trainer_state.json

{
  "best_metric": 0.9391372891020696,
  "best_model_checkpoint": "model_saves/electra-large_lemon_5k_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": 0.9390925658828922,
      "eval_loss": 0.43472719192504883,
      "eval_runtime": 3.4124,
      "eval_samples_per_second": 1284.14,
      "eval_steps_per_second": 10.257,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9391372891020696,
      "eval_loss": 0.44080281257629395,
      "eval_runtime": 3.4128,
      "eval_samples_per_second": 1284.0,
      "eval_steps_per_second": 10.256,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9388633593846085,
      "eval_loss": 0.46593591570854187,
      "eval_runtime": 3.4062,
      "eval_samples_per_second": 1286.492,
      "eval_steps_per_second": 10.275,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.3492,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9386900569102964,
      "eval_loss": 0.48738524317741394,
      "eval_runtime": 3.408,
      "eval_samples_per_second": 1285.813,
      "eval_steps_per_second": 10.27,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9373148179205939,
      "eval_loss": 0.5164535641670227,
      "eval_runtime": 3.407,
      "eval_samples_per_second": 1286.175,
      "eval_steps_per_second": 10.273,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.7386042109722624e+16,
      "train_loss": 0.32128405784493064,
      "train_runtime": 407.3672,
      "train_samples_per_second": 1263.136,
      "train_steps_per_second": 9.868
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.7386042109722624e+16,
  "trial_name": null,
  "trial_params": null
}