{
  "best_metric": 0.06105376407504082,
  "best_model_checkpoint": "deepfake-custom/checkpoint-400",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 26.546144485473633,
      "learning_rate": 8.333333333333334e-06,
      "loss": 4.7282,
      "step": 10
    },
    {
      "epoch": 0.1,
      "grad_norm": 37.319374084472656,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 4.2879,
      "step": 20
    },
    {
      "epoch": 0.15,
      "grad_norm": 10.413372039794922,
      "learning_rate": 2.5e-05,
      "loss": 1.7524,
      "step": 30
    },
    {
      "epoch": 0.2,
      "grad_norm": 6.332607269287109,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.7878,
      "step": 40
    },
    {
      "epoch": 0.25,
      "grad_norm": 11.610198020935059,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.5413,
      "step": 50
    },
    {
      "epoch": 0.3,
      "grad_norm": 4.810694694519043,
      "learning_rate": 5e-05,
      "loss": 0.4014,
      "step": 60
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.429203987121582,
      "learning_rate": 4.9074074074074075e-05,
      "loss": 0.2954,
      "step": 70
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.562802791595459,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.2239,
      "step": 80
    },
    {
      "epoch": 0.45,
      "grad_norm": 3.2474796772003174,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.3575,
      "step": 90
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.1597166061401367,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.1528,
      "step": 100
    },
    {
      "epoch": 0.55,
      "grad_norm": 8.586223602294922,
      "learning_rate": 4.5370370370370374e-05,
      "loss": 0.2503,
      "step": 110
    },
    {
      "epoch": 0.6,
      "grad_norm": 26.026302337646484,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.114,
      "step": 120
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.38507279753685,
      "learning_rate": 4.351851851851852e-05,
      "loss": 0.2078,
      "step": 130
    },
    {
      "epoch": 0.7,
      "grad_norm": 10.108200073242188,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.1432,
      "step": 140
    },
    {
      "epoch": 0.75,
      "grad_norm": 2.9426777362823486,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.1231,
      "step": 150
    },
    {
      "epoch": 0.8,
      "grad_norm": 37.13353729248047,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.1891,
      "step": 160
    },
    {
      "epoch": 0.85,
      "grad_norm": 16.757221221923828,
      "learning_rate": 3.981481481481482e-05,
      "loss": 0.2966,
      "step": 170
    },
    {
      "epoch": 0.9,
      "grad_norm": 17.378347396850586,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2156,
      "step": 180
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.16039691865444183,
      "learning_rate": 3.7962962962962964e-05,
      "loss": 0.1215,
      "step": 190
    },
    {
      "epoch": 1.0,
      "grad_norm": 32.3621940612793,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2481,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.95,
      "eval_auc": 0.9906250000000001,
      "eval_f1": 0.9512195121951219,
      "eval_loss": 0.16212879121303558,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.975,
      "eval_runtime": 117.7512,
      "eval_samples_per_second": 3.397,
      "eval_steps_per_second": 0.212,
      "step": 200
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.0691157802939415,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.114,
      "step": 210
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.07614326477050781,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.1587,
      "step": 220
    },
    {
      "epoch": 1.15,
      "grad_norm": 3.3094229698181152,
      "learning_rate": 3.425925925925926e-05,
      "loss": 0.1153,
      "step": 230
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.43230101466178894,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.17,
      "step": 240
    },
    {
      "epoch": 1.25,
      "grad_norm": 4.47321891784668,
      "learning_rate": 3.240740740740741e-05,
      "loss": 0.2543,
      "step": 250
    },
    {
      "epoch": 1.3,
      "grad_norm": 13.483596801757812,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.3613,
      "step": 260
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.24584409594535828,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.1508,
      "step": 270
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.4941992163658142,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.2375,
      "step": 280
    },
    {
      "epoch": 1.45,
      "grad_norm": 2.139481782913208,
      "learning_rate": 2.8703703703703706e-05,
      "loss": 0.1685,
      "step": 290
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.24849505722522736,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1729,
      "step": 300
    },
    {
      "epoch": 1.55,
      "grad_norm": 13.943930625915527,
      "learning_rate": 2.6851851851851855e-05,
      "loss": 0.2713,
      "step": 310
    },
    {
      "epoch": 1.6,
      "grad_norm": 18.25007438659668,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.2127,
      "step": 320
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.24169185757637024,
      "learning_rate": 2.5e-05,
      "loss": 0.0963,
      "step": 330
    },
    {
      "epoch": 1.7,
      "grad_norm": 2.021479845046997,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.1298,
      "step": 340
    },
    {
      "epoch": 1.75,
      "grad_norm": 12.028373718261719,
      "learning_rate": 2.314814814814815e-05,
      "loss": 0.1902,
      "step": 350
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.25696495175361633,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.163,
      "step": 360
    },
    {
      "epoch": 1.85,
      "grad_norm": 28.349374771118164,
      "learning_rate": 2.1296296296296296e-05,
      "loss": 0.1161,
      "step": 370
    },
    {
      "epoch": 1.9,
      "grad_norm": 18.065284729003906,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.2918,
      "step": 380
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.1438354104757309,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.158,
      "step": 390
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.04646734893321991,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.0268,
      "step": 400
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.985,
      "eval_auc": 0.998925,
      "eval_f1": 0.9850746268656716,
      "eval_loss": 0.06105376407504082,
      "eval_precision": 0.9801980198019802,
      "eval_recall": 0.99,
      "eval_runtime": 122.4824,
      "eval_samples_per_second": 3.266,
      "eval_steps_per_second": 0.204,
      "step": 400
    }
  ],
  "logging_steps": 10,
  "max_steps": 600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.479743667666944e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}