trainer_state.json

{
  "best_metric": 3.0846824645996094,
  "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_data_aug_rte_256/checkpoint-2272",
  "epoch": 7.0,
  "global_step": 7952,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 0.2703,
      "step": 1136
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4657039711191336,
      "eval_loss": 3.27677845954895,
      "eval_runtime": 0.4574,
      "eval_samples_per_second": 605.59,
      "eval_steps_per_second": 6.559,
      "step": 1136
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.0555,
      "step": 2272
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.48736462093862815,
      "eval_loss": 3.0846824645996094,
      "eval_runtime": 0.4539,
      "eval_samples_per_second": 610.31,
      "eval_steps_per_second": 6.61,
      "step": 2272
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.0253,
      "step": 3408
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5018050541516246,
      "eval_loss": 5.496804714202881,
      "eval_runtime": 0.4523,
      "eval_samples_per_second": 612.364,
      "eval_steps_per_second": 6.632,
      "step": 3408
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.0149,
      "step": 4544
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4981949458483754,
      "eval_loss": 5.601980686187744,
      "eval_runtime": 0.451,
      "eval_samples_per_second": 614.234,
      "eval_steps_per_second": 6.652,
      "step": 4544
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 0.0104,
      "step": 5680
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5090252707581228,
      "eval_loss": 6.668324947357178,
      "eval_runtime": 0.452,
      "eval_samples_per_second": 612.893,
      "eval_steps_per_second": 6.638,
      "step": 5680
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0082,
      "step": 6816
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5090252707581228,
      "eval_loss": 8.221952438354492,
      "eval_runtime": 0.4579,
      "eval_samples_per_second": 604.94,
      "eval_steps_per_second": 6.552,
      "step": 6816
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 0.0062,
      "step": 7952
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5054151624548736,
      "eval_loss": 8.21786880493164,
      "eval_runtime": 0.4496,
      "eval_samples_per_second": 616.076,
      "eval_steps_per_second": 6.672,
      "step": 7952
    },
    {
      "epoch": 7.0,
      "step": 7952,
      "total_flos": 2.6809652670889984e+16,
      "train_loss": 0.05584204784581359,
      "train_runtime": 5012.5504,
      "train_samples_per_second": 1450.34,
      "train_steps_per_second": 11.332
    }
  ],
  "max_steps": 56800,
  "num_train_epochs": 50,
  "total_flos": 2.6809652670889984e+16,
  "trial_name": null,
  "trial_params": null
}