{
  "best_metric": 0.7325463743676223,
  "best_model_checkpoint": "omarmomen/tf_babylm_1/finetune/qnli/checkpoint-3800",
  "epoch": 6.11353711790393,
  "global_step": 4200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29,
      "eval_accuracy": 0.6141732335090637,
      "eval_f1": 0.6941747572815534,
      "eval_loss": 0.6516156196594238,
      "eval_mcc": 0.22499476038642752,
      "eval_runtime": 3.0135,
      "eval_samples_per_second": 758.595,
      "eval_steps_per_second": 94.907,
      "step": 200
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.6146106719970703,
      "eval_f1": 0.6876993973768167,
      "eval_loss": 0.6433870196342468,
      "eval_mcc": 0.22324602394367388,
      "eval_runtime": 2.9954,
      "eval_samples_per_second": 763.173,
      "eval_steps_per_second": 95.48,
      "step": 400
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.636098981077147e-05,
      "loss": 0.6617,
      "step": 500
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.6132983565330505,
      "eval_f1": 0.7005420054200542,
      "eval_loss": 0.6730934977531433,
      "eval_mcc": 0.22735948653592145,
      "eval_runtime": 3.0047,
      "eval_samples_per_second": 760.815,
      "eval_steps_per_second": 95.185,
      "step": 600
    },
    {
      "epoch": 1.16,
      "eval_accuracy": 0.616797924041748,
      "eval_f1": 0.7135382603008502,
      "eval_loss": 0.7076134085655212,
      "eval_mcc": 0.24718358845250435,
      "eval_runtime": 3.0234,
      "eval_samples_per_second": 756.093,
      "eval_steps_per_second": 94.594,
      "step": 800
    },
    {
      "epoch": 1.46,
      "learning_rate": 4.272197962154294e-05,
      "loss": 0.6087,
      "step": 1000
    },
    {
      "epoch": 1.46,
      "eval_accuracy": 0.6307961344718933,
      "eval_f1": 0.7163978494623656,
      "eval_loss": 0.7313140630722046,
      "eval_mcc": 0.2710164987662102,
      "eval_runtime": 3.0006,
      "eval_samples_per_second": 761.857,
      "eval_steps_per_second": 95.315,
      "step": 1000
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.616797924041748,
      "eval_f1": 0.7183279742765273,
      "eval_loss": 0.778058648109436,
      "eval_mcc": 0.2556594651569668,
      "eval_runtime": 3.0192,
      "eval_samples_per_second": 757.142,
      "eval_steps_per_second": 94.726,
      "step": 1200
    },
    {
      "epoch": 2.04,
      "eval_accuracy": 0.6482939720153809,
      "eval_f1": 0.7280108254397835,
      "eval_loss": 0.8302698135375977,
      "eval_mcc": 0.30996246947541667,
      "eval_runtime": 3.0083,
      "eval_samples_per_second": 759.898,
      "eval_steps_per_second": 95.07,
      "step": 1400
    },
    {
      "epoch": 2.18,
      "learning_rate": 3.9082969432314415e-05,
      "loss": 0.5517,
      "step": 1500
    },
    {
      "epoch": 2.33,
      "eval_accuracy": 0.6132983565330505,
      "eval_f1": 0.7182919056724028,
      "eval_loss": 0.9632509350776672,
      "eval_mcc": 0.25168815647322607,
      "eval_runtime": 3.0009,
      "eval_samples_per_second": 761.777,
      "eval_steps_per_second": 95.305,
      "step": 1600
    },
    {
      "epoch": 2.62,
      "eval_accuracy": 0.612860918045044,
      "eval_f1": 0.7193149381541389,
      "eval_loss": 0.9746847748756409,
      "eval_mcc": 0.253685558926296,
      "eval_runtime": 3.0098,
      "eval_samples_per_second": 759.523,
      "eval_steps_per_second": 95.023,
      "step": 1800
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.544395924308588e-05,
      "loss": 0.4786,
      "step": 2000
    },
    {
      "epoch": 2.91,
      "eval_accuracy": 0.6141732335090637,
      "eval_f1": 0.7207093096896771,
      "eval_loss": 0.9755086302757263,
      "eval_mcc": 0.25849446751016225,
      "eval_runtime": 3.0063,
      "eval_samples_per_second": 760.407,
      "eval_steps_per_second": 95.134,
      "step": 2000
    },
    {
      "epoch": 3.2,
      "eval_accuracy": 0.6609798669815063,
      "eval_f1": 0.7322970639032815,
      "eval_loss": 0.9067291021347046,
      "eval_mcc": 0.33185170807279407,
      "eval_runtime": 3.0009,
      "eval_samples_per_second": 761.769,
      "eval_steps_per_second": 95.304,
      "step": 2200
    },
    {
      "epoch": 3.49,
      "eval_accuracy": 0.6303586959838867,
      "eval_f1": 0.7248453272549658,
      "eval_loss": 1.0302561521530151,
      "eval_mcc": 0.28386334628294807,
      "eval_runtime": 3.0233,
      "eval_samples_per_second": 756.116,
      "eval_steps_per_second": 94.597,
      "step": 2400
    },
    {
      "epoch": 3.64,
      "learning_rate": 3.1804949053857355e-05,
      "loss": 0.3795,
      "step": 2500
    },
    {
      "epoch": 3.78,
      "eval_accuracy": 0.6272965669631958,
      "eval_f1": 0.7233766233766233,
      "eval_loss": 1.073946237564087,
      "eval_mcc": 0.2776317996888772,
      "eval_runtime": 3.0269,
      "eval_samples_per_second": 755.222,
      "eval_steps_per_second": 94.485,
      "step": 2600
    },
    {
      "epoch": 4.08,
      "eval_accuracy": 0.6242344975471497,
      "eval_f1": 0.7193727540019601,
      "eval_loss": 1.393174648284912,
      "eval_mcc": 0.26651346871760156,
      "eval_runtime": 3.0247,
      "eval_samples_per_second": 755.772,
      "eval_steps_per_second": 94.554,
      "step": 2800
    },
    {
      "epoch": 4.37,
      "learning_rate": 2.816593886462882e-05,
      "loss": 0.3087,
      "step": 3000
    },
    {
      "epoch": 4.37,
      "eval_accuracy": 0.6299212574958801,
      "eval_f1": 0.7247885491216656,
      "eval_loss": 1.3153369426727295,
      "eval_mcc": 0.28329167906840397,
      "eval_runtime": 3.0213,
      "eval_samples_per_second": 756.639,
      "eval_steps_per_second": 94.663,
      "step": 3000
    },
    {
      "epoch": 4.66,
      "eval_accuracy": 0.6356080770492554,
      "eval_f1": 0.7273322422258592,
      "eval_loss": 1.3374916315078735,
      "eval_mcc": 0.29432211381962475,
      "eval_runtime": 3.0207,
      "eval_samples_per_second": 756.78,
      "eval_steps_per_second": 94.68,
      "step": 3200
    },
    {
      "epoch": 4.95,
      "eval_accuracy": 0.6281715035438538,
      "eval_f1": 0.723846653671215,
      "eval_loss": 1.2929773330688477,
      "eval_mcc": 0.279522957371612,
      "eval_runtime": 3.0073,
      "eval_samples_per_second": 760.149,
      "eval_steps_per_second": 95.102,
      "step": 3400
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.452692867540029e-05,
      "loss": 0.261,
      "step": 3500
    },
    {
      "epoch": 5.24,
      "eval_accuracy": 0.6552931070327759,
      "eval_f1": 0.7293956043956044,
      "eval_loss": 1.3341577053070068,
      "eval_mcc": 0.3208142617256846,
      "eval_runtime": 3.0069,
      "eval_samples_per_second": 760.245,
      "eval_steps_per_second": 95.114,
      "step": 3600
    },
    {
      "epoch": 5.53,
      "eval_accuracy": 0.6531058549880981,
      "eval_f1": 0.7325463743676223,
      "eval_loss": 1.4561353921890259,
      "eval_mcc": 0.32254298037112755,
      "eval_runtime": 3.0074,
      "eval_samples_per_second": 760.119,
      "eval_steps_per_second": 95.098,
      "step": 3800
    },
    {
      "epoch": 5.82,
      "learning_rate": 2.088791848617176e-05,
      "loss": 0.1808,
      "step": 4000
    },
    {
      "epoch": 5.82,
      "eval_accuracy": 0.6579177379608154,
      "eval_f1": 0.7292243767313019,
      "eval_loss": 1.3927106857299805,
      "eval_mcc": 0.3241513466674666,
      "eval_runtime": 3.0083,
      "eval_samples_per_second": 759.895,
      "eval_steps_per_second": 95.07,
      "step": 4000
    },
    {
      "epoch": 6.11,
      "eval_accuracy": 0.6482939720153809,
      "eval_f1": 0.7269021739130435,
      "eval_loss": 1.718584656715393,
      "eval_mcc": 0.30843932453221495,
      "eval_runtime": 3.0225,
      "eval_samples_per_second": 756.317,
      "eval_steps_per_second": 94.622,
      "step": 4200
    },
    {
      "epoch": 6.11,
      "step": 4200,
      "total_flos": 7367156844493824.0,
      "train_loss": 0.4156740043276832,
      "train_runtime": 776.4968,
      "train_samples_per_second": 565.579,
      "train_steps_per_second": 8.847
    }
  ],
  "max_steps": 6870,
  "num_train_epochs": 10,
  "total_flos": 7367156844493824.0,
  "trial_name": null,
  "trial_params": null
}