{
  "best_metric": 0.0012111830292269588,
  "best_model_checkpoint": "/leaves/checkpoint-400",
  "epoch": 50.0,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.25,
      "learning_rate": 0.000195,
      "loss": 0.2249,
      "step": 10
    },
    {
      "epoch": 1.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.03227296099066734,
      "eval_runtime": 6.7134,
      "eval_samples_per_second": 1.192,
      "eval_steps_per_second": 0.149,
      "step": 10
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.00019,
      "loss": 0.0177,
      "step": 20
    },
    {
      "epoch": 2.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.011206778697669506,
      "eval_runtime": 4.4578,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 0.224,
      "step": 20
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.00018500000000000002,
      "loss": 0.0086,
      "step": 30
    },
    {
      "epoch": 3.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.00745726190507412,
      "eval_runtime": 6.1312,
      "eval_samples_per_second": 1.305,
      "eval_steps_per_second": 0.163,
      "step": 30
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00018,
      "loss": 0.0063,
      "step": 40
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0058769905008375645,
      "eval_runtime": 4.3511,
      "eval_samples_per_second": 1.839,
      "eval_steps_per_second": 0.23,
      "step": 40
    },
    {
      "epoch": 6.25,
      "learning_rate": 0.000175,
      "loss": 0.0051,
      "step": 50
    },
    {
      "epoch": 6.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.005005857907235622,
      "eval_runtime": 6.0951,
      "eval_samples_per_second": 1.313,
      "eval_steps_per_second": 0.164,
      "step": 50
    },
    {
      "epoch": 7.5,
      "learning_rate": 0.00017,
      "loss": 0.0045,
      "step": 60
    },
    {
      "epoch": 7.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.004425144288688898,
      "eval_runtime": 4.3832,
      "eval_samples_per_second": 1.825,
      "eval_steps_per_second": 0.228,
      "step": 60
    },
    {
      "epoch": 8.75,
      "learning_rate": 0.000165,
      "loss": 0.004,
      "step": 70
    },
    {
      "epoch": 8.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0039963009767234325,
      "eval_runtime": 4.3367,
      "eval_samples_per_second": 1.845,
      "eval_steps_per_second": 0.231,
      "step": 70
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.00016,
      "loss": 0.0036,
      "step": 80
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.003639573697000742,
      "eval_runtime": 4.3969,
      "eval_samples_per_second": 1.819,
      "eval_steps_per_second": 0.227,
      "step": 80
    },
    {
      "epoch": 11.25,
      "learning_rate": 0.000155,
      "loss": 0.0033,
      "step": 90
    },
    {
      "epoch": 11.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0033505873288959265,
      "eval_runtime": 4.3672,
      "eval_samples_per_second": 1.832,
      "eval_steps_per_second": 0.229,
      "step": 90
    },
    {
      "epoch": 12.5,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.0031,
      "step": 100
    },
    {
      "epoch": 12.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.003097910899668932,
      "eval_runtime": 4.4215,
      "eval_samples_per_second": 1.809,
      "eval_steps_per_second": 0.226,
      "step": 100
    },
    {
      "epoch": 13.75,
      "learning_rate": 0.000145,
      "loss": 0.0028,
      "step": 110
    },
    {
      "epoch": 13.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0028831621166318655,
      "eval_runtime": 4.4829,
      "eval_samples_per_second": 1.785,
      "eval_steps_per_second": 0.223,
      "step": 110
    },
    {
      "epoch": 15.0,
      "learning_rate": 0.00014,
      "loss": 0.0026,
      "step": 120
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002691818866878748,
      "eval_runtime": 6.1147,
      "eval_samples_per_second": 1.308,
      "eval_steps_per_second": 0.164,
      "step": 120
    },
    {
      "epoch": 16.25,
      "learning_rate": 0.00013500000000000003,
      "loss": 0.0025,
      "step": 130
    },
    {
      "epoch": 16.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002529613673686981,
      "eval_runtime": 4.923,
      "eval_samples_per_second": 1.625,
      "eval_steps_per_second": 0.203,
      "step": 130
    },
    {
      "epoch": 17.5,
      "learning_rate": 0.00013000000000000002,
      "loss": 0.0023,
      "step": 140
    },
    {
      "epoch": 17.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.00238333223387599,
      "eval_runtime": 4.4274,
      "eval_samples_per_second": 1.807,
      "eval_steps_per_second": 0.226,
      "step": 140
    },
    {
      "epoch": 18.75,
      "learning_rate": 0.000125,
      "loss": 0.0022,
      "step": 150
    },
    {
      "epoch": 18.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002253755461424589,
      "eval_runtime": 5.8958,
      "eval_samples_per_second": 1.357,
      "eval_steps_per_second": 0.17,
      "step": 150
    },
    {
      "epoch": 20.0,
      "learning_rate": 0.00012,
      "loss": 0.0021,
      "step": 160
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002138556446880102,
      "eval_runtime": 4.3757,
      "eval_samples_per_second": 1.828,
      "eval_steps_per_second": 0.229,
      "step": 160
    },
    {
      "epoch": 21.25,
      "learning_rate": 0.00011499999999999999,
      "loss": 0.002,
      "step": 170
    },
    {
      "epoch": 21.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002034764038398862,
      "eval_runtime": 6.0111,
      "eval_samples_per_second": 1.331,
      "eval_steps_per_second": 0.166,
      "step": 170
    },
    {
      "epoch": 22.5,
      "learning_rate": 0.00011000000000000002,
      "loss": 0.0019,
      "step": 180
    },
    {
      "epoch": 22.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0019434543792158365,
      "eval_runtime": 8.0684,
      "eval_samples_per_second": 0.992,
      "eval_steps_per_second": 0.124,
      "step": 180
    },
    {
      "epoch": 23.75,
      "learning_rate": 0.000105,
      "loss": 0.0018,
      "step": 190
    },
    {
      "epoch": 23.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0018578022718429565,
      "eval_runtime": 4.356,
      "eval_samples_per_second": 1.837,
      "eval_steps_per_second": 0.23,
      "step": 190
    },
    {
      "epoch": 25.0,
      "learning_rate": 0.0001,
      "loss": 0.0017,
      "step": 200
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0017840128857642412,
      "eval_runtime": 4.4024,
      "eval_samples_per_second": 1.817,
      "eval_steps_per_second": 0.227,
      "step": 200
    },
    {
      "epoch": 26.25,
      "learning_rate": 9.5e-05,
      "loss": 0.0016,
      "step": 210
    },
    {
      "epoch": 26.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0017146510072052479,
      "eval_runtime": 6.2173,
      "eval_samples_per_second": 1.287,
      "eval_steps_per_second": 0.161,
      "step": 210
    },
    {
      "epoch": 27.5,
      "learning_rate": 9e-05,
      "loss": 0.0016,
      "step": 220
    },
    {
      "epoch": 27.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0016539423959329724,
      "eval_runtime": 4.4503,
      "eval_samples_per_second": 1.798,
      "eval_steps_per_second": 0.225,
      "step": 220
    },
    {
      "epoch": 28.75,
      "learning_rate": 8.5e-05,
      "loss": 0.0015,
      "step": 230
    },
    {
      "epoch": 28.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0015984964556992054,
      "eval_runtime": 5.2504,
      "eval_samples_per_second": 1.524,
      "eval_steps_per_second": 0.19,
      "step": 230
    },
    {
      "epoch": 30.0,
      "learning_rate": 8e-05,
      "loss": 0.0015,
      "step": 240
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0015484185423702002,
      "eval_runtime": 4.3678,
      "eval_samples_per_second": 1.832,
      "eval_steps_per_second": 0.229,
      "step": 240
    },
    {
      "epoch": 31.25,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0014,
      "step": 250
    },
    {
      "epoch": 31.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0015031290240585804,
      "eval_runtime": 4.7164,
      "eval_samples_per_second": 1.696,
      "eval_steps_per_second": 0.212,
      "step": 250
    },
    {
      "epoch": 32.5,
      "learning_rate": 7e-05,
      "loss": 0.0014,
      "step": 260
    },
    {
      "epoch": 32.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0014625097392126918,
      "eval_runtime": 5.6004,
      "eval_samples_per_second": 1.428,
      "eval_steps_per_second": 0.179,
      "step": 260
    },
    {
      "epoch": 33.75,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.0013,
      "step": 270
    },
    {
      "epoch": 33.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0014250578824430704,
      "eval_runtime": 4.4145,
      "eval_samples_per_second": 1.812,
      "eval_steps_per_second": 0.227,
      "step": 270
    },
    {
      "epoch": 35.0,
      "learning_rate": 6e-05,
      "loss": 0.0013,
      "step": 280
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0013918605400249362,
      "eval_runtime": 6.1306,
      "eval_samples_per_second": 1.305,
      "eval_steps_per_second": 0.163,
      "step": 280
    },
    {
      "epoch": 36.25,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.0013,
      "step": 290
    },
    {
      "epoch": 36.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0013616828946396708,
      "eval_runtime": 4.4127,
      "eval_samples_per_second": 1.813,
      "eval_steps_per_second": 0.227,
      "step": 290
    },
    {
      "epoch": 37.5,
      "learning_rate": 5e-05,
      "loss": 0.0013,
      "step": 300
    },
    {
      "epoch": 37.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0013343169121071696,
      "eval_runtime": 4.3912,
      "eval_samples_per_second": 1.822,
      "eval_steps_per_second": 0.228,
      "step": 300
    },
    {
      "epoch": 38.75,
      "learning_rate": 4.5e-05,
      "loss": 0.0012,
      "step": 310
    },
    {
      "epoch": 38.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0013108787825331092,
      "eval_runtime": 4.8701,
      "eval_samples_per_second": 1.643,
      "eval_steps_per_second": 0.205,
      "step": 310
    },
    {
      "epoch": 40.0,
      "learning_rate": 4e-05,
      "loss": 0.0012,
      "step": 320
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.001289389911107719,
      "eval_runtime": 5.9706,
      "eval_samples_per_second": 1.34,
      "eval_steps_per_second": 0.167,
      "step": 320
    },
    {
      "epoch": 41.25,
      "learning_rate": 3.5e-05,
      "loss": 0.0012,
      "step": 330
    },
    {
      "epoch": 41.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012705642729997635,
      "eval_runtime": 4.9479,
      "eval_samples_per_second": 1.617,
      "eval_steps_per_second": 0.202,
      "step": 330
    },
    {
      "epoch": 42.5,
      "learning_rate": 3e-05,
      "loss": 0.0012,
      "step": 340
    },
    {
      "epoch": 42.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.00125490827485919,
      "eval_runtime": 4.9186,
      "eval_samples_per_second": 1.626,
      "eval_steps_per_second": 0.203,
      "step": 340
    },
    {
      "epoch": 43.75,
      "learning_rate": 2.5e-05,
      "loss": 0.0012,
      "step": 350
    },
    {
      "epoch": 43.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012415139935910702,
      "eval_runtime": 4.3622,
      "eval_samples_per_second": 1.834,
      "eval_steps_per_second": 0.229,
      "step": 350
    },
    {
      "epoch": 45.0,
      "learning_rate": 2e-05,
      "loss": 0.0012,
      "step": 360
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012306944699957967,
      "eval_runtime": 6.2742,
      "eval_samples_per_second": 1.275,
      "eval_steps_per_second": 0.159,
      "step": 360
    },
    {
      "epoch": 46.25,
      "learning_rate": 1.5e-05,
      "loss": 0.0011,
      "step": 370
    },
    {
      "epoch": 46.25,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012223005760461092,
      "eval_runtime": 5.7273,
      "eval_samples_per_second": 1.397,
      "eval_steps_per_second": 0.175,
      "step": 370
    },
    {
      "epoch": 47.5,
      "learning_rate": 1e-05,
      "loss": 0.0012,
      "step": 380
    },
    {
      "epoch": 47.5,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012162283528596163,
      "eval_runtime": 4.7043,
      "eval_samples_per_second": 1.701,
      "eval_steps_per_second": 0.213,
      "step": 380
    },
    {
      "epoch": 48.75,
      "learning_rate": 5e-06,
      "loss": 0.0011,
      "step": 390
    },
    {
      "epoch": 48.75,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012125077191740274,
      "eval_runtime": 4.4338,
      "eval_samples_per_second": 1.804,
      "eval_steps_per_second": 0.226,
      "step": 390
    },
    {
      "epoch": 50.0,
      "learning_rate": 0.0,
      "loss": 0.0011,
      "step": 400
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012111830292269588,
      "eval_runtime": 5.2993,
      "eval_samples_per_second": 1.51,
      "eval_steps_per_second": 0.189,
      "step": 400
    },
    {
      "epoch": 50.0,
      "step": 400,
      "total_flos": 2.789711626125312e+17,
      "train_loss": 0.008226247422862798,
      "train_runtime": 8928.1619,
      "train_samples_per_second": 0.403,
      "train_steps_per_second": 0.045
    }
  ],
  "max_steps": 400,
  "num_train_epochs": 50,
  "total_flos": 2.789711626125312e+17,
  "trial_name": null,
  "trial_params": null
}