{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.928,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.096,
      "grad_norm": 2.7530338764190674,
      "learning_rate": 1e-05,
      "loss": 0.7746,
      "step": 1
    },
    {
      "epoch": 0.192,
      "grad_norm": 2.5897934436798096,
      "learning_rate": 2e-05,
      "loss": 0.7855,
      "step": 2
    },
    {
      "epoch": 0.288,
      "grad_norm": 1.4067482948303223,
      "learning_rate": 1.9848077530122083e-05,
      "loss": 0.7329,
      "step": 3
    },
    {
      "epoch": 0.384,
      "grad_norm": 1.708691954612732,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 0.6941,
      "step": 4
    },
    {
      "epoch": 0.48,
      "grad_norm": 2.313199758529663,
      "learning_rate": 1.866025403784439e-05,
      "loss": 0.6372,
      "step": 5
    },
    {
      "epoch": 0.576,
      "grad_norm": 1.4410549402236938,
      "learning_rate": 1.766044443118978e-05,
      "loss": 0.6693,
      "step": 6
    },
    {
      "epoch": 0.672,
      "grad_norm": 1.3163059949874878,
      "learning_rate": 1.6427876096865394e-05,
      "loss": 0.5952,
      "step": 7
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.9859979748725891,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.6078,
      "step": 8
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.8045167326927185,
      "learning_rate": 1.342020143325669e-05,
      "loss": 0.5778,
      "step": 9
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.8406056761741638,
      "learning_rate": 1.1736481776669307e-05,
      "loss": 0.5891,
      "step": 10
    },
    {
      "epoch": 1.064,
      "grad_norm": 1.2650195360183716,
      "learning_rate": 1e-05,
      "loss": 0.8869,
      "step": 11
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.5438147187232971,
      "learning_rate": 8.263518223330698e-06,
      "loss": 0.5072,
      "step": 12
    },
    {
      "epoch": 1.256,
      "grad_norm": 0.44890961050987244,
      "learning_rate": 6.579798566743314e-06,
      "loss": 0.4801,
      "step": 13
    },
    {
      "epoch": 1.3519999999999999,
      "grad_norm": 0.6564487814903259,
      "learning_rate": 5.000000000000003e-06,
      "loss": 0.561,
      "step": 14
    },
    {
      "epoch": 1.448,
      "grad_norm": 0.5231355428695679,
      "learning_rate": 3.5721239031346067e-06,
      "loss": 0.5224,
      "step": 15
    },
    {
      "epoch": 1.544,
      "grad_norm": 0.5453435778617859,
      "learning_rate": 2.339555568810221e-06,
      "loss": 0.4994,
      "step": 16
    },
    {
      "epoch": 1.6400000000000001,
      "grad_norm": 0.5144322514533997,
      "learning_rate": 1.339745962155613e-06,
      "loss": 0.5113,
      "step": 17
    },
    {
      "epoch": 1.736,
      "grad_norm": 0.493093878030777,
      "learning_rate": 6.030737921409169e-07,
      "loss": 0.5048,
      "step": 18
    },
    {
      "epoch": 1.8319999999999999,
      "grad_norm": 0.44925519824028015,
      "learning_rate": 1.519224698779198e-07,
      "loss": 0.4894,
      "step": 19
    },
    {
      "epoch": 1.928,
      "grad_norm": 0.4307903051376343,
      "learning_rate": 0.0,
      "loss": 0.475,
      "step": 20
    },
    {
      "epoch": 1.928,
      "step": 20,
      "total_flos": 68537506594816.0,
      "train_loss": 0.605039681494236,
      "train_runtime": 7309.421,
      "train_samples_per_second": 0.274,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 68537506594816.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}