,
      "loss": 0.3445,
      "step": 4860
    },
    {
      "epoch": 27.82857142857143,
      "grad_norm": 1.7578125,
      "learning_rate": 8e-05,
      "loss": 0.1854,
      "step": 4870
    },
    {
      "epoch": 27.885714285714286,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.2785,
      "step": 4880
    },
    {
      "epoch": 27.942857142857143,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.2414,
      "step": 4890
    },
    {
      "epoch": 28.0,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.3524,
      "step": 4900
    }
  ],
  "logging_steps": 10,
  "max_steps": 7000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 175,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3134793563897856.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}