finetuned-indian-food / trainer_state.json
Jakir057's picture
🍻 cheers
1722f1f
{
"best_metric": 0.002558160340413451,
"best_model_checkpoint": "finetuned-indian-food/checkpoint-3600",
"epoch": 4.0,
"eval_steps": 100,
"global_step": 3828,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.00019952978056426334,
"loss": 2.1561,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 0.0001990073145245559,
"loss": 2.0371,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 0.0001984848484848485,
"loss": 1.7093,
"step": 30
},
{
"epoch": 0.04,
"learning_rate": 0.00019796238244514108,
"loss": 1.4792,
"step": 40
},
{
"epoch": 0.05,
"learning_rate": 0.00019743991640543368,
"loss": 1.1808,
"step": 50
},
{
"epoch": 0.06,
"learning_rate": 0.00019691745036572625,
"loss": 0.9715,
"step": 60
},
{
"epoch": 0.07,
"learning_rate": 0.00019639498432601882,
"loss": 0.8425,
"step": 70
},
{
"epoch": 0.08,
"learning_rate": 0.0001958725182863114,
"loss": 0.8624,
"step": 80
},
{
"epoch": 0.09,
"learning_rate": 0.000195350052246604,
"loss": 0.7628,
"step": 90
},
{
"epoch": 0.1,
"learning_rate": 0.00019482758620689657,
"loss": 0.7056,
"step": 100
},
{
"epoch": 0.1,
"eval_accuracy": 0.8881481481481481,
"eval_loss": 0.5113354325294495,
"eval_runtime": 21.3674,
"eval_samples_per_second": 126.361,
"eval_steps_per_second": 15.818,
"step": 100
},
{
"epoch": 0.11,
"learning_rate": 0.00019430512016718914,
"loss": 0.6112,
"step": 110
},
{
"epoch": 0.13,
"learning_rate": 0.00019378265412748174,
"loss": 0.5859,
"step": 120
},
{
"epoch": 0.14,
"learning_rate": 0.0001932601880877743,
"loss": 0.5764,
"step": 130
},
{
"epoch": 0.15,
"learning_rate": 0.00019273772204806688,
"loss": 0.5528,
"step": 140
},
{
"epoch": 0.16,
"learning_rate": 0.00019221525600835945,
"loss": 0.4072,
"step": 150
},
{
"epoch": 0.17,
"learning_rate": 0.00019169278996865205,
"loss": 0.388,
"step": 160
},
{
"epoch": 0.18,
"learning_rate": 0.00019117032392894462,
"loss": 0.399,
"step": 170
},
{
"epoch": 0.19,
"learning_rate": 0.0001906478578892372,
"loss": 0.2888,
"step": 180
},
{
"epoch": 0.2,
"learning_rate": 0.0001901253918495298,
"loss": 0.3242,
"step": 190
},
{
"epoch": 0.21,
"learning_rate": 0.00018960292580982237,
"loss": 0.3027,
"step": 200
},
{
"epoch": 0.21,
"eval_accuracy": 0.9796296296296296,
"eval_loss": 0.12799212336540222,
"eval_runtime": 16.2334,
"eval_samples_per_second": 166.323,
"eval_steps_per_second": 20.821,
"step": 200
},
{
"epoch": 0.22,
"learning_rate": 0.00018908045977011494,
"loss": 0.2855,
"step": 210
},
{
"epoch": 0.23,
"learning_rate": 0.00018855799373040754,
"loss": 0.3776,
"step": 220
},
{
"epoch": 0.24,
"learning_rate": 0.0001880355276907001,
"loss": 0.2458,
"step": 230
},
{
"epoch": 0.25,
"learning_rate": 0.0001875130616509927,
"loss": 0.3555,
"step": 240
},
{
"epoch": 0.26,
"learning_rate": 0.00018699059561128528,
"loss": 0.4411,
"step": 250
},
{
"epoch": 0.27,
"learning_rate": 0.00018646812957157788,
"loss": 0.1867,
"step": 260
},
{
"epoch": 0.28,
"learning_rate": 0.00018594566353187045,
"loss": 0.5165,
"step": 270
},
{
"epoch": 0.29,
"learning_rate": 0.00018542319749216302,
"loss": 0.3386,
"step": 280
},
{
"epoch": 0.3,
"learning_rate": 0.0001849007314524556,
"loss": 0.2614,
"step": 290
},
{
"epoch": 0.31,
"learning_rate": 0.0001843782654127482,
"loss": 0.2823,
"step": 300
},
{
"epoch": 0.31,
"eval_accuracy": 0.9655555555555555,
"eval_loss": 0.157969132065773,
"eval_runtime": 16.9126,
"eval_samples_per_second": 159.645,
"eval_steps_per_second": 19.985,
"step": 300
},
{
"epoch": 0.32,
"learning_rate": 0.00018385579937304076,
"loss": 0.3546,
"step": 310
},
{
"epoch": 0.33,
"learning_rate": 0.00018333333333333334,
"loss": 0.2974,
"step": 320
},
{
"epoch": 0.34,
"learning_rate": 0.00018281086729362593,
"loss": 0.22,
"step": 330
},
{
"epoch": 0.36,
"learning_rate": 0.0001822884012539185,
"loss": 0.3668,
"step": 340
},
{
"epoch": 0.37,
"learning_rate": 0.00018176593521421108,
"loss": 0.2249,
"step": 350
},
{
"epoch": 0.38,
"learning_rate": 0.00018124346917450365,
"loss": 0.1652,
"step": 360
},
{
"epoch": 0.39,
"learning_rate": 0.00018072100313479625,
"loss": 0.1557,
"step": 370
},
{
"epoch": 0.4,
"learning_rate": 0.00018019853709508882,
"loss": 0.1811,
"step": 380
},
{
"epoch": 0.41,
"learning_rate": 0.0001796760710553814,
"loss": 0.189,
"step": 390
},
{
"epoch": 0.42,
"learning_rate": 0.000179153605015674,
"loss": 0.3273,
"step": 400
},
{
"epoch": 0.42,
"eval_accuracy": 0.9837037037037037,
"eval_loss": 0.0879134014248848,
"eval_runtime": 16.3211,
"eval_samples_per_second": 165.43,
"eval_steps_per_second": 20.709,
"step": 400
},
{
"epoch": 0.43,
"learning_rate": 0.00017863113897596656,
"loss": 0.1842,
"step": 410
},
{
"epoch": 0.44,
"learning_rate": 0.00017810867293625913,
"loss": 0.2524,
"step": 420
},
{
"epoch": 0.45,
"learning_rate": 0.00017758620689655173,
"loss": 0.1823,
"step": 430
},
{
"epoch": 0.46,
"learning_rate": 0.0001770637408568443,
"loss": 0.1912,
"step": 440
},
{
"epoch": 0.47,
"learning_rate": 0.0001765412748171369,
"loss": 0.2241,
"step": 450
},
{
"epoch": 0.48,
"learning_rate": 0.00017601880877742948,
"loss": 0.2356,
"step": 460
},
{
"epoch": 0.49,
"learning_rate": 0.00017549634273772208,
"loss": 0.196,
"step": 470
},
{
"epoch": 0.5,
"learning_rate": 0.00017497387669801465,
"loss": 0.1365,
"step": 480
},
{
"epoch": 0.51,
"learning_rate": 0.00017445141065830722,
"loss": 0.2863,
"step": 490
},
{
"epoch": 0.52,
"learning_rate": 0.00017392894461859982,
"loss": 0.1808,
"step": 500
},
{
"epoch": 0.52,
"eval_accuracy": 0.9822222222222222,
"eval_loss": 0.08116234838962555,
"eval_runtime": 16.1751,
"eval_samples_per_second": 166.923,
"eval_steps_per_second": 20.896,
"step": 500
},
{
"epoch": 0.53,
"learning_rate": 0.0001734064785788924,
"loss": 0.2045,
"step": 510
},
{
"epoch": 0.54,
"learning_rate": 0.00017288401253918496,
"loss": 0.1195,
"step": 520
},
{
"epoch": 0.55,
"learning_rate": 0.00017236154649947753,
"loss": 0.1668,
"step": 530
},
{
"epoch": 0.56,
"learning_rate": 0.00017183908045977013,
"loss": 0.2419,
"step": 540
},
{
"epoch": 0.57,
"learning_rate": 0.0001713166144200627,
"loss": 0.2098,
"step": 550
},
{
"epoch": 0.59,
"learning_rate": 0.00017079414838035528,
"loss": 0.1827,
"step": 560
},
{
"epoch": 0.6,
"learning_rate": 0.00017027168234064787,
"loss": 0.184,
"step": 570
},
{
"epoch": 0.61,
"learning_rate": 0.00016974921630094045,
"loss": 0.2049,
"step": 580
},
{
"epoch": 0.62,
"learning_rate": 0.00016922675026123302,
"loss": 0.118,
"step": 590
},
{
"epoch": 0.63,
"learning_rate": 0.0001687042842215256,
"loss": 0.2101,
"step": 600
},
{
"epoch": 0.63,
"eval_accuracy": 0.9937037037037038,
"eval_loss": 0.03391774371266365,
"eval_runtime": 16.2812,
"eval_samples_per_second": 165.835,
"eval_steps_per_second": 20.76,
"step": 600
},
{
"epoch": 0.64,
"learning_rate": 0.0001681818181818182,
"loss": 0.1021,
"step": 610
},
{
"epoch": 0.65,
"learning_rate": 0.00016765935214211076,
"loss": 0.3133,
"step": 620
},
{
"epoch": 0.66,
"learning_rate": 0.00016713688610240333,
"loss": 0.3214,
"step": 630
},
{
"epoch": 0.67,
"learning_rate": 0.00016661442006269593,
"loss": 0.232,
"step": 640
},
{
"epoch": 0.68,
"learning_rate": 0.0001660919540229885,
"loss": 0.0819,
"step": 650
},
{
"epoch": 0.69,
"learning_rate": 0.0001655694879832811,
"loss": 0.1256,
"step": 660
},
{
"epoch": 0.7,
"learning_rate": 0.00016504702194357367,
"loss": 0.1577,
"step": 670
},
{
"epoch": 0.71,
"learning_rate": 0.00016452455590386627,
"loss": 0.0825,
"step": 680
},
{
"epoch": 0.72,
"learning_rate": 0.00016400208986415885,
"loss": 0.1146,
"step": 690
},
{
"epoch": 0.73,
"learning_rate": 0.00016347962382445142,
"loss": 0.1495,
"step": 700
},
{
"epoch": 0.73,
"eval_accuracy": 0.9833333333333333,
"eval_loss": 0.0568382553756237,
"eval_runtime": 16.2547,
"eval_samples_per_second": 166.106,
"eval_steps_per_second": 20.794,
"step": 700
},
{
"epoch": 0.74,
"learning_rate": 0.00016295715778474402,
"loss": 0.197,
"step": 710
},
{
"epoch": 0.75,
"learning_rate": 0.0001624346917450366,
"loss": 0.1708,
"step": 720
},
{
"epoch": 0.76,
"learning_rate": 0.00016191222570532916,
"loss": 0.1702,
"step": 730
},
{
"epoch": 0.77,
"learning_rate": 0.00016138975966562173,
"loss": 0.3196,
"step": 740
},
{
"epoch": 0.78,
"learning_rate": 0.00016086729362591433,
"loss": 0.1967,
"step": 750
},
{
"epoch": 0.79,
"learning_rate": 0.0001603448275862069,
"loss": 0.1943,
"step": 760
},
{
"epoch": 0.8,
"learning_rate": 0.00015982236154649947,
"loss": 0.1489,
"step": 770
},
{
"epoch": 0.82,
"learning_rate": 0.00015929989550679207,
"loss": 0.1284,
"step": 780
},
{
"epoch": 0.83,
"learning_rate": 0.00015877742946708464,
"loss": 0.2296,
"step": 790
},
{
"epoch": 0.84,
"learning_rate": 0.00015825496342737722,
"loss": 0.1296,
"step": 800
},
{
"epoch": 0.84,
"eval_accuracy": 0.9844444444444445,
"eval_loss": 0.06289423257112503,
"eval_runtime": 16.2541,
"eval_samples_per_second": 166.112,
"eval_steps_per_second": 20.795,
"step": 800
},
{
"epoch": 0.85,
"learning_rate": 0.00015773249738766982,
"loss": 0.0969,
"step": 810
},
{
"epoch": 0.86,
"learning_rate": 0.0001572100313479624,
"loss": 0.1199,
"step": 820
},
{
"epoch": 0.87,
"learning_rate": 0.00015668756530825496,
"loss": 0.1064,
"step": 830
},
{
"epoch": 0.88,
"learning_rate": 0.00015616509926854753,
"loss": 0.2688,
"step": 840
},
{
"epoch": 0.89,
"learning_rate": 0.00015564263322884013,
"loss": 0.2549,
"step": 850
},
{
"epoch": 0.9,
"learning_rate": 0.0001551201671891327,
"loss": 0.1788,
"step": 860
},
{
"epoch": 0.91,
"learning_rate": 0.0001545977011494253,
"loss": 0.0653,
"step": 870
},
{
"epoch": 0.92,
"learning_rate": 0.00015407523510971787,
"loss": 0.3013,
"step": 880
},
{
"epoch": 0.93,
"learning_rate": 0.00015355276907001047,
"loss": 0.167,
"step": 890
},
{
"epoch": 0.94,
"learning_rate": 0.00015303030303030304,
"loss": 0.1462,
"step": 900
},
{
"epoch": 0.94,
"eval_accuracy": 0.9733333333333334,
"eval_loss": 0.08862249553203583,
"eval_runtime": 16.3153,
"eval_samples_per_second": 165.489,
"eval_steps_per_second": 20.717,
"step": 900
},
{
"epoch": 0.95,
"learning_rate": 0.00015250783699059562,
"loss": 0.1513,
"step": 910
},
{
"epoch": 0.96,
"learning_rate": 0.00015198537095088821,
"loss": 0.1546,
"step": 920
},
{
"epoch": 0.97,
"learning_rate": 0.00015146290491118079,
"loss": 0.0779,
"step": 930
},
{
"epoch": 0.98,
"learning_rate": 0.00015094043887147336,
"loss": 0.1373,
"step": 940
},
{
"epoch": 0.99,
"learning_rate": 0.00015041797283176596,
"loss": 0.0776,
"step": 950
},
{
"epoch": 1.0,
"learning_rate": 0.00014989550679205853,
"loss": 0.0995,
"step": 960
},
{
"epoch": 1.01,
"learning_rate": 0.0001493730407523511,
"loss": 0.1783,
"step": 970
},
{
"epoch": 1.02,
"learning_rate": 0.00014885057471264367,
"loss": 0.1035,
"step": 980
},
{
"epoch": 1.03,
"learning_rate": 0.00014832810867293627,
"loss": 0.1682,
"step": 990
},
{
"epoch": 1.04,
"learning_rate": 0.00014780564263322884,
"loss": 0.0519,
"step": 1000
},
{
"epoch": 1.04,
"eval_accuracy": 0.987037037037037,
"eval_loss": 0.054442230612039566,
"eval_runtime": 16.2404,
"eval_samples_per_second": 166.252,
"eval_steps_per_second": 20.812,
"step": 1000
},
{
"epoch": 1.06,
"learning_rate": 0.00014728317659352141,
"loss": 0.1676,
"step": 1010
},
{
"epoch": 1.07,
"learning_rate": 0.000146760710553814,
"loss": 0.2244,
"step": 1020
},
{
"epoch": 1.08,
"learning_rate": 0.00014623824451410659,
"loss": 0.1305,
"step": 1030
},
{
"epoch": 1.09,
"learning_rate": 0.00014571577847439916,
"loss": 0.126,
"step": 1040
},
{
"epoch": 1.1,
"learning_rate": 0.00014519331243469173,
"loss": 0.1695,
"step": 1050
},
{
"epoch": 1.11,
"learning_rate": 0.00014467084639498433,
"loss": 0.2042,
"step": 1060
},
{
"epoch": 1.12,
"learning_rate": 0.0001441483803552769,
"loss": 0.1022,
"step": 1070
},
{
"epoch": 1.13,
"learning_rate": 0.0001436259143155695,
"loss": 0.1836,
"step": 1080
},
{
"epoch": 1.14,
"learning_rate": 0.0001431034482758621,
"loss": 0.1896,
"step": 1090
},
{
"epoch": 1.15,
"learning_rate": 0.00014258098223615467,
"loss": 0.3192,
"step": 1100
},
{
"epoch": 1.15,
"eval_accuracy": 0.9725925925925926,
"eval_loss": 0.08920213580131531,
"eval_runtime": 16.309,
"eval_samples_per_second": 165.552,
"eval_steps_per_second": 20.725,
"step": 1100
},
{
"epoch": 1.16,
"learning_rate": 0.00014205851619644724,
"loss": 0.0961,
"step": 1110
},
{
"epoch": 1.17,
"learning_rate": 0.0001415360501567398,
"loss": 0.0903,
"step": 1120
},
{
"epoch": 1.18,
"learning_rate": 0.0001410135841170324,
"loss": 0.133,
"step": 1130
},
{
"epoch": 1.19,
"learning_rate": 0.00014049111807732498,
"loss": 0.1073,
"step": 1140
},
{
"epoch": 1.2,
"learning_rate": 0.00013996865203761756,
"loss": 0.1116,
"step": 1150
},
{
"epoch": 1.21,
"learning_rate": 0.00013944618599791015,
"loss": 0.1372,
"step": 1160
},
{
"epoch": 1.22,
"learning_rate": 0.00013892371995820273,
"loss": 0.1428,
"step": 1170
},
{
"epoch": 1.23,
"learning_rate": 0.0001384012539184953,
"loss": 0.1059,
"step": 1180
},
{
"epoch": 1.24,
"learning_rate": 0.0001378787878787879,
"loss": 0.0891,
"step": 1190
},
{
"epoch": 1.25,
"learning_rate": 0.00013735632183908047,
"loss": 0.158,
"step": 1200
},
{
"epoch": 1.25,
"eval_accuracy": 0.98,
"eval_loss": 0.06320571154356003,
"eval_runtime": 16.4346,
"eval_samples_per_second": 164.287,
"eval_steps_per_second": 20.566,
"step": 1200
},
{
"epoch": 1.26,
"learning_rate": 0.00013683385579937304,
"loss": 0.0856,
"step": 1210
},
{
"epoch": 1.27,
"learning_rate": 0.0001363113897596656,
"loss": 0.1741,
"step": 1220
},
{
"epoch": 1.29,
"learning_rate": 0.0001357889237199582,
"loss": 0.1105,
"step": 1230
},
{
"epoch": 1.3,
"learning_rate": 0.00013526645768025078,
"loss": 0.2382,
"step": 1240
},
{
"epoch": 1.31,
"learning_rate": 0.00013474399164054336,
"loss": 0.1309,
"step": 1250
},
{
"epoch": 1.32,
"learning_rate": 0.00013422152560083595,
"loss": 0.0947,
"step": 1260
},
{
"epoch": 1.33,
"learning_rate": 0.00013369905956112853,
"loss": 0.12,
"step": 1270
},
{
"epoch": 1.34,
"learning_rate": 0.00013317659352142112,
"loss": 0.0805,
"step": 1280
},
{
"epoch": 1.35,
"learning_rate": 0.0001326541274817137,
"loss": 0.1211,
"step": 1290
},
{
"epoch": 1.36,
"learning_rate": 0.0001321316614420063,
"loss": 0.0266,
"step": 1300
},
{
"epoch": 1.36,
"eval_accuracy": 0.9944444444444445,
"eval_loss": 0.02332974411547184,
"eval_runtime": 16.2062,
"eval_samples_per_second": 166.603,
"eval_steps_per_second": 20.856,
"step": 1300
},
{
"epoch": 1.37,
"learning_rate": 0.00013160919540229887,
"loss": 0.1716,
"step": 1310
},
{
"epoch": 1.38,
"learning_rate": 0.00013108672936259144,
"loss": 0.1854,
"step": 1320
},
{
"epoch": 1.39,
"learning_rate": 0.00013056426332288404,
"loss": 0.0916,
"step": 1330
},
{
"epoch": 1.4,
"learning_rate": 0.0001300417972831766,
"loss": 0.0378,
"step": 1340
},
{
"epoch": 1.41,
"learning_rate": 0.00012951933124346918,
"loss": 0.0995,
"step": 1350
},
{
"epoch": 1.42,
"learning_rate": 0.00012899686520376175,
"loss": 0.0803,
"step": 1360
},
{
"epoch": 1.43,
"learning_rate": 0.00012847439916405435,
"loss": 0.0585,
"step": 1370
},
{
"epoch": 1.44,
"learning_rate": 0.00012795193312434692,
"loss": 0.1277,
"step": 1380
},
{
"epoch": 1.45,
"learning_rate": 0.0001274294670846395,
"loss": 0.0938,
"step": 1390
},
{
"epoch": 1.46,
"learning_rate": 0.0001269070010449321,
"loss": 0.1832,
"step": 1400
},
{
"epoch": 1.46,
"eval_accuracy": 0.9929629629629629,
"eval_loss": 0.029181810095906258,
"eval_runtime": 16.5252,
"eval_samples_per_second": 163.387,
"eval_steps_per_second": 20.454,
"step": 1400
},
{
"epoch": 1.47,
"learning_rate": 0.00012638453500522467,
"loss": 0.1317,
"step": 1410
},
{
"epoch": 1.48,
"learning_rate": 0.00012586206896551724,
"loss": 0.176,
"step": 1420
},
{
"epoch": 1.49,
"learning_rate": 0.0001253396029258098,
"loss": 0.0925,
"step": 1430
},
{
"epoch": 1.5,
"learning_rate": 0.0001248171368861024,
"loss": 0.159,
"step": 1440
},
{
"epoch": 1.52,
"learning_rate": 0.00012429467084639498,
"loss": 0.1374,
"step": 1450
},
{
"epoch": 1.53,
"learning_rate": 0.00012377220480668755,
"loss": 0.2189,
"step": 1460
},
{
"epoch": 1.54,
"learning_rate": 0.00012324973876698015,
"loss": 0.1034,
"step": 1470
},
{
"epoch": 1.55,
"learning_rate": 0.00012272727272727272,
"loss": 0.0965,
"step": 1480
},
{
"epoch": 1.56,
"learning_rate": 0.00012220480668756532,
"loss": 0.2171,
"step": 1490
},
{
"epoch": 1.57,
"learning_rate": 0.00012168234064785788,
"loss": 0.1212,
"step": 1500
},
{
"epoch": 1.57,
"eval_accuracy": 0.9851851851851852,
"eval_loss": 0.04892929643392563,
"eval_runtime": 16.2659,
"eval_samples_per_second": 165.991,
"eval_steps_per_second": 20.78,
"step": 1500
},
{
"epoch": 1.58,
"learning_rate": 0.00012115987460815048,
"loss": 0.125,
"step": 1510
},
{
"epoch": 1.59,
"learning_rate": 0.00012063740856844305,
"loss": 0.0725,
"step": 1520
},
{
"epoch": 1.6,
"learning_rate": 0.00012011494252873562,
"loss": 0.149,
"step": 1530
},
{
"epoch": 1.61,
"learning_rate": 0.00011959247648902822,
"loss": 0.1072,
"step": 1540
},
{
"epoch": 1.62,
"learning_rate": 0.00011907001044932081,
"loss": 0.0786,
"step": 1550
},
{
"epoch": 1.63,
"learning_rate": 0.00011854754440961338,
"loss": 0.1061,
"step": 1560
},
{
"epoch": 1.64,
"learning_rate": 0.00011802507836990598,
"loss": 0.26,
"step": 1570
},
{
"epoch": 1.65,
"learning_rate": 0.00011750261233019855,
"loss": 0.0717,
"step": 1580
},
{
"epoch": 1.66,
"learning_rate": 0.00011698014629049112,
"loss": 0.1236,
"step": 1590
},
{
"epoch": 1.67,
"learning_rate": 0.0001164576802507837,
"loss": 0.0994,
"step": 1600
},
{
"epoch": 1.67,
"eval_accuracy": 0.9974074074074074,
"eval_loss": 0.014160659164190292,
"eval_runtime": 16.4275,
"eval_samples_per_second": 164.358,
"eval_steps_per_second": 20.575,
"step": 1600
},
{
"epoch": 1.68,
"learning_rate": 0.00011593521421107629,
"loss": 0.1335,
"step": 1610
},
{
"epoch": 1.69,
"learning_rate": 0.00011541274817136886,
"loss": 0.1385,
"step": 1620
},
{
"epoch": 1.7,
"learning_rate": 0.00011489028213166144,
"loss": 0.0761,
"step": 1630
},
{
"epoch": 1.71,
"learning_rate": 0.00011436781609195404,
"loss": 0.1571,
"step": 1640
},
{
"epoch": 1.72,
"learning_rate": 0.00011384535005224661,
"loss": 0.046,
"step": 1650
},
{
"epoch": 1.73,
"learning_rate": 0.00011332288401253919,
"loss": 0.1345,
"step": 1660
},
{
"epoch": 1.75,
"learning_rate": 0.00011280041797283176,
"loss": 0.0945,
"step": 1670
},
{
"epoch": 1.76,
"learning_rate": 0.00011227795193312436,
"loss": 0.1222,
"step": 1680
},
{
"epoch": 1.77,
"learning_rate": 0.00011175548589341694,
"loss": 0.1118,
"step": 1690
},
{
"epoch": 1.78,
"learning_rate": 0.00011123301985370951,
"loss": 0.0219,
"step": 1700
},
{
"epoch": 1.78,
"eval_accuracy": 0.9929629629629629,
"eval_loss": 0.02774985507130623,
"eval_runtime": 16.27,
"eval_samples_per_second": 165.949,
"eval_steps_per_second": 20.774,
"step": 1700
},
{
"epoch": 1.79,
"learning_rate": 0.0001107105538140021,
"loss": 0.1182,
"step": 1710
},
{
"epoch": 1.8,
"learning_rate": 0.00011018808777429468,
"loss": 0.0108,
"step": 1720
},
{
"epoch": 1.81,
"learning_rate": 0.00010966562173458725,
"loss": 0.0983,
"step": 1730
},
{
"epoch": 1.82,
"learning_rate": 0.00010914315569487982,
"loss": 0.0752,
"step": 1740
},
{
"epoch": 1.83,
"learning_rate": 0.00010862068965517242,
"loss": 0.0566,
"step": 1750
},
{
"epoch": 1.84,
"learning_rate": 0.000108098223615465,
"loss": 0.088,
"step": 1760
},
{
"epoch": 1.85,
"learning_rate": 0.00010757575757575758,
"loss": 0.1686,
"step": 1770
},
{
"epoch": 1.86,
"learning_rate": 0.00010705329153605018,
"loss": 0.0176,
"step": 1780
},
{
"epoch": 1.87,
"learning_rate": 0.00010653082549634275,
"loss": 0.0688,
"step": 1790
},
{
"epoch": 1.88,
"learning_rate": 0.00010600835945663532,
"loss": 0.0664,
"step": 1800
},
{
"epoch": 1.88,
"eval_accuracy": 0.9974074074074074,
"eval_loss": 0.015811724588274956,
"eval_runtime": 16.2257,
"eval_samples_per_second": 166.402,
"eval_steps_per_second": 20.831,
"step": 1800
},
{
"epoch": 1.89,
"learning_rate": 0.00010548589341692789,
"loss": 0.1437,
"step": 1810
},
{
"epoch": 1.9,
"learning_rate": 0.00010496342737722049,
"loss": 0.0091,
"step": 1820
},
{
"epoch": 1.91,
"learning_rate": 0.00010444096133751306,
"loss": 0.109,
"step": 1830
},
{
"epoch": 1.92,
"learning_rate": 0.00010391849529780563,
"loss": 0.0641,
"step": 1840
},
{
"epoch": 1.93,
"learning_rate": 0.00010339602925809823,
"loss": 0.112,
"step": 1850
},
{
"epoch": 1.94,
"learning_rate": 0.0001028735632183908,
"loss": 0.0732,
"step": 1860
},
{
"epoch": 1.95,
"learning_rate": 0.00010235109717868339,
"loss": 0.1322,
"step": 1870
},
{
"epoch": 1.96,
"learning_rate": 0.00010182863113897596,
"loss": 0.046,
"step": 1880
},
{
"epoch": 1.97,
"learning_rate": 0.00010130616509926856,
"loss": 0.1062,
"step": 1890
},
{
"epoch": 1.99,
"learning_rate": 0.00010078369905956113,
"loss": 0.0834,
"step": 1900
},
{
"epoch": 1.99,
"eval_accuracy": 0.9977777777777778,
"eval_loss": 0.012378275394439697,
"eval_runtime": 16.3569,
"eval_samples_per_second": 165.068,
"eval_steps_per_second": 20.664,
"step": 1900
},
{
"epoch": 2.0,
"learning_rate": 0.0001002612330198537,
"loss": 0.0779,
"step": 1910
},
{
"epoch": 2.01,
"learning_rate": 9.973876698014629e-05,
"loss": 0.061,
"step": 1920
},
{
"epoch": 2.02,
"learning_rate": 9.921630094043888e-05,
"loss": 0.0608,
"step": 1930
},
{
"epoch": 2.03,
"learning_rate": 9.869383490073145e-05,
"loss": 0.0975,
"step": 1940
},
{
"epoch": 2.04,
"learning_rate": 9.817136886102403e-05,
"loss": 0.0532,
"step": 1950
},
{
"epoch": 2.05,
"learning_rate": 9.764890282131662e-05,
"loss": 0.1071,
"step": 1960
},
{
"epoch": 2.06,
"learning_rate": 9.71264367816092e-05,
"loss": 0.0721,
"step": 1970
},
{
"epoch": 2.07,
"learning_rate": 9.660397074190179e-05,
"loss": 0.1187,
"step": 1980
},
{
"epoch": 2.08,
"learning_rate": 9.608150470219436e-05,
"loss": 0.0543,
"step": 1990
},
{
"epoch": 2.09,
"learning_rate": 9.555903866248695e-05,
"loss": 0.1093,
"step": 2000
},
{
"epoch": 2.09,
"eval_accuracy": 0.9974074074074074,
"eval_loss": 0.013967631384730339,
"eval_runtime": 16.4607,
"eval_samples_per_second": 164.027,
"eval_steps_per_second": 20.534,
"step": 2000
},
{
"epoch": 2.1,
"learning_rate": 9.503657262277952e-05,
"loss": 0.0528,
"step": 2010
},
{
"epoch": 2.11,
"learning_rate": 9.45141065830721e-05,
"loss": 0.0391,
"step": 2020
},
{
"epoch": 2.12,
"learning_rate": 9.399164054336469e-05,
"loss": 0.0918,
"step": 2030
},
{
"epoch": 2.13,
"learning_rate": 9.346917450365726e-05,
"loss": 0.0567,
"step": 2040
},
{
"epoch": 2.14,
"learning_rate": 9.294670846394985e-05,
"loss": 0.0128,
"step": 2050
},
{
"epoch": 2.15,
"learning_rate": 9.242424242424242e-05,
"loss": 0.0776,
"step": 2060
},
{
"epoch": 2.16,
"learning_rate": 9.1901776384535e-05,
"loss": 0.0597,
"step": 2070
},
{
"epoch": 2.17,
"learning_rate": 9.137931034482759e-05,
"loss": 0.0866,
"step": 2080
},
{
"epoch": 2.18,
"learning_rate": 9.085684430512017e-05,
"loss": 0.1428,
"step": 2090
},
{
"epoch": 2.19,
"learning_rate": 9.033437826541276e-05,
"loss": 0.1726,
"step": 2100
},
{
"epoch": 2.19,
"eval_accuracy": 0.9962962962962963,
"eval_loss": 0.01471767295151949,
"eval_runtime": 16.5946,
"eval_samples_per_second": 162.703,
"eval_steps_per_second": 20.368,
"step": 2100
},
{
"epoch": 2.2,
"learning_rate": 8.981191222570533e-05,
"loss": 0.0762,
"step": 2110
},
{
"epoch": 2.22,
"learning_rate": 8.928944618599792e-05,
"loss": 0.0663,
"step": 2120
},
{
"epoch": 2.23,
"learning_rate": 8.876698014629049e-05,
"loss": 0.0743,
"step": 2130
},
{
"epoch": 2.24,
"learning_rate": 8.824451410658307e-05,
"loss": 0.0549,
"step": 2140
},
{
"epoch": 2.25,
"learning_rate": 8.772204806687566e-05,
"loss": 0.0409,
"step": 2150
},
{
"epoch": 2.26,
"learning_rate": 8.719958202716823e-05,
"loss": 0.0222,
"step": 2160
},
{
"epoch": 2.27,
"learning_rate": 8.667711598746082e-05,
"loss": 0.0481,
"step": 2170
},
{
"epoch": 2.28,
"learning_rate": 8.61546499477534e-05,
"loss": 0.0117,
"step": 2180
},
{
"epoch": 2.29,
"learning_rate": 8.563218390804599e-05,
"loss": 0.0415,
"step": 2190
},
{
"epoch": 2.3,
"learning_rate": 8.510971786833856e-05,
"loss": 0.0476,
"step": 2200
},
{
"epoch": 2.3,
"eval_accuracy": 0.9992592592592593,
"eval_loss": 0.005821824073791504,
"eval_runtime": 16.2815,
"eval_samples_per_second": 165.832,
"eval_steps_per_second": 20.76,
"step": 2200
},
{
"epoch": 2.31,
"learning_rate": 8.458725182863114e-05,
"loss": 0.005,
"step": 2210
},
{
"epoch": 2.32,
"learning_rate": 8.406478578892373e-05,
"loss": 0.0456,
"step": 2220
},
{
"epoch": 2.33,
"learning_rate": 8.35423197492163e-05,
"loss": 0.1262,
"step": 2230
},
{
"epoch": 2.34,
"learning_rate": 8.301985370950889e-05,
"loss": 0.0189,
"step": 2240
},
{
"epoch": 2.35,
"learning_rate": 8.249738766980146e-05,
"loss": 0.1614,
"step": 2250
},
{
"epoch": 2.36,
"learning_rate": 8.197492163009404e-05,
"loss": 0.1015,
"step": 2260
},
{
"epoch": 2.37,
"learning_rate": 8.145245559038663e-05,
"loss": 0.0352,
"step": 2270
},
{
"epoch": 2.38,
"learning_rate": 8.09299895506792e-05,
"loss": 0.0751,
"step": 2280
},
{
"epoch": 2.39,
"learning_rate": 8.04075235109718e-05,
"loss": 0.0369,
"step": 2290
},
{
"epoch": 2.4,
"learning_rate": 7.988505747126437e-05,
"loss": 0.0257,
"step": 2300
},
{
"epoch": 2.4,
"eval_accuracy": 0.9911111111111112,
"eval_loss": 0.04240882024168968,
"eval_runtime": 16.3138,
"eval_samples_per_second": 165.504,
"eval_steps_per_second": 20.719,
"step": 2300
},
{
"epoch": 2.41,
"learning_rate": 7.936259143155696e-05,
"loss": 0.1677,
"step": 2310
},
{
"epoch": 2.42,
"learning_rate": 7.884012539184953e-05,
"loss": 0.0875,
"step": 2320
},
{
"epoch": 2.43,
"learning_rate": 7.831765935214211e-05,
"loss": 0.0303,
"step": 2330
},
{
"epoch": 2.45,
"learning_rate": 7.77951933124347e-05,
"loss": 0.0992,
"step": 2340
},
{
"epoch": 2.46,
"learning_rate": 7.727272727272727e-05,
"loss": 0.0388,
"step": 2350
},
{
"epoch": 2.47,
"learning_rate": 7.675026123301986e-05,
"loss": 0.1306,
"step": 2360
},
{
"epoch": 2.48,
"learning_rate": 7.622779519331243e-05,
"loss": 0.0915,
"step": 2370
},
{
"epoch": 2.49,
"learning_rate": 7.570532915360501e-05,
"loss": 0.039,
"step": 2380
},
{
"epoch": 2.5,
"learning_rate": 7.51828631138976e-05,
"loss": 0.0623,
"step": 2390
},
{
"epoch": 2.51,
"learning_rate": 7.466039707419019e-05,
"loss": 0.0215,
"step": 2400
},
{
"epoch": 2.51,
"eval_accuracy": 0.9988888888888889,
"eval_loss": 0.007595046889036894,
"eval_runtime": 16.5367,
"eval_samples_per_second": 163.274,
"eval_steps_per_second": 20.439,
"step": 2400
},
{
"epoch": 2.52,
"learning_rate": 7.413793103448277e-05,
"loss": 0.0512,
"step": 2410
},
{
"epoch": 2.53,
"learning_rate": 7.361546499477534e-05,
"loss": 0.0809,
"step": 2420
},
{
"epoch": 2.54,
"learning_rate": 7.309299895506793e-05,
"loss": 0.0449,
"step": 2430
},
{
"epoch": 2.55,
"learning_rate": 7.25705329153605e-05,
"loss": 0.0824,
"step": 2440
},
{
"epoch": 2.56,
"learning_rate": 7.204806687565309e-05,
"loss": 0.0512,
"step": 2450
},
{
"epoch": 2.57,
"learning_rate": 7.152560083594567e-05,
"loss": 0.108,
"step": 2460
},
{
"epoch": 2.58,
"learning_rate": 7.100313479623824e-05,
"loss": 0.1016,
"step": 2470
},
{
"epoch": 2.59,
"learning_rate": 7.048066875653083e-05,
"loss": 0.0212,
"step": 2480
},
{
"epoch": 2.6,
"learning_rate": 6.995820271682341e-05,
"loss": 0.1275,
"step": 2490
},
{
"epoch": 2.61,
"learning_rate": 6.9435736677116e-05,
"loss": 0.0748,
"step": 2500
},
{
"epoch": 2.61,
"eval_accuracy": 0.9974074074074074,
"eval_loss": 0.00986641738563776,
"eval_runtime": 16.1049,
"eval_samples_per_second": 167.651,
"eval_steps_per_second": 20.987,
"step": 2500
},
{
"epoch": 2.62,
"learning_rate": 6.891327063740857e-05,
"loss": 0.0733,
"step": 2510
},
{
"epoch": 2.63,
"learning_rate": 6.839080459770116e-05,
"loss": 0.0985,
"step": 2520
},
{
"epoch": 2.64,
"learning_rate": 6.786833855799374e-05,
"loss": 0.059,
"step": 2530
},
{
"epoch": 2.65,
"learning_rate": 6.734587251828631e-05,
"loss": 0.0151,
"step": 2540
},
{
"epoch": 2.66,
"learning_rate": 6.68234064785789e-05,
"loss": 0.0597,
"step": 2550
},
{
"epoch": 2.68,
"learning_rate": 6.630094043887147e-05,
"loss": 0.1076,
"step": 2560
},
{
"epoch": 2.69,
"learning_rate": 6.577847439916406e-05,
"loss": 0.007,
"step": 2570
},
{
"epoch": 2.7,
"learning_rate": 6.525600835945663e-05,
"loss": 0.0186,
"step": 2580
},
{
"epoch": 2.71,
"learning_rate": 6.473354231974921e-05,
"loss": 0.0184,
"step": 2590
},
{
"epoch": 2.72,
"learning_rate": 6.42110762800418e-05,
"loss": 0.0059,
"step": 2600
},
{
"epoch": 2.72,
"eval_accuracy": 0.9992592592592593,
"eval_loss": 0.005262466147542,
"eval_runtime": 16.3428,
"eval_samples_per_second": 165.211,
"eval_steps_per_second": 20.682,
"step": 2600
},
{
"epoch": 2.73,
"learning_rate": 6.368861024033438e-05,
"loss": 0.0509,
"step": 2610
},
{
"epoch": 2.74,
"learning_rate": 6.316614420062697e-05,
"loss": 0.0477,
"step": 2620
},
{
"epoch": 2.75,
"learning_rate": 6.264367816091954e-05,
"loss": 0.0112,
"step": 2630
},
{
"epoch": 2.76,
"learning_rate": 6.212121212121213e-05,
"loss": 0.1193,
"step": 2640
},
{
"epoch": 2.77,
"learning_rate": 6.159874608150471e-05,
"loss": 0.1131,
"step": 2650
},
{
"epoch": 2.78,
"learning_rate": 6.107628004179728e-05,
"loss": 0.0899,
"step": 2660
},
{
"epoch": 2.79,
"learning_rate": 6.055381400208987e-05,
"loss": 0.0171,
"step": 2670
},
{
"epoch": 2.8,
"learning_rate": 6.003134796238245e-05,
"loss": 0.0834,
"step": 2680
},
{
"epoch": 2.81,
"learning_rate": 5.950888192267503e-05,
"loss": 0.078,
"step": 2690
},
{
"epoch": 2.82,
"learning_rate": 5.8986415882967604e-05,
"loss": 0.0527,
"step": 2700
},
{
"epoch": 2.82,
"eval_accuracy": 0.9962962962962963,
"eval_loss": 0.014925166964530945,
"eval_runtime": 16.5271,
"eval_samples_per_second": 163.369,
"eval_steps_per_second": 20.451,
"step": 2700
},
{
"epoch": 2.83,
"learning_rate": 5.846394984326019e-05,
"loss": 0.0311,
"step": 2710
},
{
"epoch": 2.84,
"learning_rate": 5.7941483803552775e-05,
"loss": 0.0198,
"step": 2720
},
{
"epoch": 2.85,
"learning_rate": 5.7419017763845354e-05,
"loss": 0.0303,
"step": 2730
},
{
"epoch": 2.86,
"learning_rate": 5.689655172413794e-05,
"loss": 0.0612,
"step": 2740
},
{
"epoch": 2.87,
"learning_rate": 5.637408568443051e-05,
"loss": 0.0949,
"step": 2750
},
{
"epoch": 2.88,
"learning_rate": 5.5851619644723096e-05,
"loss": 0.0499,
"step": 2760
},
{
"epoch": 2.89,
"learning_rate": 5.532915360501567e-05,
"loss": 0.0596,
"step": 2770
},
{
"epoch": 2.9,
"learning_rate": 5.4806687565308254e-05,
"loss": 0.0293,
"step": 2780
},
{
"epoch": 2.92,
"learning_rate": 5.4284221525600846e-05,
"loss": 0.053,
"step": 2790
},
{
"epoch": 2.93,
"learning_rate": 5.376175548589342e-05,
"loss": 0.0203,
"step": 2800
},
{
"epoch": 2.93,
"eval_accuracy": 0.9992592592592593,
"eval_loss": 0.004079786594957113,
"eval_runtime": 16.3325,
"eval_samples_per_second": 165.315,
"eval_steps_per_second": 20.695,
"step": 2800
},
{
"epoch": 2.94,
"learning_rate": 5.3239289446186e-05,
"loss": 0.0386,
"step": 2810
},
{
"epoch": 2.95,
"learning_rate": 5.2716823406478575e-05,
"loss": 0.0208,
"step": 2820
},
{
"epoch": 2.96,
"learning_rate": 5.219435736677116e-05,
"loss": 0.1105,
"step": 2830
},
{
"epoch": 2.97,
"learning_rate": 5.1671891327063745e-05,
"loss": 0.0827,
"step": 2840
},
{
"epoch": 2.98,
"learning_rate": 5.1149425287356324e-05,
"loss": 0.0249,
"step": 2850
},
{
"epoch": 2.99,
"learning_rate": 5.062695924764891e-05,
"loss": 0.0509,
"step": 2860
},
{
"epoch": 3.0,
"learning_rate": 5.010449320794148e-05,
"loss": 0.0324,
"step": 2870
},
{
"epoch": 3.01,
"learning_rate": 4.958202716823407e-05,
"loss": 0.0158,
"step": 2880
},
{
"epoch": 3.02,
"learning_rate": 4.905956112852665e-05,
"loss": 0.0029,
"step": 2890
},
{
"epoch": 3.03,
"learning_rate": 4.853709508881923e-05,
"loss": 0.0791,
"step": 2900
},
{
"epoch": 3.03,
"eval_accuracy": 0.9988888888888889,
"eval_loss": 0.0032932416070252657,
"eval_runtime": 16.2412,
"eval_samples_per_second": 166.244,
"eval_steps_per_second": 20.811,
"step": 2900
},
{
"epoch": 3.04,
"learning_rate": 4.801462904911181e-05,
"loss": 0.0097,
"step": 2910
},
{
"epoch": 3.05,
"learning_rate": 4.749216300940439e-05,
"loss": 0.0335,
"step": 2920
},
{
"epoch": 3.06,
"learning_rate": 4.696969696969697e-05,
"loss": 0.0392,
"step": 2930
},
{
"epoch": 3.07,
"learning_rate": 4.644723092998955e-05,
"loss": 0.0222,
"step": 2940
},
{
"epoch": 3.08,
"learning_rate": 4.592476489028214e-05,
"loss": 0.059,
"step": 2950
},
{
"epoch": 3.09,
"learning_rate": 4.5402298850574716e-05,
"loss": 0.0029,
"step": 2960
},
{
"epoch": 3.1,
"learning_rate": 4.4879832810867294e-05,
"loss": 0.0562,
"step": 2970
},
{
"epoch": 3.11,
"learning_rate": 4.435736677115987e-05,
"loss": 0.0233,
"step": 2980
},
{
"epoch": 3.12,
"learning_rate": 4.383490073145246e-05,
"loss": 0.0073,
"step": 2990
},
{
"epoch": 3.13,
"learning_rate": 4.3312434691745044e-05,
"loss": 0.0389,
"step": 3000
},
{
"epoch": 3.13,
"eval_accuracy": 0.9988888888888889,
"eval_loss": 0.0033078393898904324,
"eval_runtime": 16.4559,
"eval_samples_per_second": 164.075,
"eval_steps_per_second": 20.54,
"step": 3000
},
{
"epoch": 3.15,
"learning_rate": 4.278996865203762e-05,
"loss": 0.0057,
"step": 3010
},
{
"epoch": 3.16,
"learning_rate": 4.22675026123302e-05,
"loss": 0.0355,
"step": 3020
},
{
"epoch": 3.17,
"learning_rate": 4.174503657262278e-05,
"loss": 0.0169,
"step": 3030
},
{
"epoch": 3.18,
"learning_rate": 4.122257053291536e-05,
"loss": 0.0377,
"step": 3040
},
{
"epoch": 3.19,
"learning_rate": 4.0700104493207943e-05,
"loss": 0.038,
"step": 3050
},
{
"epoch": 3.2,
"learning_rate": 4.017763845350053e-05,
"loss": 0.0647,
"step": 3060
},
{
"epoch": 3.21,
"learning_rate": 3.965517241379311e-05,
"loss": 0.038,
"step": 3070
},
{
"epoch": 3.22,
"learning_rate": 3.9132706374085686e-05,
"loss": 0.0565,
"step": 3080
},
{
"epoch": 3.23,
"learning_rate": 3.8610240334378265e-05,
"loss": 0.0445,
"step": 3090
},
{
"epoch": 3.24,
"learning_rate": 3.808777429467085e-05,
"loss": 0.0459,
"step": 3100
},
{
"epoch": 3.24,
"eval_accuracy": 0.9988888888888889,
"eval_loss": 0.004428654909133911,
"eval_runtime": 16.26,
"eval_samples_per_second": 166.052,
"eval_steps_per_second": 20.787,
"step": 3100
},
{
"epoch": 3.25,
"learning_rate": 3.756530825496343e-05,
"loss": 0.0411,
"step": 3110
},
{
"epoch": 3.26,
"learning_rate": 3.7042842215256014e-05,
"loss": 0.0815,
"step": 3120
},
{
"epoch": 3.27,
"learning_rate": 3.652037617554859e-05,
"loss": 0.0029,
"step": 3130
},
{
"epoch": 3.28,
"learning_rate": 3.599791013584117e-05,
"loss": 0.0208,
"step": 3140
},
{
"epoch": 3.29,
"learning_rate": 3.547544409613375e-05,
"loss": 0.039,
"step": 3150
},
{
"epoch": 3.3,
"learning_rate": 3.4952978056426335e-05,
"loss": 0.0322,
"step": 3160
},
{
"epoch": 3.31,
"learning_rate": 3.4430512016718914e-05,
"loss": 0.0486,
"step": 3170
},
{
"epoch": 3.32,
"learning_rate": 3.390804597701149e-05,
"loss": 0.0396,
"step": 3180
},
{
"epoch": 3.33,
"learning_rate": 3.338557993730408e-05,
"loss": 0.0591,
"step": 3190
},
{
"epoch": 3.34,
"learning_rate": 3.2863113897596656e-05,
"loss": 0.0276,
"step": 3200
},
{
"epoch": 3.34,
"eval_accuracy": 0.9996296296296296,
"eval_loss": 0.0030968771316111088,
"eval_runtime": 16.225,
"eval_samples_per_second": 166.409,
"eval_steps_per_second": 20.832,
"step": 3200
},
{
"epoch": 3.35,
"learning_rate": 3.234064785788924e-05,
"loss": 0.0438,
"step": 3210
},
{
"epoch": 3.36,
"learning_rate": 3.181818181818182e-05,
"loss": 0.0394,
"step": 3220
},
{
"epoch": 3.38,
"learning_rate": 3.12957157784744e-05,
"loss": 0.0418,
"step": 3230
},
{
"epoch": 3.39,
"learning_rate": 3.077324973876698e-05,
"loss": 0.0278,
"step": 3240
},
{
"epoch": 3.4,
"learning_rate": 3.0250783699059566e-05,
"loss": 0.0578,
"step": 3250
},
{
"epoch": 3.41,
"learning_rate": 2.9728317659352145e-05,
"loss": 0.0365,
"step": 3260
},
{
"epoch": 3.42,
"learning_rate": 2.9205851619644723e-05,
"loss": 0.0528,
"step": 3270
},
{
"epoch": 3.43,
"learning_rate": 2.8683385579937305e-05,
"loss": 0.0597,
"step": 3280
},
{
"epoch": 3.44,
"learning_rate": 2.8160919540229884e-05,
"loss": 0.0318,
"step": 3290
},
{
"epoch": 3.45,
"learning_rate": 2.7638453500522466e-05,
"loss": 0.0139,
"step": 3300
},
{
"epoch": 3.45,
"eval_accuracy": 0.9996296296296296,
"eval_loss": 0.002758789574727416,
"eval_runtime": 16.3192,
"eval_samples_per_second": 165.449,
"eval_steps_per_second": 20.712,
"step": 3300
},
{
"epoch": 3.46,
"learning_rate": 2.711598746081505e-05,
"loss": 0.0542,
"step": 3310
},
{
"epoch": 3.47,
"learning_rate": 2.659352142110763e-05,
"loss": 0.0132,
"step": 3320
},
{
"epoch": 3.48,
"learning_rate": 2.6071055381400212e-05,
"loss": 0.0025,
"step": 3330
},
{
"epoch": 3.49,
"learning_rate": 2.554858934169279e-05,
"loss": 0.0418,
"step": 3340
},
{
"epoch": 3.5,
"learning_rate": 2.5026123301985373e-05,
"loss": 0.0376,
"step": 3350
},
{
"epoch": 3.51,
"learning_rate": 2.4503657262277955e-05,
"loss": 0.0566,
"step": 3360
},
{
"epoch": 3.52,
"learning_rate": 2.3981191222570533e-05,
"loss": 0.1575,
"step": 3370
},
{
"epoch": 3.53,
"learning_rate": 2.3458725182863115e-05,
"loss": 0.0056,
"step": 3380
},
{
"epoch": 3.54,
"learning_rate": 2.2936259143155697e-05,
"loss": 0.0066,
"step": 3390
},
{
"epoch": 3.55,
"learning_rate": 2.2413793103448276e-05,
"loss": 0.0076,
"step": 3400
},
{
"epoch": 3.55,
"eval_accuracy": 0.9985185185185185,
"eval_loss": 0.005528000649064779,
"eval_runtime": 16.457,
"eval_samples_per_second": 164.064,
"eval_steps_per_second": 20.538,
"step": 3400
},
{
"epoch": 3.56,
"learning_rate": 2.1891327063740858e-05,
"loss": 0.0064,
"step": 3410
},
{
"epoch": 3.57,
"learning_rate": 2.1368861024033436e-05,
"loss": 0.0376,
"step": 3420
},
{
"epoch": 3.58,
"learning_rate": 2.0846394984326022e-05,
"loss": 0.0471,
"step": 3430
},
{
"epoch": 3.59,
"learning_rate": 2.03239289446186e-05,
"loss": 0.0029,
"step": 3440
},
{
"epoch": 3.61,
"learning_rate": 1.980146290491118e-05,
"loss": 0.0406,
"step": 3450
},
{
"epoch": 3.62,
"learning_rate": 1.9278996865203764e-05,
"loss": 0.0174,
"step": 3460
},
{
"epoch": 3.63,
"learning_rate": 1.8756530825496343e-05,
"loss": 0.027,
"step": 3470
},
{
"epoch": 3.64,
"learning_rate": 1.8234064785788925e-05,
"loss": 0.0181,
"step": 3480
},
{
"epoch": 3.65,
"learning_rate": 1.7711598746081507e-05,
"loss": 0.0319,
"step": 3490
},
{
"epoch": 3.66,
"learning_rate": 1.7189132706374085e-05,
"loss": 0.0097,
"step": 3500
},
{
"epoch": 3.66,
"eval_accuracy": 0.9996296296296296,
"eval_loss": 0.0027309097349643707,
"eval_runtime": 16.2262,
"eval_samples_per_second": 166.398,
"eval_steps_per_second": 20.831,
"step": 3500
},
{
"epoch": 3.67,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.0221,
"step": 3510
},
{
"epoch": 3.68,
"learning_rate": 1.614420062695925e-05,
"loss": 0.0124,
"step": 3520
},
{
"epoch": 3.69,
"learning_rate": 1.5621734587251828e-05,
"loss": 0.0257,
"step": 3530
},
{
"epoch": 3.7,
"learning_rate": 1.509926854754441e-05,
"loss": 0.0794,
"step": 3540
},
{
"epoch": 3.71,
"learning_rate": 1.4576802507836992e-05,
"loss": 0.0382,
"step": 3550
},
{
"epoch": 3.72,
"learning_rate": 1.4054336468129572e-05,
"loss": 0.0441,
"step": 3560
},
{
"epoch": 3.73,
"learning_rate": 1.3531870428422153e-05,
"loss": 0.0057,
"step": 3570
},
{
"epoch": 3.74,
"learning_rate": 1.3009404388714735e-05,
"loss": 0.085,
"step": 3580
},
{
"epoch": 3.75,
"learning_rate": 1.2486938349007315e-05,
"loss": 0.0658,
"step": 3590
},
{
"epoch": 3.76,
"learning_rate": 1.1964472309299895e-05,
"loss": 0.0193,
"step": 3600
},
{
"epoch": 3.76,
"eval_accuracy": 0.9996296296296296,
"eval_loss": 0.002558160340413451,
"eval_runtime": 16.2201,
"eval_samples_per_second": 166.46,
"eval_steps_per_second": 20.838,
"step": 3600
},
{
"epoch": 3.77,
"learning_rate": 1.1442006269592477e-05,
"loss": 0.0276,
"step": 3610
},
{
"epoch": 3.78,
"learning_rate": 1.091954022988506e-05,
"loss": 0.0265,
"step": 3620
},
{
"epoch": 3.79,
"learning_rate": 1.039707419017764e-05,
"loss": 0.0297,
"step": 3630
},
{
"epoch": 3.8,
"learning_rate": 9.87460815047022e-06,
"loss": 0.0288,
"step": 3640
},
{
"epoch": 3.81,
"learning_rate": 9.352142110762802e-06,
"loss": 0.0017,
"step": 3650
},
{
"epoch": 3.82,
"learning_rate": 8.829676071055382e-06,
"loss": 0.0258,
"step": 3660
},
{
"epoch": 3.83,
"learning_rate": 8.307210031347962e-06,
"loss": 0.0044,
"step": 3670
},
{
"epoch": 3.85,
"learning_rate": 7.784743991640543e-06,
"loss": 0.0484,
"step": 3680
},
{
"epoch": 3.86,
"learning_rate": 7.262277951933125e-06,
"loss": 0.0367,
"step": 3690
},
{
"epoch": 3.87,
"learning_rate": 6.739811912225706e-06,
"loss": 0.0471,
"step": 3700
},
{
"epoch": 3.87,
"eval_accuracy": 0.9996296296296296,
"eval_loss": 0.002674533985555172,
"eval_runtime": 16.1924,
"eval_samples_per_second": 166.745,
"eval_steps_per_second": 20.874,
"step": 3700
},
{
"epoch": 3.88,
"learning_rate": 6.217345872518287e-06,
"loss": 0.0245,
"step": 3710
},
{
"epoch": 3.89,
"learning_rate": 5.694879832810868e-06,
"loss": 0.0392,
"step": 3720
},
{
"epoch": 3.9,
"learning_rate": 5.172413793103448e-06,
"loss": 0.0109,
"step": 3730
},
{
"epoch": 3.91,
"learning_rate": 4.6499477533960295e-06,
"loss": 0.0022,
"step": 3740
},
{
"epoch": 3.92,
"learning_rate": 4.127481713688611e-06,
"loss": 0.0765,
"step": 3750
},
{
"epoch": 3.93,
"learning_rate": 3.6050156739811913e-06,
"loss": 0.0029,
"step": 3760
},
{
"epoch": 3.94,
"learning_rate": 3.0825496342737725e-06,
"loss": 0.0236,
"step": 3770
},
{
"epoch": 3.95,
"learning_rate": 2.560083594566353e-06,
"loss": 0.0308,
"step": 3780
},
{
"epoch": 3.96,
"learning_rate": 2.0376175548589343e-06,
"loss": 0.0288,
"step": 3790
},
{
"epoch": 3.97,
"learning_rate": 1.5151515151515152e-06,
"loss": 0.0282,
"step": 3800
},
{
"epoch": 3.97,
"eval_accuracy": 0.9996296296296296,
"eval_loss": 0.0026795254088938236,
"eval_runtime": 16.4275,
"eval_samples_per_second": 164.358,
"eval_steps_per_second": 20.575,
"step": 3800
},
{
"epoch": 3.98,
"learning_rate": 9.926854754440962e-07,
"loss": 0.0188,
"step": 3810
},
{
"epoch": 3.99,
"learning_rate": 4.7021943573667715e-07,
"loss": 0.0323,
"step": 3820
},
{
"epoch": 4.0,
"step": 3828,
"total_flos": 4.742807303717683e+18,
"train_loss": 0.13919885493931725,
"train_runtime": 1939.8119,
"train_samples_per_second": 31.549,
"train_steps_per_second": 1.973
}
],
"logging_steps": 10,
"max_steps": 3828,
"num_train_epochs": 4,
"save_steps": 100,
"total_flos": 4.742807303717683e+18,
"trial_name": null,
"trial_params": null
}