trainer_state.json

{
  "best_metric": 0.9394391708315164,
  "best_model_checkpoint": "model_saves/electra-large_lemon_10k_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": 0.9390422522613178,
      "eval_loss": 0.4451042413711548,
      "eval_runtime": 3.4593,
      "eval_samples_per_second": 1266.736,
      "eval_steps_per_second": 10.118,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9394391708315164,
      "eval_loss": 0.450582891702652,
      "eval_runtime": 3.4585,
      "eval_samples_per_second": 1267.034,
      "eval_steps_per_second": 10.12,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9390590234685092,
      "eval_loss": 0.4746180772781372,
      "eval_runtime": 3.4728,
      "eval_samples_per_second": 1261.796,
      "eval_steps_per_second": 10.078,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.3499,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9389751674325518,
      "eval_loss": 0.49700310826301575,
      "eval_runtime": 3.4755,
      "eval_samples_per_second": 1260.834,
      "eval_steps_per_second": 10.071,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9370073457887499,
      "eval_loss": 0.5279273986816406,
      "eval_runtime": 3.4601,
      "eval_samples_per_second": 1266.448,
      "eval_steps_per_second": 10.115,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.767586509357056e+16,
      "train_loss": 0.32197913554177354,
      "train_runtime": 403.0004,
      "train_samples_per_second": 1276.823,
      "train_steps_per_second": 9.975
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.767586509357056e+16,
  "trial_name": null,
  "trial_params": null
}