last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0655379574003277,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0,
      "loss": 5.4178,
      "step": 1
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.294882868674145e-05,
      "loss": 0.446,
      "step": 5
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.852558565662928e-05,
      "loss": 0.3922,
      "step": 10
    },
    {
      "epoch": 0.07,
      "learning_rate": 2.1787779359648994e-05,
      "loss": 0.3732,
      "step": 15
    }
  ],
  "max_steps": 1140,
  "num_train_epochs": 5,
  "total_flos": 1.1404413776166912e+17,
  "trial_name": null,
  "trial_params": null
}