trainer_state.json

{
  "best_metric": 0.9390478426637149,
  "best_model_checkpoint": "model_saves/electra-large_spell_5k_1_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": 0.9384440792048212,
      "eval_loss": 0.43843042850494385,
      "eval_runtime": 3.4438,
      "eval_samples_per_second": 1272.433,
      "eval_steps_per_second": 10.163,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9390478426637149,
      "eval_loss": 0.4556538760662079,
      "eval_runtime": 3.4482,
      "eval_samples_per_second": 1270.818,
      "eval_steps_per_second": 10.15,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9380751126466083,
      "eval_loss": 0.4742107391357422,
      "eval_runtime": 3.4555,
      "eval_samples_per_second": 1268.117,
      "eval_steps_per_second": 10.129,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.3211,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9376837844788067,
      "eval_loss": 0.49898871779441833,
      "eval_runtime": 3.4609,
      "eval_samples_per_second": 1266.15,
      "eval_steps_per_second": 10.113,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9373259987253882,
      "eval_loss": 0.5385605096817017,
      "eval_runtime": 3.4414,
      "eval_samples_per_second": 1273.301,
      "eval_steps_per_second": 10.17,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.7387458509406208e+16,
      "train_loss": 0.2949276369009445,
      "train_runtime": 399.6066,
      "train_samples_per_second": 1287.666,
      "train_steps_per_second": 10.06
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.7387458509406208e+16,
  "trial_name": null,
  "trial_params": null
}