{
  "best_metric": 0.8041237113402062,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-730",
  "epoch": 9.1,
  "global_step": 3650,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.3698630136986302e-06,
      "loss": 0.0423,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.7397260273972604e-06,
      "loss": 0.4626,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.10958904109589e-06,
      "loss": 0.009,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.479452054794521e-06,
      "loss": 0.1166,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.849315068493151e-06,
      "loss": 0.8124,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.21917808219178e-06,
      "loss": 0.0274,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.589041095890411e-06,
      "loss": 0.1435,
      "step": 70
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0958904109589042e-05,
      "loss": 0.159,
      "step": 80
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2328767123287671e-05,
      "loss": 0.2835,
      "step": 90
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.3698630136986302e-05,
      "loss": 0.0095,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5068493150684931e-05,
      "loss": 0.1436,
      "step": 110
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.643835616438356e-05,
      "loss": 0.0331,
      "step": 120
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.780821917808219e-05,
      "loss": 0.4701,
      "step": 130
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9178082191780822e-05,
      "loss": 0.1941,
      "step": 140
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.0547945205479453e-05,
      "loss": 0.0103,
      "step": 150
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.1917808219178083e-05,
      "loss": 0.4455,
      "step": 160
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.328767123287671e-05,
      "loss": 0.0121,
      "step": 170
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.4657534246575342e-05,
      "loss": 0.3688,
      "step": 180
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.6027397260273973e-05,
      "loss": 0.1885,
      "step": 190
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.7397260273972603e-05,
      "loss": 0.6733,
      "step": 200
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.8767123287671234e-05,
      "loss": 0.1224,
      "step": 210
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.0136986301369862e-05,
      "loss": 0.9795,
      "step": 220
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.1506849315068496e-05,
      "loss": 0.3799,
      "step": 230
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.287671232876712e-05,
      "loss": 0.7247,
      "step": 240
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.424657534246575e-05,
      "loss": 1.259,
      "step": 250
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.561643835616438e-05,
      "loss": 0.1054,
      "step": 260
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.698630136986301e-05,
      "loss": 0.0743,
      "step": 270
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.8356164383561644e-05,
      "loss": 1.3547,
      "step": 280
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.9726027397260274e-05,
      "loss": 0.7676,
      "step": 290
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.1095890410958905e-05,
      "loss": 0.8231,
      "step": 300
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.2465753424657536e-05,
      "loss": 1.247,
      "step": 310
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.383561643835617e-05,
      "loss": 0.5459,
      "step": 320
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.520547945205479e-05,
      "loss": 0.8819,
      "step": 330
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.657534246575342e-05,
      "loss": 0.664,
      "step": 340
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.794520547945205e-05,
      "loss": 1.572,
      "step": 350
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.9315068493150684e-05,
      "loss": 1.1996,
      "step": 360
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.5567010309278351,
      "eval_loss": 2.4460103511810303,
      "eval_runtime": 26.1399,
      "eval_samples_per_second": 3.711,
      "eval_steps_per_second": 3.711,
      "step": 365
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.992389649923896e-05,
      "loss": 1.3704,
      "step": 370
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.977168949771689e-05,
      "loss": 0.3425,
      "step": 380
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.961948249619483e-05,
      "loss": 0.9157,
      "step": 390
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.9467275494672755e-05,
      "loss": 1.1445,
      "step": 400
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.9315068493150684e-05,
      "loss": 1.7488,
      "step": 410
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.916286149162861e-05,
      "loss": 0.684,
      "step": 420
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.901065449010655e-05,
      "loss": 0.6746,
      "step": 430
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.8858447488584476e-05,
      "loss": 1.6721,
      "step": 440
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.8706240487062404e-05,
      "loss": 1.8393,
      "step": 450
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.855403348554033e-05,
      "loss": 1.2233,
      "step": 460
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.840182648401827e-05,
      "loss": 1.2178,
      "step": 470
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.8249619482496196e-05,
      "loss": 1.0698,
      "step": 480
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.8097412480974124e-05,
      "loss": 1.1187,
      "step": 490
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.794520547945205e-05,
      "loss": 1.1796,
      "step": 500
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.779299847792998e-05,
      "loss": 1.4509,
      "step": 510
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.7640791476407916e-05,
      "loss": 0.5875,
      "step": 520
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.7488584474885845e-05,
      "loss": 0.7167,
      "step": 530
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.733637747336377e-05,
      "loss": 0.5177,
      "step": 540
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.71841704718417e-05,
      "loss": 1.7653,
      "step": 550
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.703196347031964e-05,
      "loss": 0.693,
      "step": 560
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.6879756468797565e-05,
      "loss": 0.3745,
      "step": 570
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.6727549467275493e-05,
      "loss": 1.3281,
      "step": 580
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.657534246575342e-05,
      "loss": 1.4898,
      "step": 590
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.642313546423136e-05,
      "loss": 1.3792,
      "step": 600
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.6270928462709285e-05,
      "loss": 0.8248,
      "step": 610
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.6118721461187214e-05,
      "loss": 2.1622,
      "step": 620
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.596651445966514e-05,
      "loss": 0.8459,
      "step": 630
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.581430745814308e-05,
      "loss": 0.9529,
      "step": 640
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.5662100456621006e-05,
      "loss": 1.2896,
      "step": 650
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.5509893455098934e-05,
      "loss": 0.8538,
      "step": 660
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.535768645357686e-05,
      "loss": 1.972,
      "step": 670
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.520547945205479e-05,
      "loss": 2.0179,
      "step": 680
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.5053272450532726e-05,
      "loss": 0.9832,
      "step": 690
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.4901065449010655e-05,
      "loss": 0.5654,
      "step": 700
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.474885844748858e-05,
      "loss": 0.8281,
      "step": 710
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.459665144596651e-05,
      "loss": 0.2361,
      "step": 720
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 1.4174,
      "step": 730
    },
    {
      "epoch": 1.1,
      "eval_accuracy": 0.8041237113402062,
      "eval_loss": 0.6252075433731079,
      "eval_runtime": 26.3776,
      "eval_samples_per_second": 3.677,
      "eval_steps_per_second": 3.677,
      "step": 730
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.4292237442922375e-05,
      "loss": 0.9256,
      "step": 740
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.41400304414003e-05,
      "loss": 0.4378,
      "step": 750
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.398782343987823e-05,
      "loss": 0.4734,
      "step": 760
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.383561643835617e-05,
      "loss": 0.451,
      "step": 770
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.3683409436834095e-05,
      "loss": 0.6936,
      "step": 780
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.3531202435312024e-05,
      "loss": 0.5293,
      "step": 790
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.337899543378995e-05,
      "loss": 0.0223,
      "step": 800
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.322678843226789e-05,
      "loss": 0.1403,
      "step": 810
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.3074581430745816e-05,
      "loss": 1.3285,
      "step": 820
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.2922374429223744e-05,
      "loss": 0.0107,
      "step": 830
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.277016742770167e-05,
      "loss": 0.9825,
      "step": 840
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.26179604261796e-05,
      "loss": 0.4971,
      "step": 850
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.2465753424657536e-05,
      "loss": 0.0056,
      "step": 860
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.2313546423135464e-05,
      "loss": 2.5121,
      "step": 870
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.216133942161339e-05,
      "loss": 1.3604,
      "step": 880
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.200913242009132e-05,
      "loss": 0.1579,
      "step": 890
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.1856925418569256e-05,
      "loss": 0.4522,
      "step": 900
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.1704718417047185e-05,
      "loss": 0.1113,
      "step": 910
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.155251141552511e-05,
      "loss": 0.8716,
      "step": 920
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.140030441400304e-05,
      "loss": 0.5942,
      "step": 930
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.124809741248098e-05,
      "loss": 0.1398,
      "step": 940
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.1095890410958905e-05,
      "loss": 1.299,
      "step": 950
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.0943683409436834e-05,
      "loss": 0.7001,
      "step": 960
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.079147640791476e-05,
      "loss": 1.363,
      "step": 970
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.063926940639269e-05,
      "loss": 0.7667,
      "step": 980
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.0487062404870626e-05,
      "loss": 0.3073,
      "step": 990
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.0334855403348554e-05,
      "loss": 0.069,
      "step": 1000
    },
    {
      "epoch": 2.08,
      "learning_rate": 4.018264840182648e-05,
      "loss": 0.7712,
      "step": 1010
    },
    {
      "epoch": 2.08,
      "learning_rate": 4.003044140030441e-05,
      "loss": 0.8272,
      "step": 1020
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.9878234398782346e-05,
      "loss": 0.6164,
      "step": 1030
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.9726027397260274e-05,
      "loss": 0.0011,
      "step": 1040
    },
    {
      "epoch": 2.09,
      "learning_rate": 3.95738203957382e-05,
      "loss": 0.7969,
      "step": 1050
    },
    {
      "epoch": 2.09,
      "learning_rate": 3.942161339421613e-05,
      "loss": 0.3793,
      "step": 1060
    },
    {
      "epoch": 2.09,
      "learning_rate": 3.9269406392694066e-05,
      "loss": 1.3395,
      "step": 1070
    },
    {
      "epoch": 2.1,
      "learning_rate": 3.9117199391171995e-05,
      "loss": 0.0117,
      "step": 1080
    },
    {
      "epoch": 2.1,
      "learning_rate": 3.896499238964992e-05,
      "loss": 0.0092,
      "step": 1090
    },
    {
      "epoch": 2.1,
      "eval_accuracy": 0.7628865979381443,
      "eval_loss": 1.2255264520645142,
      "eval_runtime": 26.1764,
      "eval_samples_per_second": 3.706,
      "eval_steps_per_second": 3.706,
      "step": 1095
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.881278538812785e-05,
      "loss": 0.3452,
      "step": 1100
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.866057838660579e-05,
      "loss": 0.0055,
      "step": 1110
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.8508371385083715e-05,
      "loss": 0.0049,
      "step": 1120
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.8356164383561644e-05,
      "loss": 0.0045,
      "step": 1130
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.820395738203957e-05,
      "loss": 0.266,
      "step": 1140
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.80517503805175e-05,
      "loss": 0.709,
      "step": 1150
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.7899543378995436e-05,
      "loss": 0.0077,
      "step": 1160
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.7747336377473364e-05,
      "loss": 0.0024,
      "step": 1170
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.759512937595129e-05,
      "loss": 0.0016,
      "step": 1180
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.744292237442922e-05,
      "loss": 0.0326,
      "step": 1190
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.7290715372907156e-05,
      "loss": 0.0517,
      "step": 1200
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.7138508371385084e-05,
      "loss": 0.0405,
      "step": 1210
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.698630136986301e-05,
      "loss": 0.7458,
      "step": 1220
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.683409436834094e-05,
      "loss": 0.002,
      "step": 1230
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.6681887366818876e-05,
      "loss": 0.0044,
      "step": 1240
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.6529680365296805e-05,
      "loss": 0.6527,
      "step": 1250
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.637747336377473e-05,
      "loss": 0.1117,
      "step": 1260
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.622526636225266e-05,
      "loss": 1.9262,
      "step": 1270
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.60730593607306e-05,
      "loss": 0.0107,
      "step": 1280
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.5920852359208525e-05,
      "loss": 0.7209,
      "step": 1290
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.5768645357686453e-05,
      "loss": 0.0092,
      "step": 1300
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.561643835616438e-05,
      "loss": 1.4381,
      "step": 1310
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.546423135464231e-05,
      "loss": 0.3603,
      "step": 1320
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.5312024353120245e-05,
      "loss": 0.5943,
      "step": 1330
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.5159817351598174e-05,
      "loss": 0.0023,
      "step": 1340
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.50076103500761e-05,
      "loss": 0.3005,
      "step": 1350
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.485540334855403e-05,
      "loss": 0.1295,
      "step": 1360
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.4703196347031966e-05,
      "loss": 0.0016,
      "step": 1370
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.4550989345509894e-05,
      "loss": 0.002,
      "step": 1380
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.439878234398782e-05,
      "loss": 0.1203,
      "step": 1390
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.424657534246575e-05,
      "loss": 0.1421,
      "step": 1400
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.4094368340943686e-05,
      "loss": 0.0017,
      "step": 1410
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.3942161339421615e-05,
      "loss": 0.0011,
      "step": 1420
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.378995433789954e-05,
      "loss": 1.7634,
      "step": 1430
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.363774733637747e-05,
      "loss": 0.265,
      "step": 1440
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.34855403348554e-05,
      "loss": 0.0013,
      "step": 1450
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.765,
      "step": 1460
    },
    {
      "epoch": 3.1,
      "eval_accuracy": 0.7731958762886598,
      "eval_loss": 0.9237529039382935,
      "eval_runtime": 26.0231,
      "eval_samples_per_second": 3.727,
      "eval_steps_per_second": 3.727,
      "step": 1460
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.318112633181126e-05,
      "loss": 0.0017,
      "step": 1470
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.302891933028919e-05,
      "loss": 0.0014,
      "step": 1480
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.287671232876712e-05,
      "loss": 1.1364,
      "step": 1490
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.2724505327245055e-05,
      "loss": 0.087,
      "step": 1500
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.2572298325722984e-05,
      "loss": 0.5396,
      "step": 1510
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.242009132420091e-05,
      "loss": 0.0016,
      "step": 1520
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.226788432267884e-05,
      "loss": 0.6971,
      "step": 1530
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.2115677321156776e-05,
      "loss": 0.936,
      "step": 1540
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.1963470319634704e-05,
      "loss": 0.0014,
      "step": 1550
    },
    {
      "epoch": 4.03,
      "learning_rate": 3.181126331811263e-05,
      "loss": 0.2137,
      "step": 1560
    },
    {
      "epoch": 4.03,
      "learning_rate": 3.165905631659056e-05,
      "loss": 0.0012,
      "step": 1570
    },
    {
      "epoch": 4.03,
      "learning_rate": 3.1506849315068496e-05,
      "loss": 0.0047,
      "step": 1580
    },
    {
      "epoch": 4.04,
      "learning_rate": 3.1354642313546424e-05,
      "loss": 0.0012,
      "step": 1590
    },
    {
      "epoch": 4.04,
      "learning_rate": 3.120243531202435e-05,
      "loss": 0.0018,
      "step": 1600
    },
    {
      "epoch": 4.04,
      "learning_rate": 3.105022831050228e-05,
      "loss": 0.0288,
      "step": 1610
    },
    {
      "epoch": 4.04,
      "learning_rate": 3.089802130898021e-05,
      "loss": 0.0017,
      "step": 1620
    },
    {
      "epoch": 4.05,
      "learning_rate": 3.0745814307458145e-05,
      "loss": 0.0012,
      "step": 1630
    },
    {
      "epoch": 4.05,
      "learning_rate": 3.059360730593607e-05,
      "loss": 0.0142,
      "step": 1640
    },
    {
      "epoch": 4.05,
      "learning_rate": 3.0441400304414e-05,
      "loss": 0.1674,
      "step": 1650
    },
    {
      "epoch": 4.05,
      "learning_rate": 3.0289193302891933e-05,
      "loss": 0.4027,
      "step": 1660
    },
    {
      "epoch": 4.06,
      "learning_rate": 3.0136986301369862e-05,
      "loss": 0.0012,
      "step": 1670
    },
    {
      "epoch": 4.06,
      "learning_rate": 2.9984779299847794e-05,
      "loss": 0.0027,
      "step": 1680
    },
    {
      "epoch": 4.06,
      "learning_rate": 2.9832572298325722e-05,
      "loss": 0.0011,
      "step": 1690
    },
    {
      "epoch": 4.07,
      "learning_rate": 2.9680365296803654e-05,
      "loss": 0.001,
      "step": 1700
    },
    {
      "epoch": 4.07,
      "learning_rate": 2.9528158295281582e-05,
      "loss": 0.1668,
      "step": 1710
    },
    {
      "epoch": 4.07,
      "learning_rate": 2.9375951293759514e-05,
      "loss": 0.5638,
      "step": 1720
    },
    {
      "epoch": 4.07,
      "learning_rate": 2.9223744292237442e-05,
      "loss": 0.0009,
      "step": 1730
    },
    {
      "epoch": 4.08,
      "learning_rate": 2.9071537290715374e-05,
      "loss": 1.0001,
      "step": 1740
    },
    {
      "epoch": 4.08,
      "learning_rate": 2.8919330289193303e-05,
      "loss": 0.7653,
      "step": 1750
    },
    {
      "epoch": 4.08,
      "learning_rate": 2.8767123287671234e-05,
      "loss": 0.0047,
      "step": 1760
    },
    {
      "epoch": 4.08,
      "learning_rate": 2.8614916286149163e-05,
      "loss": 0.0018,
      "step": 1770
    },
    {
      "epoch": 4.09,
      "learning_rate": 2.846270928462709e-05,
      "loss": 0.1766,
      "step": 1780
    },
    {
      "epoch": 4.09,
      "learning_rate": 2.8310502283105023e-05,
      "loss": 0.652,
      "step": 1790
    },
    {
      "epoch": 4.09,
      "learning_rate": 2.815829528158295e-05,
      "loss": 0.8817,
      "step": 1800
    },
    {
      "epoch": 4.1,
      "learning_rate": 2.8006088280060883e-05,
      "loss": 0.001,
      "step": 1810
    },
    {
      "epoch": 4.1,
      "learning_rate": 2.785388127853881e-05,
      "loss": 0.1714,
      "step": 1820
    },
    {
      "epoch": 4.1,
      "eval_accuracy": 0.7938144329896907,
      "eval_loss": 1.044911503791809,
      "eval_runtime": 26.3617,
      "eval_samples_per_second": 3.68,
      "eval_steps_per_second": 3.68,
      "step": 1825
    },
    {
      "epoch": 5.0,
      "learning_rate": 2.7701674277016743e-05,
      "loss": 1.0092,
      "step": 1830
    },
    {
      "epoch": 5.0,
      "learning_rate": 2.7549467275494672e-05,
      "loss": 0.2696,
      "step": 1840
    },
    {
      "epoch": 5.01,
      "learning_rate": 2.7397260273972603e-05,
      "loss": 0.0089,
      "step": 1850
    },
    {
      "epoch": 5.01,
      "learning_rate": 2.7245053272450532e-05,
      "loss": 0.8207,
      "step": 1860
    },
    {
      "epoch": 5.01,
      "learning_rate": 2.7092846270928464e-05,
      "loss": 0.0021,
      "step": 1870
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.6940639269406392e-05,
      "loss": 0.5146,
      "step": 1880
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.6788432267884324e-05,
      "loss": 0.0009,
      "step": 1890
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.6636225266362252e-05,
      "loss": 0.1109,
      "step": 1900
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.6484018264840184e-05,
      "loss": 0.0012,
      "step": 1910
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.6331811263318112e-05,
      "loss": 0.0016,
      "step": 1920
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.617960426179604e-05,
      "loss": 0.8104,
      "step": 1930
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.6027397260273973e-05,
      "loss": 1.1465,
      "step": 1940
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.58751902587519e-05,
      "loss": 0.0055,
      "step": 1950
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.5722983257229833e-05,
      "loss": 0.028,
      "step": 1960
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.557077625570776e-05,
      "loss": 0.0615,
      "step": 1970
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.5418569254185693e-05,
      "loss": 0.2375,
      "step": 1980
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.526636225266362e-05,
      "loss": 0.7314,
      "step": 1990
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.5114155251141553e-05,
      "loss": 0.0128,
      "step": 2000
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.496194824961948e-05,
      "loss": 0.7367,
      "step": 2010
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.4809741248097413e-05,
      "loss": 0.001,
      "step": 2020
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.4657534246575342e-05,
      "loss": 0.0012,
      "step": 2030
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.4505327245053274e-05,
      "loss": 0.0032,
      "step": 2040
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.4353120243531202e-05,
      "loss": 0.5217,
      "step": 2050
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.4200913242009134e-05,
      "loss": 0.0011,
      "step": 2060
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.4048706240487062e-05,
      "loss": 0.0039,
      "step": 2070
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.389649923896499e-05,
      "loss": 0.0009,
      "step": 2080
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.3744292237442922e-05,
      "loss": 0.0013,
      "step": 2090
    },
    {
      "epoch": 5.08,
      "learning_rate": 2.359208523592085e-05,
      "loss": 0.0018,
      "step": 2100
    },
    {
      "epoch": 5.08,
      "learning_rate": 2.3439878234398783e-05,
      "loss": 0.0011,
      "step": 2110
    },
    {
      "epoch": 5.08,
      "learning_rate": 2.328767123287671e-05,
      "loss": 0.0262,
      "step": 2120
    },
    {
      "epoch": 5.08,
      "learning_rate": 2.3135464231354643e-05,
      "loss": 0.001,
      "step": 2130
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.298325722983257e-05,
      "loss": 0.0012,
      "step": 2140
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.2831050228310503e-05,
      "loss": 0.0012,
      "step": 2150
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.267884322678843e-05,
      "loss": 0.0112,
      "step": 2160
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.2526636225266363e-05,
      "loss": 0.0593,
      "step": 2170
    },
    {
      "epoch": 5.1,
      "learning_rate": 2.237442922374429e-05,
      "loss": 0.0009,
      "step": 2180
    },
    {
      "epoch": 5.1,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.09,
      "step": 2190
    },
    {
      "epoch": 5.1,
      "eval_accuracy": 0.7628865979381443,
      "eval_loss": 1.474695086479187,
      "eval_runtime": 27.4624,
      "eval_samples_per_second": 3.532,
      "eval_steps_per_second": 3.532,
      "step": 2190
    },
    {
      "epoch": 6.0,
      "learning_rate": 2.207001522070015e-05,
      "loss": 0.0282,
      "step": 2200
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.1917808219178083e-05,
      "loss": 0.001,
      "step": 2210
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.1765601217656012e-05,
      "loss": 0.1037,
      "step": 2220
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.1613394216133944e-05,
      "loss": 0.0009,
      "step": 2230
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.1461187214611872e-05,
      "loss": 0.0006,
      "step": 2240
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.13089802130898e-05,
      "loss": 0.0007,
      "step": 2250
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.1156773211567732e-05,
      "loss": 0.0007,
      "step": 2260
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.100456621004566e-05,
      "loss": 0.0007,
      "step": 2270
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.0852359208523592e-05,
      "loss": 0.0007,
      "step": 2280
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.070015220700152e-05,
      "loss": 0.0007,
      "step": 2290
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.0547945205479453e-05,
      "loss": 0.0007,
      "step": 2300
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.039573820395738e-05,
      "loss": 0.0031,
      "step": 2310
    },
    {
      "epoch": 6.04,
      "learning_rate": 2.0243531202435313e-05,
      "loss": 0.2555,
      "step": 2320
    },
    {
      "epoch": 6.04,
      "learning_rate": 2.009132420091324e-05,
      "loss": 0.0008,
      "step": 2330
    },
    {
      "epoch": 6.04,
      "learning_rate": 1.9939117199391173e-05,
      "loss": 0.0008,
      "step": 2340
    },
    {
      "epoch": 6.04,
      "learning_rate": 1.97869101978691e-05,
      "loss": 0.0006,
      "step": 2350
    },
    {
      "epoch": 6.05,
      "learning_rate": 1.9634703196347033e-05,
      "loss": 0.0009,
      "step": 2360
    },
    {
      "epoch": 6.05,
      "learning_rate": 1.948249619482496e-05,
      "loss": 0.0385,
      "step": 2370
    },
    {
      "epoch": 6.05,
      "learning_rate": 1.9330289193302893e-05,
      "loss": 0.0008,
      "step": 2380
    },
    {
      "epoch": 6.05,
      "learning_rate": 1.9178082191780822e-05,
      "loss": 0.0006,
      "step": 2390
    },
    {
      "epoch": 6.06,
      "learning_rate": 1.902587519025875e-05,
      "loss": 0.0009,
      "step": 2400
    },
    {
      "epoch": 6.06,
      "learning_rate": 1.8873668188736682e-05,
      "loss": 0.0006,
      "step": 2410
    },
    {
      "epoch": 6.06,
      "learning_rate": 1.872146118721461e-05,
      "loss": 0.0008,
      "step": 2420
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.8569254185692542e-05,
      "loss": 0.0008,
      "step": 2430
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.841704718417047e-05,
      "loss": 0.0007,
      "step": 2440
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.8264840182648402e-05,
      "loss": 0.0008,
      "step": 2450
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.811263318112633e-05,
      "loss": 0.0008,
      "step": 2460
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.7960426179604263e-05,
      "loss": 0.0006,
      "step": 2470
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.780821917808219e-05,
      "loss": 0.0006,
      "step": 2480
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.7656012176560123e-05,
      "loss": 0.1391,
      "step": 2490
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.750380517503805e-05,
      "loss": 0.0006,
      "step": 2500
    },
    {
      "epoch": 6.09,
      "learning_rate": 1.7351598173515983e-05,
      "loss": 0.0005,
      "step": 2510
    },
    {
      "epoch": 6.09,
      "learning_rate": 1.719939117199391e-05,
      "loss": 0.0005,
      "step": 2520
    },
    {
      "epoch": 6.09,
      "learning_rate": 1.7047184170471843e-05,
      "loss": 0.5689,
      "step": 2530
    },
    {
      "epoch": 6.1,
      "learning_rate": 1.689497716894977e-05,
      "loss": 0.0007,
      "step": 2540
    },
    {
      "epoch": 6.1,
      "learning_rate": 1.67427701674277e-05,
      "loss": 0.0006,
      "step": 2550
    },
    {
      "epoch": 6.1,
      "eval_accuracy": 0.7835051546391752,
      "eval_loss": 1.3157318830490112,
      "eval_runtime": 27.9871,
      "eval_samples_per_second": 3.466,
      "eval_steps_per_second": 3.466,
      "step": 2555
    },
    {
      "epoch": 7.0,
      "learning_rate": 1.659056316590563e-05,
      "loss": 0.0005,
      "step": 2560
    },
    {
      "epoch": 7.0,
      "learning_rate": 1.643835616438356e-05,
      "loss": 0.0006,
      "step": 2570
    },
    {
      "epoch": 7.01,
      "learning_rate": 1.6286149162861492e-05,
      "loss": 0.0006,
      "step": 2580
    },
    {
      "epoch": 7.01,
      "learning_rate": 1.613394216133942e-05,
      "loss": 0.0055,
      "step": 2590
    },
    {
      "epoch": 7.01,
      "learning_rate": 1.5981735159817352e-05,
      "loss": 0.0007,
      "step": 2600
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.582952815829528e-05,
      "loss": 0.0007,
      "step": 2610
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.5677321156773212e-05,
      "loss": 0.0008,
      "step": 2620
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.552511415525114e-05,
      "loss": 0.0008,
      "step": 2630
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.5372907153729072e-05,
      "loss": 0.0007,
      "step": 2640
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.5220700152207e-05,
      "loss": 0.0006,
      "step": 2650
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.5068493150684931e-05,
      "loss": 0.0008,
      "step": 2660
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.4916286149162861e-05,
      "loss": 0.0007,
      "step": 2670
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.4764079147640791e-05,
      "loss": 0.0132,
      "step": 2680
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.4611872146118721e-05,
      "loss": 0.0017,
      "step": 2690
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.4459665144596651e-05,
      "loss": 0.0006,
      "step": 2700
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.4307458143074581e-05,
      "loss": 0.0007,
      "step": 2710
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.4155251141552511e-05,
      "loss": 0.0005,
      "step": 2720
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.4003044140030442e-05,
      "loss": 0.8613,
      "step": 2730
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.3850837138508372e-05,
      "loss": 0.0008,
      "step": 2740
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.3698630136986302e-05,
      "loss": 0.0008,
      "step": 2750
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.3546423135464232e-05,
      "loss": 0.0007,
      "step": 2760
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.3394216133942162e-05,
      "loss": 0.0004,
      "step": 2770
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.3242009132420092e-05,
      "loss": 0.0005,
      "step": 2780
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.308980213089802e-05,
      "loss": 0.0006,
      "step": 2790
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.293759512937595e-05,
      "loss": 0.0007,
      "step": 2800
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.278538812785388e-05,
      "loss": 0.008,
      "step": 2810
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.263318112633181e-05,
      "loss": 0.0005,
      "step": 2820
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.248097412480974e-05,
      "loss": 0.7276,
      "step": 2830
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.2328767123287671e-05,
      "loss": 0.0005,
      "step": 2840
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.2176560121765601e-05,
      "loss": 0.0009,
      "step": 2850
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.2024353120243531e-05,
      "loss": 0.0005,
      "step": 2860
    },
    {
      "epoch": 7.09,
      "learning_rate": 1.1872146118721461e-05,
      "loss": 0.0006,
      "step": 2870
    },
    {
      "epoch": 7.09,
      "learning_rate": 1.1719939117199391e-05,
      "loss": 0.009,
      "step": 2880
    },
    {
      "epoch": 7.09,
      "learning_rate": 1.1567732115677321e-05,
      "loss": 0.0008,
      "step": 2890
    },
    {
      "epoch": 7.09,
      "learning_rate": 1.1415525114155251e-05,
      "loss": 0.0587,
      "step": 2900
    },
    {
      "epoch": 7.1,
      "learning_rate": 1.1263318112633182e-05,
      "loss": 0.0006,
      "step": 2910
    },
    {
      "epoch": 7.1,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0004,
      "step": 2920
    },
    {
      "epoch": 7.1,
      "eval_accuracy": 0.7216494845360825,
      "eval_loss": 1.6183393001556396,
      "eval_runtime": 27.6439,
      "eval_samples_per_second": 3.509,
      "eval_steps_per_second": 3.509,
      "step": 2920
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.0958904109589042e-05,
      "loss": 0.0006,
      "step": 2930
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.0806697108066972e-05,
      "loss": 0.0016,
      "step": 2940
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.06544901065449e-05,
      "loss": 0.0007,
      "step": 2950
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.050228310502283e-05,
      "loss": 0.8142,
      "step": 2960
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.035007610350076e-05,
      "loss": 0.0007,
      "step": 2970
    },
    {
      "epoch": 8.02,
      "learning_rate": 1.019786910197869e-05,
      "loss": 0.0007,
      "step": 2980
    },
    {
      "epoch": 8.02,
      "learning_rate": 1.004566210045662e-05,
      "loss": 0.0006,
      "step": 2990
    },
    {
      "epoch": 8.02,
      "learning_rate": 9.89345509893455e-06,
      "loss": 0.0007,
      "step": 3000
    },
    {
      "epoch": 8.02,
      "learning_rate": 9.74124809741248e-06,
      "loss": 0.0006,
      "step": 3010
    },
    {
      "epoch": 8.03,
      "learning_rate": 9.589041095890411e-06,
      "loss": 0.0006,
      "step": 3020
    },
    {
      "epoch": 8.03,
      "learning_rate": 9.436834094368341e-06,
      "loss": 0.0005,
      "step": 3030
    },
    {
      "epoch": 8.03,
      "learning_rate": 9.284627092846271e-06,
      "loss": 0.0424,
      "step": 3040
    },
    {
      "epoch": 8.04,
      "learning_rate": 9.132420091324201e-06,
      "loss": 0.0006,
      "step": 3050
    },
    {
      "epoch": 8.04,
      "learning_rate": 8.980213089802131e-06,
      "loss": 0.001,
      "step": 3060
    },
    {
      "epoch": 8.04,
      "learning_rate": 8.828006088280061e-06,
      "loss": 0.0006,
      "step": 3070
    },
    {
      "epoch": 8.04,
      "learning_rate": 8.675799086757991e-06,
      "loss": 0.0014,
      "step": 3080
    },
    {
      "epoch": 8.05,
      "learning_rate": 8.523592085235922e-06,
      "loss": 0.0005,
      "step": 3090
    },
    {
      "epoch": 8.05,
      "learning_rate": 8.37138508371385e-06,
      "loss": 0.0006,
      "step": 3100
    },
    {
      "epoch": 8.05,
      "learning_rate": 8.21917808219178e-06,
      "loss": 0.0006,
      "step": 3110
    },
    {
      "epoch": 8.05,
      "learning_rate": 8.06697108066971e-06,
      "loss": 0.0017,
      "step": 3120
    },
    {
      "epoch": 8.06,
      "learning_rate": 7.91476407914764e-06,
      "loss": 0.0005,
      "step": 3130
    },
    {
      "epoch": 8.06,
      "learning_rate": 7.76255707762557e-06,
      "loss": 0.0007,
      "step": 3140
    },
    {
      "epoch": 8.06,
      "learning_rate": 7.6103500761035e-06,
      "loss": 0.0005,
      "step": 3150
    },
    {
      "epoch": 8.07,
      "learning_rate": 7.4581430745814305e-06,
      "loss": 0.0005,
      "step": 3160
    },
    {
      "epoch": 8.07,
      "learning_rate": 7.305936073059361e-06,
      "loss": 0.0005,
      "step": 3170
    },
    {
      "epoch": 8.07,
      "learning_rate": 7.153729071537291e-06,
      "loss": 0.0006,
      "step": 3180
    },
    {
      "epoch": 8.07,
      "learning_rate": 7.001522070015221e-06,
      "loss": 0.0005,
      "step": 3190
    },
    {
      "epoch": 8.08,
      "learning_rate": 6.849315068493151e-06,
      "loss": 0.0005,
      "step": 3200
    },
    {
      "epoch": 8.08,
      "learning_rate": 6.697108066971081e-06,
      "loss": 0.0005,
      "step": 3210
    },
    {
      "epoch": 8.08,
      "learning_rate": 6.54490106544901e-06,
      "loss": 0.004,
      "step": 3220
    },
    {
      "epoch": 8.08,
      "learning_rate": 6.39269406392694e-06,
      "loss": 0.0006,
      "step": 3230
    },
    {
      "epoch": 8.09,
      "learning_rate": 6.24048706240487e-06,
      "loss": 0.0007,
      "step": 3240
    },
    {
      "epoch": 8.09,
      "learning_rate": 6.0882800608828005e-06,
      "loss": 0.0016,
      "step": 3250
    },
    {
      "epoch": 8.09,
      "learning_rate": 5.936073059360731e-06,
      "loss": 0.0005,
      "step": 3260
    },
    {
      "epoch": 8.1,
      "learning_rate": 5.783866057838661e-06,
      "loss": 0.0006,
      "step": 3270
    },
    {
      "epoch": 8.1,
      "learning_rate": 5.631659056316591e-06,
      "loss": 0.0012,
      "step": 3280
    },
    {
      "epoch": 8.1,
      "eval_accuracy": 0.7731958762886598,
      "eval_loss": 1.5373260974884033,
      "eval_runtime": 28.4296,
      "eval_samples_per_second": 3.412,
      "eval_steps_per_second": 3.412,
      "step": 3285
    },
    {
      "epoch": 9.0,
      "learning_rate": 5.479452054794521e-06,
      "loss": 0.0005,
      "step": 3290
    },
    {
      "epoch": 9.0,
      "learning_rate": 5.32724505327245e-06,
      "loss": 0.0005,
      "step": 3300
    },
    {
      "epoch": 9.01,
      "learning_rate": 5.17503805175038e-06,
      "loss": 0.0013,
      "step": 3310
    },
    {
      "epoch": 9.01,
      "learning_rate": 5.02283105022831e-06,
      "loss": 0.0005,
      "step": 3320
    },
    {
      "epoch": 9.01,
      "learning_rate": 4.87062404870624e-06,
      "loss": 0.0006,
      "step": 3330
    },
    {
      "epoch": 9.02,
      "learning_rate": 4.7184170471841705e-06,
      "loss": 0.0005,
      "step": 3340
    },
    {
      "epoch": 9.02,
      "learning_rate": 4.566210045662101e-06,
      "loss": 0.0005,
      "step": 3350
    },
    {
      "epoch": 9.02,
      "learning_rate": 4.414003044140031e-06,
      "loss": 0.0004,
      "step": 3360
    },
    {
      "epoch": 9.02,
      "learning_rate": 4.261796042617961e-06,
      "loss": 0.0008,
      "step": 3370
    },
    {
      "epoch": 9.03,
      "learning_rate": 4.10958904109589e-06,
      "loss": 0.0005,
      "step": 3380
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.95738203957382e-06,
      "loss": 0.0005,
      "step": 3390
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.80517503805175e-06,
      "loss": 0.0007,
      "step": 3400
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.6529680365296803e-06,
      "loss": 0.0008,
      "step": 3410
    },
    {
      "epoch": 9.04,
      "learning_rate": 3.5007610350076104e-06,
      "loss": 0.0005,
      "step": 3420
    },
    {
      "epoch": 9.04,
      "learning_rate": 3.3485540334855405e-06,
      "loss": 0.0005,
      "step": 3430
    },
    {
      "epoch": 9.04,
      "learning_rate": 3.19634703196347e-06,
      "loss": 0.0007,
      "step": 3440
    },
    {
      "epoch": 9.05,
      "learning_rate": 3.0441400304414002e-06,
      "loss": 0.0006,
      "step": 3450
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.8919330289193303e-06,
      "loss": 0.0004,
      "step": 3460
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.7397260273972604e-06,
      "loss": 0.0005,
      "step": 3470
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.58751902587519e-06,
      "loss": 0.0004,
      "step": 3480
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.43531202435312e-06,
      "loss": 0.0005,
      "step": 3490
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.2831050228310503e-06,
      "loss": 0.0006,
      "step": 3500
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.1308980213089804e-06,
      "loss": 0.0006,
      "step": 3510
    },
    {
      "epoch": 9.06,
      "learning_rate": 1.97869101978691e-06,
      "loss": 0.0682,
      "step": 3520
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.8264840182648401e-06,
      "loss": 0.0004,
      "step": 3530
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.6742770167427702e-06,
      "loss": 0.0005,
      "step": 3540
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.5220700152207001e-06,
      "loss": 0.0005,
      "step": 3550
    },
    {
      "epoch": 9.08,
      "learning_rate": 1.3698630136986302e-06,
      "loss": 0.0006,
      "step": 3560
    },
    {
      "epoch": 9.08,
      "learning_rate": 1.21765601217656e-06,
      "loss": 0.0007,
      "step": 3570
    },
    {
      "epoch": 9.08,
      "learning_rate": 1.0654490106544902e-06,
      "loss": 0.0005,
      "step": 3580
    },
    {
      "epoch": 9.08,
      "learning_rate": 9.132420091324201e-07,
      "loss": 0.0005,
      "step": 3590
    },
    {
      "epoch": 9.09,
      "learning_rate": 7.610350076103501e-07,
      "loss": 0.7116,
      "step": 3600
    },
    {
      "epoch": 9.09,
      "learning_rate": 6.0882800608828e-07,
      "loss": 0.0694,
      "step": 3610
    },
    {
      "epoch": 9.09,
      "learning_rate": 4.5662100456621004e-07,
      "loss": 0.0006,
      "step": 3620
    },
    {
      "epoch": 9.09,
      "learning_rate": 3.0441400304414e-07,
      "loss": 0.0006,
      "step": 3630
    },
    {
      "epoch": 9.1,
      "learning_rate": 1.5220700152207e-07,
      "loss": 0.0004,
      "step": 3640
    },
    {
      "epoch": 9.1,
      "learning_rate": 0.0,
      "loss": 0.0005,
      "step": 3650
    },
    {
      "epoch": 9.1,
      "eval_accuracy": 0.7628865979381443,
      "eval_loss": 1.67752206325531,
      "eval_runtime": 28.4688,
      "eval_samples_per_second": 3.407,
      "eval_steps_per_second": 3.407,
      "step": 3650
    },
    {
      "epoch": 9.1,
      "step": 3650,
      "total_flos": 4.5485413607079936e+18,
      "train_loss": 0.30888694136824507,
      "train_runtime": 2930.7636,
      "train_samples_per_second": 1.245,
      "train_steps_per_second": 1.245
    },
    {
      "epoch": 9.1,
      "eval_accuracy": 0.7897727272727273,
      "eval_loss": 0.6162788271903992,
      "eval_runtime": 56.0111,
      "eval_samples_per_second": 3.142,
      "eval_steps_per_second": 3.142,
      "step": 3650
    },
    {
      "epoch": 9.1,
      "eval_accuracy": 0.7885714285714286,
      "eval_loss": 0.6197737455368042,
      "eval_runtime": 54.239,
      "eval_samples_per_second": 3.226,
      "eval_steps_per_second": 3.226,
      "step": 3650
    }
  ],
  "max_steps": 3650,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 4.5485413607079936e+18,
  "trial_name": null,
  "trial_params": null
}