{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.987531172069826,
  "eval_steps": 500,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.20781379883624274,
      "grad_norm": 0.7872959971427917,
      "learning_rate": 1.917206982543641e-05,
      "loss": 0.7834,
      "step": 500
    },
    {
      "epoch": 0.41562759767248547,
      "grad_norm": 0.9406130313873291,
      "learning_rate": 1.834081463009144e-05,
      "loss": 0.4895,
      "step": 1000
    },
    {
      "epoch": 0.6234413965087282,
      "grad_norm": 0.4806763529777527,
      "learning_rate": 1.7509559434746467e-05,
      "loss": 0.4473,
      "step": 1500
    },
    {
      "epoch": 0.8312551953449709,
      "grad_norm": 0.5443118214607239,
      "learning_rate": 1.6678304239401496e-05,
      "loss": 0.4326,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.33564111590385437,
      "eval_runtime": 26.7564,
      "eval_samples_per_second": 18.463,
      "eval_steps_per_second": 4.634,
      "step": 2406
    },
    {
      "epoch": 1.0390689941812137,
      "grad_norm": 0.7427258491516113,
      "learning_rate": 1.5847049044056525e-05,
      "loss": 0.4225,
      "step": 2500
    },
    {
      "epoch": 1.2468827930174564,
      "grad_norm": 0.4491533637046814,
      "learning_rate": 1.5015793848711555e-05,
      "loss": 0.3857,
      "step": 3000
    },
    {
      "epoch": 1.4546965918536992,
      "grad_norm": 0.4072982668876648,
      "learning_rate": 1.4184538653366584e-05,
      "loss": 0.3754,
      "step": 3500
    },
    {
      "epoch": 1.6625103906899419,
      "grad_norm": 0.8853089213371277,
      "learning_rate": 1.3353283458021613e-05,
      "loss": 0.3765,
      "step": 4000
    },
    {
      "epoch": 1.8703241895261846,
      "grad_norm": 0.59534752368927,
      "learning_rate": 1.2522028262676643e-05,
      "loss": 0.3677,
      "step": 4500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.3049054443836212,
      "eval_runtime": 26.7561,
      "eval_samples_per_second": 18.463,
      "eval_steps_per_second": 4.634,
      "step": 4812
    },
    {
      "epoch": 2.0781379883624274,
      "grad_norm": 0.5839773416519165,
      "learning_rate": 1.1690773067331672e-05,
      "loss": 0.3488,
      "step": 5000
    },
    {
      "epoch": 2.28595178719867,
      "grad_norm": 0.8203754425048828,
      "learning_rate": 1.086118038237739e-05,
      "loss": 0.3526,
      "step": 5500
    },
    {
      "epoch": 2.493765586034913,
      "grad_norm": 0.7811100482940674,
      "learning_rate": 1.0029925187032419e-05,
      "loss": 0.3438,
      "step": 6000
    },
    {
      "epoch": 2.7015793848711556,
      "grad_norm": 0.823439359664917,
      "learning_rate": 9.19866999168745e-06,
      "loss": 0.3406,
      "step": 6500
    },
    {
      "epoch": 2.9093931837073983,
      "grad_norm": 0.5342724323272705,
      "learning_rate": 8.367414796342478e-06,
      "loss": 0.3333,
      "step": 7000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.2934778034687042,
      "eval_runtime": 26.7608,
      "eval_samples_per_second": 18.46,
      "eval_steps_per_second": 4.634,
      "step": 7218
    },
    {
      "epoch": 3.117206982543641,
      "grad_norm": 0.4342059791088104,
      "learning_rate": 7.536159600997507e-06,
      "loss": 0.3313,
      "step": 7500
    },
    {
      "epoch": 3.3250207813798838,
      "grad_norm": 0.9092524647712708,
      "learning_rate": 6.704904405652536e-06,
      "loss": 0.3265,
      "step": 8000
    },
    {
      "epoch": 3.5328345802161265,
      "grad_norm": 0.7057175040245056,
      "learning_rate": 5.873649210307565e-06,
      "loss": 0.3219,
      "step": 8500
    },
    {
      "epoch": 3.7406483790523692,
      "grad_norm": 0.8144080638885498,
      "learning_rate": 5.0423940149625935e-06,
      "loss": 0.316,
      "step": 9000
    },
    {
      "epoch": 3.9484621778886115,
      "grad_norm": 0.5943393707275391,
      "learning_rate": 4.212801330008313e-06,
      "loss": 0.3155,
      "step": 9500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.2871634364128113,
      "eval_runtime": 26.8766,
      "eval_samples_per_second": 18.38,
      "eval_steps_per_second": 4.614,
      "step": 9624
    },
    {
      "epoch": 4.156275976724855,
      "grad_norm": 1.265112280845642,
      "learning_rate": 3.381546134663342e-06,
      "loss": 0.3036,
      "step": 10000
    },
    {
      "epoch": 4.364089775561097,
      "grad_norm": 0.7941420078277588,
      "learning_rate": 2.550290939318371e-06,
      "loss": 0.3129,
      "step": 10500
    },
    {
      "epoch": 4.57190357439734,
      "grad_norm": 0.956358015537262,
      "learning_rate": 1.7190357439733998e-06,
      "loss": 0.3166,
      "step": 11000
    },
    {
      "epoch": 4.7797173732335825,
      "grad_norm": 0.7600696682929993,
      "learning_rate": 8.89443059019119e-07,
      "loss": 0.3117,
      "step": 11500
    },
    {
      "epoch": 4.987531172069826,
      "grad_norm": 0.9830431342124939,
      "learning_rate": 5.8187863674147975e-08,
      "loss": 0.3083,
      "step": 12000
    }
  ],
  "logging_steps": 500,
  "max_steps": 12030,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.922754291531776e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}