{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 31.0,
  "eval_steps": 500,
  "global_step": 6200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 2.9668,
      "step": 10
    },
    {
      "epoch": 0.1,
      "grad_norm": 8.75,
      "learning_rate": 8e-05,
      "loss": 1.5447,
      "step": 20
    },
    {
      "epoch": 0.15,
      "grad_norm": 18.375,
      "learning_rate": 8e-05,
      "loss": 1.3061,
      "step": 30
    },
    {
      "epoch": 0.2,
      "grad_norm": 19.5,
      "learning_rate": 8e-05,
      "loss": 1.2262,
      "step": 40
    },
    {
      "epoch": 0.25,
      "grad_norm": 19.625,
      "learning_rate": 8e-05,
      "loss": 1.106,
      "step": 50
    },
    {
      "epoch": 0.3,
      "grad_norm": 20.5,
      "learning_rate": 8e-05,
      "loss": 1.1051,
      "step": 60
    },
    {
      "epoch": 0.35,
      "grad_norm": 8.0625,
      "learning_rate": 8e-05,
      "loss": 1.0603,
      "step": 70
    },
    {
      "epoch": 0.4,
      "grad_norm": 17.625,
      "learning_rate": 8e-05,
      "loss": 1.0426,
      "step": 80
    },
    {
      "epoch": 0.45,
      "grad_norm": 18.5,
      "learning_rate": 8e-05,
      "loss": 1.0278,
      "step": 90
    },
    {
      "epoch": 0.5,
      "grad_norm": 8.0,
      "learning_rate": 8e-05,
      "loss": 0.9811,
      "step": 100
    },
    {
      "epoch": 0.55,
      "grad_norm": 8.375,
      "learning_rate": 8e-05,
      "loss": 1.0554,
      "step": 110
    },
    {
      "epoch": 0.6,
      "grad_norm": 14.1875,
      "learning_rate": 8e-05,
      "loss": 0.981,
      "step": 120
    },
    {
      "epoch": 0.65,
      "grad_norm": 13.4375,
      "learning_rate": 8e-05,
      "loss": 0.9207,
      "step": 130
    },
    {
      "epoch": 0.7,
      "grad_norm": 10.1875,
      "learning_rate": 8e-05,
      "loss": 0.9944,
      "step": 140
    },
    {
      "epoch": 0.75,
      "grad_norm": 14.4375,
      "learning_rate": 8e-05,
      "loss": 0.7818,
      "step": 150
    },
    {
      "epoch": 0.8,
      "grad_norm": 13.8125,
      "learning_rate": 8e-05,
      "loss": 0.6701,
      "step": 160
    },
    {
      "epoch": 0.85,
      "grad_norm": 7.46875,
      "learning_rate": 8e-05,
      "loss": 0.8406,
      "step": 170
    },
    {
      "epoch": 0.9,
      "grad_norm": 8.375,
      "learning_rate": 8e-05,
      "loss": 1.1126,
      "step": 180
    },
    {
      "epoch": 0.95,
      "grad_norm": 7.84375,
      "learning_rate": 8e-05,
      "loss": 0.9493,
      "step": 190
    },
    {
      "epoch": 1.0,
      "grad_norm": 15.4375,
      "learning_rate": 8e-05,
      "loss": 0.8737,
      "step": 200
    },
    {
      "epoch": 1.05,
      "grad_norm": 9.8125,
      "learning_rate": 8e-05,
      "loss": 0.8025,
      "step": 210
    },
    {
      "epoch": 1.1,
      "grad_norm": 9.4375,
      "learning_rate": 8e-05,
      "loss": 0.6186,
      "step": 220
    },
    {
      "epoch": 1.15,
      "grad_norm": 14.0625,
      "learning_rate": 8e-05,
      "loss": 0.7477,
      "step": 230
    },
    {
      "epoch": 1.2,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.6547,
      "step": 240
    },
    {
      "epoch": 1.25,
      "grad_norm": 8.75,
      "learning_rate": 8e-05,
      "loss": 0.7623,
      "step": 250
    },
    {
      "epoch": 1.3,
      "grad_norm": 7.09375,
      "learning_rate": 8e-05,
      "loss": 0.6258,
      "step": 260
    },
    {
      "epoch": 1.35,
      "grad_norm": 8.375,
      "learning_rate": 8e-05,
      "loss": 0.7048,
      "step": 270
    },
    {
      "epoch": 1.4,
      "grad_norm": 13.6875,
      "learning_rate": 8e-05,
      "loss": 0.7504,
      "step": 280
    },
    {
      "epoch": 1.45,
      "grad_norm": 8.125,
      "learning_rate": 8e-05,
      "loss": 0.7502,
      "step": 290
    },
    {
      "epoch": 1.5,
      "grad_norm": 8.25,
      "learning_rate": 8e-05,
      "loss": 0.7787,
      "step": 300
    },
    {
      "epoch": 1.55,
      "grad_norm": 7.625,
      "learning_rate": 8e-05,
      "loss": 0.7807,
      "step": 310
    },
    {
      "epoch": 1.6,
      "grad_norm": 8.0625,
      "learning_rate": 8e-05,
      "loss": 0.7065,
      "step": 320
    },
    {
      "epoch": 1.65,
      "grad_norm": 7.78125,
      "learning_rate": 8e-05,
      "loss": 0.7102,
      "step": 330
    },
    {
      "epoch": 1.7,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.6821,
      "step": 340
    },
    {
      "epoch": 1.75,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.5562,
      "step": 350
    },
    {
      "epoch": 1.8,
      "grad_norm": 13.9375,
      "learning_rate": 8e-05,
      "loss": 0.6497,
      "step": 360
    },
    {
      "epoch": 1.85,
      "grad_norm": 8.0,
      "learning_rate": 8e-05,
      "loss": 0.7489,
      "step": 370
    },
    {
      "epoch": 1.9,
      "grad_norm": 9.5,
      "learning_rate": 8e-05,
      "loss": 0.726,
      "step": 380
    },
    {
      "epoch": 1.95,
      "grad_norm": 8.375,
      "learning_rate": 8e-05,
      "loss": 0.5764,
      "step": 390
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.9375,
      "learning_rate": 8e-05,
      "loss": 0.6491,
      "step": 400
    },
    {
      "epoch": 2.05,
      "grad_norm": 8.3125,
      "learning_rate": 8e-05,
      "loss": 0.5339,
      "step": 410
    },
    {
      "epoch": 2.1,
      "grad_norm": 8.5,
      "learning_rate": 8e-05,
      "loss": 0.7454,
      "step": 420
    },
    {
      "epoch": 2.15,
      "grad_norm": 15.375,
      "learning_rate": 8e-05,
      "loss": 0.5912,
      "step": 430
    },
    {
      "epoch": 2.2,
      "grad_norm": 14.0625,
      "learning_rate": 8e-05,
      "loss": 0.6097,
      "step": 440
    },
    {
      "epoch": 2.25,
      "grad_norm": 7.65625,
      "learning_rate": 8e-05,
      "loss": 0.6417,
      "step": 450
    },
    {
      "epoch": 2.3,
      "grad_norm": 15.0625,
      "learning_rate": 8e-05,
      "loss": 0.5642,
      "step": 460
    },
    {
      "epoch": 2.35,
      "grad_norm": 14.9375,
      "learning_rate": 8e-05,
      "loss": 0.6748,
      "step": 470
    },
    {
      "epoch": 2.4,
      "grad_norm": 9.0625,
      "learning_rate": 8e-05,
      "loss": 0.6594,
      "step": 480
    },
    {
      "epoch": 2.45,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.6317,
      "step": 490
    },
    {
      "epoch": 2.5,
      "grad_norm": 7.25,
      "learning_rate": 8e-05,
      "loss": 0.5976,
      "step": 500
    },
    {
      "epoch": 2.5,
      "eval_loss": 5.992827415466309,
      "eval_runtime": 7.763,
      "eval_samples_per_second": 64.408,
      "eval_steps_per_second": 64.408,
      "step": 500
    },
    {
      "epoch": 2.55,
      "grad_norm": 13.625,
      "learning_rate": 8e-05,
      "loss": 0.6125,
      "step": 510
    },
    {
      "epoch": 2.6,
      "grad_norm": 14.6875,
      "learning_rate": 8e-05,
      "loss": 0.621,
      "step": 520
    },
    {
      "epoch": 2.65,
      "grad_norm": 7.9375,
      "learning_rate": 8e-05,
      "loss": 0.5758,
      "step": 530
    },
    {
      "epoch": 2.7,
      "grad_norm": 11.125,
      "learning_rate": 8e-05,
      "loss": 0.5952,
      "step": 540
    },
    {
      "epoch": 2.75,
      "grad_norm": 7.25,
      "learning_rate": 8e-05,
      "loss": 0.6743,
      "step": 550
    },
    {
      "epoch": 2.8,
      "grad_norm": 11.375,
      "learning_rate": 8e-05,
      "loss": 0.5858,
      "step": 560
    },
    {
      "epoch": 2.85,
      "grad_norm": 7.3125,
      "learning_rate": 8e-05,
      "loss": 0.6662,
      "step": 570
    },
    {
      "epoch": 2.9,
      "grad_norm": 6.78125,
      "learning_rate": 8e-05,
      "loss": 0.6263,
      "step": 580
    },
    {
      "epoch": 2.95,
      "grad_norm": 13.8125,
      "learning_rate": 8e-05,
      "loss": 0.5472,
      "step": 590
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.4375,
      "learning_rate": 8e-05,
      "loss": 0.574,
      "step": 600
    },
    {
      "epoch": 3.05,
      "grad_norm": 6.9375,
      "learning_rate": 8e-05,
      "loss": 0.4992,
      "step": 610
    },
    {
      "epoch": 3.1,
      "grad_norm": 6.53125,
      "learning_rate": 8e-05,
      "loss": 0.5446,
      "step": 620
    },
    {
      "epoch": 3.15,
      "grad_norm": 9.375,
      "learning_rate": 8e-05,
      "loss": 0.5207,
      "step": 630
    },
    {
      "epoch": 3.2,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.5025,
      "step": 640
    },
    {
      "epoch": 3.25,
      "grad_norm": 7.4375,
      "learning_rate": 8e-05,
      "loss": 0.524,
      "step": 650
    },
    {
      "epoch": 3.3,
      "grad_norm": 6.5,
      "learning_rate": 8e-05,
      "loss": 0.5135,
      "step": 660
    },
    {
      "epoch": 3.35,
      "grad_norm": 6.4375,
      "learning_rate": 8e-05,
      "loss": 0.4906,
      "step": 670
    },
    {
      "epoch": 3.4,
      "grad_norm": 15.8125,
      "learning_rate": 8e-05,
      "loss": 0.4418,
      "step": 680
    },
    {
      "epoch": 3.45,
      "grad_norm": 7.1875,
      "learning_rate": 8e-05,
      "loss": 0.5683,
      "step": 690
    },
    {
      "epoch": 3.5,
      "grad_norm": 6.65625,
      "learning_rate": 8e-05,
      "loss": 0.5697,
      "step": 700
    },
    {
      "epoch": 3.55,
      "grad_norm": 7.40625,
      "learning_rate": 8e-05,
      "loss": 0.4701,
      "step": 710
    },
    {
      "epoch": 3.6,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.5056,
      "step": 720
    },
    {
      "epoch": 3.65,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.498,
      "step": 730
    },
    {
      "epoch": 3.7,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.5773,
      "step": 740
    },
    {
      "epoch": 3.75,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.5411,
      "step": 750
    },
    {
      "epoch": 3.8,
      "grad_norm": 13.8125,
      "learning_rate": 8e-05,
      "loss": 0.6051,
      "step": 760
    },
    {
      "epoch": 3.85,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.5179,
      "step": 770
    },
    {
      "epoch": 3.9,
      "grad_norm": 6.65625,
      "learning_rate": 8e-05,
      "loss": 0.5243,
      "step": 780
    },
    {
      "epoch": 3.95,
      "grad_norm": 11.9375,
      "learning_rate": 8e-05,
      "loss": 0.5162,
      "step": 790
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.0625,
      "learning_rate": 8e-05,
      "loss": 0.5591,
      "step": 800
    },
    {
      "epoch": 4.05,
      "grad_norm": 10.125,
      "learning_rate": 8e-05,
      "loss": 0.4792,
      "step": 810
    },
    {
      "epoch": 4.1,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.4441,
      "step": 820
    },
    {
      "epoch": 4.15,
      "grad_norm": 6.46875,
      "learning_rate": 8e-05,
      "loss": 0.4185,
      "step": 830
    },
    {
      "epoch": 4.2,
      "grad_norm": 7.8125,
      "learning_rate": 8e-05,
      "loss": 0.4334,
      "step": 840
    },
    {
      "epoch": 4.25,
      "grad_norm": 10.125,
      "learning_rate": 8e-05,
      "loss": 0.4924,
      "step": 850
    },
    {
      "epoch": 4.3,
      "grad_norm": 9.5,
      "learning_rate": 8e-05,
      "loss": 0.4366,
      "step": 860
    },
    {
      "epoch": 4.35,
      "grad_norm": 14.4375,
      "learning_rate": 8e-05,
      "loss": 0.4441,
      "step": 870
    },
    {
      "epoch": 4.4,
      "grad_norm": 7.1875,
      "learning_rate": 8e-05,
      "loss": 0.5339,
      "step": 880
    },
    {
      "epoch": 4.45,
      "grad_norm": 13.75,
      "learning_rate": 8e-05,
      "loss": 0.4497,
      "step": 890
    },
    {
      "epoch": 4.5,
      "grad_norm": 6.0,
      "learning_rate": 8e-05,
      "loss": 0.4611,
      "step": 900
    },
    {
      "epoch": 4.55,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.4251,
      "step": 910
    },
    {
      "epoch": 4.6,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.4453,
      "step": 920
    },
    {
      "epoch": 4.65,
      "grad_norm": 14.375,
      "learning_rate": 8e-05,
      "loss": 0.4904,
      "step": 930
    },
    {
      "epoch": 4.7,
      "grad_norm": 14.0,
      "learning_rate": 8e-05,
      "loss": 0.526,
      "step": 940
    },
    {
      "epoch": 4.75,
      "grad_norm": 8.4375,
      "learning_rate": 8e-05,
      "loss": 0.4325,
      "step": 950
    },
    {
      "epoch": 4.8,
      "grad_norm": 6.46875,
      "learning_rate": 8e-05,
      "loss": 0.3979,
      "step": 960
    },
    {
      "epoch": 4.85,
      "grad_norm": 12.3125,
      "learning_rate": 8e-05,
      "loss": 0.4622,
      "step": 970
    },
    {
      "epoch": 4.9,
      "grad_norm": 12.125,
      "learning_rate": 8e-05,
      "loss": 0.4522,
      "step": 980
    },
    {
      "epoch": 4.95,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.4131,
      "step": 990
    },
    {
      "epoch": 5.0,
      "grad_norm": 13.5625,
      "learning_rate": 8e-05,
      "loss": 0.464,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_loss": 6.443065166473389,
      "eval_runtime": 7.5814,
      "eval_samples_per_second": 65.951,
      "eval_steps_per_second": 65.951,
      "step": 1000
    },
    {
      "epoch": 5.05,
      "grad_norm": 6.90625,
      "learning_rate": 8e-05,
      "loss": 0.4061,
      "step": 1010
    },
    {
      "epoch": 5.1,
      "grad_norm": 6.75,
      "learning_rate": 8e-05,
      "loss": 0.384,
      "step": 1020
    },
    {
      "epoch": 5.15,
      "grad_norm": 9.5625,
      "learning_rate": 8e-05,
      "loss": 0.3238,
      "step": 1030
    },
    {
      "epoch": 5.2,
      "grad_norm": 8.0625,
      "learning_rate": 8e-05,
      "loss": 0.4739,
      "step": 1040
    },
    {
      "epoch": 5.25,
      "grad_norm": 4.6875,
      "learning_rate": 8e-05,
      "loss": 0.3773,
      "step": 1050
    },
    {
      "epoch": 5.3,
      "grad_norm": 6.5,
      "learning_rate": 8e-05,
      "loss": 0.3674,
      "step": 1060
    },
    {
      "epoch": 5.35,
      "grad_norm": 5.71875,
      "learning_rate": 8e-05,
      "loss": 0.3371,
      "step": 1070
    },
    {
      "epoch": 5.4,
      "grad_norm": 14.1875,
      "learning_rate": 8e-05,
      "loss": 0.4022,
      "step": 1080
    },
    {
      "epoch": 5.45,
      "grad_norm": 5.125,
      "learning_rate": 8e-05,
      "loss": 0.3692,
      "step": 1090
    },
    {
      "epoch": 5.5,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.3847,
      "step": 1100
    },
    {
      "epoch": 5.55,
      "grad_norm": 13.6875,
      "learning_rate": 8e-05,
      "loss": 0.4535,
      "step": 1110
    },
    {
      "epoch": 5.6,
      "grad_norm": 7.1875,
      "learning_rate": 8e-05,
      "loss": 0.4559,
      "step": 1120
    },
    {
      "epoch": 5.65,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.3967,
      "step": 1130
    },
    {
      "epoch": 5.7,
      "grad_norm": 11.5625,
      "learning_rate": 8e-05,
      "loss": 0.3578,
      "step": 1140
    },
    {
      "epoch": 5.75,
      "grad_norm": 5.53125,
      "learning_rate": 8e-05,
      "loss": 0.4006,
      "step": 1150
    },
    {
      "epoch": 5.8,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.3664,
      "step": 1160
    },
    {
      "epoch": 5.85,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.3775,
      "step": 1170
    },
    {
      "epoch": 5.9,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.4148,
      "step": 1180
    },
    {
      "epoch": 5.95,
      "grad_norm": 5.40625,
      "learning_rate": 8e-05,
      "loss": 0.3621,
      "step": 1190
    },
    {
      "epoch": 6.0,
      "grad_norm": 13.625,
      "learning_rate": 8e-05,
      "loss": 0.5599,
      "step": 1200
    },
    {
      "epoch": 6.05,
      "grad_norm": 7.03125,
      "learning_rate": 8e-05,
      "loss": 0.4049,
      "step": 1210
    },
    {
      "epoch": 6.1,
      "grad_norm": 13.75,
      "learning_rate": 8e-05,
      "loss": 0.4218,
      "step": 1220
    },
    {
      "epoch": 6.15,
      "grad_norm": 11.0625,
      "learning_rate": 8e-05,
      "loss": 0.3504,
      "step": 1230
    },
    {
      "epoch": 6.2,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.4073,
      "step": 1240
    },
    {
      "epoch": 6.25,
      "grad_norm": 7.3125,
      "learning_rate": 8e-05,
      "loss": 0.3338,
      "step": 1250
    },
    {
      "epoch": 6.3,
      "grad_norm": 2.609375,
      "learning_rate": 8e-05,
      "loss": 0.3149,
      "step": 1260
    },
    {
      "epoch": 6.35,
      "grad_norm": 7.21875,
      "learning_rate": 8e-05,
      "loss": 0.3903,
      "step": 1270
    },
    {
      "epoch": 6.4,
      "grad_norm": 4.15625,
      "learning_rate": 8e-05,
      "loss": 0.3733,
      "step": 1280
    },
    {
      "epoch": 6.45,
      "grad_norm": 6.21875,
      "learning_rate": 8e-05,
      "loss": 0.3173,
      "step": 1290
    },
    {
      "epoch": 6.5,
      "grad_norm": 4.5,
      "learning_rate": 8e-05,
      "loss": 0.4605,
      "step": 1300
    },
    {
      "epoch": 6.55,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.4063,
      "step": 1310
    },
    {
      "epoch": 6.6,
      "grad_norm": 12.5625,
      "learning_rate": 8e-05,
      "loss": 0.3575,
      "step": 1320
    },
    {
      "epoch": 6.65,
      "grad_norm": 5.875,
      "learning_rate": 8e-05,
      "loss": 0.3622,
      "step": 1330
    },
    {
      "epoch": 6.7,
      "grad_norm": 13.5625,
      "learning_rate": 8e-05,
      "loss": 0.3387,
      "step": 1340
    },
    {
      "epoch": 6.75,
      "grad_norm": 4.96875,
      "learning_rate": 8e-05,
      "loss": 0.3373,
      "step": 1350
    },
    {
      "epoch": 6.8,
      "grad_norm": 6.375,
      "learning_rate": 8e-05,
      "loss": 0.3143,
      "step": 1360
    },
    {
      "epoch": 6.85,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.3434,
      "step": 1370
    },
    {
      "epoch": 6.9,
      "grad_norm": 4.34375,
      "learning_rate": 8e-05,
      "loss": 0.3255,
      "step": 1380
    },
    {
      "epoch": 6.95,
      "grad_norm": 6.65625,
      "learning_rate": 8e-05,
      "loss": 0.357,
      "step": 1390
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.0625,
      "learning_rate": 8e-05,
      "loss": 0.3876,
      "step": 1400
    },
    {
      "epoch": 7.05,
      "grad_norm": 2.40625,
      "learning_rate": 8e-05,
      "loss": 0.3567,
      "step": 1410
    },
    {
      "epoch": 7.1,
      "grad_norm": 10.0625,
      "learning_rate": 8e-05,
      "loss": 0.3101,
      "step": 1420
    },
    {
      "epoch": 7.15,
      "grad_norm": 7.03125,
      "learning_rate": 8e-05,
      "loss": 0.3498,
      "step": 1430
    },
    {
      "epoch": 7.2,
      "grad_norm": 2.28125,
      "learning_rate": 8e-05,
      "loss": 0.2873,
      "step": 1440
    },
    {
      "epoch": 7.25,
      "grad_norm": 3.609375,
      "learning_rate": 8e-05,
      "loss": 0.2651,
      "step": 1450
    },
    {
      "epoch": 7.3,
      "grad_norm": 2.3125,
      "learning_rate": 8e-05,
      "loss": 0.2505,
      "step": 1460
    },
    {
      "epoch": 7.35,
      "grad_norm": 6.375,
      "learning_rate": 8e-05,
      "loss": 0.2421,
      "step": 1470
    },
    {
      "epoch": 7.4,
      "grad_norm": 3.34375,
      "learning_rate": 8e-05,
      "loss": 0.3338,
      "step": 1480
    },
    {
      "epoch": 7.45,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.4454,
      "step": 1490
    },
    {
      "epoch": 7.5,
      "grad_norm": 4.0,
      "learning_rate": 8e-05,
      "loss": 0.4056,
      "step": 1500
    },
    {
      "epoch": 7.5,
      "eval_loss": 7.630816459655762,
      "eval_runtime": 7.5449,
      "eval_samples_per_second": 66.27,
      "eval_steps_per_second": 66.27,
      "step": 1500
    },
    {
      "epoch": 7.55,
      "grad_norm": 2.671875,
      "learning_rate": 8e-05,
      "loss": 0.2285,
      "step": 1510
    },
    {
      "epoch": 7.6,
      "grad_norm": 14.4375,
      "learning_rate": 8e-05,
      "loss": 0.3527,
      "step": 1520
    },
    {
      "epoch": 7.65,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.4113,
      "step": 1530
    },
    {
      "epoch": 7.7,
      "grad_norm": 10.4375,
      "learning_rate": 8e-05,
      "loss": 0.4207,
      "step": 1540
    },
    {
      "epoch": 7.75,
      "grad_norm": 2.265625,
      "learning_rate": 8e-05,
      "loss": 0.3073,
      "step": 1550
    },
    {
      "epoch": 7.8,
      "grad_norm": 9.75,
      "learning_rate": 8e-05,
      "loss": 0.3525,
      "step": 1560
    },
    {
      "epoch": 7.85,
      "grad_norm": 10.0625,
      "learning_rate": 8e-05,
      "loss": 0.4336,
      "step": 1570
    },
    {
      "epoch": 7.9,
      "grad_norm": 3.015625,
      "learning_rate": 8e-05,
      "loss": 0.3153,
      "step": 1580
    },
    {
      "epoch": 7.95,
      "grad_norm": 12.3125,
      "learning_rate": 8e-05,
      "loss": 0.311,
      "step": 1590
    },
    {
      "epoch": 8.0,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.2972,
      "step": 1600
    },
    {
      "epoch": 8.05,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.3153,
      "step": 1610
    },
    {
      "epoch": 8.1,
      "grad_norm": 4.4375,
      "learning_rate": 8e-05,
      "loss": 0.3684,
      "step": 1620
    },
    {
      "epoch": 8.15,
      "grad_norm": 11.5,
      "learning_rate": 8e-05,
      "loss": 0.1658,
      "step": 1630
    },
    {
      "epoch": 8.2,
      "grad_norm": 11.4375,
      "learning_rate": 8e-05,
      "loss": 0.317,
      "step": 1640
    },
    {
      "epoch": 8.25,
      "grad_norm": 1.9296875,
      "learning_rate": 8e-05,
      "loss": 0.2382,
      "step": 1650
    },
    {
      "epoch": 8.3,
      "grad_norm": 2.296875,
      "learning_rate": 8e-05,
      "loss": 0.3048,
      "step": 1660
    },
    {
      "epoch": 8.35,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.3565,
      "step": 1670
    },
    {
      "epoch": 8.4,
      "grad_norm": 11.25,
      "learning_rate": 8e-05,
      "loss": 0.2439,
      "step": 1680
    },
    {
      "epoch": 8.45,
      "grad_norm": 2.65625,
      "learning_rate": 8e-05,
      "loss": 0.3198,
      "step": 1690
    },
    {
      "epoch": 8.5,
      "grad_norm": 11.5,
      "learning_rate": 8e-05,
      "loss": 0.3728,
      "step": 1700
    },
    {
      "epoch": 8.55,
      "grad_norm": 5.21875,
      "learning_rate": 8e-05,
      "loss": 0.3061,
      "step": 1710
    },
    {
      "epoch": 8.6,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.5014,
      "step": 1720
    },
    {
      "epoch": 8.65,
      "grad_norm": 1.6953125,
      "learning_rate": 8e-05,
      "loss": 0.2326,
      "step": 1730
    },
    {
      "epoch": 8.7,
      "grad_norm": 4.0,
      "learning_rate": 8e-05,
      "loss": 0.2521,
      "step": 1740
    },
    {
      "epoch": 8.75,
      "grad_norm": 2.140625,
      "learning_rate": 8e-05,
      "loss": 0.3909,
      "step": 1750
    },
    {
      "epoch": 8.8,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.3824,
      "step": 1760
    },
    {
      "epoch": 8.85,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.3043,
      "step": 1770
    },
    {
      "epoch": 8.9,
      "grad_norm": 7.21875,
      "learning_rate": 8e-05,
      "loss": 0.3079,
      "step": 1780
    },
    {
      "epoch": 8.95,
      "grad_norm": 2.625,
      "learning_rate": 8e-05,
      "loss": 0.336,
      "step": 1790
    },
    {
      "epoch": 9.0,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.3156,
      "step": 1800
    },
    {
      "epoch": 9.05,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.258,
      "step": 1810
    },
    {
      "epoch": 9.1,
      "grad_norm": 3.234375,
      "learning_rate": 8e-05,
      "loss": 0.3333,
      "step": 1820
    },
    {
      "epoch": 9.15,
      "grad_norm": 12.5625,
      "learning_rate": 8e-05,
      "loss": 0.4112,
      "step": 1830
    },
    {
      "epoch": 9.2,
      "grad_norm": 3.890625,
      "learning_rate": 8e-05,
      "loss": 0.3298,
      "step": 1840
    },
    {
      "epoch": 9.25,
      "grad_norm": 2.390625,
      "learning_rate": 8e-05,
      "loss": 0.3552,
      "step": 1850
    },
    {
      "epoch": 9.3,
      "grad_norm": 11.0,
      "learning_rate": 8e-05,
      "loss": 0.196,
      "step": 1860
    },
    {
      "epoch": 9.35,
      "grad_norm": 5.09375,
      "learning_rate": 8e-05,
      "loss": 0.2146,
      "step": 1870
    },
    {
      "epoch": 9.4,
      "grad_norm": 2.5625,
      "learning_rate": 8e-05,
      "loss": 0.2204,
      "step": 1880
    },
    {
      "epoch": 9.45,
      "grad_norm": 2.90625,
      "learning_rate": 8e-05,
      "loss": 0.3717,
      "step": 1890
    },
    {
      "epoch": 9.5,
      "grad_norm": 11.1875,
      "learning_rate": 8e-05,
      "loss": 0.2952,
      "step": 1900
    },
    {
      "epoch": 9.55,
      "grad_norm": 11.375,
      "learning_rate": 8e-05,
      "loss": 0.3634,
      "step": 1910
    },
    {
      "epoch": 9.6,
      "grad_norm": 1.8515625,
      "learning_rate": 8e-05,
      "loss": 0.327,
      "step": 1920
    },
    {
      "epoch": 9.65,
      "grad_norm": 11.25,
      "learning_rate": 8e-05,
      "loss": 0.2675,
      "step": 1930
    },
    {
      "epoch": 9.7,
      "grad_norm": 13.6875,
      "learning_rate": 8e-05,
      "loss": 0.289,
      "step": 1940
    },
    {
      "epoch": 9.75,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.4264,
      "step": 1950
    },
    {
      "epoch": 9.8,
      "grad_norm": 2.21875,
      "learning_rate": 8e-05,
      "loss": 0.2596,
      "step": 1960
    },
    {
      "epoch": 9.85,
      "grad_norm": 3.546875,
      "learning_rate": 8e-05,
      "loss": 0.2564,
      "step": 1970
    },
    {
      "epoch": 9.9,
      "grad_norm": 4.0,
      "learning_rate": 8e-05,
      "loss": 0.3128,
      "step": 1980
    },
    {
      "epoch": 9.95,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.4754,
      "step": 1990
    },
    {
      "epoch": 10.0,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.3452,
      "step": 2000
    },
    {
      "epoch": 10.0,
      "eval_loss": 7.939090728759766,
      "eval_runtime": 7.5613,
      "eval_samples_per_second": 66.126,
      "eval_steps_per_second": 66.126,
      "step": 2000
    },
    {
      "epoch": 10.05,
      "grad_norm": 1.6484375,
      "learning_rate": 8e-05,
      "loss": 0.3269,
      "step": 2010
    },
    {
      "epoch": 10.1,
      "grad_norm": 3.40625,
      "learning_rate": 8e-05,
      "loss": 0.3554,
      "step": 2020
    },
    {
      "epoch": 10.15,
      "grad_norm": 11.375,
      "learning_rate": 8e-05,
      "loss": 0.3173,
      "step": 2030
    },
    {
      "epoch": 10.2,
      "grad_norm": 1.515625,
      "learning_rate": 8e-05,
      "loss": 0.2155,
      "step": 2040
    },
    {
      "epoch": 10.25,
      "grad_norm": 2.546875,
      "learning_rate": 8e-05,
      "loss": 0.2992,
      "step": 2050
    },
    {
      "epoch": 10.3,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.4483,
      "step": 2060
    },
    {
      "epoch": 10.35,
      "grad_norm": 10.75,
      "learning_rate": 8e-05,
      "loss": 0.4116,
      "step": 2070
    },
    {
      "epoch": 10.4,
      "grad_norm": 3.046875,
      "learning_rate": 8e-05,
      "loss": 0.2901,
      "step": 2080
    },
    {
      "epoch": 10.45,
      "grad_norm": 2.84375,
      "learning_rate": 8e-05,
      "loss": 0.3026,
      "step": 2090
    },
    {
      "epoch": 10.5,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.3183,
      "step": 2100
    },
    {
      "epoch": 10.55,
      "grad_norm": 12.5625,
      "learning_rate": 8e-05,
      "loss": 0.263,
      "step": 2110
    },
    {
      "epoch": 10.6,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.2742,
      "step": 2120
    },
    {
      "epoch": 10.65,
      "grad_norm": 3.6875,
      "learning_rate": 8e-05,
      "loss": 0.3583,
      "step": 2130
    },
    {
      "epoch": 10.7,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.244,
      "step": 2140
    },
    {
      "epoch": 10.75,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.2986,
      "step": 2150
    },
    {
      "epoch": 10.8,
      "grad_norm": 1.9453125,
      "learning_rate": 8e-05,
      "loss": 0.2685,
      "step": 2160
    },
    {
      "epoch": 10.85,
      "grad_norm": 1.7734375,
      "learning_rate": 8e-05,
      "loss": 0.239,
      "step": 2170
    },
    {
      "epoch": 10.9,
      "grad_norm": 2.375,
      "learning_rate": 8e-05,
      "loss": 0.25,
      "step": 2180
    },
    {
      "epoch": 10.95,
      "grad_norm": 2.0,
      "learning_rate": 8e-05,
      "loss": 0.2264,
      "step": 2190
    },
    {
      "epoch": 11.0,
      "grad_norm": 13.75,
      "learning_rate": 8e-05,
      "loss": 0.403,
      "step": 2200
    },
    {
      "epoch": 11.05,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.3125,
      "step": 2210
    },
    {
      "epoch": 11.1,
      "grad_norm": 53.25,
      "learning_rate": 8e-05,
      "loss": 0.3731,
      "step": 2220
    },
    {
      "epoch": 11.15,
      "grad_norm": 11.625,
      "learning_rate": 8e-05,
      "loss": 0.2595,
      "step": 2230
    },
    {
      "epoch": 11.2,
      "grad_norm": 3.953125,
      "learning_rate": 8e-05,
      "loss": 0.3029,
      "step": 2240
    },
    {
      "epoch": 11.25,
      "grad_norm": 13.625,
      "learning_rate": 8e-05,
      "loss": 0.2089,
      "step": 2250
    },
    {
      "epoch": 11.3,
      "grad_norm": 4.90625,
      "learning_rate": 8e-05,
      "loss": 0.3185,
      "step": 2260
    },
    {
      "epoch": 11.35,
      "grad_norm": 1.7109375,
      "learning_rate": 8e-05,
      "loss": 0.254,
      "step": 2270
    },
    {
      "epoch": 11.4,
      "grad_norm": 11.6875,
      "learning_rate": 8e-05,
      "loss": 0.2984,
      "step": 2280
    },
    {
      "epoch": 11.45,
      "grad_norm": 1.390625,
      "learning_rate": 8e-05,
      "loss": 0.3858,
      "step": 2290
    },
    {
      "epoch": 11.5,
      "grad_norm": 2.53125,
      "learning_rate": 8e-05,
      "loss": 0.3145,
      "step": 2300
    },
    {
      "epoch": 11.55,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.3565,
      "step": 2310
    },
    {
      "epoch": 11.6,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.3711,
      "step": 2320
    },
    {
      "epoch": 11.65,
      "grad_norm": 2.625,
      "learning_rate": 8e-05,
      "loss": 0.3212,
      "step": 2330
    },
    {
      "epoch": 11.7,
      "grad_norm": 1.1171875,
      "learning_rate": 8e-05,
      "loss": 0.2068,
      "step": 2340
    },
    {
      "epoch": 11.75,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.3058,
      "step": 2350
    },
    {
      "epoch": 11.8,
      "grad_norm": 3.6875,
      "learning_rate": 8e-05,
      "loss": 0.201,
      "step": 2360
    },
    {
      "epoch": 11.85,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.3403,
      "step": 2370
    },
    {
      "epoch": 11.9,
      "grad_norm": 2.796875,
      "learning_rate": 8e-05,
      "loss": 0.2613,
      "step": 2380
    },
    {
      "epoch": 11.95,
      "grad_norm": 11.5,
      "learning_rate": 8e-05,
      "loss": 0.3407,
      "step": 2390
    },
    {
      "epoch": 12.0,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.3051,
      "step": 2400
    },
    {
      "epoch": 12.05,
      "grad_norm": 1.7578125,
      "learning_rate": 8e-05,
      "loss": 0.2997,
      "step": 2410
    },
    {
      "epoch": 12.1,
      "grad_norm": 11.3125,
      "learning_rate": 8e-05,
      "loss": 0.3016,
      "step": 2420
    },
    {
      "epoch": 12.15,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.3798,
      "step": 2430
    },
    {
      "epoch": 12.2,
      "grad_norm": 2.5,
      "learning_rate": 8e-05,
      "loss": 0.2223,
      "step": 2440
    },
    {
      "epoch": 12.25,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.2249,
      "step": 2450
    },
    {
      "epoch": 12.3,
      "grad_norm": 1.9453125,
      "learning_rate": 8e-05,
      "loss": 0.2264,
      "step": 2460
    },
    {
      "epoch": 12.35,
      "grad_norm": 1.21875,
      "learning_rate": 8e-05,
      "loss": 0.3619,
      "step": 2470
    },
    {
      "epoch": 12.4,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.3665,
      "step": 2480
    },
    {
      "epoch": 12.45,
      "grad_norm": 1.46875,
      "learning_rate": 8e-05,
      "loss": 0.3139,
      "step": 2490
    },
    {
      "epoch": 12.5,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.4703,
      "step": 2500
    },
    {
      "epoch": 12.5,
      "eval_loss": 7.897547721862793,
      "eval_runtime": 7.6709,
      "eval_samples_per_second": 65.181,
      "eval_steps_per_second": 65.181,
      "step": 2500
    },
    {
      "epoch": 12.55,
      "grad_norm": 1.6484375,
      "learning_rate": 8e-05,
      "loss": 0.2518,
      "step": 2510
    },
    {
      "epoch": 12.6,
      "grad_norm": 2.390625,
      "learning_rate": 8e-05,
      "loss": 0.2905,
      "step": 2520
    },
    {
      "epoch": 12.65,
      "grad_norm": 10.5625,
      "learning_rate": 8e-05,
      "loss": 0.2674,
      "step": 2530
    },
    {
      "epoch": 12.7,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.2041,
      "step": 2540
    },
    {
      "epoch": 12.75,
      "grad_norm": 3.140625,
      "learning_rate": 8e-05,
      "loss": 0.258,
      "step": 2550
    },
    {
      "epoch": 12.8,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.2959,
      "step": 2560
    },
    {
      "epoch": 12.85,
      "grad_norm": 1.4453125,
      "learning_rate": 8e-05,
      "loss": 0.2758,
      "step": 2570
    },
    {
      "epoch": 12.9,
      "grad_norm": 1.0390625,
      "learning_rate": 8e-05,
      "loss": 0.2456,
      "step": 2580
    },
    {
      "epoch": 12.95,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.3101,
      "step": 2590
    },
    {
      "epoch": 13.0,
      "grad_norm": 2.84375,
      "learning_rate": 8e-05,
      "loss": 0.2981,
      "step": 2600
    },
    {
      "epoch": 13.05,
      "grad_norm": 1.375,
      "learning_rate": 8e-05,
      "loss": 0.2467,
      "step": 2610
    },
    {
      "epoch": 13.1,
      "grad_norm": 1.328125,
      "learning_rate": 8e-05,
      "loss": 0.249,
      "step": 2620
    },
    {
      "epoch": 13.15,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.382,
      "step": 2630
    },
    {
      "epoch": 13.2,
      "grad_norm": 1.8515625,
      "learning_rate": 8e-05,
      "loss": 0.3436,
      "step": 2640
    },
    {
      "epoch": 13.25,
      "grad_norm": 2.5,
      "learning_rate": 8e-05,
      "loss": 0.2819,
      "step": 2650
    },
    {
      "epoch": 13.3,
      "grad_norm": 3.6875,
      "learning_rate": 8e-05,
      "loss": 0.2695,
      "step": 2660
    },
    {
      "epoch": 13.35,
      "grad_norm": 1.75,
      "learning_rate": 8e-05,
      "loss": 0.1805,
      "step": 2670
    },
    {
      "epoch": 13.4,
      "grad_norm": 11.6875,
      "learning_rate": 8e-05,
      "loss": 0.3324,
      "step": 2680
    },
    {
      "epoch": 13.45,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.4207,
      "step": 2690
    },
    {
      "epoch": 13.5,
      "grad_norm": 2.015625,
      "learning_rate": 8e-05,
      "loss": 0.36,
      "step": 2700
    },
    {
      "epoch": 13.55,
      "grad_norm": 11.5625,
      "learning_rate": 8e-05,
      "loss": 0.3061,
      "step": 2710
    },
    {
      "epoch": 13.6,
      "grad_norm": 1.703125,
      "learning_rate": 8e-05,
      "loss": 0.2009,
      "step": 2720
    },
    {
      "epoch": 13.65,
      "grad_norm": 10.625,
      "learning_rate": 8e-05,
      "loss": 0.3731,
      "step": 2730
    },
    {
      "epoch": 13.7,
      "grad_norm": 11.8125,
      "learning_rate": 8e-05,
      "loss": 0.2775,
      "step": 2740
    },
    {
      "epoch": 13.75,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.2231,
      "step": 2750
    },
    {
      "epoch": 13.8,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.3238,
      "step": 2760
    },
    {
      "epoch": 13.85,
      "grad_norm": 1.1484375,
      "learning_rate": 8e-05,
      "loss": 0.2291,
      "step": 2770
    },
    {
      "epoch": 13.9,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.2505,
      "step": 2780
    },
    {
      "epoch": 13.95,
      "grad_norm": 2.703125,
      "learning_rate": 8e-05,
      "loss": 0.352,
      "step": 2790
    },
    {
      "epoch": 14.0,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.1966,
      "step": 2800
    },
    {
      "epoch": 14.05,
      "grad_norm": 11.5625,
      "learning_rate": 8e-05,
      "loss": 0.2186,
      "step": 2810
    },
    {
      "epoch": 14.1,
      "grad_norm": 2.65625,
      "learning_rate": 8e-05,
      "loss": 0.2466,
      "step": 2820
    },
    {
      "epoch": 14.15,
      "grad_norm": 1.3359375,
      "learning_rate": 8e-05,
      "loss": 0.1781,
      "step": 2830
    },
    {
      "epoch": 14.2,
      "grad_norm": 2.59375,
      "learning_rate": 8e-05,
      "loss": 0.3363,
      "step": 2840
    },
    {
      "epoch": 14.25,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.3329,
      "step": 2850
    },
    {
      "epoch": 14.3,
      "grad_norm": 1.7890625,
      "learning_rate": 8e-05,
      "loss": 0.3858,
      "step": 2860
    },
    {
      "epoch": 14.35,
      "grad_norm": 0.9765625,
      "learning_rate": 8e-05,
      "loss": 0.2778,
      "step": 2870
    },
    {
      "epoch": 14.4,
      "grad_norm": 1.25,
      "learning_rate": 8e-05,
      "loss": 0.1987,
      "step": 2880
    },
    {
      "epoch": 14.45,
      "grad_norm": 1.921875,
      "learning_rate": 8e-05,
      "loss": 0.1794,
      "step": 2890
    },
    {
      "epoch": 14.5,
      "grad_norm": 1.5859375,
      "learning_rate": 8e-05,
      "loss": 0.2878,
      "step": 2900
    },
    {
      "epoch": 14.55,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.3154,
      "step": 2910
    },
    {
      "epoch": 14.6,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.3468,
      "step": 2920
    },
    {
      "epoch": 14.65,
      "grad_norm": 1.484375,
      "learning_rate": 8e-05,
      "loss": 0.2626,
      "step": 2930
    },
    {
      "epoch": 14.7,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.2456,
      "step": 2940
    },
    {
      "epoch": 14.75,
      "grad_norm": 2.0,
      "learning_rate": 8e-05,
      "loss": 0.2751,
      "step": 2950
    },
    {
      "epoch": 14.8,
      "grad_norm": 1.96875,
      "learning_rate": 8e-05,
      "loss": 0.2335,
      "step": 2960
    },
    {
      "epoch": 14.85,
      "grad_norm": 2.203125,
      "learning_rate": 8e-05,
      "loss": 0.3561,
      "step": 2970
    },
    {
      "epoch": 14.9,
      "grad_norm": 0.98828125,
      "learning_rate": 8e-05,
      "loss": 0.3794,
      "step": 2980
    },
    {
      "epoch": 14.95,
      "grad_norm": 1.84375,
      "learning_rate": 8e-05,
      "loss": 0.2901,
      "step": 2990
    },
    {
      "epoch": 15.0,
      "grad_norm": 12.125,
      "learning_rate": 8e-05,
      "loss": 0.2891,
      "step": 3000
    },
    {
      "epoch": 15.0,
      "eval_loss": 7.215864181518555,
      "eval_runtime": 7.9685,
      "eval_samples_per_second": 62.747,
      "eval_steps_per_second": 62.747,
      "step": 3000
    },
    {
      "epoch": 15.05,
      "grad_norm": 1.609375,
      "learning_rate": 8e-05,
      "loss": 0.1937,
      "step": 3010
    },
    {
      "epoch": 15.1,
      "grad_norm": 13.4375,
      "learning_rate": 8e-05,
      "loss": 0.3008,
      "step": 3020
    },
    {
      "epoch": 15.15,
      "grad_norm": 1.5546875,
      "learning_rate": 8e-05,
      "loss": 0.1713,
      "step": 3030
    },
    {
      "epoch": 15.2,
      "grad_norm": 1.0390625,
      "learning_rate": 8e-05,
      "loss": 0.2894,
      "step": 3040
    },
    {
      "epoch": 15.25,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.404,
      "step": 3050
    },
    {
      "epoch": 15.3,
      "grad_norm": 10.75,
      "learning_rate": 8e-05,
      "loss": 0.3772,
      "step": 3060
    },
    {
      "epoch": 15.35,
      "grad_norm": 13.5625,
      "learning_rate": 8e-05,
      "loss": 0.2001,
      "step": 3070
    },
    {
      "epoch": 15.4,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.2303,
      "step": 3080
    },
    {
      "epoch": 15.45,
      "grad_norm": 1.78125,
      "learning_rate": 8e-05,
      "loss": 0.2211,
      "step": 3090
    },
    {
      "epoch": 15.5,
      "grad_norm": 11.5,
      "learning_rate": 8e-05,
      "loss": 0.3988,
      "step": 3100
    },
    {
      "epoch": 15.55,
      "grad_norm": 11.5,
      "learning_rate": 8e-05,
      "loss": 0.1864,
      "step": 3110
    },
    {
      "epoch": 15.6,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.3078,
      "step": 3120
    },
    {
      "epoch": 15.65,
      "grad_norm": 1.90625,
      "learning_rate": 8e-05,
      "loss": 0.2891,
      "step": 3130
    },
    {
      "epoch": 15.7,
      "grad_norm": 11.9375,
      "learning_rate": 8e-05,
      "loss": 0.4039,
      "step": 3140
    },
    {
      "epoch": 15.75,
      "grad_norm": 1.4609375,
      "learning_rate": 8e-05,
      "loss": 0.2887,
      "step": 3150
    },
    {
      "epoch": 15.8,
      "grad_norm": 2.359375,
      "learning_rate": 8e-05,
      "loss": 0.3268,
      "step": 3160
    },
    {
      "epoch": 15.85,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.2726,
      "step": 3170
    },
    {
      "epoch": 15.9,
      "grad_norm": 1.9453125,
      "learning_rate": 8e-05,
      "loss": 0.1886,
      "step": 3180
    },
    {
      "epoch": 15.95,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.3537,
      "step": 3190
    },
    {
      "epoch": 16.0,
      "grad_norm": 2.0625,
      "learning_rate": 8e-05,
      "loss": 0.2877,
      "step": 3200
    },
    {
      "epoch": 16.05,
      "grad_norm": 0.96484375,
      "learning_rate": 8e-05,
      "loss": 0.2848,
      "step": 3210
    },
    {
      "epoch": 16.1,
      "grad_norm": 2.390625,
      "learning_rate": 8e-05,
      "loss": 0.1388,
      "step": 3220
    },
    {
      "epoch": 16.15,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.2291,
      "step": 3230
    },
    {
      "epoch": 16.2,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.3616,
      "step": 3240
    },
    {
      "epoch": 16.25,
      "grad_norm": 1.6015625,
      "learning_rate": 8e-05,
      "loss": 0.2676,
      "step": 3250
    },
    {
      "epoch": 16.3,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.4425,
      "step": 3260
    },
    {
      "epoch": 16.35,
      "grad_norm": 13.4375,
      "learning_rate": 8e-05,
      "loss": 0.3095,
      "step": 3270
    },
    {
      "epoch": 16.4,
      "grad_norm": 2.125,
      "learning_rate": 8e-05,
      "loss": 0.2071,
      "step": 3280
    },
    {
      "epoch": 16.45,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.3159,
      "step": 3290
    },
    {
      "epoch": 16.5,
      "grad_norm": 1.765625,
      "learning_rate": 8e-05,
      "loss": 0.3892,
      "step": 3300
    },
    {
      "epoch": 16.55,
      "grad_norm": 1.15625,
      "learning_rate": 8e-05,
      "loss": 0.2224,
      "step": 3310
    },
    {
      "epoch": 16.6,
      "grad_norm": 1.6796875,
      "learning_rate": 8e-05,
      "loss": 0.2233,
      "step": 3320
    },
    {
      "epoch": 16.65,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.3661,
      "step": 3330
    },
    {
      "epoch": 16.7,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.3293,
      "step": 3340
    },
    {
      "epoch": 16.75,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.3403,
      "step": 3350
    },
    {
      "epoch": 16.8,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.2894,
      "step": 3360
    },
    {
      "epoch": 16.85,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.2419,
      "step": 3370
    },
    {
      "epoch": 16.9,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.3061,
      "step": 3380
    },
    {
      "epoch": 16.95,
      "grad_norm": 6.09375,
      "learning_rate": 8e-05,
      "loss": 0.2204,
      "step": 3390
    },
    {
      "epoch": 17.0,
      "grad_norm": 1.921875,
      "learning_rate": 8e-05,
      "loss": 0.2497,
      "step": 3400
    },
    {
      "epoch": 17.05,
      "grad_norm": 10.75,
      "learning_rate": 8e-05,
      "loss": 0.2295,
      "step": 3410
    },
    {
      "epoch": 17.1,
      "grad_norm": 1.6328125,
      "learning_rate": 8e-05,
      "loss": 0.3195,
      "step": 3420
    },
    {
      "epoch": 17.15,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.2639,
      "step": 3430
    },
    {
      "epoch": 17.2,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.283,
      "step": 3440
    },
    {
      "epoch": 17.25,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.3231,
      "step": 3450
    },
    {
      "epoch": 17.3,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.204,
      "step": 3460
    },
    {
      "epoch": 17.35,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.2669,
      "step": 3470
    },
    {
      "epoch": 17.4,
      "grad_norm": 11.9375,
      "learning_rate": 8e-05,
      "loss": 0.2475,
      "step": 3480
    },
    {
      "epoch": 17.45,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.3369,
      "step": 3490
    },
    {
      "epoch": 17.5,
      "grad_norm": 2.5,
      "learning_rate": 8e-05,
      "loss": 0.3026,
      "step": 3500
    },
    {
      "epoch": 17.5,
      "eval_loss": 8.5922269821167,
      "eval_runtime": 7.9526,
      "eval_samples_per_second": 62.873,
      "eval_steps_per_second": 62.873,
      "step": 3500
    },
    {
      "epoch": 17.55,
      "grad_norm": 2.921875,
      "learning_rate": 8e-05,
      "loss": 0.3049,
      "step": 3510
    },
    {
      "epoch": 17.6,
      "grad_norm": 1.8984375,
      "learning_rate": 8e-05,
      "loss": 0.2318,
      "step": 3520
    },
    {
      "epoch": 17.65,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.3433,
      "step": 3530
    },
    {
      "epoch": 17.7,
      "grad_norm": 1.4765625,
      "learning_rate": 8e-05,
      "loss": 0.265,
      "step": 3540
    },
    {
      "epoch": 17.75,
      "grad_norm": 2.21875,
      "learning_rate": 8e-05,
      "loss": 0.3195,
      "step": 3550
    },
    {
      "epoch": 17.8,
      "grad_norm": 11.1875,
      "learning_rate": 8e-05,
      "loss": 0.3147,
      "step": 3560
    },
    {
      "epoch": 17.85,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.2251,
      "step": 3570
    },
    {
      "epoch": 17.9,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.1943,
      "step": 3580
    },
    {
      "epoch": 17.95,
      "grad_norm": 13.6875,
      "learning_rate": 8e-05,
      "loss": 0.3524,
      "step": 3590
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.9375,
      "learning_rate": 8e-05,
      "loss": 0.2027,
      "step": 3600
    },
    {
      "epoch": 18.05,
      "grad_norm": 1.1484375,
      "learning_rate": 8e-05,
      "loss": 0.1753,
      "step": 3610
    },
    {
      "epoch": 18.1,
      "grad_norm": 1.2421875,
      "learning_rate": 8e-05,
      "loss": 0.3361,
      "step": 3620
    },
    {
      "epoch": 18.15,
      "grad_norm": 1.0390625,
      "learning_rate": 8e-05,
      "loss": 0.2835,
      "step": 3630
    },
    {
      "epoch": 18.2,
      "grad_norm": 1.6640625,
      "learning_rate": 8e-05,
      "loss": 0.247,
      "step": 3640
    },
    {
      "epoch": 18.25,
      "grad_norm": 2.15625,
      "learning_rate": 8e-05,
      "loss": 0.2648,
      "step": 3650
    },
    {
      "epoch": 18.3,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.2733,
      "step": 3660
    },
    {
      "epoch": 18.35,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.2856,
      "step": 3670
    },
    {
      "epoch": 18.4,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.2245,
      "step": 3680
    },
    {
      "epoch": 18.45,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.2948,
      "step": 3690
    },
    {
      "epoch": 18.5,
      "grad_norm": 1.265625,
      "learning_rate": 8e-05,
      "loss": 0.2622,
      "step": 3700
    },
    {
      "epoch": 18.55,
      "grad_norm": 10.9375,
      "learning_rate": 8e-05,
      "loss": 0.3569,
      "step": 3710
    },
    {
      "epoch": 18.6,
      "grad_norm": 2.953125,
      "learning_rate": 8e-05,
      "loss": 0.2437,
      "step": 3720
    },
    {
      "epoch": 18.65,
      "grad_norm": 1.9921875,
      "learning_rate": 8e-05,
      "loss": 0.3894,
      "step": 3730
    },
    {
      "epoch": 18.7,
      "grad_norm": 1.4140625,
      "learning_rate": 8e-05,
      "loss": 0.2614,
      "step": 3740
    },
    {
      "epoch": 18.75,
      "grad_norm": 1.09375,
      "learning_rate": 8e-05,
      "loss": 0.2579,
      "step": 3750
    },
    {
      "epoch": 18.8,
      "grad_norm": 1.078125,
      "learning_rate": 8e-05,
      "loss": 0.2146,
      "step": 3760
    },
    {
      "epoch": 18.85,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.3762,
      "step": 3770
    },
    {
      "epoch": 18.9,
      "grad_norm": 11.625,
      "learning_rate": 8e-05,
      "loss": 0.3321,
      "step": 3780
    },
    {
      "epoch": 18.95,
      "grad_norm": 12.0625,
      "learning_rate": 8e-05,
      "loss": 0.3717,
      "step": 3790
    },
    {
      "epoch": 19.0,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.2763,
      "step": 3800
    },
    {
      "epoch": 19.05,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.1778,
      "step": 3810
    },
    {
      "epoch": 19.1,
      "grad_norm": 1.2578125,
      "learning_rate": 8e-05,
      "loss": 0.1328,
      "step": 3820
    },
    {
      "epoch": 19.15,
      "grad_norm": 1.9765625,
      "learning_rate": 8e-05,
      "loss": 0.1773,
      "step": 3830
    },
    {
      "epoch": 19.2,
      "grad_norm": 1.6640625,
      "learning_rate": 8e-05,
      "loss": 0.2006,
      "step": 3840
    },
    {
      "epoch": 19.25,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.2921,
      "step": 3850
    },
    {
      "epoch": 19.3,
      "grad_norm": 1.65625,
      "learning_rate": 8e-05,
      "loss": 0.1976,
      "step": 3860
    },
    {
      "epoch": 19.35,
      "grad_norm": 13.8125,
      "learning_rate": 8e-05,
      "loss": 0.3143,
      "step": 3870
    },
    {
      "epoch": 19.4,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.2905,
      "step": 3880
    },
    {
      "epoch": 19.45,
      "grad_norm": 1.625,
      "learning_rate": 8e-05,
      "loss": 0.3518,
      "step": 3890
    },
    {
      "epoch": 19.5,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.3462,
      "step": 3900
    },
    {
      "epoch": 19.55,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.2707,
      "step": 3910
    },
    {
      "epoch": 19.6,
      "grad_norm": 5.4375,
      "learning_rate": 8e-05,
      "loss": 0.3661,
      "step": 3920
    },
    {
      "epoch": 19.65,
      "grad_norm": 12.125,
      "learning_rate": 8e-05,
      "loss": 0.2653,
      "step": 3930
    },
    {
      "epoch": 19.7,
      "grad_norm": 11.9375,
      "learning_rate": 8e-05,
      "loss": 0.2482,
      "step": 3940
    },
    {
      "epoch": 19.75,
      "grad_norm": 14.375,
      "learning_rate": 8e-05,
      "loss": 0.3152,
      "step": 3950
    },
    {
      "epoch": 19.8,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.4428,
      "step": 3960
    },
    {
      "epoch": 19.85,
      "grad_norm": 12.25,
      "learning_rate": 8e-05,
      "loss": 0.2869,
      "step": 3970
    },
    {
      "epoch": 19.9,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.3191,
      "step": 3980
    },
    {
      "epoch": 19.95,
      "grad_norm": 4.90625,
      "learning_rate": 8e-05,
      "loss": 0.3486,
      "step": 3990
    },
    {
      "epoch": 20.0,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.3294,
      "step": 4000
    },
    {
      "epoch": 20.0,
      "eval_loss": 8.210659980773926,
      "eval_runtime": 7.5705,
      "eval_samples_per_second": 66.046,
      "eval_steps_per_second": 66.046,
      "step": 4000
    },
    {
      "epoch": 20.05,
      "grad_norm": 11.5,
      "learning_rate": 8e-05,
      "loss": 0.2655,
      "step": 4010
    },
    {
      "epoch": 20.1,
      "grad_norm": 13.125,
      "learning_rate": 8e-05,
      "loss": 0.2366,
      "step": 4020
    },
    {
      "epoch": 20.15,
      "grad_norm": 1.03125,
      "learning_rate": 8e-05,
      "loss": 0.306,
      "step": 4030
    },
    {
      "epoch": 20.2,
      "grad_norm": 1.5078125,
      "learning_rate": 8e-05,
      "loss": 0.1857,
      "step": 4040
    },
    {
      "epoch": 20.25,
      "grad_norm": 0.96875,
      "learning_rate": 8e-05,
      "loss": 0.2701,
      "step": 4050
    },
    {
      "epoch": 20.3,
      "grad_norm": 1.296875,
      "learning_rate": 8e-05,
      "loss": 0.2085,
      "step": 4060
    },
    {
      "epoch": 20.35,
      "grad_norm": 11.375,
      "learning_rate": 8e-05,
      "loss": 0.2882,
      "step": 4070
    },
    {
      "epoch": 20.4,
      "grad_norm": 1.3359375,
      "learning_rate": 8e-05,
      "loss": 0.1754,
      "step": 4080
    },
    {
      "epoch": 20.45,
      "grad_norm": 1.265625,
      "learning_rate": 8e-05,
      "loss": 0.2088,
      "step": 4090
    },
    {
      "epoch": 20.5,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.2538,
      "step": 4100
    },
    {
      "epoch": 20.55,
      "grad_norm": 10.75,
      "learning_rate": 8e-05,
      "loss": 0.27,
      "step": 4110
    },
    {
      "epoch": 20.6,
      "grad_norm": 13.6875,
      "learning_rate": 8e-05,
      "loss": 0.3837,
      "step": 4120
    },
    {
      "epoch": 20.65,
      "grad_norm": 1.1484375,
      "learning_rate": 8e-05,
      "loss": 0.2489,
      "step": 4130
    },
    {
      "epoch": 20.7,
      "grad_norm": 1.4375,
      "learning_rate": 8e-05,
      "loss": 0.2951,
      "step": 4140
    },
    {
      "epoch": 20.75,
      "grad_norm": 10.25,
      "learning_rate": 8e-05,
      "loss": 0.3018,
      "step": 4150
    },
    {
      "epoch": 20.8,
      "grad_norm": 2.15625,
      "learning_rate": 8e-05,
      "loss": 0.3647,
      "step": 4160
    },
    {
      "epoch": 20.85,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.2971,
      "step": 4170
    },
    {
      "epoch": 20.9,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.2797,
      "step": 4180
    },
    {
      "epoch": 20.95,
      "grad_norm": 10.3125,
      "learning_rate": 8e-05,
      "loss": 0.2959,
      "step": 4190
    },
    {
      "epoch": 21.0,
      "grad_norm": 2.40625,
      "learning_rate": 8e-05,
      "loss": 0.3338,
      "step": 4200
    },
    {
      "epoch": 21.05,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.3054,
      "step": 4210
    },
    {
      "epoch": 21.1,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.2705,
      "step": 4220
    },
    {
      "epoch": 21.15,
      "grad_norm": 11.4375,
      "learning_rate": 8e-05,
      "loss": 0.3126,
      "step": 4230
    },
    {
      "epoch": 21.2,
      "grad_norm": 0.97265625,
      "learning_rate": 8e-05,
      "loss": 0.1907,
      "step": 4240
    },
    {
      "epoch": 21.25,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.3798,
      "step": 4250
    },
    {
      "epoch": 21.3,
      "grad_norm": 11.9375,
      "learning_rate": 8e-05,
      "loss": 0.2596,
      "step": 4260
    },
    {
      "epoch": 21.35,
      "grad_norm": 13.625,
      "learning_rate": 8e-05,
      "loss": 0.2034,
      "step": 4270
    },
    {
      "epoch": 21.4,
      "grad_norm": 2.046875,
      "learning_rate": 8e-05,
      "loss": 0.3202,
      "step": 4280
    },
    {
      "epoch": 21.45,
      "grad_norm": 12.25,
      "learning_rate": 8e-05,
      "loss": 0.3012,
      "step": 4290
    },
    {
      "epoch": 21.5,
      "grad_norm": 1.875,
      "learning_rate": 8e-05,
      "loss": 0.2908,
      "step": 4300
    },
    {
      "epoch": 21.55,
      "grad_norm": 1.4609375,
      "learning_rate": 8e-05,
      "loss": 0.2747,
      "step": 4310
    },
    {
      "epoch": 21.6,
      "grad_norm": 1.484375,
      "learning_rate": 8e-05,
      "loss": 0.2408,
      "step": 4320
    },
    {
      "epoch": 21.65,
      "grad_norm": 1.890625,
      "learning_rate": 8e-05,
      "loss": 0.2384,
      "step": 4330
    },
    {
      "epoch": 21.7,
      "grad_norm": 1.8671875,
      "learning_rate": 8e-05,
      "loss": 0.3229,
      "step": 4340
    },
    {
      "epoch": 21.75,
      "grad_norm": 1.046875,
      "learning_rate": 8e-05,
      "loss": 0.273,
      "step": 4350
    },
    {
      "epoch": 21.8,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.2094,
      "step": 4360
    },
    {
      "epoch": 21.85,
      "grad_norm": 1.6796875,
      "learning_rate": 8e-05,
      "loss": 0.3144,
      "step": 4370
    },
    {
      "epoch": 21.9,
      "grad_norm": 1.1328125,
      "learning_rate": 8e-05,
      "loss": 0.296,
      "step": 4380
    },
    {
      "epoch": 21.95,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.3715,
      "step": 4390
    },
    {
      "epoch": 22.0,
      "grad_norm": 1.203125,
      "learning_rate": 8e-05,
      "loss": 0.2191,
      "step": 4400
    },
    {
      "epoch": 22.05,
      "grad_norm": 3.375,
      "learning_rate": 8e-05,
      "loss": 0.2787,
      "step": 4410
    },
    {
      "epoch": 22.1,
      "grad_norm": 1.7578125,
      "learning_rate": 8e-05,
      "loss": 0.1906,
      "step": 4420
    },
    {
      "epoch": 22.15,
      "grad_norm": 1.484375,
      "learning_rate": 8e-05,
      "loss": 0.2839,
      "step": 4430
    },
    {
      "epoch": 22.2,
      "grad_norm": 0.98046875,
      "learning_rate": 8e-05,
      "loss": 0.2678,
      "step": 4440
    },
    {
      "epoch": 22.25,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.1711,
      "step": 4450
    },
    {
      "epoch": 22.3,
      "grad_norm": 1.9609375,
      "learning_rate": 8e-05,
      "loss": 0.2485,
      "step": 4460
    },
    {
      "epoch": 22.35,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.2903,
      "step": 4470
    },
    {
      "epoch": 22.4,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.3669,
      "step": 4480
    },
    {
      "epoch": 22.45,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.3196,
      "step": 4490
    },
    {
      "epoch": 22.5,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.3304,
      "step": 4500
    },
    {
      "epoch": 22.5,
      "eval_loss": 8.893681526184082,
      "eval_runtime": 7.8064,
      "eval_samples_per_second": 64.05,
      "eval_steps_per_second": 64.05,
      "step": 4500
    },
    {
      "epoch": 22.55,
      "grad_norm": 0.96875,
      "learning_rate": 8e-05,
      "loss": 0.2211,
      "step": 4510
    },
    {
      "epoch": 22.6,
      "grad_norm": 1.265625,
      "learning_rate": 8e-05,
      "loss": 0.2872,
      "step": 4520
    },
    {
      "epoch": 22.65,
      "grad_norm": 1.625,
      "learning_rate": 8e-05,
      "loss": 0.2652,
      "step": 4530
    },
    {
      "epoch": 22.7,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.3234,
      "step": 4540
    },
    {
      "epoch": 22.75,
      "grad_norm": 0.93359375,
      "learning_rate": 8e-05,
      "loss": 0.3285,
      "step": 4550
    },
    {
      "epoch": 22.8,
      "grad_norm": 1.03125,
      "learning_rate": 8e-05,
      "loss": 0.2371,
      "step": 4560
    },
    {
      "epoch": 22.85,
      "grad_norm": 1.0,
      "learning_rate": 8e-05,
      "loss": 0.1858,
      "step": 4570
    },
    {
      "epoch": 22.9,
      "grad_norm": 2.1875,
      "learning_rate": 8e-05,
      "loss": 0.2691,
      "step": 4580
    },
    {
      "epoch": 22.95,
      "grad_norm": 4.84375,
      "learning_rate": 8e-05,
      "loss": 0.4012,
      "step": 4590
    },
    {
      "epoch": 23.0,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.3667,
      "step": 4600
    },
    {
      "epoch": 23.05,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.298,
      "step": 4610
    },
    {
      "epoch": 23.1,
      "grad_norm": 12.0,
      "learning_rate": 8e-05,
      "loss": 0.3351,
      "step": 4620
    },
    {
      "epoch": 23.15,
      "grad_norm": 12.5625,
      "learning_rate": 8e-05,
      "loss": 0.298,
      "step": 4630
    },
    {
      "epoch": 23.2,
      "grad_norm": 0.875,
      "learning_rate": 8e-05,
      "loss": 0.0958,
      "step": 4640
    },
    {
      "epoch": 23.25,
      "grad_norm": 0.875,
      "learning_rate": 8e-05,
      "loss": 0.2265,
      "step": 4650
    },
    {
      "epoch": 23.3,
      "grad_norm": 1.8359375,
      "learning_rate": 8e-05,
      "loss": 0.3006,
      "step": 4660
    },
    {
      "epoch": 23.35,
      "grad_norm": 1.7734375,
      "learning_rate": 8e-05,
      "loss": 0.3545,
      "step": 4670
    },
    {
      "epoch": 23.4,
      "grad_norm": 1.109375,
      "learning_rate": 8e-05,
      "loss": 0.276,
      "step": 4680
    },
    {
      "epoch": 23.45,
      "grad_norm": 1.109375,
      "learning_rate": 8e-05,
      "loss": 0.2633,
      "step": 4690
    },
    {
      "epoch": 23.5,
      "grad_norm": 0.953125,
      "learning_rate": 8e-05,
      "loss": 0.2083,
      "step": 4700
    },
    {
      "epoch": 23.55,
      "grad_norm": 2.625,
      "learning_rate": 8e-05,
      "loss": 0.3108,
      "step": 4710
    },
    {
      "epoch": 23.6,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.3238,
      "step": 4720
    },
    {
      "epoch": 23.65,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.3359,
      "step": 4730
    },
    {
      "epoch": 23.7,
      "grad_norm": 11.6875,
      "learning_rate": 8e-05,
      "loss": 0.3277,
      "step": 4740
    },
    {
      "epoch": 23.75,
      "grad_norm": 1.6875,
      "learning_rate": 8e-05,
      "loss": 0.2861,
      "step": 4750
    },
    {
      "epoch": 23.8,
      "grad_norm": 12.125,
      "learning_rate": 8e-05,
      "loss": 0.2311,
      "step": 4760
    },
    {
      "epoch": 23.85,
      "grad_norm": 2.453125,
      "learning_rate": 8e-05,
      "loss": 0.3383,
      "step": 4770
    },
    {
      "epoch": 23.9,
      "grad_norm": 1.1875,
      "learning_rate": 8e-05,
      "loss": 0.2087,
      "step": 4780
    },
    {
      "epoch": 23.95,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.3536,
      "step": 4790
    },
    {
      "epoch": 24.0,
      "grad_norm": 1.7578125,
      "learning_rate": 8e-05,
      "loss": 0.2426,
      "step": 4800
    },
    {
      "epoch": 24.05,
      "grad_norm": 1.7734375,
      "learning_rate": 8e-05,
      "loss": 0.2743,
      "step": 4810
    },
    {
      "epoch": 24.1,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.2697,
      "step": 4820
    },
    {
      "epoch": 24.15,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.1881,
      "step": 4830
    },
    {
      "epoch": 24.2,
      "grad_norm": 0.8671875,
      "learning_rate": 8e-05,
      "loss": 0.2347,
      "step": 4840
    },
    {
      "epoch": 24.25,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.2888,
      "step": 4850
    },
    {
      "epoch": 24.3,
      "grad_norm": 1.2109375,
      "learning_rate": 8e-05,
      "loss": 0.1057,
      "step": 4860
    },
    {
      "epoch": 24.35,
      "grad_norm": 1.859375,
      "learning_rate": 8e-05,
      "loss": 0.2993,
      "step": 4870
    },
    {
      "epoch": 24.4,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.3406,
      "step": 4880
    },
    {
      "epoch": 24.45,
      "grad_norm": 1.2890625,
      "learning_rate": 8e-05,
      "loss": 0.2368,
      "step": 4890
    },
    {
      "epoch": 24.5,
      "grad_norm": 2.03125,
      "learning_rate": 8e-05,
      "loss": 0.2919,
      "step": 4900
    },
    {
      "epoch": 24.55,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.3382,
      "step": 4910
    },
    {
      "epoch": 24.6,
      "grad_norm": 1.8828125,
      "learning_rate": 8e-05,
      "loss": 0.2537,
      "step": 4920
    },
    {
      "epoch": 24.65,
      "grad_norm": 2.015625,
      "learning_rate": 8e-05,
      "loss": 0.2991,
      "step": 4930
    },
    {
      "epoch": 24.7,
      "grad_norm": 1.8984375,
      "learning_rate": 8e-05,
      "loss": 0.2682,
      "step": 4940
    },
    {
      "epoch": 24.75,
      "grad_norm": 2.859375,
      "learning_rate": 8e-05,
      "loss": 0.2845,
      "step": 4950
    },
    {
      "epoch": 24.8,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.3295,
      "step": 4960
    },
    {
      "epoch": 24.85,
      "grad_norm": 2.234375,
      "learning_rate": 8e-05,
      "loss": 0.3334,
      "step": 4970
    },
    {
      "epoch": 24.9,
      "grad_norm": 1.1015625,
      "learning_rate": 8e-05,
      "loss": 0.3461,
      "step": 4980
    },
    {
      "epoch": 24.95,
      "grad_norm": 1.7890625,
      "learning_rate": 8e-05,
      "loss": 0.211,
      "step": 4990
    },
    {
      "epoch": 25.0,
      "grad_norm": 12.8125,
      "learning_rate": 8e-05,
      "loss": 0.2913,
      "step": 5000
    },
    {
      "epoch": 25.0,
      "eval_loss": 8.996872901916504,
      "eval_runtime": 7.5487,
      "eval_samples_per_second": 66.236,
      "eval_steps_per_second": 66.236,
      "step": 5000
    },
    {
      "epoch": 25.05,
      "grad_norm": 1.703125,
      "learning_rate": 8e-05,
      "loss": 0.2376,
      "step": 5010
    },
    {
      "epoch": 25.1,
      "grad_norm": 1.6796875,
      "learning_rate": 8e-05,
      "loss": 0.175,
      "step": 5020
    },
    {
      "epoch": 25.15,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.3706,
      "step": 5030
    },
    {
      "epoch": 25.2,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.3528,
      "step": 5040
    },
    {
      "epoch": 25.25,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.44,
      "step": 5050
    },
    {
      "epoch": 25.3,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.1827,
      "step": 5060
    },
    {
      "epoch": 25.35,
      "grad_norm": 13.3125,
      "learning_rate": 8e-05,
      "loss": 0.3465,
      "step": 5070
    },
    {
      "epoch": 25.4,
      "grad_norm": 1.0625,
      "learning_rate": 8e-05,
      "loss": 0.2745,
      "step": 5080
    },
    {
      "epoch": 25.45,
      "grad_norm": 0.9453125,
      "learning_rate": 8e-05,
      "loss": 0.1351,
      "step": 5090
    },
    {
      "epoch": 25.5,
      "grad_norm": 13.625,
      "learning_rate": 8e-05,
      "loss": 0.3016,
      "step": 5100
    },
    {
      "epoch": 25.55,
      "grad_norm": 0.98828125,
      "learning_rate": 8e-05,
      "loss": 0.1977,
      "step": 5110
    },
    {
      "epoch": 25.6,
      "grad_norm": 0.890625,
      "learning_rate": 8e-05,
      "loss": 0.1249,
      "step": 5120
    },
    {
      "epoch": 25.65,
      "grad_norm": 2.515625,
      "learning_rate": 8e-05,
      "loss": 0.3311,
      "step": 5130
    },
    {
      "epoch": 25.7,
      "grad_norm": 1.125,
      "learning_rate": 8e-05,
      "loss": 0.2879,
      "step": 5140
    },
    {
      "epoch": 25.75,
      "grad_norm": 2.28125,
      "learning_rate": 8e-05,
      "loss": 0.1996,
      "step": 5150
    },
    {
      "epoch": 25.8,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.3853,
      "step": 5160
    },
    {
      "epoch": 25.85,
      "grad_norm": 1.4140625,
      "learning_rate": 8e-05,
      "loss": 0.2886,
      "step": 5170
    },
    {
      "epoch": 25.9,
      "grad_norm": 13.25,
      "learning_rate": 8e-05,
      "loss": 0.3065,
      "step": 5180
    },
    {
      "epoch": 25.95,
      "grad_norm": 12.6875,
      "learning_rate": 8e-05,
      "loss": 0.2465,
      "step": 5190
    },
    {
      "epoch": 26.0,
      "grad_norm": 1.171875,
      "learning_rate": 8e-05,
      "loss": 0.2765,
      "step": 5200
    },
    {
      "epoch": 26.05,
      "grad_norm": 2.5,
      "learning_rate": 8e-05,
      "loss": 0.2821,
      "step": 5210
    },
    {
      "epoch": 26.1,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.2685,
      "step": 5220
    },
    {
      "epoch": 26.15,
      "grad_norm": 11.6875,
      "learning_rate": 8e-05,
      "loss": 0.2782,
      "step": 5230
    },
    {
      "epoch": 26.2,
      "grad_norm": 1.4453125,
      "learning_rate": 8e-05,
      "loss": 0.2616,
      "step": 5240
    },
    {
      "epoch": 26.25,
      "grad_norm": 13.0,
      "learning_rate": 8e-05,
      "loss": 0.3842,
      "step": 5250
    },
    {
      "epoch": 26.3,
      "grad_norm": 10.9375,
      "learning_rate": 8e-05,
      "loss": 0.2202,
      "step": 5260
    },
    {
      "epoch": 26.35,
      "grad_norm": 0.9609375,
      "learning_rate": 8e-05,
      "loss": 0.1973,
      "step": 5270
    },
    {
      "epoch": 26.4,
      "grad_norm": 13.1875,
      "learning_rate": 8e-05,
      "loss": 0.3689,
      "step": 5280
    },
    {
      "epoch": 26.45,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.2377,
      "step": 5290
    },
    {
      "epoch": 26.5,
      "grad_norm": 1.8828125,
      "learning_rate": 8e-05,
      "loss": 0.2211,
      "step": 5300
    },
    {
      "epoch": 26.55,
      "grad_norm": 1.0078125,
      "learning_rate": 8e-05,
      "loss": 0.3614,
      "step": 5310
    },
    {
      "epoch": 26.6,
      "grad_norm": 1.15625,
      "learning_rate": 8e-05,
      "loss": 0.2357,
      "step": 5320
    },
    {
      "epoch": 26.65,
      "grad_norm": 13.4375,
      "learning_rate": 8e-05,
      "loss": 0.2854,
      "step": 5330
    },
    {
      "epoch": 26.7,
      "grad_norm": 10.0,
      "learning_rate": 8e-05,
      "loss": 0.2299,
      "step": 5340
    },
    {
      "epoch": 26.75,
      "grad_norm": 1.109375,
      "learning_rate": 8e-05,
      "loss": 0.3128,
      "step": 5350
    },
    {
      "epoch": 26.8,
      "grad_norm": 2.4375,
      "learning_rate": 8e-05,
      "loss": 0.2182,
      "step": 5360
    },
    {
      "epoch": 26.85,
      "grad_norm": 1.109375,
      "learning_rate": 8e-05,
      "loss": 0.1037,
      "step": 5370
    },
    {
      "epoch": 26.9,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.2412,
      "step": 5380
    },
    {
      "epoch": 26.95,
      "grad_norm": 3.40625,
      "learning_rate": 8e-05,
      "loss": 0.3913,
      "step": 5390
    },
    {
      "epoch": 27.0,
      "grad_norm": 11.3125,
      "learning_rate": 8e-05,
      "loss": 0.2973,
      "step": 5400
    },
    {
      "epoch": 27.05,
      "grad_norm": 1.8125,
      "learning_rate": 8e-05,
      "loss": 0.2552,
      "step": 5410
    },
    {
      "epoch": 27.1,
      "grad_norm": 12.5,
      "learning_rate": 8e-05,
      "loss": 0.1857,
      "step": 5420
    },
    {
      "epoch": 27.15,
      "grad_norm": 0.72265625,
      "learning_rate": 8e-05,
      "loss": 0.2013,
      "step": 5430
    },
    {
      "epoch": 27.2,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.3292,
      "step": 5440
    },
    {
      "epoch": 27.25,
      "grad_norm": 1.125,
      "learning_rate": 8e-05,
      "loss": 0.3218,
      "step": 5450
    },
    {
      "epoch": 27.3,
      "grad_norm": 1.7578125,
      "learning_rate": 8e-05,
      "loss": 0.2718,
      "step": 5460
    },
    {
      "epoch": 27.35,
      "grad_norm": 12.625,
      "learning_rate": 8e-05,
      "loss": 0.2765,
      "step": 5470
    },
    {
      "epoch": 27.4,
      "grad_norm": 12.25,
      "learning_rate": 8e-05,
      "loss": 0.3151,
      "step": 5480
    },
    {
      "epoch": 27.45,
      "grad_norm": 2.6875,
      "learning_rate": 8e-05,
      "loss": 0.2165,
      "step": 5490
    },
    {
      "epoch": 27.5,
      "grad_norm": 2.0,
      "learning_rate": 8e-05,
      "loss": 0.2343,
      "step": 5500
    },
    {
      "epoch": 27.5,
      "eval_loss": 8.912619590759277,
      "eval_runtime": 7.8779,
      "eval_samples_per_second": 63.469,
      "eval_steps_per_second": 63.469,
      "step": 5500
    },
    {
      "epoch": 27.55,
      "grad_norm": 11.4375,
      "learning_rate": 8e-05,
      "loss": 0.3697,
      "step": 5510
    },
    {
      "epoch": 27.6,
      "grad_norm": 1.578125,
      "learning_rate": 8e-05,
      "loss": 0.1904,
      "step": 5520
    },
    {
      "epoch": 27.65,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.3406,
      "step": 5530
    },
    {
      "epoch": 27.7,
      "grad_norm": 1.7265625,
      "learning_rate": 8e-05,
      "loss": 0.2541,
      "step": 5540
    },
    {
      "epoch": 27.75,
      "grad_norm": 0.9921875,
      "learning_rate": 8e-05,
      "loss": 0.2346,
      "step": 5550
    },
    {
      "epoch": 27.8,
      "grad_norm": 7.90625,
      "learning_rate": 8e-05,
      "loss": 0.2101,
      "step": 5560
    },
    {
      "epoch": 27.85,
      "grad_norm": 12.4375,
      "learning_rate": 8e-05,
      "loss": 0.2812,
      "step": 5570
    },
    {
      "epoch": 27.9,
      "grad_norm": 0.99609375,
      "learning_rate": 8e-05,
      "loss": 0.2815,
      "step": 5580
    },
    {
      "epoch": 27.95,
      "grad_norm": 3.015625,
      "learning_rate": 8e-05,
      "loss": 0.2582,
      "step": 5590
    },
    {
      "epoch": 28.0,
      "grad_norm": 11.6875,
      "learning_rate": 8e-05,
      "loss": 0.4554,
      "step": 5600
    },
    {
      "epoch": 28.05,
      "grad_norm": 10.75,
      "learning_rate": 8e-05,
      "loss": 0.3659,
      "step": 5610
    },
    {
      "epoch": 28.1,
      "grad_norm": 10.625,
      "learning_rate": 8e-05,
      "loss": 0.1781,
      "step": 5620
    },
    {
      "epoch": 28.15,
      "grad_norm": 1.015625,
      "learning_rate": 8e-05,
      "loss": 0.2764,
      "step": 5630
    },
    {
      "epoch": 28.2,
      "grad_norm": 10.6875,
      "learning_rate": 8e-05,
      "loss": 0.3022,
      "step": 5640
    },
    {
      "epoch": 28.25,
      "grad_norm": 13.5,
      "learning_rate": 8e-05,
      "loss": 0.2352,
      "step": 5650
    },
    {
      "epoch": 28.3,
      "grad_norm": 1.046875,
      "learning_rate": 8e-05,
      "loss": 0.2052,
      "step": 5660
    },
    {
      "epoch": 28.35,
      "grad_norm": 2.046875,
      "learning_rate": 8e-05,
      "loss": 0.2885,
      "step": 5670
    },
    {
      "epoch": 28.4,
      "grad_norm": 1.6015625,
      "learning_rate": 8e-05,
      "loss": 0.3102,
      "step": 5680
    },
    {
      "epoch": 28.45,
      "grad_norm": 14.125,
      "learning_rate": 8e-05,
      "loss": 0.2526,
      "step": 5690
    },
    {
      "epoch": 28.5,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.1431,
      "step": 5700
    },
    {
      "epoch": 28.55,
      "grad_norm": 10.9375,
      "learning_rate": 8e-05,
      "loss": 0.2638,
      "step": 5710
    },
    {
      "epoch": 28.6,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.3692,
      "step": 5720
    },
    {
      "epoch": 28.65,
      "grad_norm": 3.625,
      "learning_rate": 8e-05,
      "loss": 0.3725,
      "step": 5730
    },
    {
      "epoch": 28.7,
      "grad_norm": 10.8125,
      "learning_rate": 8e-05,
      "loss": 0.2955,
      "step": 5740
    },
    {
      "epoch": 28.75,
      "grad_norm": 2.421875,
      "learning_rate": 8e-05,
      "loss": 0.1972,
      "step": 5750
    },
    {
      "epoch": 28.8,
      "grad_norm": 3.046875,
      "learning_rate": 8e-05,
      "loss": 0.3902,
      "step": 5760
    },
    {
      "epoch": 28.85,
      "grad_norm": 1.7265625,
      "learning_rate": 8e-05,
      "loss": 0.2459,
      "step": 5770
    },
    {
      "epoch": 28.9,
      "grad_norm": 0.9921875,
      "learning_rate": 8e-05,
      "loss": 0.2117,
      "step": 5780
    },
    {
      "epoch": 28.95,
      "grad_norm": 11.125,
      "learning_rate": 8e-05,
      "loss": 0.3475,
      "step": 5790
    },
    {
      "epoch": 29.0,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.342,
      "step": 5800
    },
    {
      "epoch": 29.05,
      "grad_norm": 1.859375,
      "learning_rate": 8e-05,
      "loss": 0.1948,
      "step": 5810
    },
    {
      "epoch": 29.1,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.3632,
      "step": 5820
    },
    {
      "epoch": 29.15,
      "grad_norm": 12.75,
      "learning_rate": 8e-05,
      "loss": 0.3307,
      "step": 5830
    },
    {
      "epoch": 29.2,
      "grad_norm": 11.375,
      "learning_rate": 8e-05,
      "loss": 0.2275,
      "step": 5840
    },
    {
      "epoch": 29.25,
      "grad_norm": 1.5078125,
      "learning_rate": 8e-05,
      "loss": 0.2147,
      "step": 5850
    },
    {
      "epoch": 29.3,
      "grad_norm": 1.3515625,
      "learning_rate": 8e-05,
      "loss": 0.2111,
      "step": 5860
    },
    {
      "epoch": 29.35,
      "grad_norm": 11.5625,
      "learning_rate": 8e-05,
      "loss": 0.3288,
      "step": 5870
    },
    {
      "epoch": 29.4,
      "grad_norm": 1.640625,
      "learning_rate": 8e-05,
      "loss": 0.1918,
      "step": 5880
    },
    {
      "epoch": 29.45,
      "grad_norm": 14.8125,
      "learning_rate": 8e-05,
      "loss": 0.2553,
      "step": 5890
    },
    {
      "epoch": 29.5,
      "grad_norm": 1.3671875,
      "learning_rate": 8e-05,
      "loss": 0.2064,
      "step": 5900
    },
    {
      "epoch": 29.55,
      "grad_norm": 1.046875,
      "learning_rate": 8e-05,
      "loss": 0.3478,
      "step": 5910
    },
    {
      "epoch": 29.6,
      "grad_norm": 13.4375,
      "learning_rate": 8e-05,
      "loss": 0.2593,
      "step": 5920
    },
    {
      "epoch": 29.65,
      "grad_norm": 0.79296875,
      "learning_rate": 8e-05,
      "loss": 0.3139,
      "step": 5930
    },
    {
      "epoch": 29.7,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.1809,
      "step": 5940
    },
    {
      "epoch": 29.75,
      "grad_norm": 1.8359375,
      "learning_rate": 8e-05,
      "loss": 0.276,
      "step": 5950
    },
    {
      "epoch": 29.8,
      "grad_norm": 12.1875,
      "learning_rate": 8e-05,
      "loss": 0.3907,
      "step": 5960
    },
    {
      "epoch": 29.85,
      "grad_norm": 13.875,
      "learning_rate": 8e-05,
      "loss": 0.3121,
      "step": 5970
    },
    {
      "epoch": 29.9,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.3276,
      "step": 5980
    },
    {
      "epoch": 29.95,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.2805,
      "step": 5990
    },
    {
      "epoch": 30.0,
      "grad_norm": 1.78125,
      "learning_rate": 8e-05,
      "loss": 0.2789,
      "step": 6000
    },
    {
      "epoch": 30.0,
      "eval_loss": 8.706538200378418,
      "eval_runtime": 8.1518,
      "eval_samples_per_second": 61.336,
      "eval_steps_per_second": 61.336,
      "step": 6000
    },
    {
      "epoch": 30.05,
      "grad_norm": 8.25,
      "learning_rate": 8e-05,
      "loss": 0.2017,
      "step": 6010
    },
    {
      "epoch": 30.1,
      "grad_norm": 3.578125,
      "learning_rate": 8e-05,
      "loss": 0.3184,
      "step": 6020
    },
    {
      "epoch": 30.15,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.2877,
      "step": 6030
    },
    {
      "epoch": 30.2,
      "grad_norm": 12.375,
      "learning_rate": 8e-05,
      "loss": 0.2002,
      "step": 6040
    },
    {
      "epoch": 30.25,
      "grad_norm": 11.0625,
      "learning_rate": 8e-05,
      "loss": 0.2738,
      "step": 6050
    },
    {
      "epoch": 30.3,
      "grad_norm": 1.21875,
      "learning_rate": 8e-05,
      "loss": 0.2625,
      "step": 6060
    },
    {
      "epoch": 30.35,
      "grad_norm": 1.078125,
      "learning_rate": 8e-05,
      "loss": 0.2488,
      "step": 6070
    },
    {
      "epoch": 30.4,
      "grad_norm": 13.0625,
      "learning_rate": 8e-05,
      "loss": 0.3019,
      "step": 6080
    },
    {
      "epoch": 30.45,
      "grad_norm": 11.5625,
      "learning_rate": 8e-05,
      "loss": 0.2733,
      "step": 6090
    },
    {
      "epoch": 30.5,
      "grad_norm": 12.9375,
      "learning_rate": 8e-05,
      "loss": 0.2455,
      "step": 6100
    },
    {
      "epoch": 30.55,
      "grad_norm": 11.875,
      "learning_rate": 8e-05,
      "loss": 0.3836,
      "step": 6110
    },
    {
      "epoch": 30.6,
      "grad_norm": 1.875,
      "learning_rate": 8e-05,
      "loss": 0.2377,
      "step": 6120
    },
    {
      "epoch": 30.65,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.315,
      "step": 6130
    },
    {
      "epoch": 30.7,
      "grad_norm": 0.8359375,
      "learning_rate": 8e-05,
      "loss": 0.3438,
      "step": 6140
    },
    {
      "epoch": 30.75,
      "grad_norm": 12.875,
      "learning_rate": 8e-05,
      "loss": 0.4268,
      "step": 6150
    },
    {
      "epoch": 30.8,
      "grad_norm": 13.375,
      "learning_rate": 8e-05,
      "loss": 0.2822,
      "step": 6160
    },
    {
      "epoch": 30.85,
      "grad_norm": 1.03125,
      "learning_rate": 8e-05,
      "loss": 0.2607,
      "step": 6170
    },
    {
      "epoch": 30.9,
      "grad_norm": 11.75,
      "learning_rate": 8e-05,
      "loss": 0.1853,
      "step": 6180
    },
    {
      "epoch": 30.95,
      "grad_norm": 1.7890625,
      "learning_rate": 8e-05,
      "loss": 0.2499,
      "step": 6190
    },
    {
      "epoch": 31.0,
      "grad_norm": 10.1875,
      "learning_rate": 8e-05,
      "loss": 0.2878,
      "step": 6200
    }
  ],
  "logging_steps": 10,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3565855315034112.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}