trainer_state.json

{
  "best_metric": 0.9393720860027505,
  "best_model_checkpoint": "model_saves/electra-large_lemon_5k_5_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": 0.9386229720815304,
      "eval_loss": 0.4429810643196106,
      "eval_runtime": 3.4072,
      "eval_samples_per_second": 1286.087,
      "eval_steps_per_second": 10.272,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9393720860027505,
      "eval_loss": 0.45496228337287903,
      "eval_runtime": 3.4171,
      "eval_samples_per_second": 1282.392,
      "eval_steps_per_second": 10.243,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9387124185198851,
      "eval_loss": 0.480425626039505,
      "eval_runtime": 3.4294,
      "eval_samples_per_second": 1277.776,
      "eval_steps_per_second": 10.206,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.321,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9375328436140833,
      "eval_loss": 0.5025012493133545,
      "eval_runtime": 3.4073,
      "eval_samples_per_second": 1286.055,
      "eval_steps_per_second": 10.272,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9364874383658136,
      "eval_loss": 0.5369945764541626,
      "eval_runtime": 3.4228,
      "eval_samples_per_second": 1280.221,
      "eval_steps_per_second": 10.225,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.7366444845039616e+16,
      "train_loss": 0.29449602383286205,
      "train_runtime": 405.4112,
      "train_samples_per_second": 1269.23,
      "train_steps_per_second": 9.916
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.7366444845039616e+16,
  "trial_name": null,
  "trial_params": null
}