trainer_state.json

{
  "best_metric": 0.9387850937510482,
  "best_model_checkpoint": "model_saves/roberta-large_lemon-spell_5k_3_p3/checkpoint-536",
  "epoch": 5.0,
  "global_step": 1340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9387124185198851,
      "eval_loss": 0.43775027990341187,
      "eval_runtime": 3.3826,
      "eval_samples_per_second": 1295.466,
      "eval_steps_per_second": 10.347,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9387850937510482,
      "eval_loss": 0.45006421208381653,
      "eval_runtime": 3.3747,
      "eval_samples_per_second": 1298.5,
      "eval_steps_per_second": 10.371,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9381366070729771,
      "eval_loss": 0.49755820631980896,
      "eval_runtime": 3.3886,
      "eval_samples_per_second": 1293.141,
      "eval_steps_per_second": 10.329,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.272,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9381198358657856,
      "eval_loss": 0.5394700169563293,
      "eval_runtime": 3.375,
      "eval_samples_per_second": 1298.359,
      "eval_steps_per_second": 10.37,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9376390612596295,
      "eval_loss": 0.593403697013855,
      "eval_runtime": 3.3747,
      "eval_samples_per_second": 1298.477,
      "eval_steps_per_second": 10.371,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.7322435992879104e+16,
      "train_loss": 0.24598438490682573,
      "train_runtime": 398.7676,
      "train_samples_per_second": 1290.376,
      "train_steps_per_second": 10.081
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.7322435992879104e+16,
  "trial_name": null,
  "trial_params": null
}