trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.999939338792842,
  "eval_steps": 500,
  "global_step": 8242,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 2e-05,
      "loss": 1.0899,
      "step": 500
    },
    {
      "epoch": 0.12,
      "learning_rate": 2e-05,
      "loss": 1.0809,
      "step": 1000
    },
    {
      "epoch": 0.18,
      "learning_rate": 2e-05,
      "loss": 1.0725,
      "step": 1500
    },
    {
      "epoch": 0.24,
      "learning_rate": 2e-05,
      "loss": 1.0673,
      "step": 2000
    },
    {
      "epoch": 0.3,
      "learning_rate": 2e-05,
      "loss": 1.0655,
      "step": 2500
    },
    {
      "epoch": 0.36,
      "learning_rate": 2e-05,
      "loss": 1.0658,
      "step": 3000
    },
    {
      "epoch": 0.42,
      "learning_rate": 2e-05,
      "loss": 1.0719,
      "step": 3500
    },
    {
      "epoch": 0.49,
      "learning_rate": 2e-05,
      "loss": 1.0571,
      "step": 4000
    },
    {
      "epoch": 0.55,
      "learning_rate": 2e-05,
      "loss": 1.0567,
      "step": 4500
    },
    {
      "epoch": 0.61,
      "learning_rate": 2e-05,
      "loss": 1.0511,
      "step": 5000
    },
    {
      "epoch": 0.67,
      "learning_rate": 2e-05,
      "loss": 1.0541,
      "step": 5500
    },
    {
      "epoch": 0.73,
      "learning_rate": 2e-05,
      "loss": 1.0539,
      "step": 6000
    },
    {
      "epoch": 0.79,
      "learning_rate": 2e-05,
      "loss": 1.0504,
      "step": 6500
    },
    {
      "epoch": 0.85,
      "learning_rate": 2e-05,
      "loss": 1.0387,
      "step": 7000
    },
    {
      "epoch": 0.91,
      "learning_rate": 2e-05,
      "loss": 1.0517,
      "step": 7500
    },
    {
      "epoch": 0.97,
      "learning_rate": 2e-05,
      "loss": 1.0439,
      "step": 8000
    },
    {
      "epoch": 1.0,
      "step": 8242,
      "total_flos": 1.475330277017387e+18,
      "train_loss": 1.0600965963177125,
      "train_runtime": 93908.9997,
      "train_samples_per_second": 2.809,
      "train_steps_per_second": 0.088
    }
  ],
  "logging_steps": 500,
  "max_steps": 8242,
  "num_train_epochs": 1,
  "save_steps": 2000,
  "total_flos": 1.475330277017387e+18,
  "trial_name": null,
  "trial_params": null
}