{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"global_step": 143540,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.00029895499512331056,
"loss": 4.0695,
"step": 500
},
{
"epoch": 0.07,
"learning_rate": 0.00029790999024662115,
"loss": 3.433,
"step": 1000
},
{
"epoch": 0.1,
"learning_rate": 0.0002968649853699317,
"loss": 3.198,
"step": 1500
},
{
"epoch": 0.14,
"learning_rate": 0.00029581998049324227,
"loss": 3.1807,
"step": 2000
},
{
"epoch": 0.17,
"learning_rate": 0.00029477497561655286,
"loss": 3.0421,
"step": 2500
},
{
"epoch": 0.21,
"learning_rate": 0.00029372997073986344,
"loss": 3.0883,
"step": 3000
},
{
"epoch": 0.24,
"learning_rate": 0.00029268496586317403,
"loss": 3.0714,
"step": 3500
},
{
"epoch": 0.28,
"learning_rate": 0.00029163996098648457,
"loss": 2.979,
"step": 4000
},
{
"epoch": 0.31,
"learning_rate": 0.00029059495610979515,
"loss": 2.9596,
"step": 4500
},
{
"epoch": 0.35,
"learning_rate": 0.00028954995123310574,
"loss": 2.9377,
"step": 5000
},
{
"epoch": 0.35,
"eval_gen_len": 12.8717,
"eval_loss": 2.5157084465026855,
"eval_rouge1": 54.6148,
"eval_rouge2": 35.1518,
"eval_rougeL": 51.8908,
"eval_rougeLsum": 51.8957,
"eval_runtime": 122.2396,
"eval_samples_per_second": 39.668,
"eval_steps_per_second": 4.966,
"step": 5000
},
{
"epoch": 0.38,
"learning_rate": 0.0002885049463564163,
"loss": 2.9158,
"step": 5500
},
{
"epoch": 0.42,
"learning_rate": 0.0002874599414797269,
"loss": 2.9109,
"step": 6000
},
{
"epoch": 0.45,
"learning_rate": 0.00028641493660303745,
"loss": 2.857,
"step": 6500
},
{
"epoch": 0.49,
"learning_rate": 0.00028536993172634804,
"loss": 2.9129,
"step": 7000
},
{
"epoch": 0.52,
"learning_rate": 0.0002843249268496586,
"loss": 2.8741,
"step": 7500
},
{
"epoch": 0.56,
"learning_rate": 0.00028327992197296916,
"loss": 2.8337,
"step": 8000
},
{
"epoch": 0.59,
"learning_rate": 0.00028223491709627975,
"loss": 2.7539,
"step": 8500
},
{
"epoch": 0.63,
"learning_rate": 0.00028118991221959033,
"loss": 2.8348,
"step": 9000
},
{
"epoch": 0.66,
"learning_rate": 0.0002801449073429009,
"loss": 2.8118,
"step": 9500
},
{
"epoch": 0.7,
"learning_rate": 0.0002790999024662115,
"loss": 2.803,
"step": 10000
},
{
"epoch": 0.7,
"eval_gen_len": 12.7513,
"eval_loss": 2.4086406230926514,
"eval_rouge1": 55.641,
"eval_rouge2": 36.1214,
"eval_rougeL": 52.8683,
"eval_rougeLsum": 52.8572,
"eval_runtime": 124.3863,
"eval_samples_per_second": 38.983,
"eval_steps_per_second": 4.88,
"step": 10000
},
{
"epoch": 0.73,
"learning_rate": 0.00027805489758952204,
"loss": 2.7639,
"step": 10500
},
{
"epoch": 0.77,
"learning_rate": 0.00027700989271283263,
"loss": 2.7668,
"step": 11000
},
{
"epoch": 0.8,
"learning_rate": 0.0002759648878361432,
"loss": 2.7692,
"step": 11500
},
{
"epoch": 0.84,
"learning_rate": 0.0002749198829594538,
"loss": 2.7589,
"step": 12000
},
{
"epoch": 0.87,
"learning_rate": 0.00027387487808276434,
"loss": 2.7104,
"step": 12500
},
{
"epoch": 0.91,
"learning_rate": 0.0002728298732060749,
"loss": 2.7085,
"step": 13000
},
{
"epoch": 0.94,
"learning_rate": 0.0002717848683293855,
"loss": 2.734,
"step": 13500
},
{
"epoch": 0.98,
"learning_rate": 0.0002707398634526961,
"loss": 2.7221,
"step": 14000
},
{
"epoch": 1.01,
"learning_rate": 0.0002696948585760067,
"loss": 2.6748,
"step": 14500
},
{
"epoch": 1.05,
"learning_rate": 0.0002686498536993172,
"loss": 2.5483,
"step": 15000
},
{
"epoch": 1.05,
"eval_gen_len": 12.7754,
"eval_loss": 2.341980218887329,
"eval_rouge1": 55.6604,
"eval_rouge2": 36.0085,
"eval_rougeL": 52.9599,
"eval_rougeLsum": 52.9433,
"eval_runtime": 121.3998,
"eval_samples_per_second": 39.942,
"eval_steps_per_second": 5.0,
"step": 15000
},
{
"epoch": 1.08,
"learning_rate": 0.0002676048488226278,
"loss": 2.5811,
"step": 15500
},
{
"epoch": 1.11,
"learning_rate": 0.0002665598439459384,
"loss": 2.5846,
"step": 16000
},
{
"epoch": 1.15,
"learning_rate": 0.00026551483906924893,
"loss": 2.543,
"step": 16500
},
{
"epoch": 1.18,
"learning_rate": 0.00026446983419255957,
"loss": 2.5682,
"step": 17000
},
{
"epoch": 1.22,
"learning_rate": 0.0002634248293158701,
"loss": 2.5234,
"step": 17500
},
{
"epoch": 1.25,
"learning_rate": 0.0002623798244391807,
"loss": 2.5756,
"step": 18000
},
{
"epoch": 1.29,
"learning_rate": 0.0002613348195624913,
"loss": 2.5471,
"step": 18500
},
{
"epoch": 1.32,
"learning_rate": 0.0002602898146858018,
"loss": 2.5657,
"step": 19000
},
{
"epoch": 1.36,
"learning_rate": 0.00025924480980911245,
"loss": 2.536,
"step": 19500
},
{
"epoch": 1.39,
"learning_rate": 0.000258199804932423,
"loss": 2.4978,
"step": 20000
},
{
"epoch": 1.39,
"eval_gen_len": 12.8804,
"eval_loss": 2.314547061920166,
"eval_rouge1": 56.204,
"eval_rouge2": 36.5896,
"eval_rougeL": 53.338,
"eval_rougeLsum": 53.3351,
"eval_runtime": 117.8229,
"eval_samples_per_second": 41.155,
"eval_steps_per_second": 5.152,
"step": 20000
},
{
"epoch": 1.43,
"learning_rate": 0.0002571548000557336,
"loss": 2.5415,
"step": 20500
},
{
"epoch": 1.46,
"learning_rate": 0.00025610979517904416,
"loss": 2.5634,
"step": 21000
},
{
"epoch": 1.5,
"learning_rate": 0.0002550647903023547,
"loss": 2.5392,
"step": 21500
},
{
"epoch": 1.53,
"learning_rate": 0.00025401978542566534,
"loss": 2.567,
"step": 22000
},
{
"epoch": 1.57,
"learning_rate": 0.00025297478054897587,
"loss": 2.5332,
"step": 22500
},
{
"epoch": 1.6,
"learning_rate": 0.00025192977567228646,
"loss": 2.5615,
"step": 23000
},
{
"epoch": 1.64,
"learning_rate": 0.00025088477079559705,
"loss": 2.5275,
"step": 23500
},
{
"epoch": 1.67,
"learning_rate": 0.0002498397659189076,
"loss": 2.5342,
"step": 24000
},
{
"epoch": 1.71,
"learning_rate": 0.00024879476104221817,
"loss": 2.5248,
"step": 24500
},
{
"epoch": 1.74,
"learning_rate": 0.00024774975616552875,
"loss": 2.5383,
"step": 25000
},
{
"epoch": 1.74,
"eval_gen_len": 12.795,
"eval_loss": 2.2696738243103027,
"eval_rouge1": 56.1356,
"eval_rouge2": 36.6963,
"eval_rougeL": 53.3579,
"eval_rougeLsum": 53.3664,
"eval_runtime": 123.3173,
"eval_samples_per_second": 39.321,
"eval_steps_per_second": 4.922,
"step": 25000
},
{
"epoch": 1.78,
"learning_rate": 0.00024670475128883934,
"loss": 2.4931,
"step": 25500
},
{
"epoch": 1.81,
"learning_rate": 0.00024565974641214993,
"loss": 2.4933,
"step": 26000
},
{
"epoch": 1.85,
"learning_rate": 0.00024461474153546046,
"loss": 2.5195,
"step": 26500
},
{
"epoch": 1.88,
"learning_rate": 0.00024356973665877105,
"loss": 2.5158,
"step": 27000
},
{
"epoch": 1.92,
"learning_rate": 0.00024252473178208164,
"loss": 2.5311,
"step": 27500
},
{
"epoch": 1.95,
"learning_rate": 0.0002414797269053922,
"loss": 2.5037,
"step": 28000
},
{
"epoch": 1.99,
"learning_rate": 0.00024043472202870276,
"loss": 2.5134,
"step": 28500
},
{
"epoch": 2.02,
"learning_rate": 0.00023938971715201337,
"loss": 2.4096,
"step": 29000
},
{
"epoch": 2.06,
"learning_rate": 0.00023834471227532393,
"loss": 2.3211,
"step": 29500
},
{
"epoch": 2.09,
"learning_rate": 0.00023729970739863452,
"loss": 2.3368,
"step": 30000
},
{
"epoch": 2.09,
"eval_gen_len": 12.7478,
"eval_loss": 2.260253667831421,
"eval_rouge1": 56.0271,
"eval_rouge2": 36.4249,
"eval_rougeL": 53.3113,
"eval_rougeLsum": 53.3272,
"eval_runtime": 122.3509,
"eval_samples_per_second": 39.632,
"eval_steps_per_second": 4.961,
"step": 30000
},
{
"epoch": 2.12,
"learning_rate": 0.00023625470252194508,
"loss": 2.3855,
"step": 30500
},
{
"epoch": 2.16,
"learning_rate": 0.00023520969764525564,
"loss": 2.3562,
"step": 31000
},
{
"epoch": 2.19,
"learning_rate": 0.00023416469276856623,
"loss": 2.345,
"step": 31500
},
{
"epoch": 2.23,
"learning_rate": 0.00023311968789187682,
"loss": 2.371,
"step": 32000
},
{
"epoch": 2.26,
"learning_rate": 0.00023207468301518738,
"loss": 2.3653,
"step": 32500
},
{
"epoch": 2.3,
"learning_rate": 0.00023102967813849797,
"loss": 2.3069,
"step": 33000
},
{
"epoch": 2.33,
"learning_rate": 0.00022998467326180853,
"loss": 2.3499,
"step": 33500
},
{
"epoch": 2.37,
"learning_rate": 0.0002289396683851191,
"loss": 2.3479,
"step": 34000
},
{
"epoch": 2.4,
"learning_rate": 0.00022789466350842967,
"loss": 2.3473,
"step": 34500
},
{
"epoch": 2.44,
"learning_rate": 0.00022684965863174026,
"loss": 2.371,
"step": 35000
},
{
"epoch": 2.44,
"eval_gen_len": 12.8243,
"eval_loss": 2.2327780723571777,
"eval_rouge1": 56.5041,
"eval_rouge2": 36.8718,
"eval_rougeL": 53.8064,
"eval_rougeLsum": 53.7995,
"eval_runtime": 122.1132,
"eval_samples_per_second": 39.709,
"eval_steps_per_second": 4.971,
"step": 35000
},
{
"epoch": 2.47,
"learning_rate": 0.00022580465375505085,
"loss": 2.3728,
"step": 35500
},
{
"epoch": 2.51,
"learning_rate": 0.0002247596488783614,
"loss": 2.3489,
"step": 36000
},
{
"epoch": 2.54,
"learning_rate": 0.00022371464400167197,
"loss": 2.4012,
"step": 36500
},
{
"epoch": 2.58,
"learning_rate": 0.00022266963912498256,
"loss": 2.3648,
"step": 37000
},
{
"epoch": 2.61,
"learning_rate": 0.00022162463424829312,
"loss": 2.3887,
"step": 37500
},
{
"epoch": 2.65,
"learning_rate": 0.00022057962937160373,
"loss": 2.3375,
"step": 38000
},
{
"epoch": 2.68,
"learning_rate": 0.0002195346244949143,
"loss": 2.3742,
"step": 38500
},
{
"epoch": 2.72,
"learning_rate": 0.00021848961961822485,
"loss": 2.3393,
"step": 39000
},
{
"epoch": 2.75,
"learning_rate": 0.00021744461474153544,
"loss": 2.327,
"step": 39500
},
{
"epoch": 2.79,
"learning_rate": 0.000216399609864846,
"loss": 2.3567,
"step": 40000
},
{
"epoch": 2.79,
"eval_gen_len": 12.6851,
"eval_loss": 2.207930088043213,
"eval_rouge1": 56.5318,
"eval_rouge2": 36.9437,
"eval_rougeL": 53.8359,
"eval_rougeLsum": 53.8254,
"eval_runtime": 117.7084,
"eval_samples_per_second": 41.195,
"eval_steps_per_second": 5.157,
"step": 40000
},
{
"epoch": 2.82,
"learning_rate": 0.0002153546049881566,
"loss": 2.3582,
"step": 40500
},
{
"epoch": 2.86,
"learning_rate": 0.00021430960011146718,
"loss": 2.3617,
"step": 41000
},
{
"epoch": 2.89,
"learning_rate": 0.00021326459523477774,
"loss": 2.3521,
"step": 41500
},
{
"epoch": 2.93,
"learning_rate": 0.00021221959035808832,
"loss": 2.3396,
"step": 42000
},
{
"epoch": 2.96,
"learning_rate": 0.00021117458548139888,
"loss": 2.3598,
"step": 42500
},
{
"epoch": 3.0,
"learning_rate": 0.00021012958060470945,
"loss": 2.3396,
"step": 43000
},
{
"epoch": 3.03,
"learning_rate": 0.00020908457572802006,
"loss": 2.1784,
"step": 43500
},
{
"epoch": 3.07,
"learning_rate": 0.00020803957085133062,
"loss": 2.2141,
"step": 44000
},
{
"epoch": 3.1,
"learning_rate": 0.0002069945659746412,
"loss": 2.2108,
"step": 44500
},
{
"epoch": 3.14,
"learning_rate": 0.00020594956109795177,
"loss": 2.1753,
"step": 45000
},
{
"epoch": 3.14,
"eval_gen_len": 12.67,
"eval_loss": 2.216768980026245,
"eval_rouge1": 56.3831,
"eval_rouge2": 36.8896,
"eval_rougeL": 53.6542,
"eval_rougeLsum": 53.6708,
"eval_runtime": 116.3323,
"eval_samples_per_second": 41.682,
"eval_steps_per_second": 5.218,
"step": 45000
},
{
"epoch": 3.17,
"learning_rate": 0.00020490455622126233,
"loss": 2.1618,
"step": 45500
},
{
"epoch": 3.2,
"learning_rate": 0.00020385955134457294,
"loss": 2.205,
"step": 46000
},
{
"epoch": 3.24,
"learning_rate": 0.0002028145464678835,
"loss": 2.2087,
"step": 46500
},
{
"epoch": 3.27,
"learning_rate": 0.00020176954159119406,
"loss": 2.1862,
"step": 47000
},
{
"epoch": 3.31,
"learning_rate": 0.00020072453671450465,
"loss": 2.1947,
"step": 47500
},
{
"epoch": 3.34,
"learning_rate": 0.0001996795318378152,
"loss": 2.203,
"step": 48000
},
{
"epoch": 3.38,
"learning_rate": 0.00019863452696112583,
"loss": 2.2225,
"step": 48500
},
{
"epoch": 3.41,
"learning_rate": 0.0001975895220844364,
"loss": 2.2253,
"step": 49000
},
{
"epoch": 3.45,
"learning_rate": 0.00019654451720774695,
"loss": 2.2147,
"step": 49500
},
{
"epoch": 3.48,
"learning_rate": 0.00019549951233105754,
"loss": 2.2069,
"step": 50000
},
{
"epoch": 3.48,
"eval_gen_len": 12.8014,
"eval_loss": 2.2055139541625977,
"eval_rouge1": 56.7171,
"eval_rouge2": 37.1665,
"eval_rougeL": 53.9299,
"eval_rougeLsum": 53.9259,
"eval_runtime": 120.1221,
"eval_samples_per_second": 40.367,
"eval_steps_per_second": 5.053,
"step": 50000
},
{
"epoch": 3.52,
"learning_rate": 0.0001944545074543681,
"loss": 2.2185,
"step": 50500
},
{
"epoch": 3.55,
"learning_rate": 0.00019340950257767866,
"loss": 2.2145,
"step": 51000
},
{
"epoch": 3.59,
"learning_rate": 0.00019236449770098927,
"loss": 2.2661,
"step": 51500
},
{
"epoch": 3.62,
"learning_rate": 0.00019131949282429983,
"loss": 2.2281,
"step": 52000
},
{
"epoch": 3.66,
"learning_rate": 0.00019027448794761042,
"loss": 2.2344,
"step": 52500
},
{
"epoch": 3.69,
"learning_rate": 0.00018922948307092098,
"loss": 2.2283,
"step": 53000
},
{
"epoch": 3.73,
"learning_rate": 0.00018818447819423154,
"loss": 2.2424,
"step": 53500
},
{
"epoch": 3.76,
"learning_rate": 0.00018713947331754215,
"loss": 2.2265,
"step": 54000
},
{
"epoch": 3.8,
"learning_rate": 0.00018609446844085271,
"loss": 2.1928,
"step": 54500
},
{
"epoch": 3.83,
"learning_rate": 0.00018504946356416328,
"loss": 2.2396,
"step": 55000
},
{
"epoch": 3.83,
"eval_gen_len": 12.7989,
"eval_loss": 2.1801397800445557,
"eval_rouge1": 56.936,
"eval_rouge2": 37.5465,
"eval_rougeL": 54.1064,
"eval_rougeLsum": 54.1125,
"eval_runtime": 118.0218,
"eval_samples_per_second": 41.086,
"eval_steps_per_second": 5.143,
"step": 55000
},
{
"epoch": 3.87,
"learning_rate": 0.00018400445868747386,
"loss": 2.2357,
"step": 55500
},
{
"epoch": 3.9,
"learning_rate": 0.00018295945381078442,
"loss": 2.2118,
"step": 56000
},
{
"epoch": 3.94,
"learning_rate": 0.00018191444893409504,
"loss": 2.226,
"step": 56500
},
{
"epoch": 3.97,
"learning_rate": 0.0001808694440574056,
"loss": 2.213,
"step": 57000
},
{
"epoch": 4.01,
"learning_rate": 0.00017982443918071616,
"loss": 2.2106,
"step": 57500
},
{
"epoch": 4.04,
"learning_rate": 0.00017877943430402675,
"loss": 2.0718,
"step": 58000
},
{
"epoch": 4.08,
"learning_rate": 0.0001777344294273373,
"loss": 2.1013,
"step": 58500
},
{
"epoch": 4.11,
"learning_rate": 0.00017668942455064787,
"loss": 2.0646,
"step": 59000
},
{
"epoch": 4.15,
"learning_rate": 0.00017564441967395848,
"loss": 2.102,
"step": 59500
},
{
"epoch": 4.18,
"learning_rate": 0.00017459941479726904,
"loss": 2.0657,
"step": 60000
},
{
"epoch": 4.18,
"eval_gen_len": 12.6987,
"eval_loss": 2.1915152072906494,
"eval_rouge1": 56.6312,
"eval_rouge2": 37.1618,
"eval_rougeL": 53.8646,
"eval_rougeLsum": 53.8791,
"eval_runtime": 116.9251,
"eval_samples_per_second": 41.471,
"eval_steps_per_second": 5.191,
"step": 60000
},
{
"epoch": 4.21,
"learning_rate": 0.00017355440992057963,
"loss": 2.0758,
"step": 60500
},
{
"epoch": 4.25,
"learning_rate": 0.0001725094050438902,
"loss": 2.1113,
"step": 61000
},
{
"epoch": 4.28,
"learning_rate": 0.00017146440016720075,
"loss": 2.1134,
"step": 61500
},
{
"epoch": 4.32,
"learning_rate": 0.00017041939529051134,
"loss": 2.1019,
"step": 62000
},
{
"epoch": 4.35,
"learning_rate": 0.00016937439041382193,
"loss": 2.0924,
"step": 62500
},
{
"epoch": 4.39,
"learning_rate": 0.00016832938553713249,
"loss": 2.1106,
"step": 63000
},
{
"epoch": 4.42,
"learning_rate": 0.00016728438066044307,
"loss": 2.1112,
"step": 63500
},
{
"epoch": 4.46,
"learning_rate": 0.00016623937578375363,
"loss": 2.0951,
"step": 64000
},
{
"epoch": 4.49,
"learning_rate": 0.00016519437090706422,
"loss": 2.1011,
"step": 64500
},
{
"epoch": 4.53,
"learning_rate": 0.0001641493660303748,
"loss": 2.0806,
"step": 65000
},
{
"epoch": 4.53,
"eval_gen_len": 12.715,
"eval_loss": 2.180889844894409,
"eval_rouge1": 56.6599,
"eval_rouge2": 37.1282,
"eval_rougeL": 53.8838,
"eval_rougeLsum": 53.8781,
"eval_runtime": 116.2111,
"eval_samples_per_second": 41.726,
"eval_steps_per_second": 5.223,
"step": 65000
},
{
"epoch": 4.56,
"learning_rate": 0.00016310436115368537,
"loss": 2.1025,
"step": 65500
},
{
"epoch": 4.6,
"learning_rate": 0.00016205935627699596,
"loss": 2.1107,
"step": 66000
},
{
"epoch": 4.63,
"learning_rate": 0.00016101435140030652,
"loss": 2.1125,
"step": 66500
},
{
"epoch": 4.67,
"learning_rate": 0.00015996934652361708,
"loss": 2.0816,
"step": 67000
},
{
"epoch": 4.7,
"learning_rate": 0.00015892434164692767,
"loss": 2.1276,
"step": 67500
},
{
"epoch": 4.74,
"learning_rate": 0.00015787933677023825,
"loss": 2.1125,
"step": 68000
},
{
"epoch": 4.77,
"learning_rate": 0.00015683433189354884,
"loss": 2.1045,
"step": 68500
},
{
"epoch": 4.81,
"learning_rate": 0.0001557893270168594,
"loss": 2.1052,
"step": 69000
},
{
"epoch": 4.84,
"learning_rate": 0.00015474432214016996,
"loss": 2.1224,
"step": 69500
},
{
"epoch": 4.88,
"learning_rate": 0.00015369931726348055,
"loss": 2.0933,
"step": 70000
},
{
"epoch": 4.88,
"eval_gen_len": 12.6593,
"eval_loss": 2.1771466732025146,
"eval_rouge1": 56.5891,
"eval_rouge2": 36.9461,
"eval_rougeL": 53.8058,
"eval_rougeLsum": 53.8087,
"eval_runtime": 119.5299,
"eval_samples_per_second": 40.567,
"eval_steps_per_second": 5.078,
"step": 70000
},
{
"epoch": 4.91,
"learning_rate": 0.0001526543123867911,
"loss": 2.1076,
"step": 70500
},
{
"epoch": 4.95,
"learning_rate": 0.0001516093075101017,
"loss": 2.0704,
"step": 71000
},
{
"epoch": 4.98,
"learning_rate": 0.00015056430263341228,
"loss": 2.0975,
"step": 71500
},
{
"epoch": 5.02,
"learning_rate": 0.00014951929775672285,
"loss": 2.0603,
"step": 72000
},
{
"epoch": 5.05,
"learning_rate": 0.00014847429288003343,
"loss": 1.9328,
"step": 72500
},
{
"epoch": 5.09,
"learning_rate": 0.000147429288003344,
"loss": 1.9873,
"step": 73000
},
{
"epoch": 5.12,
"learning_rate": 0.00014638428312665458,
"loss": 1.9887,
"step": 73500
},
{
"epoch": 5.16,
"learning_rate": 0.00014533927824996514,
"loss": 1.9416,
"step": 74000
},
{
"epoch": 5.19,
"learning_rate": 0.00014429427337327573,
"loss": 2.0074,
"step": 74500
},
{
"epoch": 5.23,
"learning_rate": 0.00014324926849658632,
"loss": 1.9949,
"step": 75000
},
{
"epoch": 5.23,
"eval_gen_len": 12.6723,
"eval_loss": 2.1931562423706055,
"eval_rouge1": 56.4956,
"eval_rouge2": 36.9679,
"eval_rougeL": 53.7634,
"eval_rougeLsum": 53.7731,
"eval_runtime": 122.4799,
"eval_samples_per_second": 39.59,
"eval_steps_per_second": 4.956,
"step": 75000
},
{
"epoch": 5.26,
"learning_rate": 0.00014220426361989688,
"loss": 2.002,
"step": 75500
},
{
"epoch": 5.29,
"learning_rate": 0.00014115925874320744,
"loss": 2.0161,
"step": 76000
},
{
"epoch": 5.33,
"learning_rate": 0.00014011425386651802,
"loss": 2.0139,
"step": 76500
},
{
"epoch": 5.36,
"learning_rate": 0.0001390692489898286,
"loss": 1.9874,
"step": 77000
},
{
"epoch": 5.4,
"learning_rate": 0.0001380242441131392,
"loss": 1.961,
"step": 77500
},
{
"epoch": 5.43,
"learning_rate": 0.00013697923923644976,
"loss": 2.0082,
"step": 78000
},
{
"epoch": 5.47,
"learning_rate": 0.00013593423435976032,
"loss": 2.0175,
"step": 78500
},
{
"epoch": 5.5,
"learning_rate": 0.0001348892294830709,
"loss": 1.987,
"step": 79000
},
{
"epoch": 5.54,
"learning_rate": 0.0001338442246063815,
"loss": 2.0167,
"step": 79500
},
{
"epoch": 5.57,
"learning_rate": 0.00013279921972969206,
"loss": 1.9954,
"step": 80000
},
{
"epoch": 5.57,
"eval_gen_len": 12.6599,
"eval_loss": 2.181297779083252,
"eval_rouge1": 56.4827,
"eval_rouge2": 36.8319,
"eval_rougeL": 53.6397,
"eval_rougeLsum": 53.6399,
"eval_runtime": 121.225,
"eval_samples_per_second": 40.0,
"eval_steps_per_second": 5.007,
"step": 80000
},
{
"epoch": 5.61,
"learning_rate": 0.00013175421485300264,
"loss": 1.9816,
"step": 80500
},
{
"epoch": 5.64,
"learning_rate": 0.0001307092099763132,
"loss": 2.0116,
"step": 81000
},
{
"epoch": 5.68,
"learning_rate": 0.0001296642050996238,
"loss": 2.0182,
"step": 81500
},
{
"epoch": 5.71,
"learning_rate": 0.00012861920022293435,
"loss": 2.0032,
"step": 82000
},
{
"epoch": 5.75,
"learning_rate": 0.00012757419534624494,
"loss": 1.9978,
"step": 82500
},
{
"epoch": 5.78,
"learning_rate": 0.0001265291904695555,
"loss": 2.0083,
"step": 83000
},
{
"epoch": 5.82,
"learning_rate": 0.0001254841855928661,
"loss": 2.0017,
"step": 83500
},
{
"epoch": 5.85,
"learning_rate": 0.00012443918071617665,
"loss": 2.0163,
"step": 84000
},
{
"epoch": 5.89,
"learning_rate": 0.00012339417583948724,
"loss": 2.001,
"step": 84500
},
{
"epoch": 5.92,
"learning_rate": 0.00012234917096279782,
"loss": 1.9912,
"step": 85000
},
{
"epoch": 5.92,
"eval_gen_len": 12.7534,
"eval_loss": 2.1754705905914307,
"eval_rouge1": 56.6723,
"eval_rouge2": 37.0432,
"eval_rougeL": 53.8339,
"eval_rougeLsum": 53.8233,
"eval_runtime": 119.3089,
"eval_samples_per_second": 40.642,
"eval_steps_per_second": 5.088,
"step": 85000
},
{
"epoch": 5.96,
"learning_rate": 0.0001213041660861084,
"loss": 2.0087,
"step": 85500
},
{
"epoch": 5.99,
"learning_rate": 0.00012025916120941897,
"loss": 1.9864,
"step": 86000
},
{
"epoch": 6.03,
"learning_rate": 0.00011921415633272953,
"loss": 1.9169,
"step": 86500
},
{
"epoch": 6.06,
"learning_rate": 0.00011816915145604012,
"loss": 1.8964,
"step": 87000
},
{
"epoch": 6.1,
"learning_rate": 0.00011712414657935069,
"loss": 1.8886,
"step": 87500
},
{
"epoch": 6.13,
"learning_rate": 0.00011607914170266128,
"loss": 1.9246,
"step": 88000
},
{
"epoch": 6.17,
"learning_rate": 0.00011503413682597184,
"loss": 1.9266,
"step": 88500
},
{
"epoch": 6.2,
"learning_rate": 0.00011398913194928242,
"loss": 1.8718,
"step": 89000
},
{
"epoch": 6.24,
"learning_rate": 0.000112944127072593,
"loss": 1.8617,
"step": 89500
},
{
"epoch": 6.27,
"learning_rate": 0.00011189912219590358,
"loss": 1.9068,
"step": 90000
},
{
"epoch": 6.27,
"eval_gen_len": 12.7037,
"eval_loss": 2.184929847717285,
"eval_rouge1": 56.6574,
"eval_rouge2": 37.0691,
"eval_rougeL": 53.9029,
"eval_rougeLsum": 53.892,
"eval_runtime": 117.4531,
"eval_samples_per_second": 41.285,
"eval_steps_per_second": 5.168,
"step": 90000
},
{
"epoch": 6.3,
"learning_rate": 0.00011085411731921414,
"loss": 1.8786,
"step": 90500
},
{
"epoch": 6.34,
"learning_rate": 0.00010980911244252472,
"loss": 1.9071,
"step": 91000
},
{
"epoch": 6.37,
"learning_rate": 0.0001087641075658353,
"loss": 1.8807,
"step": 91500
},
{
"epoch": 6.41,
"learning_rate": 0.00010771910268914589,
"loss": 1.9267,
"step": 92000
},
{
"epoch": 6.44,
"learning_rate": 0.00010667409781245645,
"loss": 1.9136,
"step": 92500
},
{
"epoch": 6.48,
"learning_rate": 0.00010562909293576702,
"loss": 1.9075,
"step": 93000
},
{
"epoch": 6.51,
"learning_rate": 0.00010458408805907761,
"loss": 1.908,
"step": 93500
},
{
"epoch": 6.55,
"learning_rate": 0.00010353908318238818,
"loss": 1.9132,
"step": 94000
},
{
"epoch": 6.58,
"learning_rate": 0.00010249407830569874,
"loss": 1.9013,
"step": 94500
},
{
"epoch": 6.62,
"learning_rate": 0.00010144907342900933,
"loss": 1.9173,
"step": 95000
},
{
"epoch": 6.62,
"eval_gen_len": 12.6467,
"eval_loss": 2.1786956787109375,
"eval_rouge1": 56.5701,
"eval_rouge2": 36.861,
"eval_rougeL": 53.6855,
"eval_rougeLsum": 53.6699,
"eval_runtime": 120.1115,
"eval_samples_per_second": 40.371,
"eval_steps_per_second": 5.054,
"step": 95000
},
{
"epoch": 6.65,
"learning_rate": 0.0001004040685523199,
"loss": 1.9182,
"step": 95500
},
{
"epoch": 6.69,
"learning_rate": 9.935906367563048e-05,
"loss": 1.8904,
"step": 96000
},
{
"epoch": 6.72,
"learning_rate": 9.831405879894105e-05,
"loss": 1.9399,
"step": 96500
},
{
"epoch": 6.76,
"learning_rate": 9.726905392225163e-05,
"loss": 1.9188,
"step": 97000
},
{
"epoch": 6.79,
"learning_rate": 9.622404904556221e-05,
"loss": 1.8742,
"step": 97500
},
{
"epoch": 6.83,
"learning_rate": 9.517904416887279e-05,
"loss": 1.9191,
"step": 98000
},
{
"epoch": 6.86,
"learning_rate": 9.413403929218335e-05,
"loss": 1.9207,
"step": 98500
},
{
"epoch": 6.9,
"learning_rate": 9.308903441549394e-05,
"loss": 1.9416,
"step": 99000
},
{
"epoch": 6.93,
"learning_rate": 9.204402953880451e-05,
"loss": 1.9364,
"step": 99500
},
{
"epoch": 6.97,
"learning_rate": 9.099902466211508e-05,
"loss": 1.9131,
"step": 100000
},
{
"epoch": 6.97,
"eval_gen_len": 12.7072,
"eval_loss": 2.186249017715454,
"eval_rouge1": 56.7175,
"eval_rouge2": 37.0749,
"eval_rougeL": 53.8761,
"eval_rougeLsum": 53.8794,
"eval_runtime": 125.5501,
"eval_samples_per_second": 38.622,
"eval_steps_per_second": 4.835,
"step": 100000
},
{
"epoch": 7.0,
"learning_rate": 8.995401978542566e-05,
"loss": 1.9358,
"step": 100500
},
{
"epoch": 7.04,
"learning_rate": 8.890901490873623e-05,
"loss": 1.8457,
"step": 101000
},
{
"epoch": 7.07,
"learning_rate": 8.78640100320468e-05,
"loss": 1.821,
"step": 101500
},
{
"epoch": 7.11,
"learning_rate": 8.681900515535739e-05,
"loss": 1.8181,
"step": 102000
},
{
"epoch": 7.14,
"learning_rate": 8.577400027866795e-05,
"loss": 1.8372,
"step": 102500
},
{
"epoch": 7.18,
"learning_rate": 8.472899540197853e-05,
"loss": 1.8143,
"step": 103000
},
{
"epoch": 7.21,
"learning_rate": 8.368399052528912e-05,
"loss": 1.8175,
"step": 103500
},
{
"epoch": 7.25,
"learning_rate": 8.263898564859969e-05,
"loss": 1.8245,
"step": 104000
},
{
"epoch": 7.28,
"learning_rate": 8.159398077191025e-05,
"loss": 1.8187,
"step": 104500
},
{
"epoch": 7.32,
"learning_rate": 8.054897589522084e-05,
"loss": 1.8164,
"step": 105000
},
{
"epoch": 7.32,
"eval_gen_len": 12.6364,
"eval_loss": 2.1999216079711914,
"eval_rouge1": 56.6104,
"eval_rouge2": 37.0809,
"eval_rougeL": 53.8098,
"eval_rougeLsum": 53.8216,
"eval_runtime": 117.7818,
"eval_samples_per_second": 41.169,
"eval_steps_per_second": 5.154,
"step": 105000
},
{
"epoch": 7.35,
"learning_rate": 7.950397101853141e-05,
"loss": 1.8402,
"step": 105500
},
{
"epoch": 7.38,
"learning_rate": 7.8458966141842e-05,
"loss": 1.8282,
"step": 106000
},
{
"epoch": 7.42,
"learning_rate": 7.741396126515256e-05,
"loss": 1.811,
"step": 106500
},
{
"epoch": 7.45,
"learning_rate": 7.636895638846313e-05,
"loss": 1.8315,
"step": 107000
},
{
"epoch": 7.49,
"learning_rate": 7.532395151177372e-05,
"loss": 1.8636,
"step": 107500
},
{
"epoch": 7.52,
"learning_rate": 7.42789466350843e-05,
"loss": 1.8146,
"step": 108000
},
{
"epoch": 7.56,
"learning_rate": 7.323394175839487e-05,
"loss": 1.8336,
"step": 108500
},
{
"epoch": 7.59,
"learning_rate": 7.218893688170544e-05,
"loss": 1.8256,
"step": 109000
},
{
"epoch": 7.63,
"learning_rate": 7.114393200501602e-05,
"loss": 1.8438,
"step": 109500
},
{
"epoch": 7.66,
"learning_rate": 7.009892712832659e-05,
"loss": 1.8489,
"step": 110000
},
{
"epoch": 7.66,
"eval_gen_len": 12.5741,
"eval_loss": 2.1944735050201416,
"eval_rouge1": 56.6645,
"eval_rouge2": 37.1267,
"eval_rougeL": 53.9009,
"eval_rougeLsum": 53.9008,
"eval_runtime": 119.7301,
"eval_samples_per_second": 40.499,
"eval_steps_per_second": 5.07,
"step": 110000
},
{
"epoch": 7.7,
"learning_rate": 6.905392225163716e-05,
"loss": 1.7998,
"step": 110500
},
{
"epoch": 7.73,
"learning_rate": 6.800891737494774e-05,
"loss": 1.8382,
"step": 111000
},
{
"epoch": 7.77,
"learning_rate": 6.696391249825833e-05,
"loss": 1.8198,
"step": 111500
},
{
"epoch": 7.8,
"learning_rate": 6.591890762156889e-05,
"loss": 1.8558,
"step": 112000
},
{
"epoch": 7.84,
"learning_rate": 6.487390274487947e-05,
"loss": 1.8395,
"step": 112500
},
{
"epoch": 7.87,
"learning_rate": 6.382889786819005e-05,
"loss": 1.8386,
"step": 113000
},
{
"epoch": 7.91,
"learning_rate": 6.278389299150062e-05,
"loss": 1.8703,
"step": 113500
},
{
"epoch": 7.94,
"learning_rate": 6.17388881148112e-05,
"loss": 1.8576,
"step": 114000
},
{
"epoch": 7.98,
"learning_rate": 6.069388323812178e-05,
"loss": 1.8179,
"step": 114500
},
{
"epoch": 8.01,
"learning_rate": 5.9648878361432344e-05,
"loss": 1.82,
"step": 115000
},
{
"epoch": 8.01,
"eval_gen_len": 12.6428,
"eval_loss": 2.2074716091156006,
"eval_rouge1": 56.6075,
"eval_rouge2": 37.0359,
"eval_rougeL": 53.8792,
"eval_rougeLsum": 53.8833,
"eval_runtime": 121.7726,
"eval_samples_per_second": 39.82,
"eval_steps_per_second": 4.985,
"step": 115000
},
{
"epoch": 8.05,
"learning_rate": 5.8603873484742925e-05,
"loss": 1.7476,
"step": 115500
},
{
"epoch": 8.08,
"learning_rate": 5.75588686080535e-05,
"loss": 1.798,
"step": 116000
},
{
"epoch": 8.12,
"learning_rate": 5.651386373136407e-05,
"loss": 1.7595,
"step": 116500
},
{
"epoch": 8.15,
"learning_rate": 5.546885885467465e-05,
"loss": 1.7551,
"step": 117000
},
{
"epoch": 8.19,
"learning_rate": 5.442385397798523e-05,
"loss": 1.7635,
"step": 117500
},
{
"epoch": 8.22,
"learning_rate": 5.33788491012958e-05,
"loss": 1.7637,
"step": 118000
},
{
"epoch": 8.26,
"learning_rate": 5.2333844224606375e-05,
"loss": 1.7355,
"step": 118500
},
{
"epoch": 8.29,
"learning_rate": 5.128883934791695e-05,
"loss": 1.7711,
"step": 119000
},
{
"epoch": 8.33,
"learning_rate": 5.024383447122753e-05,
"loss": 1.7786,
"step": 119500
},
{
"epoch": 8.36,
"learning_rate": 4.91988295945381e-05,
"loss": 1.772,
"step": 120000
},
{
"epoch": 8.36,
"eval_gen_len": 12.6591,
"eval_loss": 2.2067320346832275,
"eval_rouge1": 56.4716,
"eval_rouge2": 36.8675,
"eval_rougeL": 53.6826,
"eval_rougeLsum": 53.6742,
"eval_runtime": 122.8181,
"eval_samples_per_second": 39.481,
"eval_steps_per_second": 4.942,
"step": 120000
},
{
"epoch": 8.39,
"learning_rate": 4.815382471784868e-05,
"loss": 1.8009,
"step": 120500
},
{
"epoch": 8.43,
"learning_rate": 4.710881984115925e-05,
"loss": 1.7549,
"step": 121000
},
{
"epoch": 8.46,
"learning_rate": 4.606381496446983e-05,
"loss": 1.7607,
"step": 121500
},
{
"epoch": 8.5,
"learning_rate": 4.50188100877804e-05,
"loss": 1.7686,
"step": 122000
},
{
"epoch": 8.53,
"learning_rate": 4.397380521109098e-05,
"loss": 1.7666,
"step": 122500
},
{
"epoch": 8.57,
"learning_rate": 4.2928800334401555e-05,
"loss": 1.7789,
"step": 123000
},
{
"epoch": 8.6,
"learning_rate": 4.1883795457712136e-05,
"loss": 1.7836,
"step": 123500
},
{
"epoch": 8.64,
"learning_rate": 4.08387905810227e-05,
"loss": 1.8019,
"step": 124000
},
{
"epoch": 8.67,
"learning_rate": 3.9793785704333284e-05,
"loss": 1.7651,
"step": 124500
},
{
"epoch": 8.71,
"learning_rate": 3.874878082764386e-05,
"loss": 1.7795,
"step": 125000
},
{
"epoch": 8.71,
"eval_gen_len": 12.608,
"eval_loss": 2.205610990524292,
"eval_rouge1": 56.4112,
"eval_rouge2": 36.9011,
"eval_rougeL": 53.6554,
"eval_rougeLsum": 53.6495,
"eval_runtime": 119.8689,
"eval_samples_per_second": 40.453,
"eval_steps_per_second": 5.064,
"step": 125000
},
{
"epoch": 8.74,
"learning_rate": 3.770377595095443e-05,
"loss": 1.7496,
"step": 125500
},
{
"epoch": 8.78,
"learning_rate": 3.665877107426501e-05,
"loss": 1.7984,
"step": 126000
},
{
"epoch": 8.81,
"learning_rate": 3.5613766197575586e-05,
"loss": 1.781,
"step": 126500
},
{
"epoch": 8.85,
"learning_rate": 3.456876132088616e-05,
"loss": 1.783,
"step": 127000
},
{
"epoch": 8.88,
"learning_rate": 3.3523756444196734e-05,
"loss": 1.7638,
"step": 127500
},
{
"epoch": 8.92,
"learning_rate": 3.2478751567507315e-05,
"loss": 1.7673,
"step": 128000
},
{
"epoch": 8.95,
"learning_rate": 3.143374669081789e-05,
"loss": 1.7821,
"step": 128500
},
{
"epoch": 8.99,
"learning_rate": 3.0388741814128463e-05,
"loss": 1.7632,
"step": 129000
},
{
"epoch": 9.02,
"learning_rate": 2.9343736937439037e-05,
"loss": 1.7608,
"step": 129500
},
{
"epoch": 9.06,
"learning_rate": 2.8298732060749614e-05,
"loss": 1.72,
"step": 130000
},
{
"epoch": 9.06,
"eval_gen_len": 12.6758,
"eval_loss": 2.2197024822235107,
"eval_rouge1": 56.4735,
"eval_rouge2": 36.9255,
"eval_rougeL": 53.6592,
"eval_rougeLsum": 53.6463,
"eval_runtime": 115.7881,
"eval_samples_per_second": 41.878,
"eval_steps_per_second": 5.242,
"step": 130000
},
{
"epoch": 9.09,
"learning_rate": 2.725372718406019e-05,
"loss": 1.7181,
"step": 130500
},
{
"epoch": 9.13,
"learning_rate": 2.6208722307370766e-05,
"loss": 1.6894,
"step": 131000
},
{
"epoch": 9.16,
"learning_rate": 2.516371743068134e-05,
"loss": 1.7012,
"step": 131500
},
{
"epoch": 9.2,
"learning_rate": 2.4118712553991917e-05,
"loss": 1.7261,
"step": 132000
},
{
"epoch": 9.23,
"learning_rate": 2.307370767730249e-05,
"loss": 1.752,
"step": 132500
},
{
"epoch": 9.27,
"learning_rate": 2.202870280061307e-05,
"loss": 1.7228,
"step": 133000
},
{
"epoch": 9.3,
"learning_rate": 2.0983697923923643e-05,
"loss": 1.7194,
"step": 133500
},
{
"epoch": 9.34,
"learning_rate": 1.9938693047234217e-05,
"loss": 1.7077,
"step": 134000
},
{
"epoch": 9.37,
"learning_rate": 1.8893688170544794e-05,
"loss": 1.7183,
"step": 134500
},
{
"epoch": 9.41,
"learning_rate": 1.7848683293855368e-05,
"loss": 1.7174,
"step": 135000
},
{
"epoch": 9.41,
"eval_gen_len": 12.6568,
"eval_loss": 2.216855764389038,
"eval_rouge1": 56.4209,
"eval_rouge2": 36.8139,
"eval_rougeL": 53.5778,
"eval_rougeLsum": 53.5685,
"eval_runtime": 116.7222,
"eval_samples_per_second": 41.543,
"eval_steps_per_second": 5.2,
"step": 135000
},
{
"epoch": 9.44,
"learning_rate": 1.6803678417165945e-05,
"loss": 1.7175,
"step": 135500
},
{
"epoch": 9.47,
"learning_rate": 1.575867354047652e-05,
"loss": 1.7225,
"step": 136000
},
{
"epoch": 9.51,
"learning_rate": 1.4713668663787095e-05,
"loss": 1.7305,
"step": 136500
},
{
"epoch": 9.54,
"learning_rate": 1.366866378709767e-05,
"loss": 1.7352,
"step": 137000
},
{
"epoch": 9.58,
"learning_rate": 1.2623658910408248e-05,
"loss": 1.727,
"step": 137500
},
{
"epoch": 9.61,
"learning_rate": 1.1578654033718824e-05,
"loss": 1.7204,
"step": 138000
},
{
"epoch": 9.65,
"learning_rate": 1.05336491570294e-05,
"loss": 1.7324,
"step": 138500
},
{
"epoch": 9.68,
"learning_rate": 9.488644280339975e-06,
"loss": 1.7106,
"step": 139000
},
{
"epoch": 9.72,
"learning_rate": 8.443639403650549e-06,
"loss": 1.7305,
"step": 139500
},
{
"epoch": 9.75,
"learning_rate": 7.398634526961125e-06,
"loss": 1.7466,
"step": 140000
},
{
"epoch": 9.75,
"eval_gen_len": 12.6416,
"eval_loss": 2.2165005207061768,
"eval_rouge1": 56.3715,
"eval_rouge2": 36.767,
"eval_rougeL": 53.555,
"eval_rougeLsum": 53.5468,
"eval_runtime": 116.5858,
"eval_samples_per_second": 41.592,
"eval_steps_per_second": 5.206,
"step": 140000
},
{
"epoch": 9.79,
"learning_rate": 6.3536296502717004e-06,
"loss": 1.7393,
"step": 140500
},
{
"epoch": 9.82,
"learning_rate": 5.308624773582276e-06,
"loss": 1.7315,
"step": 141000
},
{
"epoch": 9.86,
"learning_rate": 4.263619896892852e-06,
"loss": 1.7313,
"step": 141500
},
{
"epoch": 9.89,
"learning_rate": 3.2186150202034275e-06,
"loss": 1.741,
"step": 142000
},
{
"epoch": 9.93,
"learning_rate": 2.1736101435140028e-06,
"loss": 1.7116,
"step": 142500
},
{
"epoch": 9.96,
"learning_rate": 1.1286052668245784e-06,
"loss": 1.7329,
"step": 143000
},
{
"epoch": 10.0,
"learning_rate": 8.360039013515396e-08,
"loss": 1.736,
"step": 143500
},
{
"epoch": 10.0,
"step": 143540,
"total_flos": 2.0621473641824256e+16,
"train_loss": 2.137477010615576,
"train_runtime": 24602.576,
"train_samples_per_second": 46.674,
"train_steps_per_second": 5.834
}
],
"max_steps": 143540,
"num_train_epochs": 10,
"total_flos": 2.0621473641824256e+16,
"trial_name": null,
"trial_params": null
}