{
  "best_metric": 0.9024815820085305,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-37/checkpoint-4515",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 6450,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.3864041566848755,
      "learning_rate": 0.00016868362667848253,
      "loss": 0.2983,
      "step": 645
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8949205118262893,
      "eval_f1": 0.2322946175637394,
      "eval_loss": 0.2774158716201782,
      "eval_mcc": 0.257425944719354,
      "eval_precision": 0.5985401459854015,
      "eval_recall": 0.14411247803163443,
      "eval_runtime": 9.9039,
      "eval_samples_per_second": 520.804,
      "eval_steps_per_second": 16.357,
      "step": 645
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.49007099866867065,
      "learning_rate": 0.0001499410014919845,
      "loss": 0.2764,
      "step": 1290
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8960837533927879,
      "eval_f1": 0.1518987341772152,
      "eval_loss": 0.27505719661712646,
      "eval_mcc": 0.23128061022537116,
      "eval_precision": 0.7619047619047619,
      "eval_recall": 0.0843585237258348,
      "eval_runtime": 9.9176,
      "eval_samples_per_second": 520.084,
      "eval_steps_per_second": 16.335,
      "step": 1290
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7594336867332458,
      "learning_rate": 0.0001311983763054864,
      "loss": 0.2709,
      "step": 1935
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8978286157425359,
      "eval_f1": 0.1700787401574803,
      "eval_loss": 0.2664003074169159,
      "eval_mcc": 0.2572444823436495,
      "eval_precision": 0.8181818181818182,
      "eval_recall": 0.09490333919156414,
      "eval_runtime": 9.9289,
      "eval_samples_per_second": 519.496,
      "eval_steps_per_second": 16.316,
      "step": 1935
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.2910077571868896,
      "learning_rate": 0.00011245575111898834,
      "loss": 0.2668,
      "step": 2580
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9005428460643661,
      "eval_f1": 0.31140939597315437,
      "eval_loss": 0.27183133363723755,
      "eval_mcc": 0.3292428995850844,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.2038664323374341,
      "eval_runtime": 9.9232,
      "eval_samples_per_second": 519.793,
      "eval_steps_per_second": 16.325,
      "step": 2580
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.6549971103668213,
      "learning_rate": 9.37131259324903e-05,
      "loss": 0.2638,
      "step": 3225
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9018999612252812,
      "eval_f1": 0.36750000000000005,
      "eval_loss": 0.2644723951816559,
      "eval_mcc": 0.36358633066631857,
      "eval_precision": 0.6363636363636364,
      "eval_recall": 0.2583479789103691,
      "eval_runtime": 9.901,
      "eval_samples_per_second": 520.958,
      "eval_steps_per_second": 16.362,
      "step": 3225
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.2526729106903076,
      "learning_rate": 7.497050074599225e-05,
      "loss": 0.2609,
      "step": 3870
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9013183404420317,
      "eval_f1": 0.30178326474622774,
      "eval_loss": 0.2586013972759247,
      "eval_mcc": 0.32964293560865,
      "eval_precision": 0.6875,
      "eval_recall": 0.19332161687170474,
      "eval_runtime": 9.9174,
      "eval_samples_per_second": 520.096,
      "eval_steps_per_second": 16.335,
      "step": 3870
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.0979706048965454,
      "learning_rate": 5.622787555949417e-05,
      "loss": 0.2579,
      "step": 4515
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9024815820085305,
      "eval_f1": 0.39030303030303026,
      "eval_loss": 0.2604124844074249,
      "eval_mcc": 0.37829052544304637,
      "eval_precision": 0.62890625,
      "eval_recall": 0.28295254833040423,
      "eval_runtime": 9.9054,
      "eval_samples_per_second": 520.728,
      "eval_steps_per_second": 16.355,
      "step": 4515
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.8568099737167358,
      "learning_rate": 3.7485250372996123e-05,
      "loss": 0.2559,
      "step": 5160
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9013183404420317,
      "eval_f1": 0.3328964613368284,
      "eval_loss": 0.2584333121776581,
      "eval_mcc": 0.3434863755500925,
      "eval_precision": 0.654639175257732,
      "eval_recall": 0.22319859402460457,
      "eval_runtime": 9.9128,
      "eval_samples_per_second": 520.34,
      "eval_steps_per_second": 16.343,
      "step": 5160
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.6786758303642273,
      "learning_rate": 1.8742625186498062e-05,
      "loss": 0.2524,
      "step": 5805
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.902287708414114,
      "eval_f1": 0.3420365535248042,
      "eval_loss": 0.2598136067390442,
      "eval_mcc": 0.35281086308273785,
      "eval_precision": 0.6649746192893401,
      "eval_recall": 0.23022847100175747,
      "eval_runtime": 9.9114,
      "eval_samples_per_second": 520.412,
      "eval_steps_per_second": 16.345,
      "step": 5805
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.7956949472427368,
      "learning_rate": 0.0,
      "loss": 0.2523,
      "step": 6450
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.902287708414114,
      "eval_f1": 0.3488372093023256,
      "eval_loss": 0.258432537317276,
      "eval_mcc": 0.3560129584743689,
      "eval_precision": 0.6585365853658537,
      "eval_recall": 0.23725834797891038,
      "eval_runtime": 10.0673,
      "eval_samples_per_second": 512.352,
      "eval_steps_per_second": 16.092,
      "step": 6450
    }
  ],
  "logging_steps": 500,
  "max_steps": 6450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 15869425326000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8497931576459636,
    "learning_rate": 0.0001874262518649806,
    "num_train_epochs": 10,
    "temperature": 17
  }
}