{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00812842918106076,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 1.7533113956451416,
      "learning_rate": 0.0001,
      "loss": 1.7094,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.4007155895233154,
      "learning_rate": 0.0002,
      "loss": 1.2964,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.439706563949585,
      "learning_rate": 0.00018888888888888888,
      "loss": 1.5479,
      "step": 3
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.9308393001556396,
      "learning_rate": 0.00017777777777777779,
      "loss": 1.9081,
      "step": 4
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.1223907470703125,
      "learning_rate": 0.0001666666666666667,
      "loss": 1.7825,
      "step": 5
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.7965916395187378,
      "learning_rate": 0.00015555555555555556,
      "loss": 1.0745,
      "step": 6
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.8971462249755859,
      "learning_rate": 0.00014444444444444444,
      "loss": 0.9313,
      "step": 7
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.9314714670181274,
      "learning_rate": 0.00013333333333333334,
      "loss": 1.6707,
      "step": 8
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.2980624437332153,
      "learning_rate": 0.00012222222222222224,
      "loss": 1.8406,
      "step": 9
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.1596261262893677,
      "learning_rate": 0.00011111111111111112,
      "loss": 1.4502,
      "step": 10
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.9083566069602966,
      "learning_rate": 0.0001,
      "loss": 1.3373,
      "step": 11
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.6125576496124268,
      "learning_rate": 8.888888888888889e-05,
      "loss": 1.4743,
      "step": 12
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1022965908050537,
      "learning_rate": 7.777777777777778e-05,
      "loss": 1.4054,
      "step": 13
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1945298910140991,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.8368,
      "step": 14
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.051852226257324,
      "learning_rate": 5.555555555555556e-05,
      "loss": 1.7753,
      "step": 15
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.8756366968154907,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 1.2103,
      "step": 16
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6548067331314087,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.5789,
      "step": 17
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1712427139282227,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 1.2315,
      "step": 18
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.0134000778198242,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 1.9206,
      "step": 19
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.2952359914779663,
      "learning_rate": 0.0,
      "loss": 1.5372,
      "step": 20
    },
    {
      "epoch": 0.01,
      "step": 20,
      "total_flos": 1340174549999616.0,
      "train_loss": 1.5259597092866897,
      "train_runtime": 82.7118,
      "train_samples_per_second": 0.967,
      "train_steps_per_second": 0.242
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 1340174549999616.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}