trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 9453,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 1.8952713424309745e-05,
      "loss": 0.4879,
      "step": 500
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7894848196339786e-05,
      "loss": 0.2154,
      "step": 1000
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.683698296836983e-05,
      "loss": 0.1571,
      "step": 1500
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5779117740399874e-05,
      "loss": 0.1299,
      "step": 2000
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4721252512429918e-05,
      "loss": 0.1046,
      "step": 2500
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3665503014915903e-05,
      "loss": 0.0914,
      "step": 3000
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2607637786945943e-05,
      "loss": 0.0796,
      "step": 3500
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1549772558975987e-05,
      "loss": 0.0793,
      "step": 4000
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.049190733100603e-05,
      "loss": 0.0767,
      "step": 4500
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.436157833492013e-06,
      "loss": 0.0675,
      "step": 5000
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.378292605522057e-06,
      "loss": 0.0679,
      "step": 5500
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.3204273775521e-06,
      "loss": 0.0626,
      "step": 6000
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.264677880038084e-06,
      "loss": 0.0603,
      "step": 6500
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.206812652068127e-06,
      "loss": 0.0521,
      "step": 7000
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1489474240981706e-06,
      "loss": 0.052,
      "step": 7500
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0910821961282134e-06,
      "loss": 0.0491,
      "step": 8000
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0332169681582567e-06,
      "loss": 0.0464,
      "step": 8500
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.753517401883e-07,
      "loss": 0.0491,
      "step": 9000
    },
    {
      "epoch": 3.0,
      "step": 9453,
      "total_flos": 1.989533978433792e+16,
      "train_loss": 0.10405463833386691,
      "train_runtime": 977.5682,
      "train_samples_per_second": 309.403,
      "train_steps_per_second": 9.67
    }
  ],
  "max_steps": 9453,
  "num_train_epochs": 3,
  "total_flos": 1.989533978433792e+16,
  "trial_name": null,
  "trial_params": null
}