{
  "best_metric": 2.8627383708953857,
  "best_model_checkpoint": "cdetr-mist1-brain-gt-tumors-8ah-6l/checkpoint-2645",
  "epoch": 25.0,
  "eval_steps": 500,
  "global_step": 2875,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9.624347826086956e-06,
      "loss": 5.4149,
      "step": 115
    },
    {
      "epoch": 1.0,
      "eval_loss": 4.397365093231201,
      "eval_runtime": 5.8932,
      "eval_samples_per_second": 6.787,
      "eval_steps_per_second": 0.848,
      "step": 115
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.231304347826087e-06,
      "loss": 3.9453,
      "step": 230
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.651952028274536,
      "eval_runtime": 5.9289,
      "eval_samples_per_second": 6.747,
      "eval_steps_per_second": 0.843,
      "step": 230
    },
    {
      "epoch": 3.0,
      "learning_rate": 8.831304347826088e-06,
      "loss": 3.7269,
      "step": 345
    },
    {
      "epoch": 3.0,
      "eval_loss": 3.7602438926696777,
      "eval_runtime": 5.8799,
      "eval_samples_per_second": 6.803,
      "eval_steps_per_second": 0.85,
      "step": 345
    },
    {
      "epoch": 4.0,
      "learning_rate": 8.434782608695653e-06,
      "loss": 3.5898,
      "step": 460
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.5670828819274902,
      "eval_runtime": 5.9645,
      "eval_samples_per_second": 6.706,
      "eval_steps_per_second": 0.838,
      "step": 460
    },
    {
      "epoch": 5.0,
      "learning_rate": 8.034782608695652e-06,
      "loss": 3.486,
      "step": 575
    },
    {
      "epoch": 5.0,
      "eval_loss": 3.4911675453186035,
      "eval_runtime": 5.9241,
      "eval_samples_per_second": 6.752,
      "eval_steps_per_second": 0.844,
      "step": 575
    },
    {
      "epoch": 6.0,
      "learning_rate": 7.634782608695653e-06,
      "loss": 3.4073,
      "step": 690
    },
    {
      "epoch": 6.0,
      "eval_loss": 3.409463405609131,
      "eval_runtime": 5.8302,
      "eval_samples_per_second": 6.861,
      "eval_steps_per_second": 0.858,
      "step": 690
    },
    {
      "epoch": 7.0,
      "learning_rate": 7.234782608695653e-06,
      "loss": 3.4181,
      "step": 805
    },
    {
      "epoch": 7.0,
      "eval_loss": 3.3183083534240723,
      "eval_runtime": 5.8894,
      "eval_samples_per_second": 6.792,
      "eval_steps_per_second": 0.849,
      "step": 805
    },
    {
      "epoch": 8.0,
      "learning_rate": 6.834782608695653e-06,
      "loss": 3.3603,
      "step": 920
    },
    {
      "epoch": 8.0,
      "eval_loss": 3.1111273765563965,
      "eval_runtime": 5.9175,
      "eval_samples_per_second": 6.76,
      "eval_steps_per_second": 0.845,
      "step": 920
    },
    {
      "epoch": 9.0,
      "learning_rate": 6.434782608695652e-06,
      "loss": 3.2777,
      "step": 1035
    },
    {
      "epoch": 9.0,
      "eval_loss": 3.1992499828338623,
      "eval_runtime": 5.9275,
      "eval_samples_per_second": 6.748,
      "eval_steps_per_second": 0.844,
      "step": 1035
    },
    {
      "epoch": 10.0,
      "learning_rate": 6.034782608695653e-06,
      "loss": 3.2851,
      "step": 1150
    },
    {
      "epoch": 10.0,
      "eval_loss": 3.3997130393981934,
      "eval_runtime": 5.8939,
      "eval_samples_per_second": 6.787,
      "eval_steps_per_second": 0.848,
      "step": 1150
    },
    {
      "epoch": 11.0,
      "learning_rate": 5.634782608695652e-06,
      "loss": 3.266,
      "step": 1265
    },
    {
      "epoch": 11.0,
      "eval_loss": 3.286081314086914,
      "eval_runtime": 6.1287,
      "eval_samples_per_second": 6.527,
      "eval_steps_per_second": 0.816,
      "step": 1265
    },
    {
      "epoch": 12.0,
      "learning_rate": 5.234782608695652e-06,
      "loss": 3.2803,
      "step": 1380
    },
    {
      "epoch": 12.0,
      "eval_loss": 3.18131947517395,
      "eval_runtime": 6.1123,
      "eval_samples_per_second": 6.544,
      "eval_steps_per_second": 0.818,
      "step": 1380
    },
    {
      "epoch": 13.0,
      "learning_rate": 4.834782608695652e-06,
      "loss": 3.1733,
      "step": 1495
    },
    {
      "epoch": 13.0,
      "eval_loss": 2.9837594032287598,
      "eval_runtime": 5.925,
      "eval_samples_per_second": 6.751,
      "eval_steps_per_second": 0.844,
      "step": 1495
    },
    {
      "epoch": 14.0,
      "learning_rate": 4.434782608695653e-06,
      "loss": 3.2094,
      "step": 1610
    },
    {
      "epoch": 14.0,
      "eval_loss": 3.1175174713134766,
      "eval_runtime": 5.9137,
      "eval_samples_per_second": 6.764,
      "eval_steps_per_second": 0.845,
      "step": 1610
    },
    {
      "epoch": 15.0,
      "learning_rate": 4.034782608695652e-06,
      "loss": 3.1718,
      "step": 1725
    },
    {
      "epoch": 15.0,
      "eval_loss": 3.0063586235046387,
      "eval_runtime": 5.9077,
      "eval_samples_per_second": 6.771,
      "eval_steps_per_second": 0.846,
      "step": 1725
    },
    {
      "epoch": 16.0,
      "learning_rate": 3.6347826086956524e-06,
      "loss": 3.1303,
      "step": 1840
    },
    {
      "epoch": 16.0,
      "eval_loss": 3.086920738220215,
      "eval_runtime": 5.9305,
      "eval_samples_per_second": 6.745,
      "eval_steps_per_second": 0.843,
      "step": 1840
    },
    {
      "epoch": 17.0,
      "learning_rate": 3.2347826086956524e-06,
      "loss": 3.0897,
      "step": 1955
    },
    {
      "epoch": 17.0,
      "eval_loss": 3.0305533409118652,
      "eval_runtime": 6.0975,
      "eval_samples_per_second": 6.56,
      "eval_steps_per_second": 0.82,
      "step": 1955
    },
    {
      "epoch": 18.0,
      "learning_rate": 2.8347826086956525e-06,
      "loss": 3.0233,
      "step": 2070
    },
    {
      "epoch": 18.0,
      "eval_loss": 2.9478588104248047,
      "eval_runtime": 5.9801,
      "eval_samples_per_second": 6.689,
      "eval_steps_per_second": 0.836,
      "step": 2070
    },
    {
      "epoch": 19.0,
      "learning_rate": 2.4347826086956525e-06,
      "loss": 3.0156,
      "step": 2185
    },
    {
      "epoch": 19.0,
      "eval_loss": 2.914536952972412,
      "eval_runtime": 6.0155,
      "eval_samples_per_second": 6.649,
      "eval_steps_per_second": 0.831,
      "step": 2185
    },
    {
      "epoch": 20.0,
      "learning_rate": 2.0347826086956525e-06,
      "loss": 3.0277,
      "step": 2300
    },
    {
      "epoch": 20.0,
      "eval_loss": 2.891925811767578,
      "eval_runtime": 6.058,
      "eval_samples_per_second": 6.603,
      "eval_steps_per_second": 0.825,
      "step": 2300
    },
    {
      "epoch": 21.0,
      "learning_rate": 1.6382608695652176e-06,
      "loss": 3.0847,
      "step": 2415
    },
    {
      "epoch": 21.0,
      "eval_loss": 2.9321396350860596,
      "eval_runtime": 6.1412,
      "eval_samples_per_second": 6.513,
      "eval_steps_per_second": 0.814,
      "step": 2415
    },
    {
      "epoch": 22.0,
      "learning_rate": 1.2382608695652176e-06,
      "loss": 3.0333,
      "step": 2530
    },
    {
      "epoch": 22.0,
      "eval_loss": 2.91283917427063,
      "eval_runtime": 6.139,
      "eval_samples_per_second": 6.516,
      "eval_steps_per_second": 0.814,
      "step": 2530
    },
    {
      "epoch": 23.0,
      "learning_rate": 8.382608695652175e-07,
      "loss": 3.0126,
      "step": 2645
    },
    {
      "epoch": 23.0,
      "eval_loss": 2.8627383708953857,
      "eval_runtime": 6.0565,
      "eval_samples_per_second": 6.604,
      "eval_steps_per_second": 0.826,
      "step": 2645
    },
    {
      "epoch": 24.0,
      "learning_rate": 4.382608695652174e-07,
      "loss": 2.9701,
      "step": 2760
    },
    {
      "epoch": 24.0,
      "eval_loss": 2.868567705154419,
      "eval_runtime": 6.0333,
      "eval_samples_per_second": 6.63,
      "eval_steps_per_second": 0.829,
      "step": 2760
    },
    {
      "epoch": 25.0,
      "learning_rate": 3.82608695652174e-08,
      "loss": 2.9964,
      "step": 2875
    },
    {
      "epoch": 25.0,
      "eval_loss": 2.8917393684387207,
      "eval_runtime": 6.0377,
      "eval_samples_per_second": 6.625,
      "eval_steps_per_second": 0.828,
      "step": 2875
    },
    {
      "epoch": 25.0,
      "step": 2875,
      "total_flos": 5.73883103808e+18,
      "train_loss": 3.3358317977241847,
      "train_runtime": 2483.4302,
      "train_samples_per_second": 4.631,
      "train_steps_per_second": 1.158
    }
  ],
  "logging_steps": 500,
  "max_steps": 2875,
  "num_train_epochs": 25,
  "save_steps": 500,
  "total_flos": 5.73883103808e+18,
  "trial_name": null,
  "trial_params": null
}