{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 5.201437473297119,
      "learning_rate": 4.9e-05,
      "loss": 2.6116,
      "step": 100
    },
    {
      "epoch": 0.16,
      "grad_norm": 4.178571701049805,
      "learning_rate": 4.8e-05,
      "loss": 1.9634,
      "step": 200
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.5103917121887207,
      "learning_rate": 4.7e-05,
      "loss": 1.6752,
      "step": 300
    },
    {
      "epoch": 0.32,
      "grad_norm": 4.7221550941467285,
      "learning_rate": 4.600000000000001e-05,
      "loss": 1.483,
      "step": 400
    },
    {
      "epoch": 0.4,
      "grad_norm": 4.4923272132873535,
      "learning_rate": 4.5e-05,
      "loss": 1.3296,
      "step": 500
    },
    {
      "epoch": 0.48,
      "grad_norm": 6.077773571014404,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.1978,
      "step": 600
    },
    {
      "epoch": 0.56,
      "grad_norm": 2.388939142227173,
      "learning_rate": 4.3e-05,
      "loss": 1.1306,
      "step": 700
    },
    {
      "epoch": 0.64,
      "grad_norm": 2.460210084915161,
      "learning_rate": 4.201e-05,
      "loss": 1.0513,
      "step": 800
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.139218330383301,
      "learning_rate": 4.101e-05,
      "loss": 0.9612,
      "step": 900
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.382781505584717,
      "learning_rate": 4.0010000000000005e-05,
      "loss": 0.8977,
      "step": 1000
    },
    {
      "epoch": 0.88,
      "grad_norm": 6.498015403747559,
      "learning_rate": 3.902e-05,
      "loss": 0.8653,
      "step": 1100
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.856051445007324,
      "learning_rate": 3.802e-05,
      "loss": 0.8131,
      "step": 1200
    }
  ],
  "logging_steps": 100,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 1.917794423144448e+20,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}