{
  "best_metric": 0.6763644218444824,
  "best_model_checkpoint": "/d/hpc/projects/FRI/bb6846/run11//mt5-base/checkpoint-25861",
  "epoch": 28.0,
  "global_step": 51723,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27,
      "learning_rate": 0.0004954881790290561,
      "loss": 2.2444,
      "step": 500
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004909763580581122,
      "loss": 1.6704,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00048646453708716834,
      "loss": 1.5382,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_LaBSE similarity": 0.6048825370314684,
      "eval_bleu": 0.18207720933969782,
      "eval_loss": 1.2285538911819458,
      "eval_runtime": 727.9323,
      "eval_samples_per_second": 19.2,
      "eval_steps_per_second": 1.201,
      "step": 1847
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00048195271611622454,
      "loss": 1.4208,
      "step": 2000
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00047744089514528064,
      "loss": 1.307,
      "step": 2500
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0004729290741743368,
      "loss": 1.2675,
      "step": 3000
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0004684172532033929,
      "loss": 1.2288,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_LaBSE similarity": 0.6103176045423836,
      "eval_bleu": 0.18286165047145764,
      "eval_loss": 1.0573153495788574,
      "eval_runtime": 727.6037,
      "eval_samples_per_second": 19.208,
      "eval_steps_per_second": 1.201,
      "step": 3694
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.000463905432232449,
      "loss": 1.1399,
      "step": 4000
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0004593936112615051,
      "loss": 1.0766,
      "step": 4500
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.00045488179029056126,
      "loss": 1.0643,
      "step": 5000
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0004503699693196174,
      "loss": 1.0478,
      "step": 5500
    },
    {
      "epoch": 3.0,
      "eval_LaBSE similarity": 0.6151767824817785,
      "eval_bleu": 0.18683028946919106,
      "eval_loss": 0.9635915160179138,
      "eval_runtime": 727.6118,
      "eval_samples_per_second": 19.208,
      "eval_steps_per_second": 1.201,
      "step": 5541
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.00044585814834867355,
      "loss": 0.9375,
      "step": 6000
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.00044134632737772965,
      "loss": 0.918,
      "step": 6500
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.0004368345064067858,
      "loss": 0.9207,
      "step": 7000
    },
    {
      "epoch": 4.0,
      "eval_LaBSE similarity": 0.614242324825887,
      "eval_bleu": 0.18815387525896146,
      "eval_loss": 0.8977922797203064,
      "eval_runtime": 728.2433,
      "eval_samples_per_second": 19.191,
      "eval_steps_per_second": 1.2,
      "step": 7389
    },
    {
      "epoch": 4.06,
      "learning_rate": 0.0004323226854358419,
      "loss": 0.8848,
      "step": 7500
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.00042781086446489803,
      "loss": 0.7985,
      "step": 8000
    },
    {
      "epoch": 4.6,
      "learning_rate": 0.0004232990434939541,
      "loss": 0.8032,
      "step": 8500
    },
    {
      "epoch": 4.87,
      "learning_rate": 0.00041878722252301027,
      "loss": 0.8037,
      "step": 9000
    },
    {
      "epoch": 5.0,
      "eval_LaBSE similarity": 0.622497909277748,
      "eval_bleu": 0.19241623278663542,
      "eval_loss": 0.8346570134162903,
      "eval_runtime": 727.6058,
      "eval_samples_per_second": 19.208,
      "eval_steps_per_second": 1.201,
      "step": 9236
    },
    {
      "epoch": 5.14,
      "learning_rate": 0.0004142754015520664,
      "loss": 0.7435,
      "step": 9500
    },
    {
      "epoch": 5.41,
      "learning_rate": 0.00040976358058112257,
      "loss": 0.7011,
      "step": 10000
    },
    {
      "epoch": 5.68,
      "learning_rate": 0.00040525175961017866,
      "loss": 0.7054,
      "step": 10500
    },
    {
      "epoch": 5.95,
      "learning_rate": 0.0004007399386392348,
      "loss": 0.705,
      "step": 11000
    },
    {
      "epoch": 6.0,
      "eval_LaBSE similarity": 0.6255796529372922,
      "eval_bleu": 0.19577697504607858,
      "eval_loss": 0.790106475353241,
      "eval_runtime": 729.5559,
      "eval_samples_per_second": 19.157,
      "eval_steps_per_second": 1.198,
      "step": 11083
    },
    {
      "epoch": 6.23,
      "learning_rate": 0.0003962281176682909,
      "loss": 0.6238,
      "step": 11500
    },
    {
      "epoch": 6.5,
      "learning_rate": 0.00039171629669734704,
      "loss": 0.6194,
      "step": 12000
    },
    {
      "epoch": 6.77,
      "learning_rate": 0.00038720447572640314,
      "loss": 0.6263,
      "step": 12500
    },
    {
      "epoch": 7.0,
      "eval_LaBSE similarity": 0.6281634647360271,
      "eval_bleu": 0.1984314625604209,
      "eval_loss": 0.7558700442314148,
      "eval_runtime": 728.5022,
      "eval_samples_per_second": 19.185,
      "eval_steps_per_second": 1.2,
      "step": 12930
    },
    {
      "epoch": 7.04,
      "learning_rate": 0.00038269265475545934,
      "loss": 0.611,
      "step": 13000
    },
    {
      "epoch": 7.31,
      "learning_rate": 0.00037818083378451543,
      "loss": 0.5407,
      "step": 13500
    },
    {
      "epoch": 7.58,
      "learning_rate": 0.0003736690128135716,
      "loss": 0.5506,
      "step": 14000
    },
    {
      "epoch": 7.85,
      "learning_rate": 0.00036915719184262767,
      "loss": 0.5596,
      "step": 14500
    },
    {
      "epoch": 8.0,
      "eval_LaBSE similarity": 0.6297791211180278,
      "eval_bleu": 0.2016469323333738,
      "eval_loss": 0.7360725998878479,
      "eval_runtime": 730.4745,
      "eval_samples_per_second": 19.133,
      "eval_steps_per_second": 1.196,
      "step": 14778
    },
    {
      "epoch": 8.12,
      "learning_rate": 0.0003646453708716838,
      "loss": 0.5205,
      "step": 15000
    },
    {
      "epoch": 8.39,
      "learning_rate": 0.0003601335499007399,
      "loss": 0.4852,
      "step": 15500
    },
    {
      "epoch": 8.66,
      "learning_rate": 0.00035562172892979606,
      "loss": 0.4916,
      "step": 16000
    },
    {
      "epoch": 8.93,
      "learning_rate": 0.0003511099079588522,
      "loss": 0.4974,
      "step": 16500
    },
    {
      "epoch": 9.0,
      "eval_LaBSE similarity": 0.6329554534294927,
      "eval_bleu": 0.20453006594808004,
      "eval_loss": 0.7115353345870972,
      "eval_runtime": 731.2775,
      "eval_samples_per_second": 19.112,
      "eval_steps_per_second": 1.195,
      "step": 16625
    },
    {
      "epoch": 9.2,
      "learning_rate": 0.00034659808698790835,
      "loss": 0.4433,
      "step": 17000
    },
    {
      "epoch": 9.47,
      "learning_rate": 0.00034208626601696444,
      "loss": 0.4341,
      "step": 17500
    },
    {
      "epoch": 9.74,
      "learning_rate": 0.0003375744450460206,
      "loss": 0.447,
      "step": 18000
    },
    {
      "epoch": 10.0,
      "eval_LaBSE similarity": 0.635916732698647,
      "eval_bleu": 0.20824504916149092,
      "eval_loss": 0.6980001926422119,
      "eval_runtime": 730.5297,
      "eval_samples_per_second": 19.131,
      "eval_steps_per_second": 1.196,
      "step": 18472
    },
    {
      "epoch": 10.01,
      "learning_rate": 0.0003330626240750767,
      "loss": 0.4488,
      "step": 18500
    },
    {
      "epoch": 10.29,
      "learning_rate": 0.00032855080310413283,
      "loss": 0.3861,
      "step": 19000
    },
    {
      "epoch": 10.56,
      "learning_rate": 0.0003240389821331889,
      "loss": 0.398,
      "step": 19500
    },
    {
      "epoch": 10.83,
      "learning_rate": 0.00031952716116224507,
      "loss": 0.4006,
      "step": 20000
    },
    {
      "epoch": 11.0,
      "eval_LaBSE similarity": 0.6365891740497193,
      "eval_bleu": 0.20953963135229853,
      "eval_loss": 0.690319836139679,
      "eval_runtime": 730.2358,
      "eval_samples_per_second": 19.139,
      "eval_steps_per_second": 1.197,
      "step": 20319
    },
    {
      "epoch": 11.1,
      "learning_rate": 0.0003150153401913012,
      "loss": 0.3803,
      "step": 20500
    },
    {
      "epoch": 11.37,
      "learning_rate": 0.00031050351922035736,
      "loss": 0.3551,
      "step": 21000
    },
    {
      "epoch": 11.64,
      "learning_rate": 0.00030599169824941345,
      "loss": 0.361,
      "step": 21500
    },
    {
      "epoch": 11.91,
      "learning_rate": 0.0003014798772784696,
      "loss": 0.3662,
      "step": 22000
    },
    {
      "epoch": 12.0,
      "eval_LaBSE similarity": 0.6368156212159983,
      "eval_bleu": 0.21027391744413493,
      "eval_loss": 0.689200222492218,
      "eval_runtime": 729.7202,
      "eval_samples_per_second": 19.153,
      "eval_steps_per_second": 1.198,
      "step": 22167
    },
    {
      "epoch": 12.18,
      "learning_rate": 0.0002969680563075257,
      "loss": 0.3313,
      "step": 22500
    },
    {
      "epoch": 12.45,
      "learning_rate": 0.00029245623533658184,
      "loss": 0.3244,
      "step": 23000
    },
    {
      "epoch": 12.72,
      "learning_rate": 0.00028794441436563793,
      "loss": 0.3321,
      "step": 23500
    },
    {
      "epoch": 12.99,
      "learning_rate": 0.00028343259339469413,
      "loss": 0.3365,
      "step": 24000
    },
    {
      "epoch": 13.0,
      "eval_LaBSE similarity": 0.64150025325299,
      "eval_bleu": 0.21448433499231642,
      "eval_loss": 0.6796755790710449,
      "eval_runtime": 731.0965,
      "eval_samples_per_second": 19.116,
      "eval_steps_per_second": 1.195,
      "step": 24014
    },
    {
      "epoch": 13.26,
      "learning_rate": 0.0002789207724237502,
      "loss": 0.2878,
      "step": 24500
    },
    {
      "epoch": 13.53,
      "learning_rate": 0.00027440895145280637,
      "loss": 0.2969,
      "step": 25000
    },
    {
      "epoch": 13.8,
      "learning_rate": 0.00026989713048186246,
      "loss": 0.3052,
      "step": 25500
    },
    {
      "epoch": 14.0,
      "eval_LaBSE similarity": 0.6435598843860529,
      "eval_bleu": 0.217333486806597,
      "eval_loss": 0.6763644218444824,
      "eval_runtime": 731.6928,
      "eval_samples_per_second": 19.101,
      "eval_steps_per_second": 1.194,
      "step": 25861
    },
    {
      "epoch": 14.07,
      "learning_rate": 0.0002653853095109186,
      "loss": 0.2962,
      "step": 26000
    },
    {
      "epoch": 14.35,
      "learning_rate": 0.0002608734885399747,
      "loss": 0.2648,
      "step": 26500
    },
    {
      "epoch": 14.62,
      "learning_rate": 0.00025636166756903085,
      "loss": 0.2746,
      "step": 27000
    },
    {
      "epoch": 14.89,
      "learning_rate": 0.000251849846598087,
      "loss": 0.2811,
      "step": 27500
    },
    {
      "epoch": 15.0,
      "eval_LaBSE similarity": 0.6415143596865642,
      "eval_bleu": 0.2176019486610223,
      "eval_loss": 0.6784259676933289,
      "eval_runtime": 731.0412,
      "eval_samples_per_second": 19.118,
      "eval_steps_per_second": 1.196,
      "step": 27708
    },
    {
      "epoch": 15.16,
      "learning_rate": 0.0002473380256271431,
      "loss": 0.2585,
      "step": 28000
    },
    {
      "epoch": 15.43,
      "learning_rate": 0.00024282620465619926,
      "loss": 0.2478,
      "step": 28500
    },
    {
      "epoch": 15.7,
      "learning_rate": 0.00023831438368525538,
      "loss": 0.2554,
      "step": 29000
    },
    {
      "epoch": 15.97,
      "learning_rate": 0.0002338025627143115,
      "loss": 0.2601,
      "step": 29500
    },
    {
      "epoch": 16.0,
      "eval_LaBSE similarity": 0.6428080388896847,
      "eval_bleu": 0.2193918672969541,
      "eval_loss": 0.681696355342865,
      "eval_runtime": 731.5003,
      "eval_samples_per_second": 19.106,
      "eval_steps_per_second": 1.195,
      "step": 29556
    },
    {
      "epoch": 16.24,
      "learning_rate": 0.00022929074174336762,
      "loss": 0.227,
      "step": 30000
    },
    {
      "epoch": 16.51,
      "learning_rate": 0.00022477892077242377,
      "loss": 0.2329,
      "step": 30500
    },
    {
      "epoch": 16.78,
      "learning_rate": 0.0002202670998014799,
      "loss": 0.2379,
      "step": 31000
    },
    {
      "epoch": 17.0,
      "eval_LaBSE similarity": 0.643077788578904,
      "eval_bleu": 0.21980244649693723,
      "eval_loss": 0.6875574588775635,
      "eval_runtime": 729.8897,
      "eval_samples_per_second": 19.148,
      "eval_steps_per_second": 1.197,
      "step": 31403
    },
    {
      "epoch": 17.05,
      "learning_rate": 0.000215755278830536,
      "loss": 0.2353,
      "step": 31500
    },
    {
      "epoch": 17.32,
      "learning_rate": 0.00021124345785959216,
      "loss": 0.2097,
      "step": 32000
    },
    {
      "epoch": 17.59,
      "learning_rate": 0.00020673163688864828,
      "loss": 0.2181,
      "step": 32500
    },
    {
      "epoch": 17.86,
      "learning_rate": 0.0002022198159177044,
      "loss": 0.2219,
      "step": 33000
    },
    {
      "epoch": 18.0,
      "eval_LaBSE similarity": 0.6413028645145259,
      "eval_bleu": 0.2195570154576433,
      "eval_loss": 0.6933163404464722,
      "eval_runtime": 730.845,
      "eval_samples_per_second": 19.123,
      "eval_steps_per_second": 1.196,
      "step": 33250
    },
    {
      "epoch": 18.14,
      "learning_rate": 0.00019770799494676051,
      "loss": 0.2073,
      "step": 33500
    },
    {
      "epoch": 18.41,
      "learning_rate": 0.00019319617397581666,
      "loss": 0.1972,
      "step": 34000
    },
    {
      "epoch": 18.68,
      "learning_rate": 0.00018868435300487278,
      "loss": 0.2055,
      "step": 34500
    },
    {
      "epoch": 18.95,
      "learning_rate": 0.0001841725320339289,
      "loss": 0.2078,
      "step": 35000
    },
    {
      "epoch": 19.0,
      "eval_LaBSE similarity": 0.6431215279734293,
      "eval_bleu": 0.22102576165957888,
      "eval_loss": 0.6970334053039551,
      "eval_runtime": 730.8583,
      "eval_samples_per_second": 19.123,
      "eval_steps_per_second": 1.196,
      "step": 35097
    },
    {
      "epoch": 19.22,
      "learning_rate": 0.00017966071106298505,
      "loss": 0.1855,
      "step": 35500
    },
    {
      "epoch": 19.49,
      "learning_rate": 0.00017514889009204117,
      "loss": 0.1862,
      "step": 36000
    },
    {
      "epoch": 19.76,
      "learning_rate": 0.0001706370691210973,
      "loss": 0.1921,
      "step": 36500
    },
    {
      "epoch": 20.0,
      "eval_LaBSE similarity": 0.644736666482104,
      "eval_bleu": 0.22277214090837807,
      "eval_loss": 0.7024596333503723,
      "eval_runtime": 731.5546,
      "eval_samples_per_second": 19.105,
      "eval_steps_per_second": 1.195,
      "step": 36945
    },
    {
      "epoch": 20.03,
      "learning_rate": 0.0001661252481501534,
      "loss": 0.1931,
      "step": 37000
    },
    {
      "epoch": 20.3,
      "learning_rate": 0.00016161342717920955,
      "loss": 0.1702,
      "step": 37500
    },
    {
      "epoch": 20.57,
      "learning_rate": 0.00015710160620826567,
      "loss": 0.1786,
      "step": 38000
    },
    {
      "epoch": 20.84,
      "learning_rate": 0.0001525897852373218,
      "loss": 0.1808,
      "step": 38500
    },
    {
      "epoch": 21.0,
      "eval_LaBSE similarity": 0.6434357442408501,
      "eval_bleu": 0.22258539444788225,
      "eval_loss": 0.7120693325996399,
      "eval_runtime": 732.5597,
      "eval_samples_per_second": 19.078,
      "eval_steps_per_second": 1.193,
      "step": 38792
    },
    {
      "epoch": 21.11,
      "learning_rate": 0.0001480779642663779,
      "loss": 0.1727,
      "step": 39000
    },
    {
      "epoch": 21.38,
      "learning_rate": 0.00014356614329543406,
      "loss": 0.1606,
      "step": 39500
    },
    {
      "epoch": 21.65,
      "learning_rate": 0.00013905432232449018,
      "loss": 0.168,
      "step": 40000
    },
    {
      "epoch": 21.92,
      "learning_rate": 0.0001345425013535463,
      "loss": 0.171,
      "step": 40500
    },
    {
      "epoch": 22.0,
      "eval_LaBSE similarity": 0.6431315611951554,
      "eval_bleu": 0.22194535539014804,
      "eval_loss": 0.7173079252243042,
      "eval_runtime": 730.9028,
      "eval_samples_per_second": 19.122,
      "eval_steps_per_second": 1.196,
      "step": 40639
    },
    {
      "epoch": 22.2,
      "learning_rate": 0.00013003068038260244,
      "loss": 0.1557,
      "step": 41000
    },
    {
      "epoch": 22.47,
      "learning_rate": 0.00012551885941165856,
      "loss": 0.155,
      "step": 41500
    },
    {
      "epoch": 22.74,
      "learning_rate": 0.00012100703844071467,
      "loss": 0.1586,
      "step": 42000
    },
    {
      "epoch": 23.0,
      "eval_LaBSE similarity": 0.6457557632484162,
      "eval_bleu": 0.22429042016550227,
      "eval_loss": 0.7178816795349121,
      "eval_runtime": 733.5242,
      "eval_samples_per_second": 19.053,
      "eval_steps_per_second": 1.192,
      "step": 42486
    },
    {
      "epoch": 23.01,
      "learning_rate": 0.0001164952174697708,
      "loss": 0.1604,
      "step": 42500
    },
    {
      "epoch": 23.28,
      "learning_rate": 0.00011198339649882692,
      "loss": 0.142,
      "step": 43000
    },
    {
      "epoch": 23.55,
      "learning_rate": 0.00010747157552788306,
      "loss": 0.147,
      "step": 43500
    },
    {
      "epoch": 23.82,
      "learning_rate": 0.00010295975455693918,
      "loss": 0.1516,
      "step": 44000
    },
    {
      "epoch": 24.0,
      "eval_LaBSE similarity": 0.6450597929800783,
      "eval_bleu": 0.223409056978912,
      "eval_loss": 0.7363256216049194,
      "eval_runtime": 731.1036,
      "eval_samples_per_second": 19.116,
      "eval_steps_per_second": 1.195,
      "step": 44334
    },
    {
      "epoch": 24.09,
      "learning_rate": 9.844793358599531e-05,
      "loss": 0.1471,
      "step": 44500
    },
    {
      "epoch": 24.36,
      "learning_rate": 9.393611261505143e-05,
      "loss": 0.1376,
      "step": 45000
    },
    {
      "epoch": 24.63,
      "learning_rate": 8.942429164410756e-05,
      "loss": 0.14,
      "step": 45500
    },
    {
      "epoch": 24.9,
      "learning_rate": 8.49124706731637e-05,
      "loss": 0.1425,
      "step": 46000
    },
    {
      "epoch": 25.0,
      "eval_LaBSE similarity": 0.6440151894096452,
      "eval_bleu": 0.22351487448534282,
      "eval_loss": 0.7389833331108093,
      "eval_runtime": 731.0483,
      "eval_samples_per_second": 19.118,
      "eval_steps_per_second": 1.196,
      "step": 46181
    },
    {
      "epoch": 25.17,
      "learning_rate": 8.040064970221982e-05,
      "loss": 0.1346,
      "step": 46500
    },
    {
      "epoch": 25.44,
      "learning_rate": 7.588882873127595e-05,
      "loss": 0.1318,
      "step": 47000
    },
    {
      "epoch": 25.71,
      "learning_rate": 7.137700776033207e-05,
      "loss": 0.1342,
      "step": 47500
    },
    {
      "epoch": 25.98,
      "learning_rate": 6.68651867893882e-05,
      "loss": 0.1357,
      "step": 48000
    },
    {
      "epoch": 26.0,
      "eval_LaBSE similarity": 0.6438781539249833,
      "eval_bleu": 0.22384959032900478,
      "eval_loss": 0.7478045225143433,
      "eval_runtime": 730.26,
      "eval_samples_per_second": 19.138,
      "eval_steps_per_second": 1.197,
      "step": 48028
    },
    {
      "epoch": 26.26,
      "learning_rate": 6.235336581844433e-05,
      "loss": 0.1246,
      "step": 48500
    },
    {
      "epoch": 26.53,
      "learning_rate": 5.7841544847500454e-05,
      "loss": 0.1262,
      "step": 49000
    },
    {
      "epoch": 26.8,
      "learning_rate": 5.332972387655658e-05,
      "loss": 0.1285,
      "step": 49500
    },
    {
      "epoch": 27.0,
      "eval_LaBSE similarity": 0.6438721785361547,
      "eval_bleu": 0.22299911848016674,
      "eval_loss": 0.7585892081260681,
      "eval_runtime": 730.7966,
      "eval_samples_per_second": 19.124,
      "eval_steps_per_second": 1.196,
      "step": 49875
    },
    {
      "epoch": 27.07,
      "learning_rate": 4.881790290561271e-05,
      "loss": 0.1258,
      "step": 50000
    },
    {
      "epoch": 27.34,
      "learning_rate": 4.430608193466883e-05,
      "loss": 0.1196,
      "step": 50500
    },
    {
      "epoch": 27.61,
      "learning_rate": 3.979426096372496e-05,
      "loss": 0.1208,
      "step": 51000
    },
    {
      "epoch": 27.88,
      "learning_rate": 3.5282439992781086e-05,
      "loss": 0.1234,
      "step": 51500
    },
    {
      "epoch": 28.0,
      "eval_LaBSE similarity": 0.6445591971687343,
      "eval_bleu": 0.22332463264359148,
      "eval_loss": 0.7620986700057983,
      "eval_runtime": 729.8836,
      "eval_samples_per_second": 19.148,
      "eval_steps_per_second": 1.197,
      "step": 51723
    }
  ],
  "max_steps": 55410,
  "num_train_epochs": 30,
  "total_flos": 6.48299892475902e+17,
  "trial_name": null,
  "trial_params": null
}