{
  "best_metric": 1.8443655967712402,
  "best_model_checkpoint": "./out/checkpoint-16",
  "epoch": 0.9846153846153847,
  "eval_steps": 4,
  "global_step": 16,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.1256,
      "step": 1
    },
    {
      "epoch": 0.06,
      "eval_loss": 2.164146661758423,
      "eval_runtime": 4.1785,
      "eval_samples_per_second": 6.701,
      "eval_steps_per_second": 1.675,
      "step": 1
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.1478,
      "step": 2
    },
    {
      "epoch": 0.18,
      "learning_rate": 6e-06,
      "loss": 2.1527,
      "step": 3
    },
    {
      "epoch": 0.25,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.1049,
      "step": 4
    },
    {
      "epoch": 0.25,
      "eval_loss": 2.125430107116699,
      "eval_runtime": 4.2671,
      "eval_samples_per_second": 6.562,
      "eval_steps_per_second": 1.64,
      "step": 4
    },
    {
      "epoch": 0.31,
      "learning_rate": 1e-05,
      "loss": 2.0978,
      "step": 5
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.2e-05,
      "loss": 2.0632,
      "step": 6
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 2.0281,
      "step": 7
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.9826,
      "step": 8
    },
    {
      "epoch": 0.49,
      "eval_loss": 1.9868038892745972,
      "eval_runtime": 4.2896,
      "eval_samples_per_second": 6.527,
      "eval_steps_per_second": 1.632,
      "step": 8
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.8e-05,
      "loss": 1.964,
      "step": 9
    },
    {
      "epoch": 0.62,
      "learning_rate": 2e-05,
      "loss": 1.9839,
      "step": 10
    },
    {
      "epoch": 0.68,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 1.9244,
      "step": 11
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.4e-05,
      "loss": 1.8545,
      "step": 12
    },
    {
      "epoch": 0.74,
      "eval_loss": 1.877912163734436,
      "eval_runtime": 4.2905,
      "eval_samples_per_second": 6.526,
      "eval_steps_per_second": 1.632,
      "step": 12
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 1.894,
      "step": 13
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 1.8909,
      "step": 14
    },
    {
      "epoch": 0.92,
      "learning_rate": 3e-05,
      "loss": 1.8444,
      "step": 15
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.8597,
      "step": 16
    },
    {
      "epoch": 0.98,
      "eval_loss": 1.8443655967712402,
      "eval_runtime": 4.2861,
      "eval_samples_per_second": 6.533,
      "eval_steps_per_second": 1.633,
      "step": 16
    }
  ],
  "logging_steps": 1,
  "max_steps": 16,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 8,
  "total_flos": 4.15698116369449e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}