trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6154244542121887,
  "best_model_checkpoint": "distilbert_rand_10_v2_cola/checkpoint-34",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.1962289810180664,
      "learning_rate": 4.9e-05,
      "loss": 0.615,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6154244542121887,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4402,
      "eval_samples_per_second": 2369.255,
      "eval_steps_per_second": 11.358,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8502079248428345,
      "learning_rate": 4.8e-05,
      "loss": 0.5908,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6836050152778625,
      "eval_loss": 0.6301819682121277,
      "eval_matthews_correlation": 0.05132011407765458,
      "eval_runtime": 0.4854,
      "eval_samples_per_second": 2148.911,
      "eval_steps_per_second": 10.302,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.2110676765441895,
      "learning_rate": 4.7e-05,
      "loss": 0.5475,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6816874146461487,
      "eval_loss": 0.6317144632339478,
      "eval_matthews_correlation": 0.08034386400319438,
      "eval_runtime": 0.4663,
      "eval_samples_per_second": 2236.665,
      "eval_steps_per_second": 10.722,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.9596452713012695,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4883,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6500479578971863,
      "eval_loss": 0.7006556391716003,
      "eval_matthews_correlation": 0.09743296097822274,
      "eval_runtime": 0.4737,
      "eval_samples_per_second": 2201.646,
      "eval_steps_per_second": 10.554,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.2971408367156982,
      "learning_rate": 4.5e-05,
      "loss": 0.4326,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6471716165542603,
      "eval_loss": 0.7236018180847168,
      "eval_matthews_correlation": 0.08382215590402259,
      "eval_runtime": 0.4459,
      "eval_samples_per_second": 2339.048,
      "eval_steps_per_second": 11.213,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.464991092681885,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3826,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6433365345001221,
      "eval_loss": 0.864536702632904,
      "eval_matthews_correlation": 0.07335755495747934,
      "eval_runtime": 0.469,
      "eval_samples_per_second": 2224.069,
      "eval_steps_per_second": 10.662,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 3398186177722368.0,
      "train_loss": 0.5094801912120744,
      "train_runtime": 58.0721,
      "train_samples_per_second": 7362.398,
      "train_steps_per_second": 29.274
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "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": 3398186177722368.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}