{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 250,
  "global_step": 1445,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011072664359861591,
      "grad_norm": 7.3027024269104,
      "learning_rate": 0.00036363636363636367,
      "loss": 7.4688,
      "step": 16
    },
    {
      "epoch": 0.022145328719723183,
      "grad_norm": 3.441852331161499,
      "learning_rate": 0.0006363636363636364,
      "loss": 6.1486,
      "step": 32
    },
    {
      "epoch": 0.03321799307958478,
      "grad_norm": 0.9797470569610596,
      "learning_rate": 0.001,
      "loss": 5.1898,
      "step": 48
    },
    {
      "epoch": 0.044290657439446365,
      "grad_norm": 0.4113083481788635,
      "learning_rate": 0.0009996782216198338,
      "loss": 5.0792,
      "step": 64
    },
    {
      "epoch": 0.05536332179930796,
      "grad_norm": 0.23998336493968964,
      "learning_rate": 0.0009987133006446386,
      "loss": 4.9765,
      "step": 80
    },
    {
      "epoch": 0.06643598615916955,
      "grad_norm": 0.34719282388687134,
      "learning_rate": 0.0009971064790372482,
      "loss": 5.0589,
      "step": 96
    },
    {
      "epoch": 0.07750865051903114,
      "grad_norm": 0.28703269362449646,
      "learning_rate": 0.0009948598249594788,
      "loss": 5.0273,
      "step": 112
    },
    {
      "epoch": 0.08858131487889273,
      "grad_norm": 0.3285124599933624,
      "learning_rate": 0.00099197623011017,
      "loss": 4.9965,
      "step": 128
    },
    {
      "epoch": 0.09965397923875433,
      "grad_norm": 0.21146929264068604,
      "learning_rate": 0.0009884594060032405,
      "loss": 5.0386,
      "step": 144
    },
    {
      "epoch": 0.11072664359861592,
      "grad_norm": 0.2574313282966614,
      "learning_rate": 0.0009843138791905482,
      "loss": 4.9744,
      "step": 160
    },
    {
      "epoch": 0.12179930795847752,
      "grad_norm": 0.2936093211174011,
      "learning_rate": 0.000979544985435704,
      "loss": 5.0448,
      "step": 176
    },
    {
      "epoch": 0.1328719723183391,
      "grad_norm": 0.2813468277454376,
      "learning_rate": 0.0009741588628463382,
      "loss": 5.0399,
      "step": 192
    },
    {
      "epoch": 0.1439446366782007,
      "grad_norm": 0.27917245030403137,
      "learning_rate": 0.0009681624439736599,
      "loss": 5.0001,
      "step": 208
    },
    {
      "epoch": 0.15501730103806227,
      "grad_norm": 0.23994863033294678,
      "learning_rate": 0.0009615634468894752,
      "loss": 4.9829,
      "step": 224
    },
    {
      "epoch": 0.16608996539792387,
      "grad_norm": 0.18816818296909332,
      "learning_rate": 0.0009543703652521542,
      "loss": 4.9348,
      "step": 240
    },
    {
      "epoch": 0.17301038062283736,
      "eval_bleu": 0.07082991922728418,
      "eval_cap_loss": 1.862730330052497,
      "eval_con_loss": 2.0789193881875687,
      "eval_loss": 3.941649715827423,
      "step": 250
    },
    {
      "epoch": 0.17301038062283736,
      "eval_bleu": 0.07082991922728418,
      "eval_cap_loss": 1.862730330052497,
      "eval_con_loss": 2.0789193881875687,
      "eval_loss": 3.941649715827423,
      "eval_runtime": 809.5614,
      "eval_samples_per_second": 19.036,
      "eval_steps_per_second": 2.38,
      "step": 250
    },
    {
      "epoch": 0.17716262975778546,
      "grad_norm": 0.24259746074676514,
      "learning_rate": 0.0009465924573743279,
      "loss": 4.9856,
      "step": 256
    },
    {
      "epoch": 0.18823529411764706,
      "grad_norm": 0.2658279836177826,
      "learning_rate": 0.0009382397343063877,
      "loss": 4.9553,
      "step": 272
    },
    {
      "epoch": 0.19930795847750865,
      "grad_norm": 0.18191872537136078,
      "learning_rate": 0.0009293229469511293,
      "loss": 4.9995,
      "step": 288
    },
    {
      "epoch": 0.21038062283737025,
      "grad_norm": 0.29174473881721497,
      "learning_rate": 0.000919853572226118,
      "loss": 5.0197,
      "step": 304
    },
    {
      "epoch": 0.22145328719723184,
      "grad_norm": 0.21258646249771118,
      "learning_rate": 0.0009098437982915953,
      "loss": 4.9636,
      "step": 320
    },
    {
      "epoch": 0.23252595155709344,
      "grad_norm": 0.18101181089878082,
      "learning_rate": 0.0008993065088629304,
      "loss": 4.9274,
      "step": 336
    },
    {
      "epoch": 0.24359861591695503,
      "grad_norm": 0.2041468620300293,
      "learning_rate": 0.0008882552666278186,
      "loss": 5.0077,
      "step": 352
    },
    {
      "epoch": 0.2546712802768166,
      "grad_norm": 0.3073960244655609,
      "learning_rate": 0.0008767042957895606,
      "loss": 4.9781,
      "step": 368
    },
    {
      "epoch": 0.2657439446366782,
      "grad_norm": 0.17274291813373566,
      "learning_rate": 0.0008646684637588991,
      "loss": 4.911,
      "step": 384
    },
    {
      "epoch": 0.2768166089965398,
      "grad_norm": 0.20907478034496307,
      "learning_rate": 0.0008521632620179735,
      "loss": 5.0089,
      "step": 400
    },
    {
      "epoch": 0.2878892733564014,
      "grad_norm": 0.2937226891517639,
      "learning_rate": 0.0008392047861810229,
      "loss": 4.9306,
      "step": 416
    },
    {
      "epoch": 0.29896193771626295,
      "grad_norm": 0.35890915989875793,
      "learning_rate": 0.0008258097152775044,
      "loss": 4.9363,
      "step": 432
    },
    {
      "epoch": 0.31003460207612454,
      "grad_norm": 0.19902297854423523,
      "learning_rate": 0.0008119952902842882,
      "loss": 4.9872,
      "step": 448
    },
    {
      "epoch": 0.32110726643598614,
      "grad_norm": 0.2420748621225357,
      "learning_rate": 0.0007977792919345632,
      "loss": 5.0178,
      "step": 464
    },
    {
      "epoch": 0.33217993079584773,
      "grad_norm": 0.21633224189281464,
      "learning_rate": 0.0007831800178320152,
      "loss": 4.9588,
      "step": 480
    },
    {
      "epoch": 0.34325259515570933,
      "grad_norm": 0.2581658363342285,
      "learning_rate": 0.0007682162588997332,
      "loss": 4.9214,
      "step": 496
    },
    {
      "epoch": 0.3460207612456747,
      "eval_bleu": 0.0743834461108624,
      "eval_cap_loss": 1.8318203130285544,
      "eval_con_loss": 2.0786952485553205,
      "eval_loss": 3.9105155644914187,
      "step": 500
    },
    {
      "epoch": 0.3460207612456747,
      "eval_bleu": 0.0743834461108624,
      "eval_cap_loss": 1.8318203130285544,
      "eval_con_loss": 2.0786952485553205,
      "eval_loss": 3.9105155644914187,
      "eval_runtime": 820.0211,
      "eval_samples_per_second": 18.793,
      "eval_steps_per_second": 2.35,
      "step": 500
    },
    {
      "epoch": 0.3543252595155709,
      "grad_norm": 0.24933111667633057,
      "learning_rate": 0.0007529072751941595,
      "loss": 4.9951,
      "step": 512
    },
    {
      "epoch": 0.3653979238754325,
      "grad_norm": 0.1792280375957489,
      "learning_rate": 0.0007372727711152087,
      "loss": 4.8813,
      "step": 528
    },
    {
      "epoch": 0.3764705882352941,
      "grad_norm": 0.13637550175189972,
      "learning_rate": 0.0007213328700444696,
      "loss": 4.9125,
      "step": 544
    },
    {
      "epoch": 0.3875432525951557,
      "grad_norm": 0.3409987688064575,
      "learning_rate": 0.0007051080884441287,
      "loss": 4.9073,
      "step": 560
    },
    {
      "epoch": 0.3986159169550173,
      "grad_norm": 0.2467418760061264,
      "learning_rate": 0.0006886193094499536,
      "loss": 4.9227,
      "step": 576
    },
    {
      "epoch": 0.4096885813148789,
      "grad_norm": 0.2751147747039795,
      "learning_rate": 0.000671887755992327,
      "loss": 4.9871,
      "step": 592
    },
    {
      "epoch": 0.4207612456747405,
      "grad_norm": 0.21447473764419556,
      "learning_rate": 0.000654934963479926,
      "loss": 4.9484,
      "step": 608
    },
    {
      "epoch": 0.4318339100346021,
      "grad_norm": 0.26142337918281555,
      "learning_rate": 0.0006377827520812061,
      "loss": 4.9204,
      "step": 624
    },
    {
      "epoch": 0.4429065743944637,
      "grad_norm": 0.3584180176258087,
      "learning_rate": 0.0006204531986393678,
      "loss": 5.0,
      "step": 640
    },
    {
      "epoch": 0.4539792387543253,
      "grad_norm": 0.2877948582172394,
      "learning_rate": 0.0006029686082569537,
      "loss": 5.0003,
      "step": 656
    },
    {
      "epoch": 0.46505190311418687,
      "grad_norm": 0.2049325406551361,
      "learning_rate": 0.000585351485586648,
      "loss": 4.9349,
      "step": 672
    },
    {
      "epoch": 0.47612456747404847,
      "grad_norm": 0.200492262840271,
      "learning_rate": 0.0005676245058652349,
      "loss": 4.9385,
      "step": 688
    },
    {
      "epoch": 0.48719723183391006,
      "grad_norm": 0.24712397158145905,
      "learning_rate": 0.000549810485727994,
      "loss": 4.9516,
      "step": 704
    },
    {
      "epoch": 0.4982698961937716,
      "grad_norm": 0.380741149187088,
      "learning_rate": 0.0005319323538411021,
      "loss": 4.9121,
      "step": 720
    },
    {
      "epoch": 0.5093425605536333,
      "grad_norm": 0.2339814305305481,
      "learning_rate": 0.0005140131213898345,
      "loss": 4.9058,
      "step": 736
    },
    {
      "epoch": 0.5190311418685121,
      "eval_bleu": 0.0755992902414665,
      "eval_cap_loss": 1.8195727369484656,
      "eval_con_loss": 2.0776209530585406,
      "eval_loss": 3.897193689945144,
      "step": 750
    },
    {
      "epoch": 0.5190311418685121,
      "eval_bleu": 0.0755992902414665,
      "eval_cap_loss": 1.8195727369484656,
      "eval_con_loss": 2.0776209530585406,
      "eval_loss": 3.897193689945144,
      "eval_runtime": 819.7019,
      "eval_samples_per_second": 18.801,
      "eval_steps_per_second": 2.351,
      "step": 750
    },
    {
      "epoch": 0.5204152249134948,
      "grad_norm": 0.23871256411075592,
      "learning_rate": 0.0004960758524605593,
      "loss": 4.9217,
      "step": 752
    },
    {
      "epoch": 0.5314878892733564,
      "grad_norm": 0.19021867215633392,
      "learning_rate": 0.0004781436343546391,
      "loss": 4.9383,
      "step": 768
    },
    {
      "epoch": 0.542560553633218,
      "grad_norm": 0.39012765884399414,
      "learning_rate": 0.0004602395478724539,
      "loss": 4.8852,
      "step": 784
    },
    {
      "epoch": 0.5536332179930796,
      "grad_norm": 0.2299482375383377,
      "learning_rate": 0.00044238663760578963,
      "loss": 4.9461,
      "step": 800
    },
    {
      "epoch": 0.5647058823529412,
      "grad_norm": 0.20332716405391693,
      "learning_rate": 0.0004246078822768339,
      "loss": 4.8672,
      "step": 816
    },
    {
      "epoch": 0.5757785467128028,
      "grad_norm": 0.2742558717727661,
      "learning_rate": 0.00040692616516195134,
      "loss": 5.0016,
      "step": 832
    },
    {
      "epoch": 0.5868512110726644,
      "grad_norm": 0.18639664351940155,
      "learning_rate": 0.0003893642446383089,
      "loss": 4.8718,
      "step": 848
    },
    {
      "epoch": 0.5979238754325259,
      "grad_norm": 0.14423900842666626,
      "learning_rate": 0.00037194472489126174,
      "loss": 4.9254,
      "step": 864
    },
    {
      "epoch": 0.6089965397923875,
      "grad_norm": 0.25753453373908997,
      "learning_rate": 0.00035469002682019933,
      "loss": 5.0226,
      "step": 880
    },
    {
      "epoch": 0.6200692041522491,
      "grad_norm": 0.29433926939964294,
      "learning_rate": 0.00033762235918030425,
      "loss": 4.9075,
      "step": 896
    },
    {
      "epoch": 0.6311418685121107,
      "grad_norm": 0.16973747313022614,
      "learning_rate": 0.0003207636899973617,
      "loss": 4.9367,
      "step": 912
    },
    {
      "epoch": 0.6422145328719723,
      "grad_norm": 0.23007385432720184,
      "learning_rate": 0.0003041357182924178,
      "loss": 4.8899,
      "step": 928
    },
    {
      "epoch": 0.6532871972318339,
      "grad_norm": 0.23643143475055695,
      "learning_rate": 0.000287759846152675,
      "loss": 4.8813,
      "step": 944
    },
    {
      "epoch": 0.6643598615916955,
      "grad_norm": 0.17182067036628723,
      "learning_rate": 0.00027165715118457735,
      "loss": 4.9201,
      "step": 960
    },
    {
      "epoch": 0.6754325259515571,
      "grad_norm": 0.29925912618637085,
      "learning_rate": 0.0002558483593845372,
      "loss": 4.898,
      "step": 976
    },
    {
      "epoch": 0.6865051903114187,
      "grad_norm": 0.17474640905857086,
      "learning_rate": 0.00024035381846222555,
      "loss": 4.8839,
      "step": 992
    },
    {
      "epoch": 0.6920415224913494,
      "eval_bleu": 0.07638727198557162,
      "eval_cap_loss": 1.797022891217832,
      "eval_con_loss": 2.0767276909416923,
      "eval_loss": 3.87375058110786,
      "step": 1000
    },
    {
      "epoch": 0.6920415224913494,
      "eval_bleu": 0.07638727198557162,
      "eval_cap_loss": 1.797022891217832,
      "eval_con_loss": 2.0767276909416923,
      "eval_loss": 3.87375058110786,
      "eval_runtime": 815.983,
      "eval_samples_per_second": 18.886,
      "eval_steps_per_second": 2.362,
      "step": 1000
    },
    {
      "epoch": 0.6975778546712803,
      "grad_norm": 0.2200097143650055,
      "learning_rate": 0.00022519347165076065,
      "loss": 4.9202,
      "step": 1008
    },
    {
      "epoch": 0.7086505190311418,
      "grad_norm": 0.2346457690000534,
      "learning_rate": 0.00021038683203750092,
      "loss": 4.8698,
      "step": 1024
    },
    {
      "epoch": 0.7197231833910035,
      "grad_norm": 0.24073997139930725,
      "learning_rate": 0.00019595295744848825,
      "loss": 5.01,
      "step": 1040
    },
    {
      "epoch": 0.730795847750865,
      "grad_norm": 0.2730049192905426,
      "learning_rate": 0.00018191042591886197,
      "loss": 4.8847,
      "step": 1056
    },
    {
      "epoch": 0.7418685121107267,
      "grad_norm": 0.21411365270614624,
      "learning_rate": 0.00016827731178081822,
      "loss": 4.9802,
      "step": 1072
    },
    {
      "epoch": 0.7529411764705882,
      "grad_norm": 0.2260352373123169,
      "learning_rate": 0.0001550711623998926,
      "loss": 4.9593,
      "step": 1088
    },
    {
      "epoch": 0.7640138408304499,
      "grad_norm": 0.2655484974384308,
      "learning_rate": 0.0001423089755895095,
      "loss": 4.8097,
      "step": 1104
    },
    {
      "epoch": 0.7750865051903114,
      "grad_norm": 0.2067459672689438,
      "learning_rate": 0.0001300071777328658,
      "loss": 4.947,
      "step": 1120
    },
    {
      "epoch": 0.7861591695501731,
      "grad_norm": 0.21391142904758453,
      "learning_rate": 0.00011818160264031097,
      "loss": 4.8673,
      "step": 1136
    },
    {
      "epoch": 0.7972318339100346,
      "grad_norm": 0.3288532793521881,
      "learning_rate": 0.00010684747116943683,
      "loss": 4.9652,
      "step": 1152
    },
    {
      "epoch": 0.8083044982698961,
      "grad_norm": 0.1693842113018036,
      "learning_rate": 9.60193716341039e-05,
      "loss": 4.9003,
      "step": 1168
    },
    {
      "epoch": 0.8193771626297578,
      "grad_norm": 0.3056645691394806,
      "learning_rate": 8.571124102762767e-05,
      "loss": 4.9436,
      "step": 1184
    },
    {
      "epoch": 0.8304498269896193,
      "grad_norm": 0.3365015387535095,
      "learning_rate": 7.593634708428437e-05,
      "loss": 5.0349,
      "step": 1200
    },
    {
      "epoch": 0.841522491349481,
      "grad_norm": 0.2918214201927185,
      "learning_rate": 6.670727120223142e-05,
      "loss": 4.8812,
      "step": 1216
    },
    {
      "epoch": 0.8525951557093425,
      "grad_norm": 0.2237936407327652,
      "learning_rate": 5.8035892249820085e-05,
      "loss": 4.8842,
      "step": 1232
    },
    {
      "epoch": 0.8636678200692042,
      "grad_norm": 0.23957201838493347,
      "learning_rate": 4.993337127614273e-05,
      "loss": 4.9281,
      "step": 1248
    },
    {
      "epoch": 0.8650519031141869,
      "eval_bleu": 0.07725825664704841,
      "eval_cap_loss": 1.781457468055355,
      "eval_con_loss": 2.0761843878924693,
      "eval_loss": 3.8576418577727716,
      "step": 1250
    },
    {
      "epoch": 0.8650519031141869,
      "eval_bleu": 0.07725825664704841,
      "eval_cap_loss": 1.781457468055355,
      "eval_con_loss": 2.0761843878924693,
      "eval_loss": 3.8576418577727716,
      "eval_runtime": 819.4117,
      "eval_samples_per_second": 18.807,
      "eval_steps_per_second": 2.352,
      "step": 1250
    },
    {
      "epoch": 0.8747404844290657,
      "grad_norm": 0.09429222345352173,
      "learning_rate": 4.2410137145495964e-05,
      "loss": 4.8982,
      "step": 1264
    },
    {
      "epoch": 0.8858131487889274,
      "grad_norm": 0.23660314083099365,
      "learning_rate": 3.54758731142486e-05,
      "loss": 4.9127,
      "step": 1280
    },
    {
      "epoch": 0.8968858131487889,
      "grad_norm": 0.2092629075050354,
      "learning_rate": 2.9139504367391158e-05,
      "loss": 5.0183,
      "step": 1296
    },
    {
      "epoch": 0.9079584775086506,
      "grad_norm": 0.2870592772960663,
      "learning_rate": 2.3409186530809423e-05,
      "loss": 4.9319,
      "step": 1312
    },
    {
      "epoch": 0.9190311418685121,
      "grad_norm": 0.23483090102672577,
      "learning_rate": 1.8292295174068717e-05,
      "loss": 4.836,
      "step": 1328
    },
    {
      "epoch": 0.9301038062283737,
      "grad_norm": 0.24700886011123657,
      "learning_rate": 1.3795416317218035e-05,
      "loss": 4.9113,
      "step": 1344
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.20898666977882385,
      "learning_rate": 9.924337953834795e-06,
      "loss": 4.9519,
      "step": 1360
    },
    {
      "epoch": 0.9522491349480969,
      "grad_norm": 0.14755718410015106,
      "learning_rate": 6.684042601220186e-06,
      "loss": 4.8827,
      "step": 1376
    },
    {
      "epoch": 0.9633217993079585,
      "grad_norm": 0.2121778130531311,
      "learning_rate": 4.078700887333364e-06,
      "loss": 4.861,
      "step": 1392
    },
    {
      "epoch": 0.9743944636678201,
      "grad_norm": 0.2525366544723511,
      "learning_rate": 2.1116661827202956e-06,
      "loss": 4.8674,
      "step": 1408
    },
    {
      "epoch": 0.9854671280276817,
      "grad_norm": 0.278390109539032,
      "learning_rate": 7.854702843449468e-07,
      "loss": 4.9493,
      "step": 1424
    },
    {
      "epoch": 0.9965397923875432,
      "grad_norm": 0.22038041055202484,
      "learning_rate": 1.0182015687909552e-07,
      "loss": 4.9065,
      "step": 1440
    }
  ],
  "logging_steps": 16,
  "max_steps": 1445,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 250,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null,
  "tau_value": 0.8081
}