{
  "best_metric": 0.8703868389129639,
  "best_model_checkpoint": "/d/hpc/projects/FRI/bb6846/run11//mbart-large-50/checkpoint-11083",
  "epoch": 22.999593991067805,
  "global_step": 42486,
  "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": 5.4927,
      "step": 500
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004909763580581122,
      "loss": 3.5335,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00048646453708716834,
      "loss": 2.9081,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_LaBSE similarity": 0.5398781112980036,
      "eval_bleu": 0.09115205757865759,
      "eval_loss": 2.2135918140411377,
      "eval_runtime": 1365.0262,
      "eval_samples_per_second": 10.239,
      "eval_steps_per_second": 0.64,
      "step": 1847
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00048195271611622454,
      "loss": 2.2837,
      "step": 2000
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00047744089514528064,
      "loss": 1.801,
      "step": 2500
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0004729290741743368,
      "loss": 1.5852,
      "step": 3000
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0004684172532033929,
      "loss": 1.4323,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_LaBSE similarity": 0.6886588053033609,
      "eval_bleu": 0.25496466080419466,
      "eval_loss": 1.3433868885040283,
      "eval_runtime": 1227.9131,
      "eval_samples_per_second": 11.382,
      "eval_steps_per_second": 0.712,
      "step": 3694
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.000463905432232449,
      "loss": 1.1374,
      "step": 4000
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0004593936112615051,
      "loss": 1.0137,
      "step": 4500
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.00045488179029056126,
      "loss": 0.9937,
      "step": 5000
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0004503699693196174,
      "loss": 0.9558,
      "step": 5500
    },
    {
      "epoch": 3.0,
      "eval_LaBSE similarity": 0.7105592780650029,
      "eval_bleu": 0.2856338691080233,
      "eval_loss": 1.0794386863708496,
      "eval_runtime": 1168.6662,
      "eval_samples_per_second": 11.959,
      "eval_steps_per_second": 0.748,
      "step": 5541
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.00044585814834867355,
      "loss": 0.6749,
      "step": 6000
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.00044134632737772965,
      "loss": 0.6931,
      "step": 6500
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.0004368345064067858,
      "loss": 0.6971,
      "step": 7000
    },
    {
      "epoch": 4.0,
      "eval_LaBSE similarity": 0.7229989019695549,
      "eval_bleu": 0.3177643404347968,
      "eval_loss": 0.9699832201004028,
      "eval_runtime": 1195.5257,
      "eval_samples_per_second": 11.69,
      "eval_steps_per_second": 0.731,
      "step": 7389
    },
    {
      "epoch": 4.06,
      "learning_rate": 0.0004323226854358419,
      "loss": 0.6386,
      "step": 7500
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.00042781086446489803,
      "loss": 0.4793,
      "step": 8000
    },
    {
      "epoch": 4.6,
      "learning_rate": 0.0004232990434939541,
      "loss": 0.5143,
      "step": 8500
    },
    {
      "epoch": 4.87,
      "learning_rate": 0.00041878722252301027,
      "loss": 0.5263,
      "step": 9000
    },
    {
      "epoch": 5.0,
      "eval_LaBSE similarity": 0.7233434942747937,
      "eval_bleu": 0.3231544767102279,
      "eval_loss": 0.9114313125610352,
      "eval_runtime": 1133.9982,
      "eval_samples_per_second": 12.325,
      "eval_steps_per_second": 0.771,
      "step": 9236
    },
    {
      "epoch": 5.14,
      "learning_rate": 0.0004142754015520664,
      "loss": 0.4303,
      "step": 9500
    },
    {
      "epoch": 5.41,
      "learning_rate": 0.00040976358058112257,
      "loss": 0.3777,
      "step": 10000
    },
    {
      "epoch": 5.68,
      "learning_rate": 0.00040525175961017866,
      "loss": 0.4072,
      "step": 10500
    },
    {
      "epoch": 5.95,
      "learning_rate": 0.0004007399386392348,
      "loss": 0.419,
      "step": 11000
    },
    {
      "epoch": 6.0,
      "eval_LaBSE similarity": 0.7420425884927818,
      "eval_bleu": 0.3542640467481141,
      "eval_loss": 0.8703868389129639,
      "eval_runtime": 1176.62,
      "eval_samples_per_second": 11.878,
      "eval_steps_per_second": 0.743,
      "step": 11083
    },
    {
      "epoch": 6.23,
      "learning_rate": 0.0003962281176682909,
      "loss": 0.3006,
      "step": 11500
    },
    {
      "epoch": 6.5,
      "learning_rate": 0.00039171629669734704,
      "loss": 0.3122,
      "step": 12000
    },
    {
      "epoch": 6.77,
      "learning_rate": 0.00038720447572640314,
      "loss": 0.3367,
      "step": 12500
    },
    {
      "epoch": 7.0,
      "eval_LaBSE similarity": 0.7435614244402539,
      "eval_bleu": 0.3642121844301522,
      "eval_loss": 0.8795487880706787,
      "eval_runtime": 1222.0463,
      "eval_samples_per_second": 11.437,
      "eval_steps_per_second": 0.715,
      "step": 12930
    },
    {
      "epoch": 7.04,
      "learning_rate": 0.00038269265475545934,
      "loss": 0.3307,
      "step": 13000
    },
    {
      "epoch": 7.31,
      "learning_rate": 0.00037818083378451543,
      "loss": 0.2398,
      "step": 13500
    },
    {
      "epoch": 7.58,
      "learning_rate": 0.0003736690128135716,
      "loss": 0.2691,
      "step": 14000
    },
    {
      "epoch": 7.85,
      "learning_rate": 0.00036915719184262767,
      "loss": 0.286,
      "step": 14500
    },
    {
      "epoch": 8.0,
      "eval_LaBSE similarity": 0.7364368761335748,
      "eval_bleu": 0.34924239776356997,
      "eval_loss": 0.8712966442108154,
      "eval_runtime": 1189.2045,
      "eval_samples_per_second": 11.752,
      "eval_steps_per_second": 0.735,
      "step": 14778
    },
    {
      "epoch": 8.12,
      "learning_rate": 0.0003646453708716838,
      "loss": 0.2493,
      "step": 15000
    },
    {
      "epoch": 8.39,
      "learning_rate": 0.0003601335499007399,
      "loss": 0.2157,
      "step": 15500
    },
    {
      "epoch": 8.66,
      "learning_rate": 0.00035562172892979606,
      "loss": 0.2358,
      "step": 16000
    },
    {
      "epoch": 8.93,
      "learning_rate": 0.0003511099079588522,
      "loss": 0.2518,
      "step": 16500
    },
    {
      "epoch": 9.0,
      "eval_LaBSE similarity": 0.7470671729007166,
      "eval_bleu": 0.37070924925201876,
      "eval_loss": 0.8941549062728882,
      "eval_runtime": 1235.4354,
      "eval_samples_per_second": 11.313,
      "eval_steps_per_second": 0.707,
      "step": 16625
    },
    {
      "epoch": 9.2,
      "learning_rate": 0.00034659808698790835,
      "loss": 0.1955,
      "step": 17000
    },
    {
      "epoch": 9.47,
      "learning_rate": 0.00034208626601696444,
      "loss": 0.1953,
      "step": 17500
    },
    {
      "epoch": 9.74,
      "learning_rate": 0.0003375744450460206,
      "loss": 0.2129,
      "step": 18000
    },
    {
      "epoch": 10.0,
      "eval_LaBSE similarity": 0.7456615649679629,
      "eval_bleu": 0.3745055214526943,
      "eval_loss": 0.9131841659545898,
      "eval_runtime": 1223.5826,
      "eval_samples_per_second": 11.422,
      "eval_steps_per_second": 0.714,
      "step": 18472
    },
    {
      "epoch": 10.01,
      "learning_rate": 0.0003330626240750767,
      "loss": 0.219,
      "step": 18500
    },
    {
      "epoch": 10.29,
      "learning_rate": 0.00032855080310413283,
      "loss": 0.1602,
      "step": 19000
    },
    {
      "epoch": 10.56,
      "learning_rate": 0.0003240389821331889,
      "loss": 0.1807,
      "step": 19500
    },
    {
      "epoch": 10.83,
      "learning_rate": 0.00031952716116224507,
      "loss": 0.1925,
      "step": 20000
    },
    {
      "epoch": 11.0,
      "eval_LaBSE similarity": 0.7471288781191697,
      "eval_bleu": 0.37290758010326946,
      "eval_loss": 0.9259693622589111,
      "eval_runtime": 1200.2766,
      "eval_samples_per_second": 11.644,
      "eval_steps_per_second": 0.728,
      "step": 20319
    },
    {
      "epoch": 11.1,
      "learning_rate": 0.0003150153401913012,
      "loss": 0.1765,
      "step": 20500
    },
    {
      "epoch": 11.37,
      "learning_rate": 0.00031050351922035736,
      "loss": 0.1498,
      "step": 21000
    },
    {
      "epoch": 11.64,
      "learning_rate": 0.00030599169824941345,
      "loss": 0.1655,
      "step": 21500
    },
    {
      "epoch": 11.91,
      "learning_rate": 0.0003014798772784696,
      "loss": 0.1752,
      "step": 22000
    },
    {
      "epoch": 12.0,
      "eval_LaBSE similarity": 0.7468445703041946,
      "eval_bleu": 0.37465455155641914,
      "eval_loss": 0.9446151256561279,
      "eval_runtime": 1244.3591,
      "eval_samples_per_second": 11.231,
      "eval_steps_per_second": 0.702,
      "step": 22167
    },
    {
      "epoch": 12.18,
      "learning_rate": 0.0002969680563075257,
      "loss": 0.1418,
      "step": 22500
    },
    {
      "epoch": 12.45,
      "learning_rate": 0.00029245623533658184,
      "loss": 0.1396,
      "step": 23000
    },
    {
      "epoch": 12.72,
      "learning_rate": 0.00028794441436563793,
      "loss": 0.1526,
      "step": 23500
    },
    {
      "epoch": 12.99,
      "learning_rate": 0.00028343259339469413,
      "loss": 0.1601,
      "step": 24000
    },
    {
      "epoch": 13.0,
      "eval_LaBSE similarity": 0.7469340100011889,
      "eval_bleu": 0.37582936090048125,
      "eval_loss": 0.9434267282485962,
      "eval_runtime": 1202.1543,
      "eval_samples_per_second": 11.626,
      "eval_steps_per_second": 0.727,
      "step": 24014
    },
    {
      "epoch": 13.26,
      "learning_rate": 0.0002789207724237502,
      "loss": 0.1195,
      "step": 24500
    },
    {
      "epoch": 13.53,
      "learning_rate": 0.00027440895145280637,
      "loss": 0.1312,
      "step": 25000
    },
    {
      "epoch": 13.8,
      "learning_rate": 0.00026989713048186246,
      "loss": 0.141,
      "step": 25500
    },
    {
      "epoch": 14.0,
      "eval_LaBSE similarity": 0.7477662809688649,
      "eval_bleu": 0.3734200651221202,
      "eval_loss": 0.9459152221679688,
      "eval_runtime": 1210.207,
      "eval_samples_per_second": 11.548,
      "eval_steps_per_second": 0.722,
      "step": 25861
    },
    {
      "epoch": 14.07,
      "learning_rate": 0.0002653853095109186,
      "loss": 0.1336,
      "step": 26000
    },
    {
      "epoch": 14.35,
      "learning_rate": 0.0002608734885399747,
      "loss": 0.1123,
      "step": 26500
    },
    {
      "epoch": 14.62,
      "learning_rate": 0.00025636166756903085,
      "loss": 0.1229,
      "step": 27000
    },
    {
      "epoch": 14.89,
      "learning_rate": 0.000251849846598087,
      "loss": 0.1303,
      "step": 27500
    },
    {
      "epoch": 15.0,
      "eval_LaBSE similarity": 0.7455726656700262,
      "eval_bleu": 0.3728343171898791,
      "eval_loss": 0.9986817240715027,
      "eval_runtime": 1216.2061,
      "eval_samples_per_second": 11.491,
      "eval_steps_per_second": 0.719,
      "step": 27708
    },
    {
      "epoch": 15.16,
      "learning_rate": 0.0002473380256271431,
      "loss": 0.1126,
      "step": 28000
    },
    {
      "epoch": 15.43,
      "learning_rate": 0.00024282620465619926,
      "loss": 0.1065,
      "step": 28500
    },
    {
      "epoch": 15.7,
      "learning_rate": 0.00023831438368525538,
      "loss": 0.1155,
      "step": 29000
    },
    {
      "epoch": 15.97,
      "learning_rate": 0.0002338025627143115,
      "loss": 0.1216,
      "step": 29500
    },
    {
      "epoch": 16.0,
      "eval_LaBSE similarity": 0.7427742036105345,
      "eval_bleu": 0.36683694992363525,
      "eval_loss": 1.0019079446792603,
      "eval_runtime": 1175.8161,
      "eval_samples_per_second": 11.886,
      "eval_steps_per_second": 0.743,
      "step": 29556
    },
    {
      "epoch": 16.24,
      "learning_rate": 0.00022929074174336762,
      "loss": 0.0952,
      "step": 30000
    },
    {
      "epoch": 16.51,
      "learning_rate": 0.00022477892077242377,
      "loss": 0.1014,
      "step": 30500
    },
    {
      "epoch": 16.78,
      "learning_rate": 0.0002202670998014799,
      "loss": 0.1071,
      "step": 31000
    },
    {
      "epoch": 17.0,
      "eval_LaBSE similarity": 0.745781449845613,
      "eval_bleu": 0.372475196057264,
      "eval_loss": 1.0133806467056274,
      "eval_runtime": 1177.7522,
      "eval_samples_per_second": 11.867,
      "eval_steps_per_second": 0.742,
      "step": 31403
    },
    {
      "epoch": 17.05,
      "learning_rate": 0.000215755278830536,
      "loss": 0.1058,
      "step": 31500
    },
    {
      "epoch": 17.32,
      "learning_rate": 0.00021124345785959216,
      "loss": 0.0861,
      "step": 32000
    },
    {
      "epoch": 17.59,
      "learning_rate": 0.00020673163688864828,
      "loss": 0.0962,
      "step": 32500
    },
    {
      "epoch": 17.86,
      "learning_rate": 0.0002022198159177044,
      "loss": 0.1001,
      "step": 33000
    },
    {
      "epoch": 18.0,
      "eval_LaBSE similarity": 0.7447677442699225,
      "eval_bleu": 0.3671262927472257,
      "eval_loss": 1.0423822402954102,
      "eval_runtime": 1175.2578,
      "eval_samples_per_second": 11.892,
      "eval_steps_per_second": 0.744,
      "step": 33250
    },
    {
      "epoch": 18.14,
      "learning_rate": 0.00019770799494676051,
      "loss": 0.0892,
      "step": 33500
    },
    {
      "epoch": 18.41,
      "learning_rate": 0.00019319617397581666,
      "loss": 0.0818,
      "step": 34000
    },
    {
      "epoch": 18.68,
      "learning_rate": 0.00018868435300487278,
      "loss": 0.0898,
      "step": 34500
    },
    {
      "epoch": 18.95,
      "learning_rate": 0.0001841725320339289,
      "loss": 0.0945,
      "step": 35000
    },
    {
      "epoch": 19.0,
      "eval_LaBSE similarity": 0.7462503632276549,
      "eval_bleu": 0.3708357814718823,
      "eval_loss": 1.0612070560455322,
      "eval_runtime": 1198.6678,
      "eval_samples_per_second": 11.66,
      "eval_steps_per_second": 0.729,
      "step": 35097
    },
    {
      "epoch": 19.22,
      "learning_rate": 0.00017966071106298505,
      "loss": 0.0762,
      "step": 35500
    },
    {
      "epoch": 19.49,
      "learning_rate": 0.00017514889009204117,
      "loss": 0.0796,
      "step": 36000
    },
    {
      "epoch": 19.76,
      "learning_rate": 0.0001706370691210973,
      "loss": 0.0844,
      "step": 36500
    },
    {
      "epoch": 20.0,
      "eval_LaBSE similarity": 0.7422917604312603,
      "eval_bleu": 0.36480748501421006,
      "eval_loss": 1.0829150676727295,
      "eval_runtime": 1172.318,
      "eval_samples_per_second": 11.922,
      "eval_steps_per_second": 0.746,
      "step": 36945
    },
    {
      "epoch": 20.03,
      "learning_rate": 0.0001661252481501534,
      "loss": 0.0855,
      "step": 37000
    },
    {
      "epoch": 20.3,
      "learning_rate": 0.00016161342717920955,
      "loss": 0.067,
      "step": 37500
    },
    {
      "epoch": 20.57,
      "learning_rate": 0.00015710160620826567,
      "loss": 0.0753,
      "step": 38000
    },
    {
      "epoch": 20.84,
      "learning_rate": 0.0001525897852373218,
      "loss": 0.0793,
      "step": 38500
    },
    {
      "epoch": 21.0,
      "eval_LaBSE similarity": 0.7398600145383046,
      "eval_bleu": 0.3595059153487151,
      "eval_loss": 1.1107969284057617,
      "eval_runtime": 1170.616,
      "eval_samples_per_second": 11.939,
      "eval_steps_per_second": 0.747,
      "step": 38792
    },
    {
      "epoch": 21.11,
      "learning_rate": 0.0001480779642663779,
      "loss": 0.0731,
      "step": 39000
    },
    {
      "epoch": 21.38,
      "learning_rate": 0.00014356614329543406,
      "loss": 0.0651,
      "step": 39500
    },
    {
      "epoch": 21.65,
      "learning_rate": 0.00013905432232449018,
      "loss": 0.0716,
      "step": 40000
    },
    {
      "epoch": 21.92,
      "learning_rate": 0.0001345425013535463,
      "loss": 0.0748,
      "step": 40500
    },
    {
      "epoch": 22.0,
      "eval_LaBSE similarity": 0.7413012238766339,
      "eval_bleu": 0.36304917318364166,
      "eval_loss": 1.1223746538162231,
      "eval_runtime": 1177.2462,
      "eval_samples_per_second": 11.872,
      "eval_steps_per_second": 0.742,
      "step": 40639
    },
    {
      "epoch": 22.2,
      "learning_rate": 0.00013003068038260244,
      "loss": 0.0618,
      "step": 41000
    },
    {
      "epoch": 22.47,
      "learning_rate": 0.00012551885941165856,
      "loss": 0.0619,
      "step": 41500
    },
    {
      "epoch": 22.74,
      "learning_rate": 0.00012100703844071467,
      "loss": 0.0676,
      "step": 42000
    },
    {
      "epoch": 23.0,
      "eval_LaBSE similarity": 0.7416807065988956,
      "eval_bleu": 0.3601707131632923,
      "eval_loss": 1.1566632986068726,
      "eval_runtime": 1175.1131,
      "eval_samples_per_second": 11.893,
      "eval_steps_per_second": 0.744,
      "step": 42486
    }
  ],
  "max_steps": 55410,
  "num_train_epochs": 30,
  "total_flos": 4.22917524411777e+17,
  "trial_name": null,
  "trial_params": null
}