{
  "best_metric": 0.9052631578947369,
  "best_model_checkpoint": "Project_Class/checkpoint-24",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 72,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6842105263157895,
      "eval_loss": 1.1297107934951782,
      "eval_runtime": 11.4206,
      "eval_samples_per_second": 8.318,
      "eval_steps_per_second": 0.525,
      "step": 6
    },
    {
      "epoch": 1.67,
      "grad_norm": 3.2869439125061035,
      "learning_rate": 3.571428571428572e-05,
      "loss": 0.9171,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7684210526315789,
      "eval_loss": 0.8037226796150208,
      "eval_runtime": 11.2098,
      "eval_samples_per_second": 8.475,
      "eval_steps_per_second": 0.535,
      "step": 12
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8736842105263158,
      "eval_loss": 0.5411734580993652,
      "eval_runtime": 10.9538,
      "eval_samples_per_second": 8.673,
      "eval_steps_per_second": 0.548,
      "step": 18
    },
    {
      "epoch": 3.33,
      "grad_norm": 1.348880648612976,
      "learning_rate": 4.745762711864407e-05,
      "loss": 0.3868,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9052631578947369,
      "eval_loss": 0.40499913692474365,
      "eval_runtime": 11.3903,
      "eval_samples_per_second": 8.34,
      "eval_steps_per_second": 0.527,
      "step": 24
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6093498468399048,
      "learning_rate": 4.3220338983050854e-05,
      "loss": 0.0723,
      "step": 30
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8842105263157894,
      "eval_loss": 0.3822001516819,
      "eval_runtime": 10.5528,
      "eval_samples_per_second": 9.002,
      "eval_steps_per_second": 0.569,
      "step": 30
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9052631578947369,
      "eval_loss": 0.36712944507598877,
      "eval_runtime": 11.5724,
      "eval_samples_per_second": 8.209,
      "eval_steps_per_second": 0.518,
      "step": 36
    },
    {
      "epoch": 6.67,
      "grad_norm": 0.1124178022146225,
      "learning_rate": 3.898305084745763e-05,
      "loss": 0.0167,
      "step": 40
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9052631578947369,
      "eval_loss": 0.36179861426353455,
      "eval_runtime": 17.1325,
      "eval_samples_per_second": 5.545,
      "eval_steps_per_second": 0.35,
      "step": 42
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8947368421052632,
      "eval_loss": 0.36193060874938965,
      "eval_runtime": 10.5666,
      "eval_samples_per_second": 8.991,
      "eval_steps_per_second": 0.568,
      "step": 48
    },
    {
      "epoch": 8.33,
      "grad_norm": 0.06889068335294724,
      "learning_rate": 3.474576271186441e-05,
      "loss": 0.0074,
      "step": 50
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8947368421052632,
      "eval_loss": 0.36159735918045044,
      "eval_runtime": 10.8579,
      "eval_samples_per_second": 8.749,
      "eval_steps_per_second": 0.553,
      "step": 54
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.04152742773294449,
      "learning_rate": 3.050847457627119e-05,
      "loss": 0.0046,
      "step": 60
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8947368421052632,
      "eval_loss": 0.3610941469669342,
      "eval_runtime": 10.8273,
      "eval_samples_per_second": 8.774,
      "eval_steps_per_second": 0.554,
      "step": 60
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.8947368421052632,
      "eval_loss": 0.36108464002609253,
      "eval_runtime": 10.8242,
      "eval_samples_per_second": 8.777,
      "eval_steps_per_second": 0.554,
      "step": 66
    },
    {
      "epoch": 11.67,
      "grad_norm": 0.032729245722293854,
      "learning_rate": 2.627118644067797e-05,
      "loss": 0.0036,
      "step": 70
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8947368421052632,
      "eval_loss": 0.36162590980529785,
      "eval_runtime": 10.8377,
      "eval_samples_per_second": 8.766,
      "eval_steps_per_second": 0.554,
      "step": 72
    }
  ],
  "logging_steps": 10,
  "max_steps": 132,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 22,
  "save_steps": 500,
  "total_flos": 3.5058947142507725e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}