{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 8400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.023809523809523808,
      "grad_norm": 0.09493794292211533,
      "learning_rate": 5e-05,
      "loss": 0.2812,
      "step": 10
    },
    {
      "epoch": 0.047619047619047616,
      "grad_norm": 0.06558936089277267,
      "learning_rate": 5e-05,
      "loss": 0.0724,
      "step": 20
    },
    {
      "epoch": 0.07142857142857142,
      "grad_norm": 0.05934833735227585,
      "learning_rate": 5e-05,
      "loss": 0.0328,
      "step": 30
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 0.0350530743598938,
      "learning_rate": 5e-05,
      "loss": 0.0267,
      "step": 40
    },
    {
      "epoch": 0.11904761904761904,
      "grad_norm": 0.028712034225463867,
      "learning_rate": 5e-05,
      "loss": 0.0236,
      "step": 50
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 0.024718789383769035,
      "learning_rate": 5e-05,
      "loss": 0.023,
      "step": 60
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 0.022061526775360107,
      "learning_rate": 5e-05,
      "loss": 0.0226,
      "step": 70
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 0.025670066475868225,
      "learning_rate": 5e-05,
      "loss": 0.0218,
      "step": 80
    },
    {
      "epoch": 0.21428571428571427,
      "grad_norm": 0.02997678704559803,
      "learning_rate": 5e-05,
      "loss": 0.0209,
      "step": 90
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 0.0373968742787838,
      "learning_rate": 5e-05,
      "loss": 0.0221,
      "step": 100
    },
    {
      "epoch": 0.2619047619047619,
      "grad_norm": 0.02778824232518673,
      "learning_rate": 5e-05,
      "loss": 0.0207,
      "step": 110
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.018951816484332085,
      "learning_rate": 5e-05,
      "loss": 0.0212,
      "step": 120
    },
    {
      "epoch": 0.30952380952380953,
      "grad_norm": 0.03219084069132805,
      "learning_rate": 5e-05,
      "loss": 0.0197,
      "step": 130
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.035099539905786514,
      "learning_rate": 5e-05,
      "loss": 0.0212,
      "step": 140
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 0.022762970998883247,
      "learning_rate": 5e-05,
      "loss": 0.0205,
      "step": 150
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 0.021571127697825432,
      "learning_rate": 5e-05,
      "loss": 0.0192,
      "step": 160
    },
    {
      "epoch": 0.40476190476190477,
      "grad_norm": 0.028216874226927757,
      "learning_rate": 5e-05,
      "loss": 0.0194,
      "step": 170
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 0.018433215096592903,
      "learning_rate": 5e-05,
      "loss": 0.0201,
      "step": 180
    },
    {
      "epoch": 0.4523809523809524,
      "grad_norm": 0.03548741340637207,
      "learning_rate": 5e-05,
      "loss": 0.0193,
      "step": 190
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 0.027879556640982628,
      "learning_rate": 5e-05,
      "loss": 0.0187,
      "step": 200
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.022470738738775253,
      "learning_rate": 5e-05,
      "loss": 0.0186,
      "step": 210
    },
    {
      "epoch": 0.5238095238095238,
      "grad_norm": 0.02206714078783989,
      "learning_rate": 5e-05,
      "loss": 0.0184,
      "step": 220
    },
    {
      "epoch": 0.5476190476190477,
      "grad_norm": 0.024714656174182892,
      "learning_rate": 5e-05,
      "loss": 0.0188,
      "step": 230
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.03632698208093643,
      "learning_rate": 5e-05,
      "loss": 0.0188,
      "step": 240
    },
    {
      "epoch": 0.5952380952380952,
      "grad_norm": 0.017954664304852486,
      "learning_rate": 5e-05,
      "loss": 0.0177,
      "step": 250
    },
    {
      "epoch": 0.6190476190476191,
      "grad_norm": 0.026338977739214897,
      "learning_rate": 5e-05,
      "loss": 0.0166,
      "step": 260
    },
    {
      "epoch": 0.6428571428571429,
      "grad_norm": 0.028515340760350227,
      "learning_rate": 5e-05,
      "loss": 0.0177,
      "step": 270
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.0308682881295681,
      "learning_rate": 5e-05,
      "loss": 0.0176,
      "step": 280
    },
    {
      "epoch": 0.6904761904761905,
      "grad_norm": 0.018636852502822876,
      "learning_rate": 5e-05,
      "loss": 0.0167,
      "step": 290
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.024639299139380455,
      "learning_rate": 5e-05,
      "loss": 0.0171,
      "step": 300
    },
    {
      "epoch": 0.7380952380952381,
      "grad_norm": 0.02328549325466156,
      "learning_rate": 5e-05,
      "loss": 0.018,
      "step": 310
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.02998117357492447,
      "learning_rate": 5e-05,
      "loss": 0.0166,
      "step": 320
    },
    {
      "epoch": 0.7857142857142857,
      "grad_norm": 0.022752949967980385,
      "learning_rate": 5e-05,
      "loss": 0.017,
      "step": 330
    },
    {
      "epoch": 0.8095238095238095,
      "grad_norm": 0.02922181785106659,
      "learning_rate": 5e-05,
      "loss": 0.0176,
      "step": 340
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.026469197124242783,
      "learning_rate": 5e-05,
      "loss": 0.0183,
      "step": 350
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.03358051925897598,
      "learning_rate": 5e-05,
      "loss": 0.0188,
      "step": 360
    },
    {
      "epoch": 0.8809523809523809,
      "grad_norm": 0.0185205340385437,
      "learning_rate": 5e-05,
      "loss": 0.0173,
      "step": 370
    },
    {
      "epoch": 0.9047619047619048,
      "grad_norm": 0.018949517980217934,
      "learning_rate": 5e-05,
      "loss": 0.0169,
      "step": 380
    },
    {
      "epoch": 0.9285714285714286,
      "grad_norm": 0.022263336926698685,
      "learning_rate": 5e-05,
      "loss": 0.017,
      "step": 390
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 0.023463713005185127,
      "learning_rate": 5e-05,
      "loss": 0.0172,
      "step": 400
    },
    {
      "epoch": 0.9761904761904762,
      "grad_norm": 0.023527899757027626,
      "learning_rate": 5e-05,
      "loss": 0.0156,
      "step": 410
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.026089975610375404,
      "learning_rate": 5e-05,
      "loss": 0.0154,
      "step": 420
    },
    {
      "epoch": 1.0238095238095237,
      "grad_norm": 0.017531510442495346,
      "learning_rate": 5e-05,
      "loss": 0.0175,
      "step": 430
    },
    {
      "epoch": 1.0476190476190477,
      "grad_norm": 0.017702311277389526,
      "learning_rate": 5e-05,
      "loss": 0.0151,
      "step": 440
    },
    {
      "epoch": 1.0714285714285714,
      "grad_norm": 0.018016375601291656,
      "learning_rate": 5e-05,
      "loss": 0.0168,
      "step": 450
    },
    {
      "epoch": 1.0952380952380953,
      "grad_norm": 0.028256898745894432,
      "learning_rate": 5e-05,
      "loss": 0.0164,
      "step": 460
    },
    {
      "epoch": 1.119047619047619,
      "grad_norm": 0.017876937985420227,
      "learning_rate": 5e-05,
      "loss": 0.0148,
      "step": 470
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.018716128543019295,
      "learning_rate": 5e-05,
      "loss": 0.0144,
      "step": 480
    },
    {
      "epoch": 1.1666666666666667,
      "grad_norm": 0.02162998355925083,
      "learning_rate": 5e-05,
      "loss": 0.0148,
      "step": 490
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 0.023180391639471054,
      "learning_rate": 5e-05,
      "loss": 0.0156,
      "step": 500
    },
    {
      "epoch": 1.2142857142857142,
      "grad_norm": 0.022495009005069733,
      "learning_rate": 5e-05,
      "loss": 0.0149,
      "step": 510
    },
    {
      "epoch": 1.2380952380952381,
      "grad_norm": 0.024750931188464165,
      "learning_rate": 5e-05,
      "loss": 0.0158,
      "step": 520
    },
    {
      "epoch": 1.2619047619047619,
      "grad_norm": 0.027094174176454544,
      "learning_rate": 5e-05,
      "loss": 0.0149,
      "step": 530
    },
    {
      "epoch": 1.2857142857142856,
      "grad_norm": 0.021923832595348358,
      "learning_rate": 5e-05,
      "loss": 0.0146,
      "step": 540
    },
    {
      "epoch": 1.3095238095238095,
      "grad_norm": 0.025097332894802094,
      "learning_rate": 5e-05,
      "loss": 0.0136,
      "step": 550
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.034609246999025345,
      "learning_rate": 5e-05,
      "loss": 0.0152,
      "step": 560
    },
    {
      "epoch": 1.3571428571428572,
      "grad_norm": 0.02195250615477562,
      "learning_rate": 5e-05,
      "loss": 0.0141,
      "step": 570
    },
    {
      "epoch": 1.380952380952381,
      "grad_norm": 0.01835169643163681,
      "learning_rate": 5e-05,
      "loss": 0.0148,
      "step": 580
    },
    {
      "epoch": 1.4047619047619047,
      "grad_norm": 0.026779156178236008,
      "learning_rate": 5e-05,
      "loss": 0.0154,
      "step": 590
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.025482457131147385,
      "learning_rate": 5e-05,
      "loss": 0.0142,
      "step": 600
    },
    {
      "epoch": 1.4523809523809523,
      "grad_norm": 0.022315753623843193,
      "learning_rate": 5e-05,
      "loss": 0.0137,
      "step": 610
    },
    {
      "epoch": 1.4761904761904763,
      "grad_norm": 0.02147059701383114,
      "learning_rate": 5e-05,
      "loss": 0.0137,
      "step": 620
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.021188456565141678,
      "learning_rate": 5e-05,
      "loss": 0.0135,
      "step": 630
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 0.018250903114676476,
      "learning_rate": 5e-05,
      "loss": 0.0137,
      "step": 640
    },
    {
      "epoch": 1.5476190476190477,
      "grad_norm": 0.02014111913740635,
      "learning_rate": 5e-05,
      "loss": 0.0141,
      "step": 650
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 0.031997792422771454,
      "learning_rate": 5e-05,
      "loss": 0.0145,
      "step": 660
    },
    {
      "epoch": 1.5952380952380953,
      "grad_norm": 0.02311546355485916,
      "learning_rate": 5e-05,
      "loss": 0.0139,
      "step": 670
    },
    {
      "epoch": 1.619047619047619,
      "grad_norm": 0.022159090265631676,
      "learning_rate": 5e-05,
      "loss": 0.0127,
      "step": 680
    },
    {
      "epoch": 1.6428571428571428,
      "grad_norm": 0.023639395833015442,
      "learning_rate": 5e-05,
      "loss": 0.0139,
      "step": 690
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.025874590501189232,
      "learning_rate": 5e-05,
      "loss": 0.0142,
      "step": 700
    },
    {
      "epoch": 1.6904761904761905,
      "grad_norm": 0.015242703258991241,
      "learning_rate": 5e-05,
      "loss": 0.0131,
      "step": 710
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.01896982081234455,
      "learning_rate": 5e-05,
      "loss": 0.0129,
      "step": 720
    },
    {
      "epoch": 1.7380952380952381,
      "grad_norm": 0.019098645076155663,
      "learning_rate": 5e-05,
      "loss": 0.014,
      "step": 730
    },
    {
      "epoch": 1.7619047619047619,
      "grad_norm": 0.020913679152727127,
      "learning_rate": 5e-05,
      "loss": 0.0139,
      "step": 740
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 0.02301224134862423,
      "learning_rate": 5e-05,
      "loss": 0.0134,
      "step": 750
    },
    {
      "epoch": 1.8095238095238095,
      "grad_norm": 0.020251275971531868,
      "learning_rate": 5e-05,
      "loss": 0.0126,
      "step": 760
    },
    {
      "epoch": 1.8333333333333335,
      "grad_norm": 0.023471834138035774,
      "learning_rate": 5e-05,
      "loss": 0.0136,
      "step": 770
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 0.024632664397358894,
      "learning_rate": 5e-05,
      "loss": 0.0145,
      "step": 780
    },
    {
      "epoch": 1.880952380952381,
      "grad_norm": 0.021192265674471855,
      "learning_rate": 5e-05,
      "loss": 0.0138,
      "step": 790
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.021170521154999733,
      "learning_rate": 5e-05,
      "loss": 0.0136,
      "step": 800
    },
    {
      "epoch": 1.9285714285714286,
      "grad_norm": 0.018560606986284256,
      "learning_rate": 5e-05,
      "loss": 0.0129,
      "step": 810
    },
    {
      "epoch": 1.9523809523809523,
      "grad_norm": 0.020409613847732544,
      "learning_rate": 5e-05,
      "loss": 0.0138,
      "step": 820
    },
    {
      "epoch": 1.9761904761904763,
      "grad_norm": 0.02413373813033104,
      "learning_rate": 5e-05,
      "loss": 0.0128,
      "step": 830
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.022253163158893585,
      "learning_rate": 5e-05,
      "loss": 0.0127,
      "step": 840
    },
    {
      "epoch": 2.0238095238095237,
      "grad_norm": 0.017652267590165138,
      "learning_rate": 5e-05,
      "loss": 0.0143,
      "step": 850
    },
    {
      "epoch": 2.0476190476190474,
      "grad_norm": 0.018153948709368706,
      "learning_rate": 5e-05,
      "loss": 0.0128,
      "step": 860
    },
    {
      "epoch": 2.0714285714285716,
      "grad_norm": 0.01955491490662098,
      "learning_rate": 5e-05,
      "loss": 0.0132,
      "step": 870
    },
    {
      "epoch": 2.0952380952380953,
      "grad_norm": 0.021146738901734352,
      "learning_rate": 5e-05,
      "loss": 0.0124,
      "step": 880
    },
    {
      "epoch": 2.119047619047619,
      "grad_norm": 0.016339635476469994,
      "learning_rate": 5e-05,
      "loss": 0.0122,
      "step": 890
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 0.017886130139231682,
      "learning_rate": 5e-05,
      "loss": 0.0113,
      "step": 900
    },
    {
      "epoch": 2.1666666666666665,
      "grad_norm": 0.02060827612876892,
      "learning_rate": 5e-05,
      "loss": 0.0117,
      "step": 910
    },
    {
      "epoch": 2.1904761904761907,
      "grad_norm": 0.016803547739982605,
      "learning_rate": 5e-05,
      "loss": 0.0132,
      "step": 920
    },
    {
      "epoch": 2.2142857142857144,
      "grad_norm": 0.018075264990329742,
      "learning_rate": 5e-05,
      "loss": 0.0119,
      "step": 930
    },
    {
      "epoch": 2.238095238095238,
      "grad_norm": 0.023032259196043015,
      "learning_rate": 5e-05,
      "loss": 0.0121,
      "step": 940
    },
    {
      "epoch": 2.261904761904762,
      "grad_norm": 0.028263242915272713,
      "learning_rate": 5e-05,
      "loss": 0.0114,
      "step": 950
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.02737823873758316,
      "learning_rate": 5e-05,
      "loss": 0.0112,
      "step": 960
    },
    {
      "epoch": 2.3095238095238093,
      "grad_norm": 0.021527152508497238,
      "learning_rate": 5e-05,
      "loss": 0.0111,
      "step": 970
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.027928918600082397,
      "learning_rate": 5e-05,
      "loss": 0.0123,
      "step": 980
    },
    {
      "epoch": 2.357142857142857,
      "grad_norm": 0.024291587993502617,
      "learning_rate": 5e-05,
      "loss": 0.0112,
      "step": 990
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.024933788925409317,
      "learning_rate": 5e-05,
      "loss": 0.0122,
      "step": 1000
    },
    {
      "epoch": 2.4047619047619047,
      "grad_norm": 0.025925207883119583,
      "learning_rate": 5e-05,
      "loss": 0.0126,
      "step": 1010
    },
    {
      "epoch": 2.4285714285714284,
      "grad_norm": 0.021988635882735252,
      "learning_rate": 5e-05,
      "loss": 0.0111,
      "step": 1020
    },
    {
      "epoch": 2.4523809523809526,
      "grad_norm": 0.029331982135772705,
      "learning_rate": 5e-05,
      "loss": 0.0113,
      "step": 1030
    },
    {
      "epoch": 2.4761904761904763,
      "grad_norm": 0.023319706320762634,
      "learning_rate": 5e-05,
      "loss": 0.0111,
      "step": 1040
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.016594231128692627,
      "learning_rate": 5e-05,
      "loss": 0.0106,
      "step": 1050
    },
    {
      "epoch": 2.5238095238095237,
      "grad_norm": 0.022670956328511238,
      "learning_rate": 5e-05,
      "loss": 0.0109,
      "step": 1060
    },
    {
      "epoch": 2.5476190476190474,
      "grad_norm": 0.02389667183160782,
      "learning_rate": 5e-05,
      "loss": 0.0116,
      "step": 1070
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 0.03487304970622063,
      "learning_rate": 5e-05,
      "loss": 0.0114,
      "step": 1080
    },
    {
      "epoch": 2.5952380952380953,
      "grad_norm": 0.02180272527039051,
      "learning_rate": 5e-05,
      "loss": 0.0116,
      "step": 1090
    },
    {
      "epoch": 2.619047619047619,
      "grad_norm": 0.021106529980897903,
      "learning_rate": 5e-05,
      "loss": 0.0105,
      "step": 1100
    },
    {
      "epoch": 2.642857142857143,
      "grad_norm": 0.023611431941390038,
      "learning_rate": 5e-05,
      "loss": 0.0119,
      "step": 1110
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.024628831073641777,
      "learning_rate": 5e-05,
      "loss": 0.0114,
      "step": 1120
    },
    {
      "epoch": 2.6904761904761907,
      "grad_norm": 0.014470246620476246,
      "learning_rate": 5e-05,
      "loss": 0.0102,
      "step": 1130
    },
    {
      "epoch": 2.7142857142857144,
      "grad_norm": 0.022718733176589012,
      "learning_rate": 5e-05,
      "loss": 0.0103,
      "step": 1140
    },
    {
      "epoch": 2.738095238095238,
      "grad_norm": 0.020995931699872017,
      "learning_rate": 5e-05,
      "loss": 0.0106,
      "step": 1150
    },
    {
      "epoch": 2.761904761904762,
      "grad_norm": 0.024527721107006073,
      "learning_rate": 5e-05,
      "loss": 0.0107,
      "step": 1160
    },
    {
      "epoch": 2.7857142857142856,
      "grad_norm": 0.021473698318004608,
      "learning_rate": 5e-05,
      "loss": 0.0104,
      "step": 1170
    },
    {
      "epoch": 2.8095238095238093,
      "grad_norm": 0.018530316650867462,
      "learning_rate": 5e-05,
      "loss": 0.0101,
      "step": 1180
    },
    {
      "epoch": 2.8333333333333335,
      "grad_norm": 0.02302338182926178,
      "learning_rate": 5e-05,
      "loss": 0.0105,
      "step": 1190
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.025742432102560997,
      "learning_rate": 5e-05,
      "loss": 0.0114,
      "step": 1200
    },
    {
      "epoch": 2.880952380952381,
      "grad_norm": 0.027784395962953568,
      "learning_rate": 5e-05,
      "loss": 0.0111,
      "step": 1210
    },
    {
      "epoch": 2.9047619047619047,
      "grad_norm": 0.021398941054940224,
      "learning_rate": 5e-05,
      "loss": 0.0102,
      "step": 1220
    },
    {
      "epoch": 2.928571428571429,
      "grad_norm": 0.027173219248652458,
      "learning_rate": 5e-05,
      "loss": 0.0101,
      "step": 1230
    },
    {
      "epoch": 2.9523809523809526,
      "grad_norm": 0.02305479347705841,
      "learning_rate": 5e-05,
      "loss": 0.0113,
      "step": 1240
    },
    {
      "epoch": 2.9761904761904763,
      "grad_norm": 0.024614600464701653,
      "learning_rate": 5e-05,
      "loss": 0.0105,
      "step": 1250
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.01538562960922718,
      "learning_rate": 5e-05,
      "loss": 0.0103,
      "step": 1260
    },
    {
      "epoch": 3.0238095238095237,
      "grad_norm": 0.018940571695566177,
      "learning_rate": 5e-05,
      "loss": 0.0117,
      "step": 1270
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 0.014765114523470402,
      "learning_rate": 5e-05,
      "loss": 0.0104,
      "step": 1280
    },
    {
      "epoch": 3.0714285714285716,
      "grad_norm": 0.02444819174706936,
      "learning_rate": 5e-05,
      "loss": 0.0115,
      "step": 1290
    },
    {
      "epoch": 3.0952380952380953,
      "grad_norm": 0.02040857821702957,
      "learning_rate": 5e-05,
      "loss": 0.0102,
      "step": 1300
    },
    {
      "epoch": 3.119047619047619,
      "grad_norm": 0.01789598912000656,
      "learning_rate": 5e-05,
      "loss": 0.0095,
      "step": 1310
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 0.020275596529245377,
      "learning_rate": 5e-05,
      "loss": 0.0084,
      "step": 1320
    },
    {
      "epoch": 3.1666666666666665,
      "grad_norm": 0.02351970411837101,
      "learning_rate": 5e-05,
      "loss": 0.0098,
      "step": 1330
    },
    {
      "epoch": 3.1904761904761907,
      "grad_norm": 0.017525261268019676,
      "learning_rate": 5e-05,
      "loss": 0.0103,
      "step": 1340
    },
    {
      "epoch": 3.2142857142857144,
      "grad_norm": 0.019003944471478462,
      "learning_rate": 5e-05,
      "loss": 0.0094,
      "step": 1350
    },
    {
      "epoch": 3.238095238095238,
      "grad_norm": 0.02266862988471985,
      "learning_rate": 5e-05,
      "loss": 0.0094,
      "step": 1360
    },
    {
      "epoch": 3.261904761904762,
      "grad_norm": 0.019711384549736977,
      "learning_rate": 5e-05,
      "loss": 0.0093,
      "step": 1370
    },
    {
      "epoch": 3.2857142857142856,
      "grad_norm": 0.02152351662516594,
      "learning_rate": 5e-05,
      "loss": 0.0082,
      "step": 1380
    },
    {
      "epoch": 3.3095238095238093,
      "grad_norm": 0.018401287496089935,
      "learning_rate": 5e-05,
      "loss": 0.0082,
      "step": 1390
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.030380433425307274,
      "learning_rate": 5e-05,
      "loss": 0.0095,
      "step": 1400
    },
    {
      "epoch": 3.357142857142857,
      "grad_norm": 0.018700776621699333,
      "learning_rate": 5e-05,
      "loss": 0.0087,
      "step": 1410
    },
    {
      "epoch": 3.380952380952381,
      "grad_norm": 0.01758432574570179,
      "learning_rate": 5e-05,
      "loss": 0.0092,
      "step": 1420
    },
    {
      "epoch": 3.4047619047619047,
      "grad_norm": 0.027525080367922783,
      "learning_rate": 5e-05,
      "loss": 0.0096,
      "step": 1430
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.023578746244311333,
      "learning_rate": 5e-05,
      "loss": 0.0087,
      "step": 1440
    },
    {
      "epoch": 3.4523809523809526,
      "grad_norm": 0.020561877638101578,
      "learning_rate": 5e-05,
      "loss": 0.0086,
      "step": 1450
    },
    {
      "epoch": 3.4761904761904763,
      "grad_norm": 0.024523900821805,
      "learning_rate": 5e-05,
      "loss": 0.0089,
      "step": 1460
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.017780443653464317,
      "learning_rate": 5e-05,
      "loss": 0.0082,
      "step": 1470
    },
    {
      "epoch": 3.5238095238095237,
      "grad_norm": 0.025551529601216316,
      "learning_rate": 5e-05,
      "loss": 0.0083,
      "step": 1480
    },
    {
      "epoch": 3.5476190476190474,
      "grad_norm": 0.024370500817894936,
      "learning_rate": 5e-05,
      "loss": 0.0091,
      "step": 1490
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 0.03639150410890579,
      "learning_rate": 5e-05,
      "loss": 0.0092,
      "step": 1500
    },
    {
      "epoch": 3.5952380952380953,
      "grad_norm": 0.02289510704576969,
      "learning_rate": 5e-05,
      "loss": 0.009,
      "step": 1510
    },
    {
      "epoch": 3.619047619047619,
      "grad_norm": 0.019606849178671837,
      "learning_rate": 5e-05,
      "loss": 0.0088,
      "step": 1520
    },
    {
      "epoch": 3.642857142857143,
      "grad_norm": 0.037268321961164474,
      "learning_rate": 5e-05,
      "loss": 0.0088,
      "step": 1530
    },
    {
      "epoch": 3.6666666666666665,
      "grad_norm": 0.028575578704476357,
      "learning_rate": 5e-05,
      "loss": 0.0099,
      "step": 1540
    },
    {
      "epoch": 3.6904761904761907,
      "grad_norm": 0.013424446806311607,
      "learning_rate": 5e-05,
      "loss": 0.0076,
      "step": 1550
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 0.01701328158378601,
      "learning_rate": 5e-05,
      "loss": 0.0084,
      "step": 1560
    },
    {
      "epoch": 3.738095238095238,
      "grad_norm": 0.025259124115109444,
      "learning_rate": 5e-05,
      "loss": 0.0081,
      "step": 1570
    },
    {
      "epoch": 3.761904761904762,
      "grad_norm": 0.024821562692523003,
      "learning_rate": 5e-05,
      "loss": 0.0086,
      "step": 1580
    },
    {
      "epoch": 3.7857142857142856,
      "grad_norm": 0.024605993181467056,
      "learning_rate": 5e-05,
      "loss": 0.0084,
      "step": 1590
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 0.02455594390630722,
      "learning_rate": 5e-05,
      "loss": 0.0075,
      "step": 1600
    },
    {
      "epoch": 3.8333333333333335,
      "grad_norm": 0.025864994153380394,
      "learning_rate": 5e-05,
      "loss": 0.008,
      "step": 1610
    },
    {
      "epoch": 3.857142857142857,
      "grad_norm": 0.02395598590373993,
      "learning_rate": 5e-05,
      "loss": 0.0087,
      "step": 1620
    },
    {
      "epoch": 3.880952380952381,
      "grad_norm": 0.015691502019762993,
      "learning_rate": 5e-05,
      "loss": 0.0084,
      "step": 1630
    },
    {
      "epoch": 3.9047619047619047,
      "grad_norm": 0.0223532821983099,
      "learning_rate": 5e-05,
      "loss": 0.0077,
      "step": 1640
    },
    {
      "epoch": 3.928571428571429,
      "grad_norm": 0.028256218880414963,
      "learning_rate": 5e-05,
      "loss": 0.0076,
      "step": 1650
    },
    {
      "epoch": 3.9523809523809526,
      "grad_norm": 0.023097924888134003,
      "learning_rate": 5e-05,
      "loss": 0.0091,
      "step": 1660
    },
    {
      "epoch": 3.9761904761904763,
      "grad_norm": 0.02783946879208088,
      "learning_rate": 5e-05,
      "loss": 0.0076,
      "step": 1670
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.013011032715439796,
      "learning_rate": 5e-05,
      "loss": 0.008,
      "step": 1680
    },
    {
      "epoch": 4.023809523809524,
      "grad_norm": 0.02342393808066845,
      "learning_rate": 5e-05,
      "loss": 0.0092,
      "step": 1690
    },
    {
      "epoch": 4.0476190476190474,
      "grad_norm": 0.021954068914055824,
      "learning_rate": 5e-05,
      "loss": 0.009,
      "step": 1700
    },
    {
      "epoch": 4.071428571428571,
      "grad_norm": 0.0207274928689003,
      "learning_rate": 5e-05,
      "loss": 0.0092,
      "step": 1710
    },
    {
      "epoch": 4.095238095238095,
      "grad_norm": 0.03118881769478321,
      "learning_rate": 5e-05,
      "loss": 0.008,
      "step": 1720
    },
    {
      "epoch": 4.119047619047619,
      "grad_norm": 0.015900803729891777,
      "learning_rate": 5e-05,
      "loss": 0.0074,
      "step": 1730
    },
    {
      "epoch": 4.142857142857143,
      "grad_norm": 0.013303752988576889,
      "learning_rate": 5e-05,
      "loss": 0.0069,
      "step": 1740
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 0.022115593776106834,
      "learning_rate": 5e-05,
      "loss": 0.0078,
      "step": 1750
    },
    {
      "epoch": 4.190476190476191,
      "grad_norm": 0.021266015246510506,
      "learning_rate": 5e-05,
      "loss": 0.008,
      "step": 1760
    },
    {
      "epoch": 4.214285714285714,
      "grad_norm": 0.023192718625068665,
      "learning_rate": 5e-05,
      "loss": 0.0075,
      "step": 1770
    },
    {
      "epoch": 4.238095238095238,
      "grad_norm": 0.02218669466674328,
      "learning_rate": 5e-05,
      "loss": 0.007,
      "step": 1780
    },
    {
      "epoch": 4.261904761904762,
      "grad_norm": 0.024710968136787415,
      "learning_rate": 5e-05,
      "loss": 0.0071,
      "step": 1790
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 0.022398030385375023,
      "learning_rate": 5e-05,
      "loss": 0.0057,
      "step": 1800
    },
    {
      "epoch": 4.309523809523809,
      "grad_norm": 0.025463633239269257,
      "learning_rate": 5e-05,
      "loss": 0.0066,
      "step": 1810
    },
    {
      "epoch": 4.333333333333333,
      "grad_norm": 0.024359669536352158,
      "learning_rate": 5e-05,
      "loss": 0.0072,
      "step": 1820
    },
    {
      "epoch": 4.357142857142857,
      "grad_norm": 0.015460989437997341,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 1830
    },
    {
      "epoch": 4.380952380952381,
      "grad_norm": 0.028968673199415207,
      "learning_rate": 5e-05,
      "loss": 0.0075,
      "step": 1840
    },
    {
      "epoch": 4.404761904761905,
      "grad_norm": 0.021456323564052582,
      "learning_rate": 5e-05,
      "loss": 0.0074,
      "step": 1850
    },
    {
      "epoch": 4.428571428571429,
      "grad_norm": 0.016484368592500687,
      "learning_rate": 5e-05,
      "loss": 0.0069,
      "step": 1860
    },
    {
      "epoch": 4.4523809523809526,
      "grad_norm": 0.03759278357028961,
      "learning_rate": 5e-05,
      "loss": 0.0073,
      "step": 1870
    },
    {
      "epoch": 4.476190476190476,
      "grad_norm": 0.02179703302681446,
      "learning_rate": 5e-05,
      "loss": 0.007,
      "step": 1880
    },
    {
      "epoch": 4.5,
      "grad_norm": 0.023886868730187416,
      "learning_rate": 5e-05,
      "loss": 0.0068,
      "step": 1890
    },
    {
      "epoch": 4.523809523809524,
      "grad_norm": 0.020535631105303764,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 1900
    },
    {
      "epoch": 4.5476190476190474,
      "grad_norm": 0.024215754121541977,
      "learning_rate": 5e-05,
      "loss": 0.0065,
      "step": 1910
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 0.02726290002465248,
      "learning_rate": 5e-05,
      "loss": 0.0068,
      "step": 1920
    },
    {
      "epoch": 4.595238095238095,
      "grad_norm": 0.022762540727853775,
      "learning_rate": 5e-05,
      "loss": 0.0066,
      "step": 1930
    },
    {
      "epoch": 4.619047619047619,
      "grad_norm": 0.03385463356971741,
      "learning_rate": 5e-05,
      "loss": 0.0066,
      "step": 1940
    },
    {
      "epoch": 4.642857142857143,
      "grad_norm": 0.029403358697891235,
      "learning_rate": 5e-05,
      "loss": 0.0065,
      "step": 1950
    },
    {
      "epoch": 4.666666666666667,
      "grad_norm": 0.03025682643055916,
      "learning_rate": 5e-05,
      "loss": 0.0078,
      "step": 1960
    },
    {
      "epoch": 4.690476190476191,
      "grad_norm": 0.012874845415353775,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 1970
    },
    {
      "epoch": 4.714285714285714,
      "grad_norm": 0.021130386739969254,
      "learning_rate": 5e-05,
      "loss": 0.007,
      "step": 1980
    },
    {
      "epoch": 4.738095238095238,
      "grad_norm": 0.02132946252822876,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 1990
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 0.03074180707335472,
      "learning_rate": 5e-05,
      "loss": 0.0061,
      "step": 2000
    },
    {
      "epoch": 4.785714285714286,
      "grad_norm": 0.02292218618094921,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 2010
    },
    {
      "epoch": 4.809523809523809,
      "grad_norm": 0.01976919360458851,
      "learning_rate": 5e-05,
      "loss": 0.0061,
      "step": 2020
    },
    {
      "epoch": 4.833333333333333,
      "grad_norm": 0.030889755114912987,
      "learning_rate": 5e-05,
      "loss": 0.0064,
      "step": 2030
    },
    {
      "epoch": 4.857142857142857,
      "grad_norm": 0.02626785635948181,
      "learning_rate": 5e-05,
      "loss": 0.0071,
      "step": 2040
    },
    {
      "epoch": 4.880952380952381,
      "grad_norm": 0.016065414994955063,
      "learning_rate": 5e-05,
      "loss": 0.0067,
      "step": 2050
    },
    {
      "epoch": 4.904761904761905,
      "grad_norm": 0.022564269602298737,
      "learning_rate": 5e-05,
      "loss": 0.0057,
      "step": 2060
    },
    {
      "epoch": 4.928571428571429,
      "grad_norm": 0.03417288884520531,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 2070
    },
    {
      "epoch": 4.9523809523809526,
      "grad_norm": 0.02042423002421856,
      "learning_rate": 5e-05,
      "loss": 0.0067,
      "step": 2080
    },
    {
      "epoch": 4.976190476190476,
      "grad_norm": 0.038521867245435715,
      "learning_rate": 5e-05,
      "loss": 0.0064,
      "step": 2090
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.026785830035805702,
      "learning_rate": 5e-05,
      "loss": 0.0061,
      "step": 2100
    },
    {
      "epoch": 5.023809523809524,
      "grad_norm": 0.01977149210870266,
      "learning_rate": 5e-05,
      "loss": 0.0067,
      "step": 2110
    },
    {
      "epoch": 5.0476190476190474,
      "grad_norm": 0.028995908796787262,
      "learning_rate": 5e-05,
      "loss": 0.0065,
      "step": 2120
    },
    {
      "epoch": 5.071428571428571,
      "grad_norm": 0.027958383783698082,
      "learning_rate": 5e-05,
      "loss": 0.0067,
      "step": 2130
    },
    {
      "epoch": 5.095238095238095,
      "grad_norm": 0.024838177487254143,
      "learning_rate": 5e-05,
      "loss": 0.0062,
      "step": 2140
    },
    {
      "epoch": 5.119047619047619,
      "grad_norm": 0.019940726459026337,
      "learning_rate": 5e-05,
      "loss": 0.0056,
      "step": 2150
    },
    {
      "epoch": 5.142857142857143,
      "grad_norm": 0.023204155266284943,
      "learning_rate": 5e-05,
      "loss": 0.0055,
      "step": 2160
    },
    {
      "epoch": 5.166666666666667,
      "grad_norm": 0.02349969372153282,
      "learning_rate": 5e-05,
      "loss": 0.006,
      "step": 2170
    },
    {
      "epoch": 5.190476190476191,
      "grad_norm": 0.030205531045794487,
      "learning_rate": 5e-05,
      "loss": 0.0065,
      "step": 2180
    },
    {
      "epoch": 5.214285714285714,
      "grad_norm": 0.0321480929851532,
      "learning_rate": 5e-05,
      "loss": 0.0074,
      "step": 2190
    },
    {
      "epoch": 5.238095238095238,
      "grad_norm": 0.020824981853365898,
      "learning_rate": 5e-05,
      "loss": 0.006,
      "step": 2200
    },
    {
      "epoch": 5.261904761904762,
      "grad_norm": 0.0227160956710577,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 2210
    },
    {
      "epoch": 5.285714285714286,
      "grad_norm": 0.029231227934360504,
      "learning_rate": 5e-05,
      "loss": 0.005,
      "step": 2220
    },
    {
      "epoch": 5.309523809523809,
      "grad_norm": 0.02282622829079628,
      "learning_rate": 5e-05,
      "loss": 0.0045,
      "step": 2230
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 0.03982138633728027,
      "learning_rate": 5e-05,
      "loss": 0.0061,
      "step": 2240
    },
    {
      "epoch": 5.357142857142857,
      "grad_norm": 0.018466891720891,
      "learning_rate": 5e-05,
      "loss": 0.0048,
      "step": 2250
    },
    {
      "epoch": 5.380952380952381,
      "grad_norm": 0.024992188438773155,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 2260
    },
    {
      "epoch": 5.404761904761905,
      "grad_norm": 0.029194045811891556,
      "learning_rate": 5e-05,
      "loss": 0.0062,
      "step": 2270
    },
    {
      "epoch": 5.428571428571429,
      "grad_norm": 0.027634933590888977,
      "learning_rate": 5e-05,
      "loss": 0.0056,
      "step": 2280
    },
    {
      "epoch": 5.4523809523809526,
      "grad_norm": 0.03338102251291275,
      "learning_rate": 5e-05,
      "loss": 0.0061,
      "step": 2290
    },
    {
      "epoch": 5.476190476190476,
      "grad_norm": 0.020240414887666702,
      "learning_rate": 5e-05,
      "loss": 0.0053,
      "step": 2300
    },
    {
      "epoch": 5.5,
      "grad_norm": 0.01747560314834118,
      "learning_rate": 5e-05,
      "loss": 0.0053,
      "step": 2310
    },
    {
      "epoch": 5.523809523809524,
      "grad_norm": 0.02589523047208786,
      "learning_rate": 5e-05,
      "loss": 0.0052,
      "step": 2320
    },
    {
      "epoch": 5.5476190476190474,
      "grad_norm": 0.03360776975750923,
      "learning_rate": 5e-05,
      "loss": 0.0051,
      "step": 2330
    },
    {
      "epoch": 5.571428571428571,
      "grad_norm": 0.0385347418487072,
      "learning_rate": 5e-05,
      "loss": 0.0053,
      "step": 2340
    },
    {
      "epoch": 5.595238095238095,
      "grad_norm": 0.02222476154565811,
      "learning_rate": 5e-05,
      "loss": 0.0051,
      "step": 2350
    },
    {
      "epoch": 5.619047619047619,
      "grad_norm": 0.021084798499941826,
      "learning_rate": 5e-05,
      "loss": 0.0058,
      "step": 2360
    },
    {
      "epoch": 5.642857142857143,
      "grad_norm": 0.02974940650165081,
      "learning_rate": 5e-05,
      "loss": 0.0052,
      "step": 2370
    },
    {
      "epoch": 5.666666666666667,
      "grad_norm": 0.01862935908138752,
      "learning_rate": 5e-05,
      "loss": 0.0061,
      "step": 2380
    },
    {
      "epoch": 5.690476190476191,
      "grad_norm": 0.01248252298682928,
      "learning_rate": 5e-05,
      "loss": 0.0051,
      "step": 2390
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 0.023803608492016792,
      "learning_rate": 5e-05,
      "loss": 0.0054,
      "step": 2400
    },
    {
      "epoch": 5.738095238095238,
      "grad_norm": 0.01642240583896637,
      "learning_rate": 5e-05,
      "loss": 0.0051,
      "step": 2410
    },
    {
      "epoch": 5.761904761904762,
      "grad_norm": 0.027454275637865067,
      "learning_rate": 5e-05,
      "loss": 0.0048,
      "step": 2420
    },
    {
      "epoch": 5.785714285714286,
      "grad_norm": 0.03252597153186798,
      "learning_rate": 5e-05,
      "loss": 0.0051,
      "step": 2430
    },
    {
      "epoch": 5.809523809523809,
      "grad_norm": 0.031316157430410385,
      "learning_rate": 5e-05,
      "loss": 0.0045,
      "step": 2440
    },
    {
      "epoch": 5.833333333333333,
      "grad_norm": 0.022797273471951485,
      "learning_rate": 5e-05,
      "loss": 0.0057,
      "step": 2450
    },
    {
      "epoch": 5.857142857142857,
      "grad_norm": 0.03269220516085625,
      "learning_rate": 5e-05,
      "loss": 0.0061,
      "step": 2460
    },
    {
      "epoch": 5.880952380952381,
      "grad_norm": 0.02592204324901104,
      "learning_rate": 5e-05,
      "loss": 0.0063,
      "step": 2470
    },
    {
      "epoch": 5.904761904761905,
      "grad_norm": 0.020954346284270287,
      "learning_rate": 5e-05,
      "loss": 0.005,
      "step": 2480
    },
    {
      "epoch": 5.928571428571429,
      "grad_norm": 0.03389604389667511,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 2490
    },
    {
      "epoch": 5.9523809523809526,
      "grad_norm": 0.022789567708969116,
      "learning_rate": 5e-05,
      "loss": 0.005,
      "step": 2500
    },
    {
      "epoch": 5.976190476190476,
      "grad_norm": 0.021143564954400063,
      "learning_rate": 5e-05,
      "loss": 0.0052,
      "step": 2510
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.01593031920492649,
      "learning_rate": 5e-05,
      "loss": 0.0048,
      "step": 2520
    },
    {
      "epoch": 6.023809523809524,
      "grad_norm": 0.024208232760429382,
      "learning_rate": 5e-05,
      "loss": 0.0054,
      "step": 2530
    },
    {
      "epoch": 6.0476190476190474,
      "grad_norm": 0.020461222156882286,
      "learning_rate": 5e-05,
      "loss": 0.0052,
      "step": 2540
    },
    {
      "epoch": 6.071428571428571,
      "grad_norm": 0.027278559282422066,
      "learning_rate": 5e-05,
      "loss": 0.0057,
      "step": 2550
    },
    {
      "epoch": 6.095238095238095,
      "grad_norm": 0.02183406613767147,
      "learning_rate": 5e-05,
      "loss": 0.005,
      "step": 2560
    },
    {
      "epoch": 6.119047619047619,
      "grad_norm": 0.024204745888710022,
      "learning_rate": 5e-05,
      "loss": 0.0047,
      "step": 2570
    },
    {
      "epoch": 6.142857142857143,
      "grad_norm": 0.020112114027142525,
      "learning_rate": 5e-05,
      "loss": 0.0042,
      "step": 2580
    },
    {
      "epoch": 6.166666666666667,
      "grad_norm": 0.03133062273263931,
      "learning_rate": 5e-05,
      "loss": 0.0043,
      "step": 2590
    },
    {
      "epoch": 6.190476190476191,
      "grad_norm": 0.016416160389780998,
      "learning_rate": 5e-05,
      "loss": 0.0051,
      "step": 2600
    },
    {
      "epoch": 6.214285714285714,
      "grad_norm": 0.02232821099460125,
      "learning_rate": 5e-05,
      "loss": 0.0049,
      "step": 2610
    },
    {
      "epoch": 6.238095238095238,
      "grad_norm": 0.023397671058773994,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 2620
    },
    {
      "epoch": 6.261904761904762,
      "grad_norm": 0.021031634882092476,
      "learning_rate": 5e-05,
      "loss": 0.0048,
      "step": 2630
    },
    {
      "epoch": 6.285714285714286,
      "grad_norm": 0.05748249217867851,
      "learning_rate": 5e-05,
      "loss": 0.0043,
      "step": 2640
    },
    {
      "epoch": 6.309523809523809,
      "grad_norm": 0.030488435178995132,
      "learning_rate": 5e-05,
      "loss": 0.0039,
      "step": 2650
    },
    {
      "epoch": 6.333333333333333,
      "grad_norm": 0.019017908722162247,
      "learning_rate": 5e-05,
      "loss": 0.0041,
      "step": 2660
    },
    {
      "epoch": 6.357142857142857,
      "grad_norm": 0.021291224285960197,
      "learning_rate": 5e-05,
      "loss": 0.0046,
      "step": 2670
    },
    {
      "epoch": 6.380952380952381,
      "grad_norm": 0.02494419366121292,
      "learning_rate": 5e-05,
      "loss": 0.0047,
      "step": 2680
    },
    {
      "epoch": 6.404761904761905,
      "grad_norm": 0.02740258164703846,
      "learning_rate": 5e-05,
      "loss": 0.0051,
      "step": 2690
    },
    {
      "epoch": 6.428571428571429,
      "grad_norm": 0.02902950905263424,
      "learning_rate": 5e-05,
      "loss": 0.0046,
      "step": 2700
    },
    {
      "epoch": 6.4523809523809526,
      "grad_norm": 0.020637964829802513,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 2710
    },
    {
      "epoch": 6.476190476190476,
      "grad_norm": 0.020837081596255302,
      "learning_rate": 5e-05,
      "loss": 0.0038,
      "step": 2720
    },
    {
      "epoch": 6.5,
      "grad_norm": 0.020413970574736595,
      "learning_rate": 5e-05,
      "loss": 0.0035,
      "step": 2730
    },
    {
      "epoch": 6.523809523809524,
      "grad_norm": 0.01466563530266285,
      "learning_rate": 5e-05,
      "loss": 0.004,
      "step": 2740
    },
    {
      "epoch": 6.5476190476190474,
      "grad_norm": 0.02238455042243004,
      "learning_rate": 5e-05,
      "loss": 0.0035,
      "step": 2750
    },
    {
      "epoch": 6.571428571428571,
      "grad_norm": 0.027074338868260384,
      "learning_rate": 5e-05,
      "loss": 0.0038,
      "step": 2760
    },
    {
      "epoch": 6.595238095238095,
      "grad_norm": 0.03330984711647034,
      "learning_rate": 5e-05,
      "loss": 0.0042,
      "step": 2770
    },
    {
      "epoch": 6.619047619047619,
      "grad_norm": 0.025337036699056625,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 2780
    },
    {
      "epoch": 6.642857142857143,
      "grad_norm": 0.034126244485378265,
      "learning_rate": 5e-05,
      "loss": 0.0047,
      "step": 2790
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.02013178914785385,
      "learning_rate": 5e-05,
      "loss": 0.0053,
      "step": 2800
    },
    {
      "epoch": 6.690476190476191,
      "grad_norm": 0.012780356220901012,
      "learning_rate": 5e-05,
      "loss": 0.0041,
      "step": 2810
    },
    {
      "epoch": 6.714285714285714,
      "grad_norm": 0.01535839308053255,
      "learning_rate": 5e-05,
      "loss": 0.0042,
      "step": 2820
    },
    {
      "epoch": 6.738095238095238,
      "grad_norm": 0.018024824559688568,
      "learning_rate": 5e-05,
      "loss": 0.0039,
      "step": 2830
    },
    {
      "epoch": 6.761904761904762,
      "grad_norm": 0.02146134339272976,
      "learning_rate": 5e-05,
      "loss": 0.0041,
      "step": 2840
    },
    {
      "epoch": 6.785714285714286,
      "grad_norm": 0.015317169018089771,
      "learning_rate": 5e-05,
      "loss": 0.0034,
      "step": 2850
    },
    {
      "epoch": 6.809523809523809,
      "grad_norm": 0.029330266639590263,
      "learning_rate": 5e-05,
      "loss": 0.0037,
      "step": 2860
    },
    {
      "epoch": 6.833333333333333,
      "grad_norm": 0.032067060470581055,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 2870
    },
    {
      "epoch": 6.857142857142857,
      "grad_norm": 0.016328303143382072,
      "learning_rate": 5e-05,
      "loss": 0.0045,
      "step": 2880
    },
    {
      "epoch": 6.880952380952381,
      "grad_norm": 0.021490612998604774,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 2890
    },
    {
      "epoch": 6.904761904761905,
      "grad_norm": 0.014762201346457005,
      "learning_rate": 5e-05,
      "loss": 0.0037,
      "step": 2900
    },
    {
      "epoch": 6.928571428571429,
      "grad_norm": 0.02523043379187584,
      "learning_rate": 5e-05,
      "loss": 0.0035,
      "step": 2910
    },
    {
      "epoch": 6.9523809523809526,
      "grad_norm": 0.01821116730570793,
      "learning_rate": 5e-05,
      "loss": 0.0037,
      "step": 2920
    },
    {
      "epoch": 6.976190476190476,
      "grad_norm": 0.025133155286312103,
      "learning_rate": 5e-05,
      "loss": 0.0039,
      "step": 2930
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.01947784051299095,
      "learning_rate": 5e-05,
      "loss": 0.004,
      "step": 2940
    },
    {
      "epoch": 7.023809523809524,
      "grad_norm": 0.019346952438354492,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 2950
    },
    {
      "epoch": 7.0476190476190474,
      "grad_norm": 0.02163790725171566,
      "learning_rate": 5e-05,
      "loss": 0.0038,
      "step": 2960
    },
    {
      "epoch": 7.071428571428571,
      "grad_norm": 0.015155022032558918,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 2970
    },
    {
      "epoch": 7.095238095238095,
      "grad_norm": 0.02857312373816967,
      "learning_rate": 5e-05,
      "loss": 0.004,
      "step": 2980
    },
    {
      "epoch": 7.119047619047619,
      "grad_norm": 0.015818042680621147,
      "learning_rate": 5e-05,
      "loss": 0.0036,
      "step": 2990
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 0.01129070483148098,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3000
    },
    {
      "epoch": 7.166666666666667,
      "grad_norm": 0.040307771414518356,
      "learning_rate": 5e-05,
      "loss": 0.0037,
      "step": 3010
    },
    {
      "epoch": 7.190476190476191,
      "grad_norm": 0.02898922935128212,
      "learning_rate": 5e-05,
      "loss": 0.0036,
      "step": 3020
    },
    {
      "epoch": 7.214285714285714,
      "grad_norm": 0.022514350712299347,
      "learning_rate": 5e-05,
      "loss": 0.0037,
      "step": 3030
    },
    {
      "epoch": 7.238095238095238,
      "grad_norm": 0.013335715979337692,
      "learning_rate": 5e-05,
      "loss": 0.0035,
      "step": 3040
    },
    {
      "epoch": 7.261904761904762,
      "grad_norm": 0.031238842755556107,
      "learning_rate": 5e-05,
      "loss": 0.0039,
      "step": 3050
    },
    {
      "epoch": 7.285714285714286,
      "grad_norm": 0.025638382881879807,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3060
    },
    {
      "epoch": 7.309523809523809,
      "grad_norm": 0.046518560498952866,
      "learning_rate": 5e-05,
      "loss": 0.003,
      "step": 3070
    },
    {
      "epoch": 7.333333333333333,
      "grad_norm": 0.024228127673268318,
      "learning_rate": 5e-05,
      "loss": 0.0038,
      "step": 3080
    },
    {
      "epoch": 7.357142857142857,
      "grad_norm": 0.02967258170247078,
      "learning_rate": 5e-05,
      "loss": 0.003,
      "step": 3090
    },
    {
      "epoch": 7.380952380952381,
      "grad_norm": 0.019939076155424118,
      "learning_rate": 5e-05,
      "loss": 0.004,
      "step": 3100
    },
    {
      "epoch": 7.404761904761905,
      "grad_norm": 0.02764073945581913,
      "learning_rate": 5e-05,
      "loss": 0.0037,
      "step": 3110
    },
    {
      "epoch": 7.428571428571429,
      "grad_norm": 0.02370736561715603,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3120
    },
    {
      "epoch": 7.4523809523809526,
      "grad_norm": 0.020530644804239273,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3130
    },
    {
      "epoch": 7.476190476190476,
      "grad_norm": 0.019800107926130295,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3140
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.018785228952765465,
      "learning_rate": 5e-05,
      "loss": 0.0029,
      "step": 3150
    },
    {
      "epoch": 7.523809523809524,
      "grad_norm": 0.02633030340075493,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3160
    },
    {
      "epoch": 7.5476190476190474,
      "grad_norm": 0.08342000097036362,
      "learning_rate": 5e-05,
      "loss": 0.0031,
      "step": 3170
    },
    {
      "epoch": 7.571428571428571,
      "grad_norm": 0.008880950510501862,
      "learning_rate": 5e-05,
      "loss": 0.0026,
      "step": 3180
    },
    {
      "epoch": 7.595238095238095,
      "grad_norm": 0.02733513154089451,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3190
    },
    {
      "epoch": 7.619047619047619,
      "grad_norm": 0.03516070544719696,
      "learning_rate": 5e-05,
      "loss": 0.0032,
      "step": 3200
    },
    {
      "epoch": 7.642857142857143,
      "grad_norm": 0.03665764257311821,
      "learning_rate": 5e-05,
      "loss": 0.0035,
      "step": 3210
    },
    {
      "epoch": 7.666666666666667,
      "grad_norm": 0.03748982027173042,
      "learning_rate": 5e-05,
      "loss": 0.0039,
      "step": 3220
    },
    {
      "epoch": 7.690476190476191,
      "grad_norm": 0.014199526980519295,
      "learning_rate": 5e-05,
      "loss": 0.0036,
      "step": 3230
    },
    {
      "epoch": 7.714285714285714,
      "grad_norm": 0.01877681352198124,
      "learning_rate": 5e-05,
      "loss": 0.0036,
      "step": 3240
    },
    {
      "epoch": 7.738095238095238,
      "grad_norm": 0.013927435502409935,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3250
    },
    {
      "epoch": 7.761904761904762,
      "grad_norm": 0.012296963483095169,
      "learning_rate": 5e-05,
      "loss": 0.003,
      "step": 3260
    },
    {
      "epoch": 7.785714285714286,
      "grad_norm": 0.022298045456409454,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3270
    },
    {
      "epoch": 7.809523809523809,
      "grad_norm": 0.04318477213382721,
      "learning_rate": 5e-05,
      "loss": 0.0023,
      "step": 3280
    },
    {
      "epoch": 7.833333333333333,
      "grad_norm": 0.03864269703626633,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3290
    },
    {
      "epoch": 7.857142857142857,
      "grad_norm": 0.02180645242333412,
      "learning_rate": 5e-05,
      "loss": 0.0038,
      "step": 3300
    },
    {
      "epoch": 7.880952380952381,
      "grad_norm": 0.03274507075548172,
      "learning_rate": 5e-05,
      "loss": 0.0039,
      "step": 3310
    },
    {
      "epoch": 7.904761904761905,
      "grad_norm": 0.015056164003908634,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3320
    },
    {
      "epoch": 7.928571428571429,
      "grad_norm": 0.024232054129242897,
      "learning_rate": 5e-05,
      "loss": 0.003,
      "step": 3330
    },
    {
      "epoch": 7.9523809523809526,
      "grad_norm": 0.01742437854409218,
      "learning_rate": 5e-05,
      "loss": 0.0031,
      "step": 3340
    },
    {
      "epoch": 7.976190476190476,
      "grad_norm": 0.024546030908823013,
      "learning_rate": 5e-05,
      "loss": 0.0031,
      "step": 3350
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.020465753972530365,
      "learning_rate": 5e-05,
      "loss": 0.0037,
      "step": 3360
    },
    {
      "epoch": 8.023809523809524,
      "grad_norm": 0.02927527017891407,
      "learning_rate": 5e-05,
      "loss": 0.0041,
      "step": 3370
    },
    {
      "epoch": 8.047619047619047,
      "grad_norm": 0.01250431314110756,
      "learning_rate": 5e-05,
      "loss": 0.0032,
      "step": 3380
    },
    {
      "epoch": 8.071428571428571,
      "grad_norm": 0.014103593304753304,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3390
    },
    {
      "epoch": 8.095238095238095,
      "grad_norm": 0.018405349925160408,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3400
    },
    {
      "epoch": 8.119047619047619,
      "grad_norm": 0.020473845303058624,
      "learning_rate": 5e-05,
      "loss": 0.0029,
      "step": 3410
    },
    {
      "epoch": 8.142857142857142,
      "grad_norm": 0.035039015114307404,
      "learning_rate": 5e-05,
      "loss": 0.0025,
      "step": 3420
    },
    {
      "epoch": 8.166666666666666,
      "grad_norm": 0.032077498733997345,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3430
    },
    {
      "epoch": 8.19047619047619,
      "grad_norm": 0.037181805819272995,
      "learning_rate": 5e-05,
      "loss": 0.003,
      "step": 3440
    },
    {
      "epoch": 8.214285714285714,
      "grad_norm": 0.018256695941090584,
      "learning_rate": 5e-05,
      "loss": 0.0029,
      "step": 3450
    },
    {
      "epoch": 8.238095238095237,
      "grad_norm": 0.01342601515352726,
      "learning_rate": 5e-05,
      "loss": 0.0034,
      "step": 3460
    },
    {
      "epoch": 8.261904761904763,
      "grad_norm": 0.03634987398982048,
      "learning_rate": 5e-05,
      "loss": 0.0032,
      "step": 3470
    },
    {
      "epoch": 8.285714285714286,
      "grad_norm": 0.050673890858888626,
      "learning_rate": 5e-05,
      "loss": 0.0021,
      "step": 3480
    },
    {
      "epoch": 8.30952380952381,
      "grad_norm": 0.013780270703136921,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 3490
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 0.028823478147387505,
      "learning_rate": 5e-05,
      "loss": 0.0027,
      "step": 3500
    },
    {
      "epoch": 8.357142857142858,
      "grad_norm": 0.024434154853224754,
      "learning_rate": 5e-05,
      "loss": 0.0031,
      "step": 3510
    },
    {
      "epoch": 8.380952380952381,
      "grad_norm": 0.025538841262459755,
      "learning_rate": 5e-05,
      "loss": 0.0024,
      "step": 3520
    },
    {
      "epoch": 8.404761904761905,
      "grad_norm": 0.02040352113544941,
      "learning_rate": 5e-05,
      "loss": 0.0029,
      "step": 3530
    },
    {
      "epoch": 8.428571428571429,
      "grad_norm": 0.031944241374731064,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 3540
    },
    {
      "epoch": 8.452380952380953,
      "grad_norm": 0.017712410539388657,
      "learning_rate": 5e-05,
      "loss": 0.0025,
      "step": 3550
    },
    {
      "epoch": 8.476190476190476,
      "grad_norm": 0.030166227370500565,
      "learning_rate": 5e-05,
      "loss": 0.0023,
      "step": 3560
    },
    {
      "epoch": 8.5,
      "grad_norm": 0.027017861604690552,
      "learning_rate": 5e-05,
      "loss": 0.0027,
      "step": 3570
    },
    {
      "epoch": 8.523809523809524,
      "grad_norm": 0.025174567475914955,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3580
    },
    {
      "epoch": 8.547619047619047,
      "grad_norm": 0.02370358444750309,
      "learning_rate": 5e-05,
      "loss": 0.0029,
      "step": 3590
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 0.019355561584234238,
      "learning_rate": 5e-05,
      "loss": 0.0029,
      "step": 3600
    },
    {
      "epoch": 8.595238095238095,
      "grad_norm": 0.07427547127008438,
      "learning_rate": 5e-05,
      "loss": 0.0027,
      "step": 3610
    },
    {
      "epoch": 8.619047619047619,
      "grad_norm": 0.011759583838284016,
      "learning_rate": 5e-05,
      "loss": 0.0021,
      "step": 3620
    },
    {
      "epoch": 8.642857142857142,
      "grad_norm": 0.02567645162343979,
      "learning_rate": 5e-05,
      "loss": 0.0031,
      "step": 3630
    },
    {
      "epoch": 8.666666666666666,
      "grad_norm": 0.02148609049618244,
      "learning_rate": 5e-05,
      "loss": 0.0044,
      "step": 3640
    },
    {
      "epoch": 8.69047619047619,
      "grad_norm": 0.021046146750450134,
      "learning_rate": 5e-05,
      "loss": 0.0036,
      "step": 3650
    },
    {
      "epoch": 8.714285714285714,
      "grad_norm": 0.022530779242515564,
      "learning_rate": 5e-05,
      "loss": 0.0041,
      "step": 3660
    },
    {
      "epoch": 8.738095238095237,
      "grad_norm": 0.008970026858150959,
      "learning_rate": 5e-05,
      "loss": 0.0034,
      "step": 3670
    },
    {
      "epoch": 8.761904761904763,
      "grad_norm": 0.018136192113161087,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3680
    },
    {
      "epoch": 8.785714285714286,
      "grad_norm": 0.015567361377179623,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 3690
    },
    {
      "epoch": 8.80952380952381,
      "grad_norm": 0.0197913758456707,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 3700
    },
    {
      "epoch": 8.833333333333334,
      "grad_norm": 0.03988071531057358,
      "learning_rate": 5e-05,
      "loss": 0.0023,
      "step": 3710
    },
    {
      "epoch": 8.857142857142858,
      "grad_norm": 0.03059978410601616,
      "learning_rate": 5e-05,
      "loss": 0.0034,
      "step": 3720
    },
    {
      "epoch": 8.880952380952381,
      "grad_norm": 0.014758502133190632,
      "learning_rate": 5e-05,
      "loss": 0.0031,
      "step": 3730
    },
    {
      "epoch": 8.904761904761905,
      "grad_norm": 0.017123445868492126,
      "learning_rate": 5e-05,
      "loss": 0.0033,
      "step": 3740
    },
    {
      "epoch": 8.928571428571429,
      "grad_norm": 0.02551579289138317,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3750
    },
    {
      "epoch": 8.952380952380953,
      "grad_norm": 0.014496712014079094,
      "learning_rate": 5e-05,
      "loss": 0.0031,
      "step": 3760
    },
    {
      "epoch": 8.976190476190476,
      "grad_norm": 0.02045934647321701,
      "learning_rate": 5e-05,
      "loss": 0.0027,
      "step": 3770
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.02440367080271244,
      "learning_rate": 5e-05,
      "loss": 0.0035,
      "step": 3780
    },
    {
      "epoch": 9.023809523809524,
      "grad_norm": 0.01835225522518158,
      "learning_rate": 5e-05,
      "loss": 0.0036,
      "step": 3790
    },
    {
      "epoch": 9.047619047619047,
      "grad_norm": 0.007453398313373327,
      "learning_rate": 5e-05,
      "loss": 0.0027,
      "step": 3800
    },
    {
      "epoch": 9.071428571428571,
      "grad_norm": 0.02707696706056595,
      "learning_rate": 5e-05,
      "loss": 0.0027,
      "step": 3810
    },
    {
      "epoch": 9.095238095238095,
      "grad_norm": 0.020089933648705482,
      "learning_rate": 5e-05,
      "loss": 0.0022,
      "step": 3820
    },
    {
      "epoch": 9.119047619047619,
      "grad_norm": 0.02362995594739914,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 3830
    },
    {
      "epoch": 9.142857142857142,
      "grad_norm": 0.00633473414927721,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 3840
    },
    {
      "epoch": 9.166666666666666,
      "grad_norm": 0.013043406419456005,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 3850
    },
    {
      "epoch": 9.19047619047619,
      "grad_norm": 0.010314088314771652,
      "learning_rate": 5e-05,
      "loss": 0.0025,
      "step": 3860
    },
    {
      "epoch": 9.214285714285714,
      "grad_norm": 0.024023326113820076,
      "learning_rate": 5e-05,
      "loss": 0.0025,
      "step": 3870
    },
    {
      "epoch": 9.238095238095237,
      "grad_norm": 0.033713992685079575,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 3880
    },
    {
      "epoch": 9.261904761904763,
      "grad_norm": 0.024672770872712135,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 3890
    },
    {
      "epoch": 9.285714285714286,
      "grad_norm": 0.017616739496588707,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 3900
    },
    {
      "epoch": 9.30952380952381,
      "grad_norm": 0.023528527468442917,
      "learning_rate": 5e-05,
      "loss": 0.0021,
      "step": 3910
    },
    {
      "epoch": 9.333333333333334,
      "grad_norm": 0.04894119128584862,
      "learning_rate": 5e-05,
      "loss": 0.0026,
      "step": 3920
    },
    {
      "epoch": 9.357142857142858,
      "grad_norm": 0.019267156720161438,
      "learning_rate": 5e-05,
      "loss": 0.0025,
      "step": 3930
    },
    {
      "epoch": 9.380952380952381,
      "grad_norm": 0.020611826330423355,
      "learning_rate": 5e-05,
      "loss": 0.0026,
      "step": 3940
    },
    {
      "epoch": 9.404761904761905,
      "grad_norm": 0.020003391429781914,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 3950
    },
    {
      "epoch": 9.428571428571429,
      "grad_norm": 0.017253611236810684,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 3960
    },
    {
      "epoch": 9.452380952380953,
      "grad_norm": 0.045779161155223846,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 3970
    },
    {
      "epoch": 9.476190476190476,
      "grad_norm": 0.024241678416728973,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 3980
    },
    {
      "epoch": 9.5,
      "grad_norm": 0.017647111788392067,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 3990
    },
    {
      "epoch": 9.523809523809524,
      "grad_norm": 0.028641676530241966,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 4000
    },
    {
      "epoch": 9.547619047619047,
      "grad_norm": 0.047451332211494446,
      "learning_rate": 5e-05,
      "loss": 0.0016,
      "step": 4010
    },
    {
      "epoch": 9.571428571428571,
      "grad_norm": 0.01619701273739338,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4020
    },
    {
      "epoch": 9.595238095238095,
      "grad_norm": 0.020237745717167854,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4030
    },
    {
      "epoch": 9.619047619047619,
      "grad_norm": 0.040860798209905624,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 4040
    },
    {
      "epoch": 9.642857142857142,
      "grad_norm": 0.01412968896329403,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 4050
    },
    {
      "epoch": 9.666666666666666,
      "grad_norm": 0.05102560669183731,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 4060
    },
    {
      "epoch": 9.69047619047619,
      "grad_norm": 0.04887241870164871,
      "learning_rate": 5e-05,
      "loss": 0.0029,
      "step": 4070
    },
    {
      "epoch": 9.714285714285714,
      "grad_norm": 0.014414262026548386,
      "learning_rate": 5e-05,
      "loss": 0.0026,
      "step": 4080
    },
    {
      "epoch": 9.738095238095237,
      "grad_norm": 0.011788238771259785,
      "learning_rate": 5e-05,
      "loss": 0.0024,
      "step": 4090
    },
    {
      "epoch": 9.761904761904763,
      "grad_norm": 0.01252928376197815,
      "learning_rate": 5e-05,
      "loss": 0.0023,
      "step": 4100
    },
    {
      "epoch": 9.785714285714286,
      "grad_norm": 0.02384638413786888,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 4110
    },
    {
      "epoch": 9.80952380952381,
      "grad_norm": 0.02536243200302124,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4120
    },
    {
      "epoch": 9.833333333333334,
      "grad_norm": 0.015807705000042915,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 4130
    },
    {
      "epoch": 9.857142857142858,
      "grad_norm": 0.01973266527056694,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4140
    },
    {
      "epoch": 9.880952380952381,
      "grad_norm": 0.030758731067180634,
      "learning_rate": 5e-05,
      "loss": 0.0025,
      "step": 4150
    },
    {
      "epoch": 9.904761904761905,
      "grad_norm": 0.01584693230688572,
      "learning_rate": 5e-05,
      "loss": 0.0024,
      "step": 4160
    },
    {
      "epoch": 9.928571428571429,
      "grad_norm": 0.014015723019838333,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 4170
    },
    {
      "epoch": 9.952380952380953,
      "grad_norm": 0.03584891930222511,
      "learning_rate": 5e-05,
      "loss": 0.0025,
      "step": 4180
    },
    {
      "epoch": 9.976190476190476,
      "grad_norm": 0.047324661165475845,
      "learning_rate": 5e-05,
      "loss": 0.0028,
      "step": 4190
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.06452364474534988,
      "learning_rate": 5e-05,
      "loss": 0.0023,
      "step": 4200
    },
    {
      "epoch": 10.023809523809524,
      "grad_norm": 0.044528111815452576,
      "learning_rate": 5e-05,
      "loss": 0.0025,
      "step": 4210
    },
    {
      "epoch": 10.047619047619047,
      "grad_norm": 0.02596374787390232,
      "learning_rate": 5e-05,
      "loss": 0.0023,
      "step": 4220
    },
    {
      "epoch": 10.071428571428571,
      "grad_norm": 0.01841367408633232,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 4230
    },
    {
      "epoch": 10.095238095238095,
      "grad_norm": 0.03294382989406586,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 4240
    },
    {
      "epoch": 10.119047619047619,
      "grad_norm": 0.026590291410684586,
      "learning_rate": 5e-05,
      "loss": 0.0021,
      "step": 4250
    },
    {
      "epoch": 10.142857142857142,
      "grad_norm": 0.019535573199391365,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 4260
    },
    {
      "epoch": 10.166666666666666,
      "grad_norm": 0.04901567101478577,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4270
    },
    {
      "epoch": 10.19047619047619,
      "grad_norm": 0.023335624486207962,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 4280
    },
    {
      "epoch": 10.214285714285714,
      "grad_norm": 0.02412569336593151,
      "learning_rate": 5e-05,
      "loss": 0.0021,
      "step": 4290
    },
    {
      "epoch": 10.238095238095237,
      "grad_norm": 0.015281541272997856,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4300
    },
    {
      "epoch": 10.261904761904763,
      "grad_norm": 0.037770889699459076,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 4310
    },
    {
      "epoch": 10.285714285714286,
      "grad_norm": 0.02363361231982708,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 4320
    },
    {
      "epoch": 10.30952380952381,
      "grad_norm": 0.019674105569720268,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4330
    },
    {
      "epoch": 10.333333333333334,
      "grad_norm": 0.02023392915725708,
      "learning_rate": 5e-05,
      "loss": 0.0032,
      "step": 4340
    },
    {
      "epoch": 10.357142857142858,
      "grad_norm": 0.01189426425844431,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 4350
    },
    {
      "epoch": 10.380952380952381,
      "grad_norm": 0.041460879147052765,
      "learning_rate": 5e-05,
      "loss": 0.0024,
      "step": 4360
    },
    {
      "epoch": 10.404761904761905,
      "grad_norm": 0.0068189771845936775,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4370
    },
    {
      "epoch": 10.428571428571429,
      "grad_norm": 0.019572466611862183,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 4380
    },
    {
      "epoch": 10.452380952380953,
      "grad_norm": 0.01345143560320139,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 4390
    },
    {
      "epoch": 10.476190476190476,
      "grad_norm": 0.009295144118368626,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 4400
    },
    {
      "epoch": 10.5,
      "grad_norm": 0.02156994678080082,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4410
    },
    {
      "epoch": 10.523809523809524,
      "grad_norm": 0.020932510495185852,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4420
    },
    {
      "epoch": 10.547619047619047,
      "grad_norm": 0.01585063897073269,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 4430
    },
    {
      "epoch": 10.571428571428571,
      "grad_norm": 0.009516189806163311,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 4440
    },
    {
      "epoch": 10.595238095238095,
      "grad_norm": 0.026152120903134346,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4450
    },
    {
      "epoch": 10.619047619047619,
      "grad_norm": 0.0138620026409626,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4460
    },
    {
      "epoch": 10.642857142857142,
      "grad_norm": 0.03089461848139763,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4470
    },
    {
      "epoch": 10.666666666666666,
      "grad_norm": 0.03294903039932251,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 4480
    },
    {
      "epoch": 10.69047619047619,
      "grad_norm": 0.00928263645619154,
      "learning_rate": 5e-05,
      "loss": 0.002,
      "step": 4490
    },
    {
      "epoch": 10.714285714285714,
      "grad_norm": 0.029545562341809273,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 4500
    },
    {
      "epoch": 10.738095238095237,
      "grad_norm": 0.009242654778063297,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4510
    },
    {
      "epoch": 10.761904761904763,
      "grad_norm": 0.014272899366915226,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 4520
    },
    {
      "epoch": 10.785714285714286,
      "grad_norm": 0.02279002219438553,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 4530
    },
    {
      "epoch": 10.80952380952381,
      "grad_norm": 0.03252875804901123,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4540
    },
    {
      "epoch": 10.833333333333334,
      "grad_norm": 0.009702946059405804,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 4550
    },
    {
      "epoch": 10.857142857142858,
      "grad_norm": 0.03491836413741112,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4560
    },
    {
      "epoch": 10.880952380952381,
      "grad_norm": 0.03947118669748306,
      "learning_rate": 5e-05,
      "loss": 0.0024,
      "step": 4570
    },
    {
      "epoch": 10.904761904761905,
      "grad_norm": 0.025778034701943398,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 4580
    },
    {
      "epoch": 10.928571428571429,
      "grad_norm": 0.02018180675804615,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4590
    },
    {
      "epoch": 10.952380952380953,
      "grad_norm": 0.005483316257596016,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4600
    },
    {
      "epoch": 10.976190476190476,
      "grad_norm": 0.017275342717766762,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4610
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.019064828753471375,
      "learning_rate": 5e-05,
      "loss": 0.0022,
      "step": 4620
    },
    {
      "epoch": 11.023809523809524,
      "grad_norm": 0.013529101386666298,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4630
    },
    {
      "epoch": 11.047619047619047,
      "grad_norm": 0.01693836972117424,
      "learning_rate": 5e-05,
      "loss": 0.0022,
      "step": 4640
    },
    {
      "epoch": 11.071428571428571,
      "grad_norm": 0.02852066047489643,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4650
    },
    {
      "epoch": 11.095238095238095,
      "grad_norm": 0.018933508545160294,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4660
    },
    {
      "epoch": 11.119047619047619,
      "grad_norm": 0.021359335631132126,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4670
    },
    {
      "epoch": 11.142857142857142,
      "grad_norm": 0.019422901794314384,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4680
    },
    {
      "epoch": 11.166666666666666,
      "grad_norm": 0.013169987127184868,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4690
    },
    {
      "epoch": 11.19047619047619,
      "grad_norm": 0.03702198714017868,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 4700
    },
    {
      "epoch": 11.214285714285714,
      "grad_norm": 0.030341317877173424,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4710
    },
    {
      "epoch": 11.238095238095237,
      "grad_norm": 0.020110273733735085,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4720
    },
    {
      "epoch": 11.261904761904763,
      "grad_norm": 0.02008882723748684,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 4730
    },
    {
      "epoch": 11.285714285714286,
      "grad_norm": 0.06150029972195625,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 4740
    },
    {
      "epoch": 11.30952380952381,
      "grad_norm": 0.0267238300293684,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 4750
    },
    {
      "epoch": 11.333333333333334,
      "grad_norm": 0.008372998796403408,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 4760
    },
    {
      "epoch": 11.357142857142858,
      "grad_norm": 0.03507944568991661,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4770
    },
    {
      "epoch": 11.380952380952381,
      "grad_norm": 0.018310125917196274,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 4780
    },
    {
      "epoch": 11.404761904761905,
      "grad_norm": 0.009987925179302692,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 4790
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 0.0077149407006800175,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4800
    },
    {
      "epoch": 11.452380952380953,
      "grad_norm": 0.013563682325184345,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 4810
    },
    {
      "epoch": 11.476190476190476,
      "grad_norm": 0.026644017547369003,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 4820
    },
    {
      "epoch": 11.5,
      "grad_norm": 0.00553143722936511,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 4830
    },
    {
      "epoch": 11.523809523809524,
      "grad_norm": 0.014055385254323483,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 4840
    },
    {
      "epoch": 11.547619047619047,
      "grad_norm": 0.02219008095562458,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 4850
    },
    {
      "epoch": 11.571428571428571,
      "grad_norm": 0.021185655146837234,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 4860
    },
    {
      "epoch": 11.595238095238095,
      "grad_norm": 0.012521096505224705,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 4870
    },
    {
      "epoch": 11.619047619047619,
      "grad_norm": 0.027117032557725906,
      "learning_rate": 5e-05,
      "loss": 0.0021,
      "step": 4880
    },
    {
      "epoch": 11.642857142857142,
      "grad_norm": 0.03251611068844795,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 4890
    },
    {
      "epoch": 11.666666666666666,
      "grad_norm": 0.014405048452317715,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 4900
    },
    {
      "epoch": 11.69047619047619,
      "grad_norm": 0.037754759192466736,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 4910
    },
    {
      "epoch": 11.714285714285714,
      "grad_norm": 0.008099643513560295,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 4920
    },
    {
      "epoch": 11.738095238095237,
      "grad_norm": 0.01963508315384388,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4930
    },
    {
      "epoch": 11.761904761904763,
      "grad_norm": 0.017585638910531998,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 4940
    },
    {
      "epoch": 11.785714285714286,
      "grad_norm": 0.030090732499957085,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 4950
    },
    {
      "epoch": 11.80952380952381,
      "grad_norm": 0.018853867426514626,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 4960
    },
    {
      "epoch": 11.833333333333334,
      "grad_norm": 0.009620056487619877,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 4970
    },
    {
      "epoch": 11.857142857142858,
      "grad_norm": 0.05379125103354454,
      "learning_rate": 5e-05,
      "loss": 0.0016,
      "step": 4980
    },
    {
      "epoch": 11.880952380952381,
      "grad_norm": 0.022303573787212372,
      "learning_rate": 5e-05,
      "loss": 0.0019,
      "step": 4990
    },
    {
      "epoch": 11.904761904761905,
      "grad_norm": 0.008755918592214584,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5000
    },
    {
      "epoch": 11.928571428571429,
      "grad_norm": 0.009491810575127602,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 5010
    },
    {
      "epoch": 11.952380952380953,
      "grad_norm": 0.02040296606719494,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 5020
    },
    {
      "epoch": 11.976190476190476,
      "grad_norm": 0.012556660920381546,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5030
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.008164563216269016,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5040
    },
    {
      "epoch": 12.023809523809524,
      "grad_norm": 0.02336413785815239,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 5050
    },
    {
      "epoch": 12.047619047619047,
      "grad_norm": 0.0061961328610777855,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5060
    },
    {
      "epoch": 12.071428571428571,
      "grad_norm": 0.012313502840697765,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5070
    },
    {
      "epoch": 12.095238095238095,
      "grad_norm": 0.010624333284795284,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5080
    },
    {
      "epoch": 12.119047619047619,
      "grad_norm": 0.01863931119441986,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5090
    },
    {
      "epoch": 12.142857142857142,
      "grad_norm": 0.01563103497028351,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 5100
    },
    {
      "epoch": 12.166666666666666,
      "grad_norm": 0.014355426654219627,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5110
    },
    {
      "epoch": 12.19047619047619,
      "grad_norm": 0.011460881680250168,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5120
    },
    {
      "epoch": 12.214285714285714,
      "grad_norm": 0.01539954636245966,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5130
    },
    {
      "epoch": 12.238095238095237,
      "grad_norm": 0.022988399490714073,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 5140
    },
    {
      "epoch": 12.261904761904763,
      "grad_norm": 0.04906224086880684,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5150
    },
    {
      "epoch": 12.285714285714286,
      "grad_norm": 0.020332837477326393,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5160
    },
    {
      "epoch": 12.30952380952381,
      "grad_norm": 0.008848396129906178,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 5170
    },
    {
      "epoch": 12.333333333333334,
      "grad_norm": 0.016721338033676147,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 5180
    },
    {
      "epoch": 12.357142857142858,
      "grad_norm": 0.024787932634353638,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5190
    },
    {
      "epoch": 12.380952380952381,
      "grad_norm": 0.016609011217951775,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5200
    },
    {
      "epoch": 12.404761904761905,
      "grad_norm": 0.02383277751505375,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5210
    },
    {
      "epoch": 12.428571428571429,
      "grad_norm": 0.009788766503334045,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5220
    },
    {
      "epoch": 12.452380952380953,
      "grad_norm": 0.008273842744529247,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5230
    },
    {
      "epoch": 12.476190476190476,
      "grad_norm": 0.0200930405408144,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 5240
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.004213349893689156,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5250
    },
    {
      "epoch": 12.523809523809524,
      "grad_norm": 0.010746212676167488,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5260
    },
    {
      "epoch": 12.547619047619047,
      "grad_norm": 0.009187485091388226,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 5270
    },
    {
      "epoch": 12.571428571428571,
      "grad_norm": 0.011904279701411724,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 5280
    },
    {
      "epoch": 12.595238095238095,
      "grad_norm": 0.011605814099311829,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5290
    },
    {
      "epoch": 12.619047619047619,
      "grad_norm": 0.0053769913502037525,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 5300
    },
    {
      "epoch": 12.642857142857142,
      "grad_norm": 0.028224363923072815,
      "learning_rate": 5e-05,
      "loss": 0.0015,
      "step": 5310
    },
    {
      "epoch": 12.666666666666666,
      "grad_norm": 0.009584328159689903,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5320
    },
    {
      "epoch": 12.69047619047619,
      "grad_norm": 0.012837843038141727,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5330
    },
    {
      "epoch": 12.714285714285714,
      "grad_norm": 0.013618317432701588,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5340
    },
    {
      "epoch": 12.738095238095237,
      "grad_norm": 0.010568618774414062,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5350
    },
    {
      "epoch": 12.761904761904763,
      "grad_norm": 0.015289129689335823,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 5360
    },
    {
      "epoch": 12.785714285714286,
      "grad_norm": 0.015249863266944885,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5370
    },
    {
      "epoch": 12.80952380952381,
      "grad_norm": 0.012764991261065006,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5380
    },
    {
      "epoch": 12.833333333333334,
      "grad_norm": 0.018289143219590187,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5390
    },
    {
      "epoch": 12.857142857142858,
      "grad_norm": 0.006316890940070152,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5400
    },
    {
      "epoch": 12.880952380952381,
      "grad_norm": 0.017579294741153717,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5410
    },
    {
      "epoch": 12.904761904761905,
      "grad_norm": 0.01720363274216652,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5420
    },
    {
      "epoch": 12.928571428571429,
      "grad_norm": 0.012609842233359814,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5430
    },
    {
      "epoch": 12.952380952380953,
      "grad_norm": 0.006522722542285919,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5440
    },
    {
      "epoch": 12.976190476190476,
      "grad_norm": 0.015605674125254154,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5450
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.009054956026375294,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5460
    },
    {
      "epoch": 13.023809523809524,
      "grad_norm": 0.016910897567868233,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5470
    },
    {
      "epoch": 13.047619047619047,
      "grad_norm": 0.012559573166072369,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5480
    },
    {
      "epoch": 13.071428571428571,
      "grad_norm": 0.01342256274074316,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5490
    },
    {
      "epoch": 13.095238095238095,
      "grad_norm": 0.008237732574343681,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5500
    },
    {
      "epoch": 13.119047619047619,
      "grad_norm": 0.01400575041770935,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5510
    },
    {
      "epoch": 13.142857142857142,
      "grad_norm": 0.019757332280278206,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5520
    },
    {
      "epoch": 13.166666666666666,
      "grad_norm": 0.018210873007774353,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5530
    },
    {
      "epoch": 13.19047619047619,
      "grad_norm": 0.0070494236424565315,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5540
    },
    {
      "epoch": 13.214285714285714,
      "grad_norm": 0.0030856772791594267,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5550
    },
    {
      "epoch": 13.238095238095237,
      "grad_norm": 0.0075521948747336864,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5560
    },
    {
      "epoch": 13.261904761904763,
      "grad_norm": 0.01915762387216091,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5570
    },
    {
      "epoch": 13.285714285714286,
      "grad_norm": 0.007310226559638977,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5580
    },
    {
      "epoch": 13.30952380952381,
      "grad_norm": 0.023431280627846718,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5590
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 0.0055434429086744785,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5600
    },
    {
      "epoch": 13.357142857142858,
      "grad_norm": 0.051787495613098145,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5610
    },
    {
      "epoch": 13.380952380952381,
      "grad_norm": 0.010900055058300495,
      "learning_rate": 5e-05,
      "loss": 0.0017,
      "step": 5620
    },
    {
      "epoch": 13.404761904761905,
      "grad_norm": 0.01772812381386757,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5630
    },
    {
      "epoch": 13.428571428571429,
      "grad_norm": 0.003741204971447587,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5640
    },
    {
      "epoch": 13.452380952380953,
      "grad_norm": 0.009689460508525372,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5650
    },
    {
      "epoch": 13.476190476190476,
      "grad_norm": 0.004325892776250839,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 5660
    },
    {
      "epoch": 13.5,
      "grad_norm": 0.01864994689822197,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 5670
    },
    {
      "epoch": 13.523809523809524,
      "grad_norm": 0.011471171863377094,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5680
    },
    {
      "epoch": 13.547619047619047,
      "grad_norm": 0.022726034745573997,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 5690
    },
    {
      "epoch": 13.571428571428571,
      "grad_norm": 0.0025451872497797012,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 5700
    },
    {
      "epoch": 13.595238095238095,
      "grad_norm": 0.0019215985666960478,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 5710
    },
    {
      "epoch": 13.619047619047619,
      "grad_norm": 0.03238370642066002,
      "learning_rate": 5e-05,
      "loss": 0.0012,
      "step": 5720
    },
    {
      "epoch": 13.642857142857142,
      "grad_norm": 0.042818158864974976,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 5730
    },
    {
      "epoch": 13.666666666666666,
      "grad_norm": 0.019736241549253464,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 5740
    },
    {
      "epoch": 13.69047619047619,
      "grad_norm": 0.009512594901025295,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 5750
    },
    {
      "epoch": 13.714285714285714,
      "grad_norm": 0.01680351048707962,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5760
    },
    {
      "epoch": 13.738095238095237,
      "grad_norm": 0.02108100987970829,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5770
    },
    {
      "epoch": 13.761904761904763,
      "grad_norm": 0.01012613344937563,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5780
    },
    {
      "epoch": 13.785714285714286,
      "grad_norm": 0.012497921474277973,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 5790
    },
    {
      "epoch": 13.80952380952381,
      "grad_norm": 0.006895515602082014,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 5800
    },
    {
      "epoch": 13.833333333333334,
      "grad_norm": 0.026326004415750504,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5810
    },
    {
      "epoch": 13.857142857142858,
      "grad_norm": 0.02858036197721958,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5820
    },
    {
      "epoch": 13.880952380952381,
      "grad_norm": 0.014894322492182255,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 5830
    },
    {
      "epoch": 13.904761904761905,
      "grad_norm": 0.006153046153485775,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5840
    },
    {
      "epoch": 13.928571428571429,
      "grad_norm": 0.02046913094818592,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5850
    },
    {
      "epoch": 13.952380952380953,
      "grad_norm": 0.05420629680156708,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5860
    },
    {
      "epoch": 13.976190476190476,
      "grad_norm": 0.01419808715581894,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 5870
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.00912909209728241,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5880
    },
    {
      "epoch": 14.023809523809524,
      "grad_norm": 0.017824770882725716,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 5890
    },
    {
      "epoch": 14.047619047619047,
      "grad_norm": 0.008159978315234184,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5900
    },
    {
      "epoch": 14.071428571428571,
      "grad_norm": 0.02015644870698452,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 5910
    },
    {
      "epoch": 14.095238095238095,
      "grad_norm": 0.020154140889644623,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5920
    },
    {
      "epoch": 14.119047619047619,
      "grad_norm": 0.008028664626181126,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 5930
    },
    {
      "epoch": 14.142857142857142,
      "grad_norm": 0.01114033441990614,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 5940
    },
    {
      "epoch": 14.166666666666666,
      "grad_norm": 0.009640121832489967,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 5950
    },
    {
      "epoch": 14.19047619047619,
      "grad_norm": 0.009514660574495792,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 5960
    },
    {
      "epoch": 14.214285714285714,
      "grad_norm": 0.010505247861146927,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 5970
    },
    {
      "epoch": 14.238095238095237,
      "grad_norm": 0.0075341821648180485,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 5980
    },
    {
      "epoch": 14.261904761904763,
      "grad_norm": 0.007736085448414087,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 5990
    },
    {
      "epoch": 14.285714285714286,
      "grad_norm": 0.006500266492366791,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6000
    },
    {
      "epoch": 14.30952380952381,
      "grad_norm": 0.003755223471671343,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6010
    },
    {
      "epoch": 14.333333333333334,
      "grad_norm": 0.03438568115234375,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 6020
    },
    {
      "epoch": 14.357142857142858,
      "grad_norm": 0.003875867696478963,
      "learning_rate": 5e-05,
      "loss": 0.0002,
      "step": 6030
    },
    {
      "epoch": 14.380952380952381,
      "grad_norm": 0.041189443320035934,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6040
    },
    {
      "epoch": 14.404761904761905,
      "grad_norm": 0.044477108865976334,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 6050
    },
    {
      "epoch": 14.428571428571429,
      "grad_norm": 0.021848198026418686,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 6060
    },
    {
      "epoch": 14.452380952380953,
      "grad_norm": 0.009198046289384365,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6070
    },
    {
      "epoch": 14.476190476190476,
      "grad_norm": 0.00906557496637106,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6080
    },
    {
      "epoch": 14.5,
      "grad_norm": 0.027908943593502045,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6090
    },
    {
      "epoch": 14.523809523809524,
      "grad_norm": 0.010970670729875565,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6100
    },
    {
      "epoch": 14.547619047619047,
      "grad_norm": 0.02569267526268959,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6110
    },
    {
      "epoch": 14.571428571428571,
      "grad_norm": 0.00839224737137556,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 6120
    },
    {
      "epoch": 14.595238095238095,
      "grad_norm": 0.00571605795994401,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6130
    },
    {
      "epoch": 14.619047619047619,
      "grad_norm": 0.01305799838155508,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6140
    },
    {
      "epoch": 14.642857142857142,
      "grad_norm": 0.033176738768815994,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6150
    },
    {
      "epoch": 14.666666666666666,
      "grad_norm": 0.08488498628139496,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 6160
    },
    {
      "epoch": 14.69047619047619,
      "grad_norm": 0.008058450184762478,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6170
    },
    {
      "epoch": 14.714285714285714,
      "grad_norm": 0.028126804158091545,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 6180
    },
    {
      "epoch": 14.738095238095237,
      "grad_norm": 0.01536962017416954,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6190
    },
    {
      "epoch": 14.761904761904763,
      "grad_norm": 0.007554737385362387,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 6200
    },
    {
      "epoch": 14.785714285714286,
      "grad_norm": 0.009313613176345825,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 6210
    },
    {
      "epoch": 14.80952380952381,
      "grad_norm": 0.03818364813923836,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6220
    },
    {
      "epoch": 14.833333333333334,
      "grad_norm": 0.004172751680016518,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6230
    },
    {
      "epoch": 14.857142857142858,
      "grad_norm": 0.02367938496172428,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 6240
    },
    {
      "epoch": 14.880952380952381,
      "grad_norm": 0.029346376657485962,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 6250
    },
    {
      "epoch": 14.904761904761905,
      "grad_norm": 0.008894797414541245,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6260
    },
    {
      "epoch": 14.928571428571429,
      "grad_norm": 0.0210129227489233,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6270
    },
    {
      "epoch": 14.952380952380953,
      "grad_norm": 0.006997206248342991,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6280
    },
    {
      "epoch": 14.976190476190476,
      "grad_norm": 0.027477620169520378,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 6290
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.01882118359208107,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 6300
    },
    {
      "epoch": 15.023809523809524,
      "grad_norm": 0.004850170575082302,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6310
    },
    {
      "epoch": 15.047619047619047,
      "grad_norm": 0.03450813889503479,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 6320
    },
    {
      "epoch": 15.071428571428571,
      "grad_norm": 0.012980910018086433,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6330
    },
    {
      "epoch": 15.095238095238095,
      "grad_norm": 0.01641819067299366,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6340
    },
    {
      "epoch": 15.119047619047619,
      "grad_norm": 0.0036320008803159,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6350
    },
    {
      "epoch": 15.142857142857142,
      "grad_norm": 0.005455750040709972,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6360
    },
    {
      "epoch": 15.166666666666666,
      "grad_norm": 0.007276155520230532,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6370
    },
    {
      "epoch": 15.19047619047619,
      "grad_norm": 0.02196861244738102,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 6380
    },
    {
      "epoch": 15.214285714285714,
      "grad_norm": 0.004664530977606773,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 6390
    },
    {
      "epoch": 15.238095238095237,
      "grad_norm": 0.008527121506631374,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6400
    },
    {
      "epoch": 15.261904761904763,
      "grad_norm": 0.015898073092103004,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6410
    },
    {
      "epoch": 15.285714285714286,
      "grad_norm": 0.007485649082809687,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6420
    },
    {
      "epoch": 15.30952380952381,
      "grad_norm": 0.011031672358512878,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6430
    },
    {
      "epoch": 15.333333333333334,
      "grad_norm": 0.017634835094213486,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6440
    },
    {
      "epoch": 15.357142857142858,
      "grad_norm": 0.010040491819381714,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6450
    },
    {
      "epoch": 15.380952380952381,
      "grad_norm": 0.00803917832672596,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6460
    },
    {
      "epoch": 15.404761904761905,
      "grad_norm": 0.002537789987400174,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6470
    },
    {
      "epoch": 15.428571428571429,
      "grad_norm": 0.004426058381795883,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 6480
    },
    {
      "epoch": 15.452380952380953,
      "grad_norm": 0.004302819259464741,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6490
    },
    {
      "epoch": 15.476190476190476,
      "grad_norm": 0.006018012762069702,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6500
    },
    {
      "epoch": 15.5,
      "grad_norm": 0.00840548425912857,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6510
    },
    {
      "epoch": 15.523809523809524,
      "grad_norm": 0.011578329838812351,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6520
    },
    {
      "epoch": 15.547619047619047,
      "grad_norm": 0.013048551976680756,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 6530
    },
    {
      "epoch": 15.571428571428571,
      "grad_norm": 0.004024627152830362,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 6540
    },
    {
      "epoch": 15.595238095238095,
      "grad_norm": 0.007286364212632179,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6550
    },
    {
      "epoch": 15.619047619047619,
      "grad_norm": 0.014821533113718033,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6560
    },
    {
      "epoch": 15.642857142857142,
      "grad_norm": 0.002453850582242012,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6570
    },
    {
      "epoch": 15.666666666666666,
      "grad_norm": 0.03152019903063774,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 6580
    },
    {
      "epoch": 15.69047619047619,
      "grad_norm": 0.046064406633377075,
      "learning_rate": 5e-05,
      "loss": 0.0014,
      "step": 6590
    },
    {
      "epoch": 15.714285714285714,
      "grad_norm": 0.019682571291923523,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6600
    },
    {
      "epoch": 15.738095238095237,
      "grad_norm": 0.0024483215529471636,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6610
    },
    {
      "epoch": 15.761904761904763,
      "grad_norm": 0.008708986453711987,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6620
    },
    {
      "epoch": 15.785714285714286,
      "grad_norm": 0.006394381169229746,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6630
    },
    {
      "epoch": 15.80952380952381,
      "grad_norm": 0.03494114428758621,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6640
    },
    {
      "epoch": 15.833333333333334,
      "grad_norm": 0.02163984254002571,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6650
    },
    {
      "epoch": 15.857142857142858,
      "grad_norm": 0.00985416304320097,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6660
    },
    {
      "epoch": 15.880952380952381,
      "grad_norm": 0.008056841790676117,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 6670
    },
    {
      "epoch": 15.904761904761905,
      "grad_norm": 0.004157788585871458,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6680
    },
    {
      "epoch": 15.928571428571429,
      "grad_norm": 0.00597544526681304,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6690
    },
    {
      "epoch": 15.952380952380953,
      "grad_norm": 0.023443806916475296,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 6700
    },
    {
      "epoch": 15.976190476190476,
      "grad_norm": 0.012778760865330696,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 6710
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.015170056372880936,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6720
    },
    {
      "epoch": 16.023809523809526,
      "grad_norm": 0.01816493459045887,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 6730
    },
    {
      "epoch": 16.047619047619047,
      "grad_norm": 0.00630978075787425,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6740
    },
    {
      "epoch": 16.071428571428573,
      "grad_norm": 0.004600324667990208,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 6750
    },
    {
      "epoch": 16.095238095238095,
      "grad_norm": 0.03235204890370369,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6760
    },
    {
      "epoch": 16.11904761904762,
      "grad_norm": 0.012307102791965008,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6770
    },
    {
      "epoch": 16.142857142857142,
      "grad_norm": 0.0052837226539850235,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6780
    },
    {
      "epoch": 16.166666666666668,
      "grad_norm": 0.02251107059419155,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6790
    },
    {
      "epoch": 16.19047619047619,
      "grad_norm": 0.011189977638423443,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6800
    },
    {
      "epoch": 16.214285714285715,
      "grad_norm": 0.05799144506454468,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 6810
    },
    {
      "epoch": 16.238095238095237,
      "grad_norm": 0.023233134299516678,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6820
    },
    {
      "epoch": 16.261904761904763,
      "grad_norm": 0.005226644221693277,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6830
    },
    {
      "epoch": 16.285714285714285,
      "grad_norm": 0.022896727547049522,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6840
    },
    {
      "epoch": 16.30952380952381,
      "grad_norm": 0.009060421027243137,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 6850
    },
    {
      "epoch": 16.333333333333332,
      "grad_norm": 0.019654234871268272,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 6860
    },
    {
      "epoch": 16.357142857142858,
      "grad_norm": 0.01879570446908474,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6870
    },
    {
      "epoch": 16.38095238095238,
      "grad_norm": 0.015871930867433548,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 6880
    },
    {
      "epoch": 16.404761904761905,
      "grad_norm": 0.005459831561893225,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6890
    },
    {
      "epoch": 16.428571428571427,
      "grad_norm": 0.002002556109800935,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 6900
    },
    {
      "epoch": 16.452380952380953,
      "grad_norm": 0.00361634767614305,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6910
    },
    {
      "epoch": 16.476190476190474,
      "grad_norm": 0.008024908602237701,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6920
    },
    {
      "epoch": 16.5,
      "grad_norm": 0.0015706607373431325,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6930
    },
    {
      "epoch": 16.523809523809526,
      "grad_norm": 0.015266403555870056,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 6940
    },
    {
      "epoch": 16.547619047619047,
      "grad_norm": 0.03144041448831558,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 6950
    },
    {
      "epoch": 16.571428571428573,
      "grad_norm": 0.0020375235471874475,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6960
    },
    {
      "epoch": 16.595238095238095,
      "grad_norm": 0.03112536109983921,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 6970
    },
    {
      "epoch": 16.61904761904762,
      "grad_norm": 0.005854692310094833,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 6980
    },
    {
      "epoch": 16.642857142857142,
      "grad_norm": 0.01317546609789133,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 6990
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 0.008511519059538841,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7000
    },
    {
      "epoch": 16.69047619047619,
      "grad_norm": 0.006545855663716793,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7010
    },
    {
      "epoch": 16.714285714285715,
      "grad_norm": 0.010415365919470787,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7020
    },
    {
      "epoch": 16.738095238095237,
      "grad_norm": 0.01626661792397499,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7030
    },
    {
      "epoch": 16.761904761904763,
      "grad_norm": 0.005767187103629112,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7040
    },
    {
      "epoch": 16.785714285714285,
      "grad_norm": 0.01930147223174572,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7050
    },
    {
      "epoch": 16.80952380952381,
      "grad_norm": 0.022985486313700676,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 7060
    },
    {
      "epoch": 16.833333333333332,
      "grad_norm": 0.015567641705274582,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7070
    },
    {
      "epoch": 16.857142857142858,
      "grad_norm": 0.023370876908302307,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7080
    },
    {
      "epoch": 16.88095238095238,
      "grad_norm": 0.007470586337149143,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7090
    },
    {
      "epoch": 16.904761904761905,
      "grad_norm": 0.006150704808533192,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7100
    },
    {
      "epoch": 16.928571428571427,
      "grad_norm": 0.0042198821902275085,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7110
    },
    {
      "epoch": 16.952380952380953,
      "grad_norm": 0.017077218741178513,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7120
    },
    {
      "epoch": 16.976190476190474,
      "grad_norm": 0.023657051846385002,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7130
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.016245121136307716,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7140
    },
    {
      "epoch": 17.023809523809526,
      "grad_norm": 0.0030179405584931374,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7150
    },
    {
      "epoch": 17.047619047619047,
      "grad_norm": 0.025457536801695824,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7160
    },
    {
      "epoch": 17.071428571428573,
      "grad_norm": 0.03168899938464165,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7170
    },
    {
      "epoch": 17.095238095238095,
      "grad_norm": 0.036503903567790985,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 7180
    },
    {
      "epoch": 17.11904761904762,
      "grad_norm": 0.0242981668561697,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 7190
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 0.014298885129392147,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7200
    },
    {
      "epoch": 17.166666666666668,
      "grad_norm": 0.02301887236535549,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7210
    },
    {
      "epoch": 17.19047619047619,
      "grad_norm": 0.03424462303519249,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 7220
    },
    {
      "epoch": 17.214285714285715,
      "grad_norm": 0.023424234241247177,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7230
    },
    {
      "epoch": 17.238095238095237,
      "grad_norm": 0.023134572431445122,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 7240
    },
    {
      "epoch": 17.261904761904763,
      "grad_norm": 0.025860372930765152,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 7250
    },
    {
      "epoch": 17.285714285714285,
      "grad_norm": 0.01248745433986187,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7260
    },
    {
      "epoch": 17.30952380952381,
      "grad_norm": 0.018757270649075508,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7270
    },
    {
      "epoch": 17.333333333333332,
      "grad_norm": 0.011798521503806114,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7280
    },
    {
      "epoch": 17.357142857142858,
      "grad_norm": 0.020112033933401108,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7290
    },
    {
      "epoch": 17.38095238095238,
      "grad_norm": 0.009371660649776459,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7300
    },
    {
      "epoch": 17.404761904761905,
      "grad_norm": 0.010503299534320831,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7310
    },
    {
      "epoch": 17.428571428571427,
      "grad_norm": 0.02050066739320755,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7320
    },
    {
      "epoch": 17.452380952380953,
      "grad_norm": 0.034579671919345856,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 7330
    },
    {
      "epoch": 17.476190476190474,
      "grad_norm": 0.02491345815360546,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7340
    },
    {
      "epoch": 17.5,
      "grad_norm": 0.0013771721860393882,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7350
    },
    {
      "epoch": 17.523809523809526,
      "grad_norm": 0.012594480067491531,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7360
    },
    {
      "epoch": 17.547619047619047,
      "grad_norm": 0.025901375338435173,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7370
    },
    {
      "epoch": 17.571428571428573,
      "grad_norm": 0.004503455013036728,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7380
    },
    {
      "epoch": 17.595238095238095,
      "grad_norm": 0.01678163930773735,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7390
    },
    {
      "epoch": 17.61904761904762,
      "grad_norm": 0.01972806081175804,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7400
    },
    {
      "epoch": 17.642857142857142,
      "grad_norm": 0.005659343674778938,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7410
    },
    {
      "epoch": 17.666666666666668,
      "grad_norm": 0.010966219939291477,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7420
    },
    {
      "epoch": 17.69047619047619,
      "grad_norm": 0.01521807536482811,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7430
    },
    {
      "epoch": 17.714285714285715,
      "grad_norm": 0.0045644501224160194,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7440
    },
    {
      "epoch": 17.738095238095237,
      "grad_norm": 0.0009657703340053558,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7450
    },
    {
      "epoch": 17.761904761904763,
      "grad_norm": 0.03967520594596863,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7460
    },
    {
      "epoch": 17.785714285714285,
      "grad_norm": 0.003580782562494278,
      "learning_rate": 5e-05,
      "loss": 0.0002,
      "step": 7470
    },
    {
      "epoch": 17.80952380952381,
      "grad_norm": 0.02645096555352211,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7480
    },
    {
      "epoch": 17.833333333333332,
      "grad_norm": 0.004722685553133488,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7490
    },
    {
      "epoch": 17.857142857142858,
      "grad_norm": 0.026108304038643837,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7500
    },
    {
      "epoch": 17.88095238095238,
      "grad_norm": 0.008777507580816746,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7510
    },
    {
      "epoch": 17.904761904761905,
      "grad_norm": 0.011482530273497105,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7520
    },
    {
      "epoch": 17.928571428571427,
      "grad_norm": 0.003905653487890959,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 7530
    },
    {
      "epoch": 17.952380952380953,
      "grad_norm": 0.020037874579429626,
      "learning_rate": 5e-05,
      "loss": 0.001,
      "step": 7540
    },
    {
      "epoch": 17.976190476190474,
      "grad_norm": 0.017041349783539772,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7550
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.0016616099746897817,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7560
    },
    {
      "epoch": 18.023809523809526,
      "grad_norm": 0.021135887131094933,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7570
    },
    {
      "epoch": 18.047619047619047,
      "grad_norm": 0.010796349495649338,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7580
    },
    {
      "epoch": 18.071428571428573,
      "grad_norm": 0.018032172694802284,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7590
    },
    {
      "epoch": 18.095238095238095,
      "grad_norm": 0.028129780665040016,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 7600
    },
    {
      "epoch": 18.11904761904762,
      "grad_norm": 0.0026430708821862936,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7610
    },
    {
      "epoch": 18.142857142857142,
      "grad_norm": 0.00862652063369751,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7620
    },
    {
      "epoch": 18.166666666666668,
      "grad_norm": 0.0017266141949221492,
      "learning_rate": 5e-05,
      "loss": 0.0002,
      "step": 7630
    },
    {
      "epoch": 18.19047619047619,
      "grad_norm": 0.002414742251858115,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7640
    },
    {
      "epoch": 18.214285714285715,
      "grad_norm": 0.002050422364845872,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7650
    },
    {
      "epoch": 18.238095238095237,
      "grad_norm": 0.0194852314889431,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7660
    },
    {
      "epoch": 18.261904761904763,
      "grad_norm": 0.01717439480125904,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7670
    },
    {
      "epoch": 18.285714285714285,
      "grad_norm": 0.011901612393558025,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7680
    },
    {
      "epoch": 18.30952380952381,
      "grad_norm": 0.029567129909992218,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 7690
    },
    {
      "epoch": 18.333333333333332,
      "grad_norm": 0.006910389289259911,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 7700
    },
    {
      "epoch": 18.357142857142858,
      "grad_norm": 0.007212989963591099,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7710
    },
    {
      "epoch": 18.38095238095238,
      "grad_norm": 0.017439018934965134,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7720
    },
    {
      "epoch": 18.404761904761905,
      "grad_norm": 0.0020984190050512552,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7730
    },
    {
      "epoch": 18.428571428571427,
      "grad_norm": 0.011482183821499348,
      "learning_rate": 5e-05,
      "loss": 0.0009,
      "step": 7740
    },
    {
      "epoch": 18.452380952380953,
      "grad_norm": 0.006130084861069918,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7750
    },
    {
      "epoch": 18.476190476190474,
      "grad_norm": 0.018064623698592186,
      "learning_rate": 5e-05,
      "loss": 0.0011,
      "step": 7760
    },
    {
      "epoch": 18.5,
      "grad_norm": 0.007949289865791798,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7770
    },
    {
      "epoch": 18.523809523809526,
      "grad_norm": 0.0036398067604750395,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7780
    },
    {
      "epoch": 18.547619047619047,
      "grad_norm": 0.021670356392860413,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7790
    },
    {
      "epoch": 18.571428571428573,
      "grad_norm": 0.0020873206667602062,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7800
    },
    {
      "epoch": 18.595238095238095,
      "grad_norm": 0.007899454794824123,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7810
    },
    {
      "epoch": 18.61904761904762,
      "grad_norm": 0.03517717495560646,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7820
    },
    {
      "epoch": 18.642857142857142,
      "grad_norm": 0.010151904076337814,
      "learning_rate": 5e-05,
      "loss": 0.0002,
      "step": 7830
    },
    {
      "epoch": 18.666666666666668,
      "grad_norm": 0.001768286689184606,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7840
    },
    {
      "epoch": 18.69047619047619,
      "grad_norm": 0.005438578315079212,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7850
    },
    {
      "epoch": 18.714285714285715,
      "grad_norm": 0.01367899402976036,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7860
    },
    {
      "epoch": 18.738095238095237,
      "grad_norm": 0.005496785510331392,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7870
    },
    {
      "epoch": 18.761904761904763,
      "grad_norm": 0.010219060815870762,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7880
    },
    {
      "epoch": 18.785714285714285,
      "grad_norm": 0.005748805124312639,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7890
    },
    {
      "epoch": 18.80952380952381,
      "grad_norm": 0.0068986667320132256,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7900
    },
    {
      "epoch": 18.833333333333332,
      "grad_norm": 0.008341987617313862,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 7910
    },
    {
      "epoch": 18.857142857142858,
      "grad_norm": 0.015715043991804123,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7920
    },
    {
      "epoch": 18.88095238095238,
      "grad_norm": 0.009235808625817299,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7930
    },
    {
      "epoch": 18.904761904761905,
      "grad_norm": 0.004791698418557644,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 7940
    },
    {
      "epoch": 18.928571428571427,
      "grad_norm": 0.020587580278515816,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 7950
    },
    {
      "epoch": 18.952380952380953,
      "grad_norm": 0.0022148280404508114,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7960
    },
    {
      "epoch": 18.976190476190474,
      "grad_norm": 0.0040055266581475735,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7970
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.0041480157524347305,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 7980
    },
    {
      "epoch": 19.023809523809526,
      "grad_norm": 0.009851422160863876,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 7990
    },
    {
      "epoch": 19.047619047619047,
      "grad_norm": 0.029928863048553467,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8000
    },
    {
      "epoch": 19.071428571428573,
      "grad_norm": 0.005271303467452526,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8010
    },
    {
      "epoch": 19.095238095238095,
      "grad_norm": 0.011121473275125027,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8020
    },
    {
      "epoch": 19.11904761904762,
      "grad_norm": 0.002328911330550909,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8030
    },
    {
      "epoch": 19.142857142857142,
      "grad_norm": 0.0005802198429591954,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8040
    },
    {
      "epoch": 19.166666666666668,
      "grad_norm": 0.003331305691972375,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8050
    },
    {
      "epoch": 19.19047619047619,
      "grad_norm": 0.0036068561021238565,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8060
    },
    {
      "epoch": 19.214285714285715,
      "grad_norm": 0.008253911510109901,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8070
    },
    {
      "epoch": 19.238095238095237,
      "grad_norm": 0.023871315643191338,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8080
    },
    {
      "epoch": 19.261904761904763,
      "grad_norm": 0.004133463837206364,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8090
    },
    {
      "epoch": 19.285714285714285,
      "grad_norm": 0.0016440710751339793,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8100
    },
    {
      "epoch": 19.30952380952381,
      "grad_norm": 0.004808527417480946,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8110
    },
    {
      "epoch": 19.333333333333332,
      "grad_norm": 0.03961317241191864,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8120
    },
    {
      "epoch": 19.357142857142858,
      "grad_norm": 0.021009011194109917,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8130
    },
    {
      "epoch": 19.38095238095238,
      "grad_norm": 0.009781712666153908,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8140
    },
    {
      "epoch": 19.404761904761905,
      "grad_norm": 0.0019496449967846274,
      "learning_rate": 5e-05,
      "loss": 0.0002,
      "step": 8150
    },
    {
      "epoch": 19.428571428571427,
      "grad_norm": 0.00960745383054018,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 8160
    },
    {
      "epoch": 19.452380952380953,
      "grad_norm": 0.008466621860861778,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8170
    },
    {
      "epoch": 19.476190476190474,
      "grad_norm": 0.014924910850822926,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 8180
    },
    {
      "epoch": 19.5,
      "grad_norm": 0.006802048999816179,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 8190
    },
    {
      "epoch": 19.523809523809526,
      "grad_norm": 0.0138113833963871,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8200
    },
    {
      "epoch": 19.547619047619047,
      "grad_norm": 0.01139379758387804,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8210
    },
    {
      "epoch": 19.571428571428573,
      "grad_norm": 0.03176883980631828,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 8220
    },
    {
      "epoch": 19.595238095238095,
      "grad_norm": 0.007066241931170225,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8230
    },
    {
      "epoch": 19.61904761904762,
      "grad_norm": 0.001747833681292832,
      "learning_rate": 5e-05,
      "loss": 0.0002,
      "step": 8240
    },
    {
      "epoch": 19.642857142857142,
      "grad_norm": 0.004946041852235794,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8250
    },
    {
      "epoch": 19.666666666666668,
      "grad_norm": 0.0037244544364511967,
      "learning_rate": 5e-05,
      "loss": 0.0002,
      "step": 8260
    },
    {
      "epoch": 19.69047619047619,
      "grad_norm": 0.007341946009546518,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8270
    },
    {
      "epoch": 19.714285714285715,
      "grad_norm": 0.002960850251838565,
      "learning_rate": 5e-05,
      "loss": 0.0007,
      "step": 8280
    },
    {
      "epoch": 19.738095238095237,
      "grad_norm": 0.0019498078618198633,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8290
    },
    {
      "epoch": 19.761904761904763,
      "grad_norm": 0.017022594809532166,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 8300
    },
    {
      "epoch": 19.785714285714285,
      "grad_norm": 0.001329585094936192,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8310
    },
    {
      "epoch": 19.80952380952381,
      "grad_norm": 0.003296843497082591,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8320
    },
    {
      "epoch": 19.833333333333332,
      "grad_norm": 0.015899814665317535,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8330
    },
    {
      "epoch": 19.857142857142858,
      "grad_norm": 0.010661507025361061,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8340
    },
    {
      "epoch": 19.88095238095238,
      "grad_norm": 0.0018181676277890801,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8350
    },
    {
      "epoch": 19.904761904761905,
      "grad_norm": 0.029869144782423973,
      "learning_rate": 5e-05,
      "loss": 0.0006,
      "step": 8360
    },
    {
      "epoch": 19.928571428571427,
      "grad_norm": 0.016890952363610268,
      "learning_rate": 5e-05,
      "loss": 0.0002,
      "step": 8370
    },
    {
      "epoch": 19.952380952380953,
      "grad_norm": 0.001414968748576939,
      "learning_rate": 5e-05,
      "loss": 0.0003,
      "step": 8380
    },
    {
      "epoch": 19.976190476190474,
      "grad_norm": 0.0023363372310996056,
      "learning_rate": 5e-05,
      "loss": 0.0004,
      "step": 8390
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.001184182707220316,
      "learning_rate": 5e-05,
      "loss": 0.0005,
      "step": 8400
    }
  ],
  "logging_steps": 10,
  "max_steps": 8400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2637663864989696e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}