{
  "best_metric": 1.716707873647693,
  "best_model_checkpoint": "experiments/translation/gpt2/2025-06-23_21-13-47/checkpoint-8000",
  "epoch": 7.269626639757821,
  "global_step": 9000,
  "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": 5.5374,
      "step": 50
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.799999999999998e-06,
      "loss": 4.9125,
      "step": 100
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.4799999999999999e-05,
      "loss": 4.7963,
      "step": 150
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.98e-05,
      "loss": 4.7123,
      "step": 200
    },
    {
      "epoch": 0.2,
      "learning_rate": 2.4799999999999996e-05,
      "loss": 4.6695,
      "step": 250
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.9799999999999996e-05,
      "loss": 4.6409,
      "step": 300
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.48e-05,
      "loss": 4.5901,
      "step": 350
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.979999999999999e-05,
      "loss": 4.5638,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.48e-05,
      "loss": 4.5254,
      "step": 450
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.98e-05,
      "loss": 4.4895,
      "step": 500
    },
    {
      "epoch": 0.4,
      "eval_bleu": 0.10366108848804567,
      "eval_loss": 4.428106784820557,
      "eval_runtime": 1956.2148,
      "eval_samples_per_second": 4.503,
      "eval_steps_per_second": 0.282,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 5.48e-05,
      "loss": 4.5073,
      "step": 550
    },
    {
      "epoch": 0.48,
      "learning_rate": 5.98e-05,
      "loss": 4.4717,
      "step": 600
    },
    {
      "epoch": 0.52,
      "learning_rate": 6.479999999999999e-05,
      "loss": 4.4849,
      "step": 650
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.979999999999999e-05,
      "loss": 4.4378,
      "step": 700
    },
    {
      "epoch": 0.61,
      "learning_rate": 7.479999999999999e-05,
      "loss": 4.4473,
      "step": 750
    },
    {
      "epoch": 0.65,
      "learning_rate": 7.98e-05,
      "loss": 4.4403,
      "step": 800
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.48e-05,
      "loss": 4.4189,
      "step": 850
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.98e-05,
      "loss": 4.3842,
      "step": 900
    },
    {
      "epoch": 0.77,
      "learning_rate": 9.479999999999999e-05,
      "loss": 4.3661,
      "step": 950
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.979999999999999e-05,
      "loss": 4.3978,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "eval_bleu": 0.12101650019262844,
      "eval_loss": 4.30416202545166,
      "eval_runtime": 1952.2431,
      "eval_samples_per_second": 4.512,
      "eval_steps_per_second": 0.282,
      "step": 1000
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00010479999999999999,
      "loss": 4.353,
      "step": 1050
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00010979999999999999,
      "loss": 4.3321,
      "step": 1100
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00011479999999999999,
      "loss": 4.3258,
      "step": 1150
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00011979999999999998,
      "loss": 4.3221,
      "step": 1200
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00012479999999999997,
      "loss": 4.3282,
      "step": 1250
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00012979999999999998,
      "loss": 4.1992,
      "step": 1300
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00013479999999999997,
      "loss": 4.1949,
      "step": 1350
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00013979999999999998,
      "loss": 4.1812,
      "step": 1400
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001448,
      "loss": 4.1842,
      "step": 1450
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00014979999999999998,
      "loss": 4.1744,
      "step": 1500
    },
    {
      "epoch": 1.21,
      "eval_bleu": 0.19483428061504493,
      "eval_loss": 4.20745849609375,
      "eval_runtime": 1957.9666,
      "eval_samples_per_second": 4.499,
      "eval_steps_per_second": 0.281,
      "step": 1500
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00014999895016087306,
      "loss": 4.1746,
      "step": 1550
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00014999562388001478,
      "loss": 4.134,
      "step": 1600
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00014999001943582685,
      "loss": 4.1534,
      "step": 1650
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001499821369985571,
      "loss": 4.1424,
      "step": 1700
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001499719768076526,
      "loss": 4.1222,
      "step": 1750
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014995953917175227,
      "loss": 4.1216,
      "step": 1800
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00014994482446867774,
      "loss": 4.0985,
      "step": 1850
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00014992783314542174,
      "loss": 4.0785,
      "step": 1900
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00014990856571813448,
      "loss": 4.1087,
      "step": 1950
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001498870227721081,
      "loss": 4.0548,
      "step": 2000
    },
    {
      "epoch": 1.62,
      "eval_bleu": 0.3099045639043874,
      "eval_loss": 4.0605082511901855,
      "eval_runtime": 1963.1271,
      "eval_samples_per_second": 4.487,
      "eval_steps_per_second": 0.281,
      "step": 2000
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0001498632049617587,
      "loss": 4.0445,
      "step": 2050
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00014983711301060673,
      "loss": 4.0045,
      "step": 2100
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00014980874771125478,
      "loss": 4.0004,
      "step": 2150
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00014977810992536359,
      "loss": 4.0131,
      "step": 2200
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00014974520058362584,
      "loss": 4.0018,
      "step": 2250
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00014971002068573793,
      "loss": 3.9797,
      "step": 2300
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00014967257130036961,
      "loss": 3.9479,
      "step": 2350
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00014963285356513152,
      "loss": 3.9432,
      "step": 2400
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001495908686865405,
      "loss": 3.9119,
      "step": 2450
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.00014954661793998317,
      "loss": 3.8522,
      "step": 2500
    },
    {
      "epoch": 2.02,
      "eval_bleu": 0.41817761915800616,
      "eval_loss": 3.9211950302124023,
      "eval_runtime": 1966.167,
      "eval_samples_per_second": 4.48,
      "eval_steps_per_second": 0.28,
      "step": 2500
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0001495001026696769,
      "loss": 3.6675,
      "step": 2550
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.00014945132428862936,
      "loss": 3.6431,
      "step": 2600
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.00014940028427859524,
      "loss": 3.6479,
      "step": 2650
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.00014934698419003133,
      "loss": 3.6214,
      "step": 2700
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0001492914256420496,
      "loss": 3.6333,
      "step": 2750
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.00014923361032236776,
      "loss": 3.6243,
      "step": 2800
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.00014917353998725823,
      "loss": 3.6292,
      "step": 2850
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.00014911121646149456,
      "loss": 3.5965,
      "step": 2900
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.00014904664163829616,
      "loss": 3.5995,
      "step": 2950
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.00014897981747927076,
      "loss": 3.5954,
      "step": 3000
    },
    {
      "epoch": 2.42,
      "eval_bleu": 0.6197998214597401,
      "eval_loss": 3.80145525932312,
      "eval_runtime": 1964.8057,
      "eval_samples_per_second": 4.483,
      "eval_steps_per_second": 0.28,
      "step": 3000
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.00014891074601435482,
      "loss": 3.5926,
      "step": 3050
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.00014883942934175177,
      "loss": 3.5782,
      "step": 3100
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0001487658696278684,
      "loss": 3.5606,
      "step": 3150
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.000148690069107249,
      "loss": 3.5623,
      "step": 3200
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.00014861203008250745,
      "loss": 3.5484,
      "step": 3250
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001485317549242574,
      "loss": 3.574,
      "step": 3300
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.00014844924607104,
      "loss": 3.5422,
      "step": 3350
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.00014836450602925014,
      "loss": 3.5185,
      "step": 3400
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.00014827753737306008,
      "loss": 3.5329,
      "step": 3450
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.00014818834274434134,
      "loss": 3.5305,
      "step": 3500
    },
    {
      "epoch": 2.83,
      "eval_bleu": 0.9915510687453413,
      "eval_loss": 3.6937255859375,
      "eval_runtime": 1965.6982,
      "eval_samples_per_second": 4.481,
      "eval_steps_per_second": 0.28,
      "step": 3500
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.00014809692485258445,
      "loss": 3.4934,
      "step": 3550
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.00014800328647481662,
      "loss": 3.5064,
      "step": 3600
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.00014790743045551744,
      "loss": 3.475,
      "step": 3650
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.00014780935970653235,
      "loss": 3.4644,
      "step": 3700
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.00014770907720698426,
      "loss": 3.3422,
      "step": 3750
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.00014760658600318318,
      "loss": 3.2076,
      "step": 3800
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.00014750188920853338,
      "loss": 3.2166,
      "step": 3850
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.00014739499000343914,
      "loss": 3.214,
      "step": 3900
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0001472858916352079,
      "loss": 3.2104,
      "step": 3950
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0001471745974179517,
      "loss": 3.175,
      "step": 4000
    },
    {
      "epoch": 3.23,
      "eval_bleu": 0.8229569245913304,
      "eval_loss": 3.6438491344451904,
      "eval_runtime": 1966.943,
      "eval_samples_per_second": 4.478,
      "eval_steps_per_second": 0.28,
      "step": 4000
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.00014706111073248656,
      "loss": 3.2015,
      "step": 4050
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0001469454350262297,
      "loss": 3.2109,
      "step": 4100
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0001468275738130948,
      "loss": 3.2042,
      "step": 4150
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0001467075306733854,
      "loss": 3.1996,
      "step": 4200
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.000146585309253686,
      "loss": 3.206,
      "step": 4250
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.00014646091326675126,
      "loss": 3.1749,
      "step": 4300
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.00014633434649139344,
      "loss": 3.2049,
      "step": 4350
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.00014620561277236722,
      "loss": 3.1852,
      "step": 4400
    },
    {
      "epoch": 3.59,
      "learning_rate": 0.0001460747160202534,
      "loss": 3.1768,
      "step": 4450
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.0001459416602113397,
      "loss": 3.1863,
      "step": 4500
    },
    {
      "epoch": 3.63,
      "eval_bleu": 1.0657169333169068,
      "eval_loss": 3.550638437271118,
      "eval_runtime": 1967.2607,
      "eval_samples_per_second": 4.477,
      "eval_steps_per_second": 0.28,
      "step": 4500
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.00014580644938750012,
      "loss": 3.1681,
      "step": 4550
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.00014566908765607222,
      "loss": 3.1675,
      "step": 4600
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.00014552957918973226,
      "loss": 3.1601,
      "step": 4650
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.00014538792822636849,
      "loss": 3.154,
      "step": 4700
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.00014524413906895234,
      "loss": 3.1548,
      "step": 4750
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.00014509821608540784,
      "loss": 3.14,
      "step": 4800
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.00014495016370847882,
      "loss": 3.1584,
      "step": 4850
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.00014479998643559435,
      "loss": 3.1351,
      "step": 4900
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00014464768882873198,
      "loss": 3.1546,
      "step": 4950
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.00014449327551427935,
      "loss": 2.9153,
      "step": 5000
    },
    {
      "epoch": 4.04,
      "eval_bleu": 1.0047431556971733,
      "eval_loss": 3.5359809398651123,
      "eval_runtime": 1966.6362,
      "eval_samples_per_second": 4.479,
      "eval_steps_per_second": 0.28,
      "step": 5000
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.0001443367511828934,
      "loss": 2.8766,
      "step": 5050
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.0001441781205893582,
      "loss": 2.8748,
      "step": 5100
    },
    {
      "epoch": 4.16,
      "learning_rate": 0.00014401738855244028,
      "loss": 2.8439,
      "step": 5150
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.00014385455995474222,
      "loss": 2.8731,
      "step": 5200
    },
    {
      "epoch": 4.24,
      "learning_rate": 0.00014368963974255454,
      "loss": 2.8727,
      "step": 5250
    },
    {
      "epoch": 4.28,
      "learning_rate": 0.0001435226329257053,
      "loss": 2.8911,
      "step": 5300
    },
    {
      "epoch": 4.32,
      "learning_rate": 0.00014335354457740792,
      "loss": 2.8639,
      "step": 5350
    },
    {
      "epoch": 4.36,
      "learning_rate": 0.00014318237983410706,
      "loss": 2.8785,
      "step": 5400
    },
    {
      "epoch": 4.4,
      "learning_rate": 0.0001430091438953227,
      "loss": 2.8755,
      "step": 5450
    },
    {
      "epoch": 4.44,
      "learning_rate": 0.00014283384202349203,
      "loss": 2.8661,
      "step": 5500
    },
    {
      "epoch": 4.44,
      "eval_bleu": 1.207406468713491,
      "eval_loss": 3.5010602474212646,
      "eval_runtime": 1964.1467,
      "eval_samples_per_second": 4.484,
      "eval_steps_per_second": 0.281,
      "step": 5500
    },
    {
      "epoch": 4.48,
      "learning_rate": 0.00014265647954380976,
      "loss": 2.894,
      "step": 5550
    },
    {
      "epoch": 4.52,
      "learning_rate": 0.00014247706184406618,
      "loss": 2.8723,
      "step": 5600
    },
    {
      "epoch": 4.56,
      "learning_rate": 0.00014229559437448362,
      "loss": 2.868,
      "step": 5650
    },
    {
      "epoch": 4.6,
      "learning_rate": 0.00014211208264755092,
      "loss": 2.8675,
      "step": 5700
    },
    {
      "epoch": 4.64,
      "learning_rate": 0.00014192653223785577,
      "loss": 2.8653,
      "step": 5750
    },
    {
      "epoch": 4.68,
      "learning_rate": 0.0001417389487819156,
      "loss": 2.8709,
      "step": 5800
    },
    {
      "epoch": 4.72,
      "learning_rate": 0.00014154933797800621,
      "loss": 2.8535,
      "step": 5850
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.0001413577055859888,
      "loss": 2.8765,
      "step": 5900
    },
    {
      "epoch": 4.81,
      "learning_rate": 0.00014116405742713484,
      "loss": 2.8777,
      "step": 5950
    },
    {
      "epoch": 4.85,
      "learning_rate": 0.00014096839938394936,
      "loss": 2.8441,
      "step": 6000
    },
    {
      "epoch": 4.85,
      "eval_bleu": 1.2898999365765402,
      "eval_loss": 3.4188201427459717,
      "eval_runtime": 1959.6425,
      "eval_samples_per_second": 4.495,
      "eval_steps_per_second": 0.281,
      "step": 6000
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.00014077073739999222,
      "loss": 2.8655,
      "step": 6050
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.0001405710774796975,
      "loss": 2.8427,
      "step": 6100
    },
    {
      "epoch": 4.97,
      "learning_rate": 0.0001403694256881913,
      "loss": 2.8573,
      "step": 6150
    },
    {
      "epoch": 5.01,
      "learning_rate": 0.00014016578815110716,
      "loss": 2.8372,
      "step": 6200
    },
    {
      "epoch": 5.05,
      "learning_rate": 0.00013996017105440036,
      "loss": 2.5577,
      "step": 6250
    },
    {
      "epoch": 5.09,
      "learning_rate": 0.00013975258064415972,
      "loss": 2.5632,
      "step": 6300
    },
    {
      "epoch": 5.13,
      "learning_rate": 0.00013954302322641797,
      "loss": 2.5707,
      "step": 6350
    },
    {
      "epoch": 5.17,
      "learning_rate": 0.00013933150516696024,
      "loss": 2.5991,
      "step": 6400
    },
    {
      "epoch": 5.21,
      "learning_rate": 0.00013911803289113055,
      "loss": 2.5697,
      "step": 6450
    },
    {
      "epoch": 5.25,
      "learning_rate": 0.00013890261288363676,
      "loss": 2.5857,
      "step": 6500
    },
    {
      "epoch": 5.25,
      "eval_bleu": 1.3087062882464005,
      "eval_loss": 3.4442849159240723,
      "eval_runtime": 1957.1145,
      "eval_samples_per_second": 4.501,
      "eval_steps_per_second": 0.282,
      "step": 6500
    },
    {
      "epoch": 5.29,
      "learning_rate": 0.00013868525168835353,
      "loss": 2.614,
      "step": 6550
    },
    {
      "epoch": 5.33,
      "learning_rate": 0.0001384659559081235,
      "loss": 2.6133,
      "step": 6600
    },
    {
      "epoch": 5.37,
      "learning_rate": 0.0001382447322045568,
      "loss": 2.5898,
      "step": 6650
    },
    {
      "epoch": 5.41,
      "learning_rate": 0.0001380215872978285,
      "loss": 2.5992,
      "step": 6700
    },
    {
      "epoch": 5.45,
      "learning_rate": 0.0001377965279664748,
      "loss": 2.5873,
      "step": 6750
    },
    {
      "epoch": 5.49,
      "learning_rate": 0.0001375695610471868,
      "loss": 2.6155,
      "step": 6800
    },
    {
      "epoch": 5.53,
      "learning_rate": 0.00013734069343460293,
      "loss": 2.6044,
      "step": 6850
    },
    {
      "epoch": 5.57,
      "learning_rate": 0.0001371099320810995,
      "loss": 2.6239,
      "step": 6900
    },
    {
      "epoch": 5.61,
      "learning_rate": 0.0001368772839965797,
      "loss": 2.6267,
      "step": 6950
    },
    {
      "epoch": 5.65,
      "learning_rate": 0.00013664275624826025,
      "loss": 2.6094,
      "step": 7000
    },
    {
      "epoch": 5.65,
      "eval_bleu": 1.423644909665514,
      "eval_loss": 3.3920862674713135,
      "eval_runtime": 1956.3214,
      "eval_samples_per_second": 4.502,
      "eval_steps_per_second": 0.282,
      "step": 7000
    },
    {
      "epoch": 5.69,
      "learning_rate": 0.00013640635596045707,
      "loss": 2.6098,
      "step": 7050
    },
    {
      "epoch": 5.73,
      "learning_rate": 0.00013616809031436876,
      "loss": 2.5955,
      "step": 7100
    },
    {
      "epoch": 5.77,
      "learning_rate": 0.0001359279665478584,
      "loss": 2.5979,
      "step": 7150
    },
    {
      "epoch": 5.82,
      "learning_rate": 0.0001356859919552337,
      "loss": 2.6305,
      "step": 7200
    },
    {
      "epoch": 5.86,
      "learning_rate": 0.0001354421738870255,
      "loss": 2.6193,
      "step": 7250
    },
    {
      "epoch": 5.9,
      "learning_rate": 0.00013519651974976433,
      "loss": 2.6063,
      "step": 7300
    },
    {
      "epoch": 5.94,
      "learning_rate": 0.00013494903700575562,
      "loss": 2.6187,
      "step": 7350
    },
    {
      "epoch": 5.98,
      "learning_rate": 0.00013469973317285284,
      "loss": 2.6155,
      "step": 7400
    },
    {
      "epoch": 6.02,
      "learning_rate": 0.0001344486158242292,
      "loss": 2.5029,
      "step": 7450
    },
    {
      "epoch": 6.06,
      "learning_rate": 0.00013419569258814757,
      "loss": 2.3144,
      "step": 7500
    },
    {
      "epoch": 6.06,
      "eval_bleu": 1.4240728657820112,
      "eval_loss": 3.422335624694824,
      "eval_runtime": 1958.4348,
      "eval_samples_per_second": 4.497,
      "eval_steps_per_second": 0.281,
      "step": 7500
    },
    {
      "epoch": 6.1,
      "learning_rate": 0.00013394097114772887,
      "loss": 2.3243,
      "step": 7550
    },
    {
      "epoch": 6.14,
      "learning_rate": 0.00013368445924071844,
      "loss": 2.3288,
      "step": 7600
    },
    {
      "epoch": 6.18,
      "learning_rate": 0.00013342616465925126,
      "loss": 2.3569,
      "step": 7650
    },
    {
      "epoch": 6.22,
      "learning_rate": 0.00013316609524961502,
      "loss": 2.3593,
      "step": 7700
    },
    {
      "epoch": 6.26,
      "learning_rate": 0.00013290425891201196,
      "loss": 2.3579,
      "step": 7750
    },
    {
      "epoch": 6.3,
      "learning_rate": 0.00013264066360031872,
      "loss": 2.3623,
      "step": 7800
    },
    {
      "epoch": 6.34,
      "learning_rate": 0.0001323753173218448,
      "loss": 2.3642,
      "step": 7850
    },
    {
      "epoch": 6.38,
      "learning_rate": 0.00013210822813708936,
      "loss": 2.3494,
      "step": 7900
    },
    {
      "epoch": 6.42,
      "learning_rate": 0.0001318394041594963,
      "loss": 2.3568,
      "step": 7950
    },
    {
      "epoch": 6.46,
      "learning_rate": 0.00013156885355520778,
      "loss": 2.3675,
      "step": 8000
    },
    {
      "epoch": 6.46,
      "eval_bleu": 1.716707873647693,
      "eval_loss": 3.403074026107788,
      "eval_runtime": 1958.6783,
      "eval_samples_per_second": 4.497,
      "eval_steps_per_second": 0.281,
      "step": 8000
    },
    {
      "epoch": 6.5,
      "learning_rate": 0.0001312965845428162,
      "loss": 2.3814,
      "step": 8050
    },
    {
      "epoch": 6.54,
      "learning_rate": 0.0001310226053931146,
      "loss": 2.3737,
      "step": 8100
    },
    {
      "epoch": 6.58,
      "learning_rate": 0.00013074692442884524,
      "loss": 2.3713,
      "step": 8150
    },
    {
      "epoch": 6.62,
      "learning_rate": 0.00013046955002444697,
      "loss": 2.3989,
      "step": 8200
    },
    {
      "epoch": 6.66,
      "learning_rate": 0.00013019049060580067,
      "loss": 2.377,
      "step": 8250
    },
    {
      "epoch": 6.7,
      "learning_rate": 0.00012991538574407902,
      "loss": 2.3949,
      "step": 8300
    },
    {
      "epoch": 6.74,
      "learning_rate": 0.0001296330150553478,
      "loss": 2.3995,
      "step": 8350
    },
    {
      "epoch": 6.78,
      "learning_rate": 0.00012934898476402848,
      "loss": 2.3936,
      "step": 8400
    },
    {
      "epoch": 6.83,
      "learning_rate": 0.0001290633034981898,
      "loss": 2.3825,
      "step": 8450
    },
    {
      "epoch": 6.87,
      "learning_rate": 0.00012877597993605252,
      "loss": 2.3931,
      "step": 8500
    },
    {
      "epoch": 6.87,
      "eval_bleu": 1.5281360670298858,
      "eval_loss": 3.3454949855804443,
      "eval_runtime": 1957.7441,
      "eval_samples_per_second": 4.499,
      "eval_steps_per_second": 0.281,
      "step": 8500
    },
    {
      "epoch": 6.91,
      "learning_rate": 0.00012848702280572602,
      "loss": 2.3948,
      "step": 8550
    },
    {
      "epoch": 6.95,
      "learning_rate": 0.0001281964408849429,
      "loss": 2.4046,
      "step": 8600
    },
    {
      "epoch": 6.99,
      "learning_rate": 0.00012790424300079258,
      "loss": 2.4168,
      "step": 8650
    },
    {
      "epoch": 7.03,
      "learning_rate": 0.00012761043802945305,
      "loss": 2.238,
      "step": 8700
    },
    {
      "epoch": 7.07,
      "learning_rate": 0.00012731503489592122,
      "loss": 2.1129,
      "step": 8750
    },
    {
      "epoch": 7.11,
      "learning_rate": 0.0001270180425737418,
      "loss": 2.1072,
      "step": 8800
    },
    {
      "epoch": 7.15,
      "learning_rate": 0.00012671947008473483,
      "loss": 2.1193,
      "step": 8850
    },
    {
      "epoch": 7.19,
      "learning_rate": 0.00012641932649872138,
      "loss": 2.1408,
      "step": 8900
    },
    {
      "epoch": 7.23,
      "learning_rate": 0.00012611762093324829,
      "loss": 2.139,
      "step": 8950
    },
    {
      "epoch": 7.27,
      "learning_rate": 0.00012581436255331107,
      "loss": 2.1578,
      "step": 9000
    },
    {
      "epoch": 7.27,
      "eval_bleu": 1.6379507937528601,
      "eval_loss": 3.419518232345581,
      "eval_runtime": 1957.3561,
      "eval_samples_per_second": 4.5,
      "eval_steps_per_second": 0.282,
      "step": 9000
    }
  ],
  "max_steps": 30000,
  "num_train_epochs": 25,
  "total_flos": 5.60444958175701e+20,
  "trial_name": null,
  "trial_params": null
}