checkpoint-4498/trainer_state.json

{
  "best_metric": 0.06474132835865021,
  "best_model_checkpoint": "ai_vs_human_generated_image_detection/checkpoint-4498",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 4498,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2223210315695865,
      "grad_norm": 0.812200129032135,
      "learning_rate": 4.664028669553532e-06,
      "loss": 0.3873,
      "step": 500
    },
    {
      "epoch": 0.444642063139173,
      "grad_norm": 0.2490566074848175,
      "learning_rate": 4.290727191279678e-06,
      "loss": 0.1084,
      "step": 1000
    },
    {
      "epoch": 0.6669630947087595,
      "grad_norm": 2.643937349319458,
      "learning_rate": 3.917425713005824e-06,
      "loss": 0.0839,
      "step": 1500
    },
    {
      "epoch": 0.889284126278346,
      "grad_norm": 1.797574520111084,
      "learning_rate": 3.54412423473197e-06,
      "loss": 0.0725,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9689806128830519,
      "eval_loss": 0.0970255509018898,
      "eval_model_preparation_time": 0.0034,
      "eval_runtime": 119.9306,
      "eval_samples_per_second": 66.664,
      "eval_steps_per_second": 8.338,
      "step": 2249
    },
    {
      "epoch": 1.1116051578479325,
      "grad_norm": 1.6910322904586792,
      "learning_rate": 3.170822756458116e-06,
      "loss": 0.0647,
      "step": 2500
    },
    {
      "epoch": 1.333926189417519,
      "grad_norm": 5.562722206115723,
      "learning_rate": 2.797521278184262e-06,
      "loss": 0.0594,
      "step": 3000
    },
    {
      "epoch": 1.5562472209871054,
      "grad_norm": 7.1707763671875,
      "learning_rate": 2.424219799910408e-06,
      "loss": 0.0467,
      "step": 3500
    },
    {
      "epoch": 1.778568252556692,
      "grad_norm": 0.07602383196353912,
      "learning_rate": 2.0509183216365537e-06,
      "loss": 0.0498,
      "step": 4000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9787367104440275,
      "eval_loss": 0.06474132835865021,
      "eval_model_preparation_time": 0.0034,
      "eval_runtime": 115.2384,
      "eval_samples_per_second": 69.378,
      "eval_steps_per_second": 8.678,
      "step": 4498
    }
  ],
  "logging_steps": 500,
  "max_steps": 6747,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1151872225435935e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}