{
  "best_metric": 0.7391304347826086,
  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-ve-U13-b-80\\checkpoint-91",
  "epoch": 73.84615384615384,
  "eval_steps": 500,
  "global_step": 480,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.92,
      "eval_accuracy": 0.13043478260869565,
      "eval_loss": 1.385754942893982,
      "eval_runtime": 0.7438,
      "eval_samples_per_second": 61.847,
      "eval_steps_per_second": 2.689,
      "step": 6
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.2916666666666667e-05,
      "loss": 1.3856,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3695652173913043,
      "eval_loss": 1.377731204032898,
      "eval_runtime": 0.7263,
      "eval_samples_per_second": 63.33,
      "eval_steps_per_second": 2.753,
      "step": 13
    },
    {
      "epoch": 2.92,
      "eval_accuracy": 0.2391304347826087,
      "eval_loss": 1.348819375038147,
      "eval_runtime": 0.75,
      "eval_samples_per_second": 61.336,
      "eval_steps_per_second": 2.667,
      "step": 19
    },
    {
      "epoch": 3.08,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 1.361,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2826086956521739,
      "eval_loss": 1.2502505779266357,
      "eval_runtime": 0.7438,
      "eval_samples_per_second": 61.841,
      "eval_steps_per_second": 2.689,
      "step": 26
    },
    {
      "epoch": 4.62,
      "learning_rate": 5.4276315789473686e-05,
      "loss": 1.2088,
      "step": 30
    },
    {
      "epoch": 4.92,
      "eval_accuracy": 0.41304347826086957,
      "eval_loss": 1.1317055225372314,
      "eval_runtime": 0.7105,
      "eval_samples_per_second": 64.741,
      "eval_steps_per_second": 2.815,
      "step": 32
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.0243556499481201,
      "eval_runtime": 0.7994,
      "eval_samples_per_second": 57.543,
      "eval_steps_per_second": 2.502,
      "step": 39
    },
    {
      "epoch": 6.15,
      "learning_rate": 5.3070175438596496e-05,
      "loss": 1.0729,
      "step": 40
    },
    {
      "epoch": 6.92,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.041305422782898,
      "eval_runtime": 0.7893,
      "eval_samples_per_second": 58.282,
      "eval_steps_per_second": 2.534,
      "step": 45
    },
    {
      "epoch": 7.69,
      "learning_rate": 5.18640350877193e-05,
      "loss": 0.9554,
      "step": 50
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5652173913043478,
      "eval_loss": 0.9286284446716309,
      "eval_runtime": 0.7455,
      "eval_samples_per_second": 61.704,
      "eval_steps_per_second": 2.683,
      "step": 52
    },
    {
      "epoch": 8.92,
      "eval_accuracy": 0.5652173913043478,
      "eval_loss": 0.9103382229804993,
      "eval_runtime": 0.7349,
      "eval_samples_per_second": 62.593,
      "eval_steps_per_second": 2.721,
      "step": 58
    },
    {
      "epoch": 9.23,
      "learning_rate": 5.0657894736842104e-05,
      "loss": 0.8221,
      "step": 60
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8519319295883179,
      "eval_runtime": 0.7139,
      "eval_samples_per_second": 64.431,
      "eval_steps_per_second": 2.801,
      "step": 65
    },
    {
      "epoch": 10.77,
      "learning_rate": 4.9451754385964915e-05,
      "loss": 0.732,
      "step": 70
    },
    {
      "epoch": 10.92,
      "eval_accuracy": 0.5869565217391305,
      "eval_loss": 0.8299721479415894,
      "eval_runtime": 0.7596,
      "eval_samples_per_second": 60.555,
      "eval_steps_per_second": 2.633,
      "step": 71
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 0.810260534286499,
      "eval_runtime": 0.7348,
      "eval_samples_per_second": 62.603,
      "eval_steps_per_second": 2.722,
      "step": 78
    },
    {
      "epoch": 12.31,
      "learning_rate": 4.824561403508772e-05,
      "loss": 0.6491,
      "step": 80
    },
    {
      "epoch": 12.92,
      "eval_accuracy": 0.5869565217391305,
      "eval_loss": 0.9533290266990662,
      "eval_runtime": 0.736,
      "eval_samples_per_second": 62.5,
      "eval_steps_per_second": 2.717,
      "step": 84
    },
    {
      "epoch": 13.85,
      "learning_rate": 4.703947368421053e-05,
      "loss": 0.5724,
      "step": 90
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7391304347826086,
      "eval_loss": 0.7882330417633057,
      "eval_runtime": 0.7754,
      "eval_samples_per_second": 59.325,
      "eval_steps_per_second": 2.579,
      "step": 91
    },
    {
      "epoch": 14.92,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 0.8071939945220947,
      "eval_runtime": 0.7958,
      "eval_samples_per_second": 57.807,
      "eval_steps_per_second": 2.513,
      "step": 97
    },
    {
      "epoch": 15.38,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 0.5305,
      "step": 100
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7391304347826086,
      "eval_loss": 0.7651436924934387,
      "eval_runtime": 0.7895,
      "eval_samples_per_second": 58.267,
      "eval_steps_per_second": 2.533,
      "step": 104
    },
    {
      "epoch": 16.92,
      "learning_rate": 4.462719298245614e-05,
      "loss": 0.4879,
      "step": 110
    },
    {
      "epoch": 16.92,
      "eval_accuracy": 0.717391304347826,
      "eval_loss": 0.7378753423690796,
      "eval_runtime": 0.7421,
      "eval_samples_per_second": 61.986,
      "eval_steps_per_second": 2.695,
      "step": 110
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 0.7589738368988037,
      "eval_runtime": 0.8499,
      "eval_samples_per_second": 54.125,
      "eval_steps_per_second": 2.353,
      "step": 117
    },
    {
      "epoch": 18.46,
      "learning_rate": 4.342105263157895e-05,
      "loss": 0.4346,
      "step": 120
    },
    {
      "epoch": 18.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 0.9283233284950256,
      "eval_runtime": 0.8223,
      "eval_samples_per_second": 55.938,
      "eval_steps_per_second": 2.432,
      "step": 123
    },
    {
      "epoch": 20.0,
      "learning_rate": 4.221491228070176e-05,
      "loss": 0.3671,
      "step": 130
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.0188376903533936,
      "eval_runtime": 0.7701,
      "eval_samples_per_second": 59.73,
      "eval_steps_per_second": 2.597,
      "step": 130
    },
    {
      "epoch": 20.92,
      "eval_accuracy": 0.7391304347826086,
      "eval_loss": 0.895862340927124,
      "eval_runtime": 0.8493,
      "eval_samples_per_second": 54.165,
      "eval_steps_per_second": 2.355,
      "step": 136
    },
    {
      "epoch": 21.54,
      "learning_rate": 4.100877192982456e-05,
      "loss": 0.3725,
      "step": 140
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 0.950153112411499,
      "eval_runtime": 0.7826,
      "eval_samples_per_second": 58.782,
      "eval_steps_per_second": 2.556,
      "step": 143
    },
    {
      "epoch": 22.92,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9626518487930298,
      "eval_runtime": 0.749,
      "eval_samples_per_second": 61.418,
      "eval_steps_per_second": 2.67,
      "step": 149
    },
    {
      "epoch": 23.08,
      "learning_rate": 3.9802631578947374e-05,
      "loss": 0.3321,
      "step": 150
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 0.9619277715682983,
      "eval_runtime": 0.767,
      "eval_samples_per_second": 59.974,
      "eval_steps_per_second": 2.608,
      "step": 156
    },
    {
      "epoch": 24.62,
      "learning_rate": 3.859649122807018e-05,
      "loss": 0.3376,
      "step": 160
    },
    {
      "epoch": 24.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.0459153652191162,
      "eval_runtime": 0.7694,
      "eval_samples_per_second": 59.79,
      "eval_steps_per_second": 2.6,
      "step": 162
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.0166922807693481,
      "eval_runtime": 0.7571,
      "eval_samples_per_second": 60.757,
      "eval_steps_per_second": 2.642,
      "step": 169
    },
    {
      "epoch": 26.15,
      "learning_rate": 3.739035087719298e-05,
      "loss": 0.3699,
      "step": 170
    },
    {
      "epoch": 26.92,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 0.9949257969856262,
      "eval_runtime": 0.8034,
      "eval_samples_per_second": 57.258,
      "eval_steps_per_second": 2.489,
      "step": 175
    },
    {
      "epoch": 27.69,
      "learning_rate": 3.618421052631579e-05,
      "loss": 0.3098,
      "step": 180
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 0.9944018721580505,
      "eval_runtime": 0.7886,
      "eval_samples_per_second": 58.329,
      "eval_steps_per_second": 2.536,
      "step": 182
    },
    {
      "epoch": 28.92,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.0860100984573364,
      "eval_runtime": 0.7948,
      "eval_samples_per_second": 57.876,
      "eval_steps_per_second": 2.516,
      "step": 188
    },
    {
      "epoch": 29.23,
      "learning_rate": 3.49780701754386e-05,
      "loss": 0.253,
      "step": 190
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.1720929145812988,
      "eval_runtime": 0.8196,
      "eval_samples_per_second": 56.128,
      "eval_steps_per_second": 2.44,
      "step": 195
    },
    {
      "epoch": 30.77,
      "learning_rate": 3.377192982456141e-05,
      "loss": 0.2615,
      "step": 200
    },
    {
      "epoch": 30.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.1625701189041138,
      "eval_runtime": 0.7635,
      "eval_samples_per_second": 60.248,
      "eval_steps_per_second": 2.619,
      "step": 201
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.2463836669921875,
      "eval_runtime": 0.7858,
      "eval_samples_per_second": 58.541,
      "eval_steps_per_second": 2.545,
      "step": 208
    },
    {
      "epoch": 32.31,
      "learning_rate": 3.256578947368421e-05,
      "loss": 0.242,
      "step": 210
    },
    {
      "epoch": 32.92,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.2178901433944702,
      "eval_runtime": 0.7976,
      "eval_samples_per_second": 57.676,
      "eval_steps_per_second": 2.508,
      "step": 214
    },
    {
      "epoch": 33.85,
      "learning_rate": 3.1359649122807015e-05,
      "loss": 0.2173,
      "step": 220
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.2406569719314575,
      "eval_runtime": 0.7647,
      "eval_samples_per_second": 60.158,
      "eval_steps_per_second": 2.616,
      "step": 221
    },
    {
      "epoch": 34.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.1584811210632324,
      "eval_runtime": 0.8017,
      "eval_samples_per_second": 57.382,
      "eval_steps_per_second": 2.495,
      "step": 227
    },
    {
      "epoch": 35.38,
      "learning_rate": 3.0153508771929826e-05,
      "loss": 0.2305,
      "step": 230
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.304764747619629,
      "eval_runtime": 0.7581,
      "eval_samples_per_second": 60.68,
      "eval_steps_per_second": 2.638,
      "step": 234
    },
    {
      "epoch": 36.92,
      "learning_rate": 2.894736842105263e-05,
      "loss": 0.2114,
      "step": 240
    },
    {
      "epoch": 36.92,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.1775559186935425,
      "eval_runtime": 0.724,
      "eval_samples_per_second": 63.534,
      "eval_steps_per_second": 2.762,
      "step": 240
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.1460415124893188,
      "eval_runtime": 0.7523,
      "eval_samples_per_second": 61.147,
      "eval_steps_per_second": 2.659,
      "step": 247
    },
    {
      "epoch": 38.46,
      "learning_rate": 2.774122807017544e-05,
      "loss": 0.2243,
      "step": 250
    },
    {
      "epoch": 38.92,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.2424203157424927,
      "eval_runtime": 0.759,
      "eval_samples_per_second": 60.607,
      "eval_steps_per_second": 2.635,
      "step": 253
    },
    {
      "epoch": 40.0,
      "learning_rate": 2.6535087719298248e-05,
      "loss": 0.1822,
      "step": 260
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.2804019451141357,
      "eval_runtime": 0.7624,
      "eval_samples_per_second": 60.336,
      "eval_steps_per_second": 2.623,
      "step": 260
    },
    {
      "epoch": 40.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3472291231155396,
      "eval_runtime": 0.7688,
      "eval_samples_per_second": 59.835,
      "eval_steps_per_second": 2.602,
      "step": 266
    },
    {
      "epoch": 41.54,
      "learning_rate": 2.5328947368421052e-05,
      "loss": 0.2065,
      "step": 270
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3632473945617676,
      "eval_runtime": 0.775,
      "eval_samples_per_second": 59.355,
      "eval_steps_per_second": 2.581,
      "step": 273
    },
    {
      "epoch": 42.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.28322434425354,
      "eval_runtime": 0.7318,
      "eval_samples_per_second": 62.856,
      "eval_steps_per_second": 2.733,
      "step": 279
    },
    {
      "epoch": 43.08,
      "learning_rate": 2.412280701754386e-05,
      "loss": 0.1942,
      "step": 280
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3499536514282227,
      "eval_runtime": 0.7849,
      "eval_samples_per_second": 58.604,
      "eval_steps_per_second": 2.548,
      "step": 286
    },
    {
      "epoch": 44.62,
      "learning_rate": 2.2916666666666667e-05,
      "loss": 0.1699,
      "step": 290
    },
    {
      "epoch": 44.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3242206573486328,
      "eval_runtime": 0.7599,
      "eval_samples_per_second": 60.533,
      "eval_steps_per_second": 2.632,
      "step": 292
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.3188772201538086,
      "eval_runtime": 0.7449,
      "eval_samples_per_second": 61.753,
      "eval_steps_per_second": 2.685,
      "step": 299
    },
    {
      "epoch": 46.15,
      "learning_rate": 2.1710526315789474e-05,
      "loss": 0.1764,
      "step": 300
    },
    {
      "epoch": 46.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.2839657068252563,
      "eval_runtime": 0.723,
      "eval_samples_per_second": 63.626,
      "eval_steps_per_second": 2.766,
      "step": 305
    },
    {
      "epoch": 47.69,
      "learning_rate": 2.050438596491228e-05,
      "loss": 0.1771,
      "step": 310
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.3069181442260742,
      "eval_runtime": 0.8128,
      "eval_samples_per_second": 56.596,
      "eval_steps_per_second": 2.461,
      "step": 312
    },
    {
      "epoch": 48.92,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.1584603786468506,
      "eval_runtime": 0.7739,
      "eval_samples_per_second": 59.436,
      "eval_steps_per_second": 2.584,
      "step": 318
    },
    {
      "epoch": 49.23,
      "learning_rate": 1.929824561403509e-05,
      "loss": 0.2095,
      "step": 320
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.3702274560928345,
      "eval_runtime": 0.7551,
      "eval_samples_per_second": 60.918,
      "eval_steps_per_second": 2.649,
      "step": 325
    },
    {
      "epoch": 50.77,
      "learning_rate": 1.8092105263157896e-05,
      "loss": 0.1404,
      "step": 330
    },
    {
      "epoch": 50.92,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.3539283275604248,
      "eval_runtime": 0.7611,
      "eval_samples_per_second": 60.437,
      "eval_steps_per_second": 2.628,
      "step": 331
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.372290015220642,
      "eval_runtime": 0.7589,
      "eval_samples_per_second": 60.611,
      "eval_steps_per_second": 2.635,
      "step": 338
    },
    {
      "epoch": 52.31,
      "learning_rate": 1.6885964912280704e-05,
      "loss": 0.1449,
      "step": 340
    },
    {
      "epoch": 52.92,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.387741208076477,
      "eval_runtime": 0.7822,
      "eval_samples_per_second": 58.807,
      "eval_steps_per_second": 2.557,
      "step": 344
    },
    {
      "epoch": 53.85,
      "learning_rate": 1.5679824561403508e-05,
      "loss": 0.1348,
      "step": 350
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3381210565567017,
      "eval_runtime": 0.7641,
      "eval_samples_per_second": 60.202,
      "eval_steps_per_second": 2.617,
      "step": 351
    },
    {
      "epoch": 54.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3700389862060547,
      "eval_runtime": 0.7668,
      "eval_samples_per_second": 59.989,
      "eval_steps_per_second": 2.608,
      "step": 357
    },
    {
      "epoch": 55.38,
      "learning_rate": 1.4473684210526315e-05,
      "loss": 0.1683,
      "step": 360
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.2870872020721436,
      "eval_runtime": 0.7471,
      "eval_samples_per_second": 61.572,
      "eval_steps_per_second": 2.677,
      "step": 364
    },
    {
      "epoch": 56.92,
      "learning_rate": 1.3267543859649124e-05,
      "loss": 0.1577,
      "step": 370
    },
    {
      "epoch": 56.92,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.3214082717895508,
      "eval_runtime": 0.7501,
      "eval_samples_per_second": 61.328,
      "eval_steps_per_second": 2.666,
      "step": 370
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.3992477655410767,
      "eval_runtime": 0.7374,
      "eval_samples_per_second": 62.382,
      "eval_steps_per_second": 2.712,
      "step": 377
    },
    {
      "epoch": 58.46,
      "learning_rate": 1.206140350877193e-05,
      "loss": 0.1474,
      "step": 380
    },
    {
      "epoch": 58.92,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.379963755607605,
      "eval_runtime": 0.7819,
      "eval_samples_per_second": 58.83,
      "eval_steps_per_second": 2.558,
      "step": 383
    },
    {
      "epoch": 60.0,
      "learning_rate": 1.0855263157894737e-05,
      "loss": 0.1267,
      "step": 390
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.2534880638122559,
      "eval_runtime": 0.7531,
      "eval_samples_per_second": 61.078,
      "eval_steps_per_second": 2.656,
      "step": 390
    },
    {
      "epoch": 60.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3200269937515259,
      "eval_runtime": 0.7796,
      "eval_samples_per_second": 59.004,
      "eval_steps_per_second": 2.565,
      "step": 396
    },
    {
      "epoch": 61.54,
      "learning_rate": 9.649122807017545e-06,
      "loss": 0.1171,
      "step": 400
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.373002290725708,
      "eval_runtime": 0.7716,
      "eval_samples_per_second": 59.614,
      "eval_steps_per_second": 2.592,
      "step": 403
    },
    {
      "epoch": 62.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3677655458450317,
      "eval_runtime": 0.7677,
      "eval_samples_per_second": 59.92,
      "eval_steps_per_second": 2.605,
      "step": 409
    },
    {
      "epoch": 63.08,
      "learning_rate": 8.442982456140352e-06,
      "loss": 0.1461,
      "step": 410
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.378843069076538,
      "eval_runtime": 0.7824,
      "eval_samples_per_second": 58.791,
      "eval_steps_per_second": 2.556,
      "step": 416
    },
    {
      "epoch": 64.62,
      "learning_rate": 7.2368421052631575e-06,
      "loss": 0.1124,
      "step": 420
    },
    {
      "epoch": 64.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3944367170333862,
      "eval_runtime": 0.8361,
      "eval_samples_per_second": 55.02,
      "eval_steps_per_second": 2.392,
      "step": 422
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.372430682182312,
      "eval_runtime": 0.7514,
      "eval_samples_per_second": 61.223,
      "eval_steps_per_second": 2.662,
      "step": 429
    },
    {
      "epoch": 66.15,
      "learning_rate": 6.030701754385965e-06,
      "loss": 0.1168,
      "step": 430
    },
    {
      "epoch": 66.92,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.3552623987197876,
      "eval_runtime": 0.8385,
      "eval_samples_per_second": 54.858,
      "eval_steps_per_second": 2.385,
      "step": 435
    },
    {
      "epoch": 67.69,
      "learning_rate": 4.824561403508772e-06,
      "loss": 0.1243,
      "step": 440
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3829400539398193,
      "eval_runtime": 0.8253,
      "eval_samples_per_second": 55.74,
      "eval_steps_per_second": 2.423,
      "step": 442
    },
    {
      "epoch": 68.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.4040273427963257,
      "eval_runtime": 0.7374,
      "eval_samples_per_second": 62.382,
      "eval_steps_per_second": 2.712,
      "step": 448
    },
    {
      "epoch": 69.23,
      "learning_rate": 3.6184210526315788e-06,
      "loss": 0.1375,
      "step": 450
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.4126824140548706,
      "eval_runtime": 0.803,
      "eval_samples_per_second": 57.287,
      "eval_steps_per_second": 2.491,
      "step": 455
    },
    {
      "epoch": 70.77,
      "learning_rate": 2.412280701754386e-06,
      "loss": 0.1017,
      "step": 460
    },
    {
      "epoch": 70.92,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.4069550037384033,
      "eval_runtime": 0.7557,
      "eval_samples_per_second": 60.867,
      "eval_steps_per_second": 2.646,
      "step": 461
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.398921012878418,
      "eval_runtime": 0.7798,
      "eval_samples_per_second": 58.99,
      "eval_steps_per_second": 2.565,
      "step": 468
    },
    {
      "epoch": 72.31,
      "learning_rate": 1.206140350877193e-06,
      "loss": 0.1346,
      "step": 470
    },
    {
      "epoch": 72.92,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3995212316513062,
      "eval_runtime": 0.7372,
      "eval_samples_per_second": 62.399,
      "eval_steps_per_second": 2.713,
      "step": 474
    },
    {
      "epoch": 73.85,
      "learning_rate": 0.0,
      "loss": 0.1382,
      "step": 480
    },
    {
      "epoch": 73.85,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.3988333940505981,
      "eval_runtime": 0.8028,
      "eval_samples_per_second": 57.301,
      "eval_steps_per_second": 2.491,
      "step": 480
    },
    {
      "epoch": 73.85,
      "step": 480,
      "total_flos": 1.9681438145406566e+18,
      "train_loss": 0.36262454763054847,
      "train_runtime": 818.339,
      "train_samples_per_second": 80.065,
      "train_steps_per_second": 0.587
    }
  ],
  "logging_steps": 10,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 80,
  "save_steps": 500,
  "total_flos": 1.9681438145406566e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}