trainer_state.json

{
  "best_metric": 0.6909662485122681,
  "best_model_checkpoint": "distilbert_lda_100_v1_wnli/checkpoint-6",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.0963492393493652,
      "learning_rate": 4.9e-05,
      "loss": 0.7643,
      "step": 3
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4507042253521127,
      "eval_loss": 0.7202079892158508,
      "eval_runtime": 0.0381,
      "eval_samples_per_second": 1865.151,
      "eval_steps_per_second": 26.27,
      "step": 3
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.829197108745575,
      "learning_rate": 4.8e-05,
      "loss": 0.699,
      "step": 6
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6909662485122681,
      "eval_runtime": 0.0375,
      "eval_samples_per_second": 1895.423,
      "eval_steps_per_second": 26.696,
      "step": 6
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6926978826522827,
      "learning_rate": 4.7e-05,
      "loss": 0.7004,
      "step": 9
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4788732394366197,
      "eval_loss": 0.6951475739479065,
      "eval_runtime": 0.0377,
      "eval_samples_per_second": 1880.878,
      "eval_steps_per_second": 26.491,
      "step": 9
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.4927375316619873,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6965,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.6969354152679443,
      "eval_runtime": 0.0373,
      "eval_samples_per_second": 1902.324,
      "eval_steps_per_second": 26.793,
      "step": 12
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.2253917455673218,
      "learning_rate": 4.5e-05,
      "loss": 0.6938,
      "step": 15
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4507042253521127,
      "eval_loss": 0.7140734791755676,
      "eval_runtime": 0.039,
      "eval_samples_per_second": 1818.699,
      "eval_steps_per_second": 25.615,
      "step": 15
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.4688159227371216,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6974,
      "step": 18
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.7016118764877319,
      "eval_runtime": 0.0387,
      "eval_samples_per_second": 1835.389,
      "eval_steps_per_second": 25.851,
      "step": 18
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.0470634698867798,
      "learning_rate": 4.3e-05,
      "loss": 0.6953,
      "step": 21
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.7068936824798584,
      "eval_runtime": 0.0377,
      "eval_samples_per_second": 1885.749,
      "eval_steps_per_second": 26.56,
      "step": 21
    },
    {
      "epoch": 7.0,
      "step": 21,
      "total_flos": 294408793512960.0,
      "train_loss": 0.7066784245627267,
      "train_runtime": 16.0368,
      "train_samples_per_second": 1979.816,
      "train_steps_per_second": 9.353
    }
  ],
  "logging_steps": 1,
  "max_steps": 150,
  "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": 294408793512960.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}