{
  "best_metric": 0.035480156540870667,
  "best_model_checkpoint": "./FL-birds/checkpoint-500",
  "epoch": 3.0120481927710845,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 0.00019698795180722893,
      "loss": 1.6629,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019397590361445782,
      "loss": 1.0702,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019096385542168677,
      "loss": 0.6424,
      "step": 30
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00018795180722891569,
      "loss": 0.4989,
      "step": 40
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018493975903614458,
      "loss": 0.352,
      "step": 50
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001819277108433735,
      "loss": 0.3337,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017891566265060242,
      "loss": 0.2032,
      "step": 70
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017590361445783134,
      "loss": 0.1717,
      "step": 80
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00017289156626506026,
      "loss": 0.171,
      "step": 90
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00016987951807228917,
      "loss": 0.1569,
      "step": 100
    },
    {
      "epoch": 0.6,
      "eval_accuracy": 0.9512195121951219,
      "eval_loss": 0.18708132207393646,
      "eval_runtime": 4.0708,
      "eval_samples_per_second": 60.431,
      "eval_steps_per_second": 7.615,
      "step": 100
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001668674698795181,
      "loss": 0.1336,
      "step": 110
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016385542168674699,
      "loss": 0.1721,
      "step": 120
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001608433734939759,
      "loss": 0.0869,
      "step": 130
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00015783132530120482,
      "loss": 0.1885,
      "step": 140
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00015481927710843374,
      "loss": 0.0992,
      "step": 150
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015180722891566266,
      "loss": 0.0944,
      "step": 160
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014879518072289158,
      "loss": 0.072,
      "step": 170
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014578313253012047,
      "loss": 0.0655,
      "step": 180
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001427710843373494,
      "loss": 0.0491,
      "step": 190
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00013975903614457834,
      "loss": 0.0569,
      "step": 200
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.9796747967479674,
      "eval_loss": 0.10953661054372787,
      "eval_runtime": 4.231,
      "eval_samples_per_second": 58.142,
      "eval_steps_per_second": 7.327,
      "step": 200
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00013674698795180723,
      "loss": 0.061,
      "step": 210
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00013373493975903615,
      "loss": 0.0531,
      "step": 220
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00013072289156626507,
      "loss": 0.0277,
      "step": 230
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012771084337349396,
      "loss": 0.0236,
      "step": 240
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012469879518072288,
      "loss": 0.0215,
      "step": 250
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012168674698795181,
      "loss": 0.026,
      "step": 260
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00011867469879518073,
      "loss": 0.0672,
      "step": 270
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011566265060240964,
      "loss": 0.0278,
      "step": 280
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011265060240963856,
      "loss": 0.0186,
      "step": 290
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010963855421686749,
      "loss": 0.0176,
      "step": 300
    },
    {
      "epoch": 1.81,
      "eval_accuracy": 0.9878048780487805,
      "eval_loss": 0.049860063940286636,
      "eval_runtime": 4.1605,
      "eval_samples_per_second": 59.127,
      "eval_steps_per_second": 7.451,
      "step": 300
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00010662650602409638,
      "loss": 0.0166,
      "step": 310
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0001036144578313253,
      "loss": 0.043,
      "step": 320
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010060240963855423,
      "loss": 0.0464,
      "step": 330
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.759036144578314e-05,
      "loss": 0.0149,
      "step": 340
    },
    {
      "epoch": 2.11,
      "learning_rate": 9.457831325301205e-05,
      "loss": 0.0379,
      "step": 350
    },
    {
      "epoch": 2.17,
      "learning_rate": 9.156626506024096e-05,
      "loss": 0.0134,
      "step": 360
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.855421686746988e-05,
      "loss": 0.0126,
      "step": 370
    },
    {
      "epoch": 2.29,
      "learning_rate": 8.55421686746988e-05,
      "loss": 0.0123,
      "step": 380
    },
    {
      "epoch": 2.35,
      "learning_rate": 8.253012048192772e-05,
      "loss": 0.012,
      "step": 390
    },
    {
      "epoch": 2.41,
      "learning_rate": 7.951807228915663e-05,
      "loss": 0.0113,
      "step": 400
    },
    {
      "epoch": 2.41,
      "eval_accuracy": 0.983739837398374,
      "eval_loss": 0.04573996365070343,
      "eval_runtime": 3.9545,
      "eval_samples_per_second": 62.208,
      "eval_steps_per_second": 7.839,
      "step": 400
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.650602409638555e-05,
      "loss": 0.0113,
      "step": 410
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.349397590361447e-05,
      "loss": 0.0106,
      "step": 420
    },
    {
      "epoch": 2.59,
      "learning_rate": 7.048192771084337e-05,
      "loss": 0.0105,
      "step": 430
    },
    {
      "epoch": 2.65,
      "learning_rate": 6.746987951807229e-05,
      "loss": 0.0102,
      "step": 440
    },
    {
      "epoch": 2.71,
      "learning_rate": 6.445783132530121e-05,
      "loss": 0.0101,
      "step": 450
    },
    {
      "epoch": 2.77,
      "learning_rate": 6.144578313253012e-05,
      "loss": 0.0098,
      "step": 460
    },
    {
      "epoch": 2.83,
      "learning_rate": 5.843373493975904e-05,
      "loss": 0.0097,
      "step": 470
    },
    {
      "epoch": 2.89,
      "learning_rate": 5.5421686746987955e-05,
      "loss": 0.0094,
      "step": 480
    },
    {
      "epoch": 2.95,
      "learning_rate": 5.240963855421687e-05,
      "loss": 0.0092,
      "step": 490
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.9397590361445786e-05,
      "loss": 0.009,
      "step": 500
    },
    {
      "epoch": 3.01,
      "eval_accuracy": 0.9959349593495935,
      "eval_loss": 0.035480156540870667,
      "eval_runtime": 3.9575,
      "eval_samples_per_second": 62.16,
      "eval_steps_per_second": 7.833,
      "step": 500
    }
  ],
  "max_steps": 664,
  "num_train_epochs": 4,
  "total_flos": 6.190337528168694e+17,
  "trial_name": null,
  "trial_params": null
}