{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9891196834817013,
  "eval_steps": 20,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 1.9969173337331283e-05,
      "loss": 1.6723,
      "step": 1
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9238795325112867e-05,
      "loss": 1.4829,
      "step": 5
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.7071067811865477e-05,
      "loss": 1.3734,
      "step": 10
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.3826834323650899e-05,
      "loss": 1.3486,
      "step": 15
    },
    {
      "epoch": 0.49,
      "learning_rate": 1e-05,
      "loss": 1.3274,
      "step": 20
    },
    {
      "epoch": 0.49,
      "eval_loss": 1.258691668510437,
      "eval_runtime": 2.1716,
      "eval_samples_per_second": 4.605,
      "eval_steps_per_second": 2.302,
      "step": 20
    },
    {
      "epoch": 0.62,
      "learning_rate": 6.173165676349103e-06,
      "loss": 1.2978,
      "step": 25
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.9289321881345257e-06,
      "loss": 1.3259,
      "step": 30
    },
    {
      "epoch": 0.87,
      "learning_rate": 7.612046748871327e-07,
      "loss": 1.2801,
      "step": 35
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0,
      "loss": 1.3066,
      "step": 40
    },
    {
      "epoch": 0.99,
      "eval_loss": 1.2477926015853882,
      "eval_runtime": 2.1661,
      "eval_samples_per_second": 4.617,
      "eval_steps_per_second": 2.308,
      "step": 40
    },
    {
      "epoch": 0.99,
      "step": 40,
      "total_flos": 2.6717900760940544e+16,
      "train_loss": 1.3475643575191498,
      "train_runtime": 1594.7957,
      "train_samples_per_second": 1.268,
      "train_steps_per_second": 0.025
    }
  ],
  "logging_steps": 5,
  "max_steps": 40,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "total_flos": 2.6717900760940544e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}