{
  "best_metric": 0.20142494142055511,
  "best_model_checkpoint": "./vit-base-brain-alzheimer-detection/checkpoint-2000",
  "epoch": 13.671875,
  "eval_steps": 500,
  "global_step": 3500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1953125,
      "grad_norm": 0.01695910282433033,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0035,
      "step": 50
    },
    {
      "epoch": 0.390625,
      "grad_norm": 0.013198412954807281,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0031,
      "step": 100
    },
    {
      "epoch": 0.5859375,
      "grad_norm": 0.01613868959248066,
      "learning_rate": 6e-06,
      "loss": 0.0031,
      "step": 150
    },
    {
      "epoch": 0.78125,
      "grad_norm": 0.011624956503510475,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0031,
      "step": 200
    },
    {
      "epoch": 0.9765625,
      "grad_norm": 0.026181865483522415,
      "learning_rate": 1e-05,
      "loss": 0.0026,
      "step": 250
    },
    {
      "epoch": 1.171875,
      "grad_norm": 0.010051410645246506,
      "learning_rate": 1.2e-05,
      "loss": 0.0026,
      "step": 300
    },
    {
      "epoch": 1.3671875,
      "grad_norm": 0.009718737564980984,
      "learning_rate": 1.4e-05,
      "loss": 0.0022,
      "step": 350
    },
    {
      "epoch": 1.5625,
      "grad_norm": 0.00813582818955183,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0019,
      "step": 400
    },
    {
      "epoch": 1.7578125,
      "grad_norm": 14.344162940979004,
      "learning_rate": 1.8e-05,
      "loss": 0.3794,
      "step": 450
    },
    {
      "epoch": 1.953125,
      "grad_norm": 27.69124984741211,
      "learning_rate": 1.9960000000000002e-05,
      "loss": 0.4213,
      "step": 500
    },
    {
      "epoch": 1.953125,
      "eval_accuracy": 0.8955078125,
      "eval_loss": 0.3758355975151062,
      "eval_runtime": 7.1938,
      "eval_samples_per_second": 142.346,
      "eval_steps_per_second": 17.793,
      "step": 500
    },
    {
      "epoch": 2.1484375,
      "grad_norm": 0.10575103759765625,
      "learning_rate": 1.978787878787879e-05,
      "loss": 0.219,
      "step": 550
    },
    {
      "epoch": 2.34375,
      "grad_norm": 10.299068450927734,
      "learning_rate": 1.9571428571428572e-05,
      "loss": 0.2295,
      "step": 600
    },
    {
      "epoch": 2.5390625,
      "grad_norm": 30.670997619628906,
      "learning_rate": 1.935497835497836e-05,
      "loss": 0.0989,
      "step": 650
    },
    {
      "epoch": 2.734375,
      "grad_norm": 10.104991912841797,
      "learning_rate": 1.9138528138528142e-05,
      "loss": 0.0813,
      "step": 700
    },
    {
      "epoch": 2.9296875,
      "grad_norm": 0.01667148247361183,
      "learning_rate": 1.8922077922077925e-05,
      "loss": 0.0195,
      "step": 750
    },
    {
      "epoch": 3.125,
      "grad_norm": 4.969385147094727,
      "learning_rate": 1.8705627705627708e-05,
      "loss": 0.02,
      "step": 800
    },
    {
      "epoch": 3.3203125,
      "grad_norm": 0.010836214758455753,
      "learning_rate": 1.848917748917749e-05,
      "loss": 0.0306,
      "step": 850
    },
    {
      "epoch": 3.515625,
      "grad_norm": 0.009891550056636333,
      "learning_rate": 1.8272727272727275e-05,
      "loss": 0.0433,
      "step": 900
    },
    {
      "epoch": 3.7109375,
      "grad_norm": 0.08000622689723969,
      "learning_rate": 1.8056277056277058e-05,
      "loss": 0.0294,
      "step": 950
    },
    {
      "epoch": 3.90625,
      "grad_norm": 0.015158270485699177,
      "learning_rate": 1.783982683982684e-05,
      "loss": 0.0224,
      "step": 1000
    },
    {
      "epoch": 3.90625,
      "eval_accuracy": 0.9501953125,
      "eval_loss": 0.2260357141494751,
      "eval_runtime": 5.6203,
      "eval_samples_per_second": 182.196,
      "eval_steps_per_second": 22.774,
      "step": 1000
    },
    {
      "epoch": 4.1015625,
      "grad_norm": 0.011108304373919964,
      "learning_rate": 1.7623376623376624e-05,
      "loss": 0.0121,
      "step": 1050
    },
    {
      "epoch": 4.296875,
      "grad_norm": 0.014846593141555786,
      "learning_rate": 1.7406926406926407e-05,
      "loss": 0.0205,
      "step": 1100
    },
    {
      "epoch": 4.4921875,
      "grad_norm": 0.032500285655260086,
      "learning_rate": 1.7190476190476194e-05,
      "loss": 0.0261,
      "step": 1150
    },
    {
      "epoch": 4.6875,
      "grad_norm": 0.01257646270096302,
      "learning_rate": 1.6974025974025977e-05,
      "loss": 0.0252,
      "step": 1200
    },
    {
      "epoch": 4.8828125,
      "grad_norm": 0.06579085439443588,
      "learning_rate": 1.675757575757576e-05,
      "loss": 0.024,
      "step": 1250
    },
    {
      "epoch": 5.078125,
      "grad_norm": 2.4693167209625244,
      "learning_rate": 1.6541125541125543e-05,
      "loss": 0.0324,
      "step": 1300
    },
    {
      "epoch": 5.2734375,
      "grad_norm": 0.005182833410799503,
      "learning_rate": 1.6324675324675326e-05,
      "loss": 0.018,
      "step": 1350
    },
    {
      "epoch": 5.46875,
      "grad_norm": 0.006213045679032803,
      "learning_rate": 1.610822510822511e-05,
      "loss": 0.0479,
      "step": 1400
    },
    {
      "epoch": 5.6640625,
      "grad_norm": 50.79215621948242,
      "learning_rate": 1.5891774891774893e-05,
      "loss": 0.0251,
      "step": 1450
    },
    {
      "epoch": 5.859375,
      "grad_norm": 0.021290121600031853,
      "learning_rate": 1.5675324675324676e-05,
      "loss": 0.0012,
      "step": 1500
    },
    {
      "epoch": 5.859375,
      "eval_accuracy": 0.95703125,
      "eval_loss": 0.21266639232635498,
      "eval_runtime": 5.9949,
      "eval_samples_per_second": 170.813,
      "eval_steps_per_second": 21.352,
      "step": 1500
    },
    {
      "epoch": 6.0546875,
      "grad_norm": 12.610783576965332,
      "learning_rate": 1.545887445887446e-05,
      "loss": 0.0256,
      "step": 1550
    },
    {
      "epoch": 6.25,
      "grad_norm": 0.013099271804094315,
      "learning_rate": 1.5242424242424244e-05,
      "loss": 0.0143,
      "step": 1600
    },
    {
      "epoch": 6.4453125,
      "grad_norm": 0.04105505719780922,
      "learning_rate": 1.5025974025974027e-05,
      "loss": 0.0109,
      "step": 1650
    },
    {
      "epoch": 6.640625,
      "grad_norm": 0.005291896406561136,
      "learning_rate": 1.4809523809523812e-05,
      "loss": 0.0222,
      "step": 1700
    },
    {
      "epoch": 6.8359375,
      "grad_norm": 0.005415532272309065,
      "learning_rate": 1.4593073593073595e-05,
      "loss": 0.0052,
      "step": 1750
    },
    {
      "epoch": 7.03125,
      "grad_norm": 0.004291017074137926,
      "learning_rate": 1.4376623376623378e-05,
      "loss": 0.0077,
      "step": 1800
    },
    {
      "epoch": 7.2265625,
      "grad_norm": 0.004927146248519421,
      "learning_rate": 1.4160173160173161e-05,
      "loss": 0.0009,
      "step": 1850
    },
    {
      "epoch": 7.421875,
      "grad_norm": 0.004201130475848913,
      "learning_rate": 1.3943722943722944e-05,
      "loss": 0.0008,
      "step": 1900
    },
    {
      "epoch": 7.6171875,
      "grad_norm": 0.003896114882081747,
      "learning_rate": 1.3727272727272729e-05,
      "loss": 0.0007,
      "step": 1950
    },
    {
      "epoch": 7.8125,
      "grad_norm": 0.0028053205460309982,
      "learning_rate": 1.3510822510822512e-05,
      "loss": 0.0007,
      "step": 2000
    },
    {
      "epoch": 7.8125,
      "eval_accuracy": 0.9677734375,
      "eval_loss": 0.20142494142055511,
      "eval_runtime": 5.4298,
      "eval_samples_per_second": 188.588,
      "eval_steps_per_second": 23.573,
      "step": 2000
    },
    {
      "epoch": 8.0078125,
      "grad_norm": 0.003189160255715251,
      "learning_rate": 1.3294372294372295e-05,
      "loss": 0.0006,
      "step": 2050
    },
    {
      "epoch": 8.203125,
      "grad_norm": 0.005525187123566866,
      "learning_rate": 1.3077922077922078e-05,
      "loss": 0.0006,
      "step": 2100
    },
    {
      "epoch": 8.3984375,
      "grad_norm": 0.002676301635801792,
      "learning_rate": 1.2861471861471862e-05,
      "loss": 0.0006,
      "step": 2150
    },
    {
      "epoch": 8.59375,
      "grad_norm": 0.0033032067585736513,
      "learning_rate": 1.2645021645021646e-05,
      "loss": 0.0006,
      "step": 2200
    },
    {
      "epoch": 8.7890625,
      "grad_norm": 0.003413607832044363,
      "learning_rate": 1.242857142857143e-05,
      "loss": 0.0006,
      "step": 2250
    },
    {
      "epoch": 8.984375,
      "grad_norm": 0.00263944361358881,
      "learning_rate": 1.2212121212121213e-05,
      "loss": 0.0006,
      "step": 2300
    },
    {
      "epoch": 9.1796875,
      "grad_norm": 0.002298458479344845,
      "learning_rate": 1.1995670995670996e-05,
      "loss": 0.0006,
      "step": 2350
    },
    {
      "epoch": 9.375,
      "grad_norm": 0.002362418919801712,
      "learning_rate": 1.1779220779220779e-05,
      "loss": 0.0005,
      "step": 2400
    },
    {
      "epoch": 9.5703125,
      "grad_norm": 0.002904605120420456,
      "learning_rate": 1.1562770562770564e-05,
      "loss": 0.0005,
      "step": 2450
    },
    {
      "epoch": 9.765625,
      "grad_norm": 0.004212340340018272,
      "learning_rate": 1.1346320346320347e-05,
      "loss": 0.0005,
      "step": 2500
    },
    {
      "epoch": 9.765625,
      "eval_accuracy": 0.9697265625,
      "eval_loss": 0.20150402188301086,
      "eval_runtime": 10.8662,
      "eval_samples_per_second": 94.237,
      "eval_steps_per_second": 11.78,
      "step": 2500
    },
    {
      "epoch": 9.9609375,
      "grad_norm": 0.0020910820458084345,
      "learning_rate": 1.112987012987013e-05,
      "loss": 0.0005,
      "step": 2550
    },
    {
      "epoch": 10.15625,
      "grad_norm": 0.0021248324774205685,
      "learning_rate": 1.0913419913419913e-05,
      "loss": 0.0005,
      "step": 2600
    },
    {
      "epoch": 10.3515625,
      "grad_norm": 0.0025855612475425005,
      "learning_rate": 1.0696969696969696e-05,
      "loss": 0.0005,
      "step": 2650
    },
    {
      "epoch": 10.546875,
      "grad_norm": 0.0029177917167544365,
      "learning_rate": 1.0480519480519481e-05,
      "loss": 0.0005,
      "step": 2700
    },
    {
      "epoch": 10.7421875,
      "grad_norm": 0.002265964401885867,
      "learning_rate": 1.0264069264069264e-05,
      "loss": 0.0005,
      "step": 2750
    },
    {
      "epoch": 10.9375,
      "grad_norm": 0.0022889028768986464,
      "learning_rate": 1.0047619047619048e-05,
      "loss": 0.0004,
      "step": 2800
    },
    {
      "epoch": 11.1328125,
      "grad_norm": 0.0023099584504961967,
      "learning_rate": 9.831168831168832e-06,
      "loss": 0.0004,
      "step": 2850
    },
    {
      "epoch": 11.328125,
      "grad_norm": 0.0022109313867986202,
      "learning_rate": 9.614718614718616e-06,
      "loss": 0.0004,
      "step": 2900
    },
    {
      "epoch": 11.5234375,
      "grad_norm": 0.002484516939148307,
      "learning_rate": 9.398268398268399e-06,
      "loss": 0.0004,
      "step": 2950
    },
    {
      "epoch": 11.71875,
      "grad_norm": 0.001686307368800044,
      "learning_rate": 9.181818181818184e-06,
      "loss": 0.0004,
      "step": 3000
    },
    {
      "epoch": 11.71875,
      "eval_accuracy": 0.96875,
      "eval_loss": 0.20902524888515472,
      "eval_runtime": 6.1763,
      "eval_samples_per_second": 165.796,
      "eval_steps_per_second": 20.725,
      "step": 3000
    },
    {
      "epoch": 11.9140625,
      "grad_norm": 0.002134295180439949,
      "learning_rate": 8.965367965367967e-06,
      "loss": 0.0004,
      "step": 3050
    },
    {
      "epoch": 12.109375,
      "grad_norm": 0.002223787596449256,
      "learning_rate": 8.74891774891775e-06,
      "loss": 0.0004,
      "step": 3100
    },
    {
      "epoch": 12.3046875,
      "grad_norm": 0.0019202354596927762,
      "learning_rate": 8.532467532467533e-06,
      "loss": 0.0004,
      "step": 3150
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.0023374806623905897,
      "learning_rate": 8.316017316017316e-06,
      "loss": 0.0004,
      "step": 3200
    },
    {
      "epoch": 12.6953125,
      "grad_norm": 0.0015720732044428587,
      "learning_rate": 8.099567099567101e-06,
      "loss": 0.0004,
      "step": 3250
    },
    {
      "epoch": 12.890625,
      "grad_norm": 0.0014028995065018535,
      "learning_rate": 7.883116883116884e-06,
      "loss": 0.0004,
      "step": 3300
    },
    {
      "epoch": 13.0859375,
      "grad_norm": 0.002278986619785428,
      "learning_rate": 7.666666666666667e-06,
      "loss": 0.0004,
      "step": 3350
    },
    {
      "epoch": 13.28125,
      "grad_norm": 0.0019616258796304464,
      "learning_rate": 7.450216450216451e-06,
      "loss": 0.0003,
      "step": 3400
    },
    {
      "epoch": 13.4765625,
      "grad_norm": 0.0017822927329689264,
      "learning_rate": 7.233766233766234e-06,
      "loss": 0.0003,
      "step": 3450
    },
    {
      "epoch": 13.671875,
      "grad_norm": 0.0021076672710478306,
      "learning_rate": 7.017316017316018e-06,
      "loss": 0.0004,
      "step": 3500
    },
    {
      "epoch": 13.671875,
      "eval_accuracy": 0.96875,
      "eval_loss": 0.21559299528598785,
      "eval_runtime": 6.0759,
      "eval_samples_per_second": 168.535,
      "eval_steps_per_second": 21.067,
      "step": 3500
    },
    {
      "epoch": 13.671875,
      "step": 3500,
      "total_flos": 4.339629206470656e+18,
      "train_loss": 0.02788944040451731,
      "train_runtime": 959.4355,
      "train_samples_per_second": 85.384,
      "train_steps_per_second": 5.336
    }
  ],
  "logging_steps": 50,
  "max_steps": 5120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.339629206470656e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}