{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 8456,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23651844843897823,
      "grad_norm": 0.62890625,
      "learning_rate": 1.881740775780511e-05,
      "loss": 0.1514,
      "mean_token_accuracy": 0.9655648044347763,
      "step": 500
    },
    {
      "epoch": 0.47303689687795647,
      "grad_norm": 0.6796875,
      "learning_rate": 1.763481551561022e-05,
      "loss": 0.0923,
      "mean_token_accuracy": 0.9776788552999497,
      "step": 1000
    },
    {
      "epoch": 0.7095553453169348,
      "grad_norm": 1.296875,
      "learning_rate": 1.6452223273415328e-05,
      "loss": 0.085,
      "mean_token_accuracy": 0.9792790929079056,
      "step": 1500
    },
    {
      "epoch": 0.9460737937559129,
      "grad_norm": 0.6171875,
      "learning_rate": 1.5269631031220437e-05,
      "loss": 0.0811,
      "mean_token_accuracy": 0.9799440746307373,
      "step": 2000
    },
    {
      "epoch": 1.1825922421948911,
      "grad_norm": 0.87109375,
      "learning_rate": 1.4087038789025544e-05,
      "loss": 0.0722,
      "mean_token_accuracy": 0.981534402012825,
      "step": 2500
    },
    {
      "epoch": 1.4191106906338695,
      "grad_norm": 1.0859375,
      "learning_rate": 1.2904446546830653e-05,
      "loss": 0.068,
      "mean_token_accuracy": 0.9825713348388672,
      "step": 3000
    },
    {
      "epoch": 1.6556291390728477,
      "grad_norm": 0.482421875,
      "learning_rate": 1.1721854304635763e-05,
      "loss": 0.0684,
      "mean_token_accuracy": 0.9822493841648102,
      "step": 3500
    },
    {
      "epoch": 1.8921475875118259,
      "grad_norm": 0.9453125,
      "learning_rate": 1.0539262062440872e-05,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9828982795476914,
      "step": 4000
    },
    {
      "epoch": 2.128666035950804,
      "grad_norm": 0.9609375,
      "learning_rate": 9.35666982024598e-06,
      "loss": 0.0627,
      "mean_token_accuracy": 0.9836636617183685,
      "step": 4500
    },
    {
      "epoch": 2.3651844843897822,
      "grad_norm": 0.71484375,
      "learning_rate": 8.174077578051088e-06,
      "loss": 0.0603,
      "mean_token_accuracy": 0.9841647505760193,
      "step": 5000
    },
    {
      "epoch": 2.6017029328287604,
      "grad_norm": 0.470703125,
      "learning_rate": 6.991485335856198e-06,
      "loss": 0.0579,
      "mean_token_accuracy": 0.9847105625867844,
      "step": 5500
    },
    {
      "epoch": 2.838221381267739,
      "grad_norm": 1.0234375,
      "learning_rate": 5.808893093661306e-06,
      "loss": 0.0593,
      "mean_token_accuracy": 0.9844530954360962,
      "step": 6000
    },
    {
      "epoch": 3.074739829706717,
      "grad_norm": 0.96484375,
      "learning_rate": 4.626300851466415e-06,
      "loss": 0.0583,
      "mean_token_accuracy": 0.984543002128601,
      "step": 6500
    },
    {
      "epoch": 3.3112582781456954,
      "grad_norm": 1.09375,
      "learning_rate": 3.443708609271523e-06,
      "loss": 0.0547,
      "mean_token_accuracy": 0.9854775066375733,
      "step": 7000
    },
    {
      "epoch": 3.5477767265846736,
      "grad_norm": 0.74609375,
      "learning_rate": 2.2611163670766324e-06,
      "loss": 0.0569,
      "mean_token_accuracy": 0.9849760442972183,
      "step": 7500
    },
    {
      "epoch": 3.7842951750236518,
      "grad_norm": 1.5390625,
      "learning_rate": 1.078524124881741e-06,
      "loss": 0.0569,
      "mean_token_accuracy": 0.9850787636041641,
      "step": 8000
    }
  ],
  "logging_steps": 500,
  "max_steps": 8456,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.359009410943795e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}