{
  "best_metric": 0.872356495468278,
  "best_model_checkpoint": "./Research_paper_MLM_all_CGO_Level_2_Final_Model/checkpoint-3000",
  "epoch": 0.24,
  "eval_steps": 100,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1e-07,
      "loss": 1.1475,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 2e-07,
      "loss": 1.1355,
      "step": 100
    },
    {
      "epoch": 0.01,
      "eval_accuracy": 0.33534743202416917,
      "eval_f1": 0.28883320685557196,
      "eval_loss": 1.125103235244751,
      "eval_precision": 0.25365069358986825,
      "eval_recall": 0.33534743202416917,
      "eval_runtime": 10.8046,
      "eval_samples_per_second": 245.08,
      "eval_steps_per_second": 3.887,
      "step": 100
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.9983974358974358e-07,
      "loss": 1.1142,
      "step": 150
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9967948717948717e-07,
      "loss": 1.0879,
      "step": 200
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.5143504531722054,
      "eval_f1": 0.3499174782499206,
      "eval_loss": 1.0747710466384888,
      "eval_precision": 0.2651511602396808,
      "eval_recall": 0.5143504531722054,
      "eval_runtime": 10.7915,
      "eval_samples_per_second": 245.378,
      "eval_steps_per_second": 3.892,
      "step": 200
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9951923076923076e-07,
      "loss": 1.0575,
      "step": 250
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9935897435897435e-07,
      "loss": 1.0245,
      "step": 300
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.5173716012084593,
      "eval_f1": 0.3848590932649115,
      "eval_loss": 1.0068289041519165,
      "eval_precision": 0.5250854777959997,
      "eval_recall": 0.5173716012084593,
      "eval_runtime": 10.7938,
      "eval_samples_per_second": 245.326,
      "eval_steps_per_second": 3.891,
      "step": 300
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9919871794871794e-07,
      "loss": 0.9823,
      "step": 350
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9903846153846153e-07,
      "loss": 0.9271,
      "step": 400
    },
    {
      "epoch": 0.03,
      "eval_accuracy": 0.5664652567975831,
      "eval_f1": 0.5612424944690789,
      "eval_loss": 0.9066060781478882,
      "eval_precision": 0.5664412983318924,
      "eval_recall": 0.5664652567975831,
      "eval_runtime": 10.8011,
      "eval_samples_per_second": 245.161,
      "eval_steps_per_second": 3.889,
      "step": 400
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9887820512820513e-07,
      "loss": 0.8753,
      "step": 450
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9871794871794872e-07,
      "loss": 0.8326,
      "step": 500
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.597809667673716,
      "eval_f1": 0.5936208877467785,
      "eval_loss": 0.8094537854194641,
      "eval_precision": 0.606349562686828,
      "eval_recall": 0.597809667673716,
      "eval_runtime": 10.7899,
      "eval_samples_per_second": 245.415,
      "eval_steps_per_second": 3.893,
      "step": 500
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.985576923076923e-07,
      "loss": 0.7909,
      "step": 550
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.983974358974359e-07,
      "loss": 0.7634,
      "step": 600
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.6185800604229608,
      "eval_f1": 0.6127776437556779,
      "eval_loss": 0.7459443211555481,
      "eval_precision": 0.6315377091362361,
      "eval_recall": 0.6185800604229608,
      "eval_runtime": 16.0132,
      "eval_samples_per_second": 165.364,
      "eval_steps_per_second": 2.623,
      "step": 600
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.982371794871795e-07,
      "loss": 0.7404,
      "step": 650
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9807692307692308e-07,
      "loss": 0.7334,
      "step": 700
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.6578549848942599,
      "eval_f1": 0.6578987306609153,
      "eval_loss": 0.7136017680168152,
      "eval_precision": 0.6579792469600714,
      "eval_recall": 0.6578549848942599,
      "eval_runtime": 10.8054,
      "eval_samples_per_second": 245.064,
      "eval_steps_per_second": 3.887,
      "step": 700
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9791666666666667e-07,
      "loss": 0.7162,
      "step": 750
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9775641025641026e-07,
      "loss": 0.7052,
      "step": 800
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.68202416918429,
      "eval_f1": 0.6820909311917643,
      "eval_loss": 0.6918750405311584,
      "eval_precision": 0.6825693212542151,
      "eval_recall": 0.68202416918429,
      "eval_runtime": 10.7979,
      "eval_samples_per_second": 245.233,
      "eval_steps_per_second": 3.89,
      "step": 800
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9759615384615385e-07,
      "loss": 0.7014,
      "step": 850
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9743589743589744e-07,
      "loss": 0.6917,
      "step": 900
    },
    {
      "epoch": 0.07,
      "eval_accuracy": 0.7005287009063444,
      "eval_f1": 0.6970978787648872,
      "eval_loss": 0.6726363897323608,
      "eval_precision": 0.7060152626941981,
      "eval_recall": 0.7005287009063444,
      "eval_runtime": 10.7791,
      "eval_samples_per_second": 245.661,
      "eval_steps_per_second": 3.896,
      "step": 900
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9727564102564103e-07,
      "loss": 0.6746,
      "step": 950
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9711538461538462e-07,
      "loss": 0.669,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.722809667673716,
      "eval_f1": 0.7197029874640588,
      "eval_loss": 0.6518763303756714,
      "eval_precision": 0.7290416924889942,
      "eval_recall": 0.722809667673716,
      "eval_runtime": 10.7984,
      "eval_samples_per_second": 245.222,
      "eval_steps_per_second": 3.889,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.969551282051282e-07,
      "loss": 0.6552,
      "step": 1050
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9679487179487178e-07,
      "loss": 0.6476,
      "step": 1100
    },
    {
      "epoch": 0.09,
      "eval_accuracy": 0.7379154078549849,
      "eval_f1": 0.7351061135755582,
      "eval_loss": 0.6282221674919128,
      "eval_precision": 0.74437720980925,
      "eval_recall": 0.7379154078549849,
      "eval_runtime": 10.7935,
      "eval_samples_per_second": 245.333,
      "eval_steps_per_second": 3.891,
      "step": 1100
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9663461538461537e-07,
      "loss": 0.6325,
      "step": 1150
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9647435897435896e-07,
      "loss": 0.6252,
      "step": 1200
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.7492447129909365,
      "eval_f1": 0.7452052048147401,
      "eval_loss": 0.5985825061798096,
      "eval_precision": 0.7607901002801604,
      "eval_recall": 0.7492447129909365,
      "eval_runtime": 10.8114,
      "eval_samples_per_second": 244.926,
      "eval_steps_per_second": 3.885,
      "step": 1200
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9631410256410255e-07,
      "loss": 0.6067,
      "step": 1250
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9615384615384614e-07,
      "loss": 0.5906,
      "step": 1300
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.7775679758308157,
      "eval_f1": 0.7747144726300009,
      "eval_loss": 0.5599903464317322,
      "eval_precision": 0.7875765823937204,
      "eval_recall": 0.7775679758308157,
      "eval_runtime": 10.7946,
      "eval_samples_per_second": 245.309,
      "eval_steps_per_second": 3.891,
      "step": 1300
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9599358974358973e-07,
      "loss": 0.5622,
      "step": 1350
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9583333333333332e-07,
      "loss": 0.5391,
      "step": 1400
    },
    {
      "epoch": 0.11,
      "eval_accuracy": 0.8111782477341389,
      "eval_f1": 0.8088711339987766,
      "eval_loss": 0.5117343664169312,
      "eval_precision": 0.8220985360822193,
      "eval_recall": 0.8111782477341389,
      "eval_runtime": 10.8029,
      "eval_samples_per_second": 245.12,
      "eval_steps_per_second": 3.888,
      "step": 1400
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.956730769230769e-07,
      "loss": 0.5259,
      "step": 1450
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.955128205128205e-07,
      "loss": 0.5089,
      "step": 1500
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.8308157099697885,
      "eval_f1": 0.8288063720795289,
      "eval_loss": 0.46717309951782227,
      "eval_precision": 0.8422125041602981,
      "eval_recall": 0.8308157099697885,
      "eval_runtime": 10.7869,
      "eval_samples_per_second": 245.483,
      "eval_steps_per_second": 3.894,
      "step": 1500
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.953525641025641e-07,
      "loss": 0.4812,
      "step": 1550
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9519230769230768e-07,
      "loss": 0.4581,
      "step": 1600
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.8466767371601208,
      "eval_f1": 0.8452935464575092,
      "eval_loss": 0.42611148953437805,
      "eval_precision": 0.8553191629170025,
      "eval_recall": 0.8466767371601208,
      "eval_runtime": 10.811,
      "eval_samples_per_second": 244.936,
      "eval_steps_per_second": 3.885,
      "step": 1600
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9503205128205127e-07,
      "loss": 0.4265,
      "step": 1650
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9487179487179486e-07,
      "loss": 0.4174,
      "step": 1700
    },
    {
      "epoch": 0.14,
      "eval_accuracy": 0.8591389728096677,
      "eval_f1": 0.8581333993450322,
      "eval_loss": 0.3975095748901367,
      "eval_precision": 0.865914594919593,
      "eval_recall": 0.8591389728096677,
      "eval_runtime": 15.8476,
      "eval_samples_per_second": 167.092,
      "eval_steps_per_second": 2.65,
      "step": 1700
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9471153846153845e-07,
      "loss": 0.3979,
      "step": 1750
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9455128205128204e-07,
      "loss": 0.3897,
      "step": 1800
    },
    {
      "epoch": 0.14,
      "eval_accuracy": 0.8557401812688822,
      "eval_f1": 0.854075463670531,
      "eval_loss": 0.3941803574562073,
      "eval_precision": 0.8677897892001838,
      "eval_recall": 0.8557401812688822,
      "eval_runtime": 10.7957,
      "eval_samples_per_second": 245.283,
      "eval_steps_per_second": 3.89,
      "step": 1800
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9439102564102564e-07,
      "loss": 0.3619,
      "step": 1850
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9423076923076923e-07,
      "loss": 0.3511,
      "step": 1900
    },
    {
      "epoch": 0.15,
      "eval_accuracy": 0.8576283987915407,
      "eval_f1": 0.8557767441533479,
      "eval_loss": 0.3930197060108185,
      "eval_precision": 0.8716354305639,
      "eval_recall": 0.8576283987915407,
      "eval_runtime": 10.8129,
      "eval_samples_per_second": 244.893,
      "eval_steps_per_second": 3.884,
      "step": 1900
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9407051282051282e-07,
      "loss": 0.3537,
      "step": 1950
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.939102564102564e-07,
      "loss": 0.3466,
      "step": 2000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.8606495468277946,
      "eval_f1": 0.8590762411480044,
      "eval_loss": 0.3839268088340759,
      "eval_precision": 0.8725861729431068,
      "eval_recall": 0.8606495468277946,
      "eval_runtime": 10.8113,
      "eval_samples_per_second": 244.93,
      "eval_steps_per_second": 3.885,
      "step": 2000
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9375e-07,
      "loss": 0.3521,
      "step": 2050
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.935897435897436e-07,
      "loss": 0.3389,
      "step": 2100
    },
    {
      "epoch": 0.17,
      "eval_accuracy": 0.8606495468277946,
      "eval_f1": 0.8588371845957175,
      "eval_loss": 0.39029455184936523,
      "eval_precision": 0.8748009772925901,
      "eval_recall": 0.8606495468277946,
      "eval_runtime": 10.802,
      "eval_samples_per_second": 245.14,
      "eval_steps_per_second": 3.888,
      "step": 2100
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9342948717948718e-07,
      "loss": 0.3238,
      "step": 2150
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.9326923076923077e-07,
      "loss": 0.3075,
      "step": 2200
    },
    {
      "epoch": 0.18,
      "eval_accuracy": 0.8602719033232629,
      "eval_f1": 0.8583410955489438,
      "eval_loss": 0.3917655050754547,
      "eval_precision": 0.8754701190846773,
      "eval_recall": 0.8602719033232629,
      "eval_runtime": 10.8054,
      "eval_samples_per_second": 245.064,
      "eval_steps_per_second": 3.887,
      "step": 2200
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.9310897435897436e-07,
      "loss": 0.3102,
      "step": 2250
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.9294871794871795e-07,
      "loss": 0.3101,
      "step": 2300
    },
    {
      "epoch": 0.18,
      "eval_accuracy": 0.8666918429003021,
      "eval_f1": 0.8652087190619578,
      "eval_loss": 0.3828712999820709,
      "eval_precision": 0.878661956862829,
      "eval_recall": 0.8666918429003021,
      "eval_runtime": 15.99,
      "eval_samples_per_second": 165.604,
      "eval_steps_per_second": 2.627,
      "step": 2300
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.9278846153846154e-07,
      "loss": 0.2863,
      "step": 2350
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.9262820512820513e-07,
      "loss": 0.3009,
      "step": 2400
    },
    {
      "epoch": 0.19,
      "eval_accuracy": 0.8621601208459214,
      "eval_f1": 0.8602680217103188,
      "eval_loss": 0.4013417959213257,
      "eval_precision": 0.8773335989772154,
      "eval_recall": 0.8621601208459214,
      "eval_runtime": 10.8074,
      "eval_samples_per_second": 245.018,
      "eval_steps_per_second": 3.886,
      "step": 2400
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.9246794871794872e-07,
      "loss": 0.306,
      "step": 2450
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.9230769230769231e-07,
      "loss": 0.2866,
      "step": 2500
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.8640483383685801,
      "eval_f1": 0.862268075795537,
      "eval_loss": 0.3997219502925873,
      "eval_precision": 0.8784797318905917,
      "eval_recall": 0.8640483383685801,
      "eval_runtime": 10.796,
      "eval_samples_per_second": 245.277,
      "eval_steps_per_second": 3.89,
      "step": 2500
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.921474358974359e-07,
      "loss": 0.2747,
      "step": 2550
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.919871794871795e-07,
      "loss": 0.2856,
      "step": 2600
    },
    {
      "epoch": 0.21,
      "eval_accuracy": 0.8587613293051359,
      "eval_f1": 0.8564860641576998,
      "eval_loss": 0.4260440468788147,
      "eval_precision": 0.8769254342874869,
      "eval_recall": 0.8587613293051359,
      "eval_runtime": 10.798,
      "eval_samples_per_second": 245.231,
      "eval_steps_per_second": 3.89,
      "step": 2600
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.9182692307692309e-07,
      "loss": 0.2907,
      "step": 2650
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.9166666666666668e-07,
      "loss": 0.2871,
      "step": 2700
    },
    {
      "epoch": 0.22,
      "eval_accuracy": 0.8632930513595166,
      "eval_f1": 0.8613280138625302,
      "eval_loss": 0.4157187342643738,
      "eval_precision": 0.8793867110640158,
      "eval_recall": 0.8632930513595166,
      "eval_runtime": 10.8079,
      "eval_samples_per_second": 245.007,
      "eval_steps_per_second": 3.886,
      "step": 2700
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.9150641025641027e-07,
      "loss": 0.2585,
      "step": 2750
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.9134615384615386e-07,
      "loss": 0.2894,
      "step": 2800
    },
    {
      "epoch": 0.22,
      "eval_accuracy": 0.8644259818731118,
      "eval_f1": 0.8624898933835367,
      "eval_loss": 0.4176577627658844,
      "eval_precision": 0.8804547245205405,
      "eval_recall": 0.8644259818731118,
      "eval_runtime": 10.8056,
      "eval_samples_per_second": 245.059,
      "eval_steps_per_second": 3.887,
      "step": 2800
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.9118589743589745e-07,
      "loss": 0.2551,
      "step": 2850
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.9102564102564104e-07,
      "loss": 0.2755,
      "step": 2900
    },
    {
      "epoch": 0.23,
      "eval_accuracy": 0.8659365558912386,
      "eval_f1": 0.8639206748641556,
      "eval_loss": 0.4247892498970032,
      "eval_precision": 0.8830682801454843,
      "eval_recall": 0.8659365558912386,
      "eval_runtime": 10.8115,
      "eval_samples_per_second": 244.924,
      "eval_steps_per_second": 3.885,
      "step": 2900
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.9086538461538463e-07,
      "loss": 0.2494,
      "step": 2950
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.9070512820512822e-07,
      "loss": 0.2508,
      "step": 3000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.872356495468278,
      "eval_f1": 0.8707077161534905,
      "eval_loss": 0.41548070311546326,
      "eval_precision": 0.8869475341602814,
      "eval_recall": 0.872356495468278,
      "eval_runtime": 10.7909,
      "eval_samples_per_second": 245.393,
      "eval_steps_per_second": 3.892,
      "step": 3000
    }
  ],
  "logging_steps": 50,
  "max_steps": 62500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "total_flos": 2.5209554336064e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}