trainer_state.json

{
  "best_metric": 87.82372321450285,
  "best_model_checkpoint": "albertina/checkpoint-2806",
  "epoch": 3.0,
  "global_step": 4209,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.3357091945830365e-05,
      "loss": 0.9145,
      "step": 1403
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 76.90633869441817,
      "eval_f1": 87.45778804454862,
      "eval_runtime": 189.9155,
      "eval_samples_per_second": 57.799,
      "eval_steps_per_second": 7.23,
      "step": 1403
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.690425279163697e-06,
      "loss": 0.5008,
      "step": 2806
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 76.96310312204352,
      "eval_f1": 87.82372321450285,
      "eval_runtime": 190.0892,
      "eval_samples_per_second": 57.747,
      "eval_steps_per_second": 7.223,
      "step": 2806
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.3758612497030175e-08,
      "loss": 0.2529,
      "step": 4209
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 75.63859981078524,
      "eval_f1": 87.40646900990333,
      "eval_runtime": 189.9958,
      "eval_samples_per_second": 57.775,
      "eval_steps_per_second": 7.226,
      "step": 4209
    },
    {
      "epoch": 3.0,
      "step": 4209,
      "total_flos": 5.682949586646958e+17,
      "train_loss": 0.5560661419238351,
      "train_runtime": 17808.0437,
      "train_samples_per_second": 15.12,
      "train_steps_per_second": 0.236
    }
  ],
  "max_steps": 4209,
  "num_train_epochs": 3,
  "total_flos": 5.682949586646958e+17,
  "trial_name": null,
  "trial_params": null
}