trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.43674904108047485,
  "best_model_checkpoint": "distilbert_rand_5_v1_sst2/checkpoint-264",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1584,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 7.028470516204834,
      "learning_rate": 4.9e-05,
      "loss": 0.387,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8119266055045872,
      "eval_loss": 0.43674904108047485,
      "eval_runtime": 0.3736,
      "eval_samples_per_second": 2334.347,
      "eval_steps_per_second": 10.708,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.199335098266602,
      "learning_rate": 4.8e-05,
      "loss": 0.2193,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8153669724770642,
      "eval_loss": 0.4830808639526367,
      "eval_runtime": 0.3779,
      "eval_samples_per_second": 2307.679,
      "eval_steps_per_second": 10.586,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.804741859436035,
      "learning_rate": 4.7e-05,
      "loss": 0.1616,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8142201834862385,
      "eval_loss": 0.49482759833335876,
      "eval_runtime": 0.3945,
      "eval_samples_per_second": 2210.297,
      "eval_steps_per_second": 10.139,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.6866455078125,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1208,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8291284403669725,
      "eval_loss": 0.5718021392822266,
      "eval_runtime": 0.3672,
      "eval_samples_per_second": 2375.006,
      "eval_steps_per_second": 10.895,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.6306769847869873,
      "learning_rate": 4.5e-05,
      "loss": 0.0926,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.823394495412844,
      "eval_loss": 0.6209654211997986,
      "eval_runtime": 0.3645,
      "eval_samples_per_second": 2392.412,
      "eval_steps_per_second": 10.974,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.8861637115478516,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0736,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.801605504587156,
      "eval_loss": 0.7925794124603271,
      "eval_runtime": 0.3657,
      "eval_samples_per_second": 2384.425,
      "eval_steps_per_second": 10.938,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.1758228432048451,
      "train_runtime": 376.2554,
      "train_samples_per_second": 8949.905,
      "train_steps_per_second": 35.083
    }
  ],
  "logging_steps": 1,
  "max_steps": 13200,
  "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": 2.676464049624883e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}