trainer_state.json

{
  "best_metric": 0.5361664295196533,
  "best_model_checkpoint": "distilbert_lda_5_v1_book_mnli/checkpoint-3068",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 10738,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.8657944202423096,
      "learning_rate": 4.9e-05,
      "loss": 0.6871,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7596535914416709,
      "eval_loss": 0.5890195965766907,
      "eval_runtime": 3.0251,
      "eval_samples_per_second": 3244.487,
      "eval_steps_per_second": 12.892,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.1727097034454346,
      "learning_rate": 4.8e-05,
      "loss": 0.5082,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.785838003056546,
      "eval_loss": 0.5361664295196533,
      "eval_runtime": 3.0369,
      "eval_samples_per_second": 3231.945,
      "eval_steps_per_second": 12.842,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.4147913455963135,
      "learning_rate": 4.7e-05,
      "loss": 0.4006,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7894039735099337,
      "eval_loss": 0.562130868434906,
      "eval_runtime": 3.0528,
      "eval_samples_per_second": 3215.043,
      "eval_steps_per_second": 12.775,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.6182262897491455,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.3094,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7918492103922568,
      "eval_loss": 0.6176037192344666,
      "eval_runtime": 3.0715,
      "eval_samples_per_second": 3195.467,
      "eval_steps_per_second": 12.697,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.770238161087036,
      "learning_rate": 4.5e-05,
      "loss": 0.2381,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7872643912379012,
      "eval_loss": 0.6732258796691895,
      "eval_runtime": 3.0367,
      "eval_samples_per_second": 3232.172,
      "eval_steps_per_second": 12.843,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.009212017059326,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1841,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7832908813041264,
      "eval_loss": 0.7930567264556885,
      "eval_runtime": 3.041,
      "eval_samples_per_second": 3227.55,
      "eval_steps_per_second": 12.825,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.186721324920654,
      "learning_rate": 4.3e-05,
      "loss": 0.1471,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7869587366276108,
      "eval_loss": 0.8369612097740173,
      "eval_runtime": 3.0611,
      "eval_samples_per_second": 3206.313,
      "eval_steps_per_second": 12.74,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "step": 10738,
      "total_flos": 1.8207399032776397e+17,
      "train_loss": 0.35354660014285566,
      "train_runtime": 1823.5924,
      "train_samples_per_second": 10767.264,
      "train_steps_per_second": 42.06
    }
  ],
  "logging_steps": 1,
  "max_steps": 76700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8207399032776397e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}