trainer_state.json

{
  "best_metric": 0.32399147748947144,
  "best_model_checkpoint": "distilbert_lda_100_v1_book_sst2/checkpoint-528",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 1848,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 7.3560991287231445,
      "learning_rate": 4.9e-05,
      "loss": 0.3301,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8715596330275229,
      "eval_loss": 0.33814114332199097,
      "eval_runtime": 0.2731,
      "eval_samples_per_second": 3192.734,
      "eval_steps_per_second": 14.646,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.438405990600586,
      "learning_rate": 4.8e-05,
      "loss": 0.1797,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8830275229357798,
      "eval_loss": 0.32399147748947144,
      "eval_runtime": 0.2724,
      "eval_samples_per_second": 3200.61,
      "eval_steps_per_second": 14.682,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.700658321380615,
      "learning_rate": 4.7e-05,
      "loss": 0.1196,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8669724770642202,
      "eval_loss": 0.42807525396347046,
      "eval_runtime": 0.2783,
      "eval_samples_per_second": 3132.764,
      "eval_steps_per_second": 14.37,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.581217288970947,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.0899,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8772935779816514,
      "eval_loss": 0.4239971935749054,
      "eval_runtime": 0.3097,
      "eval_samples_per_second": 2815.249,
      "eval_steps_per_second": 12.914,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.4569306373596191,
      "learning_rate": 4.5e-05,
      "loss": 0.0665,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8841743119266054,
      "eval_loss": 0.36415040493011475,
      "eval_runtime": 0.2973,
      "eval_samples_per_second": 2932.866,
      "eval_steps_per_second": 13.454,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.0867676734924316,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0508,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8922018348623854,
      "eval_loss": 0.38988909125328064,
      "eval_runtime": 0.2999,
      "eval_samples_per_second": 2907.212,
      "eval_steps_per_second": 13.336,
      "step": 1584
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.9464542865753174,
      "learning_rate": 4.3e-05,
      "loss": 0.0412,
      "step": 1848
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8784403669724771,
      "eval_loss": 0.45416074991226196,
      "eval_runtime": 0.2686,
      "eval_samples_per_second": 3246.521,
      "eval_steps_per_second": 14.892,
      "step": 1848
    },
    {
      "epoch": 7.0,
      "step": 1848,
      "total_flos": 3.1225413912290304e+16,
      "train_loss": 0.12540515779932856,
      "train_runtime": 317.3431,
      "train_samples_per_second": 10611.385,
      "train_steps_per_second": 41.595
    }
  ],
  "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": 3.1225413912290304e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}