bliebfl's picture
Training in progress, epoch 0
18d155a
raw
history blame
52.3 kB
{
"best_metric": 0.9405358686257562,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-3660",
"epoch": 9.981561155500922,
"eval_steps": 500,
"global_step": 4060,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 1.2315270935960593e-06,
"loss": 1.8272,
"step": 10
},
{
"epoch": 0.05,
"learning_rate": 2.4630541871921186e-06,
"loss": 1.7835,
"step": 20
},
{
"epoch": 0.07,
"learning_rate": 3.6945812807881777e-06,
"loss": 1.751,
"step": 30
},
{
"epoch": 0.1,
"learning_rate": 4.926108374384237e-06,
"loss": 1.7539,
"step": 40
},
{
"epoch": 0.12,
"learning_rate": 6.157635467980296e-06,
"loss": 1.7125,
"step": 50
},
{
"epoch": 0.15,
"learning_rate": 7.3891625615763555e-06,
"loss": 1.6708,
"step": 60
},
{
"epoch": 0.17,
"learning_rate": 8.620689655172414e-06,
"loss": 1.7727,
"step": 70
},
{
"epoch": 0.2,
"learning_rate": 9.852216748768475e-06,
"loss": 1.6377,
"step": 80
},
{
"epoch": 0.22,
"learning_rate": 1.1083743842364533e-05,
"loss": 1.8024,
"step": 90
},
{
"epoch": 0.25,
"learning_rate": 1.2315270935960592e-05,
"loss": 1.7476,
"step": 100
},
{
"epoch": 0.27,
"learning_rate": 1.354679802955665e-05,
"loss": 1.6509,
"step": 110
},
{
"epoch": 0.3,
"learning_rate": 1.4778325123152711e-05,
"loss": 1.7052,
"step": 120
},
{
"epoch": 0.32,
"learning_rate": 1.6009852216748768e-05,
"loss": 1.7003,
"step": 130
},
{
"epoch": 0.34,
"learning_rate": 1.7241379310344828e-05,
"loss": 1.7424,
"step": 140
},
{
"epoch": 0.37,
"learning_rate": 1.847290640394089e-05,
"loss": 1.7223,
"step": 150
},
{
"epoch": 0.39,
"learning_rate": 1.970443349753695e-05,
"loss": 1.6932,
"step": 160
},
{
"epoch": 0.42,
"learning_rate": 2.0935960591133006e-05,
"loss": 1.6393,
"step": 170
},
{
"epoch": 0.44,
"learning_rate": 2.2167487684729066e-05,
"loss": 1.7023,
"step": 180
},
{
"epoch": 0.47,
"learning_rate": 2.3399014778325123e-05,
"loss": 1.6219,
"step": 190
},
{
"epoch": 0.49,
"learning_rate": 2.4630541871921184e-05,
"loss": 1.7031,
"step": 200
},
{
"epoch": 0.52,
"learning_rate": 2.5862068965517244e-05,
"loss": 1.7309,
"step": 210
},
{
"epoch": 0.54,
"learning_rate": 2.70935960591133e-05,
"loss": 1.7251,
"step": 220
},
{
"epoch": 0.57,
"learning_rate": 2.8325123152709358e-05,
"loss": 1.7039,
"step": 230
},
{
"epoch": 0.59,
"learning_rate": 2.9556650246305422e-05,
"loss": 1.6753,
"step": 240
},
{
"epoch": 0.61,
"learning_rate": 3.078817733990148e-05,
"loss": 1.7149,
"step": 250
},
{
"epoch": 0.64,
"learning_rate": 3.2019704433497536e-05,
"loss": 1.7101,
"step": 260
},
{
"epoch": 0.66,
"learning_rate": 3.3251231527093596e-05,
"loss": 1.6208,
"step": 270
},
{
"epoch": 0.69,
"learning_rate": 3.4482758620689657e-05,
"loss": 1.5855,
"step": 280
},
{
"epoch": 0.71,
"learning_rate": 3.571428571428572e-05,
"loss": 1.741,
"step": 290
},
{
"epoch": 0.74,
"learning_rate": 3.694581280788178e-05,
"loss": 1.6925,
"step": 300
},
{
"epoch": 0.76,
"learning_rate": 3.817733990147783e-05,
"loss": 1.6607,
"step": 310
},
{
"epoch": 0.79,
"learning_rate": 3.94088669950739e-05,
"loss": 1.5157,
"step": 320
},
{
"epoch": 0.81,
"learning_rate": 4.064039408866995e-05,
"loss": 1.5652,
"step": 330
},
{
"epoch": 0.84,
"learning_rate": 4.187192118226601e-05,
"loss": 1.6547,
"step": 340
},
{
"epoch": 0.86,
"learning_rate": 4.3103448275862066e-05,
"loss": 1.5618,
"step": 350
},
{
"epoch": 0.89,
"learning_rate": 4.433497536945813e-05,
"loss": 1.6555,
"step": 360
},
{
"epoch": 0.91,
"learning_rate": 4.5566502463054186e-05,
"loss": 1.5538,
"step": 370
},
{
"epoch": 0.93,
"learning_rate": 4.679802955665025e-05,
"loss": 1.5496,
"step": 380
},
{
"epoch": 0.96,
"learning_rate": 4.802955665024631e-05,
"loss": 1.5248,
"step": 390
},
{
"epoch": 0.98,
"learning_rate": 4.926108374384237e-05,
"loss": 1.5645,
"step": 400
},
{
"epoch": 1.0,
"eval_accuracy": 0.7904926534140018,
"eval_loss": 1.0879656076431274,
"eval_runtime": 27.6968,
"eval_samples_per_second": 208.869,
"eval_steps_per_second": 6.535,
"step": 406
},
{
"epoch": 1.01,
"learning_rate": 4.9945265462506846e-05,
"loss": 1.5933,
"step": 410
},
{
"epoch": 1.03,
"learning_rate": 4.980842911877395e-05,
"loss": 1.5284,
"step": 420
},
{
"epoch": 1.06,
"learning_rate": 4.9671592775041054e-05,
"loss": 1.5239,
"step": 430
},
{
"epoch": 1.08,
"learning_rate": 4.953475643130816e-05,
"loss": 1.4781,
"step": 440
},
{
"epoch": 1.11,
"learning_rate": 4.939792008757526e-05,
"loss": 1.468,
"step": 450
},
{
"epoch": 1.13,
"learning_rate": 4.926108374384237e-05,
"loss": 1.5165,
"step": 460
},
{
"epoch": 1.16,
"learning_rate": 4.912424740010947e-05,
"loss": 1.5641,
"step": 470
},
{
"epoch": 1.18,
"learning_rate": 4.8987411056376576e-05,
"loss": 1.53,
"step": 480
},
{
"epoch": 1.2,
"learning_rate": 4.885057471264368e-05,
"loss": 1.4946,
"step": 490
},
{
"epoch": 1.23,
"learning_rate": 4.8713738368910785e-05,
"loss": 1.5378,
"step": 500
},
{
"epoch": 1.25,
"learning_rate": 4.857690202517789e-05,
"loss": 1.4228,
"step": 510
},
{
"epoch": 1.28,
"learning_rate": 4.8440065681444994e-05,
"loss": 1.4621,
"step": 520
},
{
"epoch": 1.3,
"learning_rate": 4.83032293377121e-05,
"loss": 1.5405,
"step": 530
},
{
"epoch": 1.33,
"learning_rate": 4.81663929939792e-05,
"loss": 1.4338,
"step": 540
},
{
"epoch": 1.35,
"learning_rate": 4.802955665024631e-05,
"loss": 1.3429,
"step": 550
},
{
"epoch": 1.38,
"learning_rate": 4.789272030651341e-05,
"loss": 1.4912,
"step": 560
},
{
"epoch": 1.4,
"learning_rate": 4.7755883962780516e-05,
"loss": 1.4859,
"step": 570
},
{
"epoch": 1.43,
"learning_rate": 4.761904761904762e-05,
"loss": 1.3696,
"step": 580
},
{
"epoch": 1.45,
"learning_rate": 4.7482211275314725e-05,
"loss": 1.3818,
"step": 590
},
{
"epoch": 1.48,
"learning_rate": 4.734537493158183e-05,
"loss": 1.395,
"step": 600
},
{
"epoch": 1.5,
"learning_rate": 4.7208538587848934e-05,
"loss": 1.5301,
"step": 610
},
{
"epoch": 1.52,
"learning_rate": 4.707170224411604e-05,
"loss": 1.3008,
"step": 620
},
{
"epoch": 1.55,
"learning_rate": 4.693486590038315e-05,
"loss": 1.3089,
"step": 630
},
{
"epoch": 1.57,
"learning_rate": 4.679802955665025e-05,
"loss": 1.4767,
"step": 640
},
{
"epoch": 1.6,
"learning_rate": 4.666119321291735e-05,
"loss": 1.3715,
"step": 650
},
{
"epoch": 1.62,
"learning_rate": 4.652435686918446e-05,
"loss": 1.3359,
"step": 660
},
{
"epoch": 1.65,
"learning_rate": 4.638752052545156e-05,
"loss": 1.38,
"step": 670
},
{
"epoch": 1.67,
"learning_rate": 4.6250684181718664e-05,
"loss": 1.4252,
"step": 680
},
{
"epoch": 1.7,
"learning_rate": 4.611384783798577e-05,
"loss": 1.2751,
"step": 690
},
{
"epoch": 1.72,
"learning_rate": 4.597701149425287e-05,
"loss": 1.4558,
"step": 700
},
{
"epoch": 1.75,
"learning_rate": 4.5840175150519984e-05,
"loss": 1.3744,
"step": 710
},
{
"epoch": 1.77,
"learning_rate": 4.570333880678708e-05,
"loss": 1.3676,
"step": 720
},
{
"epoch": 1.79,
"learning_rate": 4.5566502463054186e-05,
"loss": 1.3031,
"step": 730
},
{
"epoch": 1.82,
"learning_rate": 4.54296661193213e-05,
"loss": 1.346,
"step": 740
},
{
"epoch": 1.84,
"learning_rate": 4.5292829775588395e-05,
"loss": 1.337,
"step": 750
},
{
"epoch": 1.87,
"learning_rate": 4.5155993431855506e-05,
"loss": 1.318,
"step": 760
},
{
"epoch": 1.89,
"learning_rate": 4.501915708812261e-05,
"loss": 1.4677,
"step": 770
},
{
"epoch": 1.92,
"learning_rate": 4.488232074438971e-05,
"loss": 1.2769,
"step": 780
},
{
"epoch": 1.94,
"learning_rate": 4.474548440065682e-05,
"loss": 1.2428,
"step": 790
},
{
"epoch": 1.97,
"learning_rate": 4.460864805692392e-05,
"loss": 1.2961,
"step": 800
},
{
"epoch": 1.99,
"learning_rate": 4.447181171319103e-05,
"loss": 1.3126,
"step": 810
},
{
"epoch": 2.0,
"eval_accuracy": 0.8399308556611927,
"eval_loss": 0.7987398505210876,
"eval_runtime": 27.7317,
"eval_samples_per_second": 208.606,
"eval_steps_per_second": 6.527,
"step": 813
},
{
"epoch": 2.02,
"learning_rate": 4.433497536945813e-05,
"loss": 1.271,
"step": 820
},
{
"epoch": 2.04,
"learning_rate": 4.419813902572523e-05,
"loss": 1.3539,
"step": 830
},
{
"epoch": 2.07,
"learning_rate": 4.406130268199234e-05,
"loss": 1.2938,
"step": 840
},
{
"epoch": 2.09,
"learning_rate": 4.3924466338259446e-05,
"loss": 1.2817,
"step": 850
},
{
"epoch": 2.11,
"learning_rate": 4.3787629994526544e-05,
"loss": 1.2808,
"step": 860
},
{
"epoch": 2.14,
"learning_rate": 4.3650793650793655e-05,
"loss": 1.2517,
"step": 870
},
{
"epoch": 2.16,
"learning_rate": 4.351395730706076e-05,
"loss": 1.2371,
"step": 880
},
{
"epoch": 2.19,
"learning_rate": 4.3377120963327864e-05,
"loss": 1.277,
"step": 890
},
{
"epoch": 2.21,
"learning_rate": 4.324028461959497e-05,
"loss": 1.2152,
"step": 900
},
{
"epoch": 2.24,
"learning_rate": 4.3103448275862066e-05,
"loss": 1.1733,
"step": 910
},
{
"epoch": 2.26,
"learning_rate": 4.296661193212918e-05,
"loss": 1.3563,
"step": 920
},
{
"epoch": 2.29,
"learning_rate": 4.282977558839628e-05,
"loss": 1.4114,
"step": 930
},
{
"epoch": 2.31,
"learning_rate": 4.2692939244663386e-05,
"loss": 1.1794,
"step": 940
},
{
"epoch": 2.34,
"learning_rate": 4.255610290093049e-05,
"loss": 1.2454,
"step": 950
},
{
"epoch": 2.36,
"learning_rate": 4.2419266557197594e-05,
"loss": 1.2245,
"step": 960
},
{
"epoch": 2.38,
"learning_rate": 4.22824302134647e-05,
"loss": 1.2197,
"step": 970
},
{
"epoch": 2.41,
"learning_rate": 4.21455938697318e-05,
"loss": 1.3025,
"step": 980
},
{
"epoch": 2.43,
"learning_rate": 4.200875752599891e-05,
"loss": 1.1717,
"step": 990
},
{
"epoch": 2.46,
"learning_rate": 4.187192118226601e-05,
"loss": 1.2077,
"step": 1000
},
{
"epoch": 2.48,
"learning_rate": 4.1735084838533116e-05,
"loss": 1.3137,
"step": 1010
},
{
"epoch": 2.51,
"learning_rate": 4.159824849480022e-05,
"loss": 1.2348,
"step": 1020
},
{
"epoch": 2.53,
"learning_rate": 4.1461412151067325e-05,
"loss": 1.2611,
"step": 1030
},
{
"epoch": 2.56,
"learning_rate": 4.132457580733443e-05,
"loss": 1.0912,
"step": 1040
},
{
"epoch": 2.58,
"learning_rate": 4.1187739463601534e-05,
"loss": 1.3145,
"step": 1050
},
{
"epoch": 2.61,
"learning_rate": 4.105090311986864e-05,
"loss": 1.2088,
"step": 1060
},
{
"epoch": 2.63,
"learning_rate": 4.091406677613574e-05,
"loss": 1.1283,
"step": 1070
},
{
"epoch": 2.66,
"learning_rate": 4.077723043240285e-05,
"loss": 1.1388,
"step": 1080
},
{
"epoch": 2.68,
"learning_rate": 4.064039408866995e-05,
"loss": 1.2748,
"step": 1090
},
{
"epoch": 2.7,
"learning_rate": 4.050355774493706e-05,
"loss": 1.0486,
"step": 1100
},
{
"epoch": 2.73,
"learning_rate": 4.036672140120416e-05,
"loss": 1.1514,
"step": 1110
},
{
"epoch": 2.75,
"learning_rate": 4.0229885057471265e-05,
"loss": 1.2419,
"step": 1120
},
{
"epoch": 2.78,
"learning_rate": 4.009304871373837e-05,
"loss": 1.3025,
"step": 1130
},
{
"epoch": 2.8,
"learning_rate": 3.9956212370005474e-05,
"loss": 1.1627,
"step": 1140
},
{
"epoch": 2.83,
"learning_rate": 3.981937602627258e-05,
"loss": 1.2263,
"step": 1150
},
{
"epoch": 2.85,
"learning_rate": 3.968253968253968e-05,
"loss": 1.1918,
"step": 1160
},
{
"epoch": 2.88,
"learning_rate": 3.954570333880679e-05,
"loss": 1.2337,
"step": 1170
},
{
"epoch": 2.9,
"learning_rate": 3.94088669950739e-05,
"loss": 1.325,
"step": 1180
},
{
"epoch": 2.93,
"learning_rate": 3.9272030651340996e-05,
"loss": 1.1631,
"step": 1190
},
{
"epoch": 2.95,
"learning_rate": 3.91351943076081e-05,
"loss": 1.2506,
"step": 1200
},
{
"epoch": 2.97,
"learning_rate": 3.899835796387521e-05,
"loss": 1.2831,
"step": 1210
},
{
"epoch": 3.0,
"learning_rate": 3.886152162014231e-05,
"loss": 1.2335,
"step": 1220
},
{
"epoch": 3.0,
"eval_accuracy": 0.8741573033707866,
"eval_loss": 0.6398221254348755,
"eval_runtime": 27.2608,
"eval_samples_per_second": 212.209,
"eval_steps_per_second": 6.64,
"step": 1220
},
{
"epoch": 3.02,
"learning_rate": 3.872468527640942e-05,
"loss": 1.1144,
"step": 1230
},
{
"epoch": 3.05,
"learning_rate": 3.858784893267652e-05,
"loss": 1.1746,
"step": 1240
},
{
"epoch": 3.07,
"learning_rate": 3.845101258894362e-05,
"loss": 1.1363,
"step": 1250
},
{
"epoch": 3.1,
"learning_rate": 3.831417624521073e-05,
"loss": 1.1003,
"step": 1260
},
{
"epoch": 3.12,
"learning_rate": 3.817733990147783e-05,
"loss": 1.1743,
"step": 1270
},
{
"epoch": 3.15,
"learning_rate": 3.8040503557744935e-05,
"loss": 1.1832,
"step": 1280
},
{
"epoch": 3.17,
"learning_rate": 3.7903667214012047e-05,
"loss": 1.1495,
"step": 1290
},
{
"epoch": 3.2,
"learning_rate": 3.7766830870279144e-05,
"loss": 1.1282,
"step": 1300
},
{
"epoch": 3.22,
"learning_rate": 3.7629994526546255e-05,
"loss": 1.1273,
"step": 1310
},
{
"epoch": 3.25,
"learning_rate": 3.749315818281336e-05,
"loss": 1.0586,
"step": 1320
},
{
"epoch": 3.27,
"learning_rate": 3.735632183908046e-05,
"loss": 1.1353,
"step": 1330
},
{
"epoch": 3.29,
"learning_rate": 3.721948549534757e-05,
"loss": 1.0796,
"step": 1340
},
{
"epoch": 3.32,
"learning_rate": 3.7082649151614666e-05,
"loss": 1.1132,
"step": 1350
},
{
"epoch": 3.34,
"learning_rate": 3.694581280788178e-05,
"loss": 1.0474,
"step": 1360
},
{
"epoch": 3.37,
"learning_rate": 3.680897646414888e-05,
"loss": 1.1392,
"step": 1370
},
{
"epoch": 3.39,
"learning_rate": 3.667214012041598e-05,
"loss": 1.1334,
"step": 1380
},
{
"epoch": 3.42,
"learning_rate": 3.653530377668309e-05,
"loss": 1.0506,
"step": 1390
},
{
"epoch": 3.44,
"learning_rate": 3.6398467432950195e-05,
"loss": 1.1107,
"step": 1400
},
{
"epoch": 3.47,
"learning_rate": 3.62616310892173e-05,
"loss": 1.1643,
"step": 1410
},
{
"epoch": 3.49,
"learning_rate": 3.6124794745484404e-05,
"loss": 1.0167,
"step": 1420
},
{
"epoch": 3.52,
"learning_rate": 3.598795840175151e-05,
"loss": 1.2377,
"step": 1430
},
{
"epoch": 3.54,
"learning_rate": 3.585112205801861e-05,
"loss": 1.0392,
"step": 1440
},
{
"epoch": 3.56,
"learning_rate": 3.571428571428572e-05,
"loss": 1.1304,
"step": 1450
},
{
"epoch": 3.59,
"learning_rate": 3.5577449370552815e-05,
"loss": 1.2185,
"step": 1460
},
{
"epoch": 3.61,
"learning_rate": 3.5440613026819926e-05,
"loss": 1.0848,
"step": 1470
},
{
"epoch": 3.64,
"learning_rate": 3.530377668308703e-05,
"loss": 1.1142,
"step": 1480
},
{
"epoch": 3.66,
"learning_rate": 3.5166940339354135e-05,
"loss": 1.0458,
"step": 1490
},
{
"epoch": 3.69,
"learning_rate": 3.503010399562124e-05,
"loss": 1.041,
"step": 1500
},
{
"epoch": 3.71,
"learning_rate": 3.489326765188834e-05,
"loss": 1.119,
"step": 1510
},
{
"epoch": 3.74,
"learning_rate": 3.475643130815545e-05,
"loss": 1.0259,
"step": 1520
},
{
"epoch": 3.76,
"learning_rate": 3.461959496442255e-05,
"loss": 1.14,
"step": 1530
},
{
"epoch": 3.79,
"learning_rate": 3.4482758620689657e-05,
"loss": 0.994,
"step": 1540
},
{
"epoch": 3.81,
"learning_rate": 3.434592227695676e-05,
"loss": 1.011,
"step": 1550
},
{
"epoch": 3.84,
"learning_rate": 3.4209085933223865e-05,
"loss": 1.0745,
"step": 1560
},
{
"epoch": 3.86,
"learning_rate": 3.407224958949097e-05,
"loss": 1.0777,
"step": 1570
},
{
"epoch": 3.88,
"learning_rate": 3.3935413245758074e-05,
"loss": 1.1397,
"step": 1580
},
{
"epoch": 3.91,
"learning_rate": 3.379857690202518e-05,
"loss": 1.0727,
"step": 1590
},
{
"epoch": 3.93,
"learning_rate": 3.366174055829228e-05,
"loss": 1.0662,
"step": 1600
},
{
"epoch": 3.96,
"learning_rate": 3.352490421455939e-05,
"loss": 1.0208,
"step": 1610
},
{
"epoch": 3.98,
"learning_rate": 3.338806787082649e-05,
"loss": 1.1244,
"step": 1620
},
{
"epoch": 4.0,
"eval_accuracy": 0.9040622299049266,
"eval_loss": 0.497685045003891,
"eval_runtime": 27.8668,
"eval_samples_per_second": 207.595,
"eval_steps_per_second": 6.495,
"step": 1627
},
{
"epoch": 4.01,
"learning_rate": 3.3251231527093596e-05,
"loss": 0.9674,
"step": 1630
},
{
"epoch": 4.03,
"learning_rate": 3.31143951833607e-05,
"loss": 1.0643,
"step": 1640
},
{
"epoch": 4.06,
"learning_rate": 3.297755883962781e-05,
"loss": 1.0174,
"step": 1650
},
{
"epoch": 4.08,
"learning_rate": 3.284072249589491e-05,
"loss": 1.0732,
"step": 1660
},
{
"epoch": 4.11,
"learning_rate": 3.2703886152162014e-05,
"loss": 1.0587,
"step": 1670
},
{
"epoch": 4.13,
"learning_rate": 3.256704980842912e-05,
"loss": 1.0531,
"step": 1680
},
{
"epoch": 4.15,
"learning_rate": 3.243021346469622e-05,
"loss": 1.0491,
"step": 1690
},
{
"epoch": 4.18,
"learning_rate": 3.2293377120963334e-05,
"loss": 0.9433,
"step": 1700
},
{
"epoch": 4.2,
"learning_rate": 3.215654077723043e-05,
"loss": 0.9851,
"step": 1710
},
{
"epoch": 4.23,
"learning_rate": 3.2019704433497536e-05,
"loss": 1.0559,
"step": 1720
},
{
"epoch": 4.25,
"learning_rate": 3.188286808976465e-05,
"loss": 0.9604,
"step": 1730
},
{
"epoch": 4.28,
"learning_rate": 3.1746031746031745e-05,
"loss": 0.9938,
"step": 1740
},
{
"epoch": 4.3,
"learning_rate": 3.160919540229885e-05,
"loss": 1.0587,
"step": 1750
},
{
"epoch": 4.33,
"learning_rate": 3.147235905856596e-05,
"loss": 0.9771,
"step": 1760
},
{
"epoch": 4.35,
"learning_rate": 3.133552271483306e-05,
"loss": 1.0232,
"step": 1770
},
{
"epoch": 4.38,
"learning_rate": 3.119868637110017e-05,
"loss": 1.014,
"step": 1780
},
{
"epoch": 4.4,
"learning_rate": 3.1061850027367273e-05,
"loss": 0.9797,
"step": 1790
},
{
"epoch": 4.43,
"learning_rate": 3.092501368363437e-05,
"loss": 0.9898,
"step": 1800
},
{
"epoch": 4.45,
"learning_rate": 3.078817733990148e-05,
"loss": 1.0044,
"step": 1810
},
{
"epoch": 4.47,
"learning_rate": 3.065134099616858e-05,
"loss": 1.0723,
"step": 1820
},
{
"epoch": 4.5,
"learning_rate": 3.0514504652435688e-05,
"loss": 1.0545,
"step": 1830
},
{
"epoch": 4.52,
"learning_rate": 3.0377668308702795e-05,
"loss": 0.9979,
"step": 1840
},
{
"epoch": 4.55,
"learning_rate": 3.0240831964969896e-05,
"loss": 0.9971,
"step": 1850
},
{
"epoch": 4.57,
"learning_rate": 3.0103995621237e-05,
"loss": 1.0341,
"step": 1860
},
{
"epoch": 4.6,
"learning_rate": 2.996715927750411e-05,
"loss": 0.9853,
"step": 1870
},
{
"epoch": 4.62,
"learning_rate": 2.983032293377121e-05,
"loss": 1.0434,
"step": 1880
},
{
"epoch": 4.65,
"learning_rate": 2.9693486590038317e-05,
"loss": 0.94,
"step": 1890
},
{
"epoch": 4.67,
"learning_rate": 2.9556650246305422e-05,
"loss": 1.0002,
"step": 1900
},
{
"epoch": 4.7,
"learning_rate": 2.9419813902572523e-05,
"loss": 0.9252,
"step": 1910
},
{
"epoch": 4.72,
"learning_rate": 2.928297755883963e-05,
"loss": 0.8899,
"step": 1920
},
{
"epoch": 4.74,
"learning_rate": 2.914614121510673e-05,
"loss": 1.0344,
"step": 1930
},
{
"epoch": 4.77,
"learning_rate": 2.900930487137384e-05,
"loss": 1.008,
"step": 1940
},
{
"epoch": 4.79,
"learning_rate": 2.8872468527640944e-05,
"loss": 0.9935,
"step": 1950
},
{
"epoch": 4.82,
"learning_rate": 2.8735632183908045e-05,
"loss": 1.0432,
"step": 1960
},
{
"epoch": 4.84,
"learning_rate": 2.8598795840175153e-05,
"loss": 0.9235,
"step": 1970
},
{
"epoch": 4.87,
"learning_rate": 2.8461959496442257e-05,
"loss": 0.9564,
"step": 1980
},
{
"epoch": 4.89,
"learning_rate": 2.8325123152709358e-05,
"loss": 0.9873,
"step": 1990
},
{
"epoch": 4.92,
"learning_rate": 2.8188286808976466e-05,
"loss": 1.0216,
"step": 2000
},
{
"epoch": 4.94,
"learning_rate": 2.8051450465243574e-05,
"loss": 0.9,
"step": 2010
},
{
"epoch": 4.97,
"learning_rate": 2.7914614121510675e-05,
"loss": 0.9814,
"step": 2020
},
{
"epoch": 4.99,
"learning_rate": 2.777777777777778e-05,
"loss": 0.9785,
"step": 2030
},
{
"epoch": 5.0,
"eval_accuracy": 0.9139152981849611,
"eval_loss": 0.4472927749156952,
"eval_runtime": 27.773,
"eval_samples_per_second": 208.296,
"eval_steps_per_second": 6.517,
"step": 2033
},
{
"epoch": 5.02,
"learning_rate": 2.764094143404488e-05,
"loss": 0.97,
"step": 2040
},
{
"epoch": 5.04,
"learning_rate": 2.7504105090311988e-05,
"loss": 0.9339,
"step": 2050
},
{
"epoch": 5.06,
"learning_rate": 2.7367268746579096e-05,
"loss": 0.8636,
"step": 2060
},
{
"epoch": 5.09,
"learning_rate": 2.7230432402846197e-05,
"loss": 0.8525,
"step": 2070
},
{
"epoch": 5.11,
"learning_rate": 2.70935960591133e-05,
"loss": 0.9281,
"step": 2080
},
{
"epoch": 5.14,
"learning_rate": 2.695675971538041e-05,
"loss": 0.9573,
"step": 2090
},
{
"epoch": 5.16,
"learning_rate": 2.681992337164751e-05,
"loss": 1.0129,
"step": 2100
},
{
"epoch": 5.19,
"learning_rate": 2.6683087027914618e-05,
"loss": 0.9568,
"step": 2110
},
{
"epoch": 5.21,
"learning_rate": 2.6546250684181722e-05,
"loss": 0.8739,
"step": 2120
},
{
"epoch": 5.24,
"learning_rate": 2.6409414340448823e-05,
"loss": 1.0142,
"step": 2130
},
{
"epoch": 5.26,
"learning_rate": 2.627257799671593e-05,
"loss": 0.8615,
"step": 2140
},
{
"epoch": 5.29,
"learning_rate": 2.6135741652983032e-05,
"loss": 0.9213,
"step": 2150
},
{
"epoch": 5.31,
"learning_rate": 2.5998905309250136e-05,
"loss": 0.9751,
"step": 2160
},
{
"epoch": 5.33,
"learning_rate": 2.5862068965517244e-05,
"loss": 0.9144,
"step": 2170
},
{
"epoch": 5.36,
"learning_rate": 2.5725232621784345e-05,
"loss": 0.8717,
"step": 2180
},
{
"epoch": 5.38,
"learning_rate": 2.5588396278051453e-05,
"loss": 1.0529,
"step": 2190
},
{
"epoch": 5.41,
"learning_rate": 2.5451559934318557e-05,
"loss": 0.9158,
"step": 2200
},
{
"epoch": 5.43,
"learning_rate": 2.531472359058566e-05,
"loss": 0.9584,
"step": 2210
},
{
"epoch": 5.46,
"learning_rate": 2.5177887246852766e-05,
"loss": 0.9806,
"step": 2220
},
{
"epoch": 5.48,
"learning_rate": 2.5041050903119874e-05,
"loss": 0.8941,
"step": 2230
},
{
"epoch": 5.51,
"learning_rate": 2.4904214559386975e-05,
"loss": 0.9625,
"step": 2240
},
{
"epoch": 5.53,
"learning_rate": 2.476737821565408e-05,
"loss": 0.9277,
"step": 2250
},
{
"epoch": 5.56,
"learning_rate": 2.4630541871921184e-05,
"loss": 0.9645,
"step": 2260
},
{
"epoch": 5.58,
"learning_rate": 2.4493705528188288e-05,
"loss": 0.9162,
"step": 2270
},
{
"epoch": 5.61,
"learning_rate": 2.4356869184455393e-05,
"loss": 0.8772,
"step": 2280
},
{
"epoch": 5.63,
"learning_rate": 2.4220032840722497e-05,
"loss": 0.9556,
"step": 2290
},
{
"epoch": 5.65,
"learning_rate": 2.40831964969896e-05,
"loss": 0.885,
"step": 2300
},
{
"epoch": 5.68,
"learning_rate": 2.3946360153256706e-05,
"loss": 0.7871,
"step": 2310
},
{
"epoch": 5.7,
"learning_rate": 2.380952380952381e-05,
"loss": 0.8639,
"step": 2320
},
{
"epoch": 5.73,
"learning_rate": 2.3672687465790915e-05,
"loss": 0.9446,
"step": 2330
},
{
"epoch": 5.75,
"learning_rate": 2.353585112205802e-05,
"loss": 0.8905,
"step": 2340
},
{
"epoch": 5.78,
"learning_rate": 2.3399014778325123e-05,
"loss": 0.9502,
"step": 2350
},
{
"epoch": 5.8,
"learning_rate": 2.326217843459223e-05,
"loss": 0.9842,
"step": 2360
},
{
"epoch": 5.83,
"learning_rate": 2.3125342090859332e-05,
"loss": 0.8827,
"step": 2370
},
{
"epoch": 5.85,
"learning_rate": 2.2988505747126437e-05,
"loss": 0.8861,
"step": 2380
},
{
"epoch": 5.88,
"learning_rate": 2.285166940339354e-05,
"loss": 0.9701,
"step": 2390
},
{
"epoch": 5.9,
"learning_rate": 2.271483305966065e-05,
"loss": 0.8456,
"step": 2400
},
{
"epoch": 5.93,
"learning_rate": 2.2577996715927753e-05,
"loss": 0.8858,
"step": 2410
},
{
"epoch": 5.95,
"learning_rate": 2.2441160372194854e-05,
"loss": 0.8431,
"step": 2420
},
{
"epoch": 5.97,
"learning_rate": 2.230432402846196e-05,
"loss": 0.9466,
"step": 2430
},
{
"epoch": 6.0,
"learning_rate": 2.2167487684729066e-05,
"loss": 0.9625,
"step": 2440
},
{
"epoch": 6.0,
"eval_accuracy": 0.9199654278305963,
"eval_loss": 0.3929709494113922,
"eval_runtime": 27.3929,
"eval_samples_per_second": 211.186,
"eval_steps_per_second": 6.608,
"step": 2440
},
{
"epoch": 6.02,
"learning_rate": 2.203065134099617e-05,
"loss": 0.886,
"step": 2450
},
{
"epoch": 6.05,
"learning_rate": 2.1893814997263272e-05,
"loss": 0.8862,
"step": 2460
},
{
"epoch": 6.07,
"learning_rate": 2.175697865353038e-05,
"loss": 0.8938,
"step": 2470
},
{
"epoch": 6.1,
"learning_rate": 2.1620142309797484e-05,
"loss": 0.8165,
"step": 2480
},
{
"epoch": 6.12,
"learning_rate": 2.148330596606459e-05,
"loss": 0.8117,
"step": 2490
},
{
"epoch": 6.15,
"learning_rate": 2.1346469622331693e-05,
"loss": 0.8538,
"step": 2500
},
{
"epoch": 6.17,
"learning_rate": 2.1209633278598797e-05,
"loss": 0.8646,
"step": 2510
},
{
"epoch": 6.2,
"learning_rate": 2.10727969348659e-05,
"loss": 0.8506,
"step": 2520
},
{
"epoch": 6.22,
"learning_rate": 2.0935960591133006e-05,
"loss": 0.9332,
"step": 2530
},
{
"epoch": 6.24,
"learning_rate": 2.079912424740011e-05,
"loss": 0.7804,
"step": 2540
},
{
"epoch": 6.27,
"learning_rate": 2.0662287903667215e-05,
"loss": 0.8503,
"step": 2550
},
{
"epoch": 6.29,
"learning_rate": 2.052545155993432e-05,
"loss": 0.8818,
"step": 2560
},
{
"epoch": 6.32,
"learning_rate": 2.0388615216201424e-05,
"loss": 0.7721,
"step": 2570
},
{
"epoch": 6.34,
"learning_rate": 2.025177887246853e-05,
"loss": 0.8895,
"step": 2580
},
{
"epoch": 6.37,
"learning_rate": 2.0114942528735632e-05,
"loss": 0.8575,
"step": 2590
},
{
"epoch": 6.39,
"learning_rate": 1.9978106185002737e-05,
"loss": 0.8596,
"step": 2600
},
{
"epoch": 6.42,
"learning_rate": 1.984126984126984e-05,
"loss": 0.8394,
"step": 2610
},
{
"epoch": 6.44,
"learning_rate": 1.970443349753695e-05,
"loss": 0.8782,
"step": 2620
},
{
"epoch": 6.47,
"learning_rate": 1.956759715380405e-05,
"loss": 0.7369,
"step": 2630
},
{
"epoch": 6.49,
"learning_rate": 1.9430760810071154e-05,
"loss": 0.8684,
"step": 2640
},
{
"epoch": 6.52,
"learning_rate": 1.929392446633826e-05,
"loss": 0.8494,
"step": 2650
},
{
"epoch": 6.54,
"learning_rate": 1.9157088122605367e-05,
"loss": 0.764,
"step": 2660
},
{
"epoch": 6.56,
"learning_rate": 1.9020251778872468e-05,
"loss": 0.855,
"step": 2670
},
{
"epoch": 6.59,
"learning_rate": 1.8883415435139572e-05,
"loss": 0.8766,
"step": 2680
},
{
"epoch": 6.61,
"learning_rate": 1.874657909140668e-05,
"loss": 0.8737,
"step": 2690
},
{
"epoch": 6.64,
"learning_rate": 1.8609742747673784e-05,
"loss": 0.8441,
"step": 2700
},
{
"epoch": 6.66,
"learning_rate": 1.847290640394089e-05,
"loss": 0.8257,
"step": 2710
},
{
"epoch": 6.69,
"learning_rate": 1.833607006020799e-05,
"loss": 0.8367,
"step": 2720
},
{
"epoch": 6.71,
"learning_rate": 1.8199233716475097e-05,
"loss": 0.8218,
"step": 2730
},
{
"epoch": 6.74,
"learning_rate": 1.8062397372742202e-05,
"loss": 0.7645,
"step": 2740
},
{
"epoch": 6.76,
"learning_rate": 1.7925561029009306e-05,
"loss": 0.8305,
"step": 2750
},
{
"epoch": 6.79,
"learning_rate": 1.7788724685276407e-05,
"loss": 0.8306,
"step": 2760
},
{
"epoch": 6.81,
"learning_rate": 1.7651888341543515e-05,
"loss": 0.9052,
"step": 2770
},
{
"epoch": 6.83,
"learning_rate": 1.751505199781062e-05,
"loss": 0.7477,
"step": 2780
},
{
"epoch": 6.86,
"learning_rate": 1.7378215654077724e-05,
"loss": 0.8458,
"step": 2790
},
{
"epoch": 6.88,
"learning_rate": 1.7241379310344828e-05,
"loss": 0.901,
"step": 2800
},
{
"epoch": 6.91,
"learning_rate": 1.7104542966611933e-05,
"loss": 0.8307,
"step": 2810
},
{
"epoch": 6.93,
"learning_rate": 1.6967706622879037e-05,
"loss": 0.8201,
"step": 2820
},
{
"epoch": 6.96,
"learning_rate": 1.683087027914614e-05,
"loss": 0.8514,
"step": 2830
},
{
"epoch": 6.98,
"learning_rate": 1.6694033935413246e-05,
"loss": 0.8414,
"step": 2840
},
{
"epoch": 7.0,
"eval_accuracy": 0.9327571305099395,
"eval_loss": 0.3537824749946594,
"eval_runtime": 27.9648,
"eval_samples_per_second": 206.867,
"eval_steps_per_second": 6.472,
"step": 2847
},
{
"epoch": 7.01,
"learning_rate": 1.655719759168035e-05,
"loss": 0.9236,
"step": 2850
},
{
"epoch": 7.03,
"learning_rate": 1.6420361247947455e-05,
"loss": 0.8118,
"step": 2860
},
{
"epoch": 7.06,
"learning_rate": 1.628352490421456e-05,
"loss": 0.8715,
"step": 2870
},
{
"epoch": 7.08,
"learning_rate": 1.6146688560481667e-05,
"loss": 0.8181,
"step": 2880
},
{
"epoch": 7.11,
"learning_rate": 1.6009852216748768e-05,
"loss": 0.8577,
"step": 2890
},
{
"epoch": 7.13,
"learning_rate": 1.5873015873015872e-05,
"loss": 0.7951,
"step": 2900
},
{
"epoch": 7.15,
"learning_rate": 1.573617952928298e-05,
"loss": 0.8414,
"step": 2910
},
{
"epoch": 7.18,
"learning_rate": 1.5599343185550085e-05,
"loss": 0.8405,
"step": 2920
},
{
"epoch": 7.2,
"learning_rate": 1.5462506841817186e-05,
"loss": 0.8375,
"step": 2930
},
{
"epoch": 7.23,
"learning_rate": 1.532567049808429e-05,
"loss": 0.8256,
"step": 2940
},
{
"epoch": 7.25,
"learning_rate": 1.5188834154351398e-05,
"loss": 0.9106,
"step": 2950
},
{
"epoch": 7.28,
"learning_rate": 1.50519978106185e-05,
"loss": 0.7944,
"step": 2960
},
{
"epoch": 7.3,
"learning_rate": 1.4915161466885605e-05,
"loss": 0.8647,
"step": 2970
},
{
"epoch": 7.33,
"learning_rate": 1.4778325123152711e-05,
"loss": 0.7921,
"step": 2980
},
{
"epoch": 7.35,
"learning_rate": 1.4641488779419815e-05,
"loss": 0.7978,
"step": 2990
},
{
"epoch": 7.38,
"learning_rate": 1.450465243568692e-05,
"loss": 0.8136,
"step": 3000
},
{
"epoch": 7.4,
"learning_rate": 1.4367816091954022e-05,
"loss": 0.8278,
"step": 3010
},
{
"epoch": 7.42,
"learning_rate": 1.4230979748221129e-05,
"loss": 0.8061,
"step": 3020
},
{
"epoch": 7.45,
"learning_rate": 1.4094143404488233e-05,
"loss": 0.7277,
"step": 3030
},
{
"epoch": 7.47,
"learning_rate": 1.3957307060755337e-05,
"loss": 0.809,
"step": 3040
},
{
"epoch": 7.5,
"learning_rate": 1.382047071702244e-05,
"loss": 0.8359,
"step": 3050
},
{
"epoch": 7.52,
"learning_rate": 1.3683634373289548e-05,
"loss": 0.8107,
"step": 3060
},
{
"epoch": 7.55,
"learning_rate": 1.354679802955665e-05,
"loss": 0.8016,
"step": 3070
},
{
"epoch": 7.57,
"learning_rate": 1.3409961685823755e-05,
"loss": 0.8512,
"step": 3080
},
{
"epoch": 7.6,
"learning_rate": 1.3273125342090861e-05,
"loss": 0.7948,
"step": 3090
},
{
"epoch": 7.62,
"learning_rate": 1.3136288998357965e-05,
"loss": 0.8081,
"step": 3100
},
{
"epoch": 7.65,
"learning_rate": 1.2999452654625068e-05,
"loss": 0.854,
"step": 3110
},
{
"epoch": 7.67,
"learning_rate": 1.2862616310892173e-05,
"loss": 0.8504,
"step": 3120
},
{
"epoch": 7.7,
"learning_rate": 1.2725779967159279e-05,
"loss": 0.824,
"step": 3130
},
{
"epoch": 7.72,
"learning_rate": 1.2588943623426383e-05,
"loss": 0.8119,
"step": 3140
},
{
"epoch": 7.74,
"learning_rate": 1.2452107279693487e-05,
"loss": 0.8304,
"step": 3150
},
{
"epoch": 7.77,
"learning_rate": 1.2315270935960592e-05,
"loss": 0.7737,
"step": 3160
},
{
"epoch": 7.79,
"learning_rate": 1.2178434592227696e-05,
"loss": 0.8032,
"step": 3170
},
{
"epoch": 7.82,
"learning_rate": 1.20415982484948e-05,
"loss": 0.8448,
"step": 3180
},
{
"epoch": 7.84,
"learning_rate": 1.1904761904761905e-05,
"loss": 0.755,
"step": 3190
},
{
"epoch": 7.87,
"learning_rate": 1.176792556102901e-05,
"loss": 0.812,
"step": 3200
},
{
"epoch": 7.89,
"learning_rate": 1.1631089217296116e-05,
"loss": 0.8194,
"step": 3210
},
{
"epoch": 7.92,
"learning_rate": 1.1494252873563218e-05,
"loss": 0.8348,
"step": 3220
},
{
"epoch": 7.94,
"learning_rate": 1.1357416529830324e-05,
"loss": 0.7492,
"step": 3230
},
{
"epoch": 7.97,
"learning_rate": 1.1220580186097427e-05,
"loss": 0.8227,
"step": 3240
},
{
"epoch": 7.99,
"learning_rate": 1.1083743842364533e-05,
"loss": 0.8336,
"step": 3250
},
{
"epoch": 8.0,
"eval_accuracy": 0.9334485738980121,
"eval_loss": 0.3332645893096924,
"eval_runtime": 29.4073,
"eval_samples_per_second": 196.72,
"eval_steps_per_second": 6.155,
"step": 3254
},
{
"epoch": 8.01,
"learning_rate": 1.0946907498631636e-05,
"loss": 0.7665,
"step": 3260
},
{
"epoch": 8.04,
"learning_rate": 1.0810071154898742e-05,
"loss": 0.807,
"step": 3270
},
{
"epoch": 8.06,
"learning_rate": 1.0673234811165846e-05,
"loss": 0.8521,
"step": 3280
},
{
"epoch": 8.09,
"learning_rate": 1.053639846743295e-05,
"loss": 0.7569,
"step": 3290
},
{
"epoch": 8.11,
"learning_rate": 1.0399562123700055e-05,
"loss": 0.7465,
"step": 3300
},
{
"epoch": 8.14,
"learning_rate": 1.026272577996716e-05,
"loss": 0.8432,
"step": 3310
},
{
"epoch": 8.16,
"learning_rate": 1.0125889436234266e-05,
"loss": 0.7659,
"step": 3320
},
{
"epoch": 8.19,
"learning_rate": 9.989053092501368e-06,
"loss": 0.7682,
"step": 3330
},
{
"epoch": 8.21,
"learning_rate": 9.852216748768475e-06,
"loss": 0.7285,
"step": 3340
},
{
"epoch": 8.24,
"learning_rate": 9.715380405035577e-06,
"loss": 0.8721,
"step": 3350
},
{
"epoch": 8.26,
"learning_rate": 9.578544061302683e-06,
"loss": 0.761,
"step": 3360
},
{
"epoch": 8.29,
"learning_rate": 9.441707717569786e-06,
"loss": 0.7812,
"step": 3370
},
{
"epoch": 8.31,
"learning_rate": 9.304871373836892e-06,
"loss": 0.8125,
"step": 3380
},
{
"epoch": 8.33,
"learning_rate": 9.168035030103995e-06,
"loss": 0.8344,
"step": 3390
},
{
"epoch": 8.36,
"learning_rate": 9.031198686371101e-06,
"loss": 0.7884,
"step": 3400
},
{
"epoch": 8.38,
"learning_rate": 8.894362342638204e-06,
"loss": 0.7692,
"step": 3410
},
{
"epoch": 8.41,
"learning_rate": 8.75752599890531e-06,
"loss": 0.875,
"step": 3420
},
{
"epoch": 8.43,
"learning_rate": 8.620689655172414e-06,
"loss": 0.7938,
"step": 3430
},
{
"epoch": 8.46,
"learning_rate": 8.483853311439519e-06,
"loss": 0.7749,
"step": 3440
},
{
"epoch": 8.48,
"learning_rate": 8.347016967706623e-06,
"loss": 0.7628,
"step": 3450
},
{
"epoch": 8.51,
"learning_rate": 8.210180623973727e-06,
"loss": 0.8007,
"step": 3460
},
{
"epoch": 8.53,
"learning_rate": 8.073344280240833e-06,
"loss": 0.7281,
"step": 3470
},
{
"epoch": 8.56,
"learning_rate": 7.936507936507936e-06,
"loss": 0.7633,
"step": 3480
},
{
"epoch": 8.58,
"learning_rate": 7.799671592775042e-06,
"loss": 0.8202,
"step": 3490
},
{
"epoch": 8.6,
"learning_rate": 7.662835249042145e-06,
"loss": 0.8123,
"step": 3500
},
{
"epoch": 8.63,
"learning_rate": 7.52599890530925e-06,
"loss": 0.8223,
"step": 3510
},
{
"epoch": 8.65,
"learning_rate": 7.3891625615763555e-06,
"loss": 0.8456,
"step": 3520
},
{
"epoch": 8.68,
"learning_rate": 7.25232621784346e-06,
"loss": 0.752,
"step": 3530
},
{
"epoch": 8.7,
"learning_rate": 7.115489874110564e-06,
"loss": 0.7942,
"step": 3540
},
{
"epoch": 8.73,
"learning_rate": 6.978653530377669e-06,
"loss": 0.7635,
"step": 3550
},
{
"epoch": 8.75,
"learning_rate": 6.841817186644774e-06,
"loss": 0.7014,
"step": 3560
},
{
"epoch": 8.78,
"learning_rate": 6.7049808429118775e-06,
"loss": 0.7572,
"step": 3570
},
{
"epoch": 8.8,
"learning_rate": 6.568144499178983e-06,
"loss": 0.7816,
"step": 3580
},
{
"epoch": 8.83,
"learning_rate": 6.431308155446086e-06,
"loss": 0.7336,
"step": 3590
},
{
"epoch": 8.85,
"learning_rate": 6.2944718117131915e-06,
"loss": 0.7665,
"step": 3600
},
{
"epoch": 8.88,
"learning_rate": 6.157635467980296e-06,
"loss": 0.7669,
"step": 3610
},
{
"epoch": 8.9,
"learning_rate": 6.0207991242474e-06,
"loss": 0.7587,
"step": 3620
},
{
"epoch": 8.92,
"learning_rate": 5.883962780514505e-06,
"loss": 0.8309,
"step": 3630
},
{
"epoch": 8.95,
"learning_rate": 5.747126436781609e-06,
"loss": 0.7201,
"step": 3640
},
{
"epoch": 8.97,
"learning_rate": 5.6102900930487136e-06,
"loss": 0.736,
"step": 3650
},
{
"epoch": 9.0,
"learning_rate": 5.473453749315818e-06,
"loss": 0.758,
"step": 3660
},
{
"epoch": 9.0,
"eval_accuracy": 0.9405358686257562,
"eval_loss": 0.3180586099624634,
"eval_runtime": 28.3456,
"eval_samples_per_second": 204.088,
"eval_steps_per_second": 6.385,
"step": 3660
},
{
"epoch": 9.02,
"learning_rate": 5.336617405582923e-06,
"loss": 0.6856,
"step": 3670
},
{
"epoch": 9.05,
"learning_rate": 5.199781061850028e-06,
"loss": 0.8134,
"step": 3680
},
{
"epoch": 9.07,
"learning_rate": 5.062944718117133e-06,
"loss": 0.8841,
"step": 3690
},
{
"epoch": 9.1,
"learning_rate": 4.926108374384237e-06,
"loss": 0.7623,
"step": 3700
},
{
"epoch": 9.12,
"learning_rate": 4.789272030651342e-06,
"loss": 0.7488,
"step": 3710
},
{
"epoch": 9.15,
"learning_rate": 4.652435686918446e-06,
"loss": 0.7433,
"step": 3720
},
{
"epoch": 9.17,
"learning_rate": 4.5155993431855505e-06,
"loss": 0.7512,
"step": 3730
},
{
"epoch": 9.19,
"learning_rate": 4.378762999452655e-06,
"loss": 0.7725,
"step": 3740
},
{
"epoch": 9.22,
"learning_rate": 4.241926655719759e-06,
"loss": 0.7567,
"step": 3750
},
{
"epoch": 9.24,
"learning_rate": 4.105090311986864e-06,
"loss": 0.7213,
"step": 3760
},
{
"epoch": 9.27,
"learning_rate": 3.968253968253968e-06,
"loss": 0.75,
"step": 3770
},
{
"epoch": 9.29,
"learning_rate": 3.8314176245210725e-06,
"loss": 0.7999,
"step": 3780
},
{
"epoch": 9.32,
"learning_rate": 3.6945812807881777e-06,
"loss": 0.7489,
"step": 3790
},
{
"epoch": 9.34,
"learning_rate": 3.557744937055282e-06,
"loss": 0.7591,
"step": 3800
},
{
"epoch": 9.37,
"learning_rate": 3.420908593322387e-06,
"loss": 0.769,
"step": 3810
},
{
"epoch": 9.39,
"learning_rate": 3.2840722495894914e-06,
"loss": 0.7171,
"step": 3820
},
{
"epoch": 9.42,
"learning_rate": 3.1472359058565958e-06,
"loss": 0.7158,
"step": 3830
},
{
"epoch": 9.44,
"learning_rate": 3.0103995621237e-06,
"loss": 0.7132,
"step": 3840
},
{
"epoch": 9.47,
"learning_rate": 2.8735632183908046e-06,
"loss": 0.7236,
"step": 3850
},
{
"epoch": 9.49,
"learning_rate": 2.736726874657909e-06,
"loss": 0.7361,
"step": 3860
},
{
"epoch": 9.51,
"learning_rate": 2.599890530925014e-06,
"loss": 0.6774,
"step": 3870
},
{
"epoch": 9.54,
"learning_rate": 2.4630541871921186e-06,
"loss": 0.8082,
"step": 3880
},
{
"epoch": 9.56,
"learning_rate": 2.326217843459223e-06,
"loss": 0.776,
"step": 3890
},
{
"epoch": 9.59,
"learning_rate": 2.1893814997263274e-06,
"loss": 0.7655,
"step": 3900
},
{
"epoch": 9.61,
"learning_rate": 2.052545155993432e-06,
"loss": 0.7834,
"step": 3910
},
{
"epoch": 9.64,
"learning_rate": 1.9157088122605362e-06,
"loss": 0.7236,
"step": 3920
},
{
"epoch": 9.66,
"learning_rate": 1.778872468527641e-06,
"loss": 0.6974,
"step": 3930
},
{
"epoch": 9.69,
"learning_rate": 1.6420361247947457e-06,
"loss": 0.73,
"step": 3940
},
{
"epoch": 9.71,
"learning_rate": 1.50519978106185e-06,
"loss": 0.6768,
"step": 3950
},
{
"epoch": 9.74,
"learning_rate": 1.3683634373289545e-06,
"loss": 0.7841,
"step": 3960
},
{
"epoch": 9.76,
"learning_rate": 1.2315270935960593e-06,
"loss": 0.759,
"step": 3970
},
{
"epoch": 9.78,
"learning_rate": 1.0946907498631637e-06,
"loss": 0.7244,
"step": 3980
},
{
"epoch": 9.81,
"learning_rate": 9.578544061302681e-07,
"loss": 0.7326,
"step": 3990
},
{
"epoch": 9.83,
"learning_rate": 8.210180623973728e-07,
"loss": 0.7484,
"step": 4000
},
{
"epoch": 9.86,
"learning_rate": 6.841817186644772e-07,
"loss": 0.7589,
"step": 4010
},
{
"epoch": 9.88,
"learning_rate": 5.473453749315819e-07,
"loss": 0.7169,
"step": 4020
},
{
"epoch": 9.91,
"learning_rate": 4.105090311986864e-07,
"loss": 0.6762,
"step": 4030
},
{
"epoch": 9.93,
"learning_rate": 2.7367268746579093e-07,
"loss": 0.7896,
"step": 4040
},
{
"epoch": 9.96,
"learning_rate": 1.3683634373289546e-07,
"loss": 0.8115,
"step": 4050
},
{
"epoch": 9.98,
"learning_rate": 0.0,
"loss": 0.7343,
"step": 4060
},
{
"epoch": 9.98,
"eval_accuracy": 0.9403630077787382,
"eval_loss": 0.31005963683128357,
"eval_runtime": 28.1514,
"eval_samples_per_second": 205.496,
"eval_steps_per_second": 6.43,
"step": 4060
},
{
"epoch": 9.98,
"step": 4060,
"total_flos": 1.3140619208067262e+19,
"train_loss": 1.0548507860728673,
"train_runtime": 4291.8856,
"train_samples_per_second": 121.301,
"train_steps_per_second": 0.946
}
],
"logging_steps": 10,
"max_steps": 4060,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 1.3140619208067262e+19,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}