{
  "best_metric": 0.32138827443122864,
  "best_model_checkpoint": "outputs/checkpoint-8",
  "epoch": 2.0,
  "eval_steps": 1,
  "global_step": 8,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "grad_norm": 4.28125,
      "learning_rate": 0.001,
      "loss": 4.4445,
      "step": 1
    },
    {
      "epoch": 0.25,
      "eval_loss": 3.3636646270751953,
      "eval_runtime": 0.4504,
      "eval_samples_per_second": 104.352,
      "eval_steps_per_second": 13.322,
      "step": 1
    },
    {
      "epoch": 0.5,
      "grad_norm": 3.703125,
      "learning_rate": 0.001,
      "loss": 3.3693,
      "step": 2
    },
    {
      "epoch": 0.5,
      "eval_loss": 1.7100876569747925,
      "eval_runtime": 0.4446,
      "eval_samples_per_second": 105.702,
      "eval_steps_per_second": 13.494,
      "step": 2
    },
    {
      "epoch": 0.75,
      "grad_norm": 5.15625,
      "learning_rate": 0.001,
      "loss": 1.7172,
      "step": 3
    },
    {
      "epoch": 0.75,
      "eval_loss": 1.4450939893722534,
      "eval_runtime": 0.4448,
      "eval_samples_per_second": 105.657,
      "eval_steps_per_second": 13.488,
      "step": 3
    },
    {
      "epoch": 1.0,
      "grad_norm": 20.5,
      "learning_rate": 0.001,
      "loss": 1.4513,
      "step": 4
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8471462726593018,
      "eval_runtime": 0.4422,
      "eval_samples_per_second": 106.291,
      "eval_steps_per_second": 13.569,
      "step": 4
    },
    {
      "epoch": 1.25,
      "grad_norm": 11.1875,
      "learning_rate": 0.001,
      "loss": 0.8509,
      "step": 5
    },
    {
      "epoch": 1.25,
      "eval_loss": 0.5828761458396912,
      "eval_runtime": 0.4434,
      "eval_samples_per_second": 105.988,
      "eval_steps_per_second": 13.53,
      "step": 5
    },
    {
      "epoch": 1.5,
      "grad_norm": 9.875,
      "learning_rate": 0.001,
      "loss": 0.5781,
      "step": 6
    },
    {
      "epoch": 1.5,
      "eval_loss": 0.8648306727409363,
      "eval_runtime": 0.4441,
      "eval_samples_per_second": 105.821,
      "eval_steps_per_second": 13.509,
      "step": 6
    },
    {
      "epoch": 1.75,
      "grad_norm": 15.375,
      "learning_rate": 0.001,
      "loss": 0.8642,
      "step": 7
    },
    {
      "epoch": 1.75,
      "eval_loss": 0.4214279353618622,
      "eval_runtime": 0.4424,
      "eval_samples_per_second": 106.229,
      "eval_steps_per_second": 13.561,
      "step": 7
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.78125,
      "learning_rate": 0.001,
      "loss": 0.4199,
      "step": 8
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.32138827443122864,
      "eval_runtime": 0.4418,
      "eval_samples_per_second": 106.379,
      "eval_steps_per_second": 13.58,
      "step": 8
    },
    {
      "epoch": 2.0,
      "step": 8,
      "total_flos": 1329758335401984.0,
      "train_loss": 1.7119178622961044,
      "train_runtime": 46.9732,
      "train_samples_per_second": 19.628,
      "train_steps_per_second": 0.17
    }
  ],
  "logging_steps": 1,
  "max_steps": 8,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1,
  "total_flos": 1329758335401984.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}