byt5-small-jpn-ain-mt / trainer_state.json
koki's picture
initial commit
3ed3617
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 7.0,
"global_step": 131488,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.00019997920437606474,
"loss": 2.1277,
"step": 500
},
{
"epoch": 0.05,
"learning_rate": 0.00019995840875212947,
"loss": 1.7025,
"step": 1000
},
{
"epoch": 0.08,
"learning_rate": 0.0001999376131281942,
"loss": 1.5776,
"step": 1500
},
{
"epoch": 0.11,
"learning_rate": 0.00019991681750425894,
"loss": 1.4835,
"step": 2000
},
{
"epoch": 0.13,
"learning_rate": 0.00019989602188032367,
"loss": 1.4104,
"step": 2500
},
{
"epoch": 0.16,
"learning_rate": 0.00019987522625638843,
"loss": 1.3631,
"step": 3000
},
{
"epoch": 0.19,
"learning_rate": 0.00019985443063245316,
"loss": 1.2903,
"step": 3500
},
{
"epoch": 0.21,
"learning_rate": 0.00019983363500851792,
"loss": 1.2338,
"step": 4000
},
{
"epoch": 0.24,
"learning_rate": 0.00019981283938458265,
"loss": 1.2018,
"step": 4500
},
{
"epoch": 0.27,
"learning_rate": 0.00019979204376064738,
"loss": 1.1608,
"step": 5000
},
{
"epoch": 0.29,
"learning_rate": 0.00019977124813671211,
"loss": 1.1202,
"step": 5500
},
{
"epoch": 0.32,
"learning_rate": 0.00019975045251277685,
"loss": 1.0989,
"step": 6000
},
{
"epoch": 0.35,
"learning_rate": 0.00019972965688884158,
"loss": 1.0627,
"step": 6500
},
{
"epoch": 0.37,
"learning_rate": 0.0001997088612649063,
"loss": 1.0424,
"step": 7000
},
{
"epoch": 0.4,
"learning_rate": 0.00019968806564097104,
"loss": 1.0196,
"step": 7500
},
{
"epoch": 0.43,
"learning_rate": 0.00019966727001703577,
"loss": 0.9957,
"step": 8000
},
{
"epoch": 0.45,
"learning_rate": 0.0001996464743931005,
"loss": 0.9923,
"step": 8500
},
{
"epoch": 0.48,
"learning_rate": 0.00019962567876916524,
"loss": 0.9711,
"step": 9000
},
{
"epoch": 0.51,
"learning_rate": 0.00019960488314522997,
"loss": 0.9457,
"step": 9500
},
{
"epoch": 0.53,
"learning_rate": 0.00019958408752129473,
"loss": 0.9503,
"step": 10000
},
{
"epoch": 0.56,
"learning_rate": 0.00019956329189735946,
"loss": 0.9184,
"step": 10500
},
{
"epoch": 0.59,
"learning_rate": 0.00019954249627342422,
"loss": 0.9012,
"step": 11000
},
{
"epoch": 0.61,
"learning_rate": 0.00019952170064948895,
"loss": 0.8869,
"step": 11500
},
{
"epoch": 0.64,
"learning_rate": 0.00019950090502555368,
"loss": 0.9034,
"step": 12000
},
{
"epoch": 0.67,
"learning_rate": 0.00019948010940161841,
"loss": 0.8827,
"step": 12500
},
{
"epoch": 0.69,
"learning_rate": 0.00019945931377768315,
"loss": 0.8549,
"step": 13000
},
{
"epoch": 0.72,
"learning_rate": 0.00019943851815374788,
"loss": 0.8701,
"step": 13500
},
{
"epoch": 0.75,
"learning_rate": 0.0001994177225298126,
"loss": 0.8362,
"step": 14000
},
{
"epoch": 0.77,
"learning_rate": 0.00019939692690587734,
"loss": 0.8444,
"step": 14500
},
{
"epoch": 0.8,
"learning_rate": 0.00019937613128194207,
"loss": 0.8501,
"step": 15000
},
{
"epoch": 0.83,
"learning_rate": 0.0001993553356580068,
"loss": 0.8297,
"step": 15500
},
{
"epoch": 0.85,
"learning_rate": 0.00019933454003407156,
"loss": 0.82,
"step": 16000
},
{
"epoch": 0.88,
"learning_rate": 0.0001993137444101363,
"loss": 0.8219,
"step": 16500
},
{
"epoch": 0.91,
"learning_rate": 0.00019929294878620103,
"loss": 0.8074,
"step": 17000
},
{
"epoch": 0.93,
"learning_rate": 0.00019927215316226576,
"loss": 0.7981,
"step": 17500
},
{
"epoch": 0.96,
"learning_rate": 0.00019925135753833052,
"loss": 0.7946,
"step": 18000
},
{
"epoch": 0.98,
"learning_rate": 0.00019923056191439525,
"loss": 0.7888,
"step": 18500
},
{
"epoch": 1.0,
"eval_bleu": 0.2617,
"eval_gen_len": 18.172,
"eval_loss": 0.6930029392242432,
"eval_runtime": 742.401,
"eval_samples_per_second": 12.652,
"eval_steps_per_second": 3.164,
"step": 18784
},
{
"epoch": 1.01,
"learning_rate": 0.00019920976629045998,
"loss": 0.7721,
"step": 19000
},
{
"epoch": 1.04,
"learning_rate": 0.00019918897066652471,
"loss": 0.7574,
"step": 19500
},
{
"epoch": 1.06,
"learning_rate": 0.00019916817504258945,
"loss": 0.7442,
"step": 20000
},
{
"epoch": 1.09,
"learning_rate": 0.00019914737941865418,
"loss": 0.7454,
"step": 20500
},
{
"epoch": 1.12,
"learning_rate": 0.0001991265837947189,
"loss": 0.729,
"step": 21000
},
{
"epoch": 1.14,
"learning_rate": 0.00019910578817078364,
"loss": 0.7422,
"step": 21500
},
{
"epoch": 1.17,
"learning_rate": 0.00019908499254684837,
"loss": 0.7405,
"step": 22000
},
{
"epoch": 1.2,
"learning_rate": 0.00019906419692291313,
"loss": 0.7231,
"step": 22500
},
{
"epoch": 1.22,
"learning_rate": 0.00019904340129897787,
"loss": 0.7142,
"step": 23000
},
{
"epoch": 1.25,
"learning_rate": 0.0001990226056750426,
"loss": 0.7207,
"step": 23500
},
{
"epoch": 1.28,
"learning_rate": 0.00019900181005110733,
"loss": 0.7087,
"step": 24000
},
{
"epoch": 1.3,
"learning_rate": 0.0001989810144271721,
"loss": 0.7243,
"step": 24500
},
{
"epoch": 1.33,
"learning_rate": 0.00019896021880323682,
"loss": 0.6951,
"step": 25000
},
{
"epoch": 1.36,
"learning_rate": 0.00019893942317930155,
"loss": 0.7001,
"step": 25500
},
{
"epoch": 1.38,
"learning_rate": 0.00019891862755536628,
"loss": 0.7297,
"step": 26000
},
{
"epoch": 1.41,
"learning_rate": 0.00019889783193143102,
"loss": 0.7032,
"step": 26500
},
{
"epoch": 1.44,
"learning_rate": 0.00019887703630749575,
"loss": 0.6943,
"step": 27000
},
{
"epoch": 1.46,
"learning_rate": 0.00019885624068356048,
"loss": 0.6953,
"step": 27500
},
{
"epoch": 1.49,
"learning_rate": 0.0001988354450596252,
"loss": 0.7025,
"step": 28000
},
{
"epoch": 1.52,
"learning_rate": 0.00019881464943568994,
"loss": 0.6901,
"step": 28500
},
{
"epoch": 1.54,
"learning_rate": 0.0001987938538117547,
"loss": 0.6867,
"step": 29000
},
{
"epoch": 1.57,
"learning_rate": 0.00019877305818781943,
"loss": 0.6826,
"step": 29500
},
{
"epoch": 1.6,
"learning_rate": 0.00019875226256388417,
"loss": 0.678,
"step": 30000
},
{
"epoch": 1.62,
"learning_rate": 0.0001987314669399489,
"loss": 0.6763,
"step": 30500
},
{
"epoch": 1.65,
"learning_rate": 0.00019871067131601363,
"loss": 0.6681,
"step": 31000
},
{
"epoch": 1.68,
"learning_rate": 0.0001986898756920784,
"loss": 0.6792,
"step": 31500
},
{
"epoch": 1.7,
"learning_rate": 0.00019866908006814312,
"loss": 0.6776,
"step": 32000
},
{
"epoch": 1.73,
"learning_rate": 0.00019864828444420785,
"loss": 0.677,
"step": 32500
},
{
"epoch": 1.76,
"learning_rate": 0.00019862748882027258,
"loss": 0.6787,
"step": 33000
},
{
"epoch": 1.78,
"learning_rate": 0.00019860669319633732,
"loss": 0.6695,
"step": 33500
},
{
"epoch": 1.81,
"learning_rate": 0.00019858589757240205,
"loss": 0.6555,
"step": 34000
},
{
"epoch": 1.84,
"learning_rate": 0.00019856510194846678,
"loss": 0.6721,
"step": 34500
},
{
"epoch": 1.86,
"learning_rate": 0.0001985443063245315,
"loss": 0.6617,
"step": 35000
},
{
"epoch": 1.89,
"learning_rate": 0.00019852351070059627,
"loss": 0.6562,
"step": 35500
},
{
"epoch": 1.92,
"learning_rate": 0.000198502715076661,
"loss": 0.6496,
"step": 36000
},
{
"epoch": 1.94,
"learning_rate": 0.00019848191945272573,
"loss": 0.6447,
"step": 36500
},
{
"epoch": 1.97,
"learning_rate": 0.00019846112382879047,
"loss": 0.6516,
"step": 37000
},
{
"epoch": 2.0,
"learning_rate": 0.0001984403282048552,
"loss": 0.6424,
"step": 37500
},
{
"epoch": 2.0,
"eval_bleu": 0.2865,
"eval_gen_len": 18.082,
"eval_loss": 0.5811895728111267,
"eval_runtime": 746.9324,
"eval_samples_per_second": 12.575,
"eval_steps_per_second": 3.145,
"step": 37568
},
{
"epoch": 2.02,
"learning_rate": 0.00019841953258091993,
"loss": 0.6152,
"step": 38000
},
{
"epoch": 2.05,
"learning_rate": 0.0001983987369569847,
"loss": 0.6005,
"step": 38500
},
{
"epoch": 2.08,
"learning_rate": 0.00019837794133304942,
"loss": 0.5937,
"step": 39000
},
{
"epoch": 2.1,
"learning_rate": 0.00019835714570911415,
"loss": 0.5997,
"step": 39500
},
{
"epoch": 2.13,
"learning_rate": 0.00019833635008517888,
"loss": 0.5995,
"step": 40000
},
{
"epoch": 2.16,
"learning_rate": 0.00019831555446124362,
"loss": 0.5934,
"step": 40500
},
{
"epoch": 2.18,
"learning_rate": 0.00019829475883730835,
"loss": 0.5894,
"step": 41000
},
{
"epoch": 2.21,
"learning_rate": 0.0001982739632133731,
"loss": 0.5882,
"step": 41500
},
{
"epoch": 2.24,
"learning_rate": 0.00019825316758943784,
"loss": 0.608,
"step": 42000
},
{
"epoch": 2.26,
"learning_rate": 0.00019823237196550257,
"loss": 0.5991,
"step": 42500
},
{
"epoch": 2.29,
"learning_rate": 0.0001982115763415673,
"loss": 0.5981,
"step": 43000
},
{
"epoch": 2.32,
"learning_rate": 0.00019819078071763203,
"loss": 0.5962,
"step": 43500
},
{
"epoch": 2.34,
"learning_rate": 0.00019816998509369677,
"loss": 0.5776,
"step": 44000
},
{
"epoch": 2.37,
"learning_rate": 0.0001981491894697615,
"loss": 0.5953,
"step": 44500
},
{
"epoch": 2.4,
"learning_rate": 0.00019812839384582626,
"loss": 0.6036,
"step": 45000
},
{
"epoch": 2.42,
"learning_rate": 0.000198107598221891,
"loss": 0.5877,
"step": 45500
},
{
"epoch": 2.45,
"learning_rate": 0.00019808680259795572,
"loss": 0.5864,
"step": 46000
},
{
"epoch": 2.48,
"learning_rate": 0.00019806600697402045,
"loss": 0.579,
"step": 46500
},
{
"epoch": 2.5,
"learning_rate": 0.00019804521135008518,
"loss": 0.5873,
"step": 47000
},
{
"epoch": 2.53,
"learning_rate": 0.00019802441572614992,
"loss": 0.5912,
"step": 47500
},
{
"epoch": 2.56,
"learning_rate": 0.00019800362010221468,
"loss": 0.592,
"step": 48000
},
{
"epoch": 2.58,
"learning_rate": 0.0001979828244782794,
"loss": 0.5822,
"step": 48500
},
{
"epoch": 2.61,
"learning_rate": 0.00019796202885434414,
"loss": 0.5864,
"step": 49000
},
{
"epoch": 2.64,
"learning_rate": 0.00019794123323040887,
"loss": 0.5738,
"step": 49500
},
{
"epoch": 2.66,
"learning_rate": 0.0001979204376064736,
"loss": 0.583,
"step": 50000
},
{
"epoch": 2.69,
"learning_rate": 0.00019789964198253833,
"loss": 0.5803,
"step": 50500
},
{
"epoch": 2.72,
"learning_rate": 0.00019787884635860307,
"loss": 0.5884,
"step": 51000
},
{
"epoch": 2.74,
"learning_rate": 0.0001978580507346678,
"loss": 0.5938,
"step": 51500
},
{
"epoch": 2.77,
"learning_rate": 0.00019783725511073256,
"loss": 0.5761,
"step": 52000
},
{
"epoch": 2.79,
"learning_rate": 0.0001978164594867973,
"loss": 0.5829,
"step": 52500
},
{
"epoch": 2.82,
"learning_rate": 0.00019779566386286202,
"loss": 0.5764,
"step": 53000
},
{
"epoch": 2.85,
"learning_rate": 0.00019777486823892675,
"loss": 0.5707,
"step": 53500
},
{
"epoch": 2.87,
"learning_rate": 0.00019775407261499148,
"loss": 0.5836,
"step": 54000
},
{
"epoch": 2.9,
"learning_rate": 0.00019773327699105624,
"loss": 0.5756,
"step": 54500
},
{
"epoch": 2.93,
"learning_rate": 0.00019771248136712098,
"loss": 0.5833,
"step": 55000
},
{
"epoch": 2.95,
"learning_rate": 0.0001976916857431857,
"loss": 0.5638,
"step": 55500
},
{
"epoch": 2.98,
"learning_rate": 0.00019767089011925044,
"loss": 0.5567,
"step": 56000
},
{
"epoch": 3.0,
"eval_bleu": 0.2983,
"eval_gen_len": 18.1221,
"eval_loss": 0.5350888967514038,
"eval_runtime": 744.9161,
"eval_samples_per_second": 12.609,
"eval_steps_per_second": 3.153,
"step": 56352
},
{
"epoch": 3.01,
"learning_rate": 0.00019765009449531517,
"loss": 0.5522,
"step": 56500
},
{
"epoch": 3.03,
"learning_rate": 0.0001976292988713799,
"loss": 0.513,
"step": 57000
},
{
"epoch": 3.06,
"learning_rate": 0.00019760850324744463,
"loss": 0.5227,
"step": 57500
},
{
"epoch": 3.09,
"learning_rate": 0.00019758770762350937,
"loss": 0.5269,
"step": 58000
},
{
"epoch": 3.11,
"learning_rate": 0.0001975669119995741,
"loss": 0.5244,
"step": 58500
},
{
"epoch": 3.14,
"learning_rate": 0.00019754611637563886,
"loss": 0.5225,
"step": 59000
},
{
"epoch": 3.17,
"learning_rate": 0.0001975253207517036,
"loss": 0.5256,
"step": 59500
},
{
"epoch": 3.19,
"learning_rate": 0.00019750452512776832,
"loss": 0.5258,
"step": 60000
},
{
"epoch": 3.22,
"learning_rate": 0.00019748372950383305,
"loss": 0.5243,
"step": 60500
},
{
"epoch": 3.25,
"learning_rate": 0.0001974629338798978,
"loss": 0.5217,
"step": 61000
},
{
"epoch": 3.27,
"learning_rate": 0.00019744213825596254,
"loss": 0.5202,
"step": 61500
},
{
"epoch": 3.3,
"learning_rate": 0.00019742134263202728,
"loss": 0.5272,
"step": 62000
},
{
"epoch": 3.33,
"learning_rate": 0.000197400547008092,
"loss": 0.5183,
"step": 62500
},
{
"epoch": 3.35,
"learning_rate": 0.00019737975138415674,
"loss": 0.5239,
"step": 63000
},
{
"epoch": 3.38,
"learning_rate": 0.00019735895576022147,
"loss": 0.5247,
"step": 63500
},
{
"epoch": 3.41,
"learning_rate": 0.0001973381601362862,
"loss": 0.531,
"step": 64000
},
{
"epoch": 3.43,
"learning_rate": 0.00019731736451235093,
"loss": 0.515,
"step": 64500
},
{
"epoch": 3.46,
"learning_rate": 0.00019729656888841567,
"loss": 0.5154,
"step": 65000
},
{
"epoch": 3.49,
"learning_rate": 0.00019727577326448043,
"loss": 0.5252,
"step": 65500
},
{
"epoch": 3.51,
"learning_rate": 0.00019725497764054516,
"loss": 0.5094,
"step": 66000
},
{
"epoch": 3.54,
"learning_rate": 0.0001972341820166099,
"loss": 0.526,
"step": 66500
},
{
"epoch": 3.57,
"learning_rate": 0.00019721338639267465,
"loss": 0.5175,
"step": 67000
},
{
"epoch": 3.59,
"learning_rate": 0.00019719259076873938,
"loss": 0.523,
"step": 67500
},
{
"epoch": 3.62,
"learning_rate": 0.0001971717951448041,
"loss": 0.514,
"step": 68000
},
{
"epoch": 3.65,
"learning_rate": 0.00019715099952086884,
"loss": 0.5236,
"step": 68500
},
{
"epoch": 3.67,
"learning_rate": 0.00019713020389693358,
"loss": 0.5167,
"step": 69000
},
{
"epoch": 3.7,
"learning_rate": 0.0001971094082729983,
"loss": 0.5079,
"step": 69500
},
{
"epoch": 3.73,
"learning_rate": 0.00019708861264906304,
"loss": 0.5142,
"step": 70000
},
{
"epoch": 3.75,
"learning_rate": 0.00019706781702512777,
"loss": 0.5186,
"step": 70500
},
{
"epoch": 3.78,
"learning_rate": 0.0001970470214011925,
"loss": 0.5239,
"step": 71000
},
{
"epoch": 3.81,
"learning_rate": 0.00019702622577725724,
"loss": 0.5131,
"step": 71500
},
{
"epoch": 3.83,
"learning_rate": 0.00019700543015332197,
"loss": 0.519,
"step": 72000
},
{
"epoch": 3.86,
"learning_rate": 0.00019698463452938673,
"loss": 0.5202,
"step": 72500
},
{
"epoch": 3.89,
"learning_rate": 0.00019696383890545146,
"loss": 0.5305,
"step": 73000
},
{
"epoch": 3.91,
"learning_rate": 0.00019694304328151622,
"loss": 0.5042,
"step": 73500
},
{
"epoch": 3.94,
"learning_rate": 0.00019692224765758095,
"loss": 0.513,
"step": 74000
},
{
"epoch": 3.97,
"learning_rate": 0.00019690145203364568,
"loss": 0.5272,
"step": 74500
},
{
"epoch": 3.99,
"learning_rate": 0.0001968806564097104,
"loss": 0.5119,
"step": 75000
},
{
"epoch": 4.0,
"eval_bleu": 0.3043,
"eval_gen_len": 18.1054,
"eval_loss": 0.5083436965942383,
"eval_runtime": 750.8628,
"eval_samples_per_second": 12.51,
"eval_steps_per_second": 3.128,
"step": 75136
},
{
"epoch": 4.02,
"learning_rate": 0.00019685986078577514,
"loss": 0.4702,
"step": 75500
},
{
"epoch": 4.05,
"learning_rate": 0.00019683906516183988,
"loss": 0.4581,
"step": 76000
},
{
"epoch": 4.07,
"learning_rate": 0.0001968182695379046,
"loss": 0.4664,
"step": 76500
},
{
"epoch": 4.1,
"learning_rate": 0.00019679747391396934,
"loss": 0.4596,
"step": 77000
},
{
"epoch": 4.13,
"learning_rate": 0.00019677667829003407,
"loss": 0.459,
"step": 77500
},
{
"epoch": 4.15,
"learning_rate": 0.0001967558826660988,
"loss": 0.4688,
"step": 78000
},
{
"epoch": 4.18,
"learning_rate": 0.00019673508704216354,
"loss": 0.482,
"step": 78500
},
{
"epoch": 4.21,
"learning_rate": 0.00019671429141822827,
"loss": 0.4698,
"step": 79000
},
{
"epoch": 4.23,
"learning_rate": 0.00019669349579429303,
"loss": 0.468,
"step": 79500
},
{
"epoch": 4.26,
"learning_rate": 0.00019667270017035776,
"loss": 0.4625,
"step": 80000
},
{
"epoch": 4.29,
"learning_rate": 0.00019665190454642252,
"loss": 0.4744,
"step": 80500
},
{
"epoch": 4.31,
"learning_rate": 0.00019663110892248725,
"loss": 0.4666,
"step": 81000
},
{
"epoch": 4.34,
"learning_rate": 0.00019661031329855198,
"loss": 0.4717,
"step": 81500
},
{
"epoch": 4.37,
"learning_rate": 0.0001965895176746167,
"loss": 0.4721,
"step": 82000
},
{
"epoch": 4.39,
"learning_rate": 0.00019656872205068144,
"loss": 0.4576,
"step": 82500
},
{
"epoch": 4.42,
"learning_rate": 0.00019654792642674618,
"loss": 0.4664,
"step": 83000
},
{
"epoch": 4.45,
"learning_rate": 0.0001965271308028109,
"loss": 0.4633,
"step": 83500
},
{
"epoch": 4.47,
"learning_rate": 0.00019650633517887564,
"loss": 0.4736,
"step": 84000
},
{
"epoch": 4.5,
"learning_rate": 0.00019648553955494037,
"loss": 0.4687,
"step": 84500
},
{
"epoch": 4.53,
"learning_rate": 0.0001964647439310051,
"loss": 0.4818,
"step": 85000
},
{
"epoch": 4.55,
"learning_rate": 0.00019644394830706984,
"loss": 0.4668,
"step": 85500
},
{
"epoch": 4.58,
"learning_rate": 0.0001964231526831346,
"loss": 0.4745,
"step": 86000
},
{
"epoch": 4.6,
"learning_rate": 0.00019640235705919933,
"loss": 0.4686,
"step": 86500
},
{
"epoch": 4.63,
"learning_rate": 0.00019638156143526406,
"loss": 0.4846,
"step": 87000
},
{
"epoch": 4.66,
"learning_rate": 0.00019636076581132882,
"loss": 0.4714,
"step": 87500
},
{
"epoch": 4.68,
"learning_rate": 0.00019633997018739355,
"loss": 0.4656,
"step": 88000
},
{
"epoch": 4.71,
"learning_rate": 0.00019631917456345828,
"loss": 0.4963,
"step": 88500
},
{
"epoch": 4.74,
"learning_rate": 0.000196298378939523,
"loss": 0.4754,
"step": 89000
},
{
"epoch": 4.76,
"learning_rate": 0.00019627758331558774,
"loss": 0.4596,
"step": 89500
},
{
"epoch": 4.79,
"learning_rate": 0.00019625678769165248,
"loss": 0.4693,
"step": 90000
},
{
"epoch": 4.82,
"learning_rate": 0.0001962359920677172,
"loss": 0.4833,
"step": 90500
},
{
"epoch": 4.84,
"learning_rate": 0.00019621519644378194,
"loss": 0.4847,
"step": 91000
},
{
"epoch": 4.87,
"learning_rate": 0.00019619440081984667,
"loss": 0.4752,
"step": 91500
},
{
"epoch": 4.9,
"learning_rate": 0.0001961736051959114,
"loss": 0.4805,
"step": 92000
},
{
"epoch": 4.92,
"learning_rate": 0.00019615280957197614,
"loss": 0.4869,
"step": 92500
},
{
"epoch": 4.95,
"learning_rate": 0.0001961320139480409,
"loss": 0.4731,
"step": 93000
},
{
"epoch": 4.98,
"learning_rate": 0.00019611121832410563,
"loss": 0.4624,
"step": 93500
},
{
"epoch": 5.0,
"eval_bleu": 0.3071,
"eval_gen_len": 18.1357,
"eval_loss": 0.49742114543914795,
"eval_runtime": 738.3061,
"eval_samples_per_second": 12.722,
"eval_steps_per_second": 3.182,
"step": 93920
},
{
"epoch": 5.0,
"learning_rate": 0.00019609042270017039,
"loss": 0.4743,
"step": 94000
},
{
"epoch": 5.03,
"learning_rate": 0.00019606962707623512,
"loss": 0.4185,
"step": 94500
},
{
"epoch": 5.06,
"learning_rate": 0.00019604883145229985,
"loss": 0.4121,
"step": 95000
},
{
"epoch": 5.08,
"learning_rate": 0.00019602803582836458,
"loss": 0.4053,
"step": 95500
},
{
"epoch": 5.11,
"learning_rate": 0.0001960072402044293,
"loss": 0.4169,
"step": 96000
},
{
"epoch": 5.14,
"learning_rate": 0.00019598644458049405,
"loss": 0.4232,
"step": 96500
},
{
"epoch": 5.16,
"learning_rate": 0.00019596564895655878,
"loss": 0.4324,
"step": 97000
},
{
"epoch": 5.19,
"learning_rate": 0.0001959448533326235,
"loss": 0.4204,
"step": 97500
},
{
"epoch": 5.22,
"learning_rate": 0.00019592405770868824,
"loss": 0.4275,
"step": 98000
},
{
"epoch": 5.24,
"learning_rate": 0.00019590326208475297,
"loss": 0.424,
"step": 98500
},
{
"epoch": 5.27,
"learning_rate": 0.00019588246646081773,
"loss": 0.4277,
"step": 99000
},
{
"epoch": 5.3,
"learning_rate": 0.00019586167083688246,
"loss": 0.4265,
"step": 99500
},
{
"epoch": 5.32,
"learning_rate": 0.0001958408752129472,
"loss": 0.4272,
"step": 100000
},
{
"epoch": 5.35,
"learning_rate": 0.00019582007958901193,
"loss": 0.4221,
"step": 100500
},
{
"epoch": 5.38,
"learning_rate": 0.00019579928396507669,
"loss": 0.4289,
"step": 101000
},
{
"epoch": 5.4,
"learning_rate": 0.00019577848834114142,
"loss": 0.4427,
"step": 101500
},
{
"epoch": 5.43,
"learning_rate": 0.00019575769271720615,
"loss": 0.4201,
"step": 102000
},
{
"epoch": 5.46,
"learning_rate": 0.00019573689709327088,
"loss": 0.4202,
"step": 102500
},
{
"epoch": 5.48,
"learning_rate": 0.00019571610146933561,
"loss": 0.4257,
"step": 103000
},
{
"epoch": 5.51,
"learning_rate": 0.00019569530584540035,
"loss": 0.4409,
"step": 103500
},
{
"epoch": 5.54,
"learning_rate": 0.00019567451022146508,
"loss": 0.4371,
"step": 104000
},
{
"epoch": 5.56,
"learning_rate": 0.0001956537145975298,
"loss": 0.426,
"step": 104500
},
{
"epoch": 5.59,
"learning_rate": 0.00019563291897359454,
"loss": 0.4298,
"step": 105000
},
{
"epoch": 5.62,
"learning_rate": 0.0001956121233496593,
"loss": 0.4349,
"step": 105500
},
{
"epoch": 5.64,
"learning_rate": 0.00019559132772572403,
"loss": 0.4282,
"step": 106000
},
{
"epoch": 5.67,
"learning_rate": 0.00019557053210178876,
"loss": 0.4281,
"step": 106500
},
{
"epoch": 5.7,
"learning_rate": 0.0001955497364778535,
"loss": 0.4389,
"step": 107000
},
{
"epoch": 5.72,
"learning_rate": 0.00019552894085391823,
"loss": 0.4358,
"step": 107500
},
{
"epoch": 5.75,
"learning_rate": 0.000195508145229983,
"loss": 0.4324,
"step": 108000
},
{
"epoch": 5.78,
"learning_rate": 0.00019548734960604772,
"loss": 0.4292,
"step": 108500
},
{
"epoch": 5.8,
"learning_rate": 0.00019546655398211245,
"loss": 0.442,
"step": 109000
},
{
"epoch": 5.83,
"learning_rate": 0.00019544575835817718,
"loss": 0.4322,
"step": 109500
},
{
"epoch": 5.86,
"learning_rate": 0.00019542496273424191,
"loss": 0.4261,
"step": 110000
},
{
"epoch": 5.88,
"learning_rate": 0.00019540416711030665,
"loss": 0.4313,
"step": 110500
},
{
"epoch": 5.91,
"learning_rate": 0.00019538337148637138,
"loss": 0.4275,
"step": 111000
},
{
"epoch": 5.94,
"learning_rate": 0.0001953625758624361,
"loss": 0.4447,
"step": 111500
},
{
"epoch": 5.96,
"learning_rate": 0.00019534178023850087,
"loss": 0.4365,
"step": 112000
},
{
"epoch": 5.99,
"learning_rate": 0.0001953209846145656,
"loss": 0.4339,
"step": 112500
},
{
"epoch": 6.0,
"eval_bleu": 0.3111,
"eval_gen_len": 18.0894,
"eval_loss": 0.4859704375267029,
"eval_runtime": 732.0314,
"eval_samples_per_second": 12.831,
"eval_steps_per_second": 3.209,
"step": 112704
},
{
"epoch": 6.02,
"learning_rate": 0.00019530018899063033,
"loss": 0.4059,
"step": 113000
},
{
"epoch": 6.04,
"learning_rate": 0.00019527939336669506,
"loss": 0.3739,
"step": 113500
},
{
"epoch": 6.07,
"learning_rate": 0.0001952585977427598,
"loss": 0.365,
"step": 114000
},
{
"epoch": 6.1,
"learning_rate": 0.00019523780211882456,
"loss": 0.3764,
"step": 114500
},
{
"epoch": 6.12,
"learning_rate": 0.0001952170064948893,
"loss": 0.3829,
"step": 115000
},
{
"epoch": 6.15,
"learning_rate": 0.00019519621087095402,
"loss": 0.3851,
"step": 115500
},
{
"epoch": 6.18,
"learning_rate": 0.00019517541524701875,
"loss": 0.382,
"step": 116000
},
{
"epoch": 6.2,
"learning_rate": 0.00019515461962308348,
"loss": 0.383,
"step": 116500
},
{
"epoch": 6.23,
"learning_rate": 0.00019513382399914821,
"loss": 0.3926,
"step": 117000
},
{
"epoch": 6.26,
"learning_rate": 0.00019511302837521295,
"loss": 0.3897,
"step": 117500
},
{
"epoch": 6.28,
"learning_rate": 0.00019509223275127768,
"loss": 0.3809,
"step": 118000
},
{
"epoch": 6.31,
"learning_rate": 0.00019507143712734244,
"loss": 0.3964,
"step": 118500
},
{
"epoch": 6.34,
"learning_rate": 0.00019505064150340717,
"loss": 0.3768,
"step": 119000
},
{
"epoch": 6.36,
"learning_rate": 0.0001950298458794719,
"loss": 0.3799,
"step": 119500
},
{
"epoch": 6.39,
"learning_rate": 0.00019500905025553663,
"loss": 0.3896,
"step": 120000
},
{
"epoch": 6.42,
"learning_rate": 0.00019498825463160136,
"loss": 0.3885,
"step": 120500
},
{
"epoch": 6.44,
"learning_rate": 0.0001949674590076661,
"loss": 0.3873,
"step": 121000
},
{
"epoch": 6.47,
"learning_rate": 0.00019494666338373086,
"loss": 0.386,
"step": 121500
},
{
"epoch": 6.49,
"learning_rate": 0.0001949258677597956,
"loss": 0.3842,
"step": 122000
},
{
"epoch": 6.52,
"learning_rate": 0.00019490507213586032,
"loss": 0.396,
"step": 122500
},
{
"epoch": 6.55,
"learning_rate": 0.00019488427651192505,
"loss": 0.3954,
"step": 123000
},
{
"epoch": 6.57,
"learning_rate": 0.00019486348088798978,
"loss": 0.391,
"step": 123500
},
{
"epoch": 6.6,
"learning_rate": 0.00019484268526405451,
"loss": 0.3892,
"step": 124000
},
{
"epoch": 6.63,
"learning_rate": 0.00019482188964011927,
"loss": 0.3992,
"step": 124500
},
{
"epoch": 6.65,
"learning_rate": 0.000194801094016184,
"loss": 0.4063,
"step": 125000
},
{
"epoch": 6.68,
"learning_rate": 0.00019478029839224874,
"loss": 0.4054,
"step": 125500
},
{
"epoch": 6.71,
"learning_rate": 0.00019475950276831347,
"loss": 0.3992,
"step": 126000
},
{
"epoch": 6.73,
"learning_rate": 0.0001947387071443782,
"loss": 0.397,
"step": 126500
},
{
"epoch": 6.76,
"learning_rate": 0.00019471791152044293,
"loss": 0.3971,
"step": 127000
},
{
"epoch": 6.79,
"learning_rate": 0.00019469711589650766,
"loss": 0.401,
"step": 127500
},
{
"epoch": 6.81,
"learning_rate": 0.0001946763202725724,
"loss": 0.402,
"step": 128000
},
{
"epoch": 6.84,
"learning_rate": 0.00019465552464863716,
"loss": 0.4024,
"step": 128500
},
{
"epoch": 6.87,
"learning_rate": 0.0001946347290247019,
"loss": 0.3934,
"step": 129000
},
{
"epoch": 6.89,
"learning_rate": 0.00019461393340076662,
"loss": 0.3928,
"step": 129500
},
{
"epoch": 6.92,
"learning_rate": 0.00019459313777683135,
"loss": 0.4037,
"step": 130000
},
{
"epoch": 6.95,
"learning_rate": 0.00019457234215289608,
"loss": 0.3969,
"step": 130500
},
{
"epoch": 6.97,
"learning_rate": 0.00019455154652896084,
"loss": 0.4045,
"step": 131000
},
{
"epoch": 7.0,
"eval_bleu": 0.3151,
"eval_gen_len": 18.1157,
"eval_loss": 0.4873952865600586,
"eval_runtime": 742.15,
"eval_samples_per_second": 12.656,
"eval_steps_per_second": 3.165,
"step": 131488
}
],
"max_steps": 4808704,
"num_train_epochs": 256,
"total_flos": 6.426846373591757e+16,
"trial_name": null,
"trial_params": null
}