{
  "best_metric": 0.0444386750459671,
  "best_model_checkpoint": "572project/checkpoint-939",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4695,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.53,
      "grad_norm": 0.0010740529978647828,
      "learning_rate": 1.7870074547390843e-05,
      "loss": 0.029,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.983853606027987,
      "eval_loss": 0.0444386750459671,
      "eval_runtime": 5.0208,
      "eval_samples_per_second": 187.023,
      "eval_steps_per_second": 23.502,
      "step": 939
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.029427560046315193,
      "learning_rate": 1.5740149094781685e-05,
      "loss": 0.0461,
      "step": 1000
    },
    {
      "epoch": 1.6,
      "grad_norm": 78.75341796875,
      "learning_rate": 1.3610223642172523e-05,
      "loss": 0.0114,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.9924324324324325,
      "eval_loss": 0.054377324879169464,
      "eval_runtime": 5.0274,
      "eval_samples_per_second": 186.775,
      "eval_steps_per_second": 23.471,
      "step": 1878
    },
    {
      "epoch": 2.13,
      "grad_norm": 0.0012783489655703306,
      "learning_rate": 1.1480298189563365e-05,
      "loss": 0.0062,
      "step": 2000
    },
    {
      "epoch": 2.66,
      "grad_norm": 0.00034116956521756947,
      "learning_rate": 9.350372736954207e-06,
      "loss": 0.0013,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.9934924078091107,
      "eval_loss": 0.047857534140348434,
      "eval_runtime": 5.0032,
      "eval_samples_per_second": 187.679,
      "eval_steps_per_second": 23.585,
      "step": 2817
    },
    {
      "epoch": 3.19,
      "grad_norm": 0.00018943840404972434,
      "learning_rate": 7.220447284345049e-06,
      "loss": 0.0015,
      "step": 3000
    },
    {
      "epoch": 3.73,
      "grad_norm": 0.00020874786423519254,
      "learning_rate": 5.090521831735889e-06,
      "loss": 0.002,
      "step": 3500
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.992399565689468,
      "eval_loss": 0.054781731218099594,
      "eval_runtime": 5.0055,
      "eval_samples_per_second": 187.594,
      "eval_steps_per_second": 23.574,
      "step": 3756
    },
    {
      "epoch": 4.26,
      "grad_norm": 0.00014990718045737594,
      "learning_rate": 2.9605963791267307e-06,
      "loss": 0.002,
      "step": 4000
    },
    {
      "epoch": 4.79,
      "grad_norm": 0.00011578563135117292,
      "learning_rate": 8.306709265175719e-07,
      "loss": 0.0,
      "step": 4500
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.9913419913419914,
      "eval_loss": 0.06532972306013107,
      "eval_runtime": 5.0316,
      "eval_samples_per_second": 186.621,
      "eval_steps_per_second": 23.452,
      "step": 4695
    }
  ],
  "logging_steps": 500,
  "max_steps": 4695,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1871333823963648.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}