{
  "best_metric": 0.895774647887324,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-leukemia-08-2024.v1.1\\checkpoint-2187",
  "epoch": 9.984,
  "eval_steps": 500,
  "global_step": 3120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032,
      "grad_norm": 3.326205015182495,
      "learning_rate": 1.6025641025641025e-06,
      "loss": 0.6765,
      "step": 10
    },
    {
      "epoch": 0.064,
      "grad_norm": 4.977212429046631,
      "learning_rate": 3.205128205128205e-06,
      "loss": 0.6592,
      "step": 20
    },
    {
      "epoch": 0.096,
      "grad_norm": 5.245025634765625,
      "learning_rate": 4.807692307692308e-06,
      "loss": 0.6468,
      "step": 30
    },
    {
      "epoch": 0.128,
      "grad_norm": 8.593008041381836,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.6236,
      "step": 40
    },
    {
      "epoch": 0.16,
      "grad_norm": 8.709370613098145,
      "learning_rate": 8.012820512820515e-06,
      "loss": 0.5814,
      "step": 50
    },
    {
      "epoch": 0.192,
      "grad_norm": 11.99754524230957,
      "learning_rate": 9.615384615384616e-06,
      "loss": 0.5527,
      "step": 60
    },
    {
      "epoch": 0.224,
      "grad_norm": 9.359601020812988,
      "learning_rate": 1.1217948717948719e-05,
      "loss": 0.5557,
      "step": 70
    },
    {
      "epoch": 0.256,
      "grad_norm": 8.183499336242676,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.5155,
      "step": 80
    },
    {
      "epoch": 0.288,
      "grad_norm": 8.299748420715332,
      "learning_rate": 1.4423076923076923e-05,
      "loss": 0.5315,
      "step": 90
    },
    {
      "epoch": 0.32,
      "grad_norm": 10.13196086883545,
      "learning_rate": 1.602564102564103e-05,
      "loss": 0.4867,
      "step": 100
    },
    {
      "epoch": 0.352,
      "grad_norm": 8.469902038574219,
      "learning_rate": 1.762820512820513e-05,
      "loss": 0.5046,
      "step": 110
    },
    {
      "epoch": 0.384,
      "grad_norm": 13.50936222076416,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.482,
      "step": 120
    },
    {
      "epoch": 0.416,
      "grad_norm": 12.19388198852539,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.4686,
      "step": 130
    },
    {
      "epoch": 0.448,
      "grad_norm": 12.558857917785645,
      "learning_rate": 2.2435897435897437e-05,
      "loss": 0.4377,
      "step": 140
    },
    {
      "epoch": 0.48,
      "grad_norm": 9.872008323669434,
      "learning_rate": 2.4038461538461542e-05,
      "loss": 0.3905,
      "step": 150
    },
    {
      "epoch": 0.512,
      "grad_norm": 10.574499130249023,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.4353,
      "step": 160
    },
    {
      "epoch": 0.544,
      "grad_norm": 17.03191375732422,
      "learning_rate": 2.724358974358974e-05,
      "loss": 0.4247,
      "step": 170
    },
    {
      "epoch": 0.576,
      "grad_norm": 22.064960479736328,
      "learning_rate": 2.8846153846153845e-05,
      "loss": 0.4109,
      "step": 180
    },
    {
      "epoch": 0.608,
      "grad_norm": 17.807937622070312,
      "learning_rate": 3.0448717948717947e-05,
      "loss": 0.3676,
      "step": 190
    },
    {
      "epoch": 0.64,
      "grad_norm": 65.49658966064453,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.377,
      "step": 200
    },
    {
      "epoch": 0.672,
      "grad_norm": 31.896503448486328,
      "learning_rate": 3.365384615384616e-05,
      "loss": 0.3399,
      "step": 210
    },
    {
      "epoch": 0.704,
      "grad_norm": 60.66265869140625,
      "learning_rate": 3.525641025641026e-05,
      "loss": 0.3622,
      "step": 220
    },
    {
      "epoch": 0.736,
      "grad_norm": 18.89693260192871,
      "learning_rate": 3.685897435897436e-05,
      "loss": 0.3772,
      "step": 230
    },
    {
      "epoch": 0.768,
      "grad_norm": 60.72289276123047,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.3749,
      "step": 240
    },
    {
      "epoch": 0.8,
      "grad_norm": 47.18878936767578,
      "learning_rate": 4.006410256410257e-05,
      "loss": 0.3787,
      "step": 250
    },
    {
      "epoch": 0.832,
      "grad_norm": 43.871910095214844,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.4123,
      "step": 260
    },
    {
      "epoch": 0.864,
      "grad_norm": 41.39242172241211,
      "learning_rate": 4.326923076923077e-05,
      "loss": 0.3703,
      "step": 270
    },
    {
      "epoch": 0.896,
      "grad_norm": 21.079559326171875,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.3573,
      "step": 280
    },
    {
      "epoch": 0.928,
      "grad_norm": 36.1923942565918,
      "learning_rate": 4.6474358974358976e-05,
      "loss": 0.3568,
      "step": 290
    },
    {
      "epoch": 0.96,
      "grad_norm": 23.425731658935547,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 0.334,
      "step": 300
    },
    {
      "epoch": 0.992,
      "grad_norm": 17.070802688598633,
      "learning_rate": 4.9679487179487185e-05,
      "loss": 0.3578,
      "step": 310
    },
    {
      "epoch": 0.9984,
      "eval_accuracy": 0.4262910798122066,
      "eval_loss": 1.7840968370437622,
      "eval_runtime": 15.325,
      "eval_samples_per_second": 69.494,
      "eval_steps_per_second": 2.219,
      "step": 312
    },
    {
      "epoch": 1.024,
      "grad_norm": 30.137889862060547,
      "learning_rate": 4.985754985754986e-05,
      "loss": 0.4256,
      "step": 320
    },
    {
      "epoch": 1.056,
      "grad_norm": 39.79228591918945,
      "learning_rate": 4.9679487179487185e-05,
      "loss": 0.3435,
      "step": 330
    },
    {
      "epoch": 1.088,
      "grad_norm": 13.679237365722656,
      "learning_rate": 4.95014245014245e-05,
      "loss": 0.3055,
      "step": 340
    },
    {
      "epoch": 1.12,
      "grad_norm": 37.67884826660156,
      "learning_rate": 4.932336182336182e-05,
      "loss": 0.3971,
      "step": 350
    },
    {
      "epoch": 1.152,
      "grad_norm": 51.65517044067383,
      "learning_rate": 4.9145299145299147e-05,
      "loss": 0.4286,
      "step": 360
    },
    {
      "epoch": 1.184,
      "grad_norm": 9.33079719543457,
      "learning_rate": 4.896723646723647e-05,
      "loss": 0.3771,
      "step": 370
    },
    {
      "epoch": 1.216,
      "grad_norm": 22.497329711914062,
      "learning_rate": 4.878917378917379e-05,
      "loss": 0.3503,
      "step": 380
    },
    {
      "epoch": 1.248,
      "grad_norm": 28.880863189697266,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 0.289,
      "step": 390
    },
    {
      "epoch": 1.28,
      "grad_norm": 13.440912246704102,
      "learning_rate": 4.8433048433048433e-05,
      "loss": 0.3023,
      "step": 400
    },
    {
      "epoch": 1.312,
      "grad_norm": 25.807588577270508,
      "learning_rate": 4.825498575498576e-05,
      "loss": 0.307,
      "step": 410
    },
    {
      "epoch": 1.3439999999999999,
      "grad_norm": 14.44901180267334,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 0.304,
      "step": 420
    },
    {
      "epoch": 1.376,
      "grad_norm": 27.361024856567383,
      "learning_rate": 4.78988603988604e-05,
      "loss": 0.3366,
      "step": 430
    },
    {
      "epoch": 1.408,
      "grad_norm": 36.755638122558594,
      "learning_rate": 4.772079772079772e-05,
      "loss": 0.3469,
      "step": 440
    },
    {
      "epoch": 1.44,
      "grad_norm": 13.210039138793945,
      "learning_rate": 4.7542735042735045e-05,
      "loss": 0.2735,
      "step": 450
    },
    {
      "epoch": 1.472,
      "grad_norm": 31.067184448242188,
      "learning_rate": 4.736467236467237e-05,
      "loss": 0.2682,
      "step": 460
    },
    {
      "epoch": 1.504,
      "grad_norm": 23.262245178222656,
      "learning_rate": 4.718660968660969e-05,
      "loss": 0.2862,
      "step": 470
    },
    {
      "epoch": 1.536,
      "grad_norm": 16.004291534423828,
      "learning_rate": 4.700854700854701e-05,
      "loss": 0.2777,
      "step": 480
    },
    {
      "epoch": 1.568,
      "grad_norm": 28.26585578918457,
      "learning_rate": 4.683048433048433e-05,
      "loss": 0.2514,
      "step": 490
    },
    {
      "epoch": 1.6,
      "grad_norm": 32.8377685546875,
      "learning_rate": 4.665242165242166e-05,
      "loss": 0.2514,
      "step": 500
    },
    {
      "epoch": 1.6320000000000001,
      "grad_norm": 34.51008605957031,
      "learning_rate": 4.6474358974358976e-05,
      "loss": 0.3642,
      "step": 510
    },
    {
      "epoch": 1.6640000000000001,
      "grad_norm": 20.71710968017578,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.3203,
      "step": 520
    },
    {
      "epoch": 1.696,
      "grad_norm": 11.27377986907959,
      "learning_rate": 4.611823361823362e-05,
      "loss": 0.329,
      "step": 530
    },
    {
      "epoch": 1.728,
      "grad_norm": 7.797733306884766,
      "learning_rate": 4.594017094017094e-05,
      "loss": 0.3049,
      "step": 540
    },
    {
      "epoch": 1.76,
      "grad_norm": 13.634445190429688,
      "learning_rate": 4.576210826210827e-05,
      "loss": 0.2444,
      "step": 550
    },
    {
      "epoch": 1.792,
      "grad_norm": 10.297043800354004,
      "learning_rate": 4.558404558404559e-05,
      "loss": 0.2404,
      "step": 560
    },
    {
      "epoch": 1.8239999999999998,
      "grad_norm": 10.661726951599121,
      "learning_rate": 4.5405982905982906e-05,
      "loss": 0.2777,
      "step": 570
    },
    {
      "epoch": 1.8559999999999999,
      "grad_norm": 23.705551147460938,
      "learning_rate": 4.522792022792023e-05,
      "loss": 0.3309,
      "step": 580
    },
    {
      "epoch": 1.888,
      "grad_norm": 21.21759033203125,
      "learning_rate": 4.504985754985755e-05,
      "loss": 0.2703,
      "step": 590
    },
    {
      "epoch": 1.92,
      "grad_norm": 39.91014862060547,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.2853,
      "step": 600
    },
    {
      "epoch": 1.952,
      "grad_norm": 53.51070022583008,
      "learning_rate": 4.46937321937322e-05,
      "loss": 0.258,
      "step": 610
    },
    {
      "epoch": 1.984,
      "grad_norm": 13.379941940307617,
      "learning_rate": 4.451566951566952e-05,
      "loss": 0.2403,
      "step": 620
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6807511737089202,
      "eval_loss": 0.9413667917251587,
      "eval_runtime": 15.1245,
      "eval_samples_per_second": 70.415,
      "eval_steps_per_second": 2.248,
      "step": 625
    },
    {
      "epoch": 2.016,
      "grad_norm": 9.985576629638672,
      "learning_rate": 4.4337606837606836e-05,
      "loss": 0.2208,
      "step": 630
    },
    {
      "epoch": 2.048,
      "grad_norm": 28.021074295043945,
      "learning_rate": 4.415954415954416e-05,
      "loss": 0.2653,
      "step": 640
    },
    {
      "epoch": 2.08,
      "grad_norm": 14.402697563171387,
      "learning_rate": 4.3981481481481486e-05,
      "loss": 0.2528,
      "step": 650
    },
    {
      "epoch": 2.112,
      "grad_norm": 20.3269100189209,
      "learning_rate": 4.3803418803418805e-05,
      "loss": 0.2513,
      "step": 660
    },
    {
      "epoch": 2.144,
      "grad_norm": 14.583213806152344,
      "learning_rate": 4.362535612535612e-05,
      "loss": 0.2224,
      "step": 670
    },
    {
      "epoch": 2.176,
      "grad_norm": 11.780790328979492,
      "learning_rate": 4.344729344729345e-05,
      "loss": 0.2075,
      "step": 680
    },
    {
      "epoch": 2.208,
      "grad_norm": 39.822628021240234,
      "learning_rate": 4.326923076923077e-05,
      "loss": 0.2575,
      "step": 690
    },
    {
      "epoch": 2.24,
      "grad_norm": 17.438533782958984,
      "learning_rate": 4.309116809116809e-05,
      "loss": 0.2109,
      "step": 700
    },
    {
      "epoch": 2.2720000000000002,
      "grad_norm": 21.648460388183594,
      "learning_rate": 4.291310541310542e-05,
      "loss": 0.2231,
      "step": 710
    },
    {
      "epoch": 2.304,
      "grad_norm": 15.304508209228516,
      "learning_rate": 4.2735042735042735e-05,
      "loss": 0.1973,
      "step": 720
    },
    {
      "epoch": 2.336,
      "grad_norm": 27.324260711669922,
      "learning_rate": 4.255698005698006e-05,
      "loss": 0.2586,
      "step": 730
    },
    {
      "epoch": 2.368,
      "grad_norm": 15.033935546875,
      "learning_rate": 4.2378917378917385e-05,
      "loss": 0.2142,
      "step": 740
    },
    {
      "epoch": 2.4,
      "grad_norm": 14.733280181884766,
      "learning_rate": 4.2200854700854704e-05,
      "loss": 0.1725,
      "step": 750
    },
    {
      "epoch": 2.432,
      "grad_norm": 12.364413261413574,
      "learning_rate": 4.202279202279202e-05,
      "loss": 0.2412,
      "step": 760
    },
    {
      "epoch": 2.464,
      "grad_norm": 8.037039756774902,
      "learning_rate": 4.184472934472935e-05,
      "loss": 0.2293,
      "step": 770
    },
    {
      "epoch": 2.496,
      "grad_norm": 48.51978302001953,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.2513,
      "step": 780
    },
    {
      "epoch": 2.528,
      "grad_norm": 18.824827194213867,
      "learning_rate": 4.148860398860399e-05,
      "loss": 0.2248,
      "step": 790
    },
    {
      "epoch": 2.56,
      "grad_norm": 29.346654891967773,
      "learning_rate": 4.131054131054131e-05,
      "loss": 0.215,
      "step": 800
    },
    {
      "epoch": 2.592,
      "grad_norm": 13.431859016418457,
      "learning_rate": 4.1132478632478634e-05,
      "loss": 0.2015,
      "step": 810
    },
    {
      "epoch": 2.624,
      "grad_norm": 45.57223892211914,
      "learning_rate": 4.095441595441596e-05,
      "loss": 0.1951,
      "step": 820
    },
    {
      "epoch": 2.656,
      "grad_norm": 44.39851379394531,
      "learning_rate": 4.077635327635328e-05,
      "loss": 0.1937,
      "step": 830
    },
    {
      "epoch": 2.6879999999999997,
      "grad_norm": 29.183330535888672,
      "learning_rate": 4.05982905982906e-05,
      "loss": 0.2363,
      "step": 840
    },
    {
      "epoch": 2.7199999999999998,
      "grad_norm": 14.148327827453613,
      "learning_rate": 4.042022792022792e-05,
      "loss": 0.2374,
      "step": 850
    },
    {
      "epoch": 2.752,
      "grad_norm": 31.385358810424805,
      "learning_rate": 4.024216524216524e-05,
      "loss": 0.1903,
      "step": 860
    },
    {
      "epoch": 2.784,
      "grad_norm": 23.014760971069336,
      "learning_rate": 4.006410256410257e-05,
      "loss": 0.1592,
      "step": 870
    },
    {
      "epoch": 2.816,
      "grad_norm": 37.545082092285156,
      "learning_rate": 3.988603988603989e-05,
      "loss": 0.2185,
      "step": 880
    },
    {
      "epoch": 2.848,
      "grad_norm": 43.25140380859375,
      "learning_rate": 3.970797720797721e-05,
      "loss": 0.2185,
      "step": 890
    },
    {
      "epoch": 2.88,
      "grad_norm": 18.27262306213379,
      "learning_rate": 3.952991452991453e-05,
      "loss": 0.1996,
      "step": 900
    },
    {
      "epoch": 2.912,
      "grad_norm": 24.33517837524414,
      "learning_rate": 3.935185185185186e-05,
      "loss": 0.1716,
      "step": 910
    },
    {
      "epoch": 2.944,
      "grad_norm": 9.979388236999512,
      "learning_rate": 3.9173789173789176e-05,
      "loss": 0.2035,
      "step": 920
    },
    {
      "epoch": 2.976,
      "grad_norm": 27.87533187866211,
      "learning_rate": 3.89957264957265e-05,
      "loss": 0.1815,
      "step": 930
    },
    {
      "epoch": 2.9984,
      "eval_accuracy": 0.7784037558685446,
      "eval_loss": 0.6043610572814941,
      "eval_runtime": 15.1388,
      "eval_samples_per_second": 70.349,
      "eval_steps_per_second": 2.246,
      "step": 937
    },
    {
      "epoch": 3.008,
      "grad_norm": 13.326042175292969,
      "learning_rate": 3.881766381766382e-05,
      "loss": 0.1679,
      "step": 940
    },
    {
      "epoch": 3.04,
      "grad_norm": 33.75840759277344,
      "learning_rate": 3.863960113960114e-05,
      "loss": 0.2098,
      "step": 950
    },
    {
      "epoch": 3.072,
      "grad_norm": 10.451057434082031,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.1596,
      "step": 960
    },
    {
      "epoch": 3.104,
      "grad_norm": 21.400108337402344,
      "learning_rate": 3.828347578347579e-05,
      "loss": 0.1778,
      "step": 970
    },
    {
      "epoch": 3.136,
      "grad_norm": 19.759441375732422,
      "learning_rate": 3.8105413105413106e-05,
      "loss": 0.1957,
      "step": 980
    },
    {
      "epoch": 3.168,
      "grad_norm": 19.105667114257812,
      "learning_rate": 3.7927350427350425e-05,
      "loss": 0.1649,
      "step": 990
    },
    {
      "epoch": 3.2,
      "grad_norm": 25.42116355895996,
      "learning_rate": 3.774928774928775e-05,
      "loss": 0.148,
      "step": 1000
    },
    {
      "epoch": 3.232,
      "grad_norm": 13.591997146606445,
      "learning_rate": 3.7571225071225075e-05,
      "loss": 0.1777,
      "step": 1010
    },
    {
      "epoch": 3.2640000000000002,
      "grad_norm": 47.809547424316406,
      "learning_rate": 3.739316239316239e-05,
      "loss": 0.1719,
      "step": 1020
    },
    {
      "epoch": 3.296,
      "grad_norm": 22.09874725341797,
      "learning_rate": 3.721509971509972e-05,
      "loss": 0.1795,
      "step": 1030
    },
    {
      "epoch": 3.328,
      "grad_norm": 16.385005950927734,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.1418,
      "step": 1040
    },
    {
      "epoch": 3.36,
      "grad_norm": 40.277862548828125,
      "learning_rate": 3.685897435897436e-05,
      "loss": 0.1666,
      "step": 1050
    },
    {
      "epoch": 3.392,
      "grad_norm": 19.256372451782227,
      "learning_rate": 3.668091168091169e-05,
      "loss": 0.1594,
      "step": 1060
    },
    {
      "epoch": 3.424,
      "grad_norm": 42.8471565246582,
      "learning_rate": 3.6502849002849005e-05,
      "loss": 0.2179,
      "step": 1070
    },
    {
      "epoch": 3.456,
      "grad_norm": 14.012408256530762,
      "learning_rate": 3.6324786324786323e-05,
      "loss": 0.1801,
      "step": 1080
    },
    {
      "epoch": 3.488,
      "grad_norm": 32.536285400390625,
      "learning_rate": 3.614672364672365e-05,
      "loss": 0.1385,
      "step": 1090
    },
    {
      "epoch": 3.52,
      "grad_norm": 10.804754257202148,
      "learning_rate": 3.5968660968660974e-05,
      "loss": 0.1707,
      "step": 1100
    },
    {
      "epoch": 3.552,
      "grad_norm": 14.308632850646973,
      "learning_rate": 3.579059829059829e-05,
      "loss": 0.1928,
      "step": 1110
    },
    {
      "epoch": 3.584,
      "grad_norm": 34.59251403808594,
      "learning_rate": 3.561253561253561e-05,
      "loss": 0.1592,
      "step": 1120
    },
    {
      "epoch": 3.616,
      "grad_norm": 32.08983612060547,
      "learning_rate": 3.5434472934472935e-05,
      "loss": 0.1473,
      "step": 1130
    },
    {
      "epoch": 3.648,
      "grad_norm": 23.76660919189453,
      "learning_rate": 3.525641025641026e-05,
      "loss": 0.1567,
      "step": 1140
    },
    {
      "epoch": 3.68,
      "grad_norm": 14.27629566192627,
      "learning_rate": 3.507834757834758e-05,
      "loss": 0.1491,
      "step": 1150
    },
    {
      "epoch": 3.7119999999999997,
      "grad_norm": 13.110605239868164,
      "learning_rate": 3.4900284900284904e-05,
      "loss": 0.2185,
      "step": 1160
    },
    {
      "epoch": 3.7439999999999998,
      "grad_norm": 13.250021934509277,
      "learning_rate": 3.472222222222222e-05,
      "loss": 0.2039,
      "step": 1170
    },
    {
      "epoch": 3.776,
      "grad_norm": 32.865421295166016,
      "learning_rate": 3.454415954415954e-05,
      "loss": 0.1246,
      "step": 1180
    },
    {
      "epoch": 3.808,
      "grad_norm": 17.54583740234375,
      "learning_rate": 3.436609686609687e-05,
      "loss": 0.1992,
      "step": 1190
    },
    {
      "epoch": 3.84,
      "grad_norm": 36.1751594543457,
      "learning_rate": 3.418803418803419e-05,
      "loss": 0.1216,
      "step": 1200
    },
    {
      "epoch": 3.872,
      "grad_norm": 26.110553741455078,
      "learning_rate": 3.400997150997151e-05,
      "loss": 0.1545,
      "step": 1210
    },
    {
      "epoch": 3.904,
      "grad_norm": 30.693532943725586,
      "learning_rate": 3.3831908831908834e-05,
      "loss": 0.1758,
      "step": 1220
    },
    {
      "epoch": 3.936,
      "grad_norm": 11.511937141418457,
      "learning_rate": 3.365384615384616e-05,
      "loss": 0.1584,
      "step": 1230
    },
    {
      "epoch": 3.968,
      "grad_norm": 53.82408905029297,
      "learning_rate": 3.347578347578348e-05,
      "loss": 0.1264,
      "step": 1240
    },
    {
      "epoch": 4.0,
      "grad_norm": 20.06711769104004,
      "learning_rate": 3.32977207977208e-05,
      "loss": 0.2062,
      "step": 1250
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7643192488262911,
      "eval_loss": 0.5284382700920105,
      "eval_runtime": 16.1882,
      "eval_samples_per_second": 65.789,
      "eval_steps_per_second": 2.1,
      "step": 1250
    },
    {
      "epoch": 4.032,
      "grad_norm": 11.678908348083496,
      "learning_rate": 3.311965811965812e-05,
      "loss": 0.1359,
      "step": 1260
    },
    {
      "epoch": 4.064,
      "grad_norm": 22.579870223999023,
      "learning_rate": 3.294159544159544e-05,
      "loss": 0.1529,
      "step": 1270
    },
    {
      "epoch": 4.096,
      "grad_norm": 30.36139678955078,
      "learning_rate": 3.2763532763532764e-05,
      "loss": 0.1259,
      "step": 1280
    },
    {
      "epoch": 4.128,
      "grad_norm": 35.74097442626953,
      "learning_rate": 3.258547008547009e-05,
      "loss": 0.1506,
      "step": 1290
    },
    {
      "epoch": 4.16,
      "grad_norm": 21.0123233795166,
      "learning_rate": 3.240740740740741e-05,
      "loss": 0.1513,
      "step": 1300
    },
    {
      "epoch": 4.192,
      "grad_norm": 29.89231300354004,
      "learning_rate": 3.2229344729344726e-05,
      "loss": 0.1383,
      "step": 1310
    },
    {
      "epoch": 4.224,
      "grad_norm": 9.854804039001465,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.1327,
      "step": 1320
    },
    {
      "epoch": 4.256,
      "grad_norm": 11.543468475341797,
      "learning_rate": 3.1873219373219376e-05,
      "loss": 0.1208,
      "step": 1330
    },
    {
      "epoch": 4.288,
      "grad_norm": 21.905872344970703,
      "learning_rate": 3.1695156695156695e-05,
      "loss": 0.1181,
      "step": 1340
    },
    {
      "epoch": 4.32,
      "grad_norm": 8.628751754760742,
      "learning_rate": 3.151709401709402e-05,
      "loss": 0.1449,
      "step": 1350
    },
    {
      "epoch": 4.352,
      "grad_norm": 17.885547637939453,
      "learning_rate": 3.133903133903134e-05,
      "loss": 0.1107,
      "step": 1360
    },
    {
      "epoch": 4.384,
      "grad_norm": 34.77399444580078,
      "learning_rate": 3.116096866096866e-05,
      "loss": 0.1277,
      "step": 1370
    },
    {
      "epoch": 4.416,
      "grad_norm": 21.491153717041016,
      "learning_rate": 3.098290598290599e-05,
      "loss": 0.1681,
      "step": 1380
    },
    {
      "epoch": 4.448,
      "grad_norm": 9.110980987548828,
      "learning_rate": 3.080484330484331e-05,
      "loss": 0.1071,
      "step": 1390
    },
    {
      "epoch": 4.48,
      "grad_norm": 12.070108413696289,
      "learning_rate": 3.0626780626780625e-05,
      "loss": 0.1285,
      "step": 1400
    },
    {
      "epoch": 4.5120000000000005,
      "grad_norm": 37.39555740356445,
      "learning_rate": 3.0448717948717947e-05,
      "loss": 0.1158,
      "step": 1410
    },
    {
      "epoch": 4.5440000000000005,
      "grad_norm": 54.185611724853516,
      "learning_rate": 3.0270655270655275e-05,
      "loss": 0.1301,
      "step": 1420
    },
    {
      "epoch": 4.576,
      "grad_norm": 31.016170501708984,
      "learning_rate": 3.0092592592592593e-05,
      "loss": 0.1008,
      "step": 1430
    },
    {
      "epoch": 4.608,
      "grad_norm": 7.421366214752197,
      "learning_rate": 2.9914529914529915e-05,
      "loss": 0.0934,
      "step": 1440
    },
    {
      "epoch": 4.64,
      "grad_norm": 10.641942024230957,
      "learning_rate": 2.9736467236467237e-05,
      "loss": 0.1028,
      "step": 1450
    },
    {
      "epoch": 4.672,
      "grad_norm": 27.484214782714844,
      "learning_rate": 2.9558404558404562e-05,
      "loss": 0.1084,
      "step": 1460
    },
    {
      "epoch": 4.704,
      "grad_norm": 10.595396041870117,
      "learning_rate": 2.9380341880341884e-05,
      "loss": 0.106,
      "step": 1470
    },
    {
      "epoch": 4.736,
      "grad_norm": 18.006689071655273,
      "learning_rate": 2.9202279202279202e-05,
      "loss": 0.1322,
      "step": 1480
    },
    {
      "epoch": 4.768,
      "grad_norm": 58.16054153442383,
      "learning_rate": 2.9024216524216524e-05,
      "loss": 0.1374,
      "step": 1490
    },
    {
      "epoch": 4.8,
      "grad_norm": 14.557624816894531,
      "learning_rate": 2.8846153846153845e-05,
      "loss": 0.1255,
      "step": 1500
    },
    {
      "epoch": 4.832,
      "grad_norm": 43.29716491699219,
      "learning_rate": 2.866809116809117e-05,
      "loss": 0.1147,
      "step": 1510
    },
    {
      "epoch": 4.864,
      "grad_norm": 18.05630874633789,
      "learning_rate": 2.8490028490028492e-05,
      "loss": 0.1264,
      "step": 1520
    },
    {
      "epoch": 4.896,
      "grad_norm": 15.653997421264648,
      "learning_rate": 2.8311965811965814e-05,
      "loss": 0.1058,
      "step": 1530
    },
    {
      "epoch": 4.928,
      "grad_norm": 20.052400588989258,
      "learning_rate": 2.8133903133903132e-05,
      "loss": 0.088,
      "step": 1540
    },
    {
      "epoch": 4.96,
      "grad_norm": 28.409940719604492,
      "learning_rate": 2.795584045584046e-05,
      "loss": 0.1111,
      "step": 1550
    },
    {
      "epoch": 4.992,
      "grad_norm": 15.17120361328125,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1212,
      "step": 1560
    },
    {
      "epoch": 4.9984,
      "eval_accuracy": 0.8488262910798122,
      "eval_loss": 0.4432164132595062,
      "eval_runtime": 16.1884,
      "eval_samples_per_second": 65.788,
      "eval_steps_per_second": 2.1,
      "step": 1562
    },
    {
      "epoch": 5.024,
      "grad_norm": 28.41824722290039,
      "learning_rate": 2.75997150997151e-05,
      "loss": 0.123,
      "step": 1570
    },
    {
      "epoch": 5.056,
      "grad_norm": 69.49197387695312,
      "learning_rate": 2.7421652421652423e-05,
      "loss": 0.0979,
      "step": 1580
    },
    {
      "epoch": 5.088,
      "grad_norm": 18.26475715637207,
      "learning_rate": 2.724358974358974e-05,
      "loss": 0.1562,
      "step": 1590
    },
    {
      "epoch": 5.12,
      "grad_norm": 18.032114028930664,
      "learning_rate": 2.706552706552707e-05,
      "loss": 0.105,
      "step": 1600
    },
    {
      "epoch": 5.152,
      "grad_norm": 6.473869323730469,
      "learning_rate": 2.688746438746439e-05,
      "loss": 0.1033,
      "step": 1610
    },
    {
      "epoch": 5.184,
      "grad_norm": 63.54292678833008,
      "learning_rate": 2.670940170940171e-05,
      "loss": 0.094,
      "step": 1620
    },
    {
      "epoch": 5.216,
      "grad_norm": 5.499744892120361,
      "learning_rate": 2.653133903133903e-05,
      "loss": 0.1023,
      "step": 1630
    },
    {
      "epoch": 5.248,
      "grad_norm": 15.685684204101562,
      "learning_rate": 2.6353276353276356e-05,
      "loss": 0.0925,
      "step": 1640
    },
    {
      "epoch": 5.28,
      "grad_norm": 26.418638229370117,
      "learning_rate": 2.6175213675213678e-05,
      "loss": 0.0834,
      "step": 1650
    },
    {
      "epoch": 5.312,
      "grad_norm": 27.78782844543457,
      "learning_rate": 2.5997150997151e-05,
      "loss": 0.1303,
      "step": 1660
    },
    {
      "epoch": 5.344,
      "grad_norm": 19.61688995361328,
      "learning_rate": 2.5819088319088318e-05,
      "loss": 0.1024,
      "step": 1670
    },
    {
      "epoch": 5.376,
      "grad_norm": 9.222888946533203,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.0927,
      "step": 1680
    },
    {
      "epoch": 5.408,
      "grad_norm": 22.30426788330078,
      "learning_rate": 2.5462962962962965e-05,
      "loss": 0.0801,
      "step": 1690
    },
    {
      "epoch": 5.44,
      "grad_norm": 16.88903045654297,
      "learning_rate": 2.5284900284900286e-05,
      "loss": 0.088,
      "step": 1700
    },
    {
      "epoch": 5.4719999999999995,
      "grad_norm": 29.053525924682617,
      "learning_rate": 2.5106837606837608e-05,
      "loss": 0.0836,
      "step": 1710
    },
    {
      "epoch": 5.504,
      "grad_norm": 8.875022888183594,
      "learning_rate": 2.492877492877493e-05,
      "loss": 0.0853,
      "step": 1720
    },
    {
      "epoch": 5.536,
      "grad_norm": 22.24742317199707,
      "learning_rate": 2.475071225071225e-05,
      "loss": 0.1093,
      "step": 1730
    },
    {
      "epoch": 5.568,
      "grad_norm": 22.047805786132812,
      "learning_rate": 2.4572649572649573e-05,
      "loss": 0.0921,
      "step": 1740
    },
    {
      "epoch": 5.6,
      "grad_norm": 11.919727325439453,
      "learning_rate": 2.4394586894586895e-05,
      "loss": 0.0897,
      "step": 1750
    },
    {
      "epoch": 5.632,
      "grad_norm": 20.128320693969727,
      "learning_rate": 2.4216524216524217e-05,
      "loss": 0.1007,
      "step": 1760
    },
    {
      "epoch": 5.664,
      "grad_norm": 29.007001876831055,
      "learning_rate": 2.4038461538461542e-05,
      "loss": 0.0866,
      "step": 1770
    },
    {
      "epoch": 5.696,
      "grad_norm": 16.665626525878906,
      "learning_rate": 2.386039886039886e-05,
      "loss": 0.0931,
      "step": 1780
    },
    {
      "epoch": 5.728,
      "grad_norm": 6.438042163848877,
      "learning_rate": 2.3682336182336185e-05,
      "loss": 0.0905,
      "step": 1790
    },
    {
      "epoch": 5.76,
      "grad_norm": 26.61479949951172,
      "learning_rate": 2.3504273504273504e-05,
      "loss": 0.1292,
      "step": 1800
    },
    {
      "epoch": 5.792,
      "grad_norm": 22.947494506835938,
      "learning_rate": 2.332621082621083e-05,
      "loss": 0.0807,
      "step": 1810
    },
    {
      "epoch": 5.824,
      "grad_norm": 16.805652618408203,
      "learning_rate": 2.314814814814815e-05,
      "loss": 0.0634,
      "step": 1820
    },
    {
      "epoch": 5.856,
      "grad_norm": 31.196184158325195,
      "learning_rate": 2.297008547008547e-05,
      "loss": 0.0872,
      "step": 1830
    },
    {
      "epoch": 5.888,
      "grad_norm": 10.884623527526855,
      "learning_rate": 2.2792022792022794e-05,
      "loss": 0.0958,
      "step": 1840
    },
    {
      "epoch": 5.92,
      "grad_norm": 44.18009567260742,
      "learning_rate": 2.2613960113960116e-05,
      "loss": 0.1056,
      "step": 1850
    },
    {
      "epoch": 5.952,
      "grad_norm": 14.128731727600098,
      "learning_rate": 2.2435897435897437e-05,
      "loss": 0.0946,
      "step": 1860
    },
    {
      "epoch": 5.984,
      "grad_norm": 12.602348327636719,
      "learning_rate": 2.225783475783476e-05,
      "loss": 0.0723,
      "step": 1870
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7924882629107981,
      "eval_loss": 0.8276024460792542,
      "eval_runtime": 16.2265,
      "eval_samples_per_second": 65.633,
      "eval_steps_per_second": 2.095,
      "step": 1875
    },
    {
      "epoch": 6.016,
      "grad_norm": 18.401493072509766,
      "learning_rate": 2.207977207977208e-05,
      "loss": 0.0852,
      "step": 1880
    },
    {
      "epoch": 6.048,
      "grad_norm": 7.691855430603027,
      "learning_rate": 2.1901709401709402e-05,
      "loss": 0.0585,
      "step": 1890
    },
    {
      "epoch": 6.08,
      "grad_norm": 20.08210563659668,
      "learning_rate": 2.1723646723646724e-05,
      "loss": 0.07,
      "step": 1900
    },
    {
      "epoch": 6.112,
      "grad_norm": 12.808494567871094,
      "learning_rate": 2.1545584045584046e-05,
      "loss": 0.0951,
      "step": 1910
    },
    {
      "epoch": 6.144,
      "grad_norm": 15.134050369262695,
      "learning_rate": 2.1367521367521368e-05,
      "loss": 0.0642,
      "step": 1920
    },
    {
      "epoch": 6.176,
      "grad_norm": 10.28510570526123,
      "learning_rate": 2.1189458689458693e-05,
      "loss": 0.0808,
      "step": 1930
    },
    {
      "epoch": 6.208,
      "grad_norm": 14.190322875976562,
      "learning_rate": 2.101139601139601e-05,
      "loss": 0.0702,
      "step": 1940
    },
    {
      "epoch": 6.24,
      "grad_norm": 21.010963439941406,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.079,
      "step": 1950
    },
    {
      "epoch": 6.272,
      "grad_norm": 18.102190017700195,
      "learning_rate": 2.0655270655270654e-05,
      "loss": 0.0977,
      "step": 1960
    },
    {
      "epoch": 6.304,
      "grad_norm": 14.988326072692871,
      "learning_rate": 2.047720797720798e-05,
      "loss": 0.0831,
      "step": 1970
    },
    {
      "epoch": 6.336,
      "grad_norm": 7.804359436035156,
      "learning_rate": 2.02991452991453e-05,
      "loss": 0.07,
      "step": 1980
    },
    {
      "epoch": 6.368,
      "grad_norm": 7.981973648071289,
      "learning_rate": 2.012108262108262e-05,
      "loss": 0.0953,
      "step": 1990
    },
    {
      "epoch": 6.4,
      "grad_norm": 16.495155334472656,
      "learning_rate": 1.9943019943019945e-05,
      "loss": 0.0714,
      "step": 2000
    },
    {
      "epoch": 6.432,
      "grad_norm": 20.57613182067871,
      "learning_rate": 1.9764957264957266e-05,
      "loss": 0.0729,
      "step": 2010
    },
    {
      "epoch": 6.464,
      "grad_norm": 9.313785552978516,
      "learning_rate": 1.9586894586894588e-05,
      "loss": 0.0715,
      "step": 2020
    },
    {
      "epoch": 6.496,
      "grad_norm": 23.951351165771484,
      "learning_rate": 1.940883190883191e-05,
      "loss": 0.0871,
      "step": 2030
    },
    {
      "epoch": 6.5280000000000005,
      "grad_norm": 18.672794342041016,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.0631,
      "step": 2040
    },
    {
      "epoch": 6.5600000000000005,
      "grad_norm": 15.373053550720215,
      "learning_rate": 1.9052706552706553e-05,
      "loss": 0.0767,
      "step": 2050
    },
    {
      "epoch": 6.592,
      "grad_norm": 13.307682991027832,
      "learning_rate": 1.8874643874643875e-05,
      "loss": 0.0803,
      "step": 2060
    },
    {
      "epoch": 6.624,
      "grad_norm": 15.382416725158691,
      "learning_rate": 1.8696581196581197e-05,
      "loss": 0.058,
      "step": 2070
    },
    {
      "epoch": 6.656,
      "grad_norm": 25.57972526550293,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.0866,
      "step": 2080
    },
    {
      "epoch": 6.688,
      "grad_norm": 11.426902770996094,
      "learning_rate": 1.8340455840455843e-05,
      "loss": 0.0435,
      "step": 2090
    },
    {
      "epoch": 6.72,
      "grad_norm": 10.57555866241455,
      "learning_rate": 1.8162393162393162e-05,
      "loss": 0.0532,
      "step": 2100
    },
    {
      "epoch": 6.752,
      "grad_norm": 14.774798393249512,
      "learning_rate": 1.7984330484330487e-05,
      "loss": 0.0848,
      "step": 2110
    },
    {
      "epoch": 6.784,
      "grad_norm": 22.849029541015625,
      "learning_rate": 1.7806267806267805e-05,
      "loss": 0.0698,
      "step": 2120
    },
    {
      "epoch": 6.816,
      "grad_norm": 8.238070487976074,
      "learning_rate": 1.762820512820513e-05,
      "loss": 0.074,
      "step": 2130
    },
    {
      "epoch": 6.848,
      "grad_norm": 8.715176582336426,
      "learning_rate": 1.7450142450142452e-05,
      "loss": 0.081,
      "step": 2140
    },
    {
      "epoch": 6.88,
      "grad_norm": 7.292199611663818,
      "learning_rate": 1.727207977207977e-05,
      "loss": 0.0728,
      "step": 2150
    },
    {
      "epoch": 6.912,
      "grad_norm": 10.9399995803833,
      "learning_rate": 1.7094017094017095e-05,
      "loss": 0.0742,
      "step": 2160
    },
    {
      "epoch": 6.944,
      "grad_norm": 20.793378829956055,
      "learning_rate": 1.6915954415954417e-05,
      "loss": 0.0662,
      "step": 2170
    },
    {
      "epoch": 6.976,
      "grad_norm": 11.476004600524902,
      "learning_rate": 1.673789173789174e-05,
      "loss": 0.0656,
      "step": 2180
    },
    {
      "epoch": 6.9984,
      "eval_accuracy": 0.895774647887324,
      "eval_loss": 0.3423428237438202,
      "eval_runtime": 16.2162,
      "eval_samples_per_second": 65.675,
      "eval_steps_per_second": 2.097,
      "step": 2187
    },
    {
      "epoch": 7.008,
      "grad_norm": 8.697748184204102,
      "learning_rate": 1.655982905982906e-05,
      "loss": 0.0538,
      "step": 2190
    },
    {
      "epoch": 7.04,
      "grad_norm": 11.475558280944824,
      "learning_rate": 1.6381766381766382e-05,
      "loss": 0.0595,
      "step": 2200
    },
    {
      "epoch": 7.072,
      "grad_norm": 18.916088104248047,
      "learning_rate": 1.6203703703703704e-05,
      "loss": 0.054,
      "step": 2210
    },
    {
      "epoch": 7.104,
      "grad_norm": 16.337879180908203,
      "learning_rate": 1.602564102564103e-05,
      "loss": 0.0588,
      "step": 2220
    },
    {
      "epoch": 7.136,
      "grad_norm": 50.64964294433594,
      "learning_rate": 1.5847578347578347e-05,
      "loss": 0.078,
      "step": 2230
    },
    {
      "epoch": 7.168,
      "grad_norm": 12.643454551696777,
      "learning_rate": 1.566951566951567e-05,
      "loss": 0.061,
      "step": 2240
    },
    {
      "epoch": 7.2,
      "grad_norm": 9.207147598266602,
      "learning_rate": 1.5491452991452994e-05,
      "loss": 0.0858,
      "step": 2250
    },
    {
      "epoch": 7.232,
      "grad_norm": 13.683488845825195,
      "learning_rate": 1.5313390313390312e-05,
      "loss": 0.0641,
      "step": 2260
    },
    {
      "epoch": 7.264,
      "grad_norm": 15.510147094726562,
      "learning_rate": 1.5135327635327638e-05,
      "loss": 0.0609,
      "step": 2270
    },
    {
      "epoch": 7.296,
      "grad_norm": 11.573060989379883,
      "learning_rate": 1.4957264957264958e-05,
      "loss": 0.0537,
      "step": 2280
    },
    {
      "epoch": 7.328,
      "grad_norm": 14.698716163635254,
      "learning_rate": 1.4779202279202281e-05,
      "loss": 0.0443,
      "step": 2290
    },
    {
      "epoch": 7.36,
      "grad_norm": 48.410911560058594,
      "learning_rate": 1.4601139601139601e-05,
      "loss": 0.0667,
      "step": 2300
    },
    {
      "epoch": 7.392,
      "grad_norm": 19.12745475769043,
      "learning_rate": 1.4423076923076923e-05,
      "loss": 0.0833,
      "step": 2310
    },
    {
      "epoch": 7.424,
      "grad_norm": 6.480365753173828,
      "learning_rate": 1.4245014245014246e-05,
      "loss": 0.057,
      "step": 2320
    },
    {
      "epoch": 7.456,
      "grad_norm": 18.375783920288086,
      "learning_rate": 1.4066951566951566e-05,
      "loss": 0.066,
      "step": 2330
    },
    {
      "epoch": 7.4879999999999995,
      "grad_norm": 19.94139289855957,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.057,
      "step": 2340
    },
    {
      "epoch": 7.52,
      "grad_norm": 4.120523929595947,
      "learning_rate": 1.3710826210826211e-05,
      "loss": 0.0513,
      "step": 2350
    },
    {
      "epoch": 7.552,
      "grad_norm": 12.98390007019043,
      "learning_rate": 1.3532763532763535e-05,
      "loss": 0.0409,
      "step": 2360
    },
    {
      "epoch": 7.584,
      "grad_norm": 27.580427169799805,
      "learning_rate": 1.3354700854700855e-05,
      "loss": 0.0448,
      "step": 2370
    },
    {
      "epoch": 7.616,
      "grad_norm": 6.511716842651367,
      "learning_rate": 1.3176638176638178e-05,
      "loss": 0.0715,
      "step": 2380
    },
    {
      "epoch": 7.648,
      "grad_norm": 20.728763580322266,
      "learning_rate": 1.29985754985755e-05,
      "loss": 0.06,
      "step": 2390
    },
    {
      "epoch": 7.68,
      "grad_norm": 55.55878448486328,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.0589,
      "step": 2400
    },
    {
      "epoch": 7.712,
      "grad_norm": 15.75649642944336,
      "learning_rate": 1.2642450142450143e-05,
      "loss": 0.0716,
      "step": 2410
    },
    {
      "epoch": 7.744,
      "grad_norm": 6.725178241729736,
      "learning_rate": 1.2464387464387465e-05,
      "loss": 0.048,
      "step": 2420
    },
    {
      "epoch": 7.776,
      "grad_norm": 16.96969985961914,
      "learning_rate": 1.2286324786324787e-05,
      "loss": 0.056,
      "step": 2430
    },
    {
      "epoch": 7.808,
      "grad_norm": 12.85951042175293,
      "learning_rate": 1.2108262108262108e-05,
      "loss": 0.0534,
      "step": 2440
    },
    {
      "epoch": 7.84,
      "grad_norm": 3.358398675918579,
      "learning_rate": 1.193019943019943e-05,
      "loss": 0.0522,
      "step": 2450
    },
    {
      "epoch": 7.872,
      "grad_norm": 21.787321090698242,
      "learning_rate": 1.1752136752136752e-05,
      "loss": 0.0689,
      "step": 2460
    },
    {
      "epoch": 7.904,
      "grad_norm": 11.33475399017334,
      "learning_rate": 1.1574074074074075e-05,
      "loss": 0.0492,
      "step": 2470
    },
    {
      "epoch": 7.936,
      "grad_norm": 33.44040298461914,
      "learning_rate": 1.1396011396011397e-05,
      "loss": 0.0573,
      "step": 2480
    },
    {
      "epoch": 7.968,
      "grad_norm": 13.059186935424805,
      "learning_rate": 1.1217948717948719e-05,
      "loss": 0.0443,
      "step": 2490
    },
    {
      "epoch": 8.0,
      "grad_norm": 17.491514205932617,
      "learning_rate": 1.103988603988604e-05,
      "loss": 0.0419,
      "step": 2500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8769953051643192,
      "eval_loss": 0.5878553986549377,
      "eval_runtime": 15.5797,
      "eval_samples_per_second": 68.358,
      "eval_steps_per_second": 2.182,
      "step": 2500
    },
    {
      "epoch": 8.032,
      "grad_norm": 13.065751075744629,
      "learning_rate": 1.0861823361823362e-05,
      "loss": 0.0444,
      "step": 2510
    },
    {
      "epoch": 8.064,
      "grad_norm": 5.708134174346924,
      "learning_rate": 1.0683760683760684e-05,
      "loss": 0.0526,
      "step": 2520
    },
    {
      "epoch": 8.096,
      "grad_norm": 5.087850570678711,
      "learning_rate": 1.0505698005698005e-05,
      "loss": 0.0438,
      "step": 2530
    },
    {
      "epoch": 8.128,
      "grad_norm": 13.717277526855469,
      "learning_rate": 1.0327635327635327e-05,
      "loss": 0.0436,
      "step": 2540
    },
    {
      "epoch": 8.16,
      "grad_norm": 20.8705997467041,
      "learning_rate": 1.014957264957265e-05,
      "loss": 0.0397,
      "step": 2550
    },
    {
      "epoch": 8.192,
      "grad_norm": 15.591662406921387,
      "learning_rate": 9.971509971509972e-06,
      "loss": 0.0523,
      "step": 2560
    },
    {
      "epoch": 8.224,
      "grad_norm": 25.442882537841797,
      "learning_rate": 9.793447293447294e-06,
      "loss": 0.0506,
      "step": 2570
    },
    {
      "epoch": 8.256,
      "grad_norm": 17.579599380493164,
      "learning_rate": 9.615384615384616e-06,
      "loss": 0.0431,
      "step": 2580
    },
    {
      "epoch": 8.288,
      "grad_norm": 19.1536922454834,
      "learning_rate": 9.437321937321937e-06,
      "loss": 0.0499,
      "step": 2590
    },
    {
      "epoch": 8.32,
      "grad_norm": 2.3261446952819824,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.0468,
      "step": 2600
    },
    {
      "epoch": 8.352,
      "grad_norm": 10.36978816986084,
      "learning_rate": 9.081196581196581e-06,
      "loss": 0.0501,
      "step": 2610
    },
    {
      "epoch": 8.384,
      "grad_norm": 8.601365089416504,
      "learning_rate": 8.903133903133903e-06,
      "loss": 0.0412,
      "step": 2620
    },
    {
      "epoch": 8.416,
      "grad_norm": 23.900691986083984,
      "learning_rate": 8.725071225071226e-06,
      "loss": 0.049,
      "step": 2630
    },
    {
      "epoch": 8.448,
      "grad_norm": 20.452377319335938,
      "learning_rate": 8.547008547008548e-06,
      "loss": 0.0612,
      "step": 2640
    },
    {
      "epoch": 8.48,
      "grad_norm": 13.286907196044922,
      "learning_rate": 8.36894586894587e-06,
      "loss": 0.0435,
      "step": 2650
    },
    {
      "epoch": 8.512,
      "grad_norm": 4.822972297668457,
      "learning_rate": 8.190883190883191e-06,
      "loss": 0.048,
      "step": 2660
    },
    {
      "epoch": 8.544,
      "grad_norm": 12.461104393005371,
      "learning_rate": 8.012820512820515e-06,
      "loss": 0.0457,
      "step": 2670
    },
    {
      "epoch": 8.576,
      "grad_norm": 5.80748176574707,
      "learning_rate": 7.834757834757835e-06,
      "loss": 0.0481,
      "step": 2680
    },
    {
      "epoch": 8.608,
      "grad_norm": 17.747467041015625,
      "learning_rate": 7.656695156695156e-06,
      "loss": 0.0466,
      "step": 2690
    },
    {
      "epoch": 8.64,
      "grad_norm": 10.208796501159668,
      "learning_rate": 7.478632478632479e-06,
      "loss": 0.0427,
      "step": 2700
    },
    {
      "epoch": 8.672,
      "grad_norm": 2.8920395374298096,
      "learning_rate": 7.3005698005698005e-06,
      "loss": 0.0312,
      "step": 2710
    },
    {
      "epoch": 8.704,
      "grad_norm": 15.091995239257812,
      "learning_rate": 7.122507122507123e-06,
      "loss": 0.0518,
      "step": 2720
    },
    {
      "epoch": 8.736,
      "grad_norm": 3.1350409984588623,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.0557,
      "step": 2730
    },
    {
      "epoch": 8.768,
      "grad_norm": 17.688913345336914,
      "learning_rate": 6.766381766381767e-06,
      "loss": 0.0393,
      "step": 2740
    },
    {
      "epoch": 8.8,
      "grad_norm": 6.558457851409912,
      "learning_rate": 6.588319088319089e-06,
      "loss": 0.0434,
      "step": 2750
    },
    {
      "epoch": 8.832,
      "grad_norm": 2.839106798171997,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.0425,
      "step": 2760
    },
    {
      "epoch": 8.864,
      "grad_norm": 7.894480228424072,
      "learning_rate": 6.2321937321937325e-06,
      "loss": 0.0409,
      "step": 2770
    },
    {
      "epoch": 8.896,
      "grad_norm": 17.717144012451172,
      "learning_rate": 6.054131054131054e-06,
      "loss": 0.0402,
      "step": 2780
    },
    {
      "epoch": 8.928,
      "grad_norm": 6.443696022033691,
      "learning_rate": 5.876068376068376e-06,
      "loss": 0.0463,
      "step": 2790
    },
    {
      "epoch": 8.96,
      "grad_norm": 19.47768211364746,
      "learning_rate": 5.6980056980056985e-06,
      "loss": 0.0503,
      "step": 2800
    },
    {
      "epoch": 8.992,
      "grad_norm": 15.27401351928711,
      "learning_rate": 5.51994301994302e-06,
      "loss": 0.0469,
      "step": 2810
    },
    {
      "epoch": 8.9984,
      "eval_accuracy": 0.8854460093896713,
      "eval_loss": 0.5125524997711182,
      "eval_runtime": 14.9538,
      "eval_samples_per_second": 71.219,
      "eval_steps_per_second": 2.274,
      "step": 2812
    },
    {
      "epoch": 9.024,
      "grad_norm": 4.765958786010742,
      "learning_rate": 5.341880341880342e-06,
      "loss": 0.0294,
      "step": 2820
    },
    {
      "epoch": 9.056,
      "grad_norm": 7.613553524017334,
      "learning_rate": 5.163817663817664e-06,
      "loss": 0.0431,
      "step": 2830
    },
    {
      "epoch": 9.088,
      "grad_norm": 5.009788990020752,
      "learning_rate": 4.985754985754986e-06,
      "loss": 0.0298,
      "step": 2840
    },
    {
      "epoch": 9.12,
      "grad_norm": 5.033410549163818,
      "learning_rate": 4.807692307692308e-06,
      "loss": 0.0355,
      "step": 2850
    },
    {
      "epoch": 9.152,
      "grad_norm": 14.746557235717773,
      "learning_rate": 4.6296296296296296e-06,
      "loss": 0.0335,
      "step": 2860
    },
    {
      "epoch": 9.184,
      "grad_norm": 14.290383338928223,
      "learning_rate": 4.451566951566951e-06,
      "loss": 0.0389,
      "step": 2870
    },
    {
      "epoch": 9.216,
      "grad_norm": 2.576585054397583,
      "learning_rate": 4.273504273504274e-06,
      "loss": 0.0301,
      "step": 2880
    },
    {
      "epoch": 9.248,
      "grad_norm": 10.129234313964844,
      "learning_rate": 4.0954415954415956e-06,
      "loss": 0.031,
      "step": 2890
    },
    {
      "epoch": 9.28,
      "grad_norm": 33.465911865234375,
      "learning_rate": 3.917378917378917e-06,
      "loss": 0.0506,
      "step": 2900
    },
    {
      "epoch": 9.312,
      "grad_norm": 24.806825637817383,
      "learning_rate": 3.7393162393162394e-06,
      "loss": 0.0521,
      "step": 2910
    },
    {
      "epoch": 9.344,
      "grad_norm": 4.5822248458862305,
      "learning_rate": 3.5612535612535615e-06,
      "loss": 0.0363,
      "step": 2920
    },
    {
      "epoch": 9.376,
      "grad_norm": 5.783382892608643,
      "learning_rate": 3.3831908831908837e-06,
      "loss": 0.0393,
      "step": 2930
    },
    {
      "epoch": 9.408,
      "grad_norm": 5.785057067871094,
      "learning_rate": 3.205128205128205e-06,
      "loss": 0.0449,
      "step": 2940
    },
    {
      "epoch": 9.44,
      "grad_norm": 31.970163345336914,
      "learning_rate": 3.027065527065527e-06,
      "loss": 0.0363,
      "step": 2950
    },
    {
      "epoch": 9.472,
      "grad_norm": 8.819344520568848,
      "learning_rate": 2.8490028490028492e-06,
      "loss": 0.0267,
      "step": 2960
    },
    {
      "epoch": 9.504,
      "grad_norm": 18.424638748168945,
      "learning_rate": 2.670940170940171e-06,
      "loss": 0.0382,
      "step": 2970
    },
    {
      "epoch": 9.536,
      "grad_norm": 2.756587266921997,
      "learning_rate": 2.492877492877493e-06,
      "loss": 0.0241,
      "step": 2980
    },
    {
      "epoch": 9.568,
      "grad_norm": 25.5931339263916,
      "learning_rate": 2.3148148148148148e-06,
      "loss": 0.0449,
      "step": 2990
    },
    {
      "epoch": 9.6,
      "grad_norm": 7.068244457244873,
      "learning_rate": 2.136752136752137e-06,
      "loss": 0.0443,
      "step": 3000
    },
    {
      "epoch": 9.632,
      "grad_norm": 12.082342147827148,
      "learning_rate": 1.9586894586894586e-06,
      "loss": 0.0424,
      "step": 3010
    },
    {
      "epoch": 9.664,
      "grad_norm": 9.575368881225586,
      "learning_rate": 1.7806267806267808e-06,
      "loss": 0.0329,
      "step": 3020
    },
    {
      "epoch": 9.696,
      "grad_norm": 7.043597221374512,
      "learning_rate": 1.6025641025641025e-06,
      "loss": 0.031,
      "step": 3030
    },
    {
      "epoch": 9.728,
      "grad_norm": 12.78835391998291,
      "learning_rate": 1.4245014245014246e-06,
      "loss": 0.0268,
      "step": 3040
    },
    {
      "epoch": 9.76,
      "grad_norm": 20.406585693359375,
      "learning_rate": 1.2464387464387465e-06,
      "loss": 0.0494,
      "step": 3050
    },
    {
      "epoch": 9.792,
      "grad_norm": 22.336284637451172,
      "learning_rate": 1.0683760683760685e-06,
      "loss": 0.0427,
      "step": 3060
    },
    {
      "epoch": 9.824,
      "grad_norm": 9.303606986999512,
      "learning_rate": 8.903133903133904e-07,
      "loss": 0.0304,
      "step": 3070
    },
    {
      "epoch": 9.856,
      "grad_norm": 6.416520595550537,
      "learning_rate": 7.122507122507123e-07,
      "loss": 0.0468,
      "step": 3080
    },
    {
      "epoch": 9.888,
      "grad_norm": 14.013386726379395,
      "learning_rate": 5.341880341880342e-07,
      "loss": 0.027,
      "step": 3090
    },
    {
      "epoch": 9.92,
      "grad_norm": 5.910660266876221,
      "learning_rate": 3.5612535612535615e-07,
      "loss": 0.04,
      "step": 3100
    },
    {
      "epoch": 9.952,
      "grad_norm": 11.427671432495117,
      "learning_rate": 1.7806267806267808e-07,
      "loss": 0.0239,
      "step": 3110
    },
    {
      "epoch": 9.984,
      "grad_norm": 4.808414936065674,
      "learning_rate": 0.0,
      "loss": 0.0302,
      "step": 3120
    },
    {
      "epoch": 9.984,
      "eval_accuracy": 0.8807511737089202,
      "eval_loss": 0.606827437877655,
      "eval_runtime": 15.744,
      "eval_samples_per_second": 67.645,
      "eval_steps_per_second": 2.16,
      "step": 3120
    },
    {
      "epoch": 9.984,
      "step": 3120,
      "total_flos": 9.926487761391452e+18,
      "train_loss": 0.15854326258771695,
      "train_runtime": 146927.3485,
      "train_samples_per_second": 2.722,
      "train_steps_per_second": 0.021
    }
  ],
  "logging_steps": 10,
  "max_steps": 3120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.926487761391452e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}