{
  "best_global_step": 1050,
  "best_metric": 0.014976329170167446,
  "best_model_checkpoint": "/content/drive/MyDrive/model/Roberta_Squad/roberta_qa_squad/checkpoint-1050",
  "epoch": 1.124706985466479,
  "eval_steps": 150,
  "global_step": 1200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0009376465072667605,
      "grad_norm": 15.076981544494629,
      "learning_rate": 0.0,
      "loss": 1.4739,
      "step": 1
    },
    {
      "epoch": 0.04688232536333802,
      "grad_norm": 14.212701797485352,
      "learning_rate": 3.598875351452672e-07,
      "loss": 1.4366,
      "step": 50
    },
    {
      "epoch": 0.09376465072667604,
      "grad_norm": 12.904585838317871,
      "learning_rate": 7.347703842549205e-07,
      "loss": 1.1476,
      "step": 100
    },
    {
      "epoch": 0.14064697609001406,
      "grad_norm": 5.23610258102417,
      "learning_rate": 1.1096532333645738e-06,
      "loss": 0.6204,
      "step": 150
    },
    {
      "epoch": 0.14064697609001406,
      "eval_accuracy": 0.9943728018757327,
      "eval_f1": 0.9915671414895327,
      "eval_loss": 0.2603488862514496,
      "eval_runtime": 77.8283,
      "eval_samples_per_second": 109.6,
      "eval_steps_per_second": 3.431,
      "step": 150
    },
    {
      "epoch": 0.1875293014533521,
      "grad_norm": 1.283129096031189,
      "learning_rate": 1.484536082474227e-06,
      "loss": 0.1789,
      "step": 200
    },
    {
      "epoch": 0.23441162681669012,
      "grad_norm": 0.6366870403289795,
      "learning_rate": 1.8594189315838805e-06,
      "loss": 0.066,
      "step": 250
    },
    {
      "epoch": 0.2812939521800281,
      "grad_norm": 0.34226417541503906,
      "learning_rate": 2.2343017806935337e-06,
      "loss": 0.0497,
      "step": 300
    },
    {
      "epoch": 0.2812939521800281,
      "eval_accuracy": 0.9943728018757327,
      "eval_f1": 0.9915671414895327,
      "eval_loss": 0.04236311838030815,
      "eval_runtime": 77.3474,
      "eval_samples_per_second": 110.282,
      "eval_steps_per_second": 3.452,
      "step": 300
    },
    {
      "epoch": 0.32817627754336615,
      "grad_norm": 1.778275489807129,
      "learning_rate": 2.6091846298031867e-06,
      "loss": 0.059,
      "step": 350
    },
    {
      "epoch": 0.3750586029067042,
      "grad_norm": 0.3131299316883087,
      "learning_rate": 2.98406747891284e-06,
      "loss": 0.0414,
      "step": 400
    },
    {
      "epoch": 0.4219409282700422,
      "grad_norm": 0.203192800283432,
      "learning_rate": 3.3589503280224936e-06,
      "loss": 0.0385,
      "step": 450
    },
    {
      "epoch": 0.4219409282700422,
      "eval_accuracy": 0.9943728018757327,
      "eval_f1": 0.9915671414895327,
      "eval_loss": 0.027930891141295433,
      "eval_runtime": 77.8018,
      "eval_samples_per_second": 109.638,
      "eval_steps_per_second": 3.432,
      "step": 450
    },
    {
      "epoch": 0.46882325363338023,
      "grad_norm": 0.5342408418655396,
      "learning_rate": 3.7338331771321466e-06,
      "loss": 0.0392,
      "step": 500
    },
    {
      "epoch": 0.5157055789967182,
      "grad_norm": 0.7715256214141846,
      "learning_rate": 4.1087160262418e-06,
      "loss": 0.0303,
      "step": 550
    },
    {
      "epoch": 0.5625879043600562,
      "grad_norm": 2.367220163345337,
      "learning_rate": 4.4835988753514534e-06,
      "loss": 0.0208,
      "step": 600
    },
    {
      "epoch": 0.5625879043600562,
      "eval_accuracy": 0.9967174677608441,
      "eval_f1": 0.9958795497944622,
      "eval_loss": 0.018059343099594116,
      "eval_runtime": 77.6862,
      "eval_samples_per_second": 109.801,
      "eval_steps_per_second": 3.437,
      "step": 600
    },
    {
      "epoch": 0.6094702297233943,
      "grad_norm": 5.750198841094971,
      "learning_rate": 4.8584817244611065e-06,
      "loss": 0.0211,
      "step": 650
    },
    {
      "epoch": 0.6563525550867323,
      "grad_norm": 1.2362596988677979,
      "learning_rate": 5.2333645735707595e-06,
      "loss": 0.0262,
      "step": 700
    },
    {
      "epoch": 0.7032348804500703,
      "grad_norm": 0.045022640377283096,
      "learning_rate": 5.608247422680413e-06,
      "loss": 0.0142,
      "step": 750
    },
    {
      "epoch": 0.7032348804500703,
      "eval_accuracy": 0.9963657678780774,
      "eval_f1": 0.9953991663366644,
      "eval_loss": 0.0179133377969265,
      "eval_runtime": 78.1703,
      "eval_samples_per_second": 109.121,
      "eval_steps_per_second": 3.416,
      "step": 750
    },
    {
      "epoch": 0.7501172058134083,
      "grad_norm": 0.2589455544948578,
      "learning_rate": 5.983130271790066e-06,
      "loss": 0.0229,
      "step": 800
    },
    {
      "epoch": 0.7969995311767464,
      "grad_norm": 1.0219216346740723,
      "learning_rate": 6.35801312089972e-06,
      "loss": 0.0272,
      "step": 850
    },
    {
      "epoch": 0.8438818565400844,
      "grad_norm": 5.783281326293945,
      "learning_rate": 6.732895970009373e-06,
      "loss": 0.029,
      "step": 900
    },
    {
      "epoch": 0.8438818565400844,
      "eval_accuracy": 0.9964830011723329,
      "eval_f1": 0.9955380069992853,
      "eval_loss": 0.015631183981895447,
      "eval_runtime": 77.5242,
      "eval_samples_per_second": 110.03,
      "eval_steps_per_second": 3.444,
      "step": 900
    },
    {
      "epoch": 0.8907641819034224,
      "grad_norm": 0.03242630138993263,
      "learning_rate": 7.107778819119027e-06,
      "loss": 0.0136,
      "step": 950
    },
    {
      "epoch": 0.9376465072667605,
      "grad_norm": 0.07519914209842682,
      "learning_rate": 7.48266166822868e-06,
      "loss": 0.0167,
      "step": 1000
    },
    {
      "epoch": 0.9845288326300985,
      "grad_norm": 8.501079559326172,
      "learning_rate": 7.857544517338333e-06,
      "loss": 0.0153,
      "step": 1050
    },
    {
      "epoch": 0.9845288326300985,
      "eval_accuracy": 0.9963657678780774,
      "eval_f1": 0.9954965041531265,
      "eval_loss": 0.014976329170167446,
      "eval_runtime": 77.6596,
      "eval_samples_per_second": 109.838,
      "eval_steps_per_second": 3.438,
      "step": 1050
    },
    {
      "epoch": 1.030942334739803,
      "grad_norm": 0.6377618908882141,
      "learning_rate": 8.232427366447986e-06,
      "loss": 0.0152,
      "step": 1100
    },
    {
      "epoch": 1.0778246601031412,
      "grad_norm": 0.05754481628537178,
      "learning_rate": 8.607310215557639e-06,
      "loss": 0.0225,
      "step": 1150
    },
    {
      "epoch": 1.124706985466479,
      "grad_norm": 0.14188943803310394,
      "learning_rate": 8.982193064667292e-06,
      "loss": 0.0163,
      "step": 1200
    },
    {
      "epoch": 1.124706985466479,
      "eval_accuracy": 0.9963657678780774,
      "eval_f1": 0.9954437982952228,
      "eval_loss": 0.017581254243850708,
      "eval_runtime": 77.8986,
      "eval_samples_per_second": 109.501,
      "eval_steps_per_second": 3.428,
      "step": 1200
    }
  ],
  "logging_steps": 50,
  "max_steps": 10670,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 150,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 4,
        "early_stopping_threshold": 0.001
      },
      "attributes": {
        "early_stopping_patience_counter": 2
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.019484367039693e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}