{
  "best_metric": 0.5358908772468567,
  "best_model_checkpoint": "./vit-brain-tumour-v2/checkpoint-200",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 780,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05128205128205128,
      "grad_norm": 1.8319804668426514,
      "learning_rate": 0.00019743589743589744,
      "loss": 0.8818,
      "step": 10
    },
    {
      "epoch": 0.10256410256410256,
      "grad_norm": 0.8907442092895508,
      "learning_rate": 0.00019487179487179487,
      "loss": 0.3732,
      "step": 20
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.5684014558792114,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.2491,
      "step": 30
    },
    {
      "epoch": 0.20512820512820512,
      "grad_norm": 0.4306720495223999,
      "learning_rate": 0.00018974358974358974,
      "loss": 0.1864,
      "step": 40
    },
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 5.443202018737793,
      "learning_rate": 0.0001871794871794872,
      "loss": 0.2069,
      "step": 50
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.45025336742401123,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.2037,
      "step": 60
    },
    {
      "epoch": 0.358974358974359,
      "grad_norm": 0.8258907198905945,
      "learning_rate": 0.00018205128205128207,
      "loss": 0.1452,
      "step": 70
    },
    {
      "epoch": 0.41025641025641024,
      "grad_norm": 0.5631661415100098,
      "learning_rate": 0.0001794871794871795,
      "loss": 0.2959,
      "step": 80
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.5860559940338135,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.0992,
      "step": 90
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 3.4091548919677734,
      "learning_rate": 0.00017435897435897436,
      "loss": 0.1236,
      "step": 100
    },
    {
      "epoch": 0.5128205128205128,
      "eval_accuracy": 0.8481481481481481,
      "eval_loss": 0.5989510416984558,
      "eval_runtime": 15.9913,
      "eval_samples_per_second": 50.652,
      "eval_steps_per_second": 6.378,
      "step": 100
    },
    {
      "epoch": 0.5641025641025641,
      "grad_norm": 1.3914088010787964,
      "learning_rate": 0.0001717948717948718,
      "loss": 0.1454,
      "step": 110
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 2.7670581340789795,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.1344,
      "step": 120
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.14079055190086365,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.2146,
      "step": 130
    },
    {
      "epoch": 0.717948717948718,
      "grad_norm": 0.2149924784898758,
      "learning_rate": 0.0001641025641025641,
      "loss": 0.1724,
      "step": 140
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 4.1687703132629395,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.1342,
      "step": 150
    },
    {
      "epoch": 0.8205128205128205,
      "grad_norm": 0.09244555979967117,
      "learning_rate": 0.00015897435897435896,
      "loss": 0.1174,
      "step": 160
    },
    {
      "epoch": 0.8717948717948718,
      "grad_norm": 3.8858108520507812,
      "learning_rate": 0.00015641025641025642,
      "loss": 0.1445,
      "step": 170
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.13063883781433105,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.086,
      "step": 180
    },
    {
      "epoch": 0.9743589743589743,
      "grad_norm": 0.4698440432548523,
      "learning_rate": 0.00015128205128205128,
      "loss": 0.0978,
      "step": 190
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 0.7778797149658203,
      "learning_rate": 0.00014871794871794872,
      "loss": 0.1695,
      "step": 200
    },
    {
      "epoch": 1.0256410256410255,
      "eval_accuracy": 0.8703703703703703,
      "eval_loss": 0.5358908772468567,
      "eval_runtime": 15.8447,
      "eval_samples_per_second": 51.121,
      "eval_steps_per_second": 6.437,
      "step": 200
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.09852629154920578,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.0657,
      "step": 210
    },
    {
      "epoch": 1.1282051282051282,
      "grad_norm": 3.54819655418396,
      "learning_rate": 0.0001435897435897436,
      "loss": 0.0461,
      "step": 220
    },
    {
      "epoch": 1.1794871794871795,
      "grad_norm": 0.0645141452550888,
      "learning_rate": 0.00014102564102564104,
      "loss": 0.0754,
      "step": 230
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.2788432240486145,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.0592,
      "step": 240
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 0.9937513470649719,
      "learning_rate": 0.0001358974358974359,
      "loss": 0.0509,
      "step": 250
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 1.003893494606018,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.0319,
      "step": 260
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.04893897473812103,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0348,
      "step": 270
    },
    {
      "epoch": 1.435897435897436,
      "grad_norm": 1.5978337526321411,
      "learning_rate": 0.00012820512820512823,
      "loss": 0.0434,
      "step": 280
    },
    {
      "epoch": 1.4871794871794872,
      "grad_norm": 0.044942498207092285,
      "learning_rate": 0.00012564102564102564,
      "loss": 0.0787,
      "step": 290
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.04497215896844864,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0186,
      "step": 300
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.8975308641975308,
      "eval_loss": 0.570486843585968,
      "eval_runtime": 16.7695,
      "eval_samples_per_second": 48.302,
      "eval_steps_per_second": 6.082,
      "step": 300
    },
    {
      "epoch": 1.5897435897435899,
      "grad_norm": 0.03568781167268753,
      "learning_rate": 0.00012051282051282052,
      "loss": 0.0339,
      "step": 310
    },
    {
      "epoch": 1.641025641025641,
      "grad_norm": 0.034062109887599945,
      "learning_rate": 0.00011794871794871796,
      "loss": 0.0072,
      "step": 320
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.03320033475756645,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0062,
      "step": 330
    },
    {
      "epoch": 1.7435897435897436,
      "grad_norm": 0.03138517960906029,
      "learning_rate": 0.00011282051282051283,
      "loss": 0.0063,
      "step": 340
    },
    {
      "epoch": 1.7948717948717947,
      "grad_norm": 0.04579005390405655,
      "learning_rate": 0.00011025641025641027,
      "loss": 0.0064,
      "step": 350
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.02792205847799778,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0076,
      "step": 360
    },
    {
      "epoch": 1.8974358974358974,
      "grad_norm": 0.03118390031158924,
      "learning_rate": 0.00010512820512820514,
      "loss": 0.0068,
      "step": 370
    },
    {
      "epoch": 1.9487179487179487,
      "grad_norm": 0.02970140054821968,
      "learning_rate": 0.00010256410256410256,
      "loss": 0.006,
      "step": 380
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.09769738465547562,
      "learning_rate": 0.0001,
      "loss": 0.0374,
      "step": 390
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 0.09561528265476227,
      "learning_rate": 9.743589743589744e-05,
      "loss": 0.0368,
      "step": 400
    },
    {
      "epoch": 2.051282051282051,
      "eval_accuracy": 0.8975308641975308,
      "eval_loss": 0.6135894656181335,
      "eval_runtime": 16.0452,
      "eval_samples_per_second": 50.482,
      "eval_steps_per_second": 6.357,
      "step": 400
    },
    {
      "epoch": 2.1025641025641026,
      "grad_norm": 0.02449757605791092,
      "learning_rate": 9.487179487179487e-05,
      "loss": 0.0051,
      "step": 410
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.02481868490576744,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0044,
      "step": 420
    },
    {
      "epoch": 2.2051282051282053,
      "grad_norm": 0.02576032653450966,
      "learning_rate": 8.974358974358975e-05,
      "loss": 0.0044,
      "step": 430
    },
    {
      "epoch": 2.2564102564102564,
      "grad_norm": 0.026232751086354256,
      "learning_rate": 8.717948717948718e-05,
      "loss": 0.0043,
      "step": 440
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.02455274388194084,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0039,
      "step": 450
    },
    {
      "epoch": 2.358974358974359,
      "grad_norm": 0.02121666818857193,
      "learning_rate": 8.205128205128205e-05,
      "loss": 0.007,
      "step": 460
    },
    {
      "epoch": 2.41025641025641,
      "grad_norm": 0.022693779319524765,
      "learning_rate": 7.948717948717948e-05,
      "loss": 0.0045,
      "step": 470
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.8521128296852112,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0044,
      "step": 480
    },
    {
      "epoch": 2.5128205128205128,
      "grad_norm": 0.01888178288936615,
      "learning_rate": 7.435897435897436e-05,
      "loss": 0.0117,
      "step": 490
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 0.026412444189190865,
      "learning_rate": 7.17948717948718e-05,
      "loss": 0.0036,
      "step": 500
    },
    {
      "epoch": 2.564102564102564,
      "eval_accuracy": 0.9012345679012346,
      "eval_loss": 0.6121538877487183,
      "eval_runtime": 16.7144,
      "eval_samples_per_second": 48.461,
      "eval_steps_per_second": 6.103,
      "step": 500
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.0200443584471941,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0037,
      "step": 510
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.021741095930337906,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.0162,
      "step": 520
    },
    {
      "epoch": 2.717948717948718,
      "grad_norm": 1.7510725259780884,
      "learning_rate": 6.410256410256412e-05,
      "loss": 0.0044,
      "step": 530
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.018649321049451828,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0034,
      "step": 540
    },
    {
      "epoch": 2.8205128205128203,
      "grad_norm": 0.018490077927708626,
      "learning_rate": 5.897435897435898e-05,
      "loss": 0.0032,
      "step": 550
    },
    {
      "epoch": 2.871794871794872,
      "grad_norm": 0.018017606809735298,
      "learning_rate": 5.6410256410256414e-05,
      "loss": 0.003,
      "step": 560
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.01674257032573223,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0031,
      "step": 570
    },
    {
      "epoch": 2.9743589743589745,
      "grad_norm": 0.017541637644171715,
      "learning_rate": 5.128205128205128e-05,
      "loss": 0.0031,
      "step": 580
    },
    {
      "epoch": 3.0256410256410255,
      "grad_norm": 0.016599087044596672,
      "learning_rate": 4.871794871794872e-05,
      "loss": 0.0029,
      "step": 590
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.017430290579795837,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0029,
      "step": 600
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9024691358024691,
      "eval_loss": 0.6067318320274353,
      "eval_runtime": 16.2775,
      "eval_samples_per_second": 49.762,
      "eval_steps_per_second": 6.266,
      "step": 600
    },
    {
      "epoch": 3.128205128205128,
      "grad_norm": 0.016984164714813232,
      "learning_rate": 4.358974358974359e-05,
      "loss": 0.003,
      "step": 610
    },
    {
      "epoch": 3.1794871794871793,
      "grad_norm": 0.01852291077375412,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 0.0028,
      "step": 620
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.016155986115336418,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0066,
      "step": 630
    },
    {
      "epoch": 3.282051282051282,
      "grad_norm": 0.015048054046928883,
      "learning_rate": 3.58974358974359e-05,
      "loss": 0.0027,
      "step": 640
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.014590720646083355,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0028,
      "step": 650
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.015942472964525223,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0027,
      "step": 660
    },
    {
      "epoch": 3.435897435897436,
      "grad_norm": 0.015167794190347195,
      "learning_rate": 2.8205128205128207e-05,
      "loss": 0.0031,
      "step": 670
    },
    {
      "epoch": 3.4871794871794872,
      "grad_norm": 0.015961118042469025,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.0027,
      "step": 680
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.01679452508687973,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0029,
      "step": 690
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 0.014717783778905869,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.0027,
      "step": 700
    },
    {
      "epoch": 3.58974358974359,
      "eval_accuracy": 0.9024691358024691,
      "eval_loss": 0.644864022731781,
      "eval_runtime": 16.2975,
      "eval_samples_per_second": 49.701,
      "eval_steps_per_second": 6.259,
      "step": 700
    },
    {
      "epoch": 3.641025641025641,
      "grad_norm": 0.01574333943426609,
      "learning_rate": 1.794871794871795e-05,
      "loss": 0.0028,
      "step": 710
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.014611059799790382,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0029,
      "step": 720
    },
    {
      "epoch": 3.7435897435897436,
      "grad_norm": 0.014431651681661606,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.0026,
      "step": 730
    },
    {
      "epoch": 3.7948717948717947,
      "grad_norm": 0.015641087666153908,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 0.0025,
      "step": 740
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.021887468174099922,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0027,
      "step": 750
    },
    {
      "epoch": 3.8974358974358974,
      "grad_norm": 0.015822410583496094,
      "learning_rate": 5.128205128205128e-06,
      "loss": 0.0025,
      "step": 760
    },
    {
      "epoch": 3.948717948717949,
      "grad_norm": 0.015171283856034279,
      "learning_rate": 2.564102564102564e-06,
      "loss": 0.0026,
      "step": 770
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0153021439909935,
      "learning_rate": 0.0,
      "loss": 0.0028,
      "step": 780
    },
    {
      "epoch": 4.0,
      "step": 780,
      "total_flos": 9.636990200991498e+17,
      "train_loss": 0.06401206638472967,
      "train_runtime": 946.5564,
      "train_samples_per_second": 13.138,
      "train_steps_per_second": 0.824
    }
  ],
  "logging_steps": 10,
  "max_steps": 780,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.636990200991498e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}