{
  "best_metric": 0.7191632129465166,
  "best_model_checkpoint": "Distilled-Melanoma-Classification-ResNet/checkpoint-63325",
  "epoch": 29.0,
  "eval_steps": 500,
  "global_step": 73457,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.8333333333333334e-05,
      "loss": 1.8178,
      "step": 2533
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5569370436155516,
      "eval_loss": 1.4635199308395386,
      "eval_runtime": 443.0799,
      "eval_samples_per_second": 11.436,
      "eval_steps_per_second": 1.431,
      "step": 2533
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.666666666666667e-05,
      "loss": 1.5613,
      "step": 5066
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6009473060982831,
      "eval_loss": 1.3275712728500366,
      "eval_runtime": 450.6621,
      "eval_samples_per_second": 11.243,
      "eval_steps_per_second": 1.407,
      "step": 5066
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.5e-05,
      "loss": 1.4093,
      "step": 7599
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6279850009867772,
      "eval_loss": 1.2198262214660645,
      "eval_runtime": 455.8599,
      "eval_samples_per_second": 11.115,
      "eval_steps_per_second": 1.391,
      "step": 7599
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 1.2938,
      "step": 10132
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6360765739096113,
      "eval_loss": 1.2061253786087036,
      "eval_runtime": 457.5927,
      "eval_samples_per_second": 11.073,
      "eval_steps_per_second": 1.386,
      "step": 10132
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.2145,
      "step": 12665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.640615748963884,
      "eval_loss": 1.102392554283142,
      "eval_runtime": 461.4922,
      "eval_samples_per_second": 10.98,
      "eval_steps_per_second": 1.374,
      "step": 12665
    },
    {
      "epoch": 6.0,
      "learning_rate": 4e-05,
      "loss": 1.143,
      "step": 15198
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6682455101638051,
      "eval_loss": 1.0069061517715454,
      "eval_runtime": 464.3259,
      "eval_samples_per_second": 10.913,
      "eval_steps_per_second": 1.365,
      "step": 15198
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.8333333333333334e-05,
      "loss": 1.065,
      "step": 17731
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6714031971580817,
      "eval_loss": 1.0104970932006836,
      "eval_runtime": 464.529,
      "eval_samples_per_second": 10.908,
      "eval_steps_per_second": 1.365,
      "step": 17731
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 0.9922,
      "step": 20264
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6741661732780738,
      "eval_loss": 1.0025968551635742,
      "eval_runtime": 465.5511,
      "eval_samples_per_second": 10.884,
      "eval_steps_per_second": 1.362,
      "step": 20264
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.5e-05,
      "loss": 0.8981,
      "step": 22797
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6893625419380304,
      "eval_loss": 0.9656845331192017,
      "eval_runtime": 467.5026,
      "eval_samples_per_second": 10.838,
      "eval_steps_per_second": 1.356,
      "step": 22797
    },
    {
      "epoch": 10.0,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.8036,
      "step": 25330
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6883757647523189,
      "eval_loss": 0.9617614150047302,
      "eval_runtime": 460.7588,
      "eval_samples_per_second": 10.997,
      "eval_steps_per_second": 1.376,
      "step": 25330
    },
    {
      "epoch": 11.0,
      "learning_rate": 3.1666666666666666e-05,
      "loss": 0.6933,
      "step": 27863
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6964673376751529,
      "eval_loss": 0.9330915212631226,
      "eval_runtime": 457.9554,
      "eval_samples_per_second": 11.064,
      "eval_steps_per_second": 1.384,
      "step": 27863
    },
    {
      "epoch": 12.0,
      "learning_rate": 3e-05,
      "loss": 0.5867,
      "step": 30396
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6954805604894415,
      "eval_loss": 0.9513295292854309,
      "eval_runtime": 460.1198,
      "eval_samples_per_second": 11.012,
      "eval_steps_per_second": 1.378,
      "step": 30396
    },
    {
      "epoch": 13.0,
      "learning_rate": 2.8333333333333335e-05,
      "loss": 0.4942,
      "step": 32929
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.7051509769094139,
      "eval_loss": 0.9170295596122742,
      "eval_runtime": 461.0234,
      "eval_samples_per_second": 10.991,
      "eval_steps_per_second": 1.375,
      "step": 32929
    },
    {
      "epoch": 14.0,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.4175,
      "step": 35462
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7053483323465561,
      "eval_loss": 0.9810006618499756,
      "eval_runtime": 462.5023,
      "eval_samples_per_second": 10.956,
      "eval_steps_per_second": 1.371,
      "step": 35462
    },
    {
      "epoch": 15.0,
      "learning_rate": 2.5e-05,
      "loss": 0.3539,
      "step": 37995
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7049536214722716,
      "eval_loss": 0.9560312032699585,
      "eval_runtime": 460.5217,
      "eval_samples_per_second": 11.003,
      "eval_steps_per_second": 1.377,
      "step": 37995
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 0.3072,
      "step": 40528
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7067298204065522,
      "eval_loss": 0.9344322085380554,
      "eval_runtime": 460.2866,
      "eval_samples_per_second": 11.008,
      "eval_steps_per_second": 1.377,
      "step": 40528
    },
    {
      "epoch": 17.0,
      "learning_rate": 2.1666666666666667e-05,
      "loss": 0.2729,
      "step": 43061
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.7128478389579633,
      "eval_loss": 0.8949710726737976,
      "eval_runtime": 464.9931,
      "eval_samples_per_second": 10.897,
      "eval_steps_per_second": 1.363,
      "step": 43061
    },
    {
      "epoch": 18.0,
      "learning_rate": 2e-05,
      "loss": 0.2416,
      "step": 45594
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.7049536214722716,
      "eval_loss": 0.8919546604156494,
      "eval_runtime": 462.6468,
      "eval_samples_per_second": 10.952,
      "eval_steps_per_second": 1.37,
      "step": 45594
    },
    {
      "epoch": 19.0,
      "learning_rate": 1.8333333333333333e-05,
      "loss": 0.2173,
      "step": 48127
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.7108742845865403,
      "eval_loss": 0.879001259803772,
      "eval_runtime": 461.3852,
      "eval_samples_per_second": 10.982,
      "eval_steps_per_second": 1.374,
      "step": 48127
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1973,
      "step": 50660
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7162028813893823,
      "eval_loss": 0.8711328506469727,
      "eval_runtime": 460.7389,
      "eval_samples_per_second": 10.998,
      "eval_steps_per_second": 1.376,
      "step": 50660
    },
    {
      "epoch": 21.0,
      "learning_rate": 1.5e-05,
      "loss": 0.1806,
      "step": 53193
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.7061377540951254,
      "eval_loss": 0.8637410998344421,
      "eval_runtime": 462.8193,
      "eval_samples_per_second": 10.948,
      "eval_steps_per_second": 1.37,
      "step": 53193
    },
    {
      "epoch": 22.0,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.1662,
      "step": 55726
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.717387014012236,
      "eval_loss": 0.8513094186782837,
      "eval_runtime": 457.9446,
      "eval_samples_per_second": 11.065,
      "eval_steps_per_second": 1.384,
      "step": 55726
    },
    {
      "epoch": 23.0,
      "learning_rate": 1.1666666666666668e-05,
      "loss": 0.1549,
      "step": 58259
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.7156108150779554,
      "eval_loss": 0.8369139432907104,
      "eval_runtime": 459.9692,
      "eval_samples_per_second": 11.016,
      "eval_steps_per_second": 1.378,
      "step": 58259
    },
    {
      "epoch": 24.0,
      "learning_rate": 1e-05,
      "loss": 0.1443,
      "step": 60792
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.7138346161436747,
      "eval_loss": 0.8337984681129456,
      "eval_runtime": 455.8047,
      "eval_samples_per_second": 11.117,
      "eval_steps_per_second": 1.391,
      "step": 60792
    },
    {
      "epoch": 25.0,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.1347,
      "step": 63325
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.7191632129465166,
      "eval_loss": 0.8258342146873474,
      "eval_runtime": 460.9514,
      "eval_samples_per_second": 10.992,
      "eval_steps_per_second": 1.375,
      "step": 63325
    },
    {
      "epoch": 26.0,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.1271,
      "step": 65858
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.717387014012236,
      "eval_loss": 0.8176981210708618,
      "eval_runtime": 460.4508,
      "eval_samples_per_second": 11.004,
      "eval_steps_per_second": 1.377,
      "step": 65858
    },
    {
      "epoch": 27.0,
      "learning_rate": 5e-06,
      "loss": 0.1198,
      "step": 68391
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.7165975922636668,
      "eval_loss": 0.8105741739273071,
      "eval_runtime": 459.2895,
      "eval_samples_per_second": 11.032,
      "eval_steps_per_second": 1.38,
      "step": 68391
    },
    {
      "epoch": 28.0,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.115,
      "step": 70924
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.7171896585750938,
      "eval_loss": 0.8162857890129089,
      "eval_runtime": 446.9777,
      "eval_samples_per_second": 11.336,
      "eval_steps_per_second": 1.418,
      "step": 70924
    },
    {
      "epoch": 29.0,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.1101,
      "step": 73457
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.717387014012236,
      "eval_loss": 0.8069340586662292,
      "eval_runtime": 477.0626,
      "eval_samples_per_second": 10.621,
      "eval_steps_per_second": 1.329,
      "step": 73457
    }
  ],
  "logging_steps": 500,
  "max_steps": 75990,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 1.2485637303036936e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}