{
  "best_metric": 1.0,
  "best_model_checkpoint": "vit-base-patch16-224-mascotas\\checkpoint-52",
  "epoch": 39.61904761904762,
  "eval_steps": 500,
  "global_step": 2080,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "learning_rate": 5.288461538461539e-06,
      "loss": 0.6927,
      "step": 10
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.0576923076923078e-05,
      "loss": 0.6891,
      "step": 20
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.5865384615384613e-05,
      "loss": 0.6783,
      "step": 30
    },
    {
      "epoch": 0.76,
      "learning_rate": 2.1153846153846157e-05,
      "loss": 0.6577,
      "step": 40
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.6442307692307694e-05,
      "loss": 0.6192,
      "step": 50
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.5660281181335449,
      "eval_runtime": 0.4455,
      "eval_samples_per_second": 42.651,
      "eval_steps_per_second": 6.734,
      "step": 52
    },
    {
      "epoch": 1.14,
      "learning_rate": 3.173076923076923e-05,
      "loss": 0.5503,
      "step": 60
    },
    {
      "epoch": 1.33,
      "learning_rate": 3.701923076923077e-05,
      "loss": 0.4491,
      "step": 70
    },
    {
      "epoch": 1.52,
      "learning_rate": 4.2307692307692314e-05,
      "loss": 0.4092,
      "step": 80
    },
    {
      "epoch": 1.71,
      "learning_rate": 4.759615384615385e-05,
      "loss": 0.3908,
      "step": 90
    },
    {
      "epoch": 1.9,
      "learning_rate": 5.288461538461539e-05,
      "loss": 0.2794,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.15241165459156036,
      "eval_runtime": 0.3561,
      "eval_samples_per_second": 53.359,
      "eval_steps_per_second": 8.425,
      "step": 105
    },
    {
      "epoch": 2.1,
      "learning_rate": 5.4832995951417006e-05,
      "loss": 0.2684,
      "step": 110
    },
    {
      "epoch": 2.29,
      "learning_rate": 5.455465587044535e-05,
      "loss": 0.1892,
      "step": 120
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.4276315789473686e-05,
      "loss": 0.2716,
      "step": 130
    },
    {
      "epoch": 2.67,
      "learning_rate": 5.399797570850203e-05,
      "loss": 0.2146,
      "step": 140
    },
    {
      "epoch": 2.86,
      "learning_rate": 5.3719635627530365e-05,
      "loss": 0.2006,
      "step": 150
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.9473684210526315,
      "eval_loss": 0.11429128050804138,
      "eval_runtime": 0.3671,
      "eval_samples_per_second": 51.759,
      "eval_steps_per_second": 8.173,
      "step": 157
    },
    {
      "epoch": 3.05,
      "learning_rate": 5.34412955465587e-05,
      "loss": 0.2683,
      "step": 160
    },
    {
      "epoch": 3.24,
      "learning_rate": 5.3162955465587044e-05,
      "loss": 0.2086,
      "step": 170
    },
    {
      "epoch": 3.43,
      "learning_rate": 5.288461538461539e-05,
      "loss": 0.1823,
      "step": 180
    },
    {
      "epoch": 3.62,
      "learning_rate": 5.2606275303643724e-05,
      "loss": 0.2282,
      "step": 190
    },
    {
      "epoch": 3.81,
      "learning_rate": 5.2327935222672067e-05,
      "loss": 0.2076,
      "step": 200
    },
    {
      "epoch": 4.0,
      "learning_rate": 5.204959514170041e-05,
      "loss": 0.2106,
      "step": 210
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.050596825778484344,
      "eval_runtime": 0.3601,
      "eval_samples_per_second": 52.766,
      "eval_steps_per_second": 8.331,
      "step": 210
    },
    {
      "epoch": 4.19,
      "learning_rate": 5.1771255060728746e-05,
      "loss": 0.1872,
      "step": 220
    },
    {
      "epoch": 4.38,
      "learning_rate": 5.149291497975709e-05,
      "loss": 0.0983,
      "step": 230
    },
    {
      "epoch": 4.57,
      "learning_rate": 5.1214574898785425e-05,
      "loss": 0.165,
      "step": 240
    },
    {
      "epoch": 4.76,
      "learning_rate": 5.093623481781377e-05,
      "loss": 0.0742,
      "step": 250
    },
    {
      "epoch": 4.95,
      "learning_rate": 5.0657894736842104e-05,
      "loss": 0.1077,
      "step": 260
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.03385559841990471,
      "eval_runtime": 0.3681,
      "eval_samples_per_second": 51.619,
      "eval_steps_per_second": 8.15,
      "step": 262
    },
    {
      "epoch": 5.14,
      "learning_rate": 5.037955465587045e-05,
      "loss": 0.1099,
      "step": 270
    },
    {
      "epoch": 5.33,
      "learning_rate": 5.010121457489879e-05,
      "loss": 0.1163,
      "step": 280
    },
    {
      "epoch": 5.52,
      "learning_rate": 4.982287449392713e-05,
      "loss": 0.17,
      "step": 290
    },
    {
      "epoch": 5.71,
      "learning_rate": 4.954453441295547e-05,
      "loss": 0.2067,
      "step": 300
    },
    {
      "epoch": 5.9,
      "learning_rate": 4.926619433198381e-05,
      "loss": 0.1379,
      "step": 310
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9473684210526315,
      "eval_loss": 0.13907362520694733,
      "eval_runtime": 0.3581,
      "eval_samples_per_second": 53.061,
      "eval_steps_per_second": 8.378,
      "step": 315
    },
    {
      "epoch": 6.1,
      "learning_rate": 4.898785425101215e-05,
      "loss": 0.0956,
      "step": 320
    },
    {
      "epoch": 6.29,
      "learning_rate": 4.8709514170040485e-05,
      "loss": 0.0764,
      "step": 330
    },
    {
      "epoch": 6.48,
      "learning_rate": 4.843117408906883e-05,
      "loss": 0.106,
      "step": 340
    },
    {
      "epoch": 6.67,
      "learning_rate": 4.8152834008097165e-05,
      "loss": 0.1658,
      "step": 350
    },
    {
      "epoch": 6.86,
      "learning_rate": 4.787449392712551e-05,
      "loss": 0.1387,
      "step": 360
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.9473684210526315,
      "eval_loss": 0.07111804187297821,
      "eval_runtime": 0.4206,
      "eval_samples_per_second": 45.173,
      "eval_steps_per_second": 7.133,
      "step": 367
    },
    {
      "epoch": 7.05,
      "learning_rate": 4.759615384615385e-05,
      "loss": 0.1693,
      "step": 370
    },
    {
      "epoch": 7.24,
      "learning_rate": 4.731781376518219e-05,
      "loss": 0.1271,
      "step": 380
    },
    {
      "epoch": 7.43,
      "learning_rate": 4.703947368421053e-05,
      "loss": 0.1374,
      "step": 390
    },
    {
      "epoch": 7.62,
      "learning_rate": 4.676113360323887e-05,
      "loss": 0.0917,
      "step": 400
    },
    {
      "epoch": 7.81,
      "learning_rate": 4.64827935222672e-05,
      "loss": 0.1411,
      "step": 410
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.6204453441295545e-05,
      "loss": 0.1342,
      "step": 420
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.023148125037550926,
      "eval_runtime": 0.3751,
      "eval_samples_per_second": 50.655,
      "eval_steps_per_second": 7.998,
      "step": 420
    },
    {
      "epoch": 8.19,
      "learning_rate": 4.592611336032389e-05,
      "loss": 0.0993,
      "step": 430
    },
    {
      "epoch": 8.38,
      "learning_rate": 4.5647773279352225e-05,
      "loss": 0.1042,
      "step": 440
    },
    {
      "epoch": 8.57,
      "learning_rate": 4.536943319838057e-05,
      "loss": 0.0635,
      "step": 450
    },
    {
      "epoch": 8.76,
      "learning_rate": 4.509109311740891e-05,
      "loss": 0.1124,
      "step": 460
    },
    {
      "epoch": 8.95,
      "learning_rate": 4.481275303643725e-05,
      "loss": 0.0803,
      "step": 470
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.047453682869672775,
      "eval_runtime": 0.3711,
      "eval_samples_per_second": 51.201,
      "eval_steps_per_second": 8.084,
      "step": 472
    },
    {
      "epoch": 9.14,
      "learning_rate": 4.453441295546559e-05,
      "loss": 0.1052,
      "step": 480
    },
    {
      "epoch": 9.33,
      "learning_rate": 4.4256072874493926e-05,
      "loss": 0.1562,
      "step": 490
    },
    {
      "epoch": 9.52,
      "learning_rate": 4.397773279352227e-05,
      "loss": 0.1356,
      "step": 500
    },
    {
      "epoch": 9.71,
      "learning_rate": 4.3699392712550606e-05,
      "loss": 0.1523,
      "step": 510
    },
    {
      "epoch": 9.9,
      "learning_rate": 4.342105263157895e-05,
      "loss": 0.097,
      "step": 520
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.01901022344827652,
      "eval_runtime": 0.3681,
      "eval_samples_per_second": 51.619,
      "eval_steps_per_second": 8.15,
      "step": 525
    },
    {
      "epoch": 10.1,
      "learning_rate": 4.314271255060729e-05,
      "loss": 0.0688,
      "step": 530
    },
    {
      "epoch": 10.29,
      "learning_rate": 4.286437246963563e-05,
      "loss": 0.1178,
      "step": 540
    },
    {
      "epoch": 10.48,
      "learning_rate": 4.258603238866397e-05,
      "loss": 0.0491,
      "step": 550
    },
    {
      "epoch": 10.67,
      "learning_rate": 4.2307692307692314e-05,
      "loss": 0.1043,
      "step": 560
    },
    {
      "epoch": 10.86,
      "learning_rate": 4.202935222672065e-05,
      "loss": 0.0888,
      "step": 570
    },
    {
      "epoch": 10.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.022018224000930786,
      "eval_runtime": 0.3621,
      "eval_samples_per_second": 52.474,
      "eval_steps_per_second": 8.285,
      "step": 577
    },
    {
      "epoch": 11.05,
      "learning_rate": 4.1751012145748986e-05,
      "loss": 0.0424,
      "step": 580
    },
    {
      "epoch": 11.24,
      "learning_rate": 4.147267206477733e-05,
      "loss": 0.0698,
      "step": 590
    },
    {
      "epoch": 11.43,
      "learning_rate": 4.1194331983805666e-05,
      "loss": 0.0399,
      "step": 600
    },
    {
      "epoch": 11.62,
      "learning_rate": 4.091599190283401e-05,
      "loss": 0.1481,
      "step": 610
    },
    {
      "epoch": 11.81,
      "learning_rate": 4.063765182186235e-05,
      "loss": 0.1183,
      "step": 620
    },
    {
      "epoch": 12.0,
      "learning_rate": 4.035931174089069e-05,
      "loss": 0.0668,
      "step": 630
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.00784948468208313,
      "eval_runtime": 0.3516,
      "eval_samples_per_second": 54.041,
      "eval_steps_per_second": 8.533,
      "step": 630
    },
    {
      "epoch": 12.19,
      "learning_rate": 4.008097165991903e-05,
      "loss": 0.065,
      "step": 640
    },
    {
      "epoch": 12.38,
      "learning_rate": 3.9802631578947374e-05,
      "loss": 0.0716,
      "step": 650
    },
    {
      "epoch": 12.57,
      "learning_rate": 3.952429149797571e-05,
      "loss": 0.0811,
      "step": 660
    },
    {
      "epoch": 12.76,
      "learning_rate": 3.9245951417004047e-05,
      "loss": 0.1079,
      "step": 670
    },
    {
      "epoch": 12.95,
      "learning_rate": 3.896761133603239e-05,
      "loss": 0.0559,
      "step": 680
    },
    {
      "epoch": 12.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.007285874802619219,
      "eval_runtime": 0.3904,
      "eval_samples_per_second": 48.668,
      "eval_steps_per_second": 7.684,
      "step": 682
    },
    {
      "epoch": 13.14,
      "learning_rate": 3.8689271255060726e-05,
      "loss": 0.1006,
      "step": 690
    },
    {
      "epoch": 13.33,
      "learning_rate": 3.841093117408907e-05,
      "loss": 0.1479,
      "step": 700
    },
    {
      "epoch": 13.52,
      "learning_rate": 3.813259109311741e-05,
      "loss": 0.0442,
      "step": 710
    },
    {
      "epoch": 13.71,
      "learning_rate": 3.785425101214575e-05,
      "loss": 0.0593,
      "step": 720
    },
    {
      "epoch": 13.9,
      "learning_rate": 3.757591093117409e-05,
      "loss": 0.0759,
      "step": 730
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.005547249224036932,
      "eval_runtime": 0.3706,
      "eval_samples_per_second": 51.261,
      "eval_steps_per_second": 8.094,
      "step": 735
    },
    {
      "epoch": 14.1,
      "learning_rate": 3.7297570850202434e-05,
      "loss": 0.1361,
      "step": 740
    },
    {
      "epoch": 14.29,
      "learning_rate": 3.701923076923077e-05,
      "loss": 0.087,
      "step": 750
    },
    {
      "epoch": 14.48,
      "learning_rate": 3.674089068825911e-05,
      "loss": 0.0928,
      "step": 760
    },
    {
      "epoch": 14.67,
      "learning_rate": 3.646255060728745e-05,
      "loss": 0.0357,
      "step": 770
    },
    {
      "epoch": 14.86,
      "learning_rate": 3.618421052631579e-05,
      "loss": 0.081,
      "step": 780
    },
    {
      "epoch": 14.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.005823916289955378,
      "eval_runtime": 0.4351,
      "eval_samples_per_second": 43.668,
      "eval_steps_per_second": 6.895,
      "step": 787
    },
    {
      "epoch": 15.05,
      "learning_rate": 3.590587044534413e-05,
      "loss": 0.1453,
      "step": 790
    },
    {
      "epoch": 15.24,
      "learning_rate": 3.562753036437247e-05,
      "loss": 0.0973,
      "step": 800
    },
    {
      "epoch": 15.43,
      "learning_rate": 3.5349190283400815e-05,
      "loss": 0.0781,
      "step": 810
    },
    {
      "epoch": 15.62,
      "learning_rate": 3.507085020242915e-05,
      "loss": 0.0664,
      "step": 820
    },
    {
      "epoch": 15.81,
      "learning_rate": 3.4792510121457494e-05,
      "loss": 0.1156,
      "step": 830
    },
    {
      "epoch": 16.0,
      "learning_rate": 3.451417004048583e-05,
      "loss": 0.0806,
      "step": 840
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.010181745514273643,
      "eval_runtime": 0.4006,
      "eval_samples_per_second": 47.428,
      "eval_steps_per_second": 7.489,
      "step": 840
    },
    {
      "epoch": 16.19,
      "learning_rate": 3.423582995951417e-05,
      "loss": 0.0531,
      "step": 850
    },
    {
      "epoch": 16.38,
      "learning_rate": 3.395748987854251e-05,
      "loss": 0.0504,
      "step": 860
    },
    {
      "epoch": 16.57,
      "learning_rate": 3.367914979757085e-05,
      "loss": 0.0892,
      "step": 870
    },
    {
      "epoch": 16.76,
      "learning_rate": 3.340080971659919e-05,
      "loss": 0.0664,
      "step": 880
    },
    {
      "epoch": 16.95,
      "learning_rate": 3.312246963562753e-05,
      "loss": 0.0568,
      "step": 890
    },
    {
      "epoch": 16.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.01637931726872921,
      "eval_runtime": 0.3491,
      "eval_samples_per_second": 54.429,
      "eval_steps_per_second": 8.594,
      "step": 892
    },
    {
      "epoch": 17.14,
      "learning_rate": 3.2844129554655875e-05,
      "loss": 0.1494,
      "step": 900
    },
    {
      "epoch": 17.33,
      "learning_rate": 3.256578947368421e-05,
      "loss": 0.0428,
      "step": 910
    },
    {
      "epoch": 17.52,
      "learning_rate": 3.2287449392712554e-05,
      "loss": 0.0778,
      "step": 920
    },
    {
      "epoch": 17.71,
      "learning_rate": 3.200910931174089e-05,
      "loss": 0.0495,
      "step": 930
    },
    {
      "epoch": 17.9,
      "learning_rate": 3.173076923076923e-05,
      "loss": 0.0696,
      "step": 940
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0049411519430577755,
      "eval_runtime": 0.3741,
      "eval_samples_per_second": 50.791,
      "eval_steps_per_second": 8.02,
      "step": 945
    },
    {
      "epoch": 18.1,
      "learning_rate": 3.145242914979757e-05,
      "loss": 0.0626,
      "step": 950
    },
    {
      "epoch": 18.29,
      "learning_rate": 3.117408906882591e-05,
      "loss": 0.0884,
      "step": 960
    },
    {
      "epoch": 18.48,
      "learning_rate": 3.0895748987854256e-05,
      "loss": 0.1241,
      "step": 970
    },
    {
      "epoch": 18.67,
      "learning_rate": 3.061740890688259e-05,
      "loss": 0.0409,
      "step": 980
    },
    {
      "epoch": 18.86,
      "learning_rate": 3.0339068825910932e-05,
      "loss": 0.0692,
      "step": 990
    },
    {
      "epoch": 18.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.003958892542868853,
      "eval_runtime": 0.3831,
      "eval_samples_per_second": 49.597,
      "eval_steps_per_second": 7.831,
      "step": 997
    },
    {
      "epoch": 19.05,
      "learning_rate": 3.0060728744939275e-05,
      "loss": 0.0185,
      "step": 1000
    },
    {
      "epoch": 19.24,
      "learning_rate": 2.978238866396761e-05,
      "loss": 0.0174,
      "step": 1010
    },
    {
      "epoch": 19.43,
      "learning_rate": 2.9504048582995954e-05,
      "loss": 0.0471,
      "step": 1020
    },
    {
      "epoch": 19.62,
      "learning_rate": 2.9225708502024294e-05,
      "loss": 0.0385,
      "step": 1030
    },
    {
      "epoch": 19.81,
      "learning_rate": 2.894736842105263e-05,
      "loss": 0.094,
      "step": 1040
    },
    {
      "epoch": 20.0,
      "learning_rate": 2.8669028340080973e-05,
      "loss": 0.0929,
      "step": 1050
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.003028671257197857,
      "eval_runtime": 0.3701,
      "eval_samples_per_second": 51.34,
      "eval_steps_per_second": 8.106,
      "step": 1050
    },
    {
      "epoch": 20.19,
      "learning_rate": 2.8390688259109316e-05,
      "loss": 0.0612,
      "step": 1060
    },
    {
      "epoch": 20.38,
      "learning_rate": 2.8112348178137652e-05,
      "loss": 0.1051,
      "step": 1070
    },
    {
      "epoch": 20.57,
      "learning_rate": 2.7834008097165992e-05,
      "loss": 0.1134,
      "step": 1080
    },
    {
      "epoch": 20.76,
      "learning_rate": 2.7555668016194335e-05,
      "loss": 0.0729,
      "step": 1090
    },
    {
      "epoch": 20.95,
      "learning_rate": 2.7277327935222675e-05,
      "loss": 0.1169,
      "step": 1100
    },
    {
      "epoch": 20.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.018264027312397957,
      "eval_runtime": 0.3641,
      "eval_samples_per_second": 52.186,
      "eval_steps_per_second": 8.24,
      "step": 1102
    },
    {
      "epoch": 21.14,
      "learning_rate": 2.6998987854251014e-05,
      "loss": 0.0873,
      "step": 1110
    },
    {
      "epoch": 21.33,
      "learning_rate": 2.672064777327935e-05,
      "loss": 0.0065,
      "step": 1120
    },
    {
      "epoch": 21.52,
      "learning_rate": 2.6442307692307694e-05,
      "loss": 0.0503,
      "step": 1130
    },
    {
      "epoch": 21.71,
      "learning_rate": 2.6163967611336033e-05,
      "loss": 0.0535,
      "step": 1140
    },
    {
      "epoch": 21.9,
      "learning_rate": 2.5885627530364373e-05,
      "loss": 0.0385,
      "step": 1150
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.9473684210526315,
      "eval_loss": 0.057639673352241516,
      "eval_runtime": 0.3631,
      "eval_samples_per_second": 52.33,
      "eval_steps_per_second": 8.263,
      "step": 1155
    },
    {
      "epoch": 22.1,
      "learning_rate": 2.5607287449392713e-05,
      "loss": 0.0787,
      "step": 1160
    },
    {
      "epoch": 22.29,
      "learning_rate": 2.5328947368421052e-05,
      "loss": 0.0201,
      "step": 1170
    },
    {
      "epoch": 22.48,
      "learning_rate": 2.5050607287449395e-05,
      "loss": 0.074,
      "step": 1180
    },
    {
      "epoch": 22.67,
      "learning_rate": 2.4772267206477735e-05,
      "loss": 0.0275,
      "step": 1190
    },
    {
      "epoch": 22.86,
      "learning_rate": 2.4493927125506075e-05,
      "loss": 0.0564,
      "step": 1200
    },
    {
      "epoch": 22.99,
      "eval_accuracy": 0.9473684210526315,
      "eval_loss": 0.05121718719601631,
      "eval_runtime": 0.3611,
      "eval_samples_per_second": 52.62,
      "eval_steps_per_second": 8.308,
      "step": 1207
    },
    {
      "epoch": 23.05,
      "learning_rate": 2.4215587044534414e-05,
      "loss": 0.1354,
      "step": 1210
    },
    {
      "epoch": 23.24,
      "learning_rate": 2.3937246963562754e-05,
      "loss": 0.0665,
      "step": 1220
    },
    {
      "epoch": 23.43,
      "learning_rate": 2.3658906882591093e-05,
      "loss": 0.0393,
      "step": 1230
    },
    {
      "epoch": 23.62,
      "learning_rate": 2.3380566801619436e-05,
      "loss": 0.0427,
      "step": 1240
    },
    {
      "epoch": 23.81,
      "learning_rate": 2.3102226720647773e-05,
      "loss": 0.0563,
      "step": 1250
    },
    {
      "epoch": 24.0,
      "learning_rate": 2.2823886639676112e-05,
      "loss": 0.0206,
      "step": 1260
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0025241610128432512,
      "eval_runtime": 0.3626,
      "eval_samples_per_second": 52.4,
      "eval_steps_per_second": 8.274,
      "step": 1260
    },
    {
      "epoch": 24.19,
      "learning_rate": 2.2545546558704455e-05,
      "loss": 0.1316,
      "step": 1270
    },
    {
      "epoch": 24.38,
      "learning_rate": 2.2267206477732795e-05,
      "loss": 0.0247,
      "step": 1280
    },
    {
      "epoch": 24.57,
      "learning_rate": 2.1988866396761135e-05,
      "loss": 0.0741,
      "step": 1290
    },
    {
      "epoch": 24.76,
      "learning_rate": 2.1710526315789474e-05,
      "loss": 0.0965,
      "step": 1300
    },
    {
      "epoch": 24.95,
      "learning_rate": 2.1432186234817814e-05,
      "loss": 0.0984,
      "step": 1310
    },
    {
      "epoch": 24.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002796064130961895,
      "eval_runtime": 0.3786,
      "eval_samples_per_second": 50.185,
      "eval_steps_per_second": 7.924,
      "step": 1312
    },
    {
      "epoch": 25.14,
      "learning_rate": 2.1153846153846157e-05,
      "loss": 0.0943,
      "step": 1320
    },
    {
      "epoch": 25.33,
      "learning_rate": 2.0875506072874493e-05,
      "loss": 0.0701,
      "step": 1330
    },
    {
      "epoch": 25.52,
      "learning_rate": 2.0597165991902833e-05,
      "loss": 0.0542,
      "step": 1340
    },
    {
      "epoch": 25.71,
      "learning_rate": 2.0318825910931176e-05,
      "loss": 0.0756,
      "step": 1350
    },
    {
      "epoch": 25.9,
      "learning_rate": 2.0040485829959516e-05,
      "loss": 0.0368,
      "step": 1360
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.003676516469568014,
      "eval_runtime": 0.3821,
      "eval_samples_per_second": 49.727,
      "eval_steps_per_second": 7.852,
      "step": 1365
    },
    {
      "epoch": 26.1,
      "learning_rate": 1.9762145748987855e-05,
      "loss": 0.0547,
      "step": 1370
    },
    {
      "epoch": 26.29,
      "learning_rate": 1.9483805668016195e-05,
      "loss": 0.0563,
      "step": 1380
    },
    {
      "epoch": 26.48,
      "learning_rate": 1.9205465587044534e-05,
      "loss": 0.0399,
      "step": 1390
    },
    {
      "epoch": 26.67,
      "learning_rate": 1.8927125506072874e-05,
      "loss": 0.0499,
      "step": 1400
    },
    {
      "epoch": 26.86,
      "learning_rate": 1.8648785425101217e-05,
      "loss": 0.0436,
      "step": 1410
    },
    {
      "epoch": 26.99,
      "eval_accuracy": 0.9473684210526315,
      "eval_loss": 0.06093175709247589,
      "eval_runtime": 0.3641,
      "eval_samples_per_second": 52.186,
      "eval_steps_per_second": 8.24,
      "step": 1417
    },
    {
      "epoch": 27.05,
      "learning_rate": 1.8370445344129553e-05,
      "loss": 0.1086,
      "step": 1420
    },
    {
      "epoch": 27.24,
      "learning_rate": 1.8092105263157896e-05,
      "loss": 0.0425,
      "step": 1430
    },
    {
      "epoch": 27.43,
      "learning_rate": 1.7813765182186236e-05,
      "loss": 0.0399,
      "step": 1440
    },
    {
      "epoch": 27.62,
      "learning_rate": 1.7535425101214576e-05,
      "loss": 0.0438,
      "step": 1450
    },
    {
      "epoch": 27.81,
      "learning_rate": 1.7257085020242915e-05,
      "loss": 0.0419,
      "step": 1460
    },
    {
      "epoch": 28.0,
      "learning_rate": 1.6978744939271255e-05,
      "loss": 0.0896,
      "step": 1470
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.01538935024291277,
      "eval_runtime": 0.3706,
      "eval_samples_per_second": 51.27,
      "eval_steps_per_second": 8.095,
      "step": 1470
    },
    {
      "epoch": 28.19,
      "learning_rate": 1.6700404858299595e-05,
      "loss": 0.0483,
      "step": 1480
    },
    {
      "epoch": 28.38,
      "learning_rate": 1.6422064777327938e-05,
      "loss": 0.0589,
      "step": 1490
    },
    {
      "epoch": 28.57,
      "learning_rate": 1.6143724696356277e-05,
      "loss": 0.0708,
      "step": 1500
    },
    {
      "epoch": 28.76,
      "learning_rate": 1.5865384615384613e-05,
      "loss": 0.062,
      "step": 1510
    },
    {
      "epoch": 28.95,
      "learning_rate": 1.5587044534412957e-05,
      "loss": 0.079,
      "step": 1520
    },
    {
      "epoch": 28.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0026065746787935495,
      "eval_runtime": 0.3741,
      "eval_samples_per_second": 50.791,
      "eval_steps_per_second": 8.02,
      "step": 1522
    },
    {
      "epoch": 29.14,
      "learning_rate": 1.5308704453441296e-05,
      "loss": 0.0246,
      "step": 1530
    },
    {
      "epoch": 29.33,
      "learning_rate": 1.5030364372469637e-05,
      "loss": 0.0735,
      "step": 1540
    },
    {
      "epoch": 29.52,
      "learning_rate": 1.4752024291497977e-05,
      "loss": 0.0644,
      "step": 1550
    },
    {
      "epoch": 29.71,
      "learning_rate": 1.4473684210526315e-05,
      "loss": 0.0536,
      "step": 1560
    },
    {
      "epoch": 29.9,
      "learning_rate": 1.4195344129554658e-05,
      "loss": 0.0211,
      "step": 1570
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.004464460536837578,
      "eval_runtime": 0.3691,
      "eval_samples_per_second": 51.479,
      "eval_steps_per_second": 8.128,
      "step": 1575
    },
    {
      "epoch": 30.1,
      "learning_rate": 1.3917004048582996e-05,
      "loss": 0.0095,
      "step": 1580
    },
    {
      "epoch": 30.29,
      "learning_rate": 1.3638663967611337e-05,
      "loss": 0.019,
      "step": 1590
    },
    {
      "epoch": 30.48,
      "learning_rate": 1.3360323886639675e-05,
      "loss": 0.0278,
      "step": 1600
    },
    {
      "epoch": 30.67,
      "learning_rate": 1.3081983805668017e-05,
      "loss": 0.0514,
      "step": 1610
    },
    {
      "epoch": 30.86,
      "learning_rate": 1.2803643724696356e-05,
      "loss": 0.0499,
      "step": 1620
    },
    {
      "epoch": 30.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.004182770382612944,
      "eval_runtime": 0.3758,
      "eval_samples_per_second": 50.56,
      "eval_steps_per_second": 7.983,
      "step": 1627
    },
    {
      "epoch": 31.05,
      "learning_rate": 1.2525303643724698e-05,
      "loss": 0.0344,
      "step": 1630
    },
    {
      "epoch": 31.24,
      "learning_rate": 1.2246963562753037e-05,
      "loss": 0.0442,
      "step": 1640
    },
    {
      "epoch": 31.43,
      "learning_rate": 1.1968623481781377e-05,
      "loss": 0.0602,
      "step": 1650
    },
    {
      "epoch": 31.62,
      "learning_rate": 1.1690283400809718e-05,
      "loss": 0.0516,
      "step": 1660
    },
    {
      "epoch": 31.81,
      "learning_rate": 1.1411943319838056e-05,
      "loss": 0.0129,
      "step": 1670
    },
    {
      "epoch": 32.0,
      "learning_rate": 1.1133603238866398e-05,
      "loss": 0.0137,
      "step": 1680
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.004318131599575281,
      "eval_runtime": 0.3645,
      "eval_samples_per_second": 52.132,
      "eval_steps_per_second": 8.231,
      "step": 1680
    },
    {
      "epoch": 32.19,
      "learning_rate": 1.0855263157894737e-05,
      "loss": 0.02,
      "step": 1690
    },
    {
      "epoch": 32.38,
      "learning_rate": 1.0576923076923078e-05,
      "loss": 0.0315,
      "step": 1700
    },
    {
      "epoch": 32.57,
      "learning_rate": 1.0298582995951416e-05,
      "loss": 0.0187,
      "step": 1710
    },
    {
      "epoch": 32.76,
      "learning_rate": 1.0020242914979758e-05,
      "loss": 0.0326,
      "step": 1720
    },
    {
      "epoch": 32.95,
      "learning_rate": 9.741902834008097e-06,
      "loss": 0.0711,
      "step": 1730
    },
    {
      "epoch": 32.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0019276138627901673,
      "eval_runtime": 0.3721,
      "eval_samples_per_second": 51.064,
      "eval_steps_per_second": 8.063,
      "step": 1732
    },
    {
      "epoch": 33.14,
      "learning_rate": 9.463562753036437e-06,
      "loss": 0.0639,
      "step": 1740
    },
    {
      "epoch": 33.33,
      "learning_rate": 9.185222672064777e-06,
      "loss": 0.0381,
      "step": 1750
    },
    {
      "epoch": 33.52,
      "learning_rate": 8.906882591093118e-06,
      "loss": 0.0217,
      "step": 1760
    },
    {
      "epoch": 33.71,
      "learning_rate": 8.628542510121458e-06,
      "loss": 0.024,
      "step": 1770
    },
    {
      "epoch": 33.9,
      "learning_rate": 8.350202429149797e-06,
      "loss": 0.0369,
      "step": 1780
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002073986455798149,
      "eval_runtime": 0.3691,
      "eval_samples_per_second": 51.479,
      "eval_steps_per_second": 8.128,
      "step": 1785
    },
    {
      "epoch": 34.1,
      "learning_rate": 8.071862348178139e-06,
      "loss": 0.0355,
      "step": 1790
    },
    {
      "epoch": 34.29,
      "learning_rate": 7.793522267206478e-06,
      "loss": 0.0566,
      "step": 1800
    },
    {
      "epoch": 34.48,
      "learning_rate": 7.515182186234819e-06,
      "loss": 0.0605,
      "step": 1810
    },
    {
      "epoch": 34.67,
      "learning_rate": 7.2368421052631575e-06,
      "loss": 0.0283,
      "step": 1820
    },
    {
      "epoch": 34.86,
      "learning_rate": 6.958502024291498e-06,
      "loss": 0.0382,
      "step": 1830
    },
    {
      "epoch": 34.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0030744802206754684,
      "eval_runtime": 0.3811,
      "eval_samples_per_second": 49.857,
      "eval_steps_per_second": 7.872,
      "step": 1837
    },
    {
      "epoch": 35.05,
      "learning_rate": 6.680161943319838e-06,
      "loss": 0.0175,
      "step": 1840
    },
    {
      "epoch": 35.24,
      "learning_rate": 6.401821862348178e-06,
      "loss": 0.0406,
      "step": 1850
    },
    {
      "epoch": 35.43,
      "learning_rate": 6.123481781376519e-06,
      "loss": 0.0337,
      "step": 1860
    },
    {
      "epoch": 35.62,
      "learning_rate": 5.845141700404859e-06,
      "loss": 0.0293,
      "step": 1870
    },
    {
      "epoch": 35.81,
      "learning_rate": 5.566801619433199e-06,
      "loss": 0.0492,
      "step": 1880
    },
    {
      "epoch": 36.0,
      "learning_rate": 5.288461538461539e-06,
      "loss": 0.0785,
      "step": 1890
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002902696607634425,
      "eval_runtime": 0.3681,
      "eval_samples_per_second": 51.619,
      "eval_steps_per_second": 8.15,
      "step": 1890
    },
    {
      "epoch": 36.19,
      "learning_rate": 5.010121457489879e-06,
      "loss": 0.0444,
      "step": 1900
    },
    {
      "epoch": 36.38,
      "learning_rate": 4.7317813765182185e-06,
      "loss": 0.0211,
      "step": 1910
    },
    {
      "epoch": 36.57,
      "learning_rate": 4.453441295546559e-06,
      "loss": 0.055,
      "step": 1920
    },
    {
      "epoch": 36.76,
      "learning_rate": 4.175101214574899e-06,
      "loss": 0.1123,
      "step": 1930
    },
    {
      "epoch": 36.95,
      "learning_rate": 3.896761133603239e-06,
      "loss": 0.0575,
      "step": 1940
    },
    {
      "epoch": 36.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.00197758455760777,
      "eval_runtime": 0.3731,
      "eval_samples_per_second": 50.927,
      "eval_steps_per_second": 8.041,
      "step": 1942
    },
    {
      "epoch": 37.14,
      "learning_rate": 3.6184210526315788e-06,
      "loss": 0.0501,
      "step": 1950
    },
    {
      "epoch": 37.33,
      "learning_rate": 3.340080971659919e-06,
      "loss": 0.0605,
      "step": 1960
    },
    {
      "epoch": 37.52,
      "learning_rate": 3.0617408906882593e-06,
      "loss": 0.084,
      "step": 1970
    },
    {
      "epoch": 37.71,
      "learning_rate": 2.7834008097165994e-06,
      "loss": 0.1135,
      "step": 1980
    },
    {
      "epoch": 37.9,
      "learning_rate": 2.5050607287449394e-06,
      "loss": 0.0158,
      "step": 1990
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0019377493299543858,
      "eval_runtime": 0.3741,
      "eval_samples_per_second": 50.791,
      "eval_steps_per_second": 8.02,
      "step": 1995
    },
    {
      "epoch": 38.1,
      "learning_rate": 2.2267206477732795e-06,
      "loss": 0.0538,
      "step": 2000
    },
    {
      "epoch": 38.29,
      "learning_rate": 1.9483805668016196e-06,
      "loss": 0.0355,
      "step": 2010
    },
    {
      "epoch": 38.48,
      "learning_rate": 1.6700404858299594e-06,
      "loss": 0.011,
      "step": 2020
    },
    {
      "epoch": 38.67,
      "learning_rate": 1.3917004048582997e-06,
      "loss": 0.0291,
      "step": 2030
    },
    {
      "epoch": 38.86,
      "learning_rate": 1.1133603238866398e-06,
      "loss": 0.0489,
      "step": 2040
    },
    {
      "epoch": 38.99,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002249736338853836,
      "eval_runtime": 0.3774,
      "eval_samples_per_second": 50.345,
      "eval_steps_per_second": 7.949,
      "step": 2047
    },
    {
      "epoch": 39.05,
      "learning_rate": 8.350202429149797e-07,
      "loss": 0.0618,
      "step": 2050
    },
    {
      "epoch": 39.24,
      "learning_rate": 5.566801619433199e-07,
      "loss": 0.043,
      "step": 2060
    },
    {
      "epoch": 39.43,
      "learning_rate": 2.7834008097165994e-07,
      "loss": 0.036,
      "step": 2070
    },
    {
      "epoch": 39.62,
      "learning_rate": 0.0,
      "loss": 0.0511,
      "step": 2080
    },
    {
      "epoch": 39.62,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0023234980180859566,
      "eval_runtime": 0.3844,
      "eval_samples_per_second": 49.422,
      "eval_steps_per_second": 7.804,
      "step": 2080
    },
    {
      "epoch": 39.62,
      "step": 2080,
      "total_flos": 2.5698668515887145e+18,
      "train_loss": 0.1023936996057343,
      "train_runtime": 630.9729,
      "train_samples_per_second": 53.061,
      "train_steps_per_second": 3.296
    }
  ],
  "logging_steps": 10,
  "max_steps": 2080,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 2.5698668515887145e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}