trainer_state.json

{
  "best_metric": 0.6930387616157532,
  "best_model_checkpoint": "add_BERT_no_pretrain_qnli/checkpoint-819",
  "epoch": 6.0,
  "global_step": 4914,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0.0004900366300366301,
      "loss": 0.7205,
      "step": 819
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6930387616157532,
      "eval_runtime": 8.5235,
      "eval_samples_per_second": 640.931,
      "eval_steps_per_second": 5.045,
      "step": 819
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00048003663003663005,
      "loss": 0.6935,
      "step": 1638
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6930967569351196,
      "eval_runtime": 8.519,
      "eval_samples_per_second": 641.274,
      "eval_steps_per_second": 5.048,
      "step": 1638
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00047003663003663003,
      "loss": 0.6932,
      "step": 2457
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6930704116821289,
      "eval_runtime": 8.5483,
      "eval_samples_per_second": 639.078,
      "eval_steps_per_second": 5.03,
      "step": 2457
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00046003663003663006,
      "loss": 0.6932,
      "step": 3276
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4946000366099213,
      "eval_loss": 0.6931969523429871,
      "eval_runtime": 8.5019,
      "eval_samples_per_second": 642.559,
      "eval_steps_per_second": 5.058,
      "step": 3276
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00045003663003663003,
      "loss": 0.6932,
      "step": 4095
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6931020617485046,
      "eval_runtime": 8.5439,
      "eval_samples_per_second": 639.401,
      "eval_steps_per_second": 5.033,
      "step": 4095
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.00044003663003663,
      "loss": 0.6935,
      "step": 4914
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6933435797691345,
      "eval_runtime": 8.5341,
      "eval_samples_per_second": 640.137,
      "eval_steps_per_second": 5.039,
      "step": 4914
    },
    {
      "epoch": 6.0,
      "step": 4914,
      "total_flos": 8.975886795669504e+16,
      "train_loss": 0.697838950555015,
      "train_runtime": 2914.3249,
      "train_samples_per_second": 1797.037,
      "train_steps_per_second": 14.051
    }
  ],
  "max_steps": 40950,
  "num_train_epochs": 50,
  "total_flos": 8.975886795669504e+16,
  "trial_name": null,
  "trial_params": null
}