{
  "best_metric": 2.5784292221069336,
  "best_model_checkpoint": "./outputs/checkpoint-500",
  "epoch": 0.36429872495446264,
  "eval_steps": 100,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 2.8179,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.723731517791748,
      "eval_runtime": 208.4619,
      "eval_samples_per_second": 30.097,
      "eval_steps_per_second": 3.766,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.6952,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.6728808879852295,
      "eval_runtime": 204.3784,
      "eval_samples_per_second": 30.698,
      "eval_steps_per_second": 3.841,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.6508,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.6387226581573486,
      "eval_runtime": 205.5464,
      "eval_samples_per_second": 30.524,
      "eval_steps_per_second": 3.819,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 2.6247,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 2.6044600009918213,
      "eval_runtime": 204.5115,
      "eval_samples_per_second": 30.678,
      "eval_steps_per_second": 3.838,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 2.5814,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 2.5784292221069336,
      "eval_runtime": 204.5265,
      "eval_samples_per_second": 30.676,
      "eval_steps_per_second": 3.838,
      "step": 500
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.424403231326208e+16,
  "trial_name": null,
  "trial_params": null
}