{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.867383512544803,
  "eval_steps": 2000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "grad_norm": 2.5333876609802246,
      "learning_rate": 9.9e-07,
      "loss": 0.9981,
      "step": 100
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.1556735038757324,
      "learning_rate": 9.9e-07,
      "loss": 0.7461,
      "step": 200
    },
    {
      "epoch": 0.22,
      "grad_norm": 2.5188281536102295,
      "learning_rate": 9.7989898989899e-07,
      "loss": 0.6707,
      "step": 300
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.559631824493408,
      "learning_rate": 9.697979797979798e-07,
      "loss": 0.6281,
      "step": 400
    },
    {
      "epoch": 0.36,
      "grad_norm": 3.3940675258636475,
      "learning_rate": 9.596969696969696e-07,
      "loss": 0.6437,
      "step": 500
    },
    {
      "epoch": 0.43,
      "grad_norm": 2.0749378204345703,
      "learning_rate": 9.495959595959595e-07,
      "loss": 0.6319,
      "step": 600
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.3027114868164062,
      "learning_rate": 9.394949494949495e-07,
      "loss": 0.6477,
      "step": 700
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.6429717540740967,
      "learning_rate": 9.293939393939394e-07,
      "loss": 0.6564,
      "step": 800
    },
    {
      "epoch": 0.65,
      "grad_norm": 2.391716718673706,
      "learning_rate": 9.192929292929292e-07,
      "loss": 0.6263,
      "step": 900
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.9794795513153076,
      "learning_rate": 9.091919191919192e-07,
      "loss": 0.6422,
      "step": 1000
    },
    {
      "epoch": 0.79,
      "grad_norm": 2.853337049484253,
      "learning_rate": 8.99090909090909e-07,
      "loss": 0.6043,
      "step": 1100
    },
    {
      "epoch": 0.86,
      "grad_norm": 2.453305244445801,
      "learning_rate": 8.88989898989899e-07,
      "loss": 0.5914,
      "step": 1200
    },
    {
      "epoch": 0.93,
      "grad_norm": 2.768860340118408,
      "learning_rate": 8.788888888888889e-07,
      "loss": 0.6194,
      "step": 1300
    },
    {
      "epoch": 1.0,
      "grad_norm": 4.065279483795166,
      "learning_rate": 8.687878787878787e-07,
      "loss": 0.6038,
      "step": 1400
    },
    {
      "epoch": 1.08,
      "grad_norm": 3.8275232315063477,
      "learning_rate": 8.586868686868687e-07,
      "loss": 0.6038,
      "step": 1500
    },
    {
      "epoch": 1.15,
      "grad_norm": 2.6090586185455322,
      "learning_rate": 8.485858585858586e-07,
      "loss": 0.592,
      "step": 1600
    },
    {
      "epoch": 1.22,
      "grad_norm": 1.9870526790618896,
      "learning_rate": 8.384848484848484e-07,
      "loss": 0.5885,
      "step": 1700
    },
    {
      "epoch": 1.29,
      "grad_norm": 2.2942471504211426,
      "learning_rate": 8.283838383838383e-07,
      "loss": 0.622,
      "step": 1800
    },
    {
      "epoch": 1.36,
      "grad_norm": 2.0904691219329834,
      "learning_rate": 8.182828282828283e-07,
      "loss": 0.5857,
      "step": 1900
    },
    {
      "epoch": 1.43,
      "grad_norm": 2.909604072570801,
      "learning_rate": 8.081818181818182e-07,
      "loss": 0.5841,
      "step": 2000
    },
    {
      "epoch": 1.43,
      "eval_loss": 0.7236512303352356,
      "eval_runtime": 192.8239,
      "eval_samples_per_second": 5.186,
      "eval_steps_per_second": 0.648,
      "step": 2000
    },
    {
      "epoch": 1.51,
      "grad_norm": 2.645550012588501,
      "learning_rate": 7.98080808080808e-07,
      "loss": 0.6013,
      "step": 2100
    },
    {
      "epoch": 1.58,
      "grad_norm": 2.5540435314178467,
      "learning_rate": 7.88080808080808e-07,
      "loss": 0.6049,
      "step": 2200
    },
    {
      "epoch": 1.65,
      "grad_norm": 2.3379604816436768,
      "learning_rate": 7.779797979797979e-07,
      "loss": 0.5794,
      "step": 2300
    },
    {
      "epoch": 1.72,
      "grad_norm": 2.4884698390960693,
      "learning_rate": 7.678787878787878e-07,
      "loss": 0.5832,
      "step": 2400
    },
    {
      "epoch": 1.79,
      "grad_norm": 2.4760003089904785,
      "learning_rate": 7.577777777777778e-07,
      "loss": 0.5852,
      "step": 2500
    },
    {
      "epoch": 1.86,
      "grad_norm": 2.334510087966919,
      "learning_rate": 7.476767676767676e-07,
      "loss": 0.5825,
      "step": 2600
    },
    {
      "epoch": 1.94,
      "grad_norm": 3.3062024116516113,
      "learning_rate": 7.375757575757575e-07,
      "loss": 0.5739,
      "step": 2700
    },
    {
      "epoch": 2.01,
      "grad_norm": 3.0897557735443115,
      "learning_rate": 7.274747474747475e-07,
      "loss": 0.5823,
      "step": 2800
    },
    {
      "epoch": 2.08,
      "grad_norm": 3.0633504390716553,
      "learning_rate": 7.173737373737373e-07,
      "loss": 0.5625,
      "step": 2900
    },
    {
      "epoch": 2.15,
      "grad_norm": 3.3426527976989746,
      "learning_rate": 7.072727272727272e-07,
      "loss": 0.5484,
      "step": 3000
    },
    {
      "epoch": 2.22,
      "grad_norm": 2.7486729621887207,
      "learning_rate": 6.971717171717172e-07,
      "loss": 0.5853,
      "step": 3100
    },
    {
      "epoch": 2.29,
      "grad_norm": 2.5889084339141846,
      "learning_rate": 6.87070707070707e-07,
      "loss": 0.568,
      "step": 3200
    },
    {
      "epoch": 2.37,
      "grad_norm": 2.474393606185913,
      "learning_rate": 6.76969696969697e-07,
      "loss": 0.5699,
      "step": 3300
    },
    {
      "epoch": 2.44,
      "grad_norm": 2.99786114692688,
      "learning_rate": 6.668686868686869e-07,
      "loss": 0.5723,
      "step": 3400
    },
    {
      "epoch": 2.51,
      "grad_norm": 2.888652801513672,
      "learning_rate": 6.567676767676767e-07,
      "loss": 0.5942,
      "step": 3500
    },
    {
      "epoch": 2.58,
      "grad_norm": 3.20930552482605,
      "learning_rate": 6.466666666666666e-07,
      "loss": 0.5827,
      "step": 3600
    },
    {
      "epoch": 2.65,
      "grad_norm": 2.312199592590332,
      "learning_rate": 6.365656565656565e-07,
      "loss": 0.5644,
      "step": 3700
    },
    {
      "epoch": 2.72,
      "grad_norm": 3.1894469261169434,
      "learning_rate": 6.264646464646465e-07,
      "loss": 0.5409,
      "step": 3800
    },
    {
      "epoch": 2.8,
      "grad_norm": 2.740865468978882,
      "learning_rate": 6.163636363636363e-07,
      "loss": 0.5518,
      "step": 3900
    },
    {
      "epoch": 2.87,
      "grad_norm": 4.003296375274658,
      "learning_rate": 6.062626262626262e-07,
      "loss": 0.5796,
      "step": 4000
    },
    {
      "epoch": 2.87,
      "eval_loss": 0.7056564688682556,
      "eval_runtime": 192.8797,
      "eval_samples_per_second": 5.185,
      "eval_steps_per_second": 0.648,
      "step": 4000
    }
  ],
  "logging_steps": 100,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 2000,
  "total_flos": 7.539739430769132e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}