trainer_state.json

{
  "best_metric": 0.5409021973609924,
  "best_model_checkpoint": "distilbert_lda_20_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": 2.5330288410186768,
      "learning_rate": 4.9e-05,
      "loss": 0.7024,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7646459500764137,
      "eval_loss": 0.5842658281326294,
      "eval_runtime": 4.3578,
      "eval_samples_per_second": 2252.258,
      "eval_steps_per_second": 8.949,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.305652141571045,
      "learning_rate": 4.8e-05,
      "loss": 0.5147,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7890983188996434,
      "eval_loss": 0.5409021973609924,
      "eval_runtime": 4.3076,
      "eval_samples_per_second": 2278.506,
      "eval_steps_per_second": 9.054,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.426408052444458,
      "learning_rate": 4.7e-05,
      "loss": 0.407,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7910341314314824,
      "eval_loss": 0.5485309362411499,
      "eval_runtime": 4.5464,
      "eval_samples_per_second": 2158.847,
      "eval_steps_per_second": 8.578,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.034677028656006,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.3142,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.789302088639837,
      "eval_loss": 0.6063764691352844,
      "eval_runtime": 4.3943,
      "eval_samples_per_second": 2233.584,
      "eval_steps_per_second": 8.875,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.319906711578369,
      "learning_rate": 4.5e-05,
      "loss": 0.2419,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7875700458481916,
      "eval_loss": 0.6936997771263123,
      "eval_runtime": 4.3472,
      "eval_samples_per_second": 2257.795,
      "eval_steps_per_second": 8.971,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.0567739009857178,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1866,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7874681609780948,
      "eval_loss": 0.7424802184104919,
      "eval_runtime": 4.3189,
      "eval_samples_per_second": 2272.577,
      "eval_steps_per_second": 9.03,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.2307686805725098,
      "learning_rate": 4.3e-05,
      "loss": 0.1483,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7881813550687723,
      "eval_loss": 0.7587445974349976,
      "eval_runtime": 4.2598,
      "eval_samples_per_second": 2304.108,
      "eval_steps_per_second": 9.155,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "step": 10738,
      "total_flos": 1.8207399032776397e+17,
      "train_loss": 0.3593115150251298,
      "train_runtime": 2646.1923,
      "train_samples_per_second": 7420.134,
      "train_steps_per_second": 28.985
    }
  ],
  "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
}