{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.945914738349268,
  "eval_steps": 4096,
  "global_step": 20480,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011823934229365849,
      "grad_norm": 1.1222522258758545,
      "learning_rate": 0.000498046875,
      "loss": 0.7148012518882751,
      "step": 256
    },
    {
      "epoch": 0.023647868458731697,
      "grad_norm": 1.1869175434112549,
      "learning_rate": 0.000998046875,
      "loss": 0.7245784401893616,
      "step": 512
    },
    {
      "epoch": 0.03547180268809755,
      "grad_norm": 1.4848835468292236,
      "learning_rate": 0.000999640996023194,
      "loss": 0.7249147891998291,
      "step": 768
    },
    {
      "epoch": 0.047295736917463395,
      "grad_norm": 1.7819839715957642,
      "learning_rate": 0.0009985588674043958,
      "loss": 0.7251258492469788,
      "step": 1024
    },
    {
      "epoch": 0.05911967114682925,
      "grad_norm": 0.966533899307251,
      "learning_rate": 0.0009967551747861387,
      "loss": 0.7263231873512268,
      "step": 1280
    },
    {
      "epoch": 0.0709436053761951,
      "grad_norm": 1.143156886100769,
      "learning_rate": 0.000994232528651847,
      "loss": 0.7227165699005127,
      "step": 1536
    },
    {
      "epoch": 0.08276753960556095,
      "grad_norm": 0.969417929649353,
      "learning_rate": 0.0009909945800260092,
      "loss": 0.7240616083145142,
      "step": 1792
    },
    {
      "epoch": 0.09459147383492679,
      "grad_norm": 1.5566692352294922,
      "learning_rate": 0.0009870460151900522,
      "loss": 0.714276134967804,
      "step": 2048
    },
    {
      "epoch": 0.10641540806429264,
      "grad_norm": 0.6052780151367188,
      "learning_rate": 0.0009823925488998885,
      "loss": 0.712334394454956,
      "step": 2304
    },
    {
      "epoch": 0.1182393422936585,
      "grad_norm": 1.8517088890075684,
      "learning_rate": 0.0009770409161149525,
      "loss": 0.7229606509208679,
      "step": 2560
    },
    {
      "epoch": 0.13006327652302435,
      "grad_norm": 0.7567554712295532,
      "learning_rate": 0.0009709988622506973,
      "loss": 0.7149595618247986,
      "step": 2816
    },
    {
      "epoch": 0.1418872107523902,
      "grad_norm": 1.334909439086914,
      "learning_rate": 0.000964275131968659,
      "loss": 0.7151144742965698,
      "step": 3072
    },
    {
      "epoch": 0.15371114498175603,
      "grad_norm": 1.4210789203643799,
      "learning_rate": 0.0009568794565203123,
      "loss": 0.7077174186706543,
      "step": 3328
    },
    {
      "epoch": 0.1655350792111219,
      "grad_norm": 1.216098427772522,
      "learning_rate": 0.0009488225396630347,
      "loss": 0.7150501012802124,
      "step": 3584
    },
    {
      "epoch": 0.17735901344048774,
      "grad_norm": 2.2890193462371826,
      "learning_rate": 0.0009401160421685646,
      "loss": 0.7134696841239929,
      "step": 3840
    },
    {
      "epoch": 0.18918294766985358,
      "grad_norm": 0.5908980369567871,
      "learning_rate": 0.0009307725649463714,
      "loss": 0.7133264541625977,
      "step": 4096
    },
    {
      "epoch": 0.18918294766985358,
      "eval_bleu": 0.0009799282077471887,
      "eval_loss": 0.7145318688322965,
      "eval_mse_loss": 0.7145318688322965,
      "step": 4096
    },
    {
      "epoch": 0.18918294766985358,
      "eval_bleu": 0.0009799282077471887,
      "eval_loss": 0.7145318688322965,
      "eval_mse_loss": 0.7145318688322965,
      "eval_runtime": 230.3288,
      "eval_samples_per_second": 121.535,
      "eval_steps_per_second": 1.902,
      "step": 4096
    },
    {
      "epoch": 0.20100688189921945,
      "grad_norm": 1.3493183851242065,
      "learning_rate": 0.0009208056308063659,
      "loss": 0.7063367366790771,
      "step": 4352
    },
    {
      "epoch": 0.2128308161285853,
      "grad_norm": 1.127218246459961,
      "learning_rate": 0.0009102296648873445,
      "loss": 0.7145900130271912,
      "step": 4608
    },
    {
      "epoch": 0.22465475035795113,
      "grad_norm": 1.356933832168579,
      "learning_rate": 0.0008990599737794927,
      "loss": 0.7130961418151855,
      "step": 4864
    },
    {
      "epoch": 0.236478684587317,
      "grad_norm": 0.5257934331893921,
      "learning_rate": 0.0008873127233711644,
      "loss": 0.713787853717804,
      "step": 5120
    },
    {
      "epoch": 0.24830261881668284,
      "grad_norm": 0.9926263093948364,
      "learning_rate": 0.0008750049154520011,
      "loss": 0.7070172429084778,
      "step": 5376
    },
    {
      "epoch": 0.2601265530460487,
      "grad_norm": 1.379420518875122,
      "learning_rate": 0.0008621543631062487,
      "loss": 0.7044810056686401,
      "step": 5632
    },
    {
      "epoch": 0.27195048727541454,
      "grad_norm": 1.3494232892990112,
      "learning_rate": 0.0008487796649318904,
      "loss": 0.7017611265182495,
      "step": 5888
    },
    {
      "epoch": 0.2837744215047804,
      "grad_norm": 1.3593381643295288,
      "learning_rate": 0.0008349001781229053,
      "loss": 0.7081840634346008,
      "step": 6144
    },
    {
      "epoch": 0.2955983557341462,
      "grad_norm": 1.4364413022994995,
      "learning_rate": 0.0008205359904536107,
      "loss": 0.6988816857337952,
      "step": 6400
    },
    {
      "epoch": 0.30742228996351206,
      "grad_norm": 1.7631044387817383,
      "learning_rate": 0.0008057078912056363,
      "loss": 0.7046327590942383,
      "step": 6656
    },
    {
      "epoch": 0.3192462241928779,
      "grad_norm": 1.3340634107589722,
      "learning_rate": 0.0007904373410796086,
      "loss": 0.6967505216598511,
      "step": 6912
    },
    {
      "epoch": 0.3310701584222438,
      "grad_norm": 1.0474779605865479,
      "learning_rate": 0.0007747464411350876,
      "loss": 0.6985277533531189,
      "step": 7168
    },
    {
      "epoch": 0.34289409265160964,
      "grad_norm": 1.2403302192687988,
      "learning_rate": 0.000758657900803716,
      "loss": 0.6996051073074341,
      "step": 7424
    },
    {
      "epoch": 0.3547180268809755,
      "grad_norm": 0.8496942520141602,
      "learning_rate": 0.000742195005021869,
      "loss": 0.6959736347198486,
      "step": 7680
    },
    {
      "epoch": 0.3665419611103413,
      "grad_norm": 0.9611047506332397,
      "learning_rate": 0.0007253815805303786,
      "loss": 0.6994542479515076,
      "step": 7936
    },
    {
      "epoch": 0.37836589533970716,
      "grad_norm": 0.5330133438110352,
      "learning_rate": 0.0007082419613901028,
      "loss": 0.7035479545593262,
      "step": 8192
    },
    {
      "epoch": 0.37836589533970716,
      "eval_bleu": 0.0010052073081668492,
      "eval_loss": 0.6948950419959412,
      "eval_mse_loss": 0.6948950419959412,
      "step": 8192
    },
    {
      "epoch": 0.37836589533970716,
      "eval_bleu": 0.0010052073081668492,
      "eval_loss": 0.6948950419959412,
      "eval_mse_loss": 0.6948950419959412,
      "eval_runtime": 230.6732,
      "eval_samples_per_second": 121.354,
      "eval_steps_per_second": 1.899,
      "step": 8192
    },
    {
      "epoch": 0.390189829569073,
      "grad_norm": 0.8082056045532227,
      "learning_rate": 0.0006908009537632514,
      "loss": 0.7005434036254883,
      "step": 8448
    },
    {
      "epoch": 0.4020137637984389,
      "grad_norm": 1.2990059852600098,
      "learning_rate": 0.0006730838000114403,
      "loss": 0.7007656097412109,
      "step": 8704
    },
    {
      "epoch": 0.41383769802780473,
      "grad_norm": 0.5188842415809631,
      "learning_rate": 0.0006551161421624341,
      "loss": 0.7021461129188538,
      "step": 8960
    },
    {
      "epoch": 0.4256616322571706,
      "grad_norm": 0.6316655874252319,
      "learning_rate": 0.0006369239847984517,
      "loss": 0.6990594863891602,
      "step": 9216
    },
    {
      "epoch": 0.4374855664865364,
      "grad_norm": 1.5941392183303833,
      "learning_rate": 0.0006185336574197479,
      "loss": 0.6935109496116638,
      "step": 9472
    },
    {
      "epoch": 0.44930950071590225,
      "grad_norm": 0.41938135027885437,
      "learning_rate": 0.0005999717763379407,
      "loss": 0.6998211145401001,
      "step": 9728
    },
    {
      "epoch": 0.4611334349452681,
      "grad_norm": 0.9023215770721436,
      "learning_rate": 0.0005812652061542363,
      "loss": 0.6968086361885071,
      "step": 9984
    },
    {
      "epoch": 0.472957369174634,
      "grad_norm": 1.2630209922790527,
      "learning_rate": 0.0005624410208783071,
      "loss": 0.6952697038650513,
      "step": 10240
    },
    {
      "epoch": 0.48478130340399983,
      "grad_norm": 1.2264976501464844,
      "learning_rate": 0.0005435264647440881,
      "loss": 0.6887762546539307,
      "step": 10496
    },
    {
      "epoch": 0.49660523763336567,
      "grad_norm": 0.728610098361969,
      "learning_rate": 0.000524548912779213,
      "loss": 0.6938179135322571,
      "step": 10752
    },
    {
      "epoch": 0.5084291718627315,
      "grad_norm": 0.8676101565361023,
      "learning_rate": 0.0005055358311851499,
      "loss": 0.6948530673980713,
      "step": 11008
    },
    {
      "epoch": 0.5202531060920974,
      "grad_norm": 1.1160482168197632,
      "learning_rate": 0.0004865147375853812,
      "loss": 0.691513180732727,
      "step": 11264
    },
    {
      "epoch": 0.5320770403214632,
      "grad_norm": 0.6974979043006897,
      "learning_rate": 0.0004675131611991607,
      "loss": 0.6864028573036194,
      "step": 11520
    },
    {
      "epoch": 0.5439009745508291,
      "grad_norm": 0.46719640493392944,
      "learning_rate": 0.0004485586029984899,
      "loss": 0.6954278349876404,
      "step": 11776
    },
    {
      "epoch": 0.5557249087801949,
      "grad_norm": 1.2132223844528198,
      "learning_rate": 0.00042967849590597266,
      "loss": 0.6914871335029602,
      "step": 12032
    },
    {
      "epoch": 0.5675488430095608,
      "grad_norm": 1.054893970489502,
      "learning_rate": 0.0004109001650911621,
      "loss": 0.686524510383606,
      "step": 12288
    },
    {
      "epoch": 0.5675488430095608,
      "eval_bleu": 0.0010400607906154184,
      "eval_loss": 0.6875994070752026,
      "eval_mse_loss": 0.6875994070752026,
      "step": 12288
    },
    {
      "epoch": 0.5675488430095608,
      "eval_bleu": 0.0010400607906154184,
      "eval_loss": 0.6875994070752026,
      "eval_mse_loss": 0.6875994070752026,
      "eval_runtime": 236.4224,
      "eval_samples_per_second": 118.402,
      "eval_steps_per_second": 1.853,
      "step": 12288
    },
    {
      "epoch": 0.5793727772389267,
      "grad_norm": 0.7405524253845215,
      "learning_rate": 0.0003922507884228551,
      "loss": 0.6839651465415955,
      "step": 12544
    },
    {
      "epoch": 0.5911967114682924,
      "grad_norm": 0.7894928455352783,
      "learning_rate": 0.00037375735713457723,
      "loss": 0.6843001842498779,
      "step": 12800
    },
    {
      "epoch": 0.6030206456976583,
      "grad_norm": 0.7159557938575745,
      "learning_rate": 0.00035544663676018276,
      "loss": 0.6922345161437988,
      "step": 13056
    },
    {
      "epoch": 0.6148445799270241,
      "grad_norm": 1.1562808752059937,
      "learning_rate": 0.00033734512839611255,
      "loss": 0.6860140562057495,
      "step": 13312
    },
    {
      "epoch": 0.62666851415639,
      "grad_norm": 0.5824499130249023,
      "learning_rate": 0.0003194790303463687,
      "loss": 0.6856705546379089,
      "step": 13568
    },
    {
      "epoch": 0.6384924483857558,
      "grad_norm": 3.080799102783203,
      "learning_rate": 0.00030187420020572406,
      "loss": 0.6847730278968811,
      "step": 13824
    },
    {
      "epoch": 0.6503163826151217,
      "grad_norm": 0.6743490695953369,
      "learning_rate": 0.00028455611743603626,
      "loss": 0.6856830716133118,
      "step": 14080
    },
    {
      "epoch": 0.6621403168444876,
      "grad_norm": 1.060508370399475,
      "learning_rate": 0.0002675498464898373,
      "loss": 0.6844523549079895,
      "step": 14336
    },
    {
      "epoch": 0.6739642510738534,
      "grad_norm": 1.611391544342041,
      "learning_rate": 0.0002508800005345623,
      "loss": 0.6842568516731262,
      "step": 14592
    },
    {
      "epoch": 0.6857881853032193,
      "grad_norm": 1.5996400117874146,
      "learning_rate": 0.00023457070582992562,
      "loss": 0.6826963424682617,
      "step": 14848
    },
    {
      "epoch": 0.6976121195325851,
      "grad_norm": 0.6100959181785583,
      "learning_rate": 0.00021864556680999692,
      "loss": 0.6834294199943542,
      "step": 15104
    },
    {
      "epoch": 0.709436053761951,
      "grad_norm": 1.0907484292984009,
      "learning_rate": 0.0002031276319205152,
      "loss": 0.6882811784744263,
      "step": 15360
    },
    {
      "epoch": 0.7212599879913169,
      "grad_norm": 1.3213837146759033,
      "learning_rate": 0.00018803936026088542,
      "loss": 0.6864408850669861,
      "step": 15616
    },
    {
      "epoch": 0.7330839222206826,
      "grad_norm": 0.40607210993766785,
      "learning_rate": 0.00017340258907913464,
      "loss": 0.6805769205093384,
      "step": 15872
    },
    {
      "epoch": 0.7449078564500485,
      "grad_norm": 0.48498424887657166,
      "learning_rate": 0.0001592385021668743,
      "loss": 0.6834226846694946,
      "step": 16128
    },
    {
      "epoch": 0.7567317906794143,
      "grad_norm": 0.4616319239139557,
      "learning_rate": 0.0001455675992000087,
      "loss": 0.6785019040107727,
      "step": 16384
    },
    {
      "epoch": 0.7567317906794143,
      "eval_bleu": 0.0009476330082336835,
      "eval_loss": 0.6870187982984873,
      "eval_mse_loss": 0.6870187982984873,
      "step": 16384
    },
    {
      "epoch": 0.7567317906794143,
      "eval_bleu": 0.0009476330082336835,
      "eval_loss": 0.6870187982984873,
      "eval_mse_loss": 0.6870187982984873,
      "eval_runtime": 231.2977,
      "eval_samples_per_second": 121.026,
      "eval_steps_per_second": 1.894,
      "step": 16384
    },
    {
      "epoch": 0.7685557249087802,
      "grad_norm": 1.320206642150879,
      "learning_rate": 0.000132409666069565,
      "loss": 0.6836502552032471,
      "step": 16640
    },
    {
      "epoch": 0.780379659138146,
      "grad_norm": 0.3926827311515808,
      "learning_rate": 0.0001197837462455823,
      "loss": 0.6756935119628906,
      "step": 16896
    },
    {
      "epoch": 0.7922035933675119,
      "grad_norm": 0.4777701199054718,
      "learning_rate": 0.00010770811321550749,
      "loss": 0.6764777898788452,
      "step": 17152
    },
    {
      "epoch": 0.8040275275968778,
      "grad_norm": 1.5470472574234009,
      "learning_rate": 9.620024403698591e-05,
      "loss": 0.6807876229286194,
      "step": 17408
    },
    {
      "epoch": 0.8158514618262436,
      "grad_norm": 1.4899367094039917,
      "learning_rate": 8.527679404332429e-05,
      "loss": 0.6819196343421936,
      "step": 17664
    },
    {
      "epoch": 0.8276753960556095,
      "grad_norm": 1.1178877353668213,
      "learning_rate": 7.495357273823544e-05,
      "loss": 0.6775845289230347,
      "step": 17920
    },
    {
      "epoch": 0.8394993302849753,
      "grad_norm": 1.9614344835281372,
      "learning_rate": 6.524552091475183e-05,
      "loss": 0.6840441823005676,
      "step": 18176
    },
    {
      "epoch": 0.8513232645143411,
      "grad_norm": 1.231566309928894,
      "learning_rate": 5.6166689031422024e-05,
      "loss": 0.6766948103904724,
      "step": 18432
    },
    {
      "epoch": 0.8631471987437069,
      "grad_norm": 3.0353503227233887,
      "learning_rate": 4.773021687709067e-05,
      "loss": 0.677415132522583,
      "step": 18688
    },
    {
      "epoch": 0.8749711329730728,
      "grad_norm": 2.707676887512207,
      "learning_rate": 3.994831455368719e-05,
      "loss": 0.6782501339912415,
      "step": 18944
    },
    {
      "epoch": 0.8867950672024387,
      "grad_norm": 0.49309444427490234,
      "learning_rate": 3.283224480455282e-05,
      "loss": 0.680189847946167,
      "step": 19200
    },
    {
      "epoch": 0.8986190014318045,
      "grad_norm": 1.044478178024292,
      "learning_rate": 2.639230671387627e-05,
      "loss": 0.6733763217926025,
      "step": 19456
    },
    {
      "epoch": 0.9104429356611704,
      "grad_norm": 2.4254560470581055,
      "learning_rate": 2.063782080083576e-05,
      "loss": 0.6758845448493958,
      "step": 19712
    },
    {
      "epoch": 0.9222668698905362,
      "grad_norm": 0.8764089345932007,
      "learning_rate": 1.557711553001523e-05,
      "loss": 0.6751888990402222,
      "step": 19968
    },
    {
      "epoch": 0.9340908041199021,
      "grad_norm": 0.6017519235610962,
      "learning_rate": 1.1217515257622269e-05,
      "loss": 0.6804232597351074,
      "step": 20224
    },
    {
      "epoch": 0.945914738349268,
      "grad_norm": 2.2349886894226074,
      "learning_rate": 7.565329630950746e-06,
      "loss": 0.6779721975326538,
      "step": 20480
    },
    {
      "epoch": 0.945914738349268,
      "eval_bleu": 0.0009655630043002199,
      "eval_loss": 0.675000424866807,
      "eval_mse_loss": 0.675000424866807,
      "step": 20480
    },
    {
      "epoch": 0.945914738349268,
      "eval_bleu": 0.0009655630043002199,
      "eval_loss": 0.675000424866807,
      "eval_mse_loss": 0.675000424866807,
      "eval_runtime": 236.8078,
      "eval_samples_per_second": 118.21,
      "eval_steps_per_second": 1.85,
      "step": 20480
    }
  ],
  "logging_steps": 256,
  "max_steps": 21651,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 4096,
  "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
}