{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.983079526226735,
  "eval_steps": 500,
  "global_step": 59000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 1.9830795262267346e-05,
      "loss": 2.6117,
      "step": 500
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9661590524534688e-05,
      "loss": 1.7548,
      "step": 1000
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.9492385786802032e-05,
      "loss": 1.4794,
      "step": 1500
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.9323181049069374e-05,
      "loss": 1.3088,
      "step": 2000
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.915397631133672e-05,
      "loss": 1.1959,
      "step": 2500
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.8984771573604063e-05,
      "loss": 1.1127,
      "step": 3000
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.8815566835871405e-05,
      "loss": 1.0536,
      "step": 3500
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.864636209813875e-05,
      "loss": 0.9879,
      "step": 4000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.8477157360406094e-05,
      "loss": 0.9587,
      "step": 4500
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.830795262267344e-05,
      "loss": 0.9123,
      "step": 5000
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.813874788494078e-05,
      "loss": 0.8893,
      "step": 5500
    },
    {
      "epoch": 1.0,
      "eval_bleu": 4.0777,
      "eval_gen_len": 18.1211,
      "eval_loss": 2.386338472366333,
      "eval_runtime": 52.6636,
      "eval_samples_per_second": 39.211,
      "eval_steps_per_second": 1.234,
      "step": 5910
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.7969543147208125e-05,
      "loss": 0.8853,
      "step": 6000
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.7800338409475466e-05,
      "loss": 0.8313,
      "step": 6500
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.763113367174281e-05,
      "loss": 0.8256,
      "step": 7000
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.7461928934010152e-05,
      "loss": 0.8104,
      "step": 7500
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.7292724196277497e-05,
      "loss": 0.8096,
      "step": 8000
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.7123519458544842e-05,
      "loss": 0.7792,
      "step": 8500
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.6954314720812183e-05,
      "loss": 0.7643,
      "step": 9000
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.6785109983079528e-05,
      "loss": 0.7603,
      "step": 9500
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.661590524534687e-05,
      "loss": 0.7385,
      "step": 10000
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.6446700507614214e-05,
      "loss": 0.7424,
      "step": 10500
    },
    {
      "epoch": 1.86,
      "learning_rate": 1.627749576988156e-05,
      "loss": 0.7244,
      "step": 11000
    },
    {
      "epoch": 1.95,
      "learning_rate": 1.61082910321489e-05,
      "loss": 0.7182,
      "step": 11500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 4.0885,
      "eval_gen_len": 18.231,
      "eval_loss": 2.2319083213806152,
      "eval_runtime": 49.7931,
      "eval_samples_per_second": 41.472,
      "eval_steps_per_second": 1.305,
      "step": 11820
    },
    {
      "epoch": 2.03,
      "learning_rate": 1.5939086294416245e-05,
      "loss": 0.7002,
      "step": 12000
    },
    {
      "epoch": 2.12,
      "learning_rate": 1.576988155668359e-05,
      "loss": 0.6985,
      "step": 12500
    },
    {
      "epoch": 2.2,
      "learning_rate": 1.5600676818950934e-05,
      "loss": 0.6908,
      "step": 13000
    },
    {
      "epoch": 2.28,
      "learning_rate": 1.5431472081218276e-05,
      "loss": 0.6684,
      "step": 13500
    },
    {
      "epoch": 2.37,
      "learning_rate": 1.526226734348562e-05,
      "loss": 0.6749,
      "step": 14000
    },
    {
      "epoch": 2.45,
      "learning_rate": 1.5093062605752962e-05,
      "loss": 0.6763,
      "step": 14500
    },
    {
      "epoch": 2.54,
      "learning_rate": 1.4923857868020306e-05,
      "loss": 0.6526,
      "step": 15000
    },
    {
      "epoch": 2.62,
      "learning_rate": 1.4754653130287648e-05,
      "loss": 0.6536,
      "step": 15500
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.4585448392554992e-05,
      "loss": 0.6345,
      "step": 16000
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4416243654822337e-05,
      "loss": 0.6522,
      "step": 16500
    },
    {
      "epoch": 2.88,
      "learning_rate": 1.424703891708968e-05,
      "loss": 0.6379,
      "step": 17000
    },
    {
      "epoch": 2.96,
      "learning_rate": 1.4077834179357023e-05,
      "loss": 0.6277,
      "step": 17500
    },
    {
      "epoch": 3.0,
      "eval_bleu": 4.2683,
      "eval_gen_len": 18.2155,
      "eval_loss": 2.1325912475585938,
      "eval_runtime": 49.4798,
      "eval_samples_per_second": 41.734,
      "eval_steps_per_second": 1.314,
      "step": 17730
    },
    {
      "epoch": 3.05,
      "learning_rate": 1.3908629441624366e-05,
      "loss": 0.6309,
      "step": 18000
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.3739424703891711e-05,
      "loss": 0.6427,
      "step": 18500
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.3570219966159052e-05,
      "loss": 0.614,
      "step": 19000
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.3401015228426397e-05,
      "loss": 0.6126,
      "step": 19500
    },
    {
      "epoch": 3.38,
      "learning_rate": 1.323181049069374e-05,
      "loss": 0.6118,
      "step": 20000
    },
    {
      "epoch": 3.47,
      "learning_rate": 1.3062605752961083e-05,
      "loss": 0.5909,
      "step": 20500
    },
    {
      "epoch": 3.55,
      "learning_rate": 1.2893401015228428e-05,
      "loss": 0.5972,
      "step": 21000
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.2724196277495771e-05,
      "loss": 0.6019,
      "step": 21500
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.2554991539763116e-05,
      "loss": 0.5879,
      "step": 22000
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.2385786802030457e-05,
      "loss": 0.5931,
      "step": 22500
    },
    {
      "epoch": 3.89,
      "learning_rate": 1.2216582064297802e-05,
      "loss": 0.5979,
      "step": 23000
    },
    {
      "epoch": 3.98,
      "learning_rate": 1.2047377326565143e-05,
      "loss": 0.5981,
      "step": 23500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 4.3741,
      "eval_gen_len": 18.3075,
      "eval_loss": 2.0639102458953857,
      "eval_runtime": 49.7175,
      "eval_samples_per_second": 41.535,
      "eval_steps_per_second": 1.307,
      "step": 23640
    },
    {
      "epoch": 4.06,
      "learning_rate": 1.1878172588832488e-05,
      "loss": 0.5758,
      "step": 24000
    },
    {
      "epoch": 4.15,
      "learning_rate": 1.1708967851099833e-05,
      "loss": 0.5724,
      "step": 24500
    },
    {
      "epoch": 4.23,
      "learning_rate": 1.1539763113367176e-05,
      "loss": 0.5708,
      "step": 25000
    },
    {
      "epoch": 4.31,
      "learning_rate": 1.1370558375634519e-05,
      "loss": 0.5764,
      "step": 25500
    },
    {
      "epoch": 4.4,
      "learning_rate": 1.1201353637901862e-05,
      "loss": 0.5743,
      "step": 26000
    },
    {
      "epoch": 4.48,
      "learning_rate": 1.1032148900169206e-05,
      "loss": 0.5715,
      "step": 26500
    },
    {
      "epoch": 4.57,
      "learning_rate": 1.0862944162436548e-05,
      "loss": 0.5688,
      "step": 27000
    },
    {
      "epoch": 4.65,
      "learning_rate": 1.0693739424703892e-05,
      "loss": 0.5776,
      "step": 27500
    },
    {
      "epoch": 4.74,
      "learning_rate": 1.0524534686971236e-05,
      "loss": 0.5712,
      "step": 28000
    },
    {
      "epoch": 4.82,
      "learning_rate": 1.035532994923858e-05,
      "loss": 0.5544,
      "step": 28500
    },
    {
      "epoch": 4.91,
      "learning_rate": 1.0186125211505923e-05,
      "loss": 0.5657,
      "step": 29000
    },
    {
      "epoch": 4.99,
      "learning_rate": 1.0016920473773266e-05,
      "loss": 0.5585,
      "step": 29500
    },
    {
      "epoch": 5.0,
      "eval_bleu": 4.5943,
      "eval_gen_len": 18.309,
      "eval_loss": 2.012233018875122,
      "eval_runtime": 49.6368,
      "eval_samples_per_second": 41.602,
      "eval_steps_per_second": 1.31,
      "step": 29550
    },
    {
      "epoch": 5.08,
      "learning_rate": 9.84771573604061e-06,
      "loss": 0.5618,
      "step": 30000
    },
    {
      "epoch": 5.16,
      "learning_rate": 9.678510998307954e-06,
      "loss": 0.5427,
      "step": 30500
    },
    {
      "epoch": 5.25,
      "learning_rate": 9.509306260575297e-06,
      "loss": 0.5469,
      "step": 31000
    },
    {
      "epoch": 5.33,
      "learning_rate": 9.34010152284264e-06,
      "loss": 0.5387,
      "step": 31500
    },
    {
      "epoch": 5.41,
      "learning_rate": 9.170896785109983e-06,
      "loss": 0.5454,
      "step": 32000
    },
    {
      "epoch": 5.5,
      "learning_rate": 9.001692047377328e-06,
      "loss": 0.5474,
      "step": 32500
    },
    {
      "epoch": 5.58,
      "learning_rate": 8.832487309644671e-06,
      "loss": 0.5566,
      "step": 33000
    },
    {
      "epoch": 5.67,
      "learning_rate": 8.663282571912014e-06,
      "loss": 0.5339,
      "step": 33500
    },
    {
      "epoch": 5.75,
      "learning_rate": 8.494077834179357e-06,
      "loss": 0.5437,
      "step": 34000
    },
    {
      "epoch": 5.84,
      "learning_rate": 8.324873096446702e-06,
      "loss": 0.5408,
      "step": 34500
    },
    {
      "epoch": 5.92,
      "learning_rate": 8.155668358714045e-06,
      "loss": 0.537,
      "step": 35000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 4.6946,
      "eval_gen_len": 18.3172,
      "eval_loss": 1.9772592782974243,
      "eval_runtime": 49.5084,
      "eval_samples_per_second": 41.71,
      "eval_steps_per_second": 1.313,
      "step": 35460
    },
    {
      "epoch": 6.01,
      "learning_rate": 7.986463620981388e-06,
      "loss": 0.5411,
      "step": 35500
    },
    {
      "epoch": 6.09,
      "learning_rate": 7.817258883248731e-06,
      "loss": 0.5292,
      "step": 36000
    },
    {
      "epoch": 6.18,
      "learning_rate": 7.648054145516076e-06,
      "loss": 0.5436,
      "step": 36500
    },
    {
      "epoch": 6.26,
      "learning_rate": 7.478849407783419e-06,
      "loss": 0.5317,
      "step": 37000
    },
    {
      "epoch": 6.35,
      "learning_rate": 7.309644670050762e-06,
      "loss": 0.516,
      "step": 37500
    },
    {
      "epoch": 6.43,
      "learning_rate": 7.140439932318105e-06,
      "loss": 0.522,
      "step": 38000
    },
    {
      "epoch": 6.51,
      "learning_rate": 6.9712351945854495e-06,
      "loss": 0.528,
      "step": 38500
    },
    {
      "epoch": 6.6,
      "learning_rate": 6.8020304568527926e-06,
      "loss": 0.53,
      "step": 39000
    },
    {
      "epoch": 6.68,
      "learning_rate": 6.6328257191201364e-06,
      "loss": 0.5282,
      "step": 39500
    },
    {
      "epoch": 6.77,
      "learning_rate": 6.4636209813874795e-06,
      "loss": 0.5386,
      "step": 40000
    },
    {
      "epoch": 6.85,
      "learning_rate": 6.2944162436548225e-06,
      "loss": 0.5243,
      "step": 40500
    },
    {
      "epoch": 6.94,
      "learning_rate": 6.125211505922166e-06,
      "loss": 0.513,
      "step": 41000
    },
    {
      "epoch": 7.0,
      "eval_bleu": 4.9177,
      "eval_gen_len": 18.2939,
      "eval_loss": 1.94707190990448,
      "eval_runtime": 49.7911,
      "eval_samples_per_second": 41.473,
      "eval_steps_per_second": 1.305,
      "step": 41370
    },
    {
      "epoch": 7.02,
      "learning_rate": 5.9560067681895094e-06,
      "loss": 0.5096,
      "step": 41500
    },
    {
      "epoch": 7.11,
      "learning_rate": 5.7868020304568525e-06,
      "loss": 0.5175,
      "step": 42000
    },
    {
      "epoch": 7.19,
      "learning_rate": 5.617597292724196e-06,
      "loss": 0.5205,
      "step": 42500
    },
    {
      "epoch": 7.28,
      "learning_rate": 5.44839255499154e-06,
      "loss": 0.5079,
      "step": 43000
    },
    {
      "epoch": 7.36,
      "learning_rate": 5.279187817258884e-06,
      "loss": 0.5067,
      "step": 43500
    },
    {
      "epoch": 7.45,
      "learning_rate": 5.109983079526227e-06,
      "loss": 0.5174,
      "step": 44000
    },
    {
      "epoch": 7.53,
      "learning_rate": 4.94077834179357e-06,
      "loss": 0.5153,
      "step": 44500
    },
    {
      "epoch": 7.61,
      "learning_rate": 4.771573604060914e-06,
      "loss": 0.5227,
      "step": 45000
    },
    {
      "epoch": 7.7,
      "learning_rate": 4.602368866328257e-06,
      "loss": 0.5146,
      "step": 45500
    },
    {
      "epoch": 7.78,
      "learning_rate": 4.433164128595601e-06,
      "loss": 0.5065,
      "step": 46000
    },
    {
      "epoch": 7.87,
      "learning_rate": 4.263959390862945e-06,
      "loss": 0.5143,
      "step": 46500
    },
    {
      "epoch": 7.95,
      "learning_rate": 4.094754653130288e-06,
      "loss": 0.5234,
      "step": 47000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 5.0134,
      "eval_gen_len": 18.2939,
      "eval_loss": 1.928272008895874,
      "eval_runtime": 49.7105,
      "eval_samples_per_second": 41.54,
      "eval_steps_per_second": 1.308,
      "step": 47280
    },
    {
      "epoch": 8.04,
      "learning_rate": 3.925549915397631e-06,
      "loss": 0.5043,
      "step": 47500
    },
    {
      "epoch": 8.12,
      "learning_rate": 3.756345177664975e-06,
      "loss": 0.5272,
      "step": 48000
    },
    {
      "epoch": 8.21,
      "learning_rate": 3.5871404399323183e-06,
      "loss": 0.5068,
      "step": 48500
    },
    {
      "epoch": 8.29,
      "learning_rate": 3.417935702199662e-06,
      "loss": 0.5071,
      "step": 49000
    },
    {
      "epoch": 8.38,
      "learning_rate": 3.2487309644670053e-06,
      "loss": 0.5061,
      "step": 49500
    },
    {
      "epoch": 8.46,
      "learning_rate": 3.079526226734349e-06,
      "loss": 0.4997,
      "step": 50000
    },
    {
      "epoch": 8.54,
      "learning_rate": 2.910321489001692e-06,
      "loss": 0.4953,
      "step": 50500
    },
    {
      "epoch": 8.63,
      "learning_rate": 2.7411167512690357e-06,
      "loss": 0.5072,
      "step": 51000
    },
    {
      "epoch": 8.71,
      "learning_rate": 2.571912013536379e-06,
      "loss": 0.512,
      "step": 51500
    },
    {
      "epoch": 8.8,
      "learning_rate": 2.4027072758037226e-06,
      "loss": 0.5043,
      "step": 52000
    },
    {
      "epoch": 8.88,
      "learning_rate": 2.233502538071066e-06,
      "loss": 0.499,
      "step": 52500
    },
    {
      "epoch": 8.97,
      "learning_rate": 2.0642978003384095e-06,
      "loss": 0.5133,
      "step": 53000
    },
    {
      "epoch": 9.0,
      "eval_bleu": 5.0308,
      "eval_gen_len": 18.3027,
      "eval_loss": 1.91579270362854,
      "eval_runtime": 49.6621,
      "eval_samples_per_second": 41.581,
      "eval_steps_per_second": 1.309,
      "step": 53190
    },
    {
      "epoch": 9.05,
      "learning_rate": 1.8950930626057532e-06,
      "loss": 0.5146,
      "step": 53500
    },
    {
      "epoch": 9.14,
      "learning_rate": 1.7258883248730964e-06,
      "loss": 0.4968,
      "step": 54000
    },
    {
      "epoch": 9.22,
      "learning_rate": 1.55668358714044e-06,
      "loss": 0.4806,
      "step": 54500
    },
    {
      "epoch": 9.31,
      "learning_rate": 1.3874788494077834e-06,
      "loss": 0.4979,
      "step": 55000
    },
    {
      "epoch": 9.39,
      "learning_rate": 1.218274111675127e-06,
      "loss": 0.4971,
      "step": 55500
    },
    {
      "epoch": 9.48,
      "learning_rate": 1.0490693739424705e-06,
      "loss": 0.5013,
      "step": 56000
    },
    {
      "epoch": 9.56,
      "learning_rate": 8.79864636209814e-07,
      "loss": 0.5119,
      "step": 56500
    },
    {
      "epoch": 9.64,
      "learning_rate": 7.106598984771574e-07,
      "loss": 0.5076,
      "step": 57000
    },
    {
      "epoch": 9.73,
      "learning_rate": 5.414551607445009e-07,
      "loss": 0.508,
      "step": 57500
    },
    {
      "epoch": 9.81,
      "learning_rate": 3.7225042301184434e-07,
      "loss": 0.4978,
      "step": 58000
    },
    {
      "epoch": 9.9,
      "learning_rate": 2.0304568527918783e-07,
      "loss": 0.5046,
      "step": 58500
    },
    {
      "epoch": 9.98,
      "learning_rate": 3.3840947546531305e-08,
      "loss": 0.5074,
      "step": 59000
    }
  ],
  "logging_steps": 500,
  "max_steps": 59100,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 7.765950061873152e+16,
  "trial_name": null,
  "trial_params": null
}