{
  "best_global_step": 30,
  "best_metric": 0.6022560596466064,
  "best_model_checkpoint": "distilbert_km_10_v2_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": 1.3603156805038452,
      "learning_rate": 4.9e-05,
      "loss": 0.6198,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6911764705882353,
      "eval_combined_score": 0.7459046909903202,
      "eval_f1": 0.8006329113924051,
      "eval_loss": 0.6141668558120728,
      "eval_runtime": 0.1876,
      "eval_samples_per_second": 2175.068,
      "eval_steps_per_second": 10.662,
      "step": 15
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8541057109832764,
      "learning_rate": 4.8e-05,
      "loss": 0.5544,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6911764705882353,
      "eval_combined_score": 0.7474573941726224,
      "eval_f1": 0.8037383177570093,
      "eval_loss": 0.6022560596466064,
      "eval_runtime": 0.1889,
      "eval_samples_per_second": 2159.628,
      "eval_steps_per_second": 10.586,
      "step": 30
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.4674851894378662,
      "learning_rate": 4.7e-05,
      "loss": 0.4816,
      "step": 45
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.696078431372549,
      "eval_combined_score": 0.7517659237608086,
      "eval_f1": 0.8074534161490683,
      "eval_loss": 0.6422035098075867,
      "eval_runtime": 0.1887,
      "eval_samples_per_second": 2162.392,
      "eval_steps_per_second": 10.6,
      "step": 45
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9297282695770264,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.3821,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6813725490196079,
      "eval_combined_score": 0.7297647728033193,
      "eval_f1": 0.7781569965870307,
      "eval_loss": 0.6905685067176819,
      "eval_runtime": 0.1823,
      "eval_samples_per_second": 2238.056,
      "eval_steps_per_second": 10.971,
      "step": 60
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.107450485229492,
      "learning_rate": 4.5e-05,
      "loss": 0.2574,
      "step": 75
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6299019607843137,
      "eval_combined_score": 0.6743550772264213,
      "eval_f1": 0.7188081936685289,
      "eval_loss": 0.8928166627883911,
      "eval_runtime": 0.2114,
      "eval_samples_per_second": 1929.966,
      "eval_steps_per_second": 9.461,
      "step": 75
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.032618284225464,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1376,
      "step": 90
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6642156862745098,
      "eval_combined_score": 0.7117211999035115,
      "eval_f1": 0.7592267135325131,
      "eval_loss": 1.0320990085601807,
      "eval_runtime": 0.1835,
      "eval_samples_per_second": 2223.563,
      "eval_steps_per_second": 10.9,
      "step": 90
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.734239339828491,
      "learning_rate": 4.3e-05,
      "loss": 0.0682,
      "step": 105
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6568627450980392,
      "eval_combined_score": 0.7081564584596725,
      "eval_f1": 0.7594501718213058,
      "eval_loss": 1.322534203529358,
      "eval_runtime": 0.1973,
      "eval_samples_per_second": 2067.95,
      "eval_steps_per_second": 10.137,
      "step": 105
    },
    {
      "epoch": 7.0,
      "step": 105,
      "total_flos": 1700616463945728.0,
      "train_loss": 0.35731526556469145,
      "train_runtime": 34.5225,
      "train_samples_per_second": 5312.48,
      "train_steps_per_second": 21.725
    }
  ],
  "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
}