trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 1605,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.5895,
      "step": 535
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6146270632743835,
      "eval_matthews_correlation": 0.16988650877044353,
      "eval_runtime": 7.5781,
      "eval_samples_per_second": 137.634,
      "eval_steps_per_second": 17.287,
      "step": 535
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.4656,
      "step": 1070
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5666966438293457,
      "eval_matthews_correlation": 0.30468062994701894,
      "eval_runtime": 7.5686,
      "eval_samples_per_second": 137.807,
      "eval_steps_per_second": 17.308,
      "step": 1070
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0,
      "loss": 0.3329,
      "step": 1605
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5929099917411804,
      "eval_matthews_correlation": 0.35940659235571387,
      "eval_runtime": 7.5654,
      "eval_samples_per_second": 137.865,
      "eval_steps_per_second": 17.316,
      "step": 1605
    },
    {
      "epoch": 3.0,
      "step": 1605,
      "total_flos": 4562104380880896.0,
      "train_loss": 0.46265685654874905,
      "train_runtime": 587.6397,
      "train_samples_per_second": 43.654,
      "train_steps_per_second": 2.731
    }
  ],
  "max_steps": 1605,
  "num_train_epochs": 3,
  "total_flos": 4562104380880896.0,
  "trial_name": null,
  "trial_params": null
}