{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.090909090909092,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18,
      "grad_norm": 1.9203362743829422,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.0754,
      "step": 1
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.2522979550420272,
      "learning_rate": 2e-05,
      "loss": 2.0634,
      "step": 5
    },
    {
      "epoch": 0.91,
      "eval_loss": 1.912642002105713,
      "eval_runtime": 2.1098,
      "eval_samples_per_second": 5.214,
      "eval_steps_per_second": 0.474,
      "step": 5
    },
    {
      "epoch": 1.82,
      "grad_norm": 1.0376267934668848,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 1.9281,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.8792613744735718,
      "eval_runtime": 2.0802,
      "eval_samples_per_second": 5.288,
      "eval_steps_per_second": 0.481,
      "step": 11
    },
    {
      "epoch": 2.73,
      "grad_norm": 0.807859290768886,
      "learning_rate": 1.766044443118978e-05,
      "loss": 1.7541,
      "step": 15
    },
    {
      "epoch": 2.91,
      "eval_loss": 2.2713067531585693,
      "eval_runtime": 1.9404,
      "eval_samples_per_second": 5.669,
      "eval_steps_per_second": 0.515,
      "step": 16
    },
    {
      "epoch": 3.64,
      "grad_norm": 0.6587224869045021,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 1.5669,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.2286932468414307,
      "eval_runtime": 2.057,
      "eval_samples_per_second": 5.348,
      "eval_steps_per_second": 0.486,
      "step": 22
    },
    {
      "epoch": 4.55,
      "grad_norm": 0.6858567432222655,
      "learning_rate": 1.1736481776669307e-05,
      "loss": 1.3976,
      "step": 25
    },
    {
      "epoch": 4.91,
      "eval_loss": 2.265625,
      "eval_runtime": 1.934,
      "eval_samples_per_second": 5.688,
      "eval_steps_per_second": 0.517,
      "step": 27
    },
    {
      "epoch": 5.45,
      "grad_norm": 0.8265868923051504,
      "learning_rate": 8.263518223330698e-06,
      "loss": 1.2434,
      "step": 30
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.34375,
      "eval_runtime": 2.0742,
      "eval_samples_per_second": 5.303,
      "eval_steps_per_second": 0.482,
      "step": 33
    },
    {
      "epoch": 6.36,
      "grad_norm": 0.7945639183231874,
      "learning_rate": 5.000000000000003e-06,
      "loss": 1.1083,
      "step": 35
    },
    {
      "epoch": 6.91,
      "eval_loss": 2.3551137447357178,
      "eval_runtime": 1.9306,
      "eval_samples_per_second": 5.698,
      "eval_steps_per_second": 0.518,
      "step": 38
    },
    {
      "epoch": 7.27,
      "grad_norm": 0.5788791207539581,
      "learning_rate": 2.339555568810221e-06,
      "loss": 1.0215,
      "step": 40
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.4332387447357178,
      "eval_runtime": 2.0639,
      "eval_samples_per_second": 5.33,
      "eval_steps_per_second": 0.485,
      "step": 44
    },
    {
      "epoch": 8.18,
      "grad_norm": 0.515743831937742,
      "learning_rate": 6.030737921409169e-07,
      "loss": 0.9556,
      "step": 45
    },
    {
      "epoch": 8.91,
      "eval_loss": 2.4332387447357178,
      "eval_runtime": 1.9298,
      "eval_samples_per_second": 5.7,
      "eval_steps_per_second": 0.518,
      "step": 49
    },
    {
      "epoch": 9.09,
      "grad_norm": 0.5703656649999668,
      "learning_rate": 0.0,
      "loss": 0.9465,
      "step": 50
    },
    {
      "epoch": 9.09,
      "eval_loss": 2.4275567531585693,
      "eval_runtime": 2.0663,
      "eval_samples_per_second": 5.324,
      "eval_steps_per_second": 0.484,
      "step": 50
    },
    {
      "epoch": 9.09,
      "step": 50,
      "total_flos": 11674778075136.0,
      "train_loss": 1.39877197265625,
      "train_runtime": 578.7401,
      "train_samples_per_second": 11.629,
      "train_steps_per_second": 0.086
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000000000,
  "total_flos": 11674778075136.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}