{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 15950,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.31347962382445144,
      "grad_norm": 0.9908634424209595,
      "learning_rate": 1.9840000000000003e-05,
      "loss": 5.4652,
      "step": 500
    },
    {
      "epoch": 0.6269592476489029,
      "grad_norm": 0.6833625435829163,
      "learning_rate": 1.9357928802589e-05,
      "loss": 1.7034,
      "step": 1000
    },
    {
      "epoch": 0.9404388714733543,
      "grad_norm": 0.6710793972015381,
      "learning_rate": 1.8710679611650487e-05,
      "loss": 1.6778,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_bleu": 44.355750778466295,
      "eval_loss": 1.6447367668151855,
      "eval_runtime": 282.2475,
      "eval_samples_per_second": 9.977,
      "eval_steps_per_second": 0.624,
      "step": 1595
    },
    {
      "epoch": 1.2539184952978055,
      "grad_norm": 0.4725702404975891,
      "learning_rate": 1.8063430420711977e-05,
      "loss": 1.6499,
      "step": 2000
    },
    {
      "epoch": 1.567398119122257,
      "grad_norm": 0.6504935026168823,
      "learning_rate": 1.7416181229773465e-05,
      "loss": 1.6434,
      "step": 2500
    },
    {
      "epoch": 1.8808777429467085,
      "grad_norm": 0.4929894506931305,
      "learning_rate": 1.6768932038834952e-05,
      "loss": 1.6406,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_bleu": 46.92316806208224,
      "eval_loss": 1.6343189477920532,
      "eval_runtime": 280.4311,
      "eval_samples_per_second": 10.042,
      "eval_steps_per_second": 0.628,
      "step": 3190
    },
    {
      "epoch": 2.19435736677116,
      "grad_norm": 0.4408179521560669,
      "learning_rate": 1.612168284789644e-05,
      "loss": 1.6263,
      "step": 3500
    },
    {
      "epoch": 2.507836990595611,
      "grad_norm": 0.5890961289405823,
      "learning_rate": 1.547443365695793e-05,
      "loss": 1.6184,
      "step": 4000
    },
    {
      "epoch": 2.8213166144200628,
      "grad_norm": 0.36476999521255493,
      "learning_rate": 1.482718446601942e-05,
      "loss": 1.6186,
      "step": 4500
    },
    {
      "epoch": 3.0,
      "eval_bleu": 48.0812345935905,
      "eval_loss": 1.6287521123886108,
      "eval_runtime": 277.8914,
      "eval_samples_per_second": 10.133,
      "eval_steps_per_second": 0.633,
      "step": 4785
    },
    {
      "epoch": 3.134796238244514,
      "grad_norm": 0.5234176516532898,
      "learning_rate": 1.4179935275080907e-05,
      "loss": 1.6098,
      "step": 5000
    },
    {
      "epoch": 3.4482758620689653,
      "grad_norm": 0.5117827653884888,
      "learning_rate": 1.3532686084142396e-05,
      "loss": 1.6011,
      "step": 5500
    },
    {
      "epoch": 3.761755485893417,
      "grad_norm": 0.5140109062194824,
      "learning_rate": 1.2885436893203884e-05,
      "loss": 1.6021,
      "step": 6000
    },
    {
      "epoch": 4.0,
      "eval_bleu": 48.92348524517408,
      "eval_loss": 1.6266326904296875,
      "eval_runtime": 270.5269,
      "eval_samples_per_second": 10.409,
      "eval_steps_per_second": 0.651,
      "step": 6380
    },
    {
      "epoch": 4.075235109717869,
      "grad_norm": 0.5262463688850403,
      "learning_rate": 1.2238187702265374e-05,
      "loss": 1.5999,
      "step": 6500
    },
    {
      "epoch": 4.38871473354232,
      "grad_norm": 0.4214998185634613,
      "learning_rate": 1.1590938511326861e-05,
      "loss": 1.5893,
      "step": 7000
    },
    {
      "epoch": 4.702194357366771,
      "grad_norm": 0.4427284300327301,
      "learning_rate": 1.094368932038835e-05,
      "loss": 1.59,
      "step": 7500
    },
    {
      "epoch": 5.0,
      "eval_bleu": 48.82457853670125,
      "eval_loss": 1.626197338104248,
      "eval_runtime": 273.0297,
      "eval_samples_per_second": 10.314,
      "eval_steps_per_second": 0.645,
      "step": 7975
    },
    {
      "epoch": 5.015673981191223,
      "grad_norm": 0.4961058497428894,
      "learning_rate": 1.0296440129449838e-05,
      "loss": 1.5892,
      "step": 8000
    },
    {
      "epoch": 5.329153605015674,
      "grad_norm": 0.4459807276725769,
      "learning_rate": 9.649190938511328e-06,
      "loss": 1.5793,
      "step": 8500
    },
    {
      "epoch": 5.6426332288401255,
      "grad_norm": 0.42176157236099243,
      "learning_rate": 9.001941747572817e-06,
      "loss": 1.58,
      "step": 9000
    },
    {
      "epoch": 5.956112852664576,
      "grad_norm": 0.4856497049331665,
      "learning_rate": 8.354692556634304e-06,
      "loss": 1.5805,
      "step": 9500
    },
    {
      "epoch": 6.0,
      "eval_bleu": 49.44534738257411,
      "eval_loss": 1.6256004571914673,
      "eval_runtime": 271.5073,
      "eval_samples_per_second": 10.372,
      "eval_steps_per_second": 0.648,
      "step": 9570
    },
    {
      "epoch": 6.269592476489028,
      "grad_norm": 0.34010499715805054,
      "learning_rate": 7.707443365695793e-06,
      "loss": 1.5731,
      "step": 10000
    },
    {
      "epoch": 6.58307210031348,
      "grad_norm": 0.524381160736084,
      "learning_rate": 7.060194174757282e-06,
      "loss": 1.5727,
      "step": 10500
    },
    {
      "epoch": 6.896551724137931,
      "grad_norm": 0.3996853530406952,
      "learning_rate": 6.412944983818771e-06,
      "loss": 1.5725,
      "step": 11000
    },
    {
      "epoch": 7.0,
      "eval_bleu": 49.96969056769707,
      "eval_loss": 1.6257303953170776,
      "eval_runtime": 268.9556,
      "eval_samples_per_second": 10.47,
      "eval_steps_per_second": 0.654,
      "step": 11165
    },
    {
      "epoch": 7.210031347962382,
      "grad_norm": 0.5770251750946045,
      "learning_rate": 5.765695792880259e-06,
      "loss": 1.5686,
      "step": 11500
    },
    {
      "epoch": 7.523510971786834,
      "grad_norm": 0.40356528759002686,
      "learning_rate": 5.118446601941748e-06,
      "loss": 1.5666,
      "step": 12000
    },
    {
      "epoch": 7.836990595611285,
      "grad_norm": 0.4087662100791931,
      "learning_rate": 4.471197411003236e-06,
      "loss": 1.5663,
      "step": 12500
    },
    {
      "epoch": 8.0,
      "eval_bleu": 50.12951231046901,
      "eval_loss": 1.6264179944992065,
      "eval_runtime": 269.5975,
      "eval_samples_per_second": 10.445,
      "eval_steps_per_second": 0.653,
      "step": 12760
    },
    {
      "epoch": 8.150470219435737,
      "grad_norm": 0.493051677942276,
      "learning_rate": 3.823948220064725e-06,
      "loss": 1.5639,
      "step": 13000
    },
    {
      "epoch": 8.463949843260188,
      "grad_norm": 0.45878908038139343,
      "learning_rate": 3.176699029126214e-06,
      "loss": 1.5625,
      "step": 13500
    },
    {
      "epoch": 8.77742946708464,
      "grad_norm": 0.5698862671852112,
      "learning_rate": 2.5294498381877025e-06,
      "loss": 1.5623,
      "step": 14000
    },
    {
      "epoch": 9.0,
      "eval_bleu": 50.17886329004438,
      "eval_loss": 1.626705527305603,
      "eval_runtime": 269.908,
      "eval_samples_per_second": 10.433,
      "eval_steps_per_second": 0.652,
      "step": 14355
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.4237252175807953,
      "learning_rate": 1.882200647249191e-06,
      "loss": 1.561,
      "step": 14500
    },
    {
      "epoch": 9.404388714733543,
      "grad_norm": 0.4487234354019165,
      "learning_rate": 1.2349514563106797e-06,
      "loss": 1.5594,
      "step": 15000
    },
    {
      "epoch": 9.717868338557993,
      "grad_norm": 0.39791637659072876,
      "learning_rate": 5.877022653721683e-07,
      "loss": 1.5593,
      "step": 15500
    },
    {
      "epoch": 10.0,
      "eval_bleu": 50.10578525423357,
      "eval_loss": 1.626780390739441,
      "eval_runtime": 270.1794,
      "eval_samples_per_second": 10.423,
      "eval_steps_per_second": 0.651,
      "step": 15950
    }
  ],
  "logging_steps": 500,
  "max_steps": 15950,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 1.3821793152270336e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}