{
  "best_global_step": 170,
  "best_metric": 0.9591836734693877,
  "best_model_checkpoint": "vit-defect-classifier/checkpoint-170",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 170,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 4.38879919052124,
      "learning_rate": 4.735294117647059e-05,
      "loss": 0.49321718215942384,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8392857142857143,
      "eval_defect_f1": 0.912621359223301,
      "eval_defect_precision": 0.8392857142857143,
      "eval_defect_recall": 1.0,
      "eval_loss": 0.4327879846096039,
      "eval_runtime": 3.1742,
      "eval_samples_per_second": 17.642,
      "eval_steps_per_second": 1.26,
      "step": 17
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 1.0918186902999878,
      "learning_rate": 4.4411764705882356e-05,
      "loss": 0.3930933713912964,
      "step": 20
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 1.6918683052062988,
      "learning_rate": 4.147058823529412e-05,
      "loss": 0.4969632148742676,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8392857142857143,
      "eval_defect_f1": 0.912621359223301,
      "eval_defect_precision": 0.8392857142857143,
      "eval_defect_recall": 1.0,
      "eval_loss": 0.43066737055778503,
      "eval_runtime": 3.1503,
      "eval_samples_per_second": 17.776,
      "eval_steps_per_second": 1.27,
      "step": 34
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 3.023923635482788,
      "learning_rate": 3.8529411764705886e-05,
      "loss": 0.4409238338470459,
      "step": 40
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 6.262722492218018,
      "learning_rate": 3.558823529411765e-05,
      "loss": 0.3777069807052612,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8392857142857143,
      "eval_defect_f1": 0.912621359223301,
      "eval_defect_precision": 0.8392857142857143,
      "eval_defect_recall": 1.0,
      "eval_loss": 0.37949138879776,
      "eval_runtime": 3.1536,
      "eval_samples_per_second": 17.757,
      "eval_steps_per_second": 1.268,
      "step": 51
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 4.074706077575684,
      "learning_rate": 3.2647058823529416e-05,
      "loss": 0.38631012439727785,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8392857142857143,
      "eval_defect_f1": 0.912621359223301,
      "eval_defect_precision": 0.8392857142857143,
      "eval_defect_recall": 1.0,
      "eval_loss": 0.34745684266090393,
      "eval_runtime": 3.1877,
      "eval_samples_per_second": 17.567,
      "eval_steps_per_second": 1.255,
      "step": 68
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 1.672234058380127,
      "learning_rate": 2.9705882352941177e-05,
      "loss": 0.3591782569885254,
      "step": 70
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 2.9030237197875977,
      "learning_rate": 2.6764705882352942e-05,
      "loss": 0.3000319004058838,
      "step": 80
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8392857142857143,
      "eval_defect_f1": 0.9010989010989011,
      "eval_defect_precision": 0.9318181818181818,
      "eval_defect_recall": 0.8723404255319149,
      "eval_loss": 0.3284582197666168,
      "eval_runtime": 3.1607,
      "eval_samples_per_second": 17.718,
      "eval_steps_per_second": 1.266,
      "step": 85
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 6.889004230499268,
      "learning_rate": 2.3823529411764707e-05,
      "loss": 0.24264731407165527,
      "step": 90
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 3.043811082839966,
      "learning_rate": 2.0882352941176472e-05,
      "loss": 0.31143391132354736,
      "step": 100
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8392857142857143,
      "eval_defect_f1": 0.912621359223301,
      "eval_defect_precision": 0.8392857142857143,
      "eval_defect_recall": 1.0,
      "eval_loss": 0.44226619601249695,
      "eval_runtime": 3.1806,
      "eval_samples_per_second": 17.607,
      "eval_steps_per_second": 1.258,
      "step": 102
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 5.202059268951416,
      "learning_rate": 1.7941176470588237e-05,
      "loss": 0.26354250907897947,
      "step": 110
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8571428571428571,
      "eval_defect_f1": 0.9130434782608695,
      "eval_defect_precision": 0.9333333333333333,
      "eval_defect_recall": 0.8936170212765957,
      "eval_loss": 0.33693668246269226,
      "eval_runtime": 3.1632,
      "eval_samples_per_second": 17.704,
      "eval_steps_per_second": 1.265,
      "step": 119
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 1.2519395351409912,
      "learning_rate": 1.5e-05,
      "loss": 0.2739341497421265,
      "step": 120
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 2.8088483810424805,
      "learning_rate": 1.2058823529411765e-05,
      "loss": 0.20456738471984864,
      "step": 130
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7678571428571429,
      "eval_defect_f1": 0.8505747126436781,
      "eval_defect_precision": 0.925,
      "eval_defect_recall": 0.7872340425531915,
      "eval_loss": 0.47304031252861023,
      "eval_runtime": 3.1879,
      "eval_samples_per_second": 17.566,
      "eval_steps_per_second": 1.255,
      "step": 136
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 4.974433898925781,
      "learning_rate": 9.117647058823529e-06,
      "loss": 0.19193782806396484,
      "step": 140
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 2.3693747520446777,
      "learning_rate": 6.1764705882352944e-06,
      "loss": 0.2625112533569336,
      "step": 150
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8928571428571429,
      "eval_defect_f1": 0.9361702127659575,
      "eval_defect_precision": 0.9361702127659575,
      "eval_defect_recall": 0.9361702127659575,
      "eval_loss": 0.31851115822792053,
      "eval_runtime": 3.169,
      "eval_samples_per_second": 17.671,
      "eval_steps_per_second": 1.262,
      "step": 153
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 4.599684238433838,
      "learning_rate": 3.2352941176470594e-06,
      "loss": 0.1370754361152649,
      "step": 160
    },
    {
      "epoch": 10.0,
      "grad_norm": 5.715418338775635,
      "learning_rate": 2.9411764705882356e-07,
      "loss": 0.23060293197631837,
      "step": 170
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9285714285714286,
      "eval_defect_f1": 0.9591836734693877,
      "eval_defect_precision": 0.9215686274509803,
      "eval_defect_recall": 1.0,
      "eval_loss": 0.31930261850357056,
      "eval_runtime": 3.1722,
      "eval_samples_per_second": 17.654,
      "eval_steps_per_second": 1.261,
      "step": 170
    },
    {
      "epoch": 10.0,
      "step": 170,
      "total_flos": 2.0225409289408512e+17,
      "train_loss": 0.31562809313044826,
      "train_runtime": 284.4778,
      "train_samples_per_second": 9.175,
      "train_steps_per_second": 0.598
    }
  ],
  "logging_steps": 10,
  "max_steps": 170,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.0225409289408512e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}