{
  "best_metric": 0.8934987662530843,
  "best_model_checkpoint": "./output//roberta-large_ipc1_C_5_32_5e-6_0.01_0.06_07-07-22_23-17/checkpoint-38000",
  "epoch": 0.4979296608836941,
  "global_step": 38000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 4.3677658877484174e-07,
      "loss": 0.3698,
      "step": 2000
    },
    {
      "epoch": 0.03,
      "eval_accuracy": 0.0003641602938615241,
      "eval_f1": 0.0007814160080609232,
      "eval_loss": 0.1831192523241043,
      "eval_roc_auc": 0.5001946379048023,
      "eval_runtime": 3712.996,
      "eval_samples_per_second": 34.021,
      "eval_steps_per_second": 2.126,
      "step": 2000
    },
    {
      "epoch": 0.05,
      "learning_rate": 8.735531775496835e-07,
      "loss": 0.1415,
      "step": 4000
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.6030652796909387,
      "eval_f1": 0.7397382281617269,
      "eval_loss": 0.09564058482646942,
      "eval_roc_auc": 0.8112905369107336,
      "eval_runtime": 3711.74,
      "eval_samples_per_second": 34.032,
      "eval_steps_per_second": 2.127,
      "step": 4000
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.3103297663245252e-06,
      "loss": 0.0855,
      "step": 6000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.7227869345619785,
      "eval_f1": 0.8181041339128826,
      "eval_loss": 0.06705362349748611,
      "eval_roc_auc": 0.8763586156819746,
      "eval_runtime": 3713.5565,
      "eval_samples_per_second": 34.015,
      "eval_steps_per_second": 2.126,
      "step": 6000
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.747106355099367e-06,
      "loss": 0.0643,
      "step": 8000
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.7567884228692665,
      "eval_f1": 0.841935061915005,
      "eval_loss": 0.05476009473204613,
      "eval_roc_auc": 0.8945397223068867,
      "eval_runtime": 3712.7654,
      "eval_samples_per_second": 34.023,
      "eval_steps_per_second": 2.126,
      "step": 8000
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.1838829438742083e-06,
      "loss": 0.055,
      "step": 10000
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.7719169081207745,
      "eval_f1": 0.8556000272324324,
      "eval_loss": 0.0483216755092144,
      "eval_roc_auc": 0.9065874885210222,
      "eval_runtime": 3715.8841,
      "eval_samples_per_second": 33.994,
      "eval_steps_per_second": 2.125,
      "step": 10000
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.6206595326490503e-06,
      "loss": 0.0493,
      "step": 12000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.779200113998005,
      "eval_f1": 0.8616441123005455,
      "eval_loss": 0.045345865190029144,
      "eval_roc_auc": 0.9139128753285614,
      "eval_runtime": 3715.5277,
      "eval_samples_per_second": 33.997,
      "eval_steps_per_second": 2.125,
      "step": 12000
    },
    {
      "epoch": 0.18,
      "learning_rate": 3.057436121423892e-06,
      "loss": 0.0462,
      "step": 14000
    },
    {
      "epoch": 0.18,
      "eval_accuracy": 0.7865545686283824,
      "eval_f1": 0.8685937966212471,
      "eval_loss": 0.04221072793006897,
      "eval_roc_auc": 0.9165307568850345,
      "eval_runtime": 3716.006,
      "eval_samples_per_second": 33.993,
      "eval_steps_per_second": 2.125,
      "step": 14000
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.494212710198734e-06,
      "loss": 0.0437,
      "step": 16000
    },
    {
      "epoch": 0.21,
      "eval_accuracy": 0.7878212131287703,
      "eval_f1": 0.8674021275833662,
      "eval_loss": 0.04178840294480324,
      "eval_roc_auc": 0.9127150639723773,
      "eval_runtime": 3717.2766,
      "eval_samples_per_second": 33.981,
      "eval_steps_per_second": 2.124,
      "step": 16000
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.9309892989735755e-06,
      "loss": 0.0416,
      "step": 18000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.7823746417771023,
      "eval_f1": 0.8717482944145543,
      "eval_loss": 0.04093621298670769,
      "eval_roc_auc": 0.9277326030672447,
      "eval_runtime": 3718.3054,
      "eval_samples_per_second": 33.972,
      "eval_steps_per_second": 2.123,
      "step": 18000
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.367765887748417e-06,
      "loss": 0.0401,
      "step": 20000
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.7977643724568153,
      "eval_f1": 0.8785716806040718,
      "eval_loss": 0.03799275681376457,
      "eval_roc_auc": 0.9242562472497354,
      "eval_runtime": 3718.544,
      "eval_samples_per_second": 33.97,
      "eval_steps_per_second": 2.123,
      "step": 20000
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.804542476523259e-06,
      "loss": 0.0396,
      "step": 22000
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.8008518184265109,
      "eval_f1": 0.8802290810342459,
      "eval_loss": 0.0379040502011776,
      "eval_roc_auc": 0.9258168171723399,
      "eval_runtime": 3718.4418,
      "eval_samples_per_second": 33.971,
      "eval_steps_per_second": 2.123,
      "step": 22000
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.984596512260062e-06,
      "loss": 0.0387,
      "step": 24000
    },
    {
      "epoch": 0.31,
      "eval_accuracy": 0.802775534761475,
      "eval_f1": 0.882195878143154,
      "eval_loss": 0.036688029766082764,
      "eval_roc_auc": 0.9262638395674476,
      "eval_runtime": 3717.6016,
      "eval_samples_per_second": 33.978,
      "eval_steps_per_second": 2.124,
      "step": 24000
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.956716896441168e-06,
      "loss": 0.0377,
      "step": 26000
    },
    {
      "epoch": 0.34,
      "eval_accuracy": 0.8030446967178074,
      "eval_f1": 0.8856592114236406,
      "eval_loss": 0.03543846309185028,
      "eval_roc_auc": 0.9335204908137789,
      "eval_runtime": 3718.4609,
      "eval_samples_per_second": 33.971,
      "eval_steps_per_second": 2.123,
      "step": 26000
    },
    {
      "epoch": 0.37,
      "learning_rate": 4.9288372806222735e-06,
      "loss": 0.0366,
      "step": 28000
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.8094650010291486,
      "eval_f1": 0.8885554494611922,
      "eval_loss": 0.03468663617968559,
      "eval_roc_auc": 0.9329188925134719,
      "eval_runtime": 3718.3781,
      "eval_samples_per_second": 33.971,
      "eval_steps_per_second": 2.123,
      "step": 28000
    },
    {
      "epoch": 0.39,
      "learning_rate": 4.900957664803379e-06,
      "loss": 0.0359,
      "step": 30000
    },
    {
      "epoch": 0.39,
      "eval_accuracy": 0.8062508906094144,
      "eval_f1": 0.8869252133833476,
      "eval_loss": 0.03476712107658386,
      "eval_roc_auc": 0.9315236503482228,
      "eval_runtime": 3719.6636,
      "eval_samples_per_second": 33.96,
      "eval_steps_per_second": 2.123,
      "step": 30000
    },
    {
      "epoch": 0.42,
      "learning_rate": 4.873078048984485e-06,
      "loss": 0.0356,
      "step": 32000
    },
    {
      "epoch": 0.42,
      "eval_accuracy": 0.8126395288082459,
      "eval_f1": 0.8891505277035555,
      "eval_loss": 0.03411931172013283,
      "eval_roc_auc": 0.9300795994141317,
      "eval_runtime": 3718.944,
      "eval_samples_per_second": 33.966,
      "eval_steps_per_second": 2.123,
      "step": 32000
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.845198433165592e-06,
      "loss": 0.035,
      "step": 34000
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.8100587406387055,
      "eval_f1": 0.8897779847957649,
      "eval_loss": 0.033911969512701035,
      "eval_roc_auc": 0.9360625989159235,
      "eval_runtime": 3718.5781,
      "eval_samples_per_second": 33.969,
      "eval_steps_per_second": 2.123,
      "step": 34000
    },
    {
      "epoch": 0.47,
      "learning_rate": 4.817318817346697e-06,
      "loss": 0.0353,
      "step": 36000
    },
    {
      "epoch": 0.47,
      "eval_accuracy": 0.8157190582498139,
      "eval_f1": 0.8927496417045404,
      "eval_loss": 0.03287062793970108,
      "eval_roc_auc": 0.9343416214326565,
      "eval_runtime": 3719.9265,
      "eval_samples_per_second": 33.957,
      "eval_steps_per_second": 2.122,
      "step": 36000
    },
    {
      "epoch": 0.5,
      "learning_rate": 4.789439201527804e-06,
      "loss": 0.0345,
      "step": 38000
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.8162652986906063,
      "eval_f1": 0.8934987662530843,
      "eval_loss": 0.0325697623193264,
      "eval_roc_auc": 0.935207039689335,
      "eval_runtime": 3719.4103,
      "eval_samples_per_second": 33.962,
      "eval_steps_per_second": 2.123,
      "step": 38000
    }
  ],
  "max_steps": 381580,
  "num_train_epochs": 5,
  "total_flos": 1.133301287682048e+18,
  "trial_name": null,
  "trial_params": null
}