trainer_state.json

{
  "best_metric": 0.9388969017989915,
  "best_model_checkpoint": "model_saves/electra-large_spell_10k_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": 0.9382595959257147,
      "eval_loss": 0.44925907254219055,
      "eval_runtime": 3.4981,
      "eval_samples_per_second": 1252.685,
      "eval_steps_per_second": 10.005,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9388969017989915,
      "eval_loss": 0.4679044187068939,
      "eval_runtime": 3.4989,
      "eval_samples_per_second": 1252.38,
      "eval_steps_per_second": 10.003,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9377340981003812,
      "eval_loss": 0.48627498745918274,
      "eval_runtime": 3.5462,
      "eval_samples_per_second": 1235.694,
      "eval_steps_per_second": 9.87,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.32,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9373371795301826,
      "eval_loss": 0.5085417032241821,
      "eval_runtime": 3.4976,
      "eval_samples_per_second": 1252.857,
      "eval_steps_per_second": 10.007,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9374266259685372,
      "eval_loss": 0.5530784130096436,
      "eval_runtime": 3.496,
      "eval_samples_per_second": 1253.445,
      "eval_steps_per_second": 10.012,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.7677305786662912e+16,
      "train_loss": 0.2936649664124446,
      "train_runtime": 404.1964,
      "train_samples_per_second": 1273.045,
      "train_steps_per_second": 9.946
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.7677305786662912e+16,
  "trial_name": null,
  "trial_params": null
}