{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.3333333333333335,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 2.3431689739227295,
      "learning_rate": 2e-05,
      "loss": 1.4244,
      "mean_token_accuracy": 0.6533483505249024,
      "step": 5
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 1.3162345886230469,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 0.8059,
      "mean_token_accuracy": 0.7710338115692139,
      "step": 10
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.3825139999389648,
      "learning_rate": 1.766044443118978e-05,
      "loss": 0.7155,
      "mean_token_accuracy": 0.7817506909370422,
      "step": 15
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 1.2070649862289429,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.515,
      "mean_token_accuracy": 0.8140048146247864,
      "step": 20
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 1.0796136856079102,
      "learning_rate": 1.1736481776669307e-05,
      "loss": 0.508,
      "mean_token_accuracy": 0.8207547307014466,
      "step": 25
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3636045455932617,
      "learning_rate": 8.263518223330698e-06,
      "loss": 0.4524,
      "mean_token_accuracy": 0.8510397672653198,
      "step": 30
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.7634139060974121,
      "learning_rate": 5.000000000000003e-06,
      "loss": 0.3877,
      "mean_token_accuracy": 0.8616405725479126,
      "step": 35
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.9971669316291809,
      "learning_rate": 2.339555568810221e-06,
      "loss": 0.359,
      "mean_token_accuracy": 0.8703584909439087,
      "step": 40
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.9843636751174927,
      "learning_rate": 6.030737921409169e-07,
      "loss": 0.4023,
      "mean_token_accuracy": 0.8350511193275452,
      "step": 45
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.7454453706741333,
      "learning_rate": 0.0,
      "loss": 0.3313,
      "mean_token_accuracy": 0.8710624217987061,
      "step": 50
    },
    {
      "epoch": 3.3333333333333335,
      "step": 50,
      "total_flos": 3571493896192.0,
      "train_loss": 0.5901590967178345,
      "train_runtime": 1292.5059,
      "train_samples_per_second": 0.309,
      "train_steps_per_second": 0.039
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3571493896192.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}