{
  "best_global_step": 30,
  "best_metric": 0.5905007123947144,
  "best_model_checkpoint": "distilbert_rand_50_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": 0.8302044868469238,
      "learning_rate": 4.9e-05,
      "loss": 0.6355,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6887254901960784,
      "eval_combined_score": 0.7390559457613891,
      "eval_f1": 0.7893864013266998,
      "eval_loss": 0.6204187870025635,
      "eval_runtime": 0.1935,
      "eval_samples_per_second": 2108.083,
      "eval_steps_per_second": 10.334,
      "step": 15
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8893444538116455,
      "learning_rate": 4.8e-05,
      "loss": 0.5874,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.696078431372549,
      "eval_combined_score": 0.7505549389567148,
      "eval_f1": 0.8050314465408805,
      "eval_loss": 0.5905007123947144,
      "eval_runtime": 0.1829,
      "eval_samples_per_second": 2231.173,
      "eval_steps_per_second": 10.937,
      "step": 30
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5524550676345825,
      "learning_rate": 4.7e-05,
      "loss": 0.5207,
      "step": 45
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6985294117647058,
      "eval_combined_score": 0.7472746561311092,
      "eval_f1": 0.7960199004975125,
      "eval_loss": 0.6094551682472229,
      "eval_runtime": 0.2069,
      "eval_samples_per_second": 1971.989,
      "eval_steps_per_second": 9.667,
      "step": 45
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.8295140266418457,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4109,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6617647058823529,
      "eval_combined_score": 0.7058823529411764,
      "eval_f1": 0.75,
      "eval_loss": 0.7317143082618713,
      "eval_runtime": 0.1926,
      "eval_samples_per_second": 2118.658,
      "eval_steps_per_second": 10.386,
      "step": 60
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.282629013061523,
      "learning_rate": 4.5e-05,
      "loss": 0.2525,
      "step": 75
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6740196078431373,
      "eval_combined_score": 0.7140892863614947,
      "eval_f1": 0.7541589648798521,
      "eval_loss": 0.9529955983161926,
      "eval_runtime": 0.2035,
      "eval_samples_per_second": 2005.221,
      "eval_steps_per_second": 9.83,
      "step": 75
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.4325482845306396,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1571,
      "step": 90
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6593137254901961,
      "eval_combined_score": 0.7035225614746806,
      "eval_f1": 0.7477313974591652,
      "eval_loss": 1.0933626890182495,
      "eval_runtime": 0.1841,
      "eval_samples_per_second": 2216.311,
      "eval_steps_per_second": 10.864,
      "step": 90
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.638601541519165,
      "learning_rate": 4.3e-05,
      "loss": 0.1027,
      "step": 105
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6593137254901961,
      "eval_combined_score": 0.7016642292276026,
      "eval_f1": 0.7440147329650092,
      "eval_loss": 1.2742809057235718,
      "eval_runtime": 0.1871,
      "eval_samples_per_second": 2180.928,
      "eval_steps_per_second": 10.691,
      "step": 105
    },
    {
      "epoch": 7.0,
      "step": 105,
      "total_flos": 1700616463945728.0,
      "train_loss": 0.3809672083173479,
      "train_runtime": 34.0269,
      "train_samples_per_second": 5389.851,
      "train_steps_per_second": 22.041
    }
  ],
  "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
}