{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 45.0,
  "global_step": 6300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0007936507936507937,
      "loss": 0.4795,
      "step": 10
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0015873015873015873,
      "loss": 0.4792,
      "step": 20
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.002380952380952381,
      "loss": 0.4786,
      "step": 30
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0031746031746031746,
      "loss": 0.4773,
      "step": 40
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.003968253968253968,
      "loss": 0.4743,
      "step": 50
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.004761904761904762,
      "loss": 0.4672,
      "step": 60
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.005555555555555556,
      "loss": 0.4456,
      "step": 70
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.006349206349206349,
      "loss": 0.3561,
      "step": 80
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.007142857142857143,
      "loss": 0.2459,
      "step": 90
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.007936507936507936,
      "loss": 0.2057,
      "step": 100
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00873015873015873,
      "loss": 0.1847,
      "step": 110
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.009523809523809525,
      "loss": 0.1765,
      "step": 120
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.010317460317460317,
      "loss": 0.1404,
      "step": 130
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.011111111111111112,
      "loss": 0.1251,
      "step": 140
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.011904761904761904,
      "loss": 0.1224,
      "step": 150
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.012698412698412698,
      "loss": 0.1241,
      "step": 160
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.013492063492063493,
      "loss": 0.1137,
      "step": 170
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.014285714285714285,
      "loss": 0.1178,
      "step": 180
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.01507936507936508,
      "loss": 0.1117,
      "step": 190
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.015873015873015872,
      "loss": 0.1147,
      "step": 200
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.016666666666666666,
      "loss": 0.1049,
      "step": 210
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.01746031746031746,
      "loss": 0.1153,
      "step": 220
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.018253968253968255,
      "loss": 0.1156,
      "step": 230
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.01904761904761905,
      "loss": 0.1154,
      "step": 240
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.01984126984126984,
      "loss": 0.1049,
      "step": 250
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.020634920634920634,
      "loss": 0.0982,
      "step": 260
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.02142857142857143,
      "loss": 0.1182,
      "step": 270
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.022222222222222223,
      "loss": 0.1183,
      "step": 280
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.023015873015873017,
      "loss": 0.1131,
      "step": 290
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.023809523809523808,
      "loss": 0.1063,
      "step": 300
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.024603174603174603,
      "loss": 0.0985,
      "step": 310
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.025396825396825397,
      "loss": 0.1262,
      "step": 320
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.026190476190476195,
      "loss": 0.1319,
      "step": 330
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.026984126984126985,
      "loss": 0.1123,
      "step": 340
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.02777777777777778,
      "loss": 0.1116,
      "step": 350
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.02857142857142857,
      "loss": 0.1031,
      "step": 360
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.02936507936507937,
      "loss": 0.1021,
      "step": 370
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.03015873015873016,
      "loss": 0.1067,
      "step": 380
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.030952380952380953,
      "loss": 0.1038,
      "step": 390
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.031746031746031744,
      "loss": 0.1127,
      "step": 400
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.03253968253968254,
      "loss": 0.1008,
      "step": 410
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.03333333333333333,
      "loss": 0.0965,
      "step": 420
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.03412698412698413,
      "loss": 0.0965,
      "step": 430
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.03492063492063492,
      "loss": 0.095,
      "step": 440
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.03571428571428572,
      "loss": 0.0876,
      "step": 450
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.03650793650793651,
      "loss": 0.1019,
      "step": 460
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0373015873015873,
      "loss": 0.0881,
      "step": 470
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0380952380952381,
      "loss": 0.0901,
      "step": 480
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.03888888888888889,
      "loss": 0.0878,
      "step": 490
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.03968253968253968,
      "loss": 0.0861,
      "step": 500
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.04047619047619048,
      "loss": 0.0865,
      "step": 510
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.04126984126984127,
      "loss": 0.0818,
      "step": 520
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.04206349206349207,
      "loss": 0.0789,
      "step": 530
    },
    {
      "epoch": 3.86,
      "learning_rate": 0.04285714285714286,
      "loss": 0.1004,
      "step": 540
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.043650793650793655,
      "loss": 0.0799,
      "step": 550
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.044444444444444446,
      "loss": 0.0818,
      "step": 560
    },
    {
      "epoch": 4.07,
      "learning_rate": 0.045238095238095244,
      "loss": 0.0777,
      "step": 570
    },
    {
      "epoch": 4.14,
      "learning_rate": 0.046031746031746035,
      "loss": 0.0767,
      "step": 580
    },
    {
      "epoch": 4.21,
      "learning_rate": 0.046825396825396826,
      "loss": 0.0711,
      "step": 590
    },
    {
      "epoch": 4.29,
      "learning_rate": 0.047619047619047616,
      "loss": 0.0723,
      "step": 600
    },
    {
      "epoch": 4.36,
      "learning_rate": 0.048412698412698414,
      "loss": 0.0694,
      "step": 610
    },
    {
      "epoch": 4.43,
      "learning_rate": 0.049206349206349205,
      "loss": 0.0657,
      "step": 620
    },
    {
      "epoch": 4.5,
      "learning_rate": 0.05,
      "loss": 0.0567,
      "step": 630
    },
    {
      "epoch": 4.57,
      "learning_rate": 0.049911816578483245,
      "loss": 0.0617,
      "step": 640
    },
    {
      "epoch": 4.64,
      "learning_rate": 0.049823633156966494,
      "loss": 0.0678,
      "step": 650
    },
    {
      "epoch": 4.71,
      "learning_rate": 0.04973544973544974,
      "loss": 0.0647,
      "step": 660
    },
    {
      "epoch": 4.79,
      "learning_rate": 0.049647266313932986,
      "loss": 0.0543,
      "step": 670
    },
    {
      "epoch": 4.86,
      "learning_rate": 0.04955908289241623,
      "loss": 0.0558,
      "step": 680
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.04947089947089947,
      "loss": 0.0519,
      "step": 690
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.04938271604938271,
      "loss": 0.0506,
      "step": 700
    },
    {
      "epoch": 5.07,
      "learning_rate": 0.04929453262786596,
      "loss": 0.0583,
      "step": 710
    },
    {
      "epoch": 5.14,
      "learning_rate": 0.049206349206349205,
      "loss": 0.0518,
      "step": 720
    },
    {
      "epoch": 5.21,
      "learning_rate": 0.049118165784832454,
      "loss": 0.0491,
      "step": 730
    },
    {
      "epoch": 5.29,
      "learning_rate": 0.0490299823633157,
      "loss": 0.0471,
      "step": 740
    },
    {
      "epoch": 5.36,
      "learning_rate": 0.048941798941798946,
      "loss": 0.0488,
      "step": 750
    },
    {
      "epoch": 5.43,
      "learning_rate": 0.04885361552028219,
      "loss": 0.0449,
      "step": 760
    },
    {
      "epoch": 5.5,
      "learning_rate": 0.04876543209876544,
      "loss": 0.0417,
      "step": 770
    },
    {
      "epoch": 5.57,
      "learning_rate": 0.04867724867724868,
      "loss": 0.0409,
      "step": 780
    },
    {
      "epoch": 5.64,
      "learning_rate": 0.04858906525573192,
      "loss": 0.0485,
      "step": 790
    },
    {
      "epoch": 5.71,
      "learning_rate": 0.048500881834215165,
      "loss": 0.0457,
      "step": 800
    },
    {
      "epoch": 5.79,
      "learning_rate": 0.048412698412698414,
      "loss": 0.0422,
      "step": 810
    },
    {
      "epoch": 5.86,
      "learning_rate": 0.04832451499118166,
      "loss": 0.0407,
      "step": 820
    },
    {
      "epoch": 5.93,
      "learning_rate": 0.048236331569664906,
      "loss": 0.0394,
      "step": 830
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.04814814814814815,
      "loss": 0.0417,
      "step": 840
    },
    {
      "epoch": 6.07,
      "learning_rate": 0.0480599647266314,
      "loss": 0.0358,
      "step": 850
    },
    {
      "epoch": 6.14,
      "learning_rate": 0.04797178130511464,
      "loss": 0.0352,
      "step": 860
    },
    {
      "epoch": 6.21,
      "learning_rate": 0.04788359788359789,
      "loss": 0.0389,
      "step": 870
    },
    {
      "epoch": 6.29,
      "learning_rate": 0.04779541446208113,
      "loss": 0.036,
      "step": 880
    },
    {
      "epoch": 6.36,
      "learning_rate": 0.047707231040564374,
      "loss": 0.034,
      "step": 890
    },
    {
      "epoch": 6.43,
      "learning_rate": 0.047619047619047616,
      "loss": 0.0359,
      "step": 900
    },
    {
      "epoch": 6.5,
      "learning_rate": 0.047530864197530866,
      "loss": 0.0355,
      "step": 910
    },
    {
      "epoch": 6.57,
      "learning_rate": 0.04744268077601411,
      "loss": 0.03,
      "step": 920
    },
    {
      "epoch": 6.64,
      "learning_rate": 0.04735449735449736,
      "loss": 0.0278,
      "step": 930
    },
    {
      "epoch": 6.71,
      "learning_rate": 0.0472663139329806,
      "loss": 0.0337,
      "step": 940
    },
    {
      "epoch": 6.79,
      "learning_rate": 0.04717813051146385,
      "loss": 0.0293,
      "step": 950
    },
    {
      "epoch": 6.86,
      "learning_rate": 0.04708994708994709,
      "loss": 0.0302,
      "step": 960
    },
    {
      "epoch": 6.93,
      "learning_rate": 0.04700176366843034,
      "loss": 0.0332,
      "step": 970
    },
    {
      "epoch": 7.0,
      "learning_rate": 0.04691358024691358,
      "loss": 0.0292,
      "step": 980
    },
    {
      "epoch": 7.07,
      "learning_rate": 0.046825396825396826,
      "loss": 0.0298,
      "step": 990
    },
    {
      "epoch": 7.14,
      "learning_rate": 0.04673721340388007,
      "loss": 0.0269,
      "step": 1000
    },
    {
      "epoch": 7.21,
      "learning_rate": 0.04664902998236332,
      "loss": 0.0236,
      "step": 1010
    },
    {
      "epoch": 7.29,
      "learning_rate": 0.04656084656084656,
      "loss": 0.025,
      "step": 1020
    },
    {
      "epoch": 7.36,
      "learning_rate": 0.04647266313932981,
      "loss": 0.0266,
      "step": 1030
    },
    {
      "epoch": 7.43,
      "learning_rate": 0.04638447971781305,
      "loss": 0.0229,
      "step": 1040
    },
    {
      "epoch": 7.5,
      "learning_rate": 0.0462962962962963,
      "loss": 0.0232,
      "step": 1050
    },
    {
      "epoch": 7.57,
      "learning_rate": 0.04620811287477954,
      "loss": 0.0244,
      "step": 1060
    },
    {
      "epoch": 7.64,
      "learning_rate": 0.04611992945326279,
      "loss": 0.0241,
      "step": 1070
    },
    {
      "epoch": 7.71,
      "learning_rate": 0.046031746031746035,
      "loss": 0.0256,
      "step": 1080
    },
    {
      "epoch": 7.79,
      "learning_rate": 0.04594356261022928,
      "loss": 0.0266,
      "step": 1090
    },
    {
      "epoch": 7.86,
      "learning_rate": 0.04585537918871252,
      "loss": 0.0234,
      "step": 1100
    },
    {
      "epoch": 7.93,
      "learning_rate": 0.04576719576719577,
      "loss": 0.0223,
      "step": 1110
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.04567901234567901,
      "loss": 0.02,
      "step": 1120
    },
    {
      "epoch": 8.07,
      "learning_rate": 0.04559082892416226,
      "loss": 0.022,
      "step": 1130
    },
    {
      "epoch": 8.14,
      "learning_rate": 0.0455026455026455,
      "loss": 0.0217,
      "step": 1140
    },
    {
      "epoch": 8.21,
      "learning_rate": 0.04541446208112875,
      "loss": 0.0219,
      "step": 1150
    },
    {
      "epoch": 8.29,
      "learning_rate": 0.045326278659611995,
      "loss": 0.0197,
      "step": 1160
    },
    {
      "epoch": 8.36,
      "learning_rate": 0.045238095238095244,
      "loss": 0.0209,
      "step": 1170
    },
    {
      "epoch": 8.43,
      "learning_rate": 0.045149911816578486,
      "loss": 0.0221,
      "step": 1180
    },
    {
      "epoch": 8.5,
      "learning_rate": 0.04506172839506173,
      "loss": 0.0198,
      "step": 1190
    },
    {
      "epoch": 8.57,
      "learning_rate": 0.04497354497354497,
      "loss": 0.0199,
      "step": 1200
    },
    {
      "epoch": 8.64,
      "learning_rate": 0.04488536155202822,
      "loss": 0.0196,
      "step": 1210
    },
    {
      "epoch": 8.71,
      "learning_rate": 0.04479717813051146,
      "loss": 0.019,
      "step": 1220
    },
    {
      "epoch": 8.79,
      "learning_rate": 0.04470899470899471,
      "loss": 0.0195,
      "step": 1230
    },
    {
      "epoch": 8.86,
      "learning_rate": 0.044620811287477954,
      "loss": 0.0196,
      "step": 1240
    },
    {
      "epoch": 8.93,
      "learning_rate": 0.044532627865961204,
      "loss": 0.0184,
      "step": 1250
    },
    {
      "epoch": 9.0,
      "learning_rate": 0.044444444444444446,
      "loss": 0.0205,
      "step": 1260
    },
    {
      "epoch": 9.07,
      "learning_rate": 0.044356261022927695,
      "loss": 0.0191,
      "step": 1270
    },
    {
      "epoch": 9.14,
      "learning_rate": 0.04426807760141094,
      "loss": 0.0184,
      "step": 1280
    },
    {
      "epoch": 9.21,
      "learning_rate": 0.04417989417989418,
      "loss": 0.0211,
      "step": 1290
    },
    {
      "epoch": 9.29,
      "learning_rate": 0.04409171075837742,
      "loss": 0.0184,
      "step": 1300
    },
    {
      "epoch": 9.36,
      "learning_rate": 0.04400352733686067,
      "loss": 0.0188,
      "step": 1310
    },
    {
      "epoch": 9.43,
      "learning_rate": 0.043915343915343914,
      "loss": 0.017,
      "step": 1320
    },
    {
      "epoch": 9.5,
      "learning_rate": 0.043827160493827164,
      "loss": 0.0158,
      "step": 1330
    },
    {
      "epoch": 9.57,
      "learning_rate": 0.043738977072310406,
      "loss": 0.0158,
      "step": 1340
    },
    {
      "epoch": 9.64,
      "learning_rate": 0.043650793650793655,
      "loss": 0.0165,
      "step": 1350
    },
    {
      "epoch": 9.71,
      "learning_rate": 0.0435626102292769,
      "loss": 0.0151,
      "step": 1360
    },
    {
      "epoch": 9.79,
      "learning_rate": 0.04347442680776015,
      "loss": 0.0144,
      "step": 1370
    },
    {
      "epoch": 9.86,
      "learning_rate": 0.04338624338624339,
      "loss": 0.0168,
      "step": 1380
    },
    {
      "epoch": 9.93,
      "learning_rate": 0.04329805996472663,
      "loss": 0.0162,
      "step": 1390
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.043209876543209874,
      "loss": 0.0177,
      "step": 1400
    },
    {
      "epoch": 10.07,
      "learning_rate": 0.04312169312169312,
      "loss": 0.0144,
      "step": 1410
    },
    {
      "epoch": 10.14,
      "learning_rate": 0.043033509700176366,
      "loss": 0.0145,
      "step": 1420
    },
    {
      "epoch": 10.21,
      "learning_rate": 0.042945326278659615,
      "loss": 0.0153,
      "step": 1430
    },
    {
      "epoch": 10.29,
      "learning_rate": 0.04285714285714286,
      "loss": 0.0147,
      "step": 1440
    },
    {
      "epoch": 10.36,
      "learning_rate": 0.04276895943562611,
      "loss": 0.0133,
      "step": 1450
    },
    {
      "epoch": 10.43,
      "learning_rate": 0.04268077601410935,
      "loss": 0.0139,
      "step": 1460
    },
    {
      "epoch": 10.5,
      "learning_rate": 0.0425925925925926,
      "loss": 0.0133,
      "step": 1470
    },
    {
      "epoch": 10.57,
      "learning_rate": 0.04250440917107584,
      "loss": 0.0135,
      "step": 1480
    },
    {
      "epoch": 10.64,
      "learning_rate": 0.04241622574955908,
      "loss": 0.0106,
      "step": 1490
    },
    {
      "epoch": 10.71,
      "learning_rate": 0.042328042328042326,
      "loss": 0.0132,
      "step": 1500
    },
    {
      "epoch": 10.79,
      "learning_rate": 0.042239858906525575,
      "loss": 0.0137,
      "step": 1510
    },
    {
      "epoch": 10.86,
      "learning_rate": 0.04215167548500882,
      "loss": 0.0127,
      "step": 1520
    },
    {
      "epoch": 10.93,
      "learning_rate": 0.04206349206349207,
      "loss": 0.0116,
      "step": 1530
    },
    {
      "epoch": 11.0,
      "learning_rate": 0.04197530864197531,
      "loss": 0.0123,
      "step": 1540
    },
    {
      "epoch": 11.07,
      "learning_rate": 0.04188712522045856,
      "loss": 0.0125,
      "step": 1550
    },
    {
      "epoch": 11.14,
      "learning_rate": 0.0417989417989418,
      "loss": 0.0111,
      "step": 1560
    },
    {
      "epoch": 11.21,
      "learning_rate": 0.04171075837742505,
      "loss": 0.0106,
      "step": 1570
    },
    {
      "epoch": 11.29,
      "learning_rate": 0.04162257495590829,
      "loss": 0.0105,
      "step": 1580
    },
    {
      "epoch": 11.36,
      "learning_rate": 0.041534391534391535,
      "loss": 0.0099,
      "step": 1590
    },
    {
      "epoch": 11.43,
      "learning_rate": 0.04144620811287478,
      "loss": 0.0116,
      "step": 1600
    },
    {
      "epoch": 11.5,
      "learning_rate": 0.04135802469135803,
      "loss": 0.0115,
      "step": 1610
    },
    {
      "epoch": 11.57,
      "learning_rate": 0.04126984126984127,
      "loss": 0.0113,
      "step": 1620
    },
    {
      "epoch": 11.64,
      "learning_rate": 0.04118165784832452,
      "loss": 0.012,
      "step": 1630
    },
    {
      "epoch": 11.71,
      "learning_rate": 0.04109347442680776,
      "loss": 0.0115,
      "step": 1640
    },
    {
      "epoch": 11.79,
      "learning_rate": 0.04100529100529101,
      "loss": 0.0117,
      "step": 1650
    },
    {
      "epoch": 11.86,
      "learning_rate": 0.04091710758377425,
      "loss": 0.0112,
      "step": 1660
    },
    {
      "epoch": 11.93,
      "learning_rate": 0.0408289241622575,
      "loss": 0.0095,
      "step": 1670
    },
    {
      "epoch": 12.0,
      "learning_rate": 0.040740740740740744,
      "loss": 0.0114,
      "step": 1680
    },
    {
      "epoch": 12.07,
      "learning_rate": 0.040652557319223986,
      "loss": 0.0114,
      "step": 1690
    },
    {
      "epoch": 12.14,
      "learning_rate": 0.04056437389770723,
      "loss": 0.0105,
      "step": 1700
    },
    {
      "epoch": 12.21,
      "learning_rate": 0.04047619047619048,
      "loss": 0.011,
      "step": 1710
    },
    {
      "epoch": 12.29,
      "learning_rate": 0.04038800705467372,
      "loss": 0.0113,
      "step": 1720
    },
    {
      "epoch": 12.36,
      "learning_rate": 0.04029982363315697,
      "loss": 0.0098,
      "step": 1730
    },
    {
      "epoch": 12.43,
      "learning_rate": 0.04021164021164021,
      "loss": 0.0093,
      "step": 1740
    },
    {
      "epoch": 12.5,
      "learning_rate": 0.04012345679012346,
      "loss": 0.0086,
      "step": 1750
    },
    {
      "epoch": 12.57,
      "learning_rate": 0.040035273368606704,
      "loss": 0.0086,
      "step": 1760
    },
    {
      "epoch": 12.64,
      "learning_rate": 0.03994708994708995,
      "loss": 0.0077,
      "step": 1770
    },
    {
      "epoch": 12.71,
      "learning_rate": 0.039858906525573196,
      "loss": 0.0088,
      "step": 1780
    },
    {
      "epoch": 12.79,
      "learning_rate": 0.03977072310405644,
      "loss": 0.0092,
      "step": 1790
    },
    {
      "epoch": 12.86,
      "learning_rate": 0.03968253968253968,
      "loss": 0.0097,
      "step": 1800
    },
    {
      "epoch": 12.93,
      "learning_rate": 0.03959435626102293,
      "loss": 0.0089,
      "step": 1810
    },
    {
      "epoch": 13.0,
      "learning_rate": 0.03950617283950617,
      "loss": 0.007,
      "step": 1820
    },
    {
      "epoch": 13.07,
      "learning_rate": 0.03941798941798942,
      "loss": 0.0071,
      "step": 1830
    },
    {
      "epoch": 13.14,
      "learning_rate": 0.039329805996472664,
      "loss": 0.008,
      "step": 1840
    },
    {
      "epoch": 13.21,
      "learning_rate": 0.03924162257495591,
      "loss": 0.0074,
      "step": 1850
    },
    {
      "epoch": 13.29,
      "learning_rate": 0.039153439153439155,
      "loss": 0.0084,
      "step": 1860
    },
    {
      "epoch": 13.36,
      "learning_rate": 0.039065255731922405,
      "loss": 0.0079,
      "step": 1870
    },
    {
      "epoch": 13.43,
      "learning_rate": 0.03897707231040565,
      "loss": 0.0082,
      "step": 1880
    },
    {
      "epoch": 13.5,
      "learning_rate": 0.03888888888888889,
      "loss": 0.0079,
      "step": 1890
    },
    {
      "epoch": 13.57,
      "learning_rate": 0.03880070546737213,
      "loss": 0.007,
      "step": 1900
    },
    {
      "epoch": 13.64,
      "learning_rate": 0.03871252204585538,
      "loss": 0.0075,
      "step": 1910
    },
    {
      "epoch": 13.71,
      "learning_rate": 0.038624338624338624,
      "loss": 0.0071,
      "step": 1920
    },
    {
      "epoch": 13.79,
      "learning_rate": 0.03853615520282187,
      "loss": 0.0073,
      "step": 1930
    },
    {
      "epoch": 13.86,
      "learning_rate": 0.038447971781305115,
      "loss": 0.0073,
      "step": 1940
    },
    {
      "epoch": 13.93,
      "learning_rate": 0.038359788359788365,
      "loss": 0.0069,
      "step": 1950
    },
    {
      "epoch": 14.0,
      "learning_rate": 0.03827160493827161,
      "loss": 0.0062,
      "step": 1960
    },
    {
      "epoch": 14.07,
      "learning_rate": 0.038183421516754856,
      "loss": 0.0066,
      "step": 1970
    },
    {
      "epoch": 14.14,
      "learning_rate": 0.0380952380952381,
      "loss": 0.0067,
      "step": 1980
    },
    {
      "epoch": 14.21,
      "learning_rate": 0.03800705467372134,
      "loss": 0.0086,
      "step": 1990
    },
    {
      "epoch": 14.29,
      "learning_rate": 0.03791887125220458,
      "loss": 0.0075,
      "step": 2000
    },
    {
      "epoch": 14.36,
      "learning_rate": 0.03783068783068783,
      "loss": 0.0072,
      "step": 2010
    },
    {
      "epoch": 14.43,
      "learning_rate": 0.037742504409171075,
      "loss": 0.008,
      "step": 2020
    },
    {
      "epoch": 14.5,
      "learning_rate": 0.037654320987654324,
      "loss": 0.0074,
      "step": 2030
    },
    {
      "epoch": 14.57,
      "learning_rate": 0.03756613756613757,
      "loss": 0.0054,
      "step": 2040
    },
    {
      "epoch": 14.64,
      "learning_rate": 0.037477954144620816,
      "loss": 0.0067,
      "step": 2050
    },
    {
      "epoch": 14.71,
      "learning_rate": 0.03738977072310406,
      "loss": 0.0062,
      "step": 2060
    },
    {
      "epoch": 14.79,
      "learning_rate": 0.0373015873015873,
      "loss": 0.0058,
      "step": 2070
    },
    {
      "epoch": 14.86,
      "learning_rate": 0.03721340388007054,
      "loss": 0.0064,
      "step": 2080
    },
    {
      "epoch": 14.93,
      "learning_rate": 0.03712522045855379,
      "loss": 0.0055,
      "step": 2090
    },
    {
      "epoch": 15.0,
      "learning_rate": 0.037037037037037035,
      "loss": 0.0059,
      "step": 2100
    },
    {
      "epoch": 15.07,
      "learning_rate": 0.036948853615520284,
      "loss": 0.0058,
      "step": 2110
    },
    {
      "epoch": 15.14,
      "learning_rate": 0.03686067019400353,
      "loss": 0.006,
      "step": 2120
    },
    {
      "epoch": 15.21,
      "learning_rate": 0.036772486772486776,
      "loss": 0.0055,
      "step": 2130
    },
    {
      "epoch": 15.29,
      "learning_rate": 0.03668430335097002,
      "loss": 0.0065,
      "step": 2140
    },
    {
      "epoch": 15.36,
      "learning_rate": 0.03659611992945327,
      "loss": 0.0063,
      "step": 2150
    },
    {
      "epoch": 15.43,
      "learning_rate": 0.03650793650793651,
      "loss": 0.0059,
      "step": 2160
    },
    {
      "epoch": 15.5,
      "learning_rate": 0.03641975308641975,
      "loss": 0.0052,
      "step": 2170
    },
    {
      "epoch": 15.57,
      "learning_rate": 0.036331569664902995,
      "loss": 0.0056,
      "step": 2180
    },
    {
      "epoch": 15.64,
      "learning_rate": 0.036243386243386244,
      "loss": 0.0052,
      "step": 2190
    },
    {
      "epoch": 15.71,
      "learning_rate": 0.036155202821869487,
      "loss": 0.0056,
      "step": 2200
    },
    {
      "epoch": 15.79,
      "learning_rate": 0.036067019400352736,
      "loss": 0.0051,
      "step": 2210
    },
    {
      "epoch": 15.86,
      "learning_rate": 0.03597883597883598,
      "loss": 0.0055,
      "step": 2220
    },
    {
      "epoch": 15.93,
      "learning_rate": 0.03589065255731923,
      "loss": 0.0059,
      "step": 2230
    },
    {
      "epoch": 16.0,
      "learning_rate": 0.03580246913580247,
      "loss": 0.0061,
      "step": 2240
    },
    {
      "epoch": 16.07,
      "learning_rate": 0.03571428571428572,
      "loss": 0.0061,
      "step": 2250
    },
    {
      "epoch": 16.14,
      "learning_rate": 0.03562610229276896,
      "loss": 0.0054,
      "step": 2260
    },
    {
      "epoch": 16.21,
      "learning_rate": 0.035537918871252204,
      "loss": 0.0064,
      "step": 2270
    },
    {
      "epoch": 16.29,
      "learning_rate": 0.035449735449735446,
      "loss": 0.0055,
      "step": 2280
    },
    {
      "epoch": 16.36,
      "learning_rate": 0.035361552028218696,
      "loss": 0.0052,
      "step": 2290
    },
    {
      "epoch": 16.43,
      "learning_rate": 0.03527336860670194,
      "loss": 0.0051,
      "step": 2300
    },
    {
      "epoch": 16.5,
      "learning_rate": 0.03518518518518519,
      "loss": 0.005,
      "step": 2310
    },
    {
      "epoch": 16.57,
      "learning_rate": 0.03509700176366843,
      "loss": 0.0055,
      "step": 2320
    },
    {
      "epoch": 16.64,
      "learning_rate": 0.03500881834215168,
      "loss": 0.0052,
      "step": 2330
    },
    {
      "epoch": 16.71,
      "learning_rate": 0.03492063492063492,
      "loss": 0.0051,
      "step": 2340
    },
    {
      "epoch": 16.79,
      "learning_rate": 0.03483245149911817,
      "loss": 0.0052,
      "step": 2350
    },
    {
      "epoch": 16.86,
      "learning_rate": 0.03474426807760141,
      "loss": 0.0053,
      "step": 2360
    },
    {
      "epoch": 16.93,
      "learning_rate": 0.034656084656084656,
      "loss": 0.0056,
      "step": 2370
    },
    {
      "epoch": 17.0,
      "learning_rate": 0.0345679012345679,
      "loss": 0.0054,
      "step": 2380
    },
    {
      "epoch": 17.07,
      "learning_rate": 0.03447971781305115,
      "loss": 0.0053,
      "step": 2390
    },
    {
      "epoch": 17.14,
      "learning_rate": 0.03439153439153439,
      "loss": 0.0054,
      "step": 2400
    },
    {
      "epoch": 17.21,
      "learning_rate": 0.03430335097001764,
      "loss": 0.0064,
      "step": 2410
    },
    {
      "epoch": 17.29,
      "learning_rate": 0.03421516754850088,
      "loss": 0.005,
      "step": 2420
    },
    {
      "epoch": 17.36,
      "learning_rate": 0.03412698412698413,
      "loss": 0.0054,
      "step": 2430
    },
    {
      "epoch": 17.43,
      "learning_rate": 0.03403880070546737,
      "loss": 0.0047,
      "step": 2440
    },
    {
      "epoch": 17.5,
      "learning_rate": 0.03395061728395062,
      "loss": 0.0055,
      "step": 2450
    },
    {
      "epoch": 17.57,
      "learning_rate": 0.033862433862433865,
      "loss": 0.0052,
      "step": 2460
    },
    {
      "epoch": 17.64,
      "learning_rate": 0.03377425044091711,
      "loss": 0.0048,
      "step": 2470
    },
    {
      "epoch": 17.71,
      "learning_rate": 0.03368606701940035,
      "loss": 0.0049,
      "step": 2480
    },
    {
      "epoch": 17.79,
      "learning_rate": 0.0335978835978836,
      "loss": 0.0049,
      "step": 2490
    },
    {
      "epoch": 17.86,
      "learning_rate": 0.03350970017636684,
      "loss": 0.0055,
      "step": 2500
    },
    {
      "epoch": 17.93,
      "learning_rate": 0.03342151675485009,
      "loss": 0.0052,
      "step": 2510
    },
    {
      "epoch": 18.0,
      "learning_rate": 0.03333333333333333,
      "loss": 0.0059,
      "step": 2520
    },
    {
      "epoch": 18.07,
      "learning_rate": 0.03324514991181658,
      "loss": 0.0053,
      "step": 2530
    },
    {
      "epoch": 18.14,
      "learning_rate": 0.033156966490299825,
      "loss": 0.0051,
      "step": 2540
    },
    {
      "epoch": 18.21,
      "learning_rate": 0.033068783068783074,
      "loss": 0.0043,
      "step": 2550
    },
    {
      "epoch": 18.29,
      "learning_rate": 0.032980599647266316,
      "loss": 0.0045,
      "step": 2560
    },
    {
      "epoch": 18.36,
      "learning_rate": 0.03289241622574956,
      "loss": 0.0048,
      "step": 2570
    },
    {
      "epoch": 18.43,
      "learning_rate": 0.0328042328042328,
      "loss": 0.0042,
      "step": 2580
    },
    {
      "epoch": 18.5,
      "learning_rate": 0.03271604938271605,
      "loss": 0.0048,
      "step": 2590
    },
    {
      "epoch": 18.57,
      "learning_rate": 0.03262786596119929,
      "loss": 0.0052,
      "step": 2600
    },
    {
      "epoch": 18.64,
      "learning_rate": 0.03253968253968254,
      "loss": 0.0061,
      "step": 2610
    },
    {
      "epoch": 18.71,
      "learning_rate": 0.032451499118165784,
      "loss": 0.0052,
      "step": 2620
    },
    {
      "epoch": 18.79,
      "learning_rate": 0.032363315696649034,
      "loss": 0.0045,
      "step": 2630
    },
    {
      "epoch": 18.86,
      "learning_rate": 0.032275132275132276,
      "loss": 0.0045,
      "step": 2640
    },
    {
      "epoch": 18.93,
      "learning_rate": 0.032186948853615525,
      "loss": 0.0055,
      "step": 2650
    },
    {
      "epoch": 19.0,
      "learning_rate": 0.03209876543209877,
      "loss": 0.0046,
      "step": 2660
    },
    {
      "epoch": 19.07,
      "learning_rate": 0.03201058201058201,
      "loss": 0.0047,
      "step": 2670
    },
    {
      "epoch": 19.14,
      "learning_rate": 0.03192239858906525,
      "loss": 0.0049,
      "step": 2680
    },
    {
      "epoch": 19.21,
      "learning_rate": 0.0318342151675485,
      "loss": 0.0049,
      "step": 2690
    },
    {
      "epoch": 19.29,
      "learning_rate": 0.031746031746031744,
      "loss": 0.005,
      "step": 2700
    },
    {
      "epoch": 19.36,
      "learning_rate": 0.031657848324514994,
      "loss": 0.0054,
      "step": 2710
    },
    {
      "epoch": 19.43,
      "learning_rate": 0.031569664902998236,
      "loss": 0.0049,
      "step": 2720
    },
    {
      "epoch": 19.5,
      "learning_rate": 0.031481481481481485,
      "loss": 0.0047,
      "step": 2730
    },
    {
      "epoch": 19.57,
      "learning_rate": 0.03139329805996473,
      "loss": 0.0046,
      "step": 2740
    },
    {
      "epoch": 19.64,
      "learning_rate": 0.03130511463844798,
      "loss": 0.0045,
      "step": 2750
    },
    {
      "epoch": 19.71,
      "learning_rate": 0.031216931216931216,
      "loss": 0.0043,
      "step": 2760
    },
    {
      "epoch": 19.79,
      "learning_rate": 0.031128747795414465,
      "loss": 0.0044,
      "step": 2770
    },
    {
      "epoch": 19.86,
      "learning_rate": 0.031040564373897708,
      "loss": 0.0051,
      "step": 2780
    },
    {
      "epoch": 19.93,
      "learning_rate": 0.030952380952380953,
      "loss": 0.0048,
      "step": 2790
    },
    {
      "epoch": 20.0,
      "learning_rate": 0.030864197530864196,
      "loss": 0.0044,
      "step": 2800
    },
    {
      "epoch": 20.07,
      "learning_rate": 0.030776014109347445,
      "loss": 0.0048,
      "step": 2810
    },
    {
      "epoch": 20.14,
      "learning_rate": 0.030687830687830688,
      "loss": 0.0046,
      "step": 2820
    },
    {
      "epoch": 20.21,
      "learning_rate": 0.030599647266313937,
      "loss": 0.0045,
      "step": 2830
    },
    {
      "epoch": 20.29,
      "learning_rate": 0.03051146384479718,
      "loss": 0.004,
      "step": 2840
    },
    {
      "epoch": 20.36,
      "learning_rate": 0.030423280423280425,
      "loss": 0.0043,
      "step": 2850
    },
    {
      "epoch": 20.43,
      "learning_rate": 0.030335097001763667,
      "loss": 0.0052,
      "step": 2860
    },
    {
      "epoch": 20.5,
      "learning_rate": 0.030246913580246917,
      "loss": 0.0053,
      "step": 2870
    },
    {
      "epoch": 20.57,
      "learning_rate": 0.03015873015873016,
      "loss": 0.0053,
      "step": 2880
    },
    {
      "epoch": 20.64,
      "learning_rate": 0.030070546737213405,
      "loss": 0.0048,
      "step": 2890
    },
    {
      "epoch": 20.71,
      "learning_rate": 0.029982363315696647,
      "loss": 0.0056,
      "step": 2900
    },
    {
      "epoch": 20.79,
      "learning_rate": 0.029894179894179897,
      "loss": 0.0045,
      "step": 2910
    },
    {
      "epoch": 20.86,
      "learning_rate": 0.02980599647266314,
      "loss": 0.0048,
      "step": 2920
    },
    {
      "epoch": 20.93,
      "learning_rate": 0.02971781305114639,
      "loss": 0.005,
      "step": 2930
    },
    {
      "epoch": 21.0,
      "learning_rate": 0.02962962962962963,
      "loss": 0.0047,
      "step": 2940
    },
    {
      "epoch": 21.07,
      "learning_rate": 0.029541446208112877,
      "loss": 0.0046,
      "step": 2950
    },
    {
      "epoch": 21.14,
      "learning_rate": 0.02945326278659612,
      "loss": 0.0047,
      "step": 2960
    },
    {
      "epoch": 21.21,
      "learning_rate": 0.02936507936507937,
      "loss": 0.0046,
      "step": 2970
    },
    {
      "epoch": 21.29,
      "learning_rate": 0.02927689594356261,
      "loss": 0.0046,
      "step": 2980
    },
    {
      "epoch": 21.36,
      "learning_rate": 0.029188712522045857,
      "loss": 0.0041,
      "step": 2990
    },
    {
      "epoch": 21.43,
      "learning_rate": 0.0291005291005291,
      "loss": 0.0048,
      "step": 3000
    },
    {
      "epoch": 21.5,
      "learning_rate": 0.029012345679012348,
      "loss": 0.0044,
      "step": 3010
    },
    {
      "epoch": 21.57,
      "learning_rate": 0.02892416225749559,
      "loss": 0.0045,
      "step": 3020
    },
    {
      "epoch": 21.64,
      "learning_rate": 0.02883597883597884,
      "loss": 0.0045,
      "step": 3030
    },
    {
      "epoch": 21.71,
      "learning_rate": 0.028747795414462082,
      "loss": 0.0043,
      "step": 3040
    },
    {
      "epoch": 21.79,
      "learning_rate": 0.028659611992945328,
      "loss": 0.0038,
      "step": 3050
    },
    {
      "epoch": 21.86,
      "learning_rate": 0.02857142857142857,
      "loss": 0.0044,
      "step": 3060
    },
    {
      "epoch": 21.93,
      "learning_rate": 0.02848324514991182,
      "loss": 0.0041,
      "step": 3070
    },
    {
      "epoch": 22.0,
      "learning_rate": 0.028395061728395062,
      "loss": 0.0044,
      "step": 3080
    },
    {
      "epoch": 22.07,
      "learning_rate": 0.028306878306878308,
      "loss": 0.0038,
      "step": 3090
    },
    {
      "epoch": 22.14,
      "learning_rate": 0.02821869488536155,
      "loss": 0.0039,
      "step": 3100
    },
    {
      "epoch": 22.21,
      "learning_rate": 0.0281305114638448,
      "loss": 0.0038,
      "step": 3110
    },
    {
      "epoch": 22.29,
      "learning_rate": 0.028042328042328042,
      "loss": 0.0041,
      "step": 3120
    },
    {
      "epoch": 22.36,
      "learning_rate": 0.02795414462081129,
      "loss": 0.0042,
      "step": 3130
    },
    {
      "epoch": 22.43,
      "learning_rate": 0.027865961199294534,
      "loss": 0.0044,
      "step": 3140
    },
    {
      "epoch": 22.5,
      "learning_rate": 0.02777777777777778,
      "loss": 0.0044,
      "step": 3150
    },
    {
      "epoch": 22.57,
      "learning_rate": 0.027689594356261022,
      "loss": 0.0045,
      "step": 3160
    },
    {
      "epoch": 22.64,
      "learning_rate": 0.02760141093474427,
      "loss": 0.0043,
      "step": 3170
    },
    {
      "epoch": 22.71,
      "learning_rate": 0.027513227513227514,
      "loss": 0.0042,
      "step": 3180
    },
    {
      "epoch": 22.79,
      "learning_rate": 0.02742504409171076,
      "loss": 0.0037,
      "step": 3190
    },
    {
      "epoch": 22.86,
      "learning_rate": 0.027336860670194002,
      "loss": 0.0041,
      "step": 3200
    },
    {
      "epoch": 22.93,
      "learning_rate": 0.02724867724867725,
      "loss": 0.0039,
      "step": 3210
    },
    {
      "epoch": 23.0,
      "learning_rate": 0.027160493827160494,
      "loss": 0.0048,
      "step": 3220
    },
    {
      "epoch": 23.07,
      "learning_rate": 0.027072310405643743,
      "loss": 0.004,
      "step": 3230
    },
    {
      "epoch": 23.14,
      "learning_rate": 0.026984126984126985,
      "loss": 0.0041,
      "step": 3240
    },
    {
      "epoch": 23.21,
      "learning_rate": 0.02689594356261023,
      "loss": 0.0043,
      "step": 3250
    },
    {
      "epoch": 23.29,
      "learning_rate": 0.026807760141093474,
      "loss": 0.0044,
      "step": 3260
    },
    {
      "epoch": 23.36,
      "learning_rate": 0.026719576719576723,
      "loss": 0.004,
      "step": 3270
    },
    {
      "epoch": 23.43,
      "learning_rate": 0.026631393298059965,
      "loss": 0.0041,
      "step": 3280
    },
    {
      "epoch": 23.5,
      "learning_rate": 0.02654320987654321,
      "loss": 0.0042,
      "step": 3290
    },
    {
      "epoch": 23.57,
      "learning_rate": 0.026455026455026454,
      "loss": 0.004,
      "step": 3300
    },
    {
      "epoch": 23.64,
      "learning_rate": 0.026366843033509703,
      "loss": 0.0038,
      "step": 3310
    },
    {
      "epoch": 23.71,
      "learning_rate": 0.026278659611992945,
      "loss": 0.0043,
      "step": 3320
    },
    {
      "epoch": 23.79,
      "learning_rate": 0.026190476190476195,
      "loss": 0.0043,
      "step": 3330
    },
    {
      "epoch": 23.86,
      "learning_rate": 0.026102292768959437,
      "loss": 0.0042,
      "step": 3340
    },
    {
      "epoch": 23.93,
      "learning_rate": 0.026014109347442683,
      "loss": 0.0037,
      "step": 3350
    },
    {
      "epoch": 24.0,
      "learning_rate": 0.025925925925925925,
      "loss": 0.0032,
      "step": 3360
    },
    {
      "epoch": 24.07,
      "learning_rate": 0.025837742504409174,
      "loss": 0.0038,
      "step": 3370
    },
    {
      "epoch": 24.14,
      "learning_rate": 0.025749559082892417,
      "loss": 0.0044,
      "step": 3380
    },
    {
      "epoch": 24.21,
      "learning_rate": 0.025661375661375663,
      "loss": 0.0035,
      "step": 3390
    },
    {
      "epoch": 24.29,
      "learning_rate": 0.025573192239858905,
      "loss": 0.0035,
      "step": 3400
    },
    {
      "epoch": 24.36,
      "learning_rate": 0.025485008818342154,
      "loss": 0.0036,
      "step": 3410
    },
    {
      "epoch": 24.43,
      "learning_rate": 0.025396825396825397,
      "loss": 0.0035,
      "step": 3420
    },
    {
      "epoch": 24.5,
      "learning_rate": 0.025308641975308646,
      "loss": 0.0037,
      "step": 3430
    },
    {
      "epoch": 24.57,
      "learning_rate": 0.02522045855379189,
      "loss": 0.0038,
      "step": 3440
    },
    {
      "epoch": 24.64,
      "learning_rate": 0.025132275132275134,
      "loss": 0.0042,
      "step": 3450
    },
    {
      "epoch": 24.71,
      "learning_rate": 0.025044091710758377,
      "loss": 0.0035,
      "step": 3460
    },
    {
      "epoch": 24.79,
      "learning_rate": 0.024955908289241623,
      "loss": 0.0047,
      "step": 3470
    },
    {
      "epoch": 24.86,
      "learning_rate": 0.02486772486772487,
      "loss": 0.0041,
      "step": 3480
    },
    {
      "epoch": 24.93,
      "learning_rate": 0.024779541446208114,
      "loss": 0.0043,
      "step": 3490
    },
    {
      "epoch": 25.0,
      "learning_rate": 0.024691358024691357,
      "loss": 0.0039,
      "step": 3500
    },
    {
      "epoch": 25.07,
      "learning_rate": 0.024603174603174603,
      "loss": 0.0041,
      "step": 3510
    },
    {
      "epoch": 25.14,
      "learning_rate": 0.02451499118165785,
      "loss": 0.0042,
      "step": 3520
    },
    {
      "epoch": 25.21,
      "learning_rate": 0.024426807760141094,
      "loss": 0.0041,
      "step": 3530
    },
    {
      "epoch": 25.29,
      "learning_rate": 0.02433862433862434,
      "loss": 0.0037,
      "step": 3540
    },
    {
      "epoch": 25.36,
      "learning_rate": 0.024250440917107582,
      "loss": 0.0042,
      "step": 3550
    },
    {
      "epoch": 25.43,
      "learning_rate": 0.02416225749559083,
      "loss": 0.0035,
      "step": 3560
    },
    {
      "epoch": 25.5,
      "learning_rate": 0.024074074074074074,
      "loss": 0.0042,
      "step": 3570
    },
    {
      "epoch": 25.57,
      "learning_rate": 0.02398589065255732,
      "loss": 0.0039,
      "step": 3580
    },
    {
      "epoch": 25.64,
      "learning_rate": 0.023897707231040566,
      "loss": 0.0039,
      "step": 3590
    },
    {
      "epoch": 25.71,
      "learning_rate": 0.023809523809523808,
      "loss": 0.0035,
      "step": 3600
    },
    {
      "epoch": 25.79,
      "learning_rate": 0.023721340388007054,
      "loss": 0.0035,
      "step": 3610
    },
    {
      "epoch": 25.86,
      "learning_rate": 0.0236331569664903,
      "loss": 0.0041,
      "step": 3620
    },
    {
      "epoch": 25.93,
      "learning_rate": 0.023544973544973546,
      "loss": 0.004,
      "step": 3630
    },
    {
      "epoch": 26.0,
      "learning_rate": 0.02345679012345679,
      "loss": 0.0034,
      "step": 3640
    },
    {
      "epoch": 26.07,
      "learning_rate": 0.023368606701940034,
      "loss": 0.0039,
      "step": 3650
    },
    {
      "epoch": 26.14,
      "learning_rate": 0.02328042328042328,
      "loss": 0.0037,
      "step": 3660
    },
    {
      "epoch": 26.21,
      "learning_rate": 0.023192239858906526,
      "loss": 0.0042,
      "step": 3670
    },
    {
      "epoch": 26.29,
      "learning_rate": 0.02310405643738977,
      "loss": 0.0034,
      "step": 3680
    },
    {
      "epoch": 26.36,
      "learning_rate": 0.023015873015873017,
      "loss": 0.0036,
      "step": 3690
    },
    {
      "epoch": 26.43,
      "learning_rate": 0.02292768959435626,
      "loss": 0.0034,
      "step": 3700
    },
    {
      "epoch": 26.5,
      "learning_rate": 0.022839506172839506,
      "loss": 0.004,
      "step": 3710
    },
    {
      "epoch": 26.57,
      "learning_rate": 0.02275132275132275,
      "loss": 0.0037,
      "step": 3720
    },
    {
      "epoch": 26.64,
      "learning_rate": 0.022663139329805997,
      "loss": 0.0039,
      "step": 3730
    },
    {
      "epoch": 26.71,
      "learning_rate": 0.022574955908289243,
      "loss": 0.0034,
      "step": 3740
    },
    {
      "epoch": 26.79,
      "learning_rate": 0.022486772486772486,
      "loss": 0.0035,
      "step": 3750
    },
    {
      "epoch": 26.86,
      "learning_rate": 0.02239858906525573,
      "loss": 0.0035,
      "step": 3760
    },
    {
      "epoch": 26.93,
      "learning_rate": 0.022310405643738977,
      "loss": 0.0039,
      "step": 3770
    },
    {
      "epoch": 27.0,
      "learning_rate": 0.022222222222222223,
      "loss": 0.0035,
      "step": 3780
    },
    {
      "epoch": 27.07,
      "learning_rate": 0.02213403880070547,
      "loss": 0.0038,
      "step": 3790
    },
    {
      "epoch": 27.14,
      "learning_rate": 0.02204585537918871,
      "loss": 0.0039,
      "step": 3800
    },
    {
      "epoch": 27.21,
      "learning_rate": 0.021957671957671957,
      "loss": 0.0033,
      "step": 3810
    },
    {
      "epoch": 27.29,
      "learning_rate": 0.021869488536155203,
      "loss": 0.0035,
      "step": 3820
    },
    {
      "epoch": 27.36,
      "learning_rate": 0.02178130511463845,
      "loss": 0.0038,
      "step": 3830
    },
    {
      "epoch": 27.43,
      "learning_rate": 0.021693121693121695,
      "loss": 0.0032,
      "step": 3840
    },
    {
      "epoch": 27.5,
      "learning_rate": 0.021604938271604937,
      "loss": 0.0042,
      "step": 3850
    },
    {
      "epoch": 27.57,
      "learning_rate": 0.021516754850088183,
      "loss": 0.0036,
      "step": 3860
    },
    {
      "epoch": 27.64,
      "learning_rate": 0.02142857142857143,
      "loss": 0.0034,
      "step": 3870
    },
    {
      "epoch": 27.71,
      "learning_rate": 0.021340388007054675,
      "loss": 0.0037,
      "step": 3880
    },
    {
      "epoch": 27.79,
      "learning_rate": 0.02125220458553792,
      "loss": 0.0036,
      "step": 3890
    },
    {
      "epoch": 27.86,
      "learning_rate": 0.021164021164021163,
      "loss": 0.0033,
      "step": 3900
    },
    {
      "epoch": 27.93,
      "learning_rate": 0.02107583774250441,
      "loss": 0.0035,
      "step": 3910
    },
    {
      "epoch": 28.0,
      "learning_rate": 0.020987654320987655,
      "loss": 0.0034,
      "step": 3920
    },
    {
      "epoch": 28.07,
      "learning_rate": 0.0208994708994709,
      "loss": 0.0037,
      "step": 3930
    },
    {
      "epoch": 28.14,
      "learning_rate": 0.020811287477954146,
      "loss": 0.003,
      "step": 3940
    },
    {
      "epoch": 28.21,
      "learning_rate": 0.02072310405643739,
      "loss": 0.0033,
      "step": 3950
    },
    {
      "epoch": 28.29,
      "learning_rate": 0.020634920634920634,
      "loss": 0.004,
      "step": 3960
    },
    {
      "epoch": 28.36,
      "learning_rate": 0.02054673721340388,
      "loss": 0.0036,
      "step": 3970
    },
    {
      "epoch": 28.43,
      "learning_rate": 0.020458553791887126,
      "loss": 0.0034,
      "step": 3980
    },
    {
      "epoch": 28.5,
      "learning_rate": 0.020370370370370372,
      "loss": 0.0035,
      "step": 3990
    },
    {
      "epoch": 28.57,
      "learning_rate": 0.020282186948853614,
      "loss": 0.0034,
      "step": 4000
    },
    {
      "epoch": 28.64,
      "learning_rate": 0.02019400352733686,
      "loss": 0.0035,
      "step": 4010
    },
    {
      "epoch": 28.71,
      "learning_rate": 0.020105820105820106,
      "loss": 0.0037,
      "step": 4020
    },
    {
      "epoch": 28.79,
      "learning_rate": 0.020017636684303352,
      "loss": 0.0033,
      "step": 4030
    },
    {
      "epoch": 28.86,
      "learning_rate": 0.019929453262786598,
      "loss": 0.0037,
      "step": 4040
    },
    {
      "epoch": 28.93,
      "learning_rate": 0.01984126984126984,
      "loss": 0.0033,
      "step": 4050
    },
    {
      "epoch": 29.0,
      "learning_rate": 0.019753086419753086,
      "loss": 0.0032,
      "step": 4060
    },
    {
      "epoch": 29.07,
      "learning_rate": 0.019664902998236332,
      "loss": 0.0034,
      "step": 4070
    },
    {
      "epoch": 29.14,
      "learning_rate": 0.019576719576719578,
      "loss": 0.0036,
      "step": 4080
    },
    {
      "epoch": 29.21,
      "learning_rate": 0.019488536155202824,
      "loss": 0.0032,
      "step": 4090
    },
    {
      "epoch": 29.29,
      "learning_rate": 0.019400352733686066,
      "loss": 0.003,
      "step": 4100
    },
    {
      "epoch": 29.36,
      "learning_rate": 0.019312169312169312,
      "loss": 0.0033,
      "step": 4110
    },
    {
      "epoch": 29.43,
      "learning_rate": 0.019223985890652558,
      "loss": 0.003,
      "step": 4120
    },
    {
      "epoch": 29.5,
      "learning_rate": 0.019135802469135803,
      "loss": 0.0034,
      "step": 4130
    },
    {
      "epoch": 29.57,
      "learning_rate": 0.01904761904761905,
      "loss": 0.0032,
      "step": 4140
    },
    {
      "epoch": 29.64,
      "learning_rate": 0.01895943562610229,
      "loss": 0.0031,
      "step": 4150
    },
    {
      "epoch": 29.71,
      "learning_rate": 0.018871252204585538,
      "loss": 0.0031,
      "step": 4160
    },
    {
      "epoch": 29.79,
      "learning_rate": 0.018783068783068783,
      "loss": 0.003,
      "step": 4170
    },
    {
      "epoch": 29.86,
      "learning_rate": 0.01869488536155203,
      "loss": 0.0034,
      "step": 4180
    },
    {
      "epoch": 29.93,
      "learning_rate": 0.01860670194003527,
      "loss": 0.0037,
      "step": 4190
    },
    {
      "epoch": 30.0,
      "learning_rate": 0.018518518518518517,
      "loss": 0.0031,
      "step": 4200
    },
    {
      "epoch": 30.07,
      "learning_rate": 0.018430335097001763,
      "loss": 0.0032,
      "step": 4210
    },
    {
      "epoch": 30.14,
      "learning_rate": 0.01834215167548501,
      "loss": 0.0032,
      "step": 4220
    },
    {
      "epoch": 30.21,
      "learning_rate": 0.018253968253968255,
      "loss": 0.003,
      "step": 4230
    },
    {
      "epoch": 30.29,
      "learning_rate": 0.018165784832451497,
      "loss": 0.0033,
      "step": 4240
    },
    {
      "epoch": 30.36,
      "learning_rate": 0.018077601410934743,
      "loss": 0.0033,
      "step": 4250
    },
    {
      "epoch": 30.43,
      "learning_rate": 0.01798941798941799,
      "loss": 0.003,
      "step": 4260
    },
    {
      "epoch": 30.5,
      "learning_rate": 0.017901234567901235,
      "loss": 0.0034,
      "step": 4270
    },
    {
      "epoch": 30.57,
      "learning_rate": 0.01781305114638448,
      "loss": 0.0033,
      "step": 4280
    },
    {
      "epoch": 30.64,
      "learning_rate": 0.017724867724867723,
      "loss": 0.0036,
      "step": 4290
    },
    {
      "epoch": 30.71,
      "learning_rate": 0.01763668430335097,
      "loss": 0.0035,
      "step": 4300
    },
    {
      "epoch": 30.79,
      "learning_rate": 0.017548500881834215,
      "loss": 0.0034,
      "step": 4310
    },
    {
      "epoch": 30.86,
      "learning_rate": 0.01746031746031746,
      "loss": 0.0034,
      "step": 4320
    },
    {
      "epoch": 30.93,
      "learning_rate": 0.017372134038800707,
      "loss": 0.0036,
      "step": 4330
    },
    {
      "epoch": 31.0,
      "learning_rate": 0.01728395061728395,
      "loss": 0.0036,
      "step": 4340
    },
    {
      "epoch": 31.07,
      "learning_rate": 0.017195767195767195,
      "loss": 0.0035,
      "step": 4350
    },
    {
      "epoch": 31.14,
      "learning_rate": 0.01710758377425044,
      "loss": 0.0035,
      "step": 4360
    },
    {
      "epoch": 31.21,
      "learning_rate": 0.017019400352733687,
      "loss": 0.0034,
      "step": 4370
    },
    {
      "epoch": 31.29,
      "learning_rate": 0.016931216931216932,
      "loss": 0.0035,
      "step": 4380
    },
    {
      "epoch": 31.36,
      "learning_rate": 0.016843033509700175,
      "loss": 0.0035,
      "step": 4390
    },
    {
      "epoch": 31.43,
      "learning_rate": 0.01675485008818342,
      "loss": 0.0043,
      "step": 4400
    },
    {
      "epoch": 31.5,
      "learning_rate": 0.016666666666666666,
      "loss": 0.0033,
      "step": 4410
    },
    {
      "epoch": 31.57,
      "learning_rate": 0.016578483245149912,
      "loss": 0.0032,
      "step": 4420
    },
    {
      "epoch": 31.64,
      "learning_rate": 0.016490299823633158,
      "loss": 0.003,
      "step": 4430
    },
    {
      "epoch": 31.71,
      "learning_rate": 0.0164021164021164,
      "loss": 0.0033,
      "step": 4440
    },
    {
      "epoch": 31.79,
      "learning_rate": 0.016313932980599646,
      "loss": 0.0033,
      "step": 4450
    },
    {
      "epoch": 31.86,
      "learning_rate": 0.016225749559082892,
      "loss": 0.0033,
      "step": 4460
    },
    {
      "epoch": 31.93,
      "learning_rate": 0.016137566137566138,
      "loss": 0.0033,
      "step": 4470
    },
    {
      "epoch": 32.0,
      "learning_rate": 0.016049382716049384,
      "loss": 0.0031,
      "step": 4480
    },
    {
      "epoch": 32.07,
      "learning_rate": 0.015961199294532626,
      "loss": 0.0031,
      "step": 4490
    },
    {
      "epoch": 32.14,
      "learning_rate": 0.015873015873015872,
      "loss": 0.0034,
      "step": 4500
    },
    {
      "epoch": 32.21,
      "learning_rate": 0.015784832451499118,
      "loss": 0.0033,
      "step": 4510
    },
    {
      "epoch": 32.29,
      "learning_rate": 0.015696649029982364,
      "loss": 0.003,
      "step": 4520
    },
    {
      "epoch": 32.36,
      "learning_rate": 0.015608465608465608,
      "loss": 0.0032,
      "step": 4530
    },
    {
      "epoch": 32.43,
      "learning_rate": 0.015520282186948854,
      "loss": 0.0031,
      "step": 4540
    },
    {
      "epoch": 32.5,
      "learning_rate": 0.015432098765432098,
      "loss": 0.0034,
      "step": 4550
    },
    {
      "epoch": 32.57,
      "learning_rate": 0.015343915343915344,
      "loss": 0.0037,
      "step": 4560
    },
    {
      "epoch": 32.64,
      "learning_rate": 0.01525573192239859,
      "loss": 0.0031,
      "step": 4570
    },
    {
      "epoch": 32.71,
      "learning_rate": 0.015167548500881834,
      "loss": 0.003,
      "step": 4580
    },
    {
      "epoch": 32.79,
      "learning_rate": 0.01507936507936508,
      "loss": 0.003,
      "step": 4590
    },
    {
      "epoch": 32.86,
      "learning_rate": 0.014991181657848324,
      "loss": 0.0035,
      "step": 4600
    },
    {
      "epoch": 32.93,
      "learning_rate": 0.01490299823633157,
      "loss": 0.0031,
      "step": 4610
    },
    {
      "epoch": 33.0,
      "learning_rate": 0.014814814814814815,
      "loss": 0.0033,
      "step": 4620
    },
    {
      "epoch": 33.07,
      "learning_rate": 0.01472663139329806,
      "loss": 0.0031,
      "step": 4630
    },
    {
      "epoch": 33.14,
      "learning_rate": 0.014638447971781305,
      "loss": 0.0032,
      "step": 4640
    },
    {
      "epoch": 33.21,
      "learning_rate": 0.01455026455026455,
      "loss": 0.0036,
      "step": 4650
    },
    {
      "epoch": 33.29,
      "learning_rate": 0.014462081128747795,
      "loss": 0.0035,
      "step": 4660
    },
    {
      "epoch": 33.36,
      "learning_rate": 0.014373897707231041,
      "loss": 0.0033,
      "step": 4670
    },
    {
      "epoch": 33.43,
      "learning_rate": 0.014285714285714285,
      "loss": 0.0036,
      "step": 4680
    },
    {
      "epoch": 33.5,
      "learning_rate": 0.014197530864197531,
      "loss": 0.003,
      "step": 4690
    },
    {
      "epoch": 33.57,
      "learning_rate": 0.014109347442680775,
      "loss": 0.0034,
      "step": 4700
    },
    {
      "epoch": 33.64,
      "learning_rate": 0.014021164021164021,
      "loss": 0.0033,
      "step": 4710
    },
    {
      "epoch": 33.71,
      "learning_rate": 0.013932980599647267,
      "loss": 0.0033,
      "step": 4720
    },
    {
      "epoch": 33.79,
      "learning_rate": 0.013844797178130511,
      "loss": 0.0032,
      "step": 4730
    },
    {
      "epoch": 33.86,
      "learning_rate": 0.013756613756613757,
      "loss": 0.0032,
      "step": 4740
    },
    {
      "epoch": 33.93,
      "learning_rate": 0.013668430335097001,
      "loss": 0.0033,
      "step": 4750
    },
    {
      "epoch": 34.0,
      "learning_rate": 0.013580246913580247,
      "loss": 0.0031,
      "step": 4760
    },
    {
      "epoch": 34.07,
      "learning_rate": 0.013492063492063493,
      "loss": 0.0029,
      "step": 4770
    },
    {
      "epoch": 34.14,
      "learning_rate": 0.013403880070546737,
      "loss": 0.003,
      "step": 4780
    },
    {
      "epoch": 34.21,
      "learning_rate": 0.013315696649029983,
      "loss": 0.003,
      "step": 4790
    },
    {
      "epoch": 34.29,
      "learning_rate": 0.013227513227513227,
      "loss": 0.0032,
      "step": 4800
    },
    {
      "epoch": 34.36,
      "learning_rate": 0.013139329805996473,
      "loss": 0.0029,
      "step": 4810
    },
    {
      "epoch": 34.43,
      "learning_rate": 0.013051146384479718,
      "loss": 0.0034,
      "step": 4820
    },
    {
      "epoch": 34.5,
      "learning_rate": 0.012962962962962963,
      "loss": 0.0032,
      "step": 4830
    },
    {
      "epoch": 34.57,
      "learning_rate": 0.012874779541446208,
      "loss": 0.0035,
      "step": 4840
    },
    {
      "epoch": 34.64,
      "learning_rate": 0.012786596119929453,
      "loss": 0.003,
      "step": 4850
    },
    {
      "epoch": 34.71,
      "learning_rate": 0.012698412698412698,
      "loss": 0.0029,
      "step": 4860
    },
    {
      "epoch": 34.79,
      "learning_rate": 0.012610229276895944,
      "loss": 0.0031,
      "step": 4870
    },
    {
      "epoch": 34.86,
      "learning_rate": 0.012522045855379188,
      "loss": 0.0031,
      "step": 4880
    },
    {
      "epoch": 34.93,
      "learning_rate": 0.012433862433862434,
      "loss": 0.0034,
      "step": 4890
    },
    {
      "epoch": 35.0,
      "learning_rate": 0.012345679012345678,
      "loss": 0.0028,
      "step": 4900
    },
    {
      "epoch": 35.07,
      "learning_rate": 0.012257495590828924,
      "loss": 0.0033,
      "step": 4910
    },
    {
      "epoch": 35.14,
      "learning_rate": 0.01216931216931217,
      "loss": 0.0031,
      "step": 4920
    },
    {
      "epoch": 35.21,
      "learning_rate": 0.012081128747795414,
      "loss": 0.0032,
      "step": 4930
    },
    {
      "epoch": 35.29,
      "learning_rate": 0.01199294532627866,
      "loss": 0.0034,
      "step": 4940
    },
    {
      "epoch": 35.36,
      "learning_rate": 0.011904761904761904,
      "loss": 0.0031,
      "step": 4950
    },
    {
      "epoch": 35.43,
      "learning_rate": 0.01181657848324515,
      "loss": 0.003,
      "step": 4960
    },
    {
      "epoch": 35.5,
      "learning_rate": 0.011728395061728396,
      "loss": 0.003,
      "step": 4970
    },
    {
      "epoch": 35.57,
      "learning_rate": 0.01164021164021164,
      "loss": 0.003,
      "step": 4980
    },
    {
      "epoch": 35.64,
      "learning_rate": 0.011552028218694886,
      "loss": 0.003,
      "step": 4990
    },
    {
      "epoch": 35.71,
      "learning_rate": 0.01146384479717813,
      "loss": 0.0033,
      "step": 5000
    },
    {
      "epoch": 35.79,
      "learning_rate": 0.011375661375661376,
      "loss": 0.0032,
      "step": 5010
    },
    {
      "epoch": 35.86,
      "learning_rate": 0.011287477954144622,
      "loss": 0.003,
      "step": 5020
    },
    {
      "epoch": 35.93,
      "learning_rate": 0.011199294532627866,
      "loss": 0.003,
      "step": 5030
    },
    {
      "epoch": 36.0,
      "learning_rate": 0.011111111111111112,
      "loss": 0.0032,
      "step": 5040
    },
    {
      "epoch": 36.07,
      "learning_rate": 0.011022927689594356,
      "loss": 0.0034,
      "step": 5050
    },
    {
      "epoch": 36.14,
      "learning_rate": 0.010934744268077601,
      "loss": 0.0029,
      "step": 5060
    },
    {
      "epoch": 36.21,
      "learning_rate": 0.010846560846560847,
      "loss": 0.0032,
      "step": 5070
    },
    {
      "epoch": 36.29,
      "learning_rate": 0.010758377425044091,
      "loss": 0.0029,
      "step": 5080
    },
    {
      "epoch": 36.36,
      "learning_rate": 0.010670194003527337,
      "loss": 0.0031,
      "step": 5090
    },
    {
      "epoch": 36.43,
      "learning_rate": 0.010582010582010581,
      "loss": 0.0036,
      "step": 5100
    },
    {
      "epoch": 36.5,
      "learning_rate": 0.010493827160493827,
      "loss": 0.003,
      "step": 5110
    },
    {
      "epoch": 36.57,
      "learning_rate": 0.010405643738977073,
      "loss": 0.0031,
      "step": 5120
    },
    {
      "epoch": 36.64,
      "learning_rate": 0.010317460317460317,
      "loss": 0.0031,
      "step": 5130
    },
    {
      "epoch": 36.71,
      "learning_rate": 0.010229276895943563,
      "loss": 0.0035,
      "step": 5140
    },
    {
      "epoch": 36.79,
      "learning_rate": 0.010141093474426807,
      "loss": 0.003,
      "step": 5150
    },
    {
      "epoch": 36.86,
      "learning_rate": 0.010052910052910053,
      "loss": 0.0029,
      "step": 5160
    },
    {
      "epoch": 36.93,
      "learning_rate": 0.009964726631393299,
      "loss": 0.0029,
      "step": 5170
    },
    {
      "epoch": 37.0,
      "learning_rate": 0.009876543209876543,
      "loss": 0.0031,
      "step": 5180
    },
    {
      "epoch": 37.07,
      "learning_rate": 0.009788359788359789,
      "loss": 0.0035,
      "step": 5190
    },
    {
      "epoch": 37.14,
      "learning_rate": 0.009700176366843033,
      "loss": 0.0034,
      "step": 5200
    },
    {
      "epoch": 37.21,
      "learning_rate": 0.009611992945326279,
      "loss": 0.003,
      "step": 5210
    },
    {
      "epoch": 37.29,
      "learning_rate": 0.009523809523809525,
      "loss": 0.0031,
      "step": 5220
    },
    {
      "epoch": 37.36,
      "learning_rate": 0.009435626102292769,
      "loss": 0.003,
      "step": 5230
    },
    {
      "epoch": 37.43,
      "learning_rate": 0.009347442680776015,
      "loss": 0.0032,
      "step": 5240
    },
    {
      "epoch": 37.5,
      "learning_rate": 0.009259259259259259,
      "loss": 0.003,
      "step": 5250
    },
    {
      "epoch": 37.57,
      "learning_rate": 0.009171075837742505,
      "loss": 0.0032,
      "step": 5260
    },
    {
      "epoch": 37.64,
      "learning_rate": 0.009082892416225749,
      "loss": 0.0029,
      "step": 5270
    },
    {
      "epoch": 37.71,
      "learning_rate": 0.008994708994708995,
      "loss": 0.0029,
      "step": 5280
    },
    {
      "epoch": 37.79,
      "learning_rate": 0.00890652557319224,
      "loss": 0.0034,
      "step": 5290
    },
    {
      "epoch": 37.86,
      "learning_rate": 0.008818342151675485,
      "loss": 0.0029,
      "step": 5300
    },
    {
      "epoch": 37.93,
      "learning_rate": 0.00873015873015873,
      "loss": 0.0027,
      "step": 5310
    },
    {
      "epoch": 38.0,
      "learning_rate": 0.008641975308641974,
      "loss": 0.0027,
      "step": 5320
    },
    {
      "epoch": 38.07,
      "learning_rate": 0.00855379188712522,
      "loss": 0.0029,
      "step": 5330
    },
    {
      "epoch": 38.14,
      "learning_rate": 0.008465608465608466,
      "loss": 0.0029,
      "step": 5340
    },
    {
      "epoch": 38.21,
      "learning_rate": 0.00837742504409171,
      "loss": 0.0029,
      "step": 5350
    },
    {
      "epoch": 38.29,
      "learning_rate": 0.008289241622574956,
      "loss": 0.0029,
      "step": 5360
    },
    {
      "epoch": 38.36,
      "learning_rate": 0.0082010582010582,
      "loss": 0.0033,
      "step": 5370
    },
    {
      "epoch": 38.43,
      "learning_rate": 0.008112874779541446,
      "loss": 0.0027,
      "step": 5380
    },
    {
      "epoch": 38.5,
      "learning_rate": 0.008024691358024692,
      "loss": 0.0033,
      "step": 5390
    },
    {
      "epoch": 38.57,
      "learning_rate": 0.007936507936507936,
      "loss": 0.0027,
      "step": 5400
    },
    {
      "epoch": 38.64,
      "learning_rate": 0.007848324514991182,
      "loss": 0.0029,
      "step": 5410
    },
    {
      "epoch": 38.71,
      "learning_rate": 0.007760141093474427,
      "loss": 0.0034,
      "step": 5420
    },
    {
      "epoch": 38.79,
      "learning_rate": 0.007671957671957672,
      "loss": 0.0032,
      "step": 5430
    },
    {
      "epoch": 38.86,
      "learning_rate": 0.007583774250440917,
      "loss": 0.0033,
      "step": 5440
    },
    {
      "epoch": 38.93,
      "learning_rate": 0.007495590828924162,
      "loss": 0.0031,
      "step": 5450
    },
    {
      "epoch": 39.0,
      "learning_rate": 0.007407407407407408,
      "loss": 0.0031,
      "step": 5460
    },
    {
      "epoch": 39.07,
      "learning_rate": 0.007319223985890653,
      "loss": 0.0033,
      "step": 5470
    },
    {
      "epoch": 39.14,
      "learning_rate": 0.007231040564373898,
      "loss": 0.0029,
      "step": 5480
    },
    {
      "epoch": 39.21,
      "learning_rate": 0.007142857142857143,
      "loss": 0.0029,
      "step": 5490
    },
    {
      "epoch": 39.29,
      "learning_rate": 0.007054673721340388,
      "loss": 0.003,
      "step": 5500
    },
    {
      "epoch": 39.36,
      "learning_rate": 0.0069664902998236335,
      "loss": 0.0026,
      "step": 5510
    },
    {
      "epoch": 39.43,
      "learning_rate": 0.0068783068783068784,
      "loss": 0.0029,
      "step": 5520
    },
    {
      "epoch": 39.5,
      "learning_rate": 0.006790123456790123,
      "loss": 0.0032,
      "step": 5530
    },
    {
      "epoch": 39.57,
      "learning_rate": 0.006701940035273368,
      "loss": 0.0029,
      "step": 5540
    },
    {
      "epoch": 39.64,
      "learning_rate": 0.006613756613756613,
      "loss": 0.003,
      "step": 5550
    },
    {
      "epoch": 39.71,
      "learning_rate": 0.006525573192239859,
      "loss": 0.0029,
      "step": 5560
    },
    {
      "epoch": 39.79,
      "learning_rate": 0.006437389770723104,
      "loss": 0.003,
      "step": 5570
    },
    {
      "epoch": 39.86,
      "learning_rate": 0.006349206349206349,
      "loss": 0.003,
      "step": 5580
    },
    {
      "epoch": 39.93,
      "learning_rate": 0.006261022927689594,
      "loss": 0.0028,
      "step": 5590
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.006172839506172839,
      "loss": 0.0031,
      "step": 5600
    },
    {
      "epoch": 40.07,
      "learning_rate": 0.006084656084656085,
      "loss": 0.003,
      "step": 5610
    },
    {
      "epoch": 40.14,
      "learning_rate": 0.00599647266313933,
      "loss": 0.003,
      "step": 5620
    },
    {
      "epoch": 40.21,
      "learning_rate": 0.005908289241622575,
      "loss": 0.0028,
      "step": 5630
    },
    {
      "epoch": 40.29,
      "learning_rate": 0.00582010582010582,
      "loss": 0.0033,
      "step": 5640
    },
    {
      "epoch": 40.36,
      "learning_rate": 0.005731922398589065,
      "loss": 0.003,
      "step": 5650
    },
    {
      "epoch": 40.43,
      "learning_rate": 0.005643738977072311,
      "loss": 0.0027,
      "step": 5660
    },
    {
      "epoch": 40.5,
      "learning_rate": 0.005555555555555556,
      "loss": 0.0031,
      "step": 5670
    },
    {
      "epoch": 40.57,
      "learning_rate": 0.005467372134038801,
      "loss": 0.0027,
      "step": 5680
    },
    {
      "epoch": 40.64,
      "learning_rate": 0.005379188712522046,
      "loss": 0.0031,
      "step": 5690
    },
    {
      "epoch": 40.71,
      "learning_rate": 0.005291005291005291,
      "loss": 0.0032,
      "step": 5700
    },
    {
      "epoch": 40.79,
      "learning_rate": 0.0052028218694885366,
      "loss": 0.0031,
      "step": 5710
    },
    {
      "epoch": 40.86,
      "learning_rate": 0.0051146384479717815,
      "loss": 0.0028,
      "step": 5720
    },
    {
      "epoch": 40.93,
      "learning_rate": 0.0050264550264550265,
      "loss": 0.0031,
      "step": 5730
    },
    {
      "epoch": 41.0,
      "learning_rate": 0.0049382716049382715,
      "loss": 0.0029,
      "step": 5740
    },
    {
      "epoch": 41.07,
      "learning_rate": 0.0048500881834215165,
      "loss": 0.0029,
      "step": 5750
    },
    {
      "epoch": 41.14,
      "learning_rate": 0.004761904761904762,
      "loss": 0.0027,
      "step": 5760
    },
    {
      "epoch": 41.21,
      "learning_rate": 0.004673721340388007,
      "loss": 0.0026,
      "step": 5770
    },
    {
      "epoch": 41.29,
      "learning_rate": 0.004585537918871252,
      "loss": 0.0033,
      "step": 5780
    },
    {
      "epoch": 41.36,
      "learning_rate": 0.004497354497354497,
      "loss": 0.003,
      "step": 5790
    },
    {
      "epoch": 41.43,
      "learning_rate": 0.004409171075837742,
      "loss": 0.0029,
      "step": 5800
    },
    {
      "epoch": 41.5,
      "learning_rate": 0.004320987654320987,
      "loss": 0.0029,
      "step": 5810
    },
    {
      "epoch": 41.57,
      "learning_rate": 0.004232804232804233,
      "loss": 0.0029,
      "step": 5820
    },
    {
      "epoch": 41.64,
      "learning_rate": 0.004144620811287478,
      "loss": 0.003,
      "step": 5830
    },
    {
      "epoch": 41.71,
      "learning_rate": 0.004056437389770723,
      "loss": 0.0029,
      "step": 5840
    },
    {
      "epoch": 41.79,
      "learning_rate": 0.003968253968253968,
      "loss": 0.003,
      "step": 5850
    },
    {
      "epoch": 41.86,
      "learning_rate": 0.0038800705467372134,
      "loss": 0.0028,
      "step": 5860
    },
    {
      "epoch": 41.93,
      "learning_rate": 0.0037918871252204584,
      "loss": 0.0032,
      "step": 5870
    },
    {
      "epoch": 42.0,
      "learning_rate": 0.003703703703703704,
      "loss": 0.0028,
      "step": 5880
    },
    {
      "epoch": 42.07,
      "learning_rate": 0.003615520282186949,
      "loss": 0.0029,
      "step": 5890
    },
    {
      "epoch": 42.14,
      "learning_rate": 0.003527336860670194,
      "loss": 0.0028,
      "step": 5900
    },
    {
      "epoch": 42.21,
      "learning_rate": 0.0034391534391534392,
      "loss": 0.0028,
      "step": 5910
    },
    {
      "epoch": 42.29,
      "learning_rate": 0.003350970017636684,
      "loss": 0.0027,
      "step": 5920
    },
    {
      "epoch": 42.36,
      "learning_rate": 0.0032627865961199296,
      "loss": 0.003,
      "step": 5930
    },
    {
      "epoch": 42.43,
      "learning_rate": 0.0031746031746031746,
      "loss": 0.003,
      "step": 5940
    },
    {
      "epoch": 42.5,
      "learning_rate": 0.0030864197530864196,
      "loss": 0.0029,
      "step": 5950
    },
    {
      "epoch": 42.57,
      "learning_rate": 0.002998236331569665,
      "loss": 0.0027,
      "step": 5960
    },
    {
      "epoch": 42.64,
      "learning_rate": 0.00291005291005291,
      "loss": 0.0027,
      "step": 5970
    },
    {
      "epoch": 42.71,
      "learning_rate": 0.0028218694885361554,
      "loss": 0.0031,
      "step": 5980
    },
    {
      "epoch": 42.79,
      "learning_rate": 0.0027336860670194004,
      "loss": 0.0029,
      "step": 5990
    },
    {
      "epoch": 42.86,
      "learning_rate": 0.0026455026455026454,
      "loss": 0.0028,
      "step": 6000
    },
    {
      "epoch": 42.93,
      "learning_rate": 0.0025573192239858908,
      "loss": 0.0029,
      "step": 6010
    },
    {
      "epoch": 43.0,
      "learning_rate": 0.0024691358024691358,
      "loss": 0.0028,
      "step": 6020
    },
    {
      "epoch": 43.07,
      "learning_rate": 0.002380952380952381,
      "loss": 0.0026,
      "step": 6030
    },
    {
      "epoch": 43.14,
      "learning_rate": 0.002292768959435626,
      "loss": 0.0026,
      "step": 6040
    },
    {
      "epoch": 43.21,
      "learning_rate": 0.002204585537918871,
      "loss": 0.0028,
      "step": 6050
    },
    {
      "epoch": 43.29,
      "learning_rate": 0.0021164021164021165,
      "loss": 0.003,
      "step": 6060
    },
    {
      "epoch": 43.36,
      "learning_rate": 0.0020282186948853615,
      "loss": 0.003,
      "step": 6070
    },
    {
      "epoch": 43.43,
      "learning_rate": 0.0019400352733686067,
      "loss": 0.0029,
      "step": 6080
    },
    {
      "epoch": 43.5,
      "learning_rate": 0.001851851851851852,
      "loss": 0.0028,
      "step": 6090
    },
    {
      "epoch": 43.57,
      "learning_rate": 0.001763668430335097,
      "loss": 0.0028,
      "step": 6100
    },
    {
      "epoch": 43.64,
      "learning_rate": 0.001675485008818342,
      "loss": 0.0027,
      "step": 6110
    },
    {
      "epoch": 43.71,
      "learning_rate": 0.0015873015873015873,
      "loss": 0.003,
      "step": 6120
    },
    {
      "epoch": 43.79,
      "learning_rate": 0.0014991181657848325,
      "loss": 0.0028,
      "step": 6130
    },
    {
      "epoch": 43.86,
      "learning_rate": 0.0014109347442680777,
      "loss": 0.003,
      "step": 6140
    },
    {
      "epoch": 43.93,
      "learning_rate": 0.0013227513227513227,
      "loss": 0.0028,
      "step": 6150
    },
    {
      "epoch": 44.0,
      "learning_rate": 0.0012345679012345679,
      "loss": 0.003,
      "step": 6160
    },
    {
      "epoch": 44.07,
      "learning_rate": 0.001146384479717813,
      "loss": 0.0028,
      "step": 6170
    },
    {
      "epoch": 44.14,
      "learning_rate": 0.0010582010582010583,
      "loss": 0.0028,
      "step": 6180
    },
    {
      "epoch": 44.21,
      "learning_rate": 0.0009700176366843034,
      "loss": 0.0032,
      "step": 6190
    },
    {
      "epoch": 44.29,
      "learning_rate": 0.0008818342151675485,
      "loss": 0.0027,
      "step": 6200
    },
    {
      "epoch": 44.36,
      "learning_rate": 0.0007936507936507937,
      "loss": 0.003,
      "step": 6210
    },
    {
      "epoch": 44.43,
      "learning_rate": 0.0007054673721340388,
      "loss": 0.0026,
      "step": 6220
    },
    {
      "epoch": 44.5,
      "learning_rate": 0.0006172839506172839,
      "loss": 0.0031,
      "step": 6230
    },
    {
      "epoch": 44.57,
      "learning_rate": 0.0005291005291005291,
      "loss": 0.0028,
      "step": 6240
    },
    {
      "epoch": 44.64,
      "learning_rate": 0.0004409171075837742,
      "loss": 0.003,
      "step": 6250
    },
    {
      "epoch": 44.71,
      "learning_rate": 0.0003527336860670194,
      "loss": 0.0026,
      "step": 6260
    },
    {
      "epoch": 44.79,
      "learning_rate": 0.00026455026455026457,
      "loss": 0.003,
      "step": 6270
    },
    {
      "epoch": 44.86,
      "learning_rate": 0.0001763668430335097,
      "loss": 0.0028,
      "step": 6280
    },
    {
      "epoch": 44.93,
      "learning_rate": 8.818342151675486e-05,
      "loss": 0.0029,
      "step": 6290
    },
    {
      "epoch": 45.0,
      "learning_rate": 0.0,
      "loss": 0.0026,
      "step": 6300
    }
  ],
  "max_steps": 6300,
  "num_train_epochs": 45,
  "total_flos": 7.685698226686525e+18,
  "trial_name": null,
  "trial_params": null
}