{
  "best_metric": 0.7800611413043478,
  "best_model_checkpoint": "wav2vec2-base-lang-id/checkpoint-6930",
  "epoch": 9.99891891891892,
  "eval_steps": 500,
  "global_step": 6930,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014414414414414415,
      "grad_norm": 0.6905537843704224,
      "learning_rate": 4.329004329004329e-06,
      "loss": 3.8062,
      "step": 10
    },
    {
      "epoch": 0.02882882882882883,
      "grad_norm": 0.7706511616706848,
      "learning_rate": 8.658008658008657e-06,
      "loss": 3.8042,
      "step": 20
    },
    {
      "epoch": 0.043243243243243246,
      "grad_norm": 0.6511314511299133,
      "learning_rate": 1.2987012987012986e-05,
      "loss": 3.7979,
      "step": 30
    },
    {
      "epoch": 0.05765765765765766,
      "grad_norm": 0.8054167628288269,
      "learning_rate": 1.7316017316017315e-05,
      "loss": 3.7988,
      "step": 40
    },
    {
      "epoch": 0.07207207207207207,
      "grad_norm": 0.7099344730377197,
      "learning_rate": 2.164502164502164e-05,
      "loss": 3.7913,
      "step": 50
    },
    {
      "epoch": 0.08648648648648649,
      "grad_norm": 0.7330520749092102,
      "learning_rate": 2.5974025974025972e-05,
      "loss": 3.7849,
      "step": 60
    },
    {
      "epoch": 0.1009009009009009,
      "grad_norm": 0.9357172250747681,
      "learning_rate": 3.03030303030303e-05,
      "loss": 3.7714,
      "step": 70
    },
    {
      "epoch": 0.11531531531531532,
      "grad_norm": 0.9950684309005737,
      "learning_rate": 3.463203463203463e-05,
      "loss": 3.7663,
      "step": 80
    },
    {
      "epoch": 0.12972972972972974,
      "grad_norm": 0.892503023147583,
      "learning_rate": 3.896103896103895e-05,
      "loss": 3.7587,
      "step": 90
    },
    {
      "epoch": 0.14414414414414414,
      "grad_norm": 1.1201165914535522,
      "learning_rate": 4.329004329004328e-05,
      "loss": 3.7278,
      "step": 100
    },
    {
      "epoch": 0.15855855855855855,
      "grad_norm": 3.5548183917999268,
      "learning_rate": 4.7619047619047614e-05,
      "loss": 3.7199,
      "step": 110
    },
    {
      "epoch": 0.17297297297297298,
      "grad_norm": 1.2475069761276245,
      "learning_rate": 5.1948051948051944e-05,
      "loss": 3.7173,
      "step": 120
    },
    {
      "epoch": 0.1873873873873874,
      "grad_norm": 1.2236099243164062,
      "learning_rate": 5.627705627705627e-05,
      "loss": 3.6874,
      "step": 130
    },
    {
      "epoch": 0.2018018018018018,
      "grad_norm": 2.6467490196228027,
      "learning_rate": 6.06060606060606e-05,
      "loss": 3.6812,
      "step": 140
    },
    {
      "epoch": 0.21621621621621623,
      "grad_norm": 1.6871509552001953,
      "learning_rate": 6.493506493506494e-05,
      "loss": 3.681,
      "step": 150
    },
    {
      "epoch": 0.23063063063063063,
      "grad_norm": 1.58966863155365,
      "learning_rate": 6.926406926406926e-05,
      "loss": 3.6538,
      "step": 160
    },
    {
      "epoch": 0.24504504504504504,
      "grad_norm": 1.7775413990020752,
      "learning_rate": 7.359307359307358e-05,
      "loss": 3.6318,
      "step": 170
    },
    {
      "epoch": 0.2594594594594595,
      "grad_norm": 1.705374002456665,
      "learning_rate": 7.79220779220779e-05,
      "loss": 3.6005,
      "step": 180
    },
    {
      "epoch": 0.27387387387387385,
      "grad_norm": 1.4913026094436646,
      "learning_rate": 8.225108225108224e-05,
      "loss": 3.5767,
      "step": 190
    },
    {
      "epoch": 0.2882882882882883,
      "grad_norm": 1.9860706329345703,
      "learning_rate": 8.658008658008657e-05,
      "loss": 3.5326,
      "step": 200
    },
    {
      "epoch": 0.3027027027027027,
      "grad_norm": 2.525665760040283,
      "learning_rate": 9.09090909090909e-05,
      "loss": 3.5866,
      "step": 210
    },
    {
      "epoch": 0.3171171171171171,
      "grad_norm": 2.28623366355896,
      "learning_rate": 9.523809523809523e-05,
      "loss": 3.5222,
      "step": 220
    },
    {
      "epoch": 0.33153153153153153,
      "grad_norm": 2.221623182296753,
      "learning_rate": 9.956709956709956e-05,
      "loss": 3.4756,
      "step": 230
    },
    {
      "epoch": 0.34594594594594597,
      "grad_norm": 1.997885823249817,
      "learning_rate": 0.00010389610389610389,
      "loss": 3.468,
      "step": 240
    },
    {
      "epoch": 0.36036036036036034,
      "grad_norm": 2.61895489692688,
      "learning_rate": 0.00010822510822510823,
      "loss": 3.4109,
      "step": 250
    },
    {
      "epoch": 0.3747747747747748,
      "grad_norm": 2.7327158451080322,
      "learning_rate": 0.00011255411255411254,
      "loss": 3.4612,
      "step": 260
    },
    {
      "epoch": 0.3891891891891892,
      "grad_norm": 3.259901285171509,
      "learning_rate": 0.00011688311688311687,
      "loss": 3.383,
      "step": 270
    },
    {
      "epoch": 0.4036036036036036,
      "grad_norm": 2.1811468601226807,
      "learning_rate": 0.0001212121212121212,
      "loss": 3.2895,
      "step": 280
    },
    {
      "epoch": 0.418018018018018,
      "grad_norm": 3.6430418491363525,
      "learning_rate": 0.00012554112554112555,
      "loss": 3.3342,
      "step": 290
    },
    {
      "epoch": 0.43243243243243246,
      "grad_norm": 2.8431396484375,
      "learning_rate": 0.00012987012987012987,
      "loss": 3.309,
      "step": 300
    },
    {
      "epoch": 0.44684684684684683,
      "grad_norm": 3.1655421257019043,
      "learning_rate": 0.0001341991341991342,
      "loss": 3.1867,
      "step": 310
    },
    {
      "epoch": 0.46126126126126127,
      "grad_norm": 3.9956753253936768,
      "learning_rate": 0.00013852813852813852,
      "loss": 3.1546,
      "step": 320
    },
    {
      "epoch": 0.4756756756756757,
      "grad_norm": 5.134612560272217,
      "learning_rate": 0.00014285714285714284,
      "loss": 3.2283,
      "step": 330
    },
    {
      "epoch": 0.4900900900900901,
      "grad_norm": 5.635775089263916,
      "learning_rate": 0.00014718614718614716,
      "loss": 3.2093,
      "step": 340
    },
    {
      "epoch": 0.5045045045045045,
      "grad_norm": 5.256202697753906,
      "learning_rate": 0.00015151515151515152,
      "loss": 3.0332,
      "step": 350
    },
    {
      "epoch": 0.518918918918919,
      "grad_norm": 3.5498101711273193,
      "learning_rate": 0.0001558441558441558,
      "loss": 3.144,
      "step": 360
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 5.054722785949707,
      "learning_rate": 0.00016017316017316016,
      "loss": 3.0628,
      "step": 370
    },
    {
      "epoch": 0.5477477477477477,
      "grad_norm": 3.8183388710021973,
      "learning_rate": 0.00016450216450216449,
      "loss": 3.0434,
      "step": 380
    },
    {
      "epoch": 0.5621621621621622,
      "grad_norm": 4.0891547203063965,
      "learning_rate": 0.00016883116883116884,
      "loss": 2.9923,
      "step": 390
    },
    {
      "epoch": 0.5765765765765766,
      "grad_norm": 6.926380634307861,
      "learning_rate": 0.00017316017316017313,
      "loss": 3.0016,
      "step": 400
    },
    {
      "epoch": 0.590990990990991,
      "grad_norm": 3.8529863357543945,
      "learning_rate": 0.00017748917748917746,
      "loss": 2.8764,
      "step": 410
    },
    {
      "epoch": 0.6054054054054054,
      "grad_norm": 4.50208044052124,
      "learning_rate": 0.0001818181818181818,
      "loss": 2.9239,
      "step": 420
    },
    {
      "epoch": 0.6198198198198198,
      "grad_norm": 4.260581970214844,
      "learning_rate": 0.00018614718614718616,
      "loss": 2.8714,
      "step": 430
    },
    {
      "epoch": 0.6342342342342342,
      "grad_norm": 5.0478668212890625,
      "learning_rate": 0.00019047619047619045,
      "loss": 2.8253,
      "step": 440
    },
    {
      "epoch": 0.6486486486486487,
      "grad_norm": 5.0993523597717285,
      "learning_rate": 0.00019480519480519478,
      "loss": 2.8291,
      "step": 450
    },
    {
      "epoch": 0.6630630630630631,
      "grad_norm": 6.65144681930542,
      "learning_rate": 0.00019913419913419913,
      "loss": 2.718,
      "step": 460
    },
    {
      "epoch": 0.6774774774774774,
      "grad_norm": 7.7578840255737305,
      "learning_rate": 0.00020346320346320345,
      "loss": 2.8447,
      "step": 470
    },
    {
      "epoch": 0.6918918918918919,
      "grad_norm": 5.534895896911621,
      "learning_rate": 0.00020779220779220778,
      "loss": 2.7517,
      "step": 480
    },
    {
      "epoch": 0.7063063063063063,
      "grad_norm": 5.055182933807373,
      "learning_rate": 0.0002121212121212121,
      "loss": 2.8779,
      "step": 490
    },
    {
      "epoch": 0.7207207207207207,
      "grad_norm": 5.401959419250488,
      "learning_rate": 0.00021645021645021645,
      "loss": 2.8225,
      "step": 500
    },
    {
      "epoch": 0.7351351351351352,
      "grad_norm": 5.32056188583374,
      "learning_rate": 0.00022077922077922075,
      "loss": 2.9741,
      "step": 510
    },
    {
      "epoch": 0.7495495495495496,
      "grad_norm": 6.170608043670654,
      "learning_rate": 0.00022510822510822507,
      "loss": 2.8599,
      "step": 520
    },
    {
      "epoch": 0.7639639639639639,
      "grad_norm": 4.993892192840576,
      "learning_rate": 0.00022943722943722942,
      "loss": 2.7484,
      "step": 530
    },
    {
      "epoch": 0.7783783783783784,
      "grad_norm": 4.644223690032959,
      "learning_rate": 0.00023376623376623374,
      "loss": 2.8767,
      "step": 540
    },
    {
      "epoch": 0.7927927927927928,
      "grad_norm": 4.942373275756836,
      "learning_rate": 0.00023809523809523807,
      "loss": 2.5846,
      "step": 550
    },
    {
      "epoch": 0.8072072072072072,
      "grad_norm": 5.577380657196045,
      "learning_rate": 0.0002424242424242424,
      "loss": 2.8971,
      "step": 560
    },
    {
      "epoch": 0.8216216216216217,
      "grad_norm": 6.6913838386535645,
      "learning_rate": 0.00024675324675324674,
      "loss": 2.5795,
      "step": 570
    },
    {
      "epoch": 0.836036036036036,
      "grad_norm": 6.057412624359131,
      "learning_rate": 0.0002510822510822511,
      "loss": 2.7484,
      "step": 580
    },
    {
      "epoch": 0.8504504504504504,
      "grad_norm": 5.044222831726074,
      "learning_rate": 0.0002554112554112554,
      "loss": 2.7331,
      "step": 590
    },
    {
      "epoch": 0.8648648648648649,
      "grad_norm": 5.34713888168335,
      "learning_rate": 0.00025930735930735927,
      "loss": 2.8807,
      "step": 600
    },
    {
      "epoch": 0.8792792792792793,
      "grad_norm": 4.528939723968506,
      "learning_rate": 0.0002636363636363636,
      "loss": 2.832,
      "step": 610
    },
    {
      "epoch": 0.8936936936936937,
      "grad_norm": 16.404176712036133,
      "learning_rate": 0.000267965367965368,
      "loss": 2.8185,
      "step": 620
    },
    {
      "epoch": 0.9081081081081082,
      "grad_norm": 5.983865261077881,
      "learning_rate": 0.00027229437229437227,
      "loss": 2.8584,
      "step": 630
    },
    {
      "epoch": 0.9225225225225225,
      "grad_norm": 7.11403751373291,
      "learning_rate": 0.00027662337662337657,
      "loss": 2.5527,
      "step": 640
    },
    {
      "epoch": 0.9369369369369369,
      "grad_norm": 6.044490337371826,
      "learning_rate": 0.0002809523809523809,
      "loss": 2.8024,
      "step": 650
    },
    {
      "epoch": 0.9513513513513514,
      "grad_norm": 5.0461320877075195,
      "learning_rate": 0.00028528138528138527,
      "loss": 2.7907,
      "step": 660
    },
    {
      "epoch": 0.9657657657657658,
      "grad_norm": 5.2283854484558105,
      "learning_rate": 0.00028961038961038956,
      "loss": 2.5418,
      "step": 670
    },
    {
      "epoch": 0.9801801801801802,
      "grad_norm": 7.443022727966309,
      "learning_rate": 0.0002939393939393939,
      "loss": 2.7808,
      "step": 680
    },
    {
      "epoch": 0.9945945945945946,
      "grad_norm": 5.616246700286865,
      "learning_rate": 0.00029826839826839827,
      "loss": 2.58,
      "step": 690
    },
    {
      "epoch": 0.9989189189189189,
      "eval_accuracy": 0.2899116847826087,
      "eval_loss": 2.560854434967041,
      "eval_runtime": 77.4116,
      "eval_samples_per_second": 76.061,
      "eval_steps_per_second": 76.061,
      "step": 693
    },
    {
      "epoch": 1.01009009009009,
      "grad_norm": 5.103923320770264,
      "learning_rate": 0.0002997113997113997,
      "loss": 2.8289,
      "step": 700
    },
    {
      "epoch": 1.0245045045045045,
      "grad_norm": 7.173540115356445,
      "learning_rate": 0.0002992303992303992,
      "loss": 2.6171,
      "step": 710
    },
    {
      "epoch": 1.038918918918919,
      "grad_norm": 6.066854000091553,
      "learning_rate": 0.00029874939874939874,
      "loss": 2.4981,
      "step": 720
    },
    {
      "epoch": 1.0533333333333332,
      "grad_norm": 5.758449554443359,
      "learning_rate": 0.00029826839826839827,
      "loss": 2.6331,
      "step": 730
    },
    {
      "epoch": 1.0677477477477477,
      "grad_norm": 7.043987274169922,
      "learning_rate": 0.00029778739778739773,
      "loss": 2.3763,
      "step": 740
    },
    {
      "epoch": 1.0821621621621622,
      "grad_norm": 7.852530002593994,
      "learning_rate": 0.0002973063973063973,
      "loss": 2.306,
      "step": 750
    },
    {
      "epoch": 1.0965765765765765,
      "grad_norm": 5.415818691253662,
      "learning_rate": 0.0002968253968253968,
      "loss": 2.541,
      "step": 760
    },
    {
      "epoch": 1.110990990990991,
      "grad_norm": 8.213626861572266,
      "learning_rate": 0.00029639249639249637,
      "loss": 2.5493,
      "step": 770
    },
    {
      "epoch": 1.1254054054054055,
      "grad_norm": 5.7650604248046875,
      "learning_rate": 0.0002959114959114959,
      "loss": 2.435,
      "step": 780
    },
    {
      "epoch": 1.1398198198198197,
      "grad_norm": 6.148678302764893,
      "learning_rate": 0.0002954304954304954,
      "loss": 2.2942,
      "step": 790
    },
    {
      "epoch": 1.1542342342342342,
      "grad_norm": 6.500151634216309,
      "learning_rate": 0.00029494949494949493,
      "loss": 2.4808,
      "step": 800
    },
    {
      "epoch": 1.1686486486486487,
      "grad_norm": 7.507975101470947,
      "learning_rate": 0.00029446849446849446,
      "loss": 2.3768,
      "step": 810
    },
    {
      "epoch": 1.183063063063063,
      "grad_norm": 5.838041305541992,
      "learning_rate": 0.000293987493987494,
      "loss": 2.2174,
      "step": 820
    },
    {
      "epoch": 1.1974774774774775,
      "grad_norm": 8.22685718536377,
      "learning_rate": 0.00029350649350649345,
      "loss": 2.3818,
      "step": 830
    },
    {
      "epoch": 1.211891891891892,
      "grad_norm": 7.74230432510376,
      "learning_rate": 0.000293025493025493,
      "loss": 2.373,
      "step": 840
    },
    {
      "epoch": 1.2263063063063062,
      "grad_norm": 5.325120449066162,
      "learning_rate": 0.0002925444925444925,
      "loss": 2.1088,
      "step": 850
    },
    {
      "epoch": 1.2407207207207207,
      "grad_norm": 9.974205017089844,
      "learning_rate": 0.000292063492063492,
      "loss": 2.1565,
      "step": 860
    },
    {
      "epoch": 1.2551351351351352,
      "grad_norm": 7.374603748321533,
      "learning_rate": 0.00029158249158249154,
      "loss": 2.2818,
      "step": 870
    },
    {
      "epoch": 1.2695495495495495,
      "grad_norm": 7.026382923126221,
      "learning_rate": 0.00029110149110149106,
      "loss": 2.2698,
      "step": 880
    },
    {
      "epoch": 1.283963963963964,
      "grad_norm": 5.601807117462158,
      "learning_rate": 0.0002906204906204906,
      "loss": 2.3703,
      "step": 890
    },
    {
      "epoch": 1.2983783783783784,
      "grad_norm": 8.074653625488281,
      "learning_rate": 0.0002901394901394901,
      "loss": 2.1841,
      "step": 900
    },
    {
      "epoch": 1.3127927927927927,
      "grad_norm": 6.700265407562256,
      "learning_rate": 0.00028965848965848963,
      "loss": 2.2991,
      "step": 910
    },
    {
      "epoch": 1.3272072072072072,
      "grad_norm": 6.583888530731201,
      "learning_rate": 0.00028917748917748915,
      "loss": 2.1876,
      "step": 920
    },
    {
      "epoch": 1.3416216216216217,
      "grad_norm": 6.654956340789795,
      "learning_rate": 0.0002886964886964887,
      "loss": 2.236,
      "step": 930
    },
    {
      "epoch": 1.356036036036036,
      "grad_norm": 7.209681987762451,
      "learning_rate": 0.0002882154882154882,
      "loss": 2.2905,
      "step": 940
    },
    {
      "epoch": 1.3704504504504504,
      "grad_norm": 6.927433490753174,
      "learning_rate": 0.0002877344877344877,
      "loss": 2.128,
      "step": 950
    },
    {
      "epoch": 1.384864864864865,
      "grad_norm": 6.913885593414307,
      "learning_rate": 0.00028725348725348724,
      "loss": 2.1266,
      "step": 960
    },
    {
      "epoch": 1.3992792792792792,
      "grad_norm": 7.2124104499816895,
      "learning_rate": 0.0002867724867724867,
      "loss": 2.2429,
      "step": 970
    },
    {
      "epoch": 1.4136936936936937,
      "grad_norm": 5.6674909591674805,
      "learning_rate": 0.0002862914862914863,
      "loss": 2.1822,
      "step": 980
    },
    {
      "epoch": 1.4281081081081082,
      "grad_norm": 8.133776664733887,
      "learning_rate": 0.00028581048581048576,
      "loss": 2.2497,
      "step": 990
    },
    {
      "epoch": 1.4425225225225224,
      "grad_norm": 7.798655986785889,
      "learning_rate": 0.00028532948532948533,
      "loss": 2.3079,
      "step": 1000
    },
    {
      "epoch": 1.456936936936937,
      "grad_norm": 6.923213005065918,
      "learning_rate": 0.0002848484848484848,
      "loss": 2.107,
      "step": 1010
    },
    {
      "epoch": 1.4713513513513514,
      "grad_norm": 6.921584606170654,
      "learning_rate": 0.0002843674843674843,
      "loss": 2.1889,
      "step": 1020
    },
    {
      "epoch": 1.4857657657657657,
      "grad_norm": 5.599717617034912,
      "learning_rate": 0.0002838864838864839,
      "loss": 1.8586,
      "step": 1030
    },
    {
      "epoch": 1.5001801801801802,
      "grad_norm": 7.792572975158691,
      "learning_rate": 0.00028340548340548337,
      "loss": 2.0418,
      "step": 1040
    },
    {
      "epoch": 1.5145945945945947,
      "grad_norm": 6.6426191329956055,
      "learning_rate": 0.0002829244829244829,
      "loss": 2.0239,
      "step": 1050
    },
    {
      "epoch": 1.529009009009009,
      "grad_norm": 8.6935396194458,
      "learning_rate": 0.0002824434824434824,
      "loss": 2.1719,
      "step": 1060
    },
    {
      "epoch": 1.5434234234234234,
      "grad_norm": 6.348733901977539,
      "learning_rate": 0.00028196248196248194,
      "loss": 2.0955,
      "step": 1070
    },
    {
      "epoch": 1.557837837837838,
      "grad_norm": 7.633394241333008,
      "learning_rate": 0.00028148148148148146,
      "loss": 1.9902,
      "step": 1080
    },
    {
      "epoch": 1.5722522522522522,
      "grad_norm": 7.38104248046875,
      "learning_rate": 0.000281000481000481,
      "loss": 2.2128,
      "step": 1090
    },
    {
      "epoch": 1.5866666666666667,
      "grad_norm": 10.188706398010254,
      "learning_rate": 0.0002805194805194805,
      "loss": 2.2774,
      "step": 1100
    },
    {
      "epoch": 1.6010810810810812,
      "grad_norm": 7.001370906829834,
      "learning_rate": 0.00028003848003848,
      "loss": 2.0046,
      "step": 1110
    },
    {
      "epoch": 1.6154954954954954,
      "grad_norm": 10.873891830444336,
      "learning_rate": 0.00027955747955747955,
      "loss": 1.9857,
      "step": 1120
    },
    {
      "epoch": 1.62990990990991,
      "grad_norm": 8.734395027160645,
      "learning_rate": 0.000279076479076479,
      "loss": 2.1417,
      "step": 1130
    },
    {
      "epoch": 1.6443243243243244,
      "grad_norm": 7.794427394866943,
      "learning_rate": 0.0002785954785954786,
      "loss": 2.0302,
      "step": 1140
    },
    {
      "epoch": 1.6587387387387387,
      "grad_norm": 6.246840476989746,
      "learning_rate": 0.00027811447811447806,
      "loss": 2.0361,
      "step": 1150
    },
    {
      "epoch": 1.6731531531531532,
      "grad_norm": 6.608737468719482,
      "learning_rate": 0.0002776334776334776,
      "loss": 1.934,
      "step": 1160
    },
    {
      "epoch": 1.6875675675675677,
      "grad_norm": 6.974606990814209,
      "learning_rate": 0.00027715247715247716,
      "loss": 1.9093,
      "step": 1170
    },
    {
      "epoch": 1.701981981981982,
      "grad_norm": 5.2031025886535645,
      "learning_rate": 0.00027667147667147663,
      "loss": 2.0108,
      "step": 1180
    },
    {
      "epoch": 1.7163963963963964,
      "grad_norm": 6.270247459411621,
      "learning_rate": 0.00027619047619047615,
      "loss": 1.96,
      "step": 1190
    },
    {
      "epoch": 1.730810810810811,
      "grad_norm": 7.6708269119262695,
      "learning_rate": 0.0002757094757094757,
      "loss": 2.0533,
      "step": 1200
    },
    {
      "epoch": 1.7452252252252252,
      "grad_norm": 6.110954761505127,
      "learning_rate": 0.0002752284752284752,
      "loss": 2.0165,
      "step": 1210
    },
    {
      "epoch": 1.7596396396396397,
      "grad_norm": 11.943398475646973,
      "learning_rate": 0.0002747474747474747,
      "loss": 1.9858,
      "step": 1220
    },
    {
      "epoch": 1.7740540540540541,
      "grad_norm": 9.894146919250488,
      "learning_rate": 0.00027426647426647424,
      "loss": 1.8092,
      "step": 1230
    },
    {
      "epoch": 1.7884684684684684,
      "grad_norm": 9.148683547973633,
      "learning_rate": 0.00027378547378547377,
      "loss": 2.1166,
      "step": 1240
    },
    {
      "epoch": 1.802882882882883,
      "grad_norm": 5.730564117431641,
      "learning_rate": 0.0002733044733044733,
      "loss": 2.2126,
      "step": 1250
    },
    {
      "epoch": 1.8172972972972974,
      "grad_norm": 5.5765700340271,
      "learning_rate": 0.0002728234728234728,
      "loss": 1.9804,
      "step": 1260
    },
    {
      "epoch": 1.8317117117117117,
      "grad_norm": 8.557186126708984,
      "learning_rate": 0.0002723424723424723,
      "loss": 1.8937,
      "step": 1270
    },
    {
      "epoch": 1.8461261261261261,
      "grad_norm": 9.299976348876953,
      "learning_rate": 0.00027186147186147186,
      "loss": 1.8725,
      "step": 1280
    },
    {
      "epoch": 1.8605405405405406,
      "grad_norm": 7.710302352905273,
      "learning_rate": 0.0002713804713804714,
      "loss": 1.8815,
      "step": 1290
    },
    {
      "epoch": 1.874954954954955,
      "grad_norm": 9.544638633728027,
      "learning_rate": 0.00027089947089947085,
      "loss": 1.6686,
      "step": 1300
    },
    {
      "epoch": 1.8893693693693694,
      "grad_norm": 7.977646350860596,
      "learning_rate": 0.0002704184704184704,
      "loss": 1.7806,
      "step": 1310
    },
    {
      "epoch": 1.9037837837837839,
      "grad_norm": 6.36674690246582,
      "learning_rate": 0.0002699374699374699,
      "loss": 1.901,
      "step": 1320
    },
    {
      "epoch": 1.9181981981981981,
      "grad_norm": 8.958263397216797,
      "learning_rate": 0.00026945646945646947,
      "loss": 1.7473,
      "step": 1330
    },
    {
      "epoch": 1.9326126126126126,
      "grad_norm": 6.929313659667969,
      "learning_rate": 0.00026897546897546894,
      "loss": 1.8249,
      "step": 1340
    },
    {
      "epoch": 1.9470270270270271,
      "grad_norm": 9.395625114440918,
      "learning_rate": 0.00026849446849446846,
      "loss": 1.6475,
      "step": 1350
    },
    {
      "epoch": 1.9614414414414414,
      "grad_norm": 7.027394771575928,
      "learning_rate": 0.000268013468013468,
      "loss": 1.7785,
      "step": 1360
    },
    {
      "epoch": 1.9758558558558559,
      "grad_norm": 6.9065141677856445,
      "learning_rate": 0.0002675324675324675,
      "loss": 1.7738,
      "step": 1370
    },
    {
      "epoch": 1.9902702702702704,
      "grad_norm": 6.521425724029541,
      "learning_rate": 0.00026705146705146703,
      "loss": 1.8581,
      "step": 1380
    },
    {
      "epoch": 1.998918918918919,
      "eval_accuracy": 0.4008152173913043,
      "eval_loss": 2.1485719680786133,
      "eval_runtime": 71.8289,
      "eval_samples_per_second": 81.973,
      "eval_steps_per_second": 81.973,
      "step": 1386
    },
    {
      "epoch": 2.0057657657657657,
      "grad_norm": 7.039322853088379,
      "learning_rate": 0.00026657046657046655,
      "loss": 1.7202,
      "step": 1390
    },
    {
      "epoch": 2.02018018018018,
      "grad_norm": 7.790018558502197,
      "learning_rate": 0.0002660894660894661,
      "loss": 1.5822,
      "step": 1400
    },
    {
      "epoch": 2.0345945945945947,
      "grad_norm": 5.929462909698486,
      "learning_rate": 0.0002656084656084656,
      "loss": 1.5863,
      "step": 1410
    },
    {
      "epoch": 2.049009009009009,
      "grad_norm": 7.027909755706787,
      "learning_rate": 0.0002651274651274651,
      "loss": 1.6649,
      "step": 1420
    },
    {
      "epoch": 2.063423423423423,
      "grad_norm": 8.068297386169434,
      "learning_rate": 0.00026464646464646464,
      "loss": 1.4897,
      "step": 1430
    },
    {
      "epoch": 2.077837837837838,
      "grad_norm": 5.473997592926025,
      "learning_rate": 0.00026416546416546416,
      "loss": 1.3944,
      "step": 1440
    },
    {
      "epoch": 2.092252252252252,
      "grad_norm": 7.900372505187988,
      "learning_rate": 0.0002636844636844637,
      "loss": 1.4896,
      "step": 1450
    },
    {
      "epoch": 2.1066666666666665,
      "grad_norm": 8.289496421813965,
      "learning_rate": 0.00026320346320346316,
      "loss": 1.4708,
      "step": 1460
    },
    {
      "epoch": 2.121081081081081,
      "grad_norm": 8.37695026397705,
      "learning_rate": 0.00026272246272246273,
      "loss": 1.4163,
      "step": 1470
    },
    {
      "epoch": 2.1354954954954954,
      "grad_norm": 7.39870023727417,
      "learning_rate": 0.0002622414622414622,
      "loss": 1.4042,
      "step": 1480
    },
    {
      "epoch": 2.1499099099099097,
      "grad_norm": 5.828113079071045,
      "learning_rate": 0.0002617604617604617,
      "loss": 1.4999,
      "step": 1490
    },
    {
      "epoch": 2.1643243243243244,
      "grad_norm": 9.574938774108887,
      "learning_rate": 0.00026127946127946125,
      "loss": 1.5935,
      "step": 1500
    },
    {
      "epoch": 2.1787387387387387,
      "grad_norm": 5.495487689971924,
      "learning_rate": 0.00026079846079846077,
      "loss": 1.4395,
      "step": 1510
    },
    {
      "epoch": 2.193153153153153,
      "grad_norm": 7.692873477935791,
      "learning_rate": 0.0002603174603174603,
      "loss": 1.5065,
      "step": 1520
    },
    {
      "epoch": 2.2075675675675677,
      "grad_norm": 6.093646049499512,
      "learning_rate": 0.0002598364598364598,
      "loss": 1.2458,
      "step": 1530
    },
    {
      "epoch": 2.221981981981982,
      "grad_norm": 8.352725982666016,
      "learning_rate": 0.00025935545935545934,
      "loss": 1.3617,
      "step": 1540
    },
    {
      "epoch": 2.236396396396396,
      "grad_norm": 7.136700630187988,
      "learning_rate": 0.00025887445887445886,
      "loss": 1.336,
      "step": 1550
    },
    {
      "epoch": 2.250810810810811,
      "grad_norm": 7.038355350494385,
      "learning_rate": 0.0002583934583934584,
      "loss": 1.4329,
      "step": 1560
    },
    {
      "epoch": 2.265225225225225,
      "grad_norm": 8.975871086120605,
      "learning_rate": 0.0002579124579124579,
      "loss": 1.3865,
      "step": 1570
    },
    {
      "epoch": 2.2796396396396394,
      "grad_norm": 7.062686443328857,
      "learning_rate": 0.0002574314574314574,
      "loss": 1.5105,
      "step": 1580
    },
    {
      "epoch": 2.294054054054054,
      "grad_norm": 7.809958457946777,
      "learning_rate": 0.00025695045695045695,
      "loss": 1.5264,
      "step": 1590
    },
    {
      "epoch": 2.3084684684684684,
      "grad_norm": 16.515899658203125,
      "learning_rate": 0.0002564694564694564,
      "loss": 1.3927,
      "step": 1600
    },
    {
      "epoch": 2.3228828828828827,
      "grad_norm": 7.44677209854126,
      "learning_rate": 0.000255988455988456,
      "loss": 1.455,
      "step": 1610
    },
    {
      "epoch": 2.3372972972972974,
      "grad_norm": 8.71230411529541,
      "learning_rate": 0.00025550745550745546,
      "loss": 1.3946,
      "step": 1620
    },
    {
      "epoch": 2.3517117117117117,
      "grad_norm": 13.610152244567871,
      "learning_rate": 0.000255026455026455,
      "loss": 1.518,
      "step": 1630
    },
    {
      "epoch": 2.366126126126126,
      "grad_norm": 6.677915573120117,
      "learning_rate": 0.0002545454545454545,
      "loss": 1.4105,
      "step": 1640
    },
    {
      "epoch": 2.3805405405405407,
      "grad_norm": 8.372217178344727,
      "learning_rate": 0.00025406445406445403,
      "loss": 1.4308,
      "step": 1650
    },
    {
      "epoch": 2.394954954954955,
      "grad_norm": 8.334647178649902,
      "learning_rate": 0.00025358345358345355,
      "loss": 1.5468,
      "step": 1660
    },
    {
      "epoch": 2.409369369369369,
      "grad_norm": 6.818108081817627,
      "learning_rate": 0.0002531024531024531,
      "loss": 1.2425,
      "step": 1670
    },
    {
      "epoch": 2.423783783783784,
      "grad_norm": 8.92575454711914,
      "learning_rate": 0.0002526214526214526,
      "loss": 1.4133,
      "step": 1680
    },
    {
      "epoch": 2.438198198198198,
      "grad_norm": 5.084140300750732,
      "learning_rate": 0.0002521404521404521,
      "loss": 1.4107,
      "step": 1690
    },
    {
      "epoch": 2.4526126126126124,
      "grad_norm": 8.822663307189941,
      "learning_rate": 0.00025165945165945164,
      "loss": 1.3171,
      "step": 1700
    },
    {
      "epoch": 2.467027027027027,
      "grad_norm": 8.303985595703125,
      "learning_rate": 0.00025117845117845117,
      "loss": 1.4098,
      "step": 1710
    },
    {
      "epoch": 2.4814414414414414,
      "grad_norm": 7.201075553894043,
      "learning_rate": 0.0002506974506974507,
      "loss": 1.4737,
      "step": 1720
    },
    {
      "epoch": 2.4958558558558557,
      "grad_norm": 10.009323120117188,
      "learning_rate": 0.0002502164502164502,
      "loss": 1.3885,
      "step": 1730
    },
    {
      "epoch": 2.5102702702702704,
      "grad_norm": 9.489397048950195,
      "learning_rate": 0.0002497354497354497,
      "loss": 1.4509,
      "step": 1740
    },
    {
      "epoch": 2.5246846846846847,
      "grad_norm": 8.345633506774902,
      "learning_rate": 0.00024925444925444926,
      "loss": 1.3857,
      "step": 1750
    },
    {
      "epoch": 2.539099099099099,
      "grad_norm": 7.19754695892334,
      "learning_rate": 0.0002487734487734487,
      "loss": 1.3326,
      "step": 1760
    },
    {
      "epoch": 2.5535135135135136,
      "grad_norm": 7.177884578704834,
      "learning_rate": 0.0002482924482924483,
      "loss": 1.5869,
      "step": 1770
    },
    {
      "epoch": 2.567927927927928,
      "grad_norm": 8.491632461547852,
      "learning_rate": 0.00024785954785954783,
      "loss": 1.4267,
      "step": 1780
    },
    {
      "epoch": 2.5823423423423426,
      "grad_norm": 7.570484161376953,
      "learning_rate": 0.00024737854737854736,
      "loss": 1.2296,
      "step": 1790
    },
    {
      "epoch": 2.596756756756757,
      "grad_norm": 6.172090530395508,
      "learning_rate": 0.0002468975468975469,
      "loss": 1.4785,
      "step": 1800
    },
    {
      "epoch": 2.611171171171171,
      "grad_norm": 7.379034042358398,
      "learning_rate": 0.0002464165464165464,
      "loss": 1.3001,
      "step": 1810
    },
    {
      "epoch": 2.6255855855855854,
      "grad_norm": 10.75610065460205,
      "learning_rate": 0.0002459355459355459,
      "loss": 1.2524,
      "step": 1820
    },
    {
      "epoch": 2.64,
      "grad_norm": 13.247227668762207,
      "learning_rate": 0.00024545454545454545,
      "loss": 1.3659,
      "step": 1830
    },
    {
      "epoch": 2.6544144144144144,
      "grad_norm": 5.634087562561035,
      "learning_rate": 0.00024497354497354497,
      "loss": 1.2232,
      "step": 1840
    },
    {
      "epoch": 2.668828828828829,
      "grad_norm": 10.873608589172363,
      "learning_rate": 0.00024449254449254444,
      "loss": 1.2378,
      "step": 1850
    },
    {
      "epoch": 2.6832432432432434,
      "grad_norm": 5.584901809692383,
      "learning_rate": 0.00024401154401154402,
      "loss": 1.3656,
      "step": 1860
    },
    {
      "epoch": 2.6976576576576576,
      "grad_norm": 7.405895233154297,
      "learning_rate": 0.00024357864357864358,
      "loss": 1.4302,
      "step": 1870
    },
    {
      "epoch": 2.712072072072072,
      "grad_norm": 15.734750747680664,
      "learning_rate": 0.00024309764309764307,
      "loss": 1.4022,
      "step": 1880
    },
    {
      "epoch": 2.7264864864864866,
      "grad_norm": 7.340987682342529,
      "learning_rate": 0.00024261664261664262,
      "loss": 1.6006,
      "step": 1890
    },
    {
      "epoch": 2.740900900900901,
      "grad_norm": 7.674751281738281,
      "learning_rate": 0.00024213564213564212,
      "loss": 1.2135,
      "step": 1900
    },
    {
      "epoch": 2.755315315315315,
      "grad_norm": 6.400728702545166,
      "learning_rate": 0.0002416546416546416,
      "loss": 1.3779,
      "step": 1910
    },
    {
      "epoch": 2.76972972972973,
      "grad_norm": 7.798464298248291,
      "learning_rate": 0.00024117364117364116,
      "loss": 1.2951,
      "step": 1920
    },
    {
      "epoch": 2.784144144144144,
      "grad_norm": 7.321041107177734,
      "learning_rate": 0.00024069264069264068,
      "loss": 1.2799,
      "step": 1930
    },
    {
      "epoch": 2.7985585585585584,
      "grad_norm": 5.981571674346924,
      "learning_rate": 0.00024021164021164018,
      "loss": 1.2317,
      "step": 1940
    },
    {
      "epoch": 2.812972972972973,
      "grad_norm": 8.328124046325684,
      "learning_rate": 0.00023973063973063973,
      "loss": 1.2906,
      "step": 1950
    },
    {
      "epoch": 2.8273873873873874,
      "grad_norm": 8.400471687316895,
      "learning_rate": 0.00023924963924963922,
      "loss": 1.2612,
      "step": 1960
    },
    {
      "epoch": 2.8418018018018016,
      "grad_norm": 9.944376945495605,
      "learning_rate": 0.00023876863876863872,
      "loss": 1.277,
      "step": 1970
    },
    {
      "epoch": 2.8562162162162164,
      "grad_norm": 7.158963680267334,
      "learning_rate": 0.00023828763828763827,
      "loss": 1.1926,
      "step": 1980
    },
    {
      "epoch": 2.8706306306306306,
      "grad_norm": 8.513224601745605,
      "learning_rate": 0.00023780663780663777,
      "loss": 1.3409,
      "step": 1990
    },
    {
      "epoch": 2.885045045045045,
      "grad_norm": 9.025649070739746,
      "learning_rate": 0.00023732563732563731,
      "loss": 1.2527,
      "step": 2000
    },
    {
      "epoch": 2.8994594594594596,
      "grad_norm": 7.170861721038818,
      "learning_rate": 0.00023684463684463684,
      "loss": 1.3714,
      "step": 2010
    },
    {
      "epoch": 2.913873873873874,
      "grad_norm": 8.145925521850586,
      "learning_rate": 0.00023636363636363633,
      "loss": 1.1761,
      "step": 2020
    },
    {
      "epoch": 2.928288288288288,
      "grad_norm": 8.142692565917969,
      "learning_rate": 0.00023588263588263588,
      "loss": 1.2099,
      "step": 2030
    },
    {
      "epoch": 2.942702702702703,
      "grad_norm": 8.127079010009766,
      "learning_rate": 0.00023540163540163538,
      "loss": 1.1979,
      "step": 2040
    },
    {
      "epoch": 2.957117117117117,
      "grad_norm": 11.584084510803223,
      "learning_rate": 0.00023492063492063487,
      "loss": 1.2129,
      "step": 2050
    },
    {
      "epoch": 2.9715315315315314,
      "grad_norm": 9.020166397094727,
      "learning_rate": 0.00023443963443963442,
      "loss": 1.2372,
      "step": 2060
    },
    {
      "epoch": 2.985945945945946,
      "grad_norm": 8.28217887878418,
      "learning_rate": 0.00023395863395863395,
      "loss": 1.3784,
      "step": 2070
    },
    {
      "epoch": 2.998918918918919,
      "eval_accuracy": 0.5665760869565217,
      "eval_loss": 1.5905574560165405,
      "eval_runtime": 70.8152,
      "eval_samples_per_second": 83.146,
      "eval_steps_per_second": 83.146,
      "step": 2079
    },
    {
      "epoch": 3.0014414414414414,
      "grad_norm": 11.510876655578613,
      "learning_rate": 0.00023347763347763344,
      "loss": 1.2626,
      "step": 2080
    },
    {
      "epoch": 3.0158558558558557,
      "grad_norm": 5.197935104370117,
      "learning_rate": 0.000232996632996633,
      "loss": 0.9394,
      "step": 2090
    },
    {
      "epoch": 3.0302702702702704,
      "grad_norm": 6.015242576599121,
      "learning_rate": 0.0002325156325156325,
      "loss": 1.0227,
      "step": 2100
    },
    {
      "epoch": 3.0446846846846847,
      "grad_norm": 9.21528148651123,
      "learning_rate": 0.00023203463203463204,
      "loss": 1.0463,
      "step": 2110
    },
    {
      "epoch": 3.059099099099099,
      "grad_norm": 5.743220329284668,
      "learning_rate": 0.00023155363155363153,
      "loss": 1.1438,
      "step": 2120
    },
    {
      "epoch": 3.0735135135135137,
      "grad_norm": 6.462952136993408,
      "learning_rate": 0.00023107263107263105,
      "loss": 1.0024,
      "step": 2130
    },
    {
      "epoch": 3.087927927927928,
      "grad_norm": 12.120259284973145,
      "learning_rate": 0.00023059163059163058,
      "loss": 0.9826,
      "step": 2140
    },
    {
      "epoch": 3.102342342342342,
      "grad_norm": 12.20144271850586,
      "learning_rate": 0.0002301106301106301,
      "loss": 0.8739,
      "step": 2150
    },
    {
      "epoch": 3.116756756756757,
      "grad_norm": 9.536534309387207,
      "learning_rate": 0.0002296296296296296,
      "loss": 0.9446,
      "step": 2160
    },
    {
      "epoch": 3.131171171171171,
      "grad_norm": 8.851655960083008,
      "learning_rate": 0.00022914862914862915,
      "loss": 0.9573,
      "step": 2170
    },
    {
      "epoch": 3.1455855855855854,
      "grad_norm": 4.498742580413818,
      "learning_rate": 0.00022866762866762864,
      "loss": 0.8506,
      "step": 2180
    },
    {
      "epoch": 3.16,
      "grad_norm": 6.809707164764404,
      "learning_rate": 0.00022818662818662816,
      "loss": 1.0656,
      "step": 2190
    },
    {
      "epoch": 3.1744144144144144,
      "grad_norm": 6.868171215057373,
      "learning_rate": 0.00022770562770562769,
      "loss": 0.9994,
      "step": 2200
    },
    {
      "epoch": 3.1888288288288287,
      "grad_norm": 10.197772979736328,
      "learning_rate": 0.0002272246272246272,
      "loss": 1.037,
      "step": 2210
    },
    {
      "epoch": 3.2032432432432434,
      "grad_norm": 7.1662397384643555,
      "learning_rate": 0.00022674362674362673,
      "loss": 1.012,
      "step": 2220
    },
    {
      "epoch": 3.2176576576576577,
      "grad_norm": 10.103696823120117,
      "learning_rate": 0.00022626262626262625,
      "loss": 0.991,
      "step": 2230
    },
    {
      "epoch": 3.232072072072072,
      "grad_norm": 8.78542423248291,
      "learning_rate": 0.00022578162578162575,
      "loss": 0.9302,
      "step": 2240
    },
    {
      "epoch": 3.2464864864864866,
      "grad_norm": 7.611443519592285,
      "learning_rate": 0.0002253006253006253,
      "loss": 0.8299,
      "step": 2250
    },
    {
      "epoch": 3.260900900900901,
      "grad_norm": 9.050780296325684,
      "learning_rate": 0.0002248196248196248,
      "loss": 0.9967,
      "step": 2260
    },
    {
      "epoch": 3.275315315315315,
      "grad_norm": 6.917296886444092,
      "learning_rate": 0.00022433862433862432,
      "loss": 0.8874,
      "step": 2270
    },
    {
      "epoch": 3.28972972972973,
      "grad_norm": 8.2102689743042,
      "learning_rate": 0.00022385762385762384,
      "loss": 0.9312,
      "step": 2280
    },
    {
      "epoch": 3.304144144144144,
      "grad_norm": 9.818486213684082,
      "learning_rate": 0.00022337662337662336,
      "loss": 1.025,
      "step": 2290
    },
    {
      "epoch": 3.3185585585585584,
      "grad_norm": 6.598074436187744,
      "learning_rate": 0.00022289562289562286,
      "loss": 0.9141,
      "step": 2300
    },
    {
      "epoch": 3.332972972972973,
      "grad_norm": 9.62888240814209,
      "learning_rate": 0.0002224146224146224,
      "loss": 1.0216,
      "step": 2310
    },
    {
      "epoch": 3.3473873873873874,
      "grad_norm": 6.611650466918945,
      "learning_rate": 0.0002219336219336219,
      "loss": 0.9711,
      "step": 2320
    },
    {
      "epoch": 3.3618018018018017,
      "grad_norm": 8.440096855163574,
      "learning_rate": 0.00022145262145262145,
      "loss": 0.9708,
      "step": 2330
    },
    {
      "epoch": 3.3762162162162164,
      "grad_norm": 5.923568248748779,
      "learning_rate": 0.00022097162097162095,
      "loss": 0.8819,
      "step": 2340
    },
    {
      "epoch": 3.3906306306306306,
      "grad_norm": 9.504045486450195,
      "learning_rate": 0.00022049062049062047,
      "loss": 1.0258,
      "step": 2350
    },
    {
      "epoch": 3.405045045045045,
      "grad_norm": 8.018298149108887,
      "learning_rate": 0.00022000962000962,
      "loss": 0.966,
      "step": 2360
    },
    {
      "epoch": 3.4194594594594596,
      "grad_norm": 9.065242767333984,
      "learning_rate": 0.00021952861952861952,
      "loss": 0.9424,
      "step": 2370
    },
    {
      "epoch": 3.433873873873874,
      "grad_norm": 4.324906826019287,
      "learning_rate": 0.000219047619047619,
      "loss": 0.9095,
      "step": 2380
    },
    {
      "epoch": 3.448288288288288,
      "grad_norm": 6.282894611358643,
      "learning_rate": 0.00021856661856661856,
      "loss": 0.7899,
      "step": 2390
    },
    {
      "epoch": 3.462702702702703,
      "grad_norm": 8.898022651672363,
      "learning_rate": 0.00021808561808561806,
      "loss": 1.0555,
      "step": 2400
    },
    {
      "epoch": 3.477117117117117,
      "grad_norm": 7.557389259338379,
      "learning_rate": 0.00021760461760461758,
      "loss": 1.0346,
      "step": 2410
    },
    {
      "epoch": 3.4915315315315314,
      "grad_norm": 7.2191243171691895,
      "learning_rate": 0.0002171236171236171,
      "loss": 1.1356,
      "step": 2420
    },
    {
      "epoch": 3.505945945945946,
      "grad_norm": 8.154009819030762,
      "learning_rate": 0.00021664261664261662,
      "loss": 0.8751,
      "step": 2430
    },
    {
      "epoch": 3.5203603603603604,
      "grad_norm": 8.395825386047363,
      "learning_rate": 0.00021616161616161615,
      "loss": 1.0114,
      "step": 2440
    },
    {
      "epoch": 3.5347747747747746,
      "grad_norm": 5.63392972946167,
      "learning_rate": 0.00021568061568061567,
      "loss": 1.0379,
      "step": 2450
    },
    {
      "epoch": 3.5491891891891894,
      "grad_norm": 10.38432788848877,
      "learning_rate": 0.00021519961519961517,
      "loss": 0.8813,
      "step": 2460
    },
    {
      "epoch": 3.5636036036036036,
      "grad_norm": 6.3172831535339355,
      "learning_rate": 0.00021471861471861472,
      "loss": 0.8634,
      "step": 2470
    },
    {
      "epoch": 3.578018018018018,
      "grad_norm": 5.470358848571777,
      "learning_rate": 0.0002142376142376142,
      "loss": 0.8643,
      "step": 2480
    },
    {
      "epoch": 3.5924324324324326,
      "grad_norm": 7.979945182800293,
      "learning_rate": 0.00021375661375661373,
      "loss": 0.8846,
      "step": 2490
    },
    {
      "epoch": 3.606846846846847,
      "grad_norm": 8.539430618286133,
      "learning_rate": 0.00021327561327561326,
      "loss": 0.8023,
      "step": 2500
    },
    {
      "epoch": 3.621261261261261,
      "grad_norm": 10.614852905273438,
      "learning_rate": 0.00021279461279461278,
      "loss": 0.8827,
      "step": 2510
    },
    {
      "epoch": 3.6356756756756754,
      "grad_norm": 7.796425819396973,
      "learning_rate": 0.00021231361231361227,
      "loss": 0.9302,
      "step": 2520
    },
    {
      "epoch": 3.65009009009009,
      "grad_norm": 7.7120513916015625,
      "learning_rate": 0.00021183261183261182,
      "loss": 0.9051,
      "step": 2530
    },
    {
      "epoch": 3.6645045045045044,
      "grad_norm": 8.441040992736816,
      "learning_rate": 0.00021135161135161132,
      "loss": 0.8423,
      "step": 2540
    },
    {
      "epoch": 3.678918918918919,
      "grad_norm": 9.811504364013672,
      "learning_rate": 0.00021087061087061087,
      "loss": 1.0583,
      "step": 2550
    },
    {
      "epoch": 3.6933333333333334,
      "grad_norm": 7.54157018661499,
      "learning_rate": 0.00021038961038961036,
      "loss": 0.951,
      "step": 2560
    },
    {
      "epoch": 3.7077477477477476,
      "grad_norm": 7.79876184463501,
      "learning_rate": 0.0002099086099086099,
      "loss": 0.9965,
      "step": 2570
    },
    {
      "epoch": 3.722162162162162,
      "grad_norm": 5.9531025886535645,
      "learning_rate": 0.0002094276094276094,
      "loss": 0.9167,
      "step": 2580
    },
    {
      "epoch": 3.7365765765765766,
      "grad_norm": 8.99783706665039,
      "learning_rate": 0.00020894660894660893,
      "loss": 0.8638,
      "step": 2590
    },
    {
      "epoch": 3.750990990990991,
      "grad_norm": 9.056802749633789,
      "learning_rate": 0.00020846560846560843,
      "loss": 0.8231,
      "step": 2600
    },
    {
      "epoch": 3.7654054054054056,
      "grad_norm": 7.3464226722717285,
      "learning_rate": 0.00020798460798460798,
      "loss": 0.8209,
      "step": 2610
    },
    {
      "epoch": 3.77981981981982,
      "grad_norm": 8.303426742553711,
      "learning_rate": 0.00020750360750360747,
      "loss": 0.8386,
      "step": 2620
    },
    {
      "epoch": 3.794234234234234,
      "grad_norm": 10.655036926269531,
      "learning_rate": 0.000207022607022607,
      "loss": 0.9418,
      "step": 2630
    },
    {
      "epoch": 3.8086486486486484,
      "grad_norm": 7.482151508331299,
      "learning_rate": 0.00020654160654160652,
      "loss": 0.8695,
      "step": 2640
    },
    {
      "epoch": 3.823063063063063,
      "grad_norm": 5.685608863830566,
      "learning_rate": 0.00020606060606060604,
      "loss": 0.8369,
      "step": 2650
    },
    {
      "epoch": 3.8374774774774774,
      "grad_norm": 10.795586585998535,
      "learning_rate": 0.0002055796055796056,
      "loss": 1.0052,
      "step": 2660
    },
    {
      "epoch": 3.851891891891892,
      "grad_norm": 6.713515281677246,
      "learning_rate": 0.00020509860509860509,
      "loss": 0.8769,
      "step": 2670
    },
    {
      "epoch": 3.8663063063063063,
      "grad_norm": 7.321280479431152,
      "learning_rate": 0.00020461760461760458,
      "loss": 0.7935,
      "step": 2680
    },
    {
      "epoch": 3.8807207207207206,
      "grad_norm": 10.69791316986084,
      "learning_rate": 0.00020413660413660413,
      "loss": 0.8147,
      "step": 2690
    },
    {
      "epoch": 3.895135135135135,
      "grad_norm": 5.368622779846191,
      "learning_rate": 0.00020365560365560363,
      "loss": 0.73,
      "step": 2700
    },
    {
      "epoch": 3.9095495495495496,
      "grad_norm": 7.696519374847412,
      "learning_rate": 0.00020317460317460315,
      "loss": 0.999,
      "step": 2710
    },
    {
      "epoch": 3.923963963963964,
      "grad_norm": 6.5068278312683105,
      "learning_rate": 0.0002026936026936027,
      "loss": 0.9845,
      "step": 2720
    },
    {
      "epoch": 3.9383783783783786,
      "grad_norm": 11.021740913391113,
      "learning_rate": 0.0002022126022126022,
      "loss": 0.799,
      "step": 2730
    },
    {
      "epoch": 3.952792792792793,
      "grad_norm": 10.628512382507324,
      "learning_rate": 0.0002017316017316017,
      "loss": 0.8632,
      "step": 2740
    },
    {
      "epoch": 3.967207207207207,
      "grad_norm": 4.902791500091553,
      "learning_rate": 0.00020125060125060124,
      "loss": 0.8372,
      "step": 2750
    },
    {
      "epoch": 3.9816216216216214,
      "grad_norm": 4.3632612228393555,
      "learning_rate": 0.00020076960076960074,
      "loss": 0.7952,
      "step": 2760
    },
    {
      "epoch": 3.996036036036036,
      "grad_norm": 7.273792266845703,
      "learning_rate": 0.00020028860028860029,
      "loss": 0.976,
      "step": 2770
    },
    {
      "epoch": 3.998918918918919,
      "eval_accuracy": 0.6317934782608695,
      "eval_loss": 1.4036318063735962,
      "eval_runtime": 71.1319,
      "eval_samples_per_second": 82.776,
      "eval_steps_per_second": 82.776,
      "step": 2772
    },
    {
      "epoch": 4.011531531531531,
      "grad_norm": 6.909998893737793,
      "learning_rate": 0.00019980759980759978,
      "loss": 0.6602,
      "step": 2780
    },
    {
      "epoch": 4.025945945945946,
      "grad_norm": 8.993280410766602,
      "learning_rate": 0.0001993265993265993,
      "loss": 0.6195,
      "step": 2790
    },
    {
      "epoch": 4.04036036036036,
      "grad_norm": 7.916188716888428,
      "learning_rate": 0.00019884559884559885,
      "loss": 0.7457,
      "step": 2800
    },
    {
      "epoch": 4.054774774774775,
      "grad_norm": 7.465798854827881,
      "learning_rate": 0.00019836459836459835,
      "loss": 0.6078,
      "step": 2810
    },
    {
      "epoch": 4.069189189189189,
      "grad_norm": 6.875505447387695,
      "learning_rate": 0.00019788359788359784,
      "loss": 0.6549,
      "step": 2820
    },
    {
      "epoch": 4.083603603603604,
      "grad_norm": 3.8479785919189453,
      "learning_rate": 0.0001974025974025974,
      "loss": 0.6418,
      "step": 2830
    },
    {
      "epoch": 4.098018018018018,
      "grad_norm": 6.455437183380127,
      "learning_rate": 0.0001969215969215969,
      "loss": 0.5954,
      "step": 2840
    },
    {
      "epoch": 4.112432432432432,
      "grad_norm": 6.476559162139893,
      "learning_rate": 0.0001964405964405964,
      "loss": 0.6301,
      "step": 2850
    },
    {
      "epoch": 4.126846846846846,
      "grad_norm": 7.31296968460083,
      "learning_rate": 0.00019595959595959596,
      "loss": 0.6207,
      "step": 2860
    },
    {
      "epoch": 4.141261261261262,
      "grad_norm": 5.897990703582764,
      "learning_rate": 0.00019547859547859546,
      "loss": 0.665,
      "step": 2870
    },
    {
      "epoch": 4.155675675675676,
      "grad_norm": 11.098051071166992,
      "learning_rate": 0.000194997594997595,
      "loss": 0.7145,
      "step": 2880
    },
    {
      "epoch": 4.17009009009009,
      "grad_norm": 8.448944091796875,
      "learning_rate": 0.0001945165945165945,
      "loss": 0.6393,
      "step": 2890
    },
    {
      "epoch": 4.184504504504504,
      "grad_norm": 13.370219230651855,
      "learning_rate": 0.000194035594035594,
      "loss": 0.758,
      "step": 2900
    },
    {
      "epoch": 4.198918918918919,
      "grad_norm": 7.902467727661133,
      "learning_rate": 0.00019355459355459355,
      "loss": 0.6265,
      "step": 2910
    },
    {
      "epoch": 4.213333333333333,
      "grad_norm": 7.578878879547119,
      "learning_rate": 0.00019307359307359307,
      "loss": 0.6021,
      "step": 2920
    },
    {
      "epoch": 4.227747747747748,
      "grad_norm": 4.44959831237793,
      "learning_rate": 0.00019259259259259257,
      "loss": 0.5569,
      "step": 2930
    },
    {
      "epoch": 4.242162162162162,
      "grad_norm": 6.23007345199585,
      "learning_rate": 0.00019211159211159212,
      "loss": 0.5864,
      "step": 2940
    },
    {
      "epoch": 4.256576576576577,
      "grad_norm": 6.711118698120117,
      "learning_rate": 0.0001916305916305916,
      "loss": 0.762,
      "step": 2950
    },
    {
      "epoch": 4.270990990990991,
      "grad_norm": 6.0345892906188965,
      "learning_rate": 0.0001911495911495911,
      "loss": 0.6951,
      "step": 2960
    },
    {
      "epoch": 4.285405405405405,
      "grad_norm": 9.019355773925781,
      "learning_rate": 0.00019066859066859066,
      "loss": 0.6783,
      "step": 2970
    },
    {
      "epoch": 4.299819819819819,
      "grad_norm": 3.4161036014556885,
      "learning_rate": 0.00019018759018759018,
      "loss": 0.5314,
      "step": 2980
    },
    {
      "epoch": 4.314234234234235,
      "grad_norm": 14.645369529724121,
      "learning_rate": 0.00018970658970658967,
      "loss": 0.727,
      "step": 2990
    },
    {
      "epoch": 4.328648648648649,
      "grad_norm": 9.72807502746582,
      "learning_rate": 0.00018922558922558922,
      "loss": 0.8081,
      "step": 3000
    },
    {
      "epoch": 4.343063063063063,
      "grad_norm": 9.566605567932129,
      "learning_rate": 0.00018874458874458872,
      "loss": 0.8377,
      "step": 3010
    },
    {
      "epoch": 4.357477477477477,
      "grad_norm": 6.285200595855713,
      "learning_rate": 0.00018826358826358827,
      "loss": 0.5785,
      "step": 3020
    },
    {
      "epoch": 4.371891891891892,
      "grad_norm": 9.00932788848877,
      "learning_rate": 0.00018778258778258776,
      "loss": 0.6128,
      "step": 3030
    },
    {
      "epoch": 4.386306306306306,
      "grad_norm": 7.807287693023682,
      "learning_rate": 0.0001873015873015873,
      "loss": 0.5746,
      "step": 3040
    },
    {
      "epoch": 4.400720720720721,
      "grad_norm": 9.012112617492676,
      "learning_rate": 0.0001868205868205868,
      "loss": 0.7083,
      "step": 3050
    },
    {
      "epoch": 4.415135135135135,
      "grad_norm": 7.376474380493164,
      "learning_rate": 0.00018633958633958633,
      "loss": 0.5449,
      "step": 3060
    },
    {
      "epoch": 4.42954954954955,
      "grad_norm": 10.350975036621094,
      "learning_rate": 0.00018585858585858583,
      "loss": 0.5831,
      "step": 3070
    },
    {
      "epoch": 4.443963963963964,
      "grad_norm": 12.924885749816895,
      "learning_rate": 0.00018537758537758538,
      "loss": 0.643,
      "step": 3080
    },
    {
      "epoch": 4.458378378378378,
      "grad_norm": 9.871654510498047,
      "learning_rate": 0.00018489658489658487,
      "loss": 0.5769,
      "step": 3090
    },
    {
      "epoch": 4.472792792792792,
      "grad_norm": 10.68826675415039,
      "learning_rate": 0.0001844155844155844,
      "loss": 0.5935,
      "step": 3100
    },
    {
      "epoch": 4.487207207207208,
      "grad_norm": 3.832447052001953,
      "learning_rate": 0.00018393458393458392,
      "loss": 0.6972,
      "step": 3110
    },
    {
      "epoch": 4.501621621621622,
      "grad_norm": 11.571698188781738,
      "learning_rate": 0.00018345358345358344,
      "loss": 0.5323,
      "step": 3120
    },
    {
      "epoch": 4.516036036036036,
      "grad_norm": 6.816562175750732,
      "learning_rate": 0.00018297258297258296,
      "loss": 0.6748,
      "step": 3130
    },
    {
      "epoch": 4.53045045045045,
      "grad_norm": 8.589786529541016,
      "learning_rate": 0.00018249158249158249,
      "loss": 0.5692,
      "step": 3140
    },
    {
      "epoch": 4.544864864864865,
      "grad_norm": 5.961371421813965,
      "learning_rate": 0.00018201058201058198,
      "loss": 0.6498,
      "step": 3150
    },
    {
      "epoch": 4.559279279279279,
      "grad_norm": 5.627412796020508,
      "learning_rate": 0.00018152958152958153,
      "loss": 0.5188,
      "step": 3160
    },
    {
      "epoch": 4.573693693693694,
      "grad_norm": 7.16035270690918,
      "learning_rate": 0.00018104858104858103,
      "loss": 0.7355,
      "step": 3170
    },
    {
      "epoch": 4.588108108108108,
      "grad_norm": 10.82986831665039,
      "learning_rate": 0.00018056758056758055,
      "loss": 0.5609,
      "step": 3180
    },
    {
      "epoch": 4.602522522522523,
      "grad_norm": 4.972907066345215,
      "learning_rate": 0.00018008658008658007,
      "loss": 0.657,
      "step": 3190
    },
    {
      "epoch": 4.616936936936937,
      "grad_norm": 7.4952921867370605,
      "learning_rate": 0.0001796055796055796,
      "loss": 0.7709,
      "step": 3200
    },
    {
      "epoch": 4.631351351351351,
      "grad_norm": 10.458305358886719,
      "learning_rate": 0.0001791245791245791,
      "loss": 0.7469,
      "step": 3210
    },
    {
      "epoch": 4.645765765765765,
      "grad_norm": 7.345424652099609,
      "learning_rate": 0.00017864357864357864,
      "loss": 0.6626,
      "step": 3220
    },
    {
      "epoch": 4.6601801801801805,
      "grad_norm": 8.653913497924805,
      "learning_rate": 0.00017816257816257814,
      "loss": 0.6976,
      "step": 3230
    },
    {
      "epoch": 4.674594594594595,
      "grad_norm": 10.380864143371582,
      "learning_rate": 0.00017768157768157769,
      "loss": 0.717,
      "step": 3240
    },
    {
      "epoch": 4.689009009009009,
      "grad_norm": 7.697448253631592,
      "learning_rate": 0.00017720057720057718,
      "loss": 0.5699,
      "step": 3250
    },
    {
      "epoch": 4.703423423423423,
      "grad_norm": 8.309391975402832,
      "learning_rate": 0.0001767195767195767,
      "loss": 0.6343,
      "step": 3260
    },
    {
      "epoch": 4.717837837837838,
      "grad_norm": 9.288497924804688,
      "learning_rate": 0.00017623857623857623,
      "loss": 0.6441,
      "step": 3270
    },
    {
      "epoch": 4.732252252252252,
      "grad_norm": 6.459592342376709,
      "learning_rate": 0.00017575757575757575,
      "loss": 0.6189,
      "step": 3280
    },
    {
      "epoch": 4.746666666666667,
      "grad_norm": 10.57323932647705,
      "learning_rate": 0.00017527657527657524,
      "loss": 0.6367,
      "step": 3290
    },
    {
      "epoch": 4.761081081081081,
      "grad_norm": 12.577940940856934,
      "learning_rate": 0.0001747955747955748,
      "loss": 0.7408,
      "step": 3300
    },
    {
      "epoch": 4.775495495495496,
      "grad_norm": 3.959545373916626,
      "learning_rate": 0.0001743145743145743,
      "loss": 0.7011,
      "step": 3310
    },
    {
      "epoch": 4.78990990990991,
      "grad_norm": 6.3194074630737305,
      "learning_rate": 0.0001738335738335738,
      "loss": 0.6087,
      "step": 3320
    },
    {
      "epoch": 4.804324324324324,
      "grad_norm": 6.57200288772583,
      "learning_rate": 0.00017335257335257333,
      "loss": 0.5534,
      "step": 3330
    },
    {
      "epoch": 4.818738738738738,
      "grad_norm": 10.793018341064453,
      "learning_rate": 0.00017287157287157286,
      "loss": 0.6547,
      "step": 3340
    },
    {
      "epoch": 4.8331531531531535,
      "grad_norm": 8.22946834564209,
      "learning_rate": 0.00017239057239057238,
      "loss": 0.54,
      "step": 3350
    },
    {
      "epoch": 4.847567567567568,
      "grad_norm": 6.648707389831543,
      "learning_rate": 0.0001719095719095719,
      "loss": 0.6513,
      "step": 3360
    },
    {
      "epoch": 4.861981981981982,
      "grad_norm": 9.310436248779297,
      "learning_rate": 0.0001714285714285714,
      "loss": 0.6666,
      "step": 3370
    },
    {
      "epoch": 4.876396396396396,
      "grad_norm": 7.90936803817749,
      "learning_rate": 0.00017094757094757095,
      "loss": 0.5307,
      "step": 3380
    },
    {
      "epoch": 4.890810810810811,
      "grad_norm": 11.999992370605469,
      "learning_rate": 0.00017046657046657044,
      "loss": 0.6419,
      "step": 3390
    },
    {
      "epoch": 4.905225225225225,
      "grad_norm": 6.814235210418701,
      "learning_rate": 0.00016998556998556997,
      "loss": 0.6351,
      "step": 3400
    },
    {
      "epoch": 4.91963963963964,
      "grad_norm": 5.307042598724365,
      "learning_rate": 0.0001695045695045695,
      "loss": 0.4863,
      "step": 3410
    },
    {
      "epoch": 4.934054054054054,
      "grad_norm": 6.333554744720459,
      "learning_rate": 0.000169023569023569,
      "loss": 0.5675,
      "step": 3420
    },
    {
      "epoch": 4.9484684684684686,
      "grad_norm": 11.619277000427246,
      "learning_rate": 0.0001685425685425685,
      "loss": 0.6109,
      "step": 3430
    },
    {
      "epoch": 4.962882882882883,
      "grad_norm": 9.019371032714844,
      "learning_rate": 0.00016806156806156806,
      "loss": 0.6733,
      "step": 3440
    },
    {
      "epoch": 4.977297297297297,
      "grad_norm": 11.165461540222168,
      "learning_rate": 0.00016758056758056755,
      "loss": 0.6316,
      "step": 3450
    },
    {
      "epoch": 4.991711711711711,
      "grad_norm": 7.608892917633057,
      "learning_rate": 0.0001670995670995671,
      "loss": 0.6109,
      "step": 3460
    },
    {
      "epoch": 4.998918918918919,
      "eval_accuracy": 0.6694972826086957,
      "eval_loss": 1.3022359609603882,
      "eval_runtime": 71.2632,
      "eval_samples_per_second": 82.623,
      "eval_steps_per_second": 82.623,
      "step": 3465
    },
    {
      "epoch": 5.007207207207207,
      "grad_norm": 9.452139854431152,
      "learning_rate": 0.0001666185666185666,
      "loss": 0.5941,
      "step": 3470
    },
    {
      "epoch": 5.021621621621621,
      "grad_norm": 11.466656684875488,
      "learning_rate": 0.00016613756613756612,
      "loss": 0.4384,
      "step": 3480
    },
    {
      "epoch": 5.036036036036036,
      "grad_norm": 4.653087615966797,
      "learning_rate": 0.00016565656565656564,
      "loss": 0.3431,
      "step": 3490
    },
    {
      "epoch": 5.050450450450451,
      "grad_norm": 7.519618034362793,
      "learning_rate": 0.00016517556517556516,
      "loss": 0.3838,
      "step": 3500
    },
    {
      "epoch": 5.064864864864865,
      "grad_norm": 4.581979274749756,
      "learning_rate": 0.00016469456469456466,
      "loss": 0.4935,
      "step": 3510
    },
    {
      "epoch": 5.079279279279279,
      "grad_norm": 6.805906772613525,
      "learning_rate": 0.0001642135642135642,
      "loss": 0.5017,
      "step": 3520
    },
    {
      "epoch": 5.093693693693694,
      "grad_norm": 4.4706130027771,
      "learning_rate": 0.0001637325637325637,
      "loss": 0.3694,
      "step": 3530
    },
    {
      "epoch": 5.108108108108108,
      "grad_norm": 12.303053855895996,
      "learning_rate": 0.00016325156325156323,
      "loss": 0.5018,
      "step": 3540
    },
    {
      "epoch": 5.122522522522522,
      "grad_norm": 8.26274585723877,
      "learning_rate": 0.00016277056277056275,
      "loss": 0.4137,
      "step": 3550
    },
    {
      "epoch": 5.136936936936937,
      "grad_norm": 8.698760986328125,
      "learning_rate": 0.00016228956228956227,
      "loss": 0.4075,
      "step": 3560
    },
    {
      "epoch": 5.151351351351352,
      "grad_norm": 4.8723464012146,
      "learning_rate": 0.0001618085618085618,
      "loss": 0.3868,
      "step": 3570
    },
    {
      "epoch": 5.165765765765766,
      "grad_norm": 15.168949127197266,
      "learning_rate": 0.00016132756132756132,
      "loss": 0.5866,
      "step": 3580
    },
    {
      "epoch": 5.18018018018018,
      "grad_norm": 3.4054131507873535,
      "learning_rate": 0.00016084656084656081,
      "loss": 0.4335,
      "step": 3590
    },
    {
      "epoch": 5.194594594594594,
      "grad_norm": 6.454439640045166,
      "learning_rate": 0.00016036556036556036,
      "loss": 0.3651,
      "step": 3600
    },
    {
      "epoch": 5.209009009009009,
      "grad_norm": 9.079266548156738,
      "learning_rate": 0.00015988455988455986,
      "loss": 0.4362,
      "step": 3610
    },
    {
      "epoch": 5.223423423423424,
      "grad_norm": 5.144644737243652,
      "learning_rate": 0.00015940355940355938,
      "loss": 0.3995,
      "step": 3620
    },
    {
      "epoch": 5.237837837837838,
      "grad_norm": 3.268798828125,
      "learning_rate": 0.0001589225589225589,
      "loss": 0.3449,
      "step": 3630
    },
    {
      "epoch": 5.252252252252252,
      "grad_norm": 8.07151985168457,
      "learning_rate": 0.00015844155844155843,
      "loss": 0.4208,
      "step": 3640
    },
    {
      "epoch": 5.266666666666667,
      "grad_norm": 12.767313003540039,
      "learning_rate": 0.00015796055796055792,
      "loss": 0.6022,
      "step": 3650
    },
    {
      "epoch": 5.281081081081081,
      "grad_norm": 6.390374183654785,
      "learning_rate": 0.00015747955747955747,
      "loss": 0.4398,
      "step": 3660
    },
    {
      "epoch": 5.295495495495495,
      "grad_norm": 9.433565139770508,
      "learning_rate": 0.00015699855699855697,
      "loss": 0.4173,
      "step": 3670
    },
    {
      "epoch": 5.30990990990991,
      "grad_norm": 5.595968723297119,
      "learning_rate": 0.00015651755651755652,
      "loss": 0.3901,
      "step": 3680
    },
    {
      "epoch": 5.324324324324325,
      "grad_norm": 10.680915832519531,
      "learning_rate": 0.000156036556036556,
      "loss": 0.4181,
      "step": 3690
    },
    {
      "epoch": 5.338738738738739,
      "grad_norm": 7.522782325744629,
      "learning_rate": 0.00015555555555555554,
      "loss": 0.4297,
      "step": 3700
    },
    {
      "epoch": 5.353153153153153,
      "grad_norm": 7.204060077667236,
      "learning_rate": 0.00015507455507455509,
      "loss": 0.5376,
      "step": 3710
    },
    {
      "epoch": 5.367567567567567,
      "grad_norm": 5.985680103302002,
      "learning_rate": 0.00015459355459355458,
      "loss": 0.4228,
      "step": 3720
    },
    {
      "epoch": 5.381981981981982,
      "grad_norm": 6.600807189941406,
      "learning_rate": 0.00015411255411255408,
      "loss": 0.385,
      "step": 3730
    },
    {
      "epoch": 5.396396396396397,
      "grad_norm": 7.420990943908691,
      "learning_rate": 0.00015363155363155363,
      "loss": 0.3998,
      "step": 3740
    },
    {
      "epoch": 5.410810810810811,
      "grad_norm": 9.45582389831543,
      "learning_rate": 0.00015315055315055312,
      "loss": 0.4617,
      "step": 3750
    },
    {
      "epoch": 5.425225225225225,
      "grad_norm": 7.066572189331055,
      "learning_rate": 0.00015266955266955264,
      "loss": 0.5216,
      "step": 3760
    },
    {
      "epoch": 5.43963963963964,
      "grad_norm": 8.458218574523926,
      "learning_rate": 0.0001521885521885522,
      "loss": 0.4398,
      "step": 3770
    },
    {
      "epoch": 5.454054054054054,
      "grad_norm": 8.65445327758789,
      "learning_rate": 0.0001517075517075517,
      "loss": 0.412,
      "step": 3780
    },
    {
      "epoch": 5.468468468468468,
      "grad_norm": 6.245201587677002,
      "learning_rate": 0.00015122655122655124,
      "loss": 0.5225,
      "step": 3790
    },
    {
      "epoch": 5.482882882882883,
      "grad_norm": 6.9201226234436035,
      "learning_rate": 0.00015074555074555073,
      "loss": 0.4083,
      "step": 3800
    },
    {
      "epoch": 5.4972972972972975,
      "grad_norm": 5.687572479248047,
      "learning_rate": 0.00015026455026455023,
      "loss": 0.3771,
      "step": 3810
    },
    {
      "epoch": 5.511711711711712,
      "grad_norm": 12.578248977661133,
      "learning_rate": 0.00014978354978354978,
      "loss": 0.3696,
      "step": 3820
    },
    {
      "epoch": 5.526126126126126,
      "grad_norm": 10.053037643432617,
      "learning_rate": 0.0001493025493025493,
      "loss": 0.4423,
      "step": 3830
    },
    {
      "epoch": 5.54054054054054,
      "grad_norm": 8.149930000305176,
      "learning_rate": 0.00014882154882154883,
      "loss": 0.3522,
      "step": 3840
    },
    {
      "epoch": 5.554954954954955,
      "grad_norm": 6.910628318786621,
      "learning_rate": 0.00014834054834054832,
      "loss": 0.4902,
      "step": 3850
    },
    {
      "epoch": 5.569369369369369,
      "grad_norm": 14.113895416259766,
      "learning_rate": 0.00014785954785954784,
      "loss": 0.3764,
      "step": 3860
    },
    {
      "epoch": 5.583783783783784,
      "grad_norm": 4.280562877655029,
      "learning_rate": 0.00014737854737854737,
      "loss": 0.3926,
      "step": 3870
    },
    {
      "epoch": 5.598198198198198,
      "grad_norm": 4.793622970581055,
      "learning_rate": 0.0001468975468975469,
      "loss": 0.4178,
      "step": 3880
    },
    {
      "epoch": 5.612612612612613,
      "grad_norm": 6.7690629959106445,
      "learning_rate": 0.0001464165464165464,
      "loss": 0.4741,
      "step": 3890
    },
    {
      "epoch": 5.627027027027027,
      "grad_norm": 8.833131790161133,
      "learning_rate": 0.00014593554593554593,
      "loss": 0.4319,
      "step": 3900
    },
    {
      "epoch": 5.641441441441441,
      "grad_norm": 9.301470756530762,
      "learning_rate": 0.00014545454545454546,
      "loss": 0.4336,
      "step": 3910
    },
    {
      "epoch": 5.655855855855856,
      "grad_norm": 3.9089972972869873,
      "learning_rate": 0.00014497354497354495,
      "loss": 0.4098,
      "step": 3920
    },
    {
      "epoch": 5.6702702702702705,
      "grad_norm": 4.024796485900879,
      "learning_rate": 0.00014449254449254447,
      "loss": 0.3569,
      "step": 3930
    },
    {
      "epoch": 5.684684684684685,
      "grad_norm": 8.658758163452148,
      "learning_rate": 0.000144011544011544,
      "loss": 0.5245,
      "step": 3940
    },
    {
      "epoch": 5.699099099099099,
      "grad_norm": 9.45702075958252,
      "learning_rate": 0.00014353054353054352,
      "loss": 0.5006,
      "step": 3950
    },
    {
      "epoch": 5.713513513513513,
      "grad_norm": 10.026135444641113,
      "learning_rate": 0.00014304954304954304,
      "loss": 0.4733,
      "step": 3960
    },
    {
      "epoch": 5.727927927927928,
      "grad_norm": 3.080054998397827,
      "learning_rate": 0.00014256854256854257,
      "loss": 0.4647,
      "step": 3970
    },
    {
      "epoch": 5.742342342342342,
      "grad_norm": 11.083988189697266,
      "learning_rate": 0.0001420875420875421,
      "loss": 0.4103,
      "step": 3980
    },
    {
      "epoch": 5.756756756756757,
      "grad_norm": 8.7739839553833,
      "learning_rate": 0.0001416065416065416,
      "loss": 0.4172,
      "step": 3990
    },
    {
      "epoch": 5.771171171171171,
      "grad_norm": 7.252143383026123,
      "learning_rate": 0.0001411255411255411,
      "loss": 0.4011,
      "step": 4000
    },
    {
      "epoch": 5.7855855855855856,
      "grad_norm": 3.579751491546631,
      "learning_rate": 0.00014064454064454063,
      "loss": 0.3757,
      "step": 4010
    },
    {
      "epoch": 5.8,
      "grad_norm": 5.969137191772461,
      "learning_rate": 0.00014016354016354015,
      "loss": 0.4846,
      "step": 4020
    },
    {
      "epoch": 5.814414414414414,
      "grad_norm": 11.756107330322266,
      "learning_rate": 0.00013968253968253967,
      "loss": 0.4652,
      "step": 4030
    },
    {
      "epoch": 5.828828828828829,
      "grad_norm": 6.106459617614746,
      "learning_rate": 0.0001392015392015392,
      "loss": 0.4863,
      "step": 4040
    },
    {
      "epoch": 5.8432432432432435,
      "grad_norm": 7.70770788192749,
      "learning_rate": 0.00013872053872053872,
      "loss": 0.3973,
      "step": 4050
    },
    {
      "epoch": 5.857657657657658,
      "grad_norm": 7.212626934051514,
      "learning_rate": 0.00013823953823953824,
      "loss": 0.496,
      "step": 4060
    },
    {
      "epoch": 5.872072072072072,
      "grad_norm": 7.486503601074219,
      "learning_rate": 0.00013775853775853774,
      "loss": 0.3921,
      "step": 4070
    },
    {
      "epoch": 5.886486486486486,
      "grad_norm": 6.690408706665039,
      "learning_rate": 0.00013727753727753726,
      "loss": 0.3831,
      "step": 4080
    },
    {
      "epoch": 5.900900900900901,
      "grad_norm": 7.277036190032959,
      "learning_rate": 0.00013679653679653678,
      "loss": 0.3916,
      "step": 4090
    },
    {
      "epoch": 5.915315315315315,
      "grad_norm": 9.00048828125,
      "learning_rate": 0.00013636363636363634,
      "loss": 0.5376,
      "step": 4100
    },
    {
      "epoch": 5.92972972972973,
      "grad_norm": 4.73638916015625,
      "learning_rate": 0.00013588263588263586,
      "loss": 0.3302,
      "step": 4110
    },
    {
      "epoch": 5.944144144144144,
      "grad_norm": 9.437019348144531,
      "learning_rate": 0.0001354016354016354,
      "loss": 0.3571,
      "step": 4120
    },
    {
      "epoch": 5.9585585585585585,
      "grad_norm": 4.46668004989624,
      "learning_rate": 0.0001349206349206349,
      "loss": 0.5002,
      "step": 4130
    },
    {
      "epoch": 5.972972972972973,
      "grad_norm": 5.139366626739502,
      "learning_rate": 0.00013443963443963443,
      "loss": 0.3792,
      "step": 4140
    },
    {
      "epoch": 5.987387387387387,
      "grad_norm": 8.221098899841309,
      "learning_rate": 0.00013395863395863395,
      "loss": 0.4357,
      "step": 4150
    },
    {
      "epoch": 5.998918918918919,
      "eval_accuracy": 0.7138247282608695,
      "eval_loss": 1.2386276721954346,
      "eval_runtime": 74.8324,
      "eval_samples_per_second": 78.683,
      "eval_steps_per_second": 78.683,
      "step": 4158
    },
    {
      "epoch": 6.002882882882883,
      "grad_norm": 4.90937614440918,
      "learning_rate": 0.00013347763347763348,
      "loss": 0.3666,
      "step": 4160
    },
    {
      "epoch": 6.017297297297297,
      "grad_norm": 8.738027572631836,
      "learning_rate": 0.00013299663299663297,
      "loss": 0.4189,
      "step": 4170
    },
    {
      "epoch": 6.031711711711711,
      "grad_norm": 7.700376033782959,
      "learning_rate": 0.0001325156325156325,
      "loss": 0.2986,
      "step": 4180
    },
    {
      "epoch": 6.0461261261261265,
      "grad_norm": 7.535659313201904,
      "learning_rate": 0.00013203463203463202,
      "loss": 0.2416,
      "step": 4190
    },
    {
      "epoch": 6.060540540540541,
      "grad_norm": 6.323885440826416,
      "learning_rate": 0.00013155363155363154,
      "loss": 0.2035,
      "step": 4200
    },
    {
      "epoch": 6.074954954954955,
      "grad_norm": 5.111292839050293,
      "learning_rate": 0.00013107263107263106,
      "loss": 0.259,
      "step": 4210
    },
    {
      "epoch": 6.089369369369369,
      "grad_norm": 4.008244037628174,
      "learning_rate": 0.00013059163059163059,
      "loss": 0.2349,
      "step": 4220
    },
    {
      "epoch": 6.103783783783784,
      "grad_norm": 6.876310348510742,
      "learning_rate": 0.0001301106301106301,
      "loss": 0.3054,
      "step": 4230
    },
    {
      "epoch": 6.118198198198198,
      "grad_norm": 7.2992401123046875,
      "learning_rate": 0.0001296296296296296,
      "loss": 0.2878,
      "step": 4240
    },
    {
      "epoch": 6.132612612612613,
      "grad_norm": 4.963804244995117,
      "learning_rate": 0.00012914862914862913,
      "loss": 0.276,
      "step": 4250
    },
    {
      "epoch": 6.147027027027027,
      "grad_norm": 6.088598251342773,
      "learning_rate": 0.00012866762866762865,
      "loss": 0.324,
      "step": 4260
    },
    {
      "epoch": 6.161441441441442,
      "grad_norm": 8.737565040588379,
      "learning_rate": 0.00012818662818662817,
      "loss": 0.2386,
      "step": 4270
    },
    {
      "epoch": 6.175855855855856,
      "grad_norm": 5.8639678955078125,
      "learning_rate": 0.0001277056277056277,
      "loss": 0.3121,
      "step": 4280
    },
    {
      "epoch": 6.19027027027027,
      "grad_norm": 5.675809383392334,
      "learning_rate": 0.00012722462722462722,
      "loss": 0.2746,
      "step": 4290
    },
    {
      "epoch": 6.204684684684684,
      "grad_norm": 5.486160755157471,
      "learning_rate": 0.00012674362674362674,
      "loss": 0.315,
      "step": 4300
    },
    {
      "epoch": 6.2190990990990995,
      "grad_norm": 6.656479835510254,
      "learning_rate": 0.00012626262626262626,
      "loss": 0.2825,
      "step": 4310
    },
    {
      "epoch": 6.233513513513514,
      "grad_norm": 6.042985916137695,
      "learning_rate": 0.00012578162578162576,
      "loss": 0.3193,
      "step": 4320
    },
    {
      "epoch": 6.247927927927928,
      "grad_norm": 9.522024154663086,
      "learning_rate": 0.00012530062530062528,
      "loss": 0.3962,
      "step": 4330
    },
    {
      "epoch": 6.262342342342342,
      "grad_norm": 6.633049964904785,
      "learning_rate": 0.0001248196248196248,
      "loss": 0.3154,
      "step": 4340
    },
    {
      "epoch": 6.276756756756757,
      "grad_norm": 8.572668075561523,
      "learning_rate": 0.00012433862433862433,
      "loss": 0.3352,
      "step": 4350
    },
    {
      "epoch": 6.291171171171171,
      "grad_norm": 8.13693618774414,
      "learning_rate": 0.00012385762385762385,
      "loss": 0.2994,
      "step": 4360
    },
    {
      "epoch": 6.305585585585586,
      "grad_norm": 6.939904689788818,
      "learning_rate": 0.00012337662337662337,
      "loss": 0.2414,
      "step": 4370
    },
    {
      "epoch": 6.32,
      "grad_norm": 5.574888706207275,
      "learning_rate": 0.0001228956228956229,
      "loss": 0.2875,
      "step": 4380
    },
    {
      "epoch": 6.3344144144144146,
      "grad_norm": 5.0649824142456055,
      "learning_rate": 0.0001224146224146224,
      "loss": 0.2667,
      "step": 4390
    },
    {
      "epoch": 6.348828828828829,
      "grad_norm": 7.2256317138671875,
      "learning_rate": 0.00012193362193362193,
      "loss": 0.2765,
      "step": 4400
    },
    {
      "epoch": 6.363243243243243,
      "grad_norm": 4.355258464813232,
      "learning_rate": 0.00012145262145262145,
      "loss": 0.356,
      "step": 4410
    },
    {
      "epoch": 6.377657657657657,
      "grad_norm": 5.110248565673828,
      "learning_rate": 0.00012097162097162096,
      "loss": 0.3033,
      "step": 4420
    },
    {
      "epoch": 6.392072072072072,
      "grad_norm": 7.9375319480896,
      "learning_rate": 0.00012049062049062048,
      "loss": 0.2961,
      "step": 4430
    },
    {
      "epoch": 6.406486486486487,
      "grad_norm": 6.898852825164795,
      "learning_rate": 0.00012000962000962,
      "loss": 0.2319,
      "step": 4440
    },
    {
      "epoch": 6.420900900900901,
      "grad_norm": 6.887888431549072,
      "learning_rate": 0.00011952861952861952,
      "loss": 0.2813,
      "step": 4450
    },
    {
      "epoch": 6.435315315315315,
      "grad_norm": 6.109614849090576,
      "learning_rate": 0.00011904761904761903,
      "loss": 0.1929,
      "step": 4460
    },
    {
      "epoch": 6.44972972972973,
      "grad_norm": 11.821638107299805,
      "learning_rate": 0.00011856661856661856,
      "loss": 0.3329,
      "step": 4470
    },
    {
      "epoch": 6.464144144144144,
      "grad_norm": 2.5805349349975586,
      "learning_rate": 0.00011808561808561808,
      "loss": 0.2021,
      "step": 4480
    },
    {
      "epoch": 6.478558558558559,
      "grad_norm": 6.847564220428467,
      "learning_rate": 0.0001176046176046176,
      "loss": 0.2514,
      "step": 4490
    },
    {
      "epoch": 6.492972972972973,
      "grad_norm": 11.351963996887207,
      "learning_rate": 0.00011712361712361711,
      "loss": 0.2194,
      "step": 4500
    },
    {
      "epoch": 6.5073873873873875,
      "grad_norm": 7.810890197753906,
      "learning_rate": 0.00011664261664261663,
      "loss": 0.2414,
      "step": 4510
    },
    {
      "epoch": 6.521801801801802,
      "grad_norm": 5.470864772796631,
      "learning_rate": 0.00011616161616161616,
      "loss": 0.3492,
      "step": 4520
    },
    {
      "epoch": 6.536216216216216,
      "grad_norm": 6.755211353302002,
      "learning_rate": 0.00011568061568061567,
      "loss": 0.2824,
      "step": 4530
    },
    {
      "epoch": 6.55063063063063,
      "grad_norm": 16.999385833740234,
      "learning_rate": 0.00011519961519961519,
      "loss": 0.3492,
      "step": 4540
    },
    {
      "epoch": 6.565045045045045,
      "grad_norm": 8.815607070922852,
      "learning_rate": 0.00011471861471861471,
      "loss": 0.3004,
      "step": 4550
    },
    {
      "epoch": 6.57945945945946,
      "grad_norm": 3.8612780570983887,
      "learning_rate": 0.00011423761423761423,
      "loss": 0.3197,
      "step": 4560
    },
    {
      "epoch": 6.593873873873874,
      "grad_norm": 6.251058101654053,
      "learning_rate": 0.00011375661375661374,
      "loss": 0.3294,
      "step": 4570
    },
    {
      "epoch": 6.608288288288288,
      "grad_norm": 4.983285903930664,
      "learning_rate": 0.00011327561327561326,
      "loss": 0.257,
      "step": 4580
    },
    {
      "epoch": 6.622702702702703,
      "grad_norm": 6.4629669189453125,
      "learning_rate": 0.00011279461279461279,
      "loss": 0.3786,
      "step": 4590
    },
    {
      "epoch": 6.637117117117117,
      "grad_norm": 5.6129231452941895,
      "learning_rate": 0.00011231361231361231,
      "loss": 0.2482,
      "step": 4600
    },
    {
      "epoch": 6.651531531531532,
      "grad_norm": 5.438108921051025,
      "learning_rate": 0.00011183261183261182,
      "loss": 0.239,
      "step": 4610
    },
    {
      "epoch": 6.665945945945946,
      "grad_norm": 3.0037171840667725,
      "learning_rate": 0.00011135161135161134,
      "loss": 0.2049,
      "step": 4620
    },
    {
      "epoch": 6.6803603603603605,
      "grad_norm": 4.352254390716553,
      "learning_rate": 0.00011087061087061086,
      "loss": 0.3345,
      "step": 4630
    },
    {
      "epoch": 6.694774774774775,
      "grad_norm": 5.797568321228027,
      "learning_rate": 0.00011038961038961037,
      "loss": 0.2837,
      "step": 4640
    },
    {
      "epoch": 6.709189189189189,
      "grad_norm": 9.520432472229004,
      "learning_rate": 0.0001099086099086099,
      "loss": 0.2468,
      "step": 4650
    },
    {
      "epoch": 6.723603603603603,
      "grad_norm": 20.87645149230957,
      "learning_rate": 0.00010942760942760942,
      "loss": 0.3407,
      "step": 4660
    },
    {
      "epoch": 6.738018018018018,
      "grad_norm": 7.918701648712158,
      "learning_rate": 0.00010894660894660894,
      "loss": 0.343,
      "step": 4670
    },
    {
      "epoch": 6.752432432432433,
      "grad_norm": 10.435689926147461,
      "learning_rate": 0.00010846560846560845,
      "loss": 0.1987,
      "step": 4680
    },
    {
      "epoch": 6.766846846846847,
      "grad_norm": 6.764944076538086,
      "learning_rate": 0.00010798460798460797,
      "loss": 0.3136,
      "step": 4690
    },
    {
      "epoch": 6.781261261261261,
      "grad_norm": 4.383102893829346,
      "learning_rate": 0.0001075036075036075,
      "loss": 0.2833,
      "step": 4700
    },
    {
      "epoch": 6.7956756756756755,
      "grad_norm": 4.218920707702637,
      "learning_rate": 0.00010702260702260702,
      "loss": 0.3008,
      "step": 4710
    },
    {
      "epoch": 6.81009009009009,
      "grad_norm": 7.9592084884643555,
      "learning_rate": 0.00010654160654160653,
      "loss": 0.2569,
      "step": 4720
    },
    {
      "epoch": 6.824504504504505,
      "grad_norm": 6.064927577972412,
      "learning_rate": 0.00010606060606060605,
      "loss": 0.2501,
      "step": 4730
    },
    {
      "epoch": 6.838918918918919,
      "grad_norm": 4.929797172546387,
      "learning_rate": 0.00010557960557960557,
      "loss": 0.3194,
      "step": 4740
    },
    {
      "epoch": 6.8533333333333335,
      "grad_norm": 6.3819122314453125,
      "learning_rate": 0.00010509860509860508,
      "loss": 0.2613,
      "step": 4750
    },
    {
      "epoch": 6.867747747747748,
      "grad_norm": 5.554884910583496,
      "learning_rate": 0.0001046176046176046,
      "loss": 0.3246,
      "step": 4760
    },
    {
      "epoch": 6.882162162162162,
      "grad_norm": 4.798628330230713,
      "learning_rate": 0.00010413660413660413,
      "loss": 0.3085,
      "step": 4770
    },
    {
      "epoch": 6.896576576576576,
      "grad_norm": 5.209802150726318,
      "learning_rate": 0.00010365560365560365,
      "loss": 0.2897,
      "step": 4780
    },
    {
      "epoch": 6.910990990990991,
      "grad_norm": 8.532238960266113,
      "learning_rate": 0.00010317460317460316,
      "loss": 0.2757,
      "step": 4790
    },
    {
      "epoch": 6.925405405405406,
      "grad_norm": 5.6922760009765625,
      "learning_rate": 0.00010269360269360268,
      "loss": 0.2209,
      "step": 4800
    },
    {
      "epoch": 6.93981981981982,
      "grad_norm": 3.573279619216919,
      "learning_rate": 0.0001022126022126022,
      "loss": 0.2065,
      "step": 4810
    },
    {
      "epoch": 6.954234234234234,
      "grad_norm": 10.121118545532227,
      "learning_rate": 0.00010173160173160173,
      "loss": 0.2554,
      "step": 4820
    },
    {
      "epoch": 6.9686486486486485,
      "grad_norm": 8.642844200134277,
      "learning_rate": 0.00010125060125060124,
      "loss": 0.2555,
      "step": 4830
    },
    {
      "epoch": 6.983063063063063,
      "grad_norm": 9.771698951721191,
      "learning_rate": 0.00010076960076960076,
      "loss": 0.2784,
      "step": 4840
    },
    {
      "epoch": 6.997477477477478,
      "grad_norm": 5.332848072052002,
      "learning_rate": 0.00010028860028860028,
      "loss": 0.23,
      "step": 4850
    },
    {
      "epoch": 6.998918918918919,
      "eval_accuracy": 0.7221467391304348,
      "eval_loss": 1.3078279495239258,
      "eval_runtime": 72.8787,
      "eval_samples_per_second": 80.792,
      "eval_steps_per_second": 80.792,
      "step": 4851
    },
    {
      "epoch": 7.012972972972973,
      "grad_norm": 6.101404190063477,
      "learning_rate": 9.980759980759979e-05,
      "loss": 0.115,
      "step": 4860
    },
    {
      "epoch": 7.027387387387387,
      "grad_norm": 1.9845027923583984,
      "learning_rate": 9.932659932659931e-05,
      "loss": 0.1413,
      "step": 4870
    },
    {
      "epoch": 7.041801801801801,
      "grad_norm": 4.760300159454346,
      "learning_rate": 9.884559884559883e-05,
      "loss": 0.1826,
      "step": 4880
    },
    {
      "epoch": 7.0562162162162165,
      "grad_norm": 7.768408298492432,
      "learning_rate": 9.836459836459836e-05,
      "loss": 0.1403,
      "step": 4890
    },
    {
      "epoch": 7.070630630630631,
      "grad_norm": 3.1166656017303467,
      "learning_rate": 9.788359788359787e-05,
      "loss": 0.1885,
      "step": 4900
    },
    {
      "epoch": 7.085045045045045,
      "grad_norm": 8.536203384399414,
      "learning_rate": 9.740259740259739e-05,
      "loss": 0.1814,
      "step": 4910
    },
    {
      "epoch": 7.099459459459459,
      "grad_norm": 6.595261096954346,
      "learning_rate": 9.692159692159691e-05,
      "loss": 0.1951,
      "step": 4920
    },
    {
      "epoch": 7.113873873873874,
      "grad_norm": 3.152899980545044,
      "learning_rate": 9.644059644059643e-05,
      "loss": 0.1598,
      "step": 4930
    },
    {
      "epoch": 7.128288288288288,
      "grad_norm": 3.0426697731018066,
      "learning_rate": 9.595959595959594e-05,
      "loss": 0.1868,
      "step": 4940
    },
    {
      "epoch": 7.142702702702703,
      "grad_norm": 6.639180660247803,
      "learning_rate": 9.547859547859547e-05,
      "loss": 0.1429,
      "step": 4950
    },
    {
      "epoch": 7.157117117117117,
      "grad_norm": 6.044598579406738,
      "learning_rate": 9.499759499759499e-05,
      "loss": 0.2205,
      "step": 4960
    },
    {
      "epoch": 7.1715315315315316,
      "grad_norm": 8.558441162109375,
      "learning_rate": 9.45165945165945e-05,
      "loss": 0.1698,
      "step": 4970
    },
    {
      "epoch": 7.185945945945946,
      "grad_norm": 6.437906265258789,
      "learning_rate": 9.403559403559402e-05,
      "loss": 0.1071,
      "step": 4980
    },
    {
      "epoch": 7.20036036036036,
      "grad_norm": 3.9079666137695312,
      "learning_rate": 9.355459355459354e-05,
      "loss": 0.1602,
      "step": 4990
    },
    {
      "epoch": 7.214774774774774,
      "grad_norm": 5.485486030578613,
      "learning_rate": 9.307359307359308e-05,
      "loss": 0.1252,
      "step": 5000
    },
    {
      "epoch": 7.2291891891891895,
      "grad_norm": 1.008537769317627,
      "learning_rate": 9.259259259259257e-05,
      "loss": 0.228,
      "step": 5010
    },
    {
      "epoch": 7.243603603603604,
      "grad_norm": 7.288844585418701,
      "learning_rate": 9.21115921115921e-05,
      "loss": 0.1422,
      "step": 5020
    },
    {
      "epoch": 7.258018018018018,
      "grad_norm": 6.236993312835693,
      "learning_rate": 9.163059163059163e-05,
      "loss": 0.1498,
      "step": 5030
    },
    {
      "epoch": 7.272432432432432,
      "grad_norm": 19.82891845703125,
      "learning_rate": 9.114959114959113e-05,
      "loss": 0.1163,
      "step": 5040
    },
    {
      "epoch": 7.286846846846847,
      "grad_norm": 6.393248081207275,
      "learning_rate": 9.066859066859065e-05,
      "loss": 0.2079,
      "step": 5050
    },
    {
      "epoch": 7.301261261261261,
      "grad_norm": 8.03089714050293,
      "learning_rate": 9.018759018759019e-05,
      "loss": 0.2521,
      "step": 5060
    },
    {
      "epoch": 7.315675675675676,
      "grad_norm": 9.314800262451172,
      "learning_rate": 8.970658970658971e-05,
      "loss": 0.1911,
      "step": 5070
    },
    {
      "epoch": 7.33009009009009,
      "grad_norm": 2.4367496967315674,
      "learning_rate": 8.92255892255892e-05,
      "loss": 0.1687,
      "step": 5080
    },
    {
      "epoch": 7.3445045045045045,
      "grad_norm": 4.91244649887085,
      "learning_rate": 8.874458874458873e-05,
      "loss": 0.1607,
      "step": 5090
    },
    {
      "epoch": 7.358918918918919,
      "grad_norm": 0.31859567761421204,
      "learning_rate": 8.826358826358826e-05,
      "loss": 0.1477,
      "step": 5100
    },
    {
      "epoch": 7.373333333333333,
      "grad_norm": 4.724694728851318,
      "learning_rate": 8.778258778258779e-05,
      "loss": 0.1316,
      "step": 5110
    },
    {
      "epoch": 7.387747747747747,
      "grad_norm": 11.684412956237793,
      "learning_rate": 8.730158730158728e-05,
      "loss": 0.2144,
      "step": 5120
    },
    {
      "epoch": 7.4021621621621625,
      "grad_norm": 3.9598135948181152,
      "learning_rate": 8.682058682058682e-05,
      "loss": 0.1382,
      "step": 5130
    },
    {
      "epoch": 7.416576576576577,
      "grad_norm": 11.097365379333496,
      "learning_rate": 8.633958633958634e-05,
      "loss": 0.1959,
      "step": 5140
    },
    {
      "epoch": 7.430990990990991,
      "grad_norm": 6.260383129119873,
      "learning_rate": 8.585858585858584e-05,
      "loss": 0.138,
      "step": 5150
    },
    {
      "epoch": 7.445405405405405,
      "grad_norm": 10.03337574005127,
      "learning_rate": 8.537758537758537e-05,
      "loss": 0.1688,
      "step": 5160
    },
    {
      "epoch": 7.45981981981982,
      "grad_norm": 12.707324981689453,
      "learning_rate": 8.48965848965849e-05,
      "loss": 0.2599,
      "step": 5170
    },
    {
      "epoch": 7.474234234234234,
      "grad_norm": 11.656435012817383,
      "learning_rate": 8.441558441558442e-05,
      "loss": 0.2025,
      "step": 5180
    },
    {
      "epoch": 7.488648648648649,
      "grad_norm": 7.975742816925049,
      "learning_rate": 8.393458393458393e-05,
      "loss": 0.1548,
      "step": 5190
    },
    {
      "epoch": 7.503063063063063,
      "grad_norm": 7.300506114959717,
      "learning_rate": 8.345358345358345e-05,
      "loss": 0.1394,
      "step": 5200
    },
    {
      "epoch": 7.5174774774774775,
      "grad_norm": 5.89246129989624,
      "learning_rate": 8.297258297258297e-05,
      "loss": 0.1493,
      "step": 5210
    },
    {
      "epoch": 7.531891891891892,
      "grad_norm": 5.40004825592041,
      "learning_rate": 8.24915824915825e-05,
      "loss": 0.1824,
      "step": 5220
    },
    {
      "epoch": 7.546306306306306,
      "grad_norm": 1.3980309963226318,
      "learning_rate": 8.2010582010582e-05,
      "loss": 0.1331,
      "step": 5230
    },
    {
      "epoch": 7.56072072072072,
      "grad_norm": 8.449721336364746,
      "learning_rate": 8.152958152958153e-05,
      "loss": 0.1978,
      "step": 5240
    },
    {
      "epoch": 7.5751351351351355,
      "grad_norm": 6.612245559692383,
      "learning_rate": 8.104858104858105e-05,
      "loss": 0.161,
      "step": 5250
    },
    {
      "epoch": 7.58954954954955,
      "grad_norm": 6.418858051300049,
      "learning_rate": 8.056758056758056e-05,
      "loss": 0.2178,
      "step": 5260
    },
    {
      "epoch": 7.603963963963964,
      "grad_norm": 4.36170768737793,
      "learning_rate": 8.008658008658008e-05,
      "loss": 0.2027,
      "step": 5270
    },
    {
      "epoch": 7.618378378378378,
      "grad_norm": 4.875021934509277,
      "learning_rate": 7.96055796055796e-05,
      "loss": 0.1214,
      "step": 5280
    },
    {
      "epoch": 7.6327927927927925,
      "grad_norm": 6.8560051918029785,
      "learning_rate": 7.912457912457913e-05,
      "loss": 0.2311,
      "step": 5290
    },
    {
      "epoch": 7.647207207207208,
      "grad_norm": 5.509363651275635,
      "learning_rate": 7.864357864357864e-05,
      "loss": 0.2163,
      "step": 5300
    },
    {
      "epoch": 7.661621621621622,
      "grad_norm": 4.560415267944336,
      "learning_rate": 7.816257816257816e-05,
      "loss": 0.2231,
      "step": 5310
    },
    {
      "epoch": 7.676036036036036,
      "grad_norm": 6.805414199829102,
      "learning_rate": 7.768157768157768e-05,
      "loss": 0.1784,
      "step": 5320
    },
    {
      "epoch": 7.6904504504504505,
      "grad_norm": 6.668816566467285,
      "learning_rate": 7.72005772005772e-05,
      "loss": 0.2719,
      "step": 5330
    },
    {
      "epoch": 7.704864864864865,
      "grad_norm": 5.495980262756348,
      "learning_rate": 7.671957671957671e-05,
      "loss": 0.1169,
      "step": 5340
    },
    {
      "epoch": 7.719279279279279,
      "grad_norm": 8.417407989501953,
      "learning_rate": 7.623857623857623e-05,
      "loss": 0.2062,
      "step": 5350
    },
    {
      "epoch": 7.733693693693693,
      "grad_norm": 8.49002456665039,
      "learning_rate": 7.575757575757576e-05,
      "loss": 0.1368,
      "step": 5360
    },
    {
      "epoch": 7.7481081081081085,
      "grad_norm": 4.653470516204834,
      "learning_rate": 7.527657527657527e-05,
      "loss": 0.1647,
      "step": 5370
    },
    {
      "epoch": 7.762522522522523,
      "grad_norm": 2.8873279094696045,
      "learning_rate": 7.479557479557479e-05,
      "loss": 0.1487,
      "step": 5380
    },
    {
      "epoch": 7.776936936936937,
      "grad_norm": 2.0582430362701416,
      "learning_rate": 7.431457431457431e-05,
      "loss": 0.0748,
      "step": 5390
    },
    {
      "epoch": 7.791351351351351,
      "grad_norm": 6.6209821701049805,
      "learning_rate": 7.383357383357382e-05,
      "loss": 0.1052,
      "step": 5400
    },
    {
      "epoch": 7.8057657657657655,
      "grad_norm": 3.5858404636383057,
      "learning_rate": 7.335257335257334e-05,
      "loss": 0.141,
      "step": 5410
    },
    {
      "epoch": 7.82018018018018,
      "grad_norm": 9.415162086486816,
      "learning_rate": 7.287157287157287e-05,
      "loss": 0.1208,
      "step": 5420
    },
    {
      "epoch": 7.834594594594595,
      "grad_norm": 3.74204421043396,
      "learning_rate": 7.239057239057239e-05,
      "loss": 0.1811,
      "step": 5430
    },
    {
      "epoch": 7.849009009009009,
      "grad_norm": 4.825153827667236,
      "learning_rate": 7.19095719095719e-05,
      "loss": 0.1508,
      "step": 5440
    },
    {
      "epoch": 7.8634234234234235,
      "grad_norm": 6.1774492263793945,
      "learning_rate": 7.142857142857142e-05,
      "loss": 0.1552,
      "step": 5450
    },
    {
      "epoch": 7.877837837837838,
      "grad_norm": 7.0234832763671875,
      "learning_rate": 7.094757094757094e-05,
      "loss": 0.1617,
      "step": 5460
    },
    {
      "epoch": 7.892252252252252,
      "grad_norm": 5.158196449279785,
      "learning_rate": 7.046657046657045e-05,
      "loss": 0.1554,
      "step": 5470
    },
    {
      "epoch": 7.906666666666666,
      "grad_norm": 12.370686531066895,
      "learning_rate": 6.998556998556997e-05,
      "loss": 0.1444,
      "step": 5480
    },
    {
      "epoch": 7.921081081081081,
      "grad_norm": 5.763493537902832,
      "learning_rate": 6.95045695045695e-05,
      "loss": 0.1556,
      "step": 5490
    },
    {
      "epoch": 7.935495495495496,
      "grad_norm": 4.092106342315674,
      "learning_rate": 6.902356902356902e-05,
      "loss": 0.1576,
      "step": 5500
    },
    {
      "epoch": 7.94990990990991,
      "grad_norm": 1.8451523780822754,
      "learning_rate": 6.854256854256853e-05,
      "loss": 0.1095,
      "step": 5510
    },
    {
      "epoch": 7.964324324324324,
      "grad_norm": 2.030266761779785,
      "learning_rate": 6.806156806156807e-05,
      "loss": 0.0973,
      "step": 5520
    },
    {
      "epoch": 7.9787387387387385,
      "grad_norm": 2.9432976245880127,
      "learning_rate": 6.758056758056757e-05,
      "loss": 0.1427,
      "step": 5530
    },
    {
      "epoch": 7.993153153153153,
      "grad_norm": 3.8728318214416504,
      "learning_rate": 6.70995670995671e-05,
      "loss": 0.1461,
      "step": 5540
    },
    {
      "epoch": 7.998918918918919,
      "eval_accuracy": 0.7533967391304348,
      "eval_loss": 1.2246962785720825,
      "eval_runtime": 72.782,
      "eval_samples_per_second": 80.899,
      "eval_steps_per_second": 80.899,
      "step": 5544
    },
    {
      "epoch": 8.008648648648649,
      "grad_norm": 10.597118377685547,
      "learning_rate": 6.661856661856662e-05,
      "loss": 0.1667,
      "step": 5550
    },
    {
      "epoch": 8.023063063063063,
      "grad_norm": 6.5098419189453125,
      "learning_rate": 6.613756613756613e-05,
      "loss": 0.1044,
      "step": 5560
    },
    {
      "epoch": 8.037477477477477,
      "grad_norm": 7.034832954406738,
      "learning_rate": 6.565656565656565e-05,
      "loss": 0.0552,
      "step": 5570
    },
    {
      "epoch": 8.051891891891891,
      "grad_norm": 3.8795289993286133,
      "learning_rate": 6.517556517556517e-05,
      "loss": 0.0905,
      "step": 5580
    },
    {
      "epoch": 8.066306306306306,
      "grad_norm": 10.699118614196777,
      "learning_rate": 6.46945646945647e-05,
      "loss": 0.0921,
      "step": 5590
    },
    {
      "epoch": 8.08072072072072,
      "grad_norm": 1.3635293245315552,
      "learning_rate": 6.42135642135642e-05,
      "loss": 0.0467,
      "step": 5600
    },
    {
      "epoch": 8.095135135135136,
      "grad_norm": 1.773106336593628,
      "learning_rate": 6.373256373256373e-05,
      "loss": 0.0923,
      "step": 5610
    },
    {
      "epoch": 8.10954954954955,
      "grad_norm": 2.9646942615509033,
      "learning_rate": 6.325156325156325e-05,
      "loss": 0.0542,
      "step": 5620
    },
    {
      "epoch": 8.123963963963964,
      "grad_norm": 5.6579437255859375,
      "learning_rate": 6.277056277056277e-05,
      "loss": 0.0801,
      "step": 5630
    },
    {
      "epoch": 8.138378378378379,
      "grad_norm": 3.9888789653778076,
      "learning_rate": 6.228956228956228e-05,
      "loss": 0.0978,
      "step": 5640
    },
    {
      "epoch": 8.152792792792793,
      "grad_norm": 9.667696952819824,
      "learning_rate": 6.18085618085618e-05,
      "loss": 0.1466,
      "step": 5650
    },
    {
      "epoch": 8.167207207207207,
      "grad_norm": 6.377363681793213,
      "learning_rate": 6.132756132756133e-05,
      "loss": 0.098,
      "step": 5660
    },
    {
      "epoch": 8.181621621621622,
      "grad_norm": 6.708555698394775,
      "learning_rate": 6.0846560846560837e-05,
      "loss": 0.1135,
      "step": 5670
    },
    {
      "epoch": 8.196036036036036,
      "grad_norm": 4.443999767303467,
      "learning_rate": 6.036556036556036e-05,
      "loss": 0.0771,
      "step": 5680
    },
    {
      "epoch": 8.21045045045045,
      "grad_norm": 1.0287158489227295,
      "learning_rate": 5.9884559884559875e-05,
      "loss": 0.1132,
      "step": 5690
    },
    {
      "epoch": 8.224864864864864,
      "grad_norm": 0.35798829793930054,
      "learning_rate": 5.94035594035594e-05,
      "loss": 0.096,
      "step": 5700
    },
    {
      "epoch": 8.239279279279279,
      "grad_norm": 3.696572780609131,
      "learning_rate": 5.8922558922558913e-05,
      "loss": 0.0582,
      "step": 5710
    },
    {
      "epoch": 8.253693693693693,
      "grad_norm": 0.7742457985877991,
      "learning_rate": 5.8441558441558436e-05,
      "loss": 0.0645,
      "step": 5720
    },
    {
      "epoch": 8.268108108108109,
      "grad_norm": 0.7144525647163391,
      "learning_rate": 5.796055796055795e-05,
      "loss": 0.0874,
      "step": 5730
    },
    {
      "epoch": 8.282522522522523,
      "grad_norm": 8.79375171661377,
      "learning_rate": 5.747955747955748e-05,
      "loss": 0.1046,
      "step": 5740
    },
    {
      "epoch": 8.296936936936937,
      "grad_norm": 1.4914956092834473,
      "learning_rate": 5.699855699855699e-05,
      "loss": 0.0975,
      "step": 5750
    },
    {
      "epoch": 8.311351351351352,
      "grad_norm": 5.983489036560059,
      "learning_rate": 5.6517556517556506e-05,
      "loss": 0.1445,
      "step": 5760
    },
    {
      "epoch": 8.325765765765766,
      "grad_norm": 9.840721130371094,
      "learning_rate": 5.6036556036556036e-05,
      "loss": 0.0712,
      "step": 5770
    },
    {
      "epoch": 8.34018018018018,
      "grad_norm": 9.522501945495605,
      "learning_rate": 5.5555555555555545e-05,
      "loss": 0.0927,
      "step": 5780
    },
    {
      "epoch": 8.354594594594595,
      "grad_norm": 8.089164733886719,
      "learning_rate": 5.5074555074555074e-05,
      "loss": 0.0852,
      "step": 5790
    },
    {
      "epoch": 8.369009009009009,
      "grad_norm": 2.0056850910186768,
      "learning_rate": 5.459355459355458e-05,
      "loss": 0.0462,
      "step": 5800
    },
    {
      "epoch": 8.383423423423423,
      "grad_norm": 7.016843795776367,
      "learning_rate": 5.411255411255411e-05,
      "loss": 0.1229,
      "step": 5810
    },
    {
      "epoch": 8.397837837837837,
      "grad_norm": 8.507891654968262,
      "learning_rate": 5.363155363155363e-05,
      "loss": 0.0685,
      "step": 5820
    },
    {
      "epoch": 8.412252252252252,
      "grad_norm": 5.983791828155518,
      "learning_rate": 5.315055315055315e-05,
      "loss": 0.1028,
      "step": 5830
    },
    {
      "epoch": 8.426666666666666,
      "grad_norm": 1.4701759815216064,
      "learning_rate": 5.266955266955267e-05,
      "loss": 0.0849,
      "step": 5840
    },
    {
      "epoch": 8.441081081081082,
      "grad_norm": 4.4105329513549805,
      "learning_rate": 5.218855218855219e-05,
      "loss": 0.0766,
      "step": 5850
    },
    {
      "epoch": 8.455495495495496,
      "grad_norm": 1.2428067922592163,
      "learning_rate": 5.1707551707551705e-05,
      "loss": 0.0975,
      "step": 5860
    },
    {
      "epoch": 8.46990990990991,
      "grad_norm": 3.908950090408325,
      "learning_rate": 5.122655122655122e-05,
      "loss": 0.0829,
      "step": 5870
    },
    {
      "epoch": 8.484324324324325,
      "grad_norm": 11.43326187133789,
      "learning_rate": 5.0745550745550744e-05,
      "loss": 0.0958,
      "step": 5880
    },
    {
      "epoch": 8.498738738738739,
      "grad_norm": 5.840853691101074,
      "learning_rate": 5.026455026455026e-05,
      "loss": 0.1068,
      "step": 5890
    },
    {
      "epoch": 8.513153153153153,
      "grad_norm": 5.802908897399902,
      "learning_rate": 4.978354978354978e-05,
      "loss": 0.0815,
      "step": 5900
    },
    {
      "epoch": 8.527567567567568,
      "grad_norm": 6.045281887054443,
      "learning_rate": 4.93025493025493e-05,
      "loss": 0.0838,
      "step": 5910
    },
    {
      "epoch": 8.541981981981982,
      "grad_norm": 5.020217418670654,
      "learning_rate": 4.882154882154882e-05,
      "loss": 0.1123,
      "step": 5920
    },
    {
      "epoch": 8.556396396396396,
      "grad_norm": 8.907532691955566,
      "learning_rate": 4.8340548340548336e-05,
      "loss": 0.1186,
      "step": 5930
    },
    {
      "epoch": 8.57081081081081,
      "grad_norm": 7.421210289001465,
      "learning_rate": 4.785954785954786e-05,
      "loss": 0.0647,
      "step": 5940
    },
    {
      "epoch": 8.585225225225225,
      "grad_norm": 6.697930812835693,
      "learning_rate": 4.7378547378547375e-05,
      "loss": 0.0439,
      "step": 5950
    },
    {
      "epoch": 8.599639639639639,
      "grad_norm": 0.615060031414032,
      "learning_rate": 4.689754689754689e-05,
      "loss": 0.0602,
      "step": 5960
    },
    {
      "epoch": 8.614054054054055,
      "grad_norm": 3.175257444381714,
      "learning_rate": 4.641654641654641e-05,
      "loss": 0.1027,
      "step": 5970
    },
    {
      "epoch": 8.62846846846847,
      "grad_norm": 5.708411693572998,
      "learning_rate": 4.593554593554593e-05,
      "loss": 0.0767,
      "step": 5980
    },
    {
      "epoch": 8.642882882882883,
      "grad_norm": 0.5937002301216125,
      "learning_rate": 4.545454545454545e-05,
      "loss": 0.0418,
      "step": 5990
    },
    {
      "epoch": 8.657297297297298,
      "grad_norm": 3.6142399311065674,
      "learning_rate": 4.497354497354497e-05,
      "loss": 0.0568,
      "step": 6000
    },
    {
      "epoch": 8.671711711711712,
      "grad_norm": 0.24151787161827087,
      "learning_rate": 4.449254449254449e-05,
      "loss": 0.1095,
      "step": 6010
    },
    {
      "epoch": 8.686126126126126,
      "grad_norm": 3.8185176849365234,
      "learning_rate": 4.4011544011544006e-05,
      "loss": 0.1003,
      "step": 6020
    },
    {
      "epoch": 8.70054054054054,
      "grad_norm": 6.915360927581787,
      "learning_rate": 4.353054353054353e-05,
      "loss": 0.1352,
      "step": 6030
    },
    {
      "epoch": 8.714954954954955,
      "grad_norm": 0.8058310747146606,
      "learning_rate": 4.3049543049543045e-05,
      "loss": 0.0439,
      "step": 6040
    },
    {
      "epoch": 8.729369369369369,
      "grad_norm": 4.404812335968018,
      "learning_rate": 4.256854256854257e-05,
      "loss": 0.0485,
      "step": 6050
    },
    {
      "epoch": 8.743783783783783,
      "grad_norm": 3.9927244186401367,
      "learning_rate": 4.208754208754208e-05,
      "loss": 0.0669,
      "step": 6060
    },
    {
      "epoch": 8.758198198198198,
      "grad_norm": 3.6742093563079834,
      "learning_rate": 4.16065416065416e-05,
      "loss": 0.0295,
      "step": 6070
    },
    {
      "epoch": 8.772612612612612,
      "grad_norm": 7.703578948974609,
      "learning_rate": 4.112554112554112e-05,
      "loss": 0.0787,
      "step": 6080
    },
    {
      "epoch": 8.787027027027026,
      "grad_norm": 6.004012107849121,
      "learning_rate": 4.064454064454064e-05,
      "loss": 0.0858,
      "step": 6090
    },
    {
      "epoch": 8.801441441441442,
      "grad_norm": 4.368682384490967,
      "learning_rate": 4.016354016354016e-05,
      "loss": 0.0691,
      "step": 6100
    },
    {
      "epoch": 8.815855855855856,
      "grad_norm": 4.08095645904541,
      "learning_rate": 3.9682539682539676e-05,
      "loss": 0.0469,
      "step": 6110
    },
    {
      "epoch": 8.83027027027027,
      "grad_norm": 2.1851658821105957,
      "learning_rate": 3.92015392015392e-05,
      "loss": 0.0874,
      "step": 6120
    },
    {
      "epoch": 8.844684684684685,
      "grad_norm": 1.3691933155059814,
      "learning_rate": 3.8720538720538714e-05,
      "loss": 0.0871,
      "step": 6130
    },
    {
      "epoch": 8.8590990990991,
      "grad_norm": 11.695643424987793,
      "learning_rate": 3.823953823953824e-05,
      "loss": 0.0722,
      "step": 6140
    },
    {
      "epoch": 8.873513513513513,
      "grad_norm": 2.1198318004608154,
      "learning_rate": 3.775853775853775e-05,
      "loss": 0.0537,
      "step": 6150
    },
    {
      "epoch": 8.887927927927928,
      "grad_norm": 10.203764915466309,
      "learning_rate": 3.7277537277537275e-05,
      "loss": 0.0453,
      "step": 6160
    },
    {
      "epoch": 8.902342342342342,
      "grad_norm": 6.853752136230469,
      "learning_rate": 3.679653679653679e-05,
      "loss": 0.043,
      "step": 6170
    },
    {
      "epoch": 8.916756756756756,
      "grad_norm": 1.940738558769226,
      "learning_rate": 3.6315536315536314e-05,
      "loss": 0.0587,
      "step": 6180
    },
    {
      "epoch": 8.93117117117117,
      "grad_norm": 5.2090277671813965,
      "learning_rate": 3.583453583453583e-05,
      "loss": 0.0413,
      "step": 6190
    },
    {
      "epoch": 8.945585585585585,
      "grad_norm": 4.248027324676514,
      "learning_rate": 3.535353535353535e-05,
      "loss": 0.1331,
      "step": 6200
    },
    {
      "epoch": 8.96,
      "grad_norm": 4.932956218719482,
      "learning_rate": 3.487253487253487e-05,
      "loss": 0.0326,
      "step": 6210
    },
    {
      "epoch": 8.974414414414415,
      "grad_norm": 2.045835494995117,
      "learning_rate": 3.4391534391534384e-05,
      "loss": 0.0285,
      "step": 6220
    },
    {
      "epoch": 8.98882882882883,
      "grad_norm": 2.2289116382598877,
      "learning_rate": 3.3910533910533906e-05,
      "loss": 0.0567,
      "step": 6230
    },
    {
      "epoch": 8.99891891891892,
      "eval_accuracy": 0.7646059782608695,
      "eval_loss": 1.3279215097427368,
      "eval_runtime": 71.3947,
      "eval_samples_per_second": 82.471,
      "eval_steps_per_second": 82.471,
      "step": 6237
    },
    {
      "epoch": 9.004324324324324,
      "grad_norm": 4.529558181762695,
      "learning_rate": 3.342953342953342e-05,
      "loss": 0.0456,
      "step": 6240
    },
    {
      "epoch": 9.018738738738739,
      "grad_norm": 0.142603799700737,
      "learning_rate": 3.2948532948532945e-05,
      "loss": 0.0271,
      "step": 6250
    },
    {
      "epoch": 9.033153153153153,
      "grad_norm": 4.183549404144287,
      "learning_rate": 3.246753246753247e-05,
      "loss": 0.0232,
      "step": 6260
    },
    {
      "epoch": 9.047567567567567,
      "grad_norm": 5.978784084320068,
      "learning_rate": 3.198653198653198e-05,
      "loss": 0.0567,
      "step": 6270
    },
    {
      "epoch": 9.061981981981981,
      "grad_norm": 2.3992292881011963,
      "learning_rate": 3.1505531505531506e-05,
      "loss": 0.0322,
      "step": 6280
    },
    {
      "epoch": 9.076396396396396,
      "grad_norm": 3.95800518989563,
      "learning_rate": 3.102453102453102e-05,
      "loss": 0.0175,
      "step": 6290
    },
    {
      "epoch": 9.090810810810812,
      "grad_norm": 0.18957476317882538,
      "learning_rate": 3.0543530543530544e-05,
      "loss": 0.0495,
      "step": 6300
    },
    {
      "epoch": 9.105225225225226,
      "grad_norm": 0.8902443647384644,
      "learning_rate": 3.006253006253006e-05,
      "loss": 0.0288,
      "step": 6310
    },
    {
      "epoch": 9.11963963963964,
      "grad_norm": 5.375404357910156,
      "learning_rate": 2.9581529581529576e-05,
      "loss": 0.0343,
      "step": 6320
    },
    {
      "epoch": 9.134054054054054,
      "grad_norm": 3.743821859359741,
      "learning_rate": 2.9100529100529095e-05,
      "loss": 0.03,
      "step": 6330
    },
    {
      "epoch": 9.148468468468469,
      "grad_norm": 7.710704326629639,
      "learning_rate": 2.8619528619528615e-05,
      "loss": 0.0368,
      "step": 6340
    },
    {
      "epoch": 9.162882882882883,
      "grad_norm": 0.506182074546814,
      "learning_rate": 2.8138528138528134e-05,
      "loss": 0.0056,
      "step": 6350
    },
    {
      "epoch": 9.177297297297297,
      "grad_norm": 3.3490614891052246,
      "learning_rate": 2.7657527657527656e-05,
      "loss": 0.0478,
      "step": 6360
    },
    {
      "epoch": 9.191711711711712,
      "grad_norm": 0.08678112179040909,
      "learning_rate": 2.7176527176527176e-05,
      "loss": 0.0406,
      "step": 6370
    },
    {
      "epoch": 9.206126126126126,
      "grad_norm": 2.2751471996307373,
      "learning_rate": 2.6695526695526695e-05,
      "loss": 0.0189,
      "step": 6380
    },
    {
      "epoch": 9.22054054054054,
      "grad_norm": 9.320930480957031,
      "learning_rate": 2.6214526214526214e-05,
      "loss": 0.0406,
      "step": 6390
    },
    {
      "epoch": 9.234954954954954,
      "grad_norm": 3.804448366165161,
      "learning_rate": 2.5733525733525733e-05,
      "loss": 0.0459,
      "step": 6400
    },
    {
      "epoch": 9.249369369369369,
      "grad_norm": 1.5775264501571655,
      "learning_rate": 2.5252525252525253e-05,
      "loss": 0.0479,
      "step": 6410
    },
    {
      "epoch": 9.263783783783785,
      "grad_norm": 6.620452880859375,
      "learning_rate": 2.4771524771524772e-05,
      "loss": 0.0243,
      "step": 6420
    },
    {
      "epoch": 9.278198198198199,
      "grad_norm": 4.424402236938477,
      "learning_rate": 2.4290524290524288e-05,
      "loss": 0.0693,
      "step": 6430
    },
    {
      "epoch": 9.292612612612613,
      "grad_norm": 9.379152297973633,
      "learning_rate": 2.3809523809523807e-05,
      "loss": 0.0607,
      "step": 6440
    },
    {
      "epoch": 9.307027027027027,
      "grad_norm": 1.238783597946167,
      "learning_rate": 2.3328523328523326e-05,
      "loss": 0.0095,
      "step": 6450
    },
    {
      "epoch": 9.321441441441442,
      "grad_norm": 0.010547132231295109,
      "learning_rate": 2.2847522847522845e-05,
      "loss": 0.028,
      "step": 6460
    },
    {
      "epoch": 9.335855855855856,
      "grad_norm": 13.737777709960938,
      "learning_rate": 2.2366522366522364e-05,
      "loss": 0.0337,
      "step": 6470
    },
    {
      "epoch": 9.35027027027027,
      "grad_norm": 0.22517243027687073,
      "learning_rate": 2.1885521885521884e-05,
      "loss": 0.0247,
      "step": 6480
    },
    {
      "epoch": 9.364684684684685,
      "grad_norm": 0.18141353130340576,
      "learning_rate": 2.1404521404521403e-05,
      "loss": 0.0468,
      "step": 6490
    },
    {
      "epoch": 9.379099099099099,
      "grad_norm": 1.2880609035491943,
      "learning_rate": 2.0923520923520922e-05,
      "loss": 0.0195,
      "step": 6500
    },
    {
      "epoch": 9.393513513513513,
      "grad_norm": 1.4243581295013428,
      "learning_rate": 2.044252044252044e-05,
      "loss": 0.033,
      "step": 6510
    },
    {
      "epoch": 9.407927927927927,
      "grad_norm": 1.5414937734603882,
      "learning_rate": 1.996151996151996e-05,
      "loss": 0.0481,
      "step": 6520
    },
    {
      "epoch": 9.422342342342342,
      "grad_norm": 5.485969543457031,
      "learning_rate": 1.9480519480519476e-05,
      "loss": 0.0442,
      "step": 6530
    },
    {
      "epoch": 9.436756756756758,
      "grad_norm": 5.689090251922607,
      "learning_rate": 1.8999518999518996e-05,
      "loss": 0.0202,
      "step": 6540
    },
    {
      "epoch": 9.451171171171172,
      "grad_norm": 11.852117538452148,
      "learning_rate": 1.8518518518518515e-05,
      "loss": 0.0639,
      "step": 6550
    },
    {
      "epoch": 9.465585585585586,
      "grad_norm": 5.110032558441162,
      "learning_rate": 1.8037518037518034e-05,
      "loss": 0.0176,
      "step": 6560
    },
    {
      "epoch": 9.48,
      "grad_norm": 0.014017878100275993,
      "learning_rate": 1.7556517556517553e-05,
      "loss": 0.0331,
      "step": 6570
    },
    {
      "epoch": 9.494414414414415,
      "grad_norm": 0.42061179876327515,
      "learning_rate": 1.7075517075517073e-05,
      "loss": 0.025,
      "step": 6580
    },
    {
      "epoch": 9.508828828828829,
      "grad_norm": 0.6986181139945984,
      "learning_rate": 1.6594516594516595e-05,
      "loss": 0.0492,
      "step": 6590
    },
    {
      "epoch": 9.523243243243243,
      "grad_norm": 6.918154239654541,
      "learning_rate": 1.6113516113516114e-05,
      "loss": 0.0605,
      "step": 6600
    },
    {
      "epoch": 9.537657657657657,
      "grad_norm": 1.6587908267974854,
      "learning_rate": 1.563251563251563e-05,
      "loss": 0.0267,
      "step": 6610
    },
    {
      "epoch": 9.552072072072072,
      "grad_norm": 0.39499762654304504,
      "learning_rate": 1.515151515151515e-05,
      "loss": 0.0266,
      "step": 6620
    },
    {
      "epoch": 9.566486486486486,
      "grad_norm": 3.649427890777588,
      "learning_rate": 1.4670514670514669e-05,
      "loss": 0.0345,
      "step": 6630
    },
    {
      "epoch": 9.5809009009009,
      "grad_norm": 0.8373857140541077,
      "learning_rate": 1.418951418951419e-05,
      "loss": 0.0403,
      "step": 6640
    },
    {
      "epoch": 9.595315315315315,
      "grad_norm": 1.028454303741455,
      "learning_rate": 1.3708513708513709e-05,
      "loss": 0.012,
      "step": 6650
    },
    {
      "epoch": 9.609729729729729,
      "grad_norm": 5.6264448165893555,
      "learning_rate": 1.3227513227513226e-05,
      "loss": 0.0784,
      "step": 6660
    },
    {
      "epoch": 9.624144144144145,
      "grad_norm": 7.463881015777588,
      "learning_rate": 1.2746512746512746e-05,
      "loss": 0.0194,
      "step": 6670
    },
    {
      "epoch": 9.63855855855856,
      "grad_norm": 10.843091011047363,
      "learning_rate": 1.2265512265512265e-05,
      "loss": 0.0406,
      "step": 6680
    },
    {
      "epoch": 9.652972972972973,
      "grad_norm": 0.4320850372314453,
      "learning_rate": 1.1784511784511784e-05,
      "loss": 0.0208,
      "step": 6690
    },
    {
      "epoch": 9.667387387387388,
      "grad_norm": 0.8611360788345337,
      "learning_rate": 1.1303511303511303e-05,
      "loss": 0.023,
      "step": 6700
    },
    {
      "epoch": 9.681801801801802,
      "grad_norm": 4.806682586669922,
      "learning_rate": 1.082251082251082e-05,
      "loss": 0.0163,
      "step": 6710
    },
    {
      "epoch": 9.696216216216216,
      "grad_norm": 0.03333412855863571,
      "learning_rate": 1.034151034151034e-05,
      "loss": 0.0504,
      "step": 6720
    },
    {
      "epoch": 9.71063063063063,
      "grad_norm": 0.18930605053901672,
      "learning_rate": 9.86050986050986e-06,
      "loss": 0.0342,
      "step": 6730
    },
    {
      "epoch": 9.725045045045045,
      "grad_norm": 9.080709457397461,
      "learning_rate": 9.379509379509378e-06,
      "loss": 0.072,
      "step": 6740
    },
    {
      "epoch": 9.739459459459459,
      "grad_norm": 0.48559027910232544,
      "learning_rate": 8.898508898508898e-06,
      "loss": 0.0408,
      "step": 6750
    },
    {
      "epoch": 9.753873873873873,
      "grad_norm": 0.10552722215652466,
      "learning_rate": 8.417508417508417e-06,
      "loss": 0.0089,
      "step": 6760
    },
    {
      "epoch": 9.768288288288288,
      "grad_norm": 0.5119392275810242,
      "learning_rate": 7.936507936507936e-06,
      "loss": 0.0085,
      "step": 6770
    },
    {
      "epoch": 9.782702702702704,
      "grad_norm": 2.2527782917022705,
      "learning_rate": 7.4555074555074545e-06,
      "loss": 0.0459,
      "step": 6780
    },
    {
      "epoch": 9.797117117117118,
      "grad_norm": 0.280809611082077,
      "learning_rate": 6.974506974506975e-06,
      "loss": 0.0343,
      "step": 6790
    },
    {
      "epoch": 9.811531531531532,
      "grad_norm": 9.776128768920898,
      "learning_rate": 6.493506493506493e-06,
      "loss": 0.0385,
      "step": 6800
    },
    {
      "epoch": 9.825945945945946,
      "grad_norm": 0.9988490343093872,
      "learning_rate": 6.012506012506012e-06,
      "loss": 0.0235,
      "step": 6810
    },
    {
      "epoch": 9.84036036036036,
      "grad_norm": 0.36588847637176514,
      "learning_rate": 5.531505531505531e-06,
      "loss": 0.0474,
      "step": 6820
    },
    {
      "epoch": 9.854774774774775,
      "grad_norm": 0.32074180245399475,
      "learning_rate": 5.05050505050505e-06,
      "loss": 0.0072,
      "step": 6830
    },
    {
      "epoch": 9.86918918918919,
      "grad_norm": 0.024474648758769035,
      "learning_rate": 4.569504569504569e-06,
      "loss": 0.043,
      "step": 6840
    },
    {
      "epoch": 9.883603603603603,
      "grad_norm": 0.17253172397613525,
      "learning_rate": 4.088504088504088e-06,
      "loss": 0.0391,
      "step": 6850
    },
    {
      "epoch": 9.898018018018018,
      "grad_norm": 2.319399356842041,
      "learning_rate": 3.607503607503607e-06,
      "loss": 0.033,
      "step": 6860
    },
    {
      "epoch": 9.912432432432432,
      "grad_norm": 0.1351654827594757,
      "learning_rate": 3.1265031265031263e-06,
      "loss": 0.029,
      "step": 6870
    },
    {
      "epoch": 9.926846846846846,
      "grad_norm": 0.049245912581682205,
      "learning_rate": 2.645502645502645e-06,
      "loss": 0.0098,
      "step": 6880
    },
    {
      "epoch": 9.94126126126126,
      "grad_norm": 1.022704005241394,
      "learning_rate": 2.1645021645021643e-06,
      "loss": 0.018,
      "step": 6890
    },
    {
      "epoch": 9.955675675675675,
      "grad_norm": 9.319286346435547,
      "learning_rate": 1.6835016835016831e-06,
      "loss": 0.0696,
      "step": 6900
    },
    {
      "epoch": 9.97009009009009,
      "grad_norm": 0.25702571868896484,
      "learning_rate": 1.2025012025012024e-06,
      "loss": 0.0401,
      "step": 6910
    },
    {
      "epoch": 9.984504504504505,
      "grad_norm": 0.4501936435699463,
      "learning_rate": 7.215007215007215e-07,
      "loss": 0.0493,
      "step": 6920
    },
    {
      "epoch": 9.99891891891892,
      "grad_norm": 0.11143568903207779,
      "learning_rate": 2.405002405002405e-07,
      "loss": 0.0375,
      "step": 6930
    },
    {
      "epoch": 9.99891891891892,
      "eval_accuracy": 0.7800611413043478,
      "eval_loss": 1.2554279565811157,
      "eval_runtime": 71.5553,
      "eval_samples_per_second": 82.286,
      "eval_steps_per_second": 82.286,
      "step": 6930
    },
    {
      "epoch": 9.99891891891892,
      "step": 6930,
      "total_flos": 1.5503579786196265e+19,
      "train_loss": 0.927044793691427,
      "train_runtime": 4273.8828,
      "train_samples_per_second": 51.929,
      "train_steps_per_second": 1.621
    }
  ],
  "logging_steps": 10,
  "max_steps": 6930,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5503579786196265e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}