trainer_state.json

{
  "best_metric": 0.9412560516105949,
  "best_model_checkpoint": "model_saves/deberta-v3-large_lemon-spell_10k_2_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": 0.9412504612081978,
      "eval_loss": 0.4289807677268982,
      "eval_runtime": 4.5999,
      "eval_samples_per_second": 952.625,
      "eval_steps_per_second": 7.609,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9412560516105949,
      "eval_loss": 0.4463995695114136,
      "eval_runtime": 4.5763,
      "eval_samples_per_second": 957.55,
      "eval_steps_per_second": 7.648,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9404286720558146,
      "eval_loss": 0.4729045033454895,
      "eval_runtime": 4.5746,
      "eval_samples_per_second": 957.902,
      "eval_steps_per_second": 7.651,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2621,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9397354621585662,
      "eval_loss": 0.5097540616989136,
      "eval_runtime": 4.5699,
      "eval_samples_per_second": 958.888,
      "eval_steps_per_second": 7.659,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9393944476123391,
      "eval_loss": 0.550991415977478,
      "eval_runtime": 4.5697,
      "eval_samples_per_second": 958.919,
      "eval_steps_per_second": 7.659,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.7253490090835968e+16,
      "train_loss": 0.23529494556028452,
      "train_runtime": 594.4617,
      "train_samples_per_second": 865.59,
      "train_steps_per_second": 6.762
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.7253490090835968e+16,
  "trial_name": null,
  "trial_params": null
}