{
  "best_metric": 0.6819557547569275,
  "best_model_checkpoint": "prostate-mri-T2w-v01/checkpoint-96",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 96,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0625,
      "grad_norm": 6.9218926429748535,
      "learning_rate": 6.25e-07,
      "loss": 0.7026,
      "step": 1
    },
    {
      "epoch": 0.125,
      "grad_norm": 8.463187217712402,
      "learning_rate": 1.25e-06,
      "loss": 0.6897,
      "step": 2
    },
    {
      "epoch": 0.1875,
      "grad_norm": 7.885867118835449,
      "learning_rate": 1.875e-06,
      "loss": 0.6738,
      "step": 3
    },
    {
      "epoch": 0.25,
      "grad_norm": 12.486954689025879,
      "learning_rate": 2.5e-06,
      "loss": 0.6963,
      "step": 4
    },
    {
      "epoch": 0.3125,
      "grad_norm": 9.765836715698242,
      "learning_rate": 3.125e-06,
      "loss": 0.6987,
      "step": 5
    },
    {
      "epoch": 0.375,
      "grad_norm": 8.386995315551758,
      "learning_rate": 3.75e-06,
      "loss": 0.7004,
      "step": 6
    },
    {
      "epoch": 0.4375,
      "grad_norm": 9.164388656616211,
      "learning_rate": 4.375e-06,
      "loss": 0.668,
      "step": 7
    },
    {
      "epoch": 0.5,
      "grad_norm": 4.771695137023926,
      "learning_rate": 5e-06,
      "loss": 0.6904,
      "step": 8
    },
    {
      "epoch": 0.5625,
      "grad_norm": 7.426454544067383,
      "learning_rate": 5.625e-06,
      "loss": 0.7024,
      "step": 9
    },
    {
      "epoch": 0.625,
      "grad_norm": 5.934999465942383,
      "learning_rate": 6.25e-06,
      "loss": 0.6982,
      "step": 10
    },
    {
      "epoch": 0.6875,
      "grad_norm": 5.554957389831543,
      "learning_rate": 6.875000000000001e-06,
      "loss": 0.698,
      "step": 11
    },
    {
      "epoch": 0.75,
      "grad_norm": 5.586301326751709,
      "learning_rate": 7.5e-06,
      "loss": 0.686,
      "step": 12
    },
    {
      "epoch": 0.8125,
      "grad_norm": 6.572750091552734,
      "learning_rate": 8.125000000000001e-06,
      "loss": 0.6807,
      "step": 13
    },
    {
      "epoch": 0.875,
      "grad_norm": 5.920419692993164,
      "learning_rate": 8.75e-06,
      "loss": 0.6921,
      "step": 14
    },
    {
      "epoch": 0.9375,
      "grad_norm": 10.7252779006958,
      "learning_rate": 9.375000000000001e-06,
      "loss": 0.6975,
      "step": 15
    },
    {
      "epoch": 1.0,
      "grad_norm": 21.941551208496094,
      "learning_rate": 1e-05,
      "loss": 0.6868,
      "step": 16
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5,
      "eval_auc": 0.4831932773109244,
      "eval_f1": 0.4918032786885246,
      "eval_loss": 0.6948714852333069,
      "eval_precision": 0.5555555555555556,
      "eval_recall": 0.4411764705882353,
      "eval_runtime": 1.0055,
      "eval_samples_per_second": 61.659,
      "eval_steps_per_second": 1.989,
      "step": 16
    },
    {
      "epoch": 1.0625,
      "grad_norm": 7.140624046325684,
      "learning_rate": 1.0625e-05,
      "loss": 0.6921,
      "step": 17
    },
    {
      "epoch": 1.125,
      "grad_norm": 7.36794900894165,
      "learning_rate": 1.125e-05,
      "loss": 0.6877,
      "step": 18
    },
    {
      "epoch": 1.1875,
      "grad_norm": 10.255731582641602,
      "learning_rate": 1.1875e-05,
      "loss": 0.6946,
      "step": 19
    },
    {
      "epoch": 1.25,
      "grad_norm": 7.5590667724609375,
      "learning_rate": 1.25e-05,
      "loss": 0.6873,
      "step": 20
    },
    {
      "epoch": 1.3125,
      "grad_norm": 10.777117729187012,
      "learning_rate": 1.3125e-05,
      "loss": 0.6907,
      "step": 21
    },
    {
      "epoch": 1.375,
      "grad_norm": 6.053137302398682,
      "learning_rate": 1.3750000000000002e-05,
      "loss": 0.677,
      "step": 22
    },
    {
      "epoch": 1.4375,
      "grad_norm": 5.271788120269775,
      "learning_rate": 1.4374999999999999e-05,
      "loss": 0.6919,
      "step": 23
    },
    {
      "epoch": 1.5,
      "grad_norm": 6.248932838439941,
      "learning_rate": 1.5e-05,
      "loss": 0.6938,
      "step": 24
    },
    {
      "epoch": 1.5625,
      "grad_norm": 7.127931118011475,
      "learning_rate": 1.5625e-05,
      "loss": 0.6804,
      "step": 25
    },
    {
      "epoch": 1.625,
      "grad_norm": 5.395265579223633,
      "learning_rate": 1.6250000000000002e-05,
      "loss": 0.6938,
      "step": 26
    },
    {
      "epoch": 1.6875,
      "grad_norm": 8.488815307617188,
      "learning_rate": 1.6875000000000004e-05,
      "loss": 0.6909,
      "step": 27
    },
    {
      "epoch": 1.75,
      "grad_norm": 6.858600616455078,
      "learning_rate": 1.75e-05,
      "loss": 0.6907,
      "step": 28
    },
    {
      "epoch": 1.8125,
      "grad_norm": 7.674286842346191,
      "learning_rate": 1.8125e-05,
      "loss": 0.6912,
      "step": 29
    },
    {
      "epoch": 1.875,
      "grad_norm": 6.898157596588135,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.6782,
      "step": 30
    },
    {
      "epoch": 1.9375,
      "grad_norm": 5.758520603179932,
      "learning_rate": 1.9375e-05,
      "loss": 0.7087,
      "step": 31
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.98084831237793,
      "learning_rate": 2e-05,
      "loss": 0.7129,
      "step": 32
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6290322580645161,
      "eval_auc": 0.5273109243697479,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.6898942589759827,
      "eval_precision": 0.6571428571428571,
      "eval_recall": 0.6764705882352942,
      "eval_runtime": 0.9907,
      "eval_samples_per_second": 62.581,
      "eval_steps_per_second": 2.019,
      "step": 32
    },
    {
      "epoch": 2.0625,
      "grad_norm": 5.439036846160889,
      "learning_rate": 2.0625e-05,
      "loss": 0.6921,
      "step": 33
    },
    {
      "epoch": 2.125,
      "grad_norm": 5.089821815490723,
      "learning_rate": 2.125e-05,
      "loss": 0.7075,
      "step": 34
    },
    {
      "epoch": 2.1875,
      "grad_norm": 5.559844017028809,
      "learning_rate": 2.1875e-05,
      "loss": 0.6758,
      "step": 35
    },
    {
      "epoch": 2.25,
      "grad_norm": 7.2915496826171875,
      "learning_rate": 2.25e-05,
      "loss": 0.6877,
      "step": 36
    },
    {
      "epoch": 2.3125,
      "grad_norm": 6.99595308303833,
      "learning_rate": 2.3125000000000003e-05,
      "loss": 0.6921,
      "step": 37
    },
    {
      "epoch": 2.375,
      "grad_norm": 5.015499114990234,
      "learning_rate": 2.375e-05,
      "loss": 0.6711,
      "step": 38
    },
    {
      "epoch": 2.4375,
      "grad_norm": 6.5360517501831055,
      "learning_rate": 2.4375e-05,
      "loss": 0.6804,
      "step": 39
    },
    {
      "epoch": 2.5,
      "grad_norm": 6.66691780090332,
      "learning_rate": 2.5e-05,
      "loss": 0.6912,
      "step": 40
    },
    {
      "epoch": 2.5625,
      "grad_norm": 10.940932273864746,
      "learning_rate": 2.5625e-05,
      "loss": 0.699,
      "step": 41
    },
    {
      "epoch": 2.625,
      "grad_norm": 4.8793816566467285,
      "learning_rate": 2.625e-05,
      "loss": 0.6914,
      "step": 42
    },
    {
      "epoch": 2.6875,
      "grad_norm": 5.5832343101501465,
      "learning_rate": 2.6875e-05,
      "loss": 0.6953,
      "step": 43
    },
    {
      "epoch": 2.75,
      "grad_norm": 13.049569129943848,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 0.6973,
      "step": 44
    },
    {
      "epoch": 2.8125,
      "grad_norm": 8.290694236755371,
      "learning_rate": 2.8125000000000003e-05,
      "loss": 0.6978,
      "step": 45
    },
    {
      "epoch": 2.875,
      "grad_norm": 8.824386596679688,
      "learning_rate": 2.8749999999999997e-05,
      "loss": 0.689,
      "step": 46
    },
    {
      "epoch": 2.9375,
      "grad_norm": 8.694557189941406,
      "learning_rate": 2.9375000000000003e-05,
      "loss": 0.7039,
      "step": 47
    },
    {
      "epoch": 3.0,
      "grad_norm": 13.123662948608398,
      "learning_rate": 3e-05,
      "loss": 0.7129,
      "step": 48
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5806451612903226,
      "eval_auc": 0.5189075630252101,
      "eval_f1": 0.6976744186046512,
      "eval_loss": 0.6869328618049622,
      "eval_precision": 0.5769230769230769,
      "eval_recall": 0.8823529411764706,
      "eval_runtime": 0.9817,
      "eval_samples_per_second": 63.153,
      "eval_steps_per_second": 2.037,
      "step": 48
    },
    {
      "epoch": 3.0625,
      "grad_norm": 8.405984878540039,
      "learning_rate": 3.0625000000000006e-05,
      "loss": 0.6912,
      "step": 49
    },
    {
      "epoch": 3.125,
      "grad_norm": 11.393486976623535,
      "learning_rate": 3.125e-05,
      "loss": 0.679,
      "step": 50
    },
    {
      "epoch": 3.1875,
      "grad_norm": 7.367499828338623,
      "learning_rate": 3.1875e-05,
      "loss": 0.7205,
      "step": 51
    },
    {
      "epoch": 3.25,
      "grad_norm": 7.851259231567383,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 0.6877,
      "step": 52
    },
    {
      "epoch": 3.3125,
      "grad_norm": 15.551215171813965,
      "learning_rate": 3.3125e-05,
      "loss": 0.6865,
      "step": 53
    },
    {
      "epoch": 3.375,
      "grad_norm": 6.169222354888916,
      "learning_rate": 3.375000000000001e-05,
      "loss": 0.6853,
      "step": 54
    },
    {
      "epoch": 3.4375,
      "grad_norm": 5.225166320800781,
      "learning_rate": 3.4375e-05,
      "loss": 0.6951,
      "step": 55
    },
    {
      "epoch": 3.5,
      "grad_norm": 6.697177886962891,
      "learning_rate": 3.5e-05,
      "loss": 0.6738,
      "step": 56
    },
    {
      "epoch": 3.5625,
      "grad_norm": 8.565425872802734,
      "learning_rate": 3.5625000000000005e-05,
      "loss": 0.6936,
      "step": 57
    },
    {
      "epoch": 3.625,
      "grad_norm": 4.674345970153809,
      "learning_rate": 3.625e-05,
      "loss": 0.7007,
      "step": 58
    },
    {
      "epoch": 3.6875,
      "grad_norm": 6.7598371505737305,
      "learning_rate": 3.6875e-05,
      "loss": 0.7031,
      "step": 59
    },
    {
      "epoch": 3.75,
      "grad_norm": 5.33363151550293,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.6978,
      "step": 60
    },
    {
      "epoch": 3.8125,
      "grad_norm": 4.975074291229248,
      "learning_rate": 3.8125e-05,
      "loss": 0.6992,
      "step": 61
    },
    {
      "epoch": 3.875,
      "grad_norm": 5.9460673332214355,
      "learning_rate": 3.875e-05,
      "loss": 0.698,
      "step": 62
    },
    {
      "epoch": 3.9375,
      "grad_norm": 6.137637615203857,
      "learning_rate": 3.9375e-05,
      "loss": 0.6721,
      "step": 63
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.683951377868652,
      "learning_rate": 4e-05,
      "loss": 0.6458,
      "step": 64
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5806451612903226,
      "eval_auc": 0.4690126050420168,
      "eval_f1": 0.675,
      "eval_loss": 0.6907132267951965,
      "eval_precision": 0.5869565217391305,
      "eval_recall": 0.7941176470588235,
      "eval_runtime": 1.0012,
      "eval_samples_per_second": 61.926,
      "eval_steps_per_second": 1.998,
      "step": 64
    },
    {
      "epoch": 4.0625,
      "grad_norm": 6.240651607513428,
      "learning_rate": 4.0625000000000005e-05,
      "loss": 0.6936,
      "step": 65
    },
    {
      "epoch": 4.125,
      "grad_norm": 6.195248603820801,
      "learning_rate": 4.125e-05,
      "loss": 0.7053,
      "step": 66
    },
    {
      "epoch": 4.1875,
      "grad_norm": 5.479919910430908,
      "learning_rate": 4.1875e-05,
      "loss": 0.6921,
      "step": 67
    },
    {
      "epoch": 4.25,
      "grad_norm": 6.832111835479736,
      "learning_rate": 4.25e-05,
      "loss": 0.6721,
      "step": 68
    },
    {
      "epoch": 4.3125,
      "grad_norm": 7.457818984985352,
      "learning_rate": 4.3125000000000005e-05,
      "loss": 0.6838,
      "step": 69
    },
    {
      "epoch": 4.375,
      "grad_norm": 9.509450912475586,
      "learning_rate": 4.375e-05,
      "loss": 0.6978,
      "step": 70
    },
    {
      "epoch": 4.4375,
      "grad_norm": 4.208181381225586,
      "learning_rate": 4.4375e-05,
      "loss": 0.6863,
      "step": 71
    },
    {
      "epoch": 4.5,
      "grad_norm": 5.488190174102783,
      "learning_rate": 4.5e-05,
      "loss": 0.6948,
      "step": 72
    },
    {
      "epoch": 4.5625,
      "grad_norm": 6.613465309143066,
      "learning_rate": 4.5625e-05,
      "loss": 0.7004,
      "step": 73
    },
    {
      "epoch": 4.625,
      "grad_norm": 7.014777660369873,
      "learning_rate": 4.6250000000000006e-05,
      "loss": 0.6562,
      "step": 74
    },
    {
      "epoch": 4.6875,
      "grad_norm": 9.429577827453613,
      "learning_rate": 4.6875e-05,
      "loss": 0.7031,
      "step": 75
    },
    {
      "epoch": 4.75,
      "grad_norm": 4.446032524108887,
      "learning_rate": 4.75e-05,
      "loss": 0.6848,
      "step": 76
    },
    {
      "epoch": 4.8125,
      "grad_norm": 8.153448104858398,
      "learning_rate": 4.8125000000000004e-05,
      "loss": 0.699,
      "step": 77
    },
    {
      "epoch": 4.875,
      "grad_norm": 7.079260349273682,
      "learning_rate": 4.875e-05,
      "loss": 0.6843,
      "step": 78
    },
    {
      "epoch": 4.9375,
      "grad_norm": 4.8226213455200195,
      "learning_rate": 4.937500000000001e-05,
      "loss": 0.6763,
      "step": 79
    },
    {
      "epoch": 5.0,
      "grad_norm": 16.139135360717773,
      "learning_rate": 5e-05,
      "loss": 0.651,
      "step": 80
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5,
      "eval_auc": 0.5246848739495799,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.6864917278289795,
      "eval_precision": 0.5254237288135594,
      "eval_recall": 0.9117647058823529,
      "eval_runtime": 1.0013,
      "eval_samples_per_second": 61.918,
      "eval_steps_per_second": 1.997,
      "step": 80
    },
    {
      "epoch": 5.0625,
      "grad_norm": 6.197685718536377,
      "learning_rate": 4.993055555555556e-05,
      "loss": 0.678,
      "step": 81
    },
    {
      "epoch": 5.125,
      "grad_norm": 5.559277057647705,
      "learning_rate": 4.986111111111111e-05,
      "loss": 0.6677,
      "step": 82
    },
    {
      "epoch": 5.1875,
      "grad_norm": 11.058768272399902,
      "learning_rate": 4.979166666666667e-05,
      "loss": 0.6851,
      "step": 83
    },
    {
      "epoch": 5.25,
      "grad_norm": 7.209768772125244,
      "learning_rate": 4.972222222222223e-05,
      "loss": 0.6838,
      "step": 84
    },
    {
      "epoch": 5.3125,
      "grad_norm": 5.192998886108398,
      "learning_rate": 4.965277777777778e-05,
      "loss": 0.6899,
      "step": 85
    },
    {
      "epoch": 5.375,
      "grad_norm": 6.706368923187256,
      "learning_rate": 4.958333333333334e-05,
      "loss": 0.6956,
      "step": 86
    },
    {
      "epoch": 5.4375,
      "grad_norm": 8.244791030883789,
      "learning_rate": 4.951388888888889e-05,
      "loss": 0.6785,
      "step": 87
    },
    {
      "epoch": 5.5,
      "grad_norm": 10.724773406982422,
      "learning_rate": 4.9444444444444446e-05,
      "loss": 0.6863,
      "step": 88
    },
    {
      "epoch": 5.5625,
      "grad_norm": 5.7804484367370605,
      "learning_rate": 4.937500000000001e-05,
      "loss": 0.6404,
      "step": 89
    },
    {
      "epoch": 5.625,
      "grad_norm": 8.586164474487305,
      "learning_rate": 4.930555555555556e-05,
      "loss": 0.6904,
      "step": 90
    },
    {
      "epoch": 5.6875,
      "grad_norm": 7.401065826416016,
      "learning_rate": 4.923611111111112e-05,
      "loss": 0.6985,
      "step": 91
    },
    {
      "epoch": 5.75,
      "grad_norm": 6.9926042556762695,
      "learning_rate": 4.9166666666666665e-05,
      "loss": 0.667,
      "step": 92
    },
    {
      "epoch": 5.8125,
      "grad_norm": 6.541015625,
      "learning_rate": 4.909722222222222e-05,
      "loss": 0.6868,
      "step": 93
    },
    {
      "epoch": 5.875,
      "grad_norm": 8.960890769958496,
      "learning_rate": 4.902777777777778e-05,
      "loss": 0.7009,
      "step": 94
    },
    {
      "epoch": 5.9375,
      "grad_norm": 6.148426532745361,
      "learning_rate": 4.8958333333333335e-05,
      "loss": 0.6677,
      "step": 95
    },
    {
      "epoch": 6.0,
      "grad_norm": 18.49876594543457,
      "learning_rate": 4.888888888888889e-05,
      "loss": 0.6842,
      "step": 96
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5161290322580645,
      "eval_auc": 0.6039915966386554,
      "eval_f1": 0.6808510638297872,
      "eval_loss": 0.6819557547569275,
      "eval_precision": 0.5333333333333333,
      "eval_recall": 0.9411764705882353,
      "eval_runtime": 0.9944,
      "eval_samples_per_second": 62.35,
      "eval_steps_per_second": 2.011,
      "step": 96
    }
  ],
  "logging_steps": 1,
  "max_steps": 800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.01
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.134345545949184e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}