{
  "best_metric": 1.98361849433367,
  "best_model_checkpoint": "experiments/translation/gpt2/2025-06-24_10-08-42/checkpoint-8000",
  "epoch": 6.4617558022199795,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 4.8e-06,
      "loss": 5.4804,
      "step": 50
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.799999999999998e-06,
      "loss": 4.9145,
      "step": 100
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.4799999999999999e-05,
      "loss": 4.7992,
      "step": 150
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.98e-05,
      "loss": 4.7128,
      "step": 200
    },
    {
      "epoch": 0.2,
      "learning_rate": 2.4799999999999996e-05,
      "loss": 4.6691,
      "step": 250
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.9799999999999996e-05,
      "loss": 4.6397,
      "step": 300
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.48e-05,
      "loss": 4.591,
      "step": 350
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.979999999999999e-05,
      "loss": 4.5646,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.48e-05,
      "loss": 4.5257,
      "step": 450
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.98e-05,
      "loss": 4.4901,
      "step": 500
    },
    {
      "epoch": 0.4,
      "eval_bleu": 0.11633427622704212,
      "eval_loss": 4.419718265533447,
      "eval_runtime": 1914.9377,
      "eval_samples_per_second": 4.6,
      "eval_steps_per_second": 0.144,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 5.48e-05,
      "loss": 4.5062,
      "step": 550
    },
    {
      "epoch": 0.48,
      "learning_rate": 5.98e-05,
      "loss": 4.4699,
      "step": 600
    },
    {
      "epoch": 0.52,
      "learning_rate": 6.479999999999999e-05,
      "loss": 4.4781,
      "step": 650
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.979999999999999e-05,
      "loss": 4.433,
      "step": 700
    },
    {
      "epoch": 0.61,
      "learning_rate": 7.479999999999999e-05,
      "loss": 4.4437,
      "step": 750
    },
    {
      "epoch": 0.65,
      "learning_rate": 7.98e-05,
      "loss": 4.4278,
      "step": 800
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.48e-05,
      "loss": 4.4062,
      "step": 850
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.98e-05,
      "loss": 4.3717,
      "step": 900
    },
    {
      "epoch": 0.77,
      "learning_rate": 9.479999999999999e-05,
      "loss": 4.3586,
      "step": 950
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.979999999999999e-05,
      "loss": 4.3929,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "eval_bleu": 0.15546068077729897,
      "eval_loss": 4.284366607666016,
      "eval_runtime": 1911.4105,
      "eval_samples_per_second": 4.608,
      "eval_steps_per_second": 0.144,
      "step": 1000
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00010479999999999999,
      "loss": 4.3491,
      "step": 1050
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00010979999999999999,
      "loss": 4.3284,
      "step": 1100
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00011479999999999999,
      "loss": 4.3209,
      "step": 1150
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00011979999999999998,
      "loss": 4.3172,
      "step": 1200
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00012479999999999997,
      "loss": 4.3263,
      "step": 1250
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00012979999999999998,
      "loss": 4.1922,
      "step": 1300
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00013479999999999997,
      "loss": 4.1936,
      "step": 1350
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00013979999999999998,
      "loss": 4.1718,
      "step": 1400
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001448,
      "loss": 4.1767,
      "step": 1450
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00014979999999999998,
      "loss": 4.1589,
      "step": 1500
    },
    {
      "epoch": 1.21,
      "eval_bleu": 0.28584924145284585,
      "eval_loss": 4.1763997077941895,
      "eval_runtime": 1912.3927,
      "eval_samples_per_second": 4.606,
      "eval_steps_per_second": 0.144,
      "step": 1500
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00014999895016087306,
      "loss": 4.1576,
      "step": 1550
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00014999562388001478,
      "loss": 4.1006,
      "step": 1600
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00014999001943582685,
      "loss": 4.1227,
      "step": 1650
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001499821369985571,
      "loss": 4.1087,
      "step": 1700
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001499719768076526,
      "loss": 4.0728,
      "step": 1750
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014995953917175227,
      "loss": 4.0705,
      "step": 1800
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00014994482446867774,
      "loss": 4.0425,
      "step": 1850
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00014992783314542174,
      "loss": 4.0156,
      "step": 1900
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00014990856571813448,
      "loss": 4.0317,
      "step": 1950
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001498870227721081,
      "loss": 3.9819,
      "step": 2000
    },
    {
      "epoch": 1.62,
      "eval_bleu": 0.39689971475030256,
      "eval_loss": 3.9646406173706055,
      "eval_runtime": 1910.9822,
      "eval_samples_per_second": 4.609,
      "eval_steps_per_second": 0.144,
      "step": 2000
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0001498632049617587,
      "loss": 3.9671,
      "step": 2050
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00014983711301060673,
      "loss": 3.9195,
      "step": 2100
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00014980874771125478,
      "loss": 3.9144,
      "step": 2150
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00014977810992536359,
      "loss": 3.9092,
      "step": 2200
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00014974520058362584,
      "loss": 3.9027,
      "step": 2250
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00014971002068573793,
      "loss": 3.8732,
      "step": 2300
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00014967257130036961,
      "loss": 3.8439,
      "step": 2350
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00014963285356513152,
      "loss": 3.8375,
      "step": 2400
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001495908686865405,
      "loss": 3.8019,
      "step": 2450
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.00014954661793998317,
      "loss": 3.7382,
      "step": 2500
    },
    {
      "epoch": 2.02,
      "eval_bleu": 0.5913530218314237,
      "eval_loss": 3.787797212600708,
      "eval_runtime": 1914.1909,
      "eval_samples_per_second": 4.601,
      "eval_steps_per_second": 0.144,
      "step": 2500
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0001495001026696769,
      "loss": 3.5496,
      "step": 2550
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.00014945132428862936,
      "loss": 3.5282,
      "step": 2600
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.00014940028427859524,
      "loss": 3.5239,
      "step": 2650
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.00014934698419003133,
      "loss": 3.4925,
      "step": 2700
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0001492914256420496,
      "loss": 3.5009,
      "step": 2750
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.00014923361032236776,
      "loss": 3.4963,
      "step": 2800
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.00014917353998725823,
      "loss": 3.491,
      "step": 2850
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.00014911121646149456,
      "loss": 3.4631,
      "step": 2900
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.00014904664163829616,
      "loss": 3.4539,
      "step": 2950
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.00014897981747927076,
      "loss": 3.4567,
      "step": 3000
    },
    {
      "epoch": 2.42,
      "eval_bleu": 0.7641516646931262,
      "eval_loss": 3.635390520095825,
      "eval_runtime": 1917.6351,
      "eval_samples_per_second": 4.593,
      "eval_steps_per_second": 0.144,
      "step": 3000
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.00014891074601435482,
      "loss": 3.4458,
      "step": 3050
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.00014883942934175177,
      "loss": 3.4317,
      "step": 3100
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0001487658696278684,
      "loss": 3.4017,
      "step": 3150
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.000148690069107249,
      "loss": 3.4068,
      "step": 3200
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.00014861203008250745,
      "loss": 3.3839,
      "step": 3250
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001485317549242574,
      "loss": 3.4197,
      "step": 3300
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.00014844924607104,
      "loss": 3.3836,
      "step": 3350
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.00014836450602925014,
      "loss": 3.3611,
      "step": 3400
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.00014827753737306008,
      "loss": 3.3541,
      "step": 3450
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.00014818834274434134,
      "loss": 3.3587,
      "step": 3500
    },
    {
      "epoch": 2.83,
      "eval_bleu": 1.1491131179238592,
      "eval_loss": 3.5020267963409424,
      "eval_runtime": 1914.8223,
      "eval_samples_per_second": 4.6,
      "eval_steps_per_second": 0.144,
      "step": 3500
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.00014809692485258445,
      "loss": 3.3124,
      "step": 3550
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.00014800328647481662,
      "loss": 3.3311,
      "step": 3600
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.00014790743045551744,
      "loss": 3.302,
      "step": 3650
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.00014780935970653235,
      "loss": 3.2845,
      "step": 3700
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.00014770907720698426,
      "loss": 3.165,
      "step": 3750
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.00014760658600318318,
      "loss": 3.0311,
      "step": 3800
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.00014750188920853338,
      "loss": 3.0383,
      "step": 3850
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.00014739499000343914,
      "loss": 3.0399,
      "step": 3900
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0001472858916352079,
      "loss": 3.0327,
      "step": 3950
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0001471745974179517,
      "loss": 3.007,
      "step": 4000
    },
    {
      "epoch": 3.23,
      "eval_bleu": 1.106589199275075,
      "eval_loss": 3.435346841812134,
      "eval_runtime": 1918.0445,
      "eval_samples_per_second": 4.592,
      "eval_steps_per_second": 0.144,
      "step": 4000
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.00014706111073248656,
      "loss": 3.0285,
      "step": 4050
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0001469454350262297,
      "loss": 3.0335,
      "step": 4100
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0001468275738130948,
      "loss": 3.0331,
      "step": 4150
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0001467075306733854,
      "loss": 3.024,
      "step": 4200
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.000146585309253686,
      "loss": 3.0199,
      "step": 4250
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.00014646091326675126,
      "loss": 2.9867,
      "step": 4300
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.00014633434649139344,
      "loss": 3.0297,
      "step": 4350
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.00014620561277236722,
      "loss": 2.9971,
      "step": 4400
    },
    {
      "epoch": 3.59,
      "learning_rate": 0.0001460747160202534,
      "loss": 2.9872,
      "step": 4450
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.0001459416602113397,
      "loss": 2.9901,
      "step": 4500
    },
    {
      "epoch": 3.63,
      "eval_bleu": 1.4837165956404195,
      "eval_loss": 3.3302574157714844,
      "eval_runtime": 1928.4326,
      "eval_samples_per_second": 4.567,
      "eval_steps_per_second": 0.143,
      "step": 4500
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.00014580644938750012,
      "loss": 2.9842,
      "step": 4550
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.00014566908765607222,
      "loss": 2.9818,
      "step": 4600
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.00014552957918973226,
      "loss": 2.9784,
      "step": 4650
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.00014538792822636849,
      "loss": 2.974,
      "step": 4700
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.00014524413906895234,
      "loss": 2.9592,
      "step": 4750
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.00014509821608540784,
      "loss": 2.9519,
      "step": 4800
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.00014495016370847882,
      "loss": 2.9672,
      "step": 4850
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.00014479998643559435,
      "loss": 2.944,
      "step": 4900
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00014464768882873198,
      "loss": 2.9653,
      "step": 4950
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.00014449327551427935,
      "loss": 2.7312,
      "step": 5000
    },
    {
      "epoch": 4.04,
      "eval_bleu": 1.3796115530850706,
      "eval_loss": 3.2999939918518066,
      "eval_runtime": 1915.5145,
      "eval_samples_per_second": 4.598,
      "eval_steps_per_second": 0.144,
      "step": 5000
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.0001443367511828934,
      "loss": 2.6928,
      "step": 5050
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.0001441781205893582,
      "loss": 2.6899,
      "step": 5100
    },
    {
      "epoch": 4.16,
      "learning_rate": 0.00014401738855244028,
      "loss": 2.6697,
      "step": 5150
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.00014385455995474222,
      "loss": 2.6839,
      "step": 5200
    },
    {
      "epoch": 4.24,
      "learning_rate": 0.00014368963974255454,
      "loss": 2.6911,
      "step": 5250
    },
    {
      "epoch": 4.28,
      "learning_rate": 0.0001435226329257053,
      "loss": 2.7043,
      "step": 5300
    },
    {
      "epoch": 4.32,
      "learning_rate": 0.00014335354457740792,
      "loss": 2.6771,
      "step": 5350
    },
    {
      "epoch": 4.36,
      "learning_rate": 0.00014318237983410706,
      "loss": 2.6987,
      "step": 5400
    },
    {
      "epoch": 4.4,
      "learning_rate": 0.0001430091438953227,
      "loss": 2.6854,
      "step": 5450
    },
    {
      "epoch": 4.44,
      "learning_rate": 0.00014283384202349203,
      "loss": 2.6822,
      "step": 5500
    },
    {
      "epoch": 4.44,
      "eval_bleu": 1.572420203309693,
      "eval_loss": 3.2495412826538086,
      "eval_runtime": 1922.8389,
      "eval_samples_per_second": 4.581,
      "eval_steps_per_second": 0.144,
      "step": 5500
    },
    {
      "epoch": 4.48,
      "learning_rate": 0.00014265647954380976,
      "loss": 2.7064,
      "step": 5550
    },
    {
      "epoch": 4.52,
      "learning_rate": 0.00014247706184406618,
      "loss": 2.6905,
      "step": 5600
    },
    {
      "epoch": 4.56,
      "learning_rate": 0.00014229559437448362,
      "loss": 2.6847,
      "step": 5650
    },
    {
      "epoch": 4.6,
      "learning_rate": 0.00014211208264755092,
      "loss": 2.6825,
      "step": 5700
    },
    {
      "epoch": 4.64,
      "learning_rate": 0.00014192653223785577,
      "loss": 2.6759,
      "step": 5750
    },
    {
      "epoch": 4.68,
      "learning_rate": 0.0001417389487819156,
      "loss": 2.6888,
      "step": 5800
    },
    {
      "epoch": 4.72,
      "learning_rate": 0.00014154933797800621,
      "loss": 2.6722,
      "step": 5850
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.0001413577055859888,
      "loss": 2.688,
      "step": 5900
    },
    {
      "epoch": 4.81,
      "learning_rate": 0.00014116405742713484,
      "loss": 2.6934,
      "step": 5950
    },
    {
      "epoch": 4.85,
      "learning_rate": 0.00014096839938394936,
      "loss": 2.6577,
      "step": 6000
    },
    {
      "epoch": 4.85,
      "eval_bleu": 1.6301048703044616,
      "eval_loss": 3.176309823989868,
      "eval_runtime": 1919.9365,
      "eval_samples_per_second": 4.588,
      "eval_steps_per_second": 0.144,
      "step": 6000
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.00014077073739999222,
      "loss": 2.6749,
      "step": 6050
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.0001405710774796975,
      "loss": 2.6639,
      "step": 6100
    },
    {
      "epoch": 4.97,
      "learning_rate": 0.0001403694256881913,
      "loss": 2.6828,
      "step": 6150
    },
    {
      "epoch": 5.01,
      "learning_rate": 0.00014016578815110716,
      "loss": 2.645,
      "step": 6200
    },
    {
      "epoch": 5.05,
      "learning_rate": 0.00013996017105440036,
      "loss": 2.3837,
      "step": 6250
    },
    {
      "epoch": 5.09,
      "learning_rate": 0.00013975258064415972,
      "loss": 2.3875,
      "step": 6300
    },
    {
      "epoch": 5.13,
      "learning_rate": 0.00013954302322641797,
      "loss": 2.3975,
      "step": 6350
    },
    {
      "epoch": 5.17,
      "learning_rate": 0.00013933150516696024,
      "loss": 2.4332,
      "step": 6400
    },
    {
      "epoch": 5.21,
      "learning_rate": 0.00013911803289113055,
      "loss": 2.3976,
      "step": 6450
    },
    {
      "epoch": 5.25,
      "learning_rate": 0.00013890261288363676,
      "loss": 2.4118,
      "step": 6500
    },
    {
      "epoch": 5.25,
      "eval_bleu": 1.5739999205693251,
      "eval_loss": 3.2138876914978027,
      "eval_runtime": 1926.1281,
      "eval_samples_per_second": 4.573,
      "eval_steps_per_second": 0.143,
      "step": 6500
    },
    {
      "epoch": 5.29,
      "learning_rate": 0.00013868525168835353,
      "loss": 2.4436,
      "step": 6550
    },
    {
      "epoch": 5.33,
      "learning_rate": 0.0001384659559081235,
      "loss": 2.428,
      "step": 6600
    },
    {
      "epoch": 5.37,
      "learning_rate": 0.0001382447322045568,
      "loss": 2.4192,
      "step": 6650
    },
    {
      "epoch": 5.41,
      "learning_rate": 0.0001380215872978285,
      "loss": 2.4337,
      "step": 6700
    },
    {
      "epoch": 5.45,
      "learning_rate": 0.0001377965279664748,
      "loss": 2.4193,
      "step": 6750
    },
    {
      "epoch": 5.49,
      "learning_rate": 0.0001375695610471868,
      "loss": 2.438,
      "step": 6800
    },
    {
      "epoch": 5.53,
      "learning_rate": 0.00013734069343460293,
      "loss": 2.4385,
      "step": 6850
    },
    {
      "epoch": 5.57,
      "learning_rate": 0.0001371099320810995,
      "loss": 2.4556,
      "step": 6900
    },
    {
      "epoch": 5.61,
      "learning_rate": 0.0001368772839965797,
      "loss": 2.4518,
      "step": 6950
    },
    {
      "epoch": 5.65,
      "learning_rate": 0.00013664275624826025,
      "loss": 2.4345,
      "step": 7000
    },
    {
      "epoch": 5.65,
      "eval_bleu": 1.9002943282055358,
      "eval_loss": 3.1631877422332764,
      "eval_runtime": 1923.9555,
      "eval_samples_per_second": 4.578,
      "eval_steps_per_second": 0.143,
      "step": 7000
    },
    {
      "epoch": 5.69,
      "learning_rate": 0.00013640635596045707,
      "loss": 2.4438,
      "step": 7050
    },
    {
      "epoch": 5.73,
      "learning_rate": 0.00013616809031436876,
      "loss": 2.4239,
      "step": 7100
    },
    {
      "epoch": 5.77,
      "learning_rate": 0.0001359279665478584,
      "loss": 2.4293,
      "step": 7150
    },
    {
      "epoch": 5.82,
      "learning_rate": 0.0001356859919552337,
      "loss": 2.4569,
      "step": 7200
    },
    {
      "epoch": 5.86,
      "learning_rate": 0.0001354421738870255,
      "loss": 2.4437,
      "step": 7250
    },
    {
      "epoch": 5.9,
      "learning_rate": 0.00013519651974976433,
      "loss": 2.4373,
      "step": 7300
    },
    {
      "epoch": 5.94,
      "learning_rate": 0.00013494903700575562,
      "loss": 2.451,
      "step": 7350
    },
    {
      "epoch": 5.98,
      "learning_rate": 0.00013469973317285284,
      "loss": 2.4468,
      "step": 7400
    },
    {
      "epoch": 6.02,
      "learning_rate": 0.0001344486158242292,
      "loss": 2.333,
      "step": 7450
    },
    {
      "epoch": 6.06,
      "learning_rate": 0.00013419569258814757,
      "loss": 2.158,
      "step": 7500
    },
    {
      "epoch": 6.06,
      "eval_bleu": 1.8111395024627068,
      "eval_loss": 3.179304361343384,
      "eval_runtime": 1921.8485,
      "eval_samples_per_second": 4.583,
      "eval_steps_per_second": 0.144,
      "step": 7500
    },
    {
      "epoch": 6.1,
      "learning_rate": 0.00013394097114772887,
      "loss": 2.1613,
      "step": 7550
    },
    {
      "epoch": 6.14,
      "learning_rate": 0.00013368445924071844,
      "loss": 2.1807,
      "step": 7600
    },
    {
      "epoch": 6.18,
      "learning_rate": 0.00013342616465925126,
      "loss": 2.1873,
      "step": 7650
    },
    {
      "epoch": 6.22,
      "learning_rate": 0.00013316609524961502,
      "loss": 2.1955,
      "step": 7700
    },
    {
      "epoch": 6.26,
      "learning_rate": 0.00013290425891201196,
      "loss": 2.2115,
      "step": 7750
    },
    {
      "epoch": 6.3,
      "learning_rate": 0.00013264066360031872,
      "loss": 2.1964,
      "step": 7800
    },
    {
      "epoch": 6.34,
      "learning_rate": 0.0001323753173218448,
      "loss": 2.2045,
      "step": 7850
    },
    {
      "epoch": 6.38,
      "learning_rate": 0.00013210822813708936,
      "loss": 2.1907,
      "step": 7900
    },
    {
      "epoch": 6.42,
      "learning_rate": 0.0001318394041594963,
      "loss": 2.2066,
      "step": 7950
    },
    {
      "epoch": 6.46,
      "learning_rate": 0.00013156885355520778,
      "loss": 2.2051,
      "step": 8000
    },
    {
      "epoch": 6.46,
      "eval_bleu": 1.98361849433367,
      "eval_loss": 3.160503625869751,
      "eval_runtime": 1922.8791,
      "eval_samples_per_second": 4.581,
      "eval_steps_per_second": 0.144,
      "step": 8000
    }
  ],
  "max_steps": 30000,
  "num_train_epochs": 25,
  "total_flos": 4.9816432750591004e+20,
  "trial_name": null,
  "trial_params": null
}