checkpoint-7138/trainer_state.json

{
  "best_metric": 0.49518314003944397,
  "best_model_checkpoint": "deepfake_vs_real_image_detection/checkpoint-7138",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 7138,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14009526478005044,
      "grad_norm": 0.9985737800598145,
      "learning_rate": 9.365124153498872e-07,
      "loss": 0.672,
      "step": 500
    },
    {
      "epoch": 0.2801905295601009,
      "grad_norm": 1.7435837984085083,
      "learning_rate": 8.659706546275395e-07,
      "loss": 0.544,
      "step": 1000
    },
    {
      "epoch": 0.4202857943401513,
      "grad_norm": 2.39518141746521,
      "learning_rate": 7.954288939051918e-07,
      "loss": 0.3963,
      "step": 1500
    },
    {
      "epoch": 0.5603810591202018,
      "grad_norm": 3.065082550048828,
      "learning_rate": 7.248871331828442e-07,
      "loss": 0.3141,
      "step": 2000
    },
    {
      "epoch": 0.7004763239002522,
      "grad_norm": 3.2442855834960938,
      "learning_rate": 6.543453724604966e-07,
      "loss": 0.2678,
      "step": 2500
    },
    {
      "epoch": 0.8405715886803026,
      "grad_norm": 2.15069580078125,
      "learning_rate": 5.83803611738149e-07,
      "loss": 0.2495,
      "step": 3000
    },
    {
      "epoch": 0.9806668534603531,
      "grad_norm": 6.248366355895996,
      "learning_rate": 5.132618510158013e-07,
      "loss": 0.2278,
      "step": 3500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.691609530564531,
      "eval_loss": 0.5145022869110107,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 381.2789,
      "eval_samples_per_second": 199.681,
      "eval_steps_per_second": 24.961,
      "step": 3569
    },
    {
      "epoch": 1.1207621182404035,
      "grad_norm": 5.071241855621338,
      "learning_rate": 4.427200902934537e-07,
      "loss": 0.2198,
      "step": 4000
    },
    {
      "epoch": 1.260857383020454,
      "grad_norm": 4.649421215057373,
      "learning_rate": 3.7217832957110605e-07,
      "loss": 0.2144,
      "step": 4500
    },
    {
      "epoch": 1.4009526478005043,
      "grad_norm": 5.81755256652832,
      "learning_rate": 3.0163656884875844e-07,
      "loss": 0.2037,
      "step": 5000
    },
    {
      "epoch": 1.5410479125805547,
      "grad_norm": 0.6763508915901184,
      "learning_rate": 2.3109480812641084e-07,
      "loss": 0.1985,
      "step": 5500
    },
    {
      "epoch": 1.6811431773606054,
      "grad_norm": 4.193052291870117,
      "learning_rate": 1.605530474040632e-07,
      "loss": 0.1958,
      "step": 6000
    },
    {
      "epoch": 1.8212384421406558,
      "grad_norm": 10.019765853881836,
      "learning_rate": 9.001128668171557e-08,
      "loss": 0.1859,
      "step": 6500
    },
    {
      "epoch": 1.9613337069207062,
      "grad_norm": 4.139777660369873,
      "learning_rate": 1.9469525959367945e-08,
      "loss": 0.1865,
      "step": 7000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6998712795859932,
      "eval_loss": 0.49518314003944397,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 362.7947,
      "eval_samples_per_second": 209.854,
      "eval_steps_per_second": 26.232,
      "step": 7138
    }
  ],
  "logging_steps": 500,
  "max_steps": 7138,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 1.7699325411952042e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}