{
  "best_metric": 0.07713755965232849,
  "best_model_checkpoint": "/nasty/data/tpid/vizwiz/ViTGPT2_VW/checkpoint-14000",
  "epoch": 0.04404233526800956,
  "global_step": 14645,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 1.9863839666769097e-05,
      "loss": 0.1256,
      "step": 1000
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.09279213845729828,
      "eval_runtime": 1047.3083,
      "eval_samples_per_second": 37.0,
      "eval_steps_per_second": 9.25,
      "step": 1000
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9727269623408106e-05,
      "loss": 0.0947,
      "step": 2000
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.0897236242890358,
      "eval_runtime": 864.4123,
      "eval_samples_per_second": 44.828,
      "eval_steps_per_second": 11.208,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9590699580047118e-05,
      "loss": 0.0889,
      "step": 3000
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.08590871840715408,
      "eval_runtime": 857.9482,
      "eval_samples_per_second": 45.166,
      "eval_steps_per_second": 11.292,
      "step": 3000
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.945412953668613e-05,
      "loss": 0.0888,
      "step": 4000
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.08419814705848694,
      "eval_runtime": 866.4231,
      "eval_samples_per_second": 44.724,
      "eval_steps_per_second": 11.182,
      "step": 4000
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.931755949332514e-05,
      "loss": 0.0866,
      "step": 5000
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.08309577405452728,
      "eval_runtime": 864.2485,
      "eval_samples_per_second": 44.837,
      "eval_steps_per_second": 11.21,
      "step": 5000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.918098944996415e-05,
      "loss": 0.0852,
      "step": 6000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.08189540356397629,
      "eval_runtime": 865.5776,
      "eval_samples_per_second": 44.768,
      "eval_steps_per_second": 11.193,
      "step": 6000
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.9044419406603162e-05,
      "loss": 0.0833,
      "step": 7000
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.08095835894346237,
      "eval_runtime": 864.7144,
      "eval_samples_per_second": 44.812,
      "eval_steps_per_second": 11.204,
      "step": 7000
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8907849363242174e-05,
      "loss": 0.0835,
      "step": 8000
    },
    {
      "epoch": 0.27,
      "eval_loss": 0.08023638278245926,
      "eval_runtime": 866.728,
      "eval_samples_per_second": 44.708,
      "eval_steps_per_second": 11.178,
      "step": 8000
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8771279319881186e-05,
      "loss": 0.081,
      "step": 9000
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.07963699847459793,
      "eval_runtime": 865.0337,
      "eval_samples_per_second": 44.796,
      "eval_steps_per_second": 11.2,
      "step": 9000
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.86347092765202e-05,
      "loss": 0.0803,
      "step": 10000
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.07886938005685806,
      "eval_runtime": 864.2901,
      "eval_samples_per_second": 44.834,
      "eval_steps_per_second": 11.209,
      "step": 10000
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.849813923315921e-05,
      "loss": 0.0814,
      "step": 11000
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.07851768285036087,
      "eval_runtime": 863.0619,
      "eval_samples_per_second": 44.898,
      "eval_steps_per_second": 11.225,
      "step": 11000
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.836156918979822e-05,
      "loss": 0.0799,
      "step": 12000
    },
    {
      "epoch": 0.41,
      "eval_loss": 0.0779803916811943,
      "eval_runtime": 862.8886,
      "eval_samples_per_second": 44.907,
      "eval_steps_per_second": 11.227,
      "step": 12000
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.822513571648059e-05,
      "loss": 0.0786,
      "step": 13000
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.07762513309717178,
      "eval_runtime": 861.7094,
      "eval_samples_per_second": 44.969,
      "eval_steps_per_second": 11.243,
      "step": 13000
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.8088565673119603e-05,
      "loss": 0.0796,
      "step": 14000
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.07713755965232849,
      "eval_runtime": 866.4253,
      "eval_samples_per_second": 44.724,
      "eval_steps_per_second": 11.182,
      "step": 14000
    },
    {
      "epoch": 0.04,
      "step": 14645,
      "total_flos": 7.904602466056929e+18,
      "train_loss": 0.0033295607640092873,
      "train_runtime": 303.4327,
      "train_samples_per_second": 386.099,
      "train_steps_per_second": 48.264
    }
  ],
  "max_steps": 14645,
  "num_train_epochs": 1,
  "total_flos": 7.904602466056929e+18,
  "trial_name": null,
  "trial_params": null
}