{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001837053366400294,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.674106732800588e-05,
      "eval_loss": 3.2056524753570557,
      "eval_runtime": 712.7003,
      "eval_samples_per_second": 16.081,
      "eval_steps_per_second": 8.041,
      "step": 1
    },
    {
      "epoch": 0.0001837053366400294,
      "grad_norm": 0.7953497767448425,
      "learning_rate": 5e-05,
      "loss": 3.0528,
      "step": 5
    },
    {
      "epoch": 0.0003674106732800588,
      "grad_norm": 0.9207623600959778,
      "learning_rate": 0.0001,
      "loss": 2.8572,
      "step": 10
    },
    {
      "epoch": 0.0003674106732800588,
      "eval_loss": 3.0712854862213135,
      "eval_runtime": 714.1932,
      "eval_samples_per_second": 16.047,
      "eval_steps_per_second": 8.024,
      "step": 10
    },
    {
      "epoch": 0.0005511160099200882,
      "grad_norm": 0.83929044008255,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.8086,
      "step": 15
    },
    {
      "epoch": 0.0007348213465601176,
      "grad_norm": 1.0277793407440186,
      "learning_rate": 8.535533905932738e-05,
      "loss": 2.6758,
      "step": 20
    },
    {
      "epoch": 0.0007348213465601176,
      "eval_loss": 2.7239773273468018,
      "eval_runtime": 715.5185,
      "eval_samples_per_second": 16.018,
      "eval_steps_per_second": 8.01,
      "step": 20
    },
    {
      "epoch": 0.000918526683200147,
      "grad_norm": 1.7762820720672607,
      "learning_rate": 6.91341716182545e-05,
      "loss": 2.546,
      "step": 25
    },
    {
      "epoch": 0.0011022320198401765,
      "grad_norm": 1.726959228515625,
      "learning_rate": 5e-05,
      "loss": 2.3199,
      "step": 30
    },
    {
      "epoch": 0.0011022320198401765,
      "eval_loss": 2.395400285720825,
      "eval_runtime": 715.3917,
      "eval_samples_per_second": 16.021,
      "eval_steps_per_second": 8.011,
      "step": 30
    },
    {
      "epoch": 0.0012859373564802058,
      "grad_norm": 1.23936128616333,
      "learning_rate": 3.086582838174551e-05,
      "loss": 2.2287,
      "step": 35
    },
    {
      "epoch": 0.0014696426931202352,
      "grad_norm": 1.6855039596557617,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 2.2798,
      "step": 40
    },
    {
      "epoch": 0.0014696426931202352,
      "eval_loss": 2.334246873855591,
      "eval_runtime": 716.0904,
      "eval_samples_per_second": 16.005,
      "eval_steps_per_second": 8.003,
      "step": 40
    },
    {
      "epoch": 0.0016533480297602646,
      "grad_norm": 1.3344709873199463,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 2.2068,
      "step": 45
    },
    {
      "epoch": 0.001837053366400294,
      "grad_norm": 1.6728562116622925,
      "learning_rate": 0.0,
      "loss": 2.4558,
      "step": 50
    },
    {
      "epoch": 0.001837053366400294,
      "eval_loss": 2.327793598175049,
      "eval_runtime": 713.9009,
      "eval_samples_per_second": 16.054,
      "eval_steps_per_second": 8.028,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8713180230451200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}