{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 16.0,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8,
      "learning_rate": 4e-05,
      "loss": 3.2717,
      "step": 1
    },
    {
      "epoch": 1.6,
      "learning_rate": 3.972722606805445e-05,
      "loss": 3.1418,
      "step": 2
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.8916344834012695e-05,
      "loss": 1.8546,
      "step": 3
    },
    {
      "epoch": 3.2,
      "learning_rate": 3.758947502412978e-05,
      "loss": 1.7453,
      "step": 4
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.578281018792788e-05,
      "loss": 0.9852,
      "step": 5
    },
    {
      "epoch": 4.8,
      "learning_rate": 3.354563143251483e-05,
      "loss": 1.2799,
      "step": 6
    },
    {
      "epoch": 5.6,
      "learning_rate": 3.093896316244855e-05,
      "loss": 0.6405,
      "step": 7
    },
    {
      "epoch": 6.4,
      "learning_rate": 2.8033908493059394e-05,
      "loss": 0.5262,
      "step": 8
    },
    {
      "epoch": 7.2,
      "learning_rate": 2.4909709742815986e-05,
      "loss": 0.3255,
      "step": 9
    },
    {
      "epoch": 8.0,
      "learning_rate": 2.165158690944665e-05,
      "loss": 0.2122,
      "step": 10
    },
    {
      "epoch": 8.8,
      "learning_rate": 1.8348413090553356e-05,
      "loss": 0.1912,
      "step": 11
    },
    {
      "epoch": 9.6,
      "learning_rate": 1.5090290257184019e-05,
      "loss": 0.2122,
      "step": 12
    },
    {
      "epoch": 10.4,
      "learning_rate": 1.1966091506940616e-05,
      "loss": 0.1126,
      "step": 13
    },
    {
      "epoch": 11.2,
      "learning_rate": 9.061036837551467e-06,
      "loss": 0.0647,
      "step": 14
    },
    {
      "epoch": 12.0,
      "learning_rate": 6.4543685674851834e-06,
      "loss": 0.0087,
      "step": 15
    },
    {
      "epoch": 12.8,
      "learning_rate": 4.217189812072131e-06,
      "loss": 0.0087,
      "step": 16
    },
    {
      "epoch": 13.6,
      "learning_rate": 2.410524975870221e-06,
      "loss": 0.0036,
      "step": 17
    },
    {
      "epoch": 14.4,
      "learning_rate": 1.0836551659873073e-06,
      "loss": 0.0028,
      "step": 18
    },
    {
      "epoch": 15.2,
      "learning_rate": 2.7277393194555357e-07,
      "loss": 0.0016,
      "step": 19
    },
    {
      "epoch": 16.0,
      "learning_rate": 0.0,
      "loss": 0.0006,
      "step": 20
    },
    {
      "epoch": 16.0,
      "step": 20,
      "total_flos": 1.4917407062462824e+19,
      "train_loss": 0.7294653365388513,
      "train_runtime": 705.176,
      "train_samples_per_second": 0.567,
      "train_steps_per_second": 0.028
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 4,
  "total_flos": 1.4917407062462824e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}