{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.090909090909092,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18,
      "grad_norm": 3220.643835599575,
      "learning_rate": 4.000000000000001e-06,
      "loss": 14.9516,
      "step": 1
    },
    {
      "epoch": 0.91,
      "grad_norm": 411.2907564885413,
      "learning_rate": 2e-05,
      "loss": 10.4256,
      "step": 5
    },
    {
      "epoch": 0.91,
      "eval_loss": 47.00014877319336,
      "eval_runtime": 2.5065,
      "eval_samples_per_second": 4.389,
      "eval_steps_per_second": 0.399,
      "step": 5
    },
    {
      "epoch": 1.82,
      "grad_norm": 87.6083201371645,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 6.0419,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 43.96914291381836,
      "eval_runtime": 2.6958,
      "eval_samples_per_second": 4.08,
      "eval_steps_per_second": 0.371,
      "step": 11
    },
    {
      "epoch": 2.73,
      "grad_norm": 328.84229322130653,
      "learning_rate": 1.766044443118978e-05,
      "loss": 5.2838,
      "step": 15
    },
    {
      "epoch": 2.91,
      "eval_loss": 40.78567123413086,
      "eval_runtime": 2.4992,
      "eval_samples_per_second": 4.401,
      "eval_steps_per_second": 0.4,
      "step": 16
    },
    {
      "epoch": 3.64,
      "grad_norm": 138.82218027827275,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 4.8705,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_loss": 33.928184509277344,
      "eval_runtime": 2.6726,
      "eval_samples_per_second": 4.116,
      "eval_steps_per_second": 0.374,
      "step": 22
    },
    {
      "epoch": 4.55,
      "grad_norm": 77.94936474819474,
      "learning_rate": 1.1736481776669307e-05,
      "loss": 4.196,
      "step": 25
    },
    {
      "epoch": 4.91,
      "eval_loss": 17.533571243286133,
      "eval_runtime": 2.5096,
      "eval_samples_per_second": 4.383,
      "eval_steps_per_second": 0.398,
      "step": 27
    },
    {
      "epoch": 5.45,
      "grad_norm": 64.81716616301149,
      "learning_rate": 8.263518223330698e-06,
      "loss": 3.0724,
      "step": 30
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.7088449001312256,
      "eval_runtime": 2.6734,
      "eval_samples_per_second": 4.115,
      "eval_steps_per_second": 0.374,
      "step": 33
    },
    {
      "epoch": 6.36,
      "grad_norm": 14.389103750009635,
      "learning_rate": 5.000000000000003e-06,
      "loss": 2.1966,
      "step": 35
    },
    {
      "epoch": 6.91,
      "eval_loss": 2.7433879375457764,
      "eval_runtime": 2.4975,
      "eval_samples_per_second": 4.404,
      "eval_steps_per_second": 0.4,
      "step": 38
    },
    {
      "epoch": 7.27,
      "grad_norm": 18.34112469972439,
      "learning_rate": 2.339555568810221e-06,
      "loss": 2.1116,
      "step": 40
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.7265045642852783,
      "eval_runtime": 2.6806,
      "eval_samples_per_second": 4.104,
      "eval_steps_per_second": 0.373,
      "step": 44
    },
    {
      "epoch": 8.18,
      "grad_norm": 8.674981290289173,
      "learning_rate": 6.030737921409169e-07,
      "loss": 2.0641,
      "step": 45
    },
    {
      "epoch": 8.91,
      "eval_loss": 2.716779947280884,
      "eval_runtime": 2.4827,
      "eval_samples_per_second": 4.431,
      "eval_steps_per_second": 0.403,
      "step": 49
    },
    {
      "epoch": 9.09,
      "grad_norm": 9.524348161964365,
      "learning_rate": 0.0,
      "loss": 2.0467,
      "step": 50
    },
    {
      "epoch": 9.09,
      "eval_loss": 2.725921869277954,
      "eval_runtime": 2.669,
      "eval_samples_per_second": 4.121,
      "eval_steps_per_second": 0.375,
      "step": 50
    },
    {
      "epoch": 9.09,
      "step": 50,
      "total_flos": 6816515751936.0,
      "train_loss": 4.321450042724609,
      "train_runtime": 737.8878,
      "train_samples_per_second": 8.863,
      "train_steps_per_second": 0.068
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000000000,
  "total_flos": 6816515751936.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}