{
  "best_global_step": 4000,
  "best_metric": 0.7960173130694123,
  "best_model_checkpoint": "./results/run-0/checkpoint-4000",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 2.039576768875122,
      "learning_rate": 1.447537100215313e-06,
      "loss": 0.0,
      "step": 100
    },
    {
      "epoch": 0.1,
      "grad_norm": 4.460484504699707,
      "learning_rate": 2.9096957873014878e-06,
      "loss": 0.0821,
      "step": 200
    },
    {
      "epoch": 0.15,
      "grad_norm": 26.84711456298828,
      "learning_rate": 4.371854474387663e-06,
      "loss": 0.0519,
      "step": 300
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6646966934204102,
      "learning_rate": 5.834013161473838e-06,
      "loss": 0.0,
      "step": 400
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.8385223150253296,
      "learning_rate": 7.296171848560012e-06,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.9234638810157776,
      "learning_rate": 8.758330535646188e-06,
      "loss": 0.0,
      "step": 600
    },
    {
      "epoch": 0.35,
      "grad_norm": 38.24382400512695,
      "learning_rate": 9.9933989317167e-06,
      "loss": 0.0,
      "step": 700
    },
    {
      "epoch": 0.4,
      "grad_norm": 42.83884811401367,
      "learning_rate": 9.94162234536513e-06,
      "loss": 0.0,
      "step": 800
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.8020399808883667,
      "learning_rate": 9.889845759013559e-06,
      "loss": 0.0,
      "step": 900
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.86091148853302,
      "learning_rate": 9.838069172661988e-06,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.16439934074878693,
      "learning_rate": 9.786292586310415e-06,
      "loss": 0.0,
      "step": 1100
    },
    {
      "epoch": 0.6,
      "grad_norm": 2.752115488052368,
      "learning_rate": 9.734515999958846e-06,
      "loss": 0.0,
      "step": 1200
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.5777802467346191,
      "learning_rate": 9.682739413607275e-06,
      "loss": 0.0,
      "step": 1300
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.708035945892334,
      "learning_rate": 9.630962827255703e-06,
      "loss": 0.0598,
      "step": 1400
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3160113990306854,
      "learning_rate": 9.579186240904132e-06,
      "loss": 0.0441,
      "step": 1500
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.2273201942443848,
      "learning_rate": 9.527409654552563e-06,
      "loss": 0.0,
      "step": 1600
    },
    {
      "epoch": 0.85,
      "grad_norm": 29.215551376342773,
      "learning_rate": 9.47563306820099e-06,
      "loss": 0.0,
      "step": 1700
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.18806450068950653,
      "learning_rate": 9.42385648184942e-06,
      "loss": 0.0,
      "step": 1800
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.41723617911338806,
      "learning_rate": 9.372079895497849e-06,
      "loss": 0.0183,
      "step": 1900
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.9550623893737793,
      "learning_rate": 9.32030330914628e-06,
      "loss": 0.0,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_economic_inequality_accuracy": 0.8047858942065491,
      "eval_economic_inequality_f1": 0.7536914695213144,
      "eval_economic_policy_benefits_accuracy": 0.8282828282828283,
      "eval_economic_policy_benefits_f1": 0.7567090529234769,
      "eval_ethnic_boundaries_accuracy": 0.9526143790849673,
      "eval_ethnic_boundaries_f1": 0.9344984933861545,
      "eval_language_policy_accuracy": 0.7058823529411765,
      "eval_language_policy_f1": 0.6071428571428571,
      "eval_loss": NaN,
      "eval_mother_tongue_education_accuracy": 0.8888888888888888,
      "eval_mother_tongue_education_f1": 0.837037037037037,
      "eval_overall_accuracy": 0.8392311075242039,
      "eval_overall_f1": 0.782435168791527,
      "eval_religion_ethnic_identity_accuracy": 0.8549323017408124,
      "eval_religion_ethnic_identity_f1": 0.8055321027383222,
      "eval_runtime": 4.4874,
      "eval_samples_per_second": 891.378,
      "eval_steps_per_second": 55.711,
      "step": 2000
    },
    {
      "epoch": 1.05,
      "grad_norm": 2.762986183166504,
      "learning_rate": 9.268526722794707e-06,
      "loss": 0.0,
      "step": 2100
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.5487369298934937,
      "learning_rate": 9.216750136443136e-06,
      "loss": 0.0,
      "step": 2200
    },
    {
      "epoch": 1.15,
      "grad_norm": 4.76751184463501,
      "learning_rate": 9.164973550091565e-06,
      "loss": 0.0102,
      "step": 2300
    },
    {
      "epoch": 1.2,
      "grad_norm": 2.681756019592285,
      "learning_rate": 9.113196963739995e-06,
      "loss": 0.0,
      "step": 2400
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.24124185740947723,
      "learning_rate": 9.061420377388424e-06,
      "loss": 0.0,
      "step": 2500
    },
    {
      "epoch": 1.3,
      "grad_norm": 1.943154215812683,
      "learning_rate": 9.009643791036853e-06,
      "loss": 0.0,
      "step": 2600
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.28053101897239685,
      "learning_rate": 8.957867204685282e-06,
      "loss": 0.0347,
      "step": 2700
    },
    {
      "epoch": 1.4,
      "grad_norm": 1.7952568531036377,
      "learning_rate": 8.906090618333711e-06,
      "loss": 0.4436,
      "step": 2800
    },
    {
      "epoch": 1.45,
      "grad_norm": 3.859933614730835,
      "learning_rate": 8.85431403198214e-06,
      "loss": 0.0,
      "step": 2900
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.094464898109436,
      "learning_rate": 8.80253744563057e-06,
      "loss": 0.0,
      "step": 3000
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.5670285820960999,
      "learning_rate": 8.750760859278999e-06,
      "loss": 0.0,
      "step": 3100
    },
    {
      "epoch": 1.6,
      "grad_norm": 2.8412418365478516,
      "learning_rate": 8.698984272927428e-06,
      "loss": 0.0,
      "step": 3200
    },
    {
      "epoch": 1.65,
      "grad_norm": 65.82475280761719,
      "learning_rate": 8.647207686575855e-06,
      "loss": 0.0,
      "step": 3300
    },
    {
      "epoch": 1.7,
      "grad_norm": 8.976615905761719,
      "learning_rate": 8.595431100224286e-06,
      "loss": 0.0048,
      "step": 3400
    },
    {
      "epoch": 1.75,
      "grad_norm": 126.71548461914062,
      "learning_rate": 8.543654513872716e-06,
      "loss": 0.0,
      "step": 3500
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.595015823841095,
      "learning_rate": 8.491877927521143e-06,
      "loss": 0.0033,
      "step": 3600
    },
    {
      "epoch": 1.85,
      "grad_norm": 3.98215651512146,
      "learning_rate": 8.440101341169572e-06,
      "loss": 0.1482,
      "step": 3700
    },
    {
      "epoch": 1.9,
      "grad_norm": 3.0417540073394775,
      "learning_rate": 8.388324754818003e-06,
      "loss": 0.0031,
      "step": 3800
    },
    {
      "epoch": 1.95,
      "grad_norm": 27.85649299621582,
      "learning_rate": 8.33654816846643e-06,
      "loss": 0.0589,
      "step": 3900
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.3094232082366943,
      "learning_rate": 8.28477158211486e-06,
      "loss": 0.0,
      "step": 4000
    },
    {
      "epoch": 2.0,
      "eval_economic_inequality_accuracy": 0.8236775818639799,
      "eval_economic_inequality_f1": 0.8161521311449569,
      "eval_economic_policy_benefits_accuracy": 0.835016835016835,
      "eval_economic_policy_benefits_f1": 0.81117677081242,
      "eval_ethnic_boundaries_accuracy": 0.9477124183006536,
      "eval_ethnic_boundaries_f1": 0.9421262334088456,
      "eval_language_policy_accuracy": 0.7058823529411765,
      "eval_language_policy_f1": 0.6071428571428571,
      "eval_loss": NaN,
      "eval_mother_tongue_education_accuracy": 0.7777777777777778,
      "eval_mother_tongue_education_f1": 0.7407407407407408,
      "eval_overall_accuracy": 0.8277126438976526,
      "eval_overall_f1": 0.7960173130694123,
      "eval_religion_ethnic_identity_accuracy": 0.8762088974854932,
      "eval_religion_ethnic_identity_f1": 0.8587651451666537,
      "eval_runtime": 4.4654,
      "eval_samples_per_second": 895.775,
      "eval_steps_per_second": 55.986,
      "step": 4000
    }
  ],
  "logging_steps": 100,
  "max_steps": 20000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8420763303936000.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "gradient_accumulation_steps": 1,
    "learning_rate": 1.0001165419669436e-05,
    "num_train_epochs": 10,
    "per_device_train_batch_size": 8,
    "warmup_steps": 684,
    "weight_decay": 0.07522641729243515
  }
}