{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5675488430095608,
  "eval_steps": 1024,
  "global_step": 12288,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011823934229365849,
      "grad_norm": 0.0007859937031753361,
      "learning_rate": 0.000498046875,
      "loss": 0.6385964751243591,
      "step": 256
    },
    {
      "epoch": 0.023647868458731697,
      "grad_norm": 0.004059702157974243,
      "learning_rate": 0.000998046875,
      "loss": 0.6255754828453064,
      "step": 512
    },
    {
      "epoch": 0.03547180268809755,
      "grad_norm": 0.0074360910803079605,
      "learning_rate": 0.000999640996023194,
      "loss": 0.5881161093711853,
      "step": 768
    },
    {
      "epoch": 0.047295736917463395,
      "grad_norm": 0.014354221522808075,
      "learning_rate": 0.0009985588674043958,
      "loss": 0.5408114194869995,
      "step": 1024
    },
    {
      "epoch": 0.047295736917463395,
      "eval_bleu": 0.01917810496562663,
      "eval_ce_loss": 0.49575697606829205,
      "eval_loss": 0.49575697606829205,
      "step": 1024
    },
    {
      "epoch": 0.047295736917463395,
      "eval_bleu": 0.01917810496562663,
      "eval_ce_loss": 0.49575697606829205,
      "eval_loss": 0.49575697606829205,
      "eval_runtime": 123.2741,
      "eval_samples_per_second": 227.079,
      "eval_steps_per_second": 3.553,
      "step": 1024
    },
    {
      "epoch": 0.05911967114682925,
      "grad_norm": 0.01720331236720085,
      "learning_rate": 0.0009967551747861387,
      "loss": 0.41951486468315125,
      "step": 1280
    },
    {
      "epoch": 0.0709436053761951,
      "grad_norm": 0.019904080778360367,
      "learning_rate": 0.000994232528651847,
      "loss": 0.32046523690223694,
      "step": 1536
    },
    {
      "epoch": 0.08276753960556095,
      "grad_norm": 0.023004669696092606,
      "learning_rate": 0.0009909945800260092,
      "loss": 0.2763057351112366,
      "step": 1792
    },
    {
      "epoch": 0.09459147383492679,
      "grad_norm": 0.025376707315444946,
      "learning_rate": 0.0009870460151900522,
      "loss": 0.25197452306747437,
      "step": 2048
    },
    {
      "epoch": 0.09459147383492679,
      "eval_bleu": 0.15934969580648434,
      "eval_ce_loss": 0.24096488823357237,
      "eval_loss": 0.24096488823357237,
      "step": 2048
    },
    {
      "epoch": 0.09459147383492679,
      "eval_bleu": 0.15934969580648434,
      "eval_ce_loss": 0.24096488823357237,
      "eval_loss": 0.24096488823357237,
      "eval_runtime": 123.5565,
      "eval_samples_per_second": 226.56,
      "eval_steps_per_second": 3.545,
      "step": 2048
    },
    {
      "epoch": 0.10641540806429264,
      "grad_norm": 0.02607455477118492,
      "learning_rate": 0.0009823925488998885,
      "loss": 0.23232555389404297,
      "step": 2304
    },
    {
      "epoch": 0.1182393422936585,
      "grad_norm": 0.023672524839639664,
      "learning_rate": 0.0009770409161149525,
      "loss": 0.2173057347536087,
      "step": 2560
    },
    {
      "epoch": 0.13006327652302435,
      "grad_norm": 0.03465030714869499,
      "learning_rate": 0.0009709988622506973,
      "loss": 0.20285777747631073,
      "step": 2816
    },
    {
      "epoch": 0.1418872107523902,
      "grad_norm": 0.03454975038766861,
      "learning_rate": 0.000964275131968659,
      "loss": 0.19273696839809418,
      "step": 3072
    },
    {
      "epoch": 0.1418872107523902,
      "eval_bleu": 0.18685399237963718,
      "eval_ce_loss": 0.18767488618555678,
      "eval_loss": 0.18767488618555678,
      "step": 3072
    },
    {
      "epoch": 0.1418872107523902,
      "eval_bleu": 0.18685399237963718,
      "eval_ce_loss": 0.18767488618555678,
      "eval_loss": 0.18767488618555678,
      "eval_runtime": 124.1463,
      "eval_samples_per_second": 225.484,
      "eval_steps_per_second": 3.528,
      "step": 3072
    },
    {
      "epoch": 0.15371114498175603,
      "grad_norm": 0.028874596580863,
      "learning_rate": 0.0009568794565203123,
      "loss": 0.18194623291492462,
      "step": 3328
    },
    {
      "epoch": 0.1655350792111219,
      "grad_norm": 0.026963606476783752,
      "learning_rate": 0.0009488225396630347,
      "loss": 0.1741044968366623,
      "step": 3584
    },
    {
      "epoch": 0.17735901344048774,
      "grad_norm": 0.03203282505273819,
      "learning_rate": 0.0009401160421685646,
      "loss": 0.16683174669742584,
      "step": 3840
    },
    {
      "epoch": 0.18918294766985358,
      "grad_norm": 0.02940264903008938,
      "learning_rate": 0.0009307725649463714,
      "loss": 0.1600237637758255,
      "step": 4096
    },
    {
      "epoch": 0.18918294766985358,
      "eval_bleu": 0.2041856952873211,
      "eval_ce_loss": 0.15758018299392915,
      "eval_loss": 0.15758018299392915,
      "step": 4096
    },
    {
      "epoch": 0.18918294766985358,
      "eval_bleu": 0.2041856952873211,
      "eval_ce_loss": 0.15758018299392915,
      "eval_loss": 0.15758018299392915,
      "eval_runtime": 122.9443,
      "eval_samples_per_second": 227.688,
      "eval_steps_per_second": 3.563,
      "step": 4096
    },
    {
      "epoch": 0.20100688189921945,
      "grad_norm": 0.03609512746334076,
      "learning_rate": 0.0009208056308063659,
      "loss": 0.15560653805732727,
      "step": 4352
    },
    {
      "epoch": 0.2128308161285853,
      "grad_norm": 0.03296591714024544,
      "learning_rate": 0.0009102296648873445,
      "loss": 0.14899571239948273,
      "step": 4608
    },
    {
      "epoch": 0.22465475035795113,
      "grad_norm": 0.034538134932518005,
      "learning_rate": 0.0008990599737794927,
      "loss": 0.14445939660072327,
      "step": 4864
    },
    {
      "epoch": 0.236478684587317,
      "grad_norm": 0.032769329845905304,
      "learning_rate": 0.0008873127233711644,
      "loss": 0.13995994627475739,
      "step": 5120
    },
    {
      "epoch": 0.236478684587317,
      "eval_bleu": 0.21739622472708592,
      "eval_ce_loss": 0.13755980873965237,
      "eval_loss": 0.13755980873965237,
      "step": 5120
    },
    {
      "epoch": 0.236478684587317,
      "eval_bleu": 0.21739622472708592,
      "eval_ce_loss": 0.13755980873965237,
      "eval_loss": 0.13755980873965237,
      "eval_runtime": 123.4264,
      "eval_samples_per_second": 226.799,
      "eval_steps_per_second": 3.549,
      "step": 5120
    },
    {
      "epoch": 0.24830261881668284,
      "grad_norm": 0.03575948625802994,
      "learning_rate": 0.0008750049154520011,
      "loss": 0.13656434416770935,
      "step": 5376
    },
    {
      "epoch": 0.2601265530460487,
      "grad_norm": 0.032076139003038406,
      "learning_rate": 0.0008621543631062487,
      "loss": 0.13251279294490814,
      "step": 5632
    },
    {
      "epoch": 0.27195048727541454,
      "grad_norm": 0.03560580685734749,
      "learning_rate": 0.0008487796649318904,
      "loss": 0.12999674677848816,
      "step": 5888
    },
    {
      "epoch": 0.2837744215047804,
      "grad_norm": 0.03557967767119408,
      "learning_rate": 0.0008349001781229053,
      "loss": 0.12719185650348663,
      "step": 6144
    },
    {
      "epoch": 0.2837744215047804,
      "eval_bleu": 0.22349092475951987,
      "eval_ce_loss": 0.1254711460032964,
      "eval_loss": 0.1254711460032964,
      "step": 6144
    },
    {
      "epoch": 0.2837744215047804,
      "eval_bleu": 0.22349092475951987,
      "eval_ce_loss": 0.1254711460032964,
      "eval_loss": 0.1254711460032964,
      "eval_runtime": 124.846,
      "eval_samples_per_second": 224.22,
      "eval_steps_per_second": 3.508,
      "step": 6144
    },
    {
      "epoch": 0.2955983557341462,
      "grad_norm": 0.03403434529900551,
      "learning_rate": 0.0008205359904536107,
      "loss": 0.12399476021528244,
      "step": 6400
    },
    {
      "epoch": 0.30742228996351206,
      "grad_norm": 0.03340385481715202,
      "learning_rate": 0.0008057078912056363,
      "loss": 0.12178181111812592,
      "step": 6656
    },
    {
      "epoch": 0.3192462241928779,
      "grad_norm": 0.033506471663713455,
      "learning_rate": 0.0007904373410796086,
      "loss": 0.11948449909687042,
      "step": 6912
    },
    {
      "epoch": 0.3310701584222438,
      "grad_norm": 0.0328289233148098,
      "learning_rate": 0.0007747464411350876,
      "loss": 0.11748082935810089,
      "step": 7168
    },
    {
      "epoch": 0.3310701584222438,
      "eval_bleu": 0.22974485130663763,
      "eval_ce_loss": 0.11694975512008689,
      "eval_loss": 0.11694975512008689,
      "step": 7168
    },
    {
      "epoch": 0.3310701584222438,
      "eval_bleu": 0.22974485130663763,
      "eval_ce_loss": 0.11694975512008689,
      "eval_loss": 0.11694975512008689,
      "eval_runtime": 126.6834,
      "eval_samples_per_second": 220.968,
      "eval_steps_per_second": 3.457,
      "step": 7168
    },
    {
      "epoch": 0.34289409265160964,
      "grad_norm": 0.04062662646174431,
      "learning_rate": 0.000758657900803716,
      "loss": 0.11517409980297089,
      "step": 7424
    },
    {
      "epoch": 0.3547180268809755,
      "grad_norm": 0.03233740106225014,
      "learning_rate": 0.000742195005021869,
      "loss": 0.11353136599063873,
      "step": 7680
    },
    {
      "epoch": 0.3665419611103413,
      "grad_norm": 0.04508234187960625,
      "learning_rate": 0.0007253815805303786,
      "loss": 0.11149852722883224,
      "step": 7936
    },
    {
      "epoch": 0.37836589533970716,
      "grad_norm": 0.0339302197098732,
      "learning_rate": 0.0007082419613901028,
      "loss": 0.10967232286930084,
      "step": 8192
    },
    {
      "epoch": 0.37836589533970716,
      "eval_bleu": 0.23402933156555222,
      "eval_ce_loss": 0.10964651824268576,
      "eval_loss": 0.10964651824268576,
      "step": 8192
    },
    {
      "epoch": 0.37836589533970716,
      "eval_bleu": 0.23402933156555222,
      "eval_ce_loss": 0.10964651824268576,
      "eval_loss": 0.10964651824268576,
      "eval_runtime": 126.1015,
      "eval_samples_per_second": 221.988,
      "eval_steps_per_second": 3.473,
      "step": 8192
    },
    {
      "epoch": 0.390189829569073,
      "grad_norm": 0.03507415950298309,
      "learning_rate": 0.0006908009537632514,
      "loss": 0.10870054364204407,
      "step": 8448
    },
    {
      "epoch": 0.4020137637984389,
      "grad_norm": 0.03351368010044098,
      "learning_rate": 0.0006730838000114403,
      "loss": 0.10755178332328796,
      "step": 8704
    },
    {
      "epoch": 0.41383769802780473,
      "grad_norm": 0.03829547390341759,
      "learning_rate": 0.0006551161421624341,
      "loss": 0.1055288016796112,
      "step": 8960
    },
    {
      "epoch": 0.4256616322571706,
      "grad_norm": 0.029935523867607117,
      "learning_rate": 0.0006369239847984517,
      "loss": 0.10448558628559113,
      "step": 9216
    },
    {
      "epoch": 0.4256616322571706,
      "eval_bleu": 0.23801999524762174,
      "eval_ce_loss": 0.10415493887445154,
      "eval_loss": 0.10415493887445154,
      "step": 9216
    },
    {
      "epoch": 0.4256616322571706,
      "eval_bleu": 0.23801999524762174,
      "eval_ce_loss": 0.10415493887445154,
      "eval_loss": 0.10415493887445154,
      "eval_runtime": 127.389,
      "eval_samples_per_second": 219.744,
      "eval_steps_per_second": 3.438,
      "step": 9216
    },
    {
      "epoch": 0.4374855664865364,
      "grad_norm": 0.029414288699626923,
      "learning_rate": 0.0006185336574197479,
      "loss": 0.10328588634729385,
      "step": 9472
    },
    {
      "epoch": 0.44930950071590225,
      "grad_norm": 0.03512301295995712,
      "learning_rate": 0.0005999717763379407,
      "loss": 0.10299594700336456,
      "step": 9728
    },
    {
      "epoch": 0.4611334349452681,
      "grad_norm": 0.04324057698249817,
      "learning_rate": 0.0005812652061542363,
      "loss": 0.10201191902160645,
      "step": 9984
    },
    {
      "epoch": 0.472957369174634,
      "grad_norm": 0.032921064645051956,
      "learning_rate": 0.0005624410208783071,
      "loss": 0.10033029317855835,
      "step": 10240
    },
    {
      "epoch": 0.472957369174634,
      "eval_bleu": 0.23928028701230103,
      "eval_ce_loss": 0.10052529907866156,
      "eval_loss": 0.10052529907866156,
      "step": 10240
    },
    {
      "epoch": 0.472957369174634,
      "eval_bleu": 0.23928028701230103,
      "eval_ce_loss": 0.10052529907866156,
      "eval_loss": 0.10052529907866156,
      "eval_runtime": 126.6952,
      "eval_samples_per_second": 220.948,
      "eval_steps_per_second": 3.457,
      "step": 10240
    },
    {
      "epoch": 0.48478130340399983,
      "grad_norm": 0.038171831518411636,
      "learning_rate": 0.0005435264647440881,
      "loss": 0.09996124356985092,
      "step": 10496
    },
    {
      "epoch": 0.49660523763336567,
      "grad_norm": 0.03536969795823097,
      "learning_rate": 0.000524548912779213,
      "loss": 0.09928364306688309,
      "step": 10752
    },
    {
      "epoch": 0.5084291718627315,
      "grad_norm": 0.03323720768094063,
      "learning_rate": 0.0005055358311851499,
      "loss": 0.09906301647424698,
      "step": 11008
    },
    {
      "epoch": 0.5202531060920974,
      "grad_norm": 0.033888787031173706,
      "learning_rate": 0.0004865147375853812,
      "loss": 0.09757085144519806,
      "step": 11264
    },
    {
      "epoch": 0.5202531060920974,
      "eval_bleu": 0.24364428948892838,
      "eval_ce_loss": 0.09666734056113517,
      "eval_loss": 0.09666734056113517,
      "step": 11264
    },
    {
      "epoch": 0.5202531060920974,
      "eval_bleu": 0.24364428948892838,
      "eval_ce_loss": 0.09666734056113517,
      "eval_loss": 0.09666734056113517,
      "eval_runtime": 124.874,
      "eval_samples_per_second": 224.17,
      "eval_steps_per_second": 3.508,
      "step": 11264
    },
    {
      "epoch": 0.5320770403214632,
      "grad_norm": 0.03228915110230446,
      "learning_rate": 0.0004675131611991607,
      "loss": 0.09719787538051605,
      "step": 11520
    },
    {
      "epoch": 0.5439009745508291,
      "grad_norm": 0.03569363057613373,
      "learning_rate": 0.0004485586029984899,
      "loss": 0.09681548178195953,
      "step": 11776
    },
    {
      "epoch": 0.5557249087801949,
      "grad_norm": 0.03645534813404083,
      "learning_rate": 0.00042967849590597266,
      "loss": 0.09616198390722275,
      "step": 12032
    },
    {
      "epoch": 0.5675488430095608,
      "grad_norm": 0.03306020796298981,
      "learning_rate": 0.0004109001650911621,
      "loss": 0.09529497474431992,
      "step": 12288
    },
    {
      "epoch": 0.5675488430095608,
      "eval_bleu": 0.24421693548361179,
      "eval_ce_loss": 0.09502403213553233,
      "eval_loss": 0.09502403213553233,
      "step": 12288
    },
    {
      "epoch": 0.5675488430095608,
      "eval_bleu": 0.24421693548361179,
      "eval_ce_loss": 0.09502403213553233,
      "eval_loss": 0.09502403213553233,
      "eval_runtime": 126.0135,
      "eval_samples_per_second": 222.143,
      "eval_steps_per_second": 3.476,
      "step": 12288
    }
  ],
  "logging_steps": 256,
  "max_steps": 21651,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1024,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}