trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 42.30769230769231,
  "eval_steps": 260,
  "global_step": 1100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 0,
      "loss": 0.5435,
      "step": 1
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.5191032886505127,
      "eval_runtime": 1.1808,
      "eval_samples_per_second": 123.649,
      "eval_steps_per_second": 16.091,
      "step": 1
    },
    {
      "epoch": 10.0,
      "learning_rate": 4.51693610469592e-05,
      "loss": 0.4328,
      "step": 260
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.26642337441444397,
      "eval_runtime": 1.1901,
      "eval_samples_per_second": 122.677,
      "eval_steps_per_second": 15.965,
      "step": 260
    },
    {
      "epoch": 20.0,
      "learning_rate": 4.018475750577367e-05,
      "loss": 0.2398,
      "step": 520
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.14411863684654236,
      "eval_runtime": 1.1868,
      "eval_samples_per_second": 123.022,
      "eval_steps_per_second": 16.01,
      "step": 520
    },
    {
      "epoch": 30.0,
      "learning_rate": 3.518090839107005e-05,
      "loss": 0.1471,
      "step": 780
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.08720379322767258,
      "eval_runtime": 1.1923,
      "eval_samples_per_second": 122.453,
      "eval_steps_per_second": 15.936,
      "step": 780
    },
    {
      "epoch": 40.0,
      "learning_rate": 3.0177059276366438e-05,
      "loss": 0.0796,
      "step": 1040
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.04842868819832802,
      "eval_runtime": 1.1905,
      "eval_samples_per_second": 122.635,
      "eval_steps_per_second": 15.959,
      "step": 1040
    }
  ],
  "logging_steps": 260,
  "max_steps": 2600,
  "num_train_epochs": 100,
  "save_steps": 100,
  "total_flos": 8.931019832531354e+16,
  "trial_name": null,
  "trial_params": null
}