trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "global_step": 235639,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 2e-05,
      "loss": 1.6042,
      "step": 23564
    },
    {
      "epoch": 0.2,
      "learning_rate": 2e-05,
      "loss": 1.4558,
      "step": 47128
    },
    {
      "epoch": 0.3,
      "learning_rate": 2e-05,
      "loss": 1.4046,
      "step": 70692
    },
    {
      "epoch": 0.4,
      "learning_rate": 2e-05,
      "loss": 1.3729,
      "step": 94256
    },
    {
      "epoch": 0.5,
      "learning_rate": 2e-05,
      "loss": 1.3504,
      "step": 117820
    },
    {
      "epoch": 0.6,
      "learning_rate": 2e-05,
      "loss": 1.3312,
      "step": 141384
    },
    {
      "epoch": 0.7,
      "learning_rate": 2e-05,
      "loss": 1.3171,
      "step": 164948
    },
    {
      "epoch": 0.8,
      "learning_rate": 2e-05,
      "loss": 1.3048,
      "step": 188512
    },
    {
      "epoch": 0.9,
      "learning_rate": 2e-05,
      "loss": 1.2958,
      "step": 212076
    },
    {
      "epoch": 1.0,
      "step": 235639,
      "total_flos": 3.9693813820965274e+18,
      "train_loss": 1.3722278172331406,
      "train_runtime": 228465.2602,
      "train_samples_per_second": 66.009,
      "train_steps_per_second": 1.031
    }
  ],
  "max_steps": 235639,
  "num_train_epochs": 1,
  "total_flos": 3.9693813820965274e+18,
  "trial_name": null,
  "trial_params": null
}