{
  "best_global_step": 30,
  "best_metric": 0.5877290964126587,
  "best_model_checkpoint": "distilbert_rand_5_v1_mrpc/checkpoint-30",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 105,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.2239353656768799,
      "learning_rate": 4.9e-05,
      "loss": 0.6291,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6911764705882353,
      "eval_combined_score": 0.7462191501206161,
      "eval_f1": 0.8012618296529969,
      "eval_loss": 0.5983169078826904,
      "eval_runtime": 0.1987,
      "eval_samples_per_second": 2053.82,
      "eval_steps_per_second": 10.068,
      "step": 15
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7938200235366821,
      "learning_rate": 4.8e-05,
      "loss": 0.5735,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6936274509803921,
      "eval_combined_score": 0.7445223997946151,
      "eval_f1": 0.795417348608838,
      "eval_loss": 0.5877290964126587,
      "eval_runtime": 0.1932,
      "eval_samples_per_second": 2112.17,
      "eval_steps_per_second": 10.354,
      "step": 30
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.7242307662963867,
      "learning_rate": 4.7e-05,
      "loss": 0.5095,
      "step": 45
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6936274509803921,
      "eval_combined_score": 0.7474496555378909,
      "eval_f1": 0.8012718600953895,
      "eval_loss": 0.6555451154708862,
      "eval_runtime": 0.1808,
      "eval_samples_per_second": 2257.26,
      "eval_steps_per_second": 11.065,
      "step": 45
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.557063102722168,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4245,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_combined_score": 0.7144522144522144,
      "eval_f1": 0.7622377622377622,
      "eval_loss": 0.7126985192298889,
      "eval_runtime": 0.1801,
      "eval_samples_per_second": 2264.899,
      "eval_steps_per_second": 11.102,
      "step": 60
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.274909973144531,
      "learning_rate": 4.5e-05,
      "loss": 0.2889,
      "step": 75
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6004901960784313,
      "eval_combined_score": 0.6419926708547496,
      "eval_f1": 0.683495145631068,
      "eval_loss": 0.9767698645591736,
      "eval_runtime": 0.1796,
      "eval_samples_per_second": 2271.486,
      "eval_steps_per_second": 11.135,
      "step": 75
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.6112961769104,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1826,
      "step": 90
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5882352941176471,
      "eval_combined_score": 0.6287633163501621,
      "eval_f1": 0.6692913385826772,
      "eval_loss": 1.1932697296142578,
      "eval_runtime": 0.1786,
      "eval_samples_per_second": 2284.37,
      "eval_steps_per_second": 11.198,
      "step": 90
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.048973798751831,
      "learning_rate": 4.3e-05,
      "loss": 0.1083,
      "step": 105
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6495098039215687,
      "eval_combined_score": 0.6977566781597186,
      "eval_f1": 0.7460035523978685,
      "eval_loss": 1.4463649988174438,
      "eval_runtime": 0.1781,
      "eval_samples_per_second": 2290.733,
      "eval_steps_per_second": 11.229,
      "step": 105
    },
    {
      "epoch": 7.0,
      "step": 105,
      "total_flos": 1700616463945728.0,
      "train_loss": 0.38803887594313846,
      "train_runtime": 33.6267,
      "train_samples_per_second": 5454.007,
      "train_steps_per_second": 22.304
    }
  ],
  "logging_steps": 1,
  "max_steps": 750,
  "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": 1700616463945728.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}