{
  "best_global_step": 1000,
  "best_metric": 0.5416666666666666,
  "best_model_checkpoint": "models/checkpoints/C2/checkpoint-1000",
  "epoch": 14.285714285714286,
  "eval_steps": 200,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 2.232532024383545,
      "learning_rate": 2.45e-05,
      "loss": 0.4005,
      "step": 50
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 3.893237829208374,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 0.3697,
      "step": 100
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 1.5132582187652588,
      "learning_rate": 4.727777777777778e-05,
      "loss": 0.4108,
      "step": 150
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 1.0016103982925415,
      "learning_rate": 4.4500000000000004e-05,
      "loss": 0.4142,
      "step": 200
    },
    {
      "epoch": 2.857142857142857,
      "eval_best_pair_accuracy": 0.5,
      "eval_best_threshold": 0.09933634847402573,
      "eval_loss": 0.43502190709114075,
      "eval_negative_score_mean": 0.6703577458392829,
      "eval_pair_accuracy": 0.45161290322580644,
      "eval_positive_score_mean": 0.611747910330693,
      "eval_roc_auc": 0.3854166666666667,
      "eval_runtime": 1.0717,
      "eval_samples_per_second": 57.849,
      "eval_score_margin_mean": -0.05860983550858989,
      "eval_steps_per_second": 57.849,
      "step": 200
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 2.422534942626953,
      "learning_rate": 4.172222222222222e-05,
      "loss": 0.3177,
      "step": 250
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 3.0248260498046875,
      "learning_rate": 3.894444444444444e-05,
      "loss": 0.3118,
      "step": 300
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.678547739982605,
      "learning_rate": 3.6166666666666674e-05,
      "loss": 0.285,
      "step": 350
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 2.3976991176605225,
      "learning_rate": 3.338888888888889e-05,
      "loss": 0.2414,
      "step": 400
    },
    {
      "epoch": 5.714285714285714,
      "eval_best_pair_accuracy": 0.532258064516129,
      "eval_best_threshold": 0.9629303216934204,
      "eval_loss": 0.441215455532074,
      "eval_negative_score_mean": 0.712315920041874,
      "eval_pair_accuracy": 0.4032258064516129,
      "eval_positive_score_mean": 0.6422562948738535,
      "eval_roc_auc": 0.4322916666666667,
      "eval_runtime": 1.0274,
      "eval_samples_per_second": 60.347,
      "eval_score_margin_mean": -0.07005962516802056,
      "eval_steps_per_second": 60.347,
      "step": 400
    },
    {
      "epoch": 6.428571428571429,
      "grad_norm": 2.6827869415283203,
      "learning_rate": 3.061111111111111e-05,
      "loss": 0.2355,
      "step": 450
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 2.9197797775268555,
      "learning_rate": 2.7833333333333333e-05,
      "loss": 0.2149,
      "step": 500
    },
    {
      "epoch": 7.857142857142857,
      "grad_norm": 3.7313168048858643,
      "learning_rate": 2.5055555555555555e-05,
      "loss": 0.1902,
      "step": 550
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 1.4907903671264648,
      "learning_rate": 2.2277777777777778e-05,
      "loss": 0.1796,
      "step": 600
    },
    {
      "epoch": 8.571428571428571,
      "eval_best_pair_accuracy": 0.5806451612903226,
      "eval_best_threshold": 0.21383827924728394,
      "eval_loss": 0.41378307342529297,
      "eval_negative_score_mean": 0.4248530250042677,
      "eval_pair_accuracy": 0.4838709677419355,
      "eval_positive_score_mean": 0.44192742751911285,
      "eval_roc_auc": 0.5270833333333333,
      "eval_runtime": 1.0315,
      "eval_samples_per_second": 60.107,
      "eval_score_margin_mean": 0.017074402514845155,
      "eval_steps_per_second": 60.107,
      "step": 600
    },
    {
      "epoch": 9.285714285714286,
      "grad_norm": 1.8731682300567627,
      "learning_rate": 1.9500000000000003e-05,
      "loss": 0.1417,
      "step": 650
    },
    {
      "epoch": 10.0,
      "grad_norm": 2.522392511367798,
      "learning_rate": 1.6722222222222222e-05,
      "loss": 0.1315,
      "step": 700
    },
    {
      "epoch": 10.714285714285714,
      "grad_norm": 3.156856060028076,
      "learning_rate": 1.3944444444444446e-05,
      "loss": 0.1164,
      "step": 750
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 3.711172103881836,
      "learning_rate": 1.1166666666666668e-05,
      "loss": 0.1103,
      "step": 800
    },
    {
      "epoch": 11.428571428571429,
      "eval_best_pair_accuracy": 0.5806451612903226,
      "eval_best_threshold": 0.2008795142173767,
      "eval_loss": 0.43724265694618225,
      "eval_negative_score_mean": 0.434278360393364,
      "eval_pair_accuracy": 0.46774193548387094,
      "eval_positive_score_mean": 0.4260471633945902,
      "eval_roc_auc": 0.5,
      "eval_runtime": 1.0607,
      "eval_samples_per_second": 58.451,
      "eval_score_margin_mean": -0.008231196998773782,
      "eval_steps_per_second": 58.451,
      "step": 800
    },
    {
      "epoch": 12.142857142857142,
      "grad_norm": 1.9964823722839355,
      "learning_rate": 8.38888888888889e-06,
      "loss": 0.0953,
      "step": 850
    },
    {
      "epoch": 12.857142857142858,
      "grad_norm": 2.4619648456573486,
      "learning_rate": 5.611111111111112e-06,
      "loss": 0.1526,
      "step": 900
    },
    {
      "epoch": 13.571428571428571,
      "grad_norm": 2.315446615219116,
      "learning_rate": 2.8333333333333335e-06,
      "loss": 0.0924,
      "step": 950
    },
    {
      "epoch": 14.285714285714286,
      "grad_norm": 1.2681019306182861,
      "learning_rate": 5.555555555555556e-08,
      "loss": 0.0886,
      "step": 1000
    },
    {
      "epoch": 14.285714285714286,
      "eval_best_pair_accuracy": 0.5967741935483871,
      "eval_best_threshold": 0.2927139103412628,
      "eval_loss": 0.4323210120201111,
      "eval_negative_score_mean": 0.3917496839421801,
      "eval_pair_accuracy": 0.4838709677419355,
      "eval_positive_score_mean": 0.42978216651827095,
      "eval_roc_auc": 0.5416666666666666,
      "eval_runtime": 1.0486,
      "eval_samples_per_second": 59.129,
      "eval_score_margin_mean": 0.03803248257609082,
      "eval_steps_per_second": 59.129,
      "step": 1000
    }
  ],
  "logging_steps": 50,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}