{
  "best_global_step": 7815,
  "best_metric": 0.24368683993816376,
  "best_model_checkpoint": "./bert-imdb-cp\\checkpoint-7815",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 7815,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3198976327575176,
      "grad_norm": 1.5820907354354858,
      "learning_rate": 9.361484325015996e-06,
      "loss": 0.2699,
      "step": 500
    },
    {
      "epoch": 0.6397952655150352,
      "grad_norm": 1.875124454498291,
      "learning_rate": 8.72168905950096e-06,
      "loss": 0.2901,
      "step": 1000
    },
    {
      "epoch": 0.9596928982725528,
      "grad_norm": 2.596395492553711,
      "learning_rate": 8.081893793985925e-06,
      "loss": 0.2749,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.2548043727874756,
      "eval_runtime": 276.7267,
      "eval_samples_per_second": 90.342,
      "eval_steps_per_second": 5.648,
      "step": 1563
    },
    {
      "epoch": 1.2795905310300704,
      "grad_norm": 1.6743276119232178,
      "learning_rate": 7.44209852847089e-06,
      "loss": 0.2704,
      "step": 2000
    },
    {
      "epoch": 1.599488163787588,
      "grad_norm": 2.1094255447387695,
      "learning_rate": 6.802303262955855e-06,
      "loss": 0.2619,
      "step": 2500
    },
    {
      "epoch": 1.9193857965451055,
      "grad_norm": 1.2479547262191772,
      "learning_rate": 6.16250799744082e-06,
      "loss": 0.2723,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.2492910772562027,
      "eval_runtime": 271.0728,
      "eval_samples_per_second": 92.226,
      "eval_steps_per_second": 5.766,
      "step": 3126
    },
    {
      "epoch": 2.239283429302623,
      "grad_norm": 2.0263354778289795,
      "learning_rate": 5.522712731925784e-06,
      "loss": 0.2574,
      "step": 3500
    },
    {
      "epoch": 2.5591810620601407,
      "grad_norm": 0.6032302379608154,
      "learning_rate": 4.882917466410749e-06,
      "loss": 0.2674,
      "step": 4000
    },
    {
      "epoch": 2.8790786948176583,
      "grad_norm": 0.584612250328064,
      "learning_rate": 4.2444017914267435e-06,
      "loss": 0.271,
      "step": 4500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.2437499761581421,
      "eval_runtime": 265.4633,
      "eval_samples_per_second": 94.175,
      "eval_steps_per_second": 5.888,
      "step": 4689
    },
    {
      "epoch": 3.198976327575176,
      "grad_norm": 1.4420491456985474,
      "learning_rate": 3.6046065259117086e-06,
      "loss": 0.27,
      "step": 5000
    },
    {
      "epoch": 3.5188739603326935,
      "grad_norm": 1.480359673500061,
      "learning_rate": 2.9648112603966733e-06,
      "loss": 0.2678,
      "step": 5500
    },
    {
      "epoch": 3.838771593090211,
      "grad_norm": 1.3382619619369507,
      "learning_rate": 2.325015994881638e-06,
      "loss": 0.2581,
      "step": 6000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.2446114867925644,
      "eval_runtime": 271.3697,
      "eval_samples_per_second": 92.125,
      "eval_steps_per_second": 5.76,
      "step": 6252
    },
    {
      "epoch": 4.158669225847729,
      "grad_norm": 1.3025797605514526,
      "learning_rate": 1.6865003198976329e-06,
      "loss": 0.2558,
      "step": 6500
    },
    {
      "epoch": 4.478566858605246,
      "grad_norm": 1.6497241258621216,
      "learning_rate": 1.0467050543825978e-06,
      "loss": 0.2623,
      "step": 7000
    },
    {
      "epoch": 4.798464491362764,
      "grad_norm": 1.4709057807922363,
      "learning_rate": 4.0690978886756244e-07,
      "loss": 0.2743,
      "step": 7500
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.24368683993816376,
      "eval_runtime": 271.3467,
      "eval_samples_per_second": 92.133,
      "eval_steps_per_second": 5.76,
      "step": 7815
    }
  ],
  "logging_steps": 500,
  "max_steps": 7815,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 3.253872632752848e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}