{
  "best_metric": 0.9857142857142858,
  "best_model_checkpoint": "document-spoof-clip/checkpoint-52",
  "epoch": 16.842105263157894,
  "eval_steps": 500,
  "global_step": 80,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8421052631578947,
      "eval_accuracy": 0.8285714285714286,
      "eval_loss": 0.5162299275398254,
      "eval_runtime": 4.4211,
      "eval_samples_per_second": 15.833,
      "eval_steps_per_second": 1.131,
      "step": 4
    },
    {
      "epoch": 1.8947368421052633,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.5837788581848145,
      "eval_runtime": 4.4097,
      "eval_samples_per_second": 15.874,
      "eval_steps_per_second": 1.134,
      "step": 9
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 50.87126541137695,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 0.4911,
      "step": 10
    },
    {
      "epoch": 2.9473684210526314,
      "eval_accuracy": 0.9428571428571428,
      "eval_loss": 0.15884919464588165,
      "eval_runtime": 4.3778,
      "eval_samples_per_second": 15.99,
      "eval_steps_per_second": 1.142,
      "step": 14
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9428571428571428,
      "eval_loss": 0.20427295565605164,
      "eval_runtime": 4.3631,
      "eval_samples_per_second": 16.044,
      "eval_steps_per_second": 1.146,
      "step": 19
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 0.6421628594398499,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.0819,
      "step": 20
    },
    {
      "epoch": 4.842105263157895,
      "eval_accuracy": 0.9714285714285714,
      "eval_loss": 0.12607304751873016,
      "eval_runtime": 4.3716,
      "eval_samples_per_second": 16.013,
      "eval_steps_per_second": 1.144,
      "step": 23
    },
    {
      "epoch": 5.894736842105263,
      "eval_accuracy": 0.9142857142857143,
      "eval_loss": 0.6375396251678467,
      "eval_runtime": 4.4518,
      "eval_samples_per_second": 15.724,
      "eval_steps_per_second": 1.123,
      "step": 28
    },
    {
      "epoch": 6.315789473684211,
      "grad_norm": 27.80014991760254,
      "learning_rate": 3.472222222222222e-05,
      "loss": 0.1739,
      "step": 30
    },
    {
      "epoch": 6.947368421052632,
      "eval_accuracy": 0.8428571428571429,
      "eval_loss": 1.199052095413208,
      "eval_runtime": 4.4441,
      "eval_samples_per_second": 15.751,
      "eval_steps_per_second": 1.125,
      "step": 33
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8571428571428571,
      "eval_loss": 0.8591070175170898,
      "eval_runtime": 4.4267,
      "eval_samples_per_second": 15.813,
      "eval_steps_per_second": 1.13,
      "step": 38
    },
    {
      "epoch": 8.421052631578947,
      "grad_norm": 0.22448770701885223,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.2177,
      "step": 40
    },
    {
      "epoch": 8.842105263157894,
      "eval_accuracy": 0.9571428571428572,
      "eval_loss": 0.25587812066078186,
      "eval_runtime": 4.4536,
      "eval_samples_per_second": 15.718,
      "eval_steps_per_second": 1.123,
      "step": 42
    },
    {
      "epoch": 9.894736842105264,
      "eval_accuracy": 0.9285714285714286,
      "eval_loss": 0.14307884871959686,
      "eval_runtime": 4.4003,
      "eval_samples_per_second": 15.908,
      "eval_steps_per_second": 1.136,
      "step": 47
    },
    {
      "epoch": 10.526315789473685,
      "grad_norm": 20.779386520385742,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.022,
      "step": 50
    },
    {
      "epoch": 10.947368421052632,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.10236798971891403,
      "eval_runtime": 4.4339,
      "eval_samples_per_second": 15.788,
      "eval_steps_per_second": 1.128,
      "step": 52
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.11362212896347046,
      "eval_runtime": 4.4292,
      "eval_samples_per_second": 15.804,
      "eval_steps_per_second": 1.129,
      "step": 57
    },
    {
      "epoch": 12.631578947368421,
      "grad_norm": 1.5777080059051514,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.0007,
      "step": 60
    },
    {
      "epoch": 12.842105263157894,
      "eval_accuracy": 0.9714285714285714,
      "eval_loss": 0.16094441711902618,
      "eval_runtime": 4.4586,
      "eval_samples_per_second": 15.7,
      "eval_steps_per_second": 1.121,
      "step": 61
    },
    {
      "epoch": 13.894736842105264,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.1507013887166977,
      "eval_runtime": 4.409,
      "eval_samples_per_second": 15.877,
      "eval_steps_per_second": 1.134,
      "step": 66
    },
    {
      "epoch": 14.736842105263158,
      "grad_norm": 0.21966008841991425,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.0006,
      "step": 70
    },
    {
      "epoch": 14.947368421052632,
      "eval_accuracy": 0.9714285714285714,
      "eval_loss": 0.22763238847255707,
      "eval_runtime": 4.4029,
      "eval_samples_per_second": 15.899,
      "eval_steps_per_second": 1.136,
      "step": 71
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9714285714285714,
      "eval_loss": 0.17074808478355408,
      "eval_runtime": 4.3943,
      "eval_samples_per_second": 15.93,
      "eval_steps_per_second": 1.138,
      "step": 76
    },
    {
      "epoch": 16.842105263157894,
      "grad_norm": 0.0005645102355629206,
      "learning_rate": 0.0,
      "loss": 0.0002,
      "step": 80
    },
    {
      "epoch": 16.842105263157894,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.1557733714580536,
      "eval_runtime": 4.4555,
      "eval_samples_per_second": 15.711,
      "eval_steps_per_second": 1.122,
      "step": 80
    },
    {
      "epoch": 16.842105263157894,
      "step": 80,
      "total_flos": 3.991922735555543e+17,
      "train_loss": 0.12353107887029183,
      "train_runtime": 494.7238,
      "train_samples_per_second": 12.128,
      "train_steps_per_second": 0.162
    }
  ],
  "logging_steps": 10,
  "max_steps": 80,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 3.991922735555543e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}