diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,11359 @@ +{ + "best_metric": 0.9235026926819592, + "best_model_checkpoint": "./models/bart-base-spelling-nl/checkpoint-18000", + "epoch": 1.9998254717297765, + "global_step": 18620, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0, + "learning_rate": 0.00029983888292158965, + "loss": 3.3995, + "step": 10 + }, + { + "epoch": 0.0, + "learning_rate": 0.0002996777658431794, + "loss": 1.3033, + "step": 20 + }, + { + "epoch": 0.0, + "learning_rate": 0.00029951664876476906, + "loss": 0.9799, + "step": 30 + }, + { + "epoch": 0.0, + "learning_rate": 0.00029935553168635874, + "loss": 0.8155, + "step": 40 + }, + { + "epoch": 0.01, + "learning_rate": 0.0002991944146079484, + "loss": 0.7491, + "step": 50 + }, + { + "epoch": 0.01, + "learning_rate": 0.0002990332975295381, + "loss": 0.6689, + "step": 60 + }, + { + "epoch": 0.01, + "learning_rate": 0.0002988721804511278, + "loss": 0.6257, + "step": 70 + }, + { + "epoch": 0.01, + "learning_rate": 0.0002987110633727175, + "loss": 0.5962, + "step": 80 + }, + { + "epoch": 0.01, + "learning_rate": 0.0002985499462943072, + "loss": 0.5594, + "step": 90 + }, + { + "epoch": 0.01, + "learning_rate": 0.00029838882921589686, + "loss": 0.5641, + "step": 100 + }, + { + "epoch": 0.01, + "learning_rate": 0.00029822771213748654, + "loss": 0.528, + "step": 110 + }, + { + "epoch": 0.01, + "learning_rate": 0.0002980665950590762, + "loss": 0.4985, + "step": 120 + }, + { + "epoch": 0.01, + "learning_rate": 0.0002979054779806659, + "loss": 0.4914, + "step": 130 + }, + { + "epoch": 0.02, + "learning_rate": 0.0002977443609022556, + "loss": 0.4723, + "step": 140 + }, + { + "epoch": 0.02, + "learning_rate": 0.0002975832438238453, + "loss": 0.4396, + "step": 150 + }, + { + "epoch": 0.02, + "learning_rate": 0.000297422126745435, + "loss": 0.4234, + "step": 160 + }, + { + "epoch": 0.02, + "learning_rate": 0.00029726100966702467, + "loss": 0.4096, + "step": 170 + }, + { + "epoch": 0.02, + "learning_rate": 0.0002970998925886144, + "loss": 0.4114, + "step": 180 + }, + { + "epoch": 0.02, + "learning_rate": 0.0002969387755102041, + "loss": 0.3845, + "step": 190 + }, + { + "epoch": 0.02, + "learning_rate": 0.00029677765843179376, + "loss": 0.391, + "step": 200 + }, + { + "epoch": 0.02, + "learning_rate": 0.00029661654135338343, + "loss": 0.3617, + "step": 210 + }, + { + "epoch": 0.02, + "learning_rate": 0.0002964554242749731, + "loss": 0.3666, + "step": 220 + }, + { + "epoch": 0.02, + "learning_rate": 0.0002962943071965628, + "loss": 0.3587, + "step": 230 + }, + { + "epoch": 0.03, + "learning_rate": 0.0002961331901181525, + "loss": 0.3373, + "step": 240 + }, + { + "epoch": 0.03, + "learning_rate": 0.0002959720730397422, + "loss": 0.3485, + "step": 250 + }, + { + "epoch": 0.03, + "learning_rate": 0.0002958109559613319, + "loss": 0.3216, + "step": 260 + }, + { + "epoch": 0.03, + "learning_rate": 0.00029564983888292156, + "loss": 0.312, + "step": 270 + }, + { + "epoch": 0.03, + "learning_rate": 0.00029548872180451124, + "loss": 0.31, + "step": 280 + }, + { + "epoch": 0.03, + "learning_rate": 0.0002953276047261009, + "loss": 0.3065, + "step": 290 + }, + { + "epoch": 0.03, + "learning_rate": 0.00029516648764769065, + "loss": 0.3132, + "step": 300 + }, + { + "epoch": 0.03, + "learning_rate": 0.0002950053705692803, + "loss": 0.2903, + "step": 310 + }, + { + "epoch": 0.03, + "learning_rate": 0.00029484425349087, + "loss": 0.2843, + "step": 320 + }, + { + "epoch": 0.04, + "learning_rate": 0.00029468313641245974, + "loss": 0.2673, + "step": 330 + }, + { + "epoch": 0.04, + "learning_rate": 0.00029452201933404936, + "loss": 0.2844, + "step": 340 + }, + { + "epoch": 0.04, + "learning_rate": 0.00029436090225563904, + "loss": 0.2661, + "step": 350 + }, + { + "epoch": 0.04, + "learning_rate": 0.00029419978517722877, + "loss": 0.2634, + "step": 360 + }, + { + "epoch": 0.04, + "learning_rate": 0.00029403866809881845, + "loss": 0.2632, + "step": 370 + }, + { + "epoch": 0.04, + "learning_rate": 0.0002938775510204081, + "loss": 0.2609, + "step": 380 + }, + { + "epoch": 0.04, + "learning_rate": 0.00029371643394199786, + "loss": 0.2575, + "step": 390 + }, + { + "epoch": 0.04, + "learning_rate": 0.00029355531686358754, + "loss": 0.2638, + "step": 400 + }, + { + "epoch": 0.04, + "learning_rate": 0.0002933941997851772, + "loss": 0.2723, + "step": 410 + }, + { + "epoch": 0.05, + "learning_rate": 0.0002932330827067669, + "loss": 0.2607, + "step": 420 + }, + { + "epoch": 0.05, + "learning_rate": 0.00029307196562835657, + "loss": 0.2553, + "step": 430 + }, + { + "epoch": 0.05, + "learning_rate": 0.00029291084854994625, + "loss": 0.2455, + "step": 440 + }, + { + "epoch": 0.05, + "learning_rate": 0.00029274973147153593, + "loss": 0.2255, + "step": 450 + }, + { + "epoch": 0.05, + "learning_rate": 0.00029258861439312566, + "loss": 0.2303, + "step": 460 + }, + { + "epoch": 0.05, + "learning_rate": 0.00029242749731471534, + "loss": 0.2274, + "step": 470 + }, + { + "epoch": 0.05, + "learning_rate": 0.000292266380236305, + "loss": 0.229, + "step": 480 + }, + { + "epoch": 0.05, + "learning_rate": 0.0002921052631578947, + "loss": 0.2394, + "step": 490 + }, + { + "epoch": 0.05, + "learning_rate": 0.0002919441460794844, + "loss": 0.2418, + "step": 500 + }, + { + "epoch": 0.05, + "learning_rate": 0.00029178302900107405, + "loss": 0.2087, + "step": 510 + }, + { + "epoch": 0.06, + "learning_rate": 0.0002916219119226638, + "loss": 0.2234, + "step": 520 + }, + { + "epoch": 0.06, + "learning_rate": 0.00029146079484425346, + "loss": 0.2128, + "step": 530 + }, + { + "epoch": 0.06, + "learning_rate": 0.00029129967776584314, + "loss": 0.206, + "step": 540 + }, + { + "epoch": 0.06, + "learning_rate": 0.0002911385606874329, + "loss": 0.223, + "step": 550 + }, + { + "epoch": 0.06, + "learning_rate": 0.00029097744360902255, + "loss": 0.218, + "step": 560 + }, + { + "epoch": 0.06, + "learning_rate": 0.00029081632653061223, + "loss": 0.1962, + "step": 570 + }, + { + "epoch": 0.06, + "learning_rate": 0.0002906552094522019, + "loss": 0.1958, + "step": 580 + }, + { + "epoch": 0.06, + "learning_rate": 0.0002904940923737916, + "loss": 0.1971, + "step": 590 + }, + { + "epoch": 0.06, + "learning_rate": 0.00029033297529538126, + "loss": 0.2048, + "step": 600 + }, + { + "epoch": 0.07, + "learning_rate": 0.000290171858216971, + "loss": 0.1772, + "step": 610 + }, + { + "epoch": 0.07, + "learning_rate": 0.0002900107411385607, + "loss": 0.2031, + "step": 620 + }, + { + "epoch": 0.07, + "learning_rate": 0.00028984962406015035, + "loss": 0.1767, + "step": 630 + }, + { + "epoch": 0.07, + "learning_rate": 0.00028968850698174003, + "loss": 0.1969, + "step": 640 + }, + { + "epoch": 0.07, + "learning_rate": 0.0002895273899033297, + "loss": 0.1853, + "step": 650 + }, + { + "epoch": 0.07, + "learning_rate": 0.0002893662728249194, + "loss": 0.2021, + "step": 660 + }, + { + "epoch": 0.07, + "learning_rate": 0.0002892051557465091, + "loss": 0.1833, + "step": 670 + }, + { + "epoch": 0.07, + "learning_rate": 0.0002890440386680988, + "loss": 0.1677, + "step": 680 + }, + { + "epoch": 0.07, + "learning_rate": 0.0002888829215896885, + "loss": 0.1773, + "step": 690 + }, + { + "epoch": 0.08, + "learning_rate": 0.00028872180451127816, + "loss": 0.169, + "step": 700 + }, + { + "epoch": 0.08, + "learning_rate": 0.0002885606874328679, + "loss": 0.1727, + "step": 710 + }, + { + "epoch": 0.08, + "learning_rate": 0.00028839957035445757, + "loss": 0.1679, + "step": 720 + }, + { + "epoch": 0.08, + "learning_rate": 0.00028823845327604724, + "loss": 0.1702, + "step": 730 + }, + { + "epoch": 0.08, + "learning_rate": 0.0002880773361976369, + "loss": 0.183, + "step": 740 + }, + { + "epoch": 0.08, + "learning_rate": 0.0002879162191192266, + "loss": 0.1653, + "step": 750 + }, + { + "epoch": 0.08, + "learning_rate": 0.0002877551020408163, + "loss": 0.1816, + "step": 760 + }, + { + "epoch": 0.08, + "learning_rate": 0.000287593984962406, + "loss": 0.1666, + "step": 770 + }, + { + "epoch": 0.08, + "learning_rate": 0.0002874328678839957, + "loss": 0.1701, + "step": 780 + }, + { + "epoch": 0.08, + "learning_rate": 0.00028727175080558537, + "loss": 0.1685, + "step": 790 + }, + { + "epoch": 0.09, + "learning_rate": 0.00028711063372717505, + "loss": 0.1598, + "step": 800 + }, + { + "epoch": 0.09, + "learning_rate": 0.0002869495166487647, + "loss": 0.1671, + "step": 810 + }, + { + "epoch": 0.09, + "learning_rate": 0.0002867883995703544, + "loss": 0.1749, + "step": 820 + }, + { + "epoch": 0.09, + "learning_rate": 0.00028662728249194414, + "loss": 0.1565, + "step": 830 + }, + { + "epoch": 0.09, + "learning_rate": 0.0002864661654135338, + "loss": 0.1806, + "step": 840 + }, + { + "epoch": 0.09, + "learning_rate": 0.0002863050483351235, + "loss": 0.1769, + "step": 850 + }, + { + "epoch": 0.09, + "learning_rate": 0.0002861439312567132, + "loss": 0.1533, + "step": 860 + }, + { + "epoch": 0.09, + "learning_rate": 0.0002859828141783029, + "loss": 0.1618, + "step": 870 + }, + { + "epoch": 0.09, + "learning_rate": 0.0002858216970998926, + "loss": 0.1711, + "step": 880 + }, + { + "epoch": 0.1, + "learning_rate": 0.00028566058002148226, + "loss": 0.1572, + "step": 890 + }, + { + "epoch": 0.1, + "learning_rate": 0.00028549946294307194, + "loss": 0.1529, + "step": 900 + }, + { + "epoch": 0.1, + "learning_rate": 0.0002853383458646616, + "loss": 0.1694, + "step": 910 + }, + { + "epoch": 0.1, + "learning_rate": 0.00028517722878625135, + "loss": 0.1399, + "step": 920 + }, + { + "epoch": 0.1, + "learning_rate": 0.000285016111707841, + "loss": 0.1515, + "step": 930 + }, + { + "epoch": 0.1, + "learning_rate": 0.0002848549946294307, + "loss": 0.1526, + "step": 940 + }, + { + "epoch": 0.1, + "learning_rate": 0.0002846938775510204, + "loss": 0.1528, + "step": 950 + }, + { + "epoch": 0.1, + "learning_rate": 0.00028453276047261006, + "loss": 0.1499, + "step": 960 + }, + { + "epoch": 0.1, + "learning_rate": 0.00028437164339419974, + "loss": 0.1633, + "step": 970 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002842105263157894, + "loss": 0.1453, + "step": 980 + }, + { + "epoch": 0.11, + "learning_rate": 0.00028404940923737915, + "loss": 0.1434, + "step": 990 + }, + { + "epoch": 0.11, + "learning_rate": 0.00028388829215896883, + "loss": 0.1617, + "step": 1000 + }, + { + "epoch": 0.11, + "eval_cer": 0.9240525162766596, + "eval_loss": 0.09863634407520294, + "eval_runtime": 119.1412, + "eval_samples_per_second": 16.77, + "eval_steps_per_second": 4.197, + "step": 1000 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002837271750805585, + "loss": 0.1476, + "step": 1010 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002835660580021482, + "loss": 0.1549, + "step": 1020 + }, + { + "epoch": 0.11, + "learning_rate": 0.00028340494092373786, + "loss": 0.1531, + "step": 1030 + }, + { + "epoch": 0.11, + "learning_rate": 0.00028324382384532754, + "loss": 0.1462, + "step": 1040 + }, + { + "epoch": 0.11, + "learning_rate": 0.0002830827067669173, + "loss": 0.1269, + "step": 1050 + }, + { + "epoch": 0.11, + "learning_rate": 0.00028292158968850695, + "loss": 0.1414, + "step": 1060 + }, + { + "epoch": 0.11, + "learning_rate": 0.00028276047261009663, + "loss": 0.1357, + "step": 1070 + }, + { + "epoch": 0.12, + "learning_rate": 0.00028259935553168636, + "loss": 0.1336, + "step": 1080 + }, + { + "epoch": 0.12, + "learning_rate": 0.00028243823845327604, + "loss": 0.1358, + "step": 1090 + }, + { + "epoch": 0.12, + "learning_rate": 0.0002822771213748657, + "loss": 0.1354, + "step": 1100 + }, + { + "epoch": 0.12, + "learning_rate": 0.0002821160042964554, + "loss": 0.1385, + "step": 1110 + }, + { + "epoch": 0.12, + "learning_rate": 0.0002819548872180451, + "loss": 0.1426, + "step": 1120 + }, + { + "epoch": 0.12, + "learning_rate": 0.00028179377013963475, + "loss": 0.1369, + "step": 1130 + }, + { + "epoch": 0.12, + "learning_rate": 0.0002816326530612245, + "loss": 0.1422, + "step": 1140 + }, + { + "epoch": 0.12, + "learning_rate": 0.00028147153598281416, + "loss": 0.1173, + "step": 1150 + }, + { + "epoch": 0.12, + "learning_rate": 0.00028131041890440384, + "loss": 0.1344, + "step": 1160 + }, + { + "epoch": 0.13, + "learning_rate": 0.0002811493018259935, + "loss": 0.1308, + "step": 1170 + }, + { + "epoch": 0.13, + "learning_rate": 0.0002809881847475832, + "loss": 0.125, + "step": 1180 + }, + { + "epoch": 0.13, + "learning_rate": 0.0002808270676691729, + "loss": 0.1329, + "step": 1190 + }, + { + "epoch": 0.13, + "learning_rate": 0.0002806659505907626, + "loss": 0.1316, + "step": 1200 + }, + { + "epoch": 0.13, + "learning_rate": 0.0002805048335123523, + "loss": 0.1369, + "step": 1210 + }, + { + "epoch": 0.13, + "learning_rate": 0.00028034371643394197, + "loss": 0.1274, + "step": 1220 + }, + { + "epoch": 0.13, + "learning_rate": 0.0002801825993555317, + "loss": 0.1411, + "step": 1230 + }, + { + "epoch": 0.13, + "learning_rate": 0.0002800214822771214, + "loss": 0.1387, + "step": 1240 + }, + { + "epoch": 0.13, + "learning_rate": 0.00027986036519871106, + "loss": 0.1367, + "step": 1250 + }, + { + "epoch": 0.14, + "learning_rate": 0.00027969924812030073, + "loss": 0.1264, + "step": 1260 + }, + { + "epoch": 0.14, + "learning_rate": 0.0002795381310418904, + "loss": 0.1392, + "step": 1270 + }, + { + "epoch": 0.14, + "learning_rate": 0.0002793770139634801, + "loss": 0.1539, + "step": 1280 + }, + { + "epoch": 0.14, + "learning_rate": 0.00027921589688506977, + "loss": 0.1337, + "step": 1290 + }, + { + "epoch": 0.14, + "learning_rate": 0.0002790547798066595, + "loss": 0.127, + "step": 1300 + }, + { + "epoch": 0.14, + "learning_rate": 0.0002788936627282492, + "loss": 0.131, + "step": 1310 + }, + { + "epoch": 0.14, + "learning_rate": 0.00027873254564983886, + "loss": 0.13, + "step": 1320 + }, + { + "epoch": 0.14, + "learning_rate": 0.00027857142857142854, + "loss": 0.1284, + "step": 1330 + }, + { + "epoch": 0.14, + "learning_rate": 0.0002784103114930182, + "loss": 0.134, + "step": 1340 + }, + { + "epoch": 0.14, + "learning_rate": 0.0002782491944146079, + "loss": 0.1252, + "step": 1350 + }, + { + "epoch": 0.15, + "learning_rate": 0.0002780880773361976, + "loss": 0.129, + "step": 1360 + }, + { + "epoch": 0.15, + "learning_rate": 0.0002779269602577873, + "loss": 0.1209, + "step": 1370 + }, + { + "epoch": 0.15, + "learning_rate": 0.000277765843179377, + "loss": 0.1267, + "step": 1380 + }, + { + "epoch": 0.15, + "learning_rate": 0.0002776047261009667, + "loss": 0.1292, + "step": 1390 + }, + { + "epoch": 0.15, + "learning_rate": 0.0002774436090225564, + "loss": 0.1247, + "step": 1400 + }, + { + "epoch": 0.15, + "learning_rate": 0.00027728249194414607, + "loss": 0.1347, + "step": 1410 + }, + { + "epoch": 0.15, + "learning_rate": 0.00027712137486573575, + "loss": 0.1298, + "step": 1420 + }, + { + "epoch": 0.15, + "learning_rate": 0.0002769602577873254, + "loss": 0.1339, + "step": 1430 + }, + { + "epoch": 0.15, + "learning_rate": 0.0002767991407089151, + "loss": 0.1212, + "step": 1440 + }, + { + "epoch": 0.16, + "learning_rate": 0.00027663802363050484, + "loss": 0.1208, + "step": 1450 + }, + { + "epoch": 0.16, + "learning_rate": 0.0002764769065520945, + "loss": 0.1383, + "step": 1460 + }, + { + "epoch": 0.16, + "learning_rate": 0.0002763157894736842, + "loss": 0.1338, + "step": 1470 + }, + { + "epoch": 0.16, + "learning_rate": 0.00027615467239527387, + "loss": 0.1332, + "step": 1480 + }, + { + "epoch": 0.16, + "learning_rate": 0.00027599355531686355, + "loss": 0.1183, + "step": 1490 + }, + { + "epoch": 0.16, + "learning_rate": 0.00027583243823845323, + "loss": 0.1165, + "step": 1500 + }, + { + "epoch": 0.16, + "learning_rate": 0.0002756713211600429, + "loss": 0.121, + "step": 1510 + }, + { + "epoch": 0.16, + "learning_rate": 0.00027551020408163264, + "loss": 0.1258, + "step": 1520 + }, + { + "epoch": 0.16, + "learning_rate": 0.0002753490870032223, + "loss": 0.1183, + "step": 1530 + }, + { + "epoch": 0.17, + "learning_rate": 0.000275187969924812, + "loss": 0.1173, + "step": 1540 + }, + { + "epoch": 0.17, + "learning_rate": 0.00027502685284640173, + "loss": 0.1375, + "step": 1550 + }, + { + "epoch": 0.17, + "learning_rate": 0.0002748657357679914, + "loss": 0.1234, + "step": 1560 + }, + { + "epoch": 0.17, + "learning_rate": 0.0002747046186895811, + "loss": 0.1251, + "step": 1570 + }, + { + "epoch": 0.17, + "learning_rate": 0.00027454350161117076, + "loss": 0.1072, + "step": 1580 + }, + { + "epoch": 0.17, + "learning_rate": 0.00027438238453276044, + "loss": 0.1234, + "step": 1590 + }, + { + "epoch": 0.17, + "learning_rate": 0.0002742212674543501, + "loss": 0.1234, + "step": 1600 + }, + { + "epoch": 0.17, + "learning_rate": 0.00027406015037593985, + "loss": 0.1087, + "step": 1610 + }, + { + "epoch": 0.17, + "learning_rate": 0.00027389903329752953, + "loss": 0.1094, + "step": 1620 + }, + { + "epoch": 0.18, + "learning_rate": 0.0002737379162191192, + "loss": 0.1095, + "step": 1630 + }, + { + "epoch": 0.18, + "learning_rate": 0.0002735767991407089, + "loss": 0.122, + "step": 1640 + }, + { + "epoch": 0.18, + "learning_rate": 0.00027341568206229856, + "loss": 0.1001, + "step": 1650 + }, + { + "epoch": 0.18, + "learning_rate": 0.00027325456498388824, + "loss": 0.1097, + "step": 1660 + }, + { + "epoch": 0.18, + "learning_rate": 0.000273093447905478, + "loss": 0.1126, + "step": 1670 + }, + { + "epoch": 0.18, + "learning_rate": 0.00027293233082706765, + "loss": 0.1214, + "step": 1680 + }, + { + "epoch": 0.18, + "learning_rate": 0.00027277121374865733, + "loss": 0.1129, + "step": 1690 + }, + { + "epoch": 0.18, + "learning_rate": 0.000272610096670247, + "loss": 0.1197, + "step": 1700 + }, + { + "epoch": 0.18, + "learning_rate": 0.0002724489795918367, + "loss": 0.1195, + "step": 1710 + }, + { + "epoch": 0.18, + "learning_rate": 0.00027228786251342637, + "loss": 0.113, + "step": 1720 + }, + { + "epoch": 0.19, + "learning_rate": 0.0002721267454350161, + "loss": 0.1124, + "step": 1730 + }, + { + "epoch": 0.19, + "learning_rate": 0.0002719656283566058, + "loss": 0.1193, + "step": 1740 + }, + { + "epoch": 0.19, + "learning_rate": 0.00027180451127819546, + "loss": 0.1131, + "step": 1750 + }, + { + "epoch": 0.19, + "learning_rate": 0.0002716433941997852, + "loss": 0.107, + "step": 1760 + }, + { + "epoch": 0.19, + "learning_rate": 0.00027148227712137487, + "loss": 0.1159, + "step": 1770 + }, + { + "epoch": 0.19, + "learning_rate": 0.00027132116004296454, + "loss": 0.1028, + "step": 1780 + }, + { + "epoch": 0.19, + "learning_rate": 0.0002711600429645542, + "loss": 0.117, + "step": 1790 + }, + { + "epoch": 0.19, + "learning_rate": 0.0002709989258861439, + "loss": 0.1126, + "step": 1800 + }, + { + "epoch": 0.19, + "learning_rate": 0.0002708378088077336, + "loss": 0.0996, + "step": 1810 + }, + { + "epoch": 0.2, + "learning_rate": 0.00027067669172932326, + "loss": 0.1003, + "step": 1820 + }, + { + "epoch": 0.2, + "learning_rate": 0.000270515574650913, + "loss": 0.104, + "step": 1830 + }, + { + "epoch": 0.2, + "learning_rate": 0.00027035445757250267, + "loss": 0.1059, + "step": 1840 + }, + { + "epoch": 0.2, + "learning_rate": 0.00027019334049409235, + "loss": 0.1078, + "step": 1850 + }, + { + "epoch": 0.2, + "learning_rate": 0.000270032223415682, + "loss": 0.1105, + "step": 1860 + }, + { + "epoch": 0.2, + "learning_rate": 0.0002698711063372717, + "loss": 0.0925, + "step": 1870 + }, + { + "epoch": 0.2, + "learning_rate": 0.0002697099892588614, + "loss": 0.1024, + "step": 1880 + }, + { + "epoch": 0.2, + "learning_rate": 0.0002695488721804511, + "loss": 0.0929, + "step": 1890 + }, + { + "epoch": 0.2, + "learning_rate": 0.0002693877551020408, + "loss": 0.101, + "step": 1900 + }, + { + "epoch": 0.21, + "learning_rate": 0.00026922663802363047, + "loss": 0.1105, + "step": 1910 + }, + { + "epoch": 0.21, + "learning_rate": 0.0002690655209452202, + "loss": 0.1156, + "step": 1920 + }, + { + "epoch": 0.21, + "learning_rate": 0.0002689044038668099, + "loss": 0.1042, + "step": 1930 + }, + { + "epoch": 0.21, + "learning_rate": 0.00026874328678839956, + "loss": 0.0993, + "step": 1940 + }, + { + "epoch": 0.21, + "learning_rate": 0.00026858216970998924, + "loss": 0.1004, + "step": 1950 + }, + { + "epoch": 0.21, + "learning_rate": 0.0002684210526315789, + "loss": 0.1021, + "step": 1960 + }, + { + "epoch": 0.21, + "learning_rate": 0.0002682599355531686, + "loss": 0.115, + "step": 1970 + }, + { + "epoch": 0.21, + "learning_rate": 0.0002680988184747583, + "loss": 0.112, + "step": 1980 + }, + { + "epoch": 0.21, + "learning_rate": 0.000267937701396348, + "loss": 0.1057, + "step": 1990 + }, + { + "epoch": 0.21, + "learning_rate": 0.0002677765843179377, + "loss": 0.1326, + "step": 2000 + }, + { + "epoch": 0.21, + "eval_cer": 0.9240131820848148, + "eval_loss": 0.0675765872001648, + "eval_runtime": 119.2298, + "eval_samples_per_second": 16.758, + "eval_steps_per_second": 4.194, + "step": 2000 + }, + { + "epoch": 0.22, + "learning_rate": 0.00026761546723952736, + "loss": 0.1176, + "step": 2010 + }, + { + "epoch": 0.22, + "learning_rate": 0.00026745435016111704, + "loss": 0.111, + "step": 2020 + }, + { + "epoch": 0.22, + "learning_rate": 0.0002672932330827067, + "loss": 0.1068, + "step": 2030 + }, + { + "epoch": 0.22, + "learning_rate": 0.00026713211600429645, + "loss": 0.1056, + "step": 2040 + }, + { + "epoch": 0.22, + "learning_rate": 0.00026697099892588613, + "loss": 0.1058, + "step": 2050 + }, + { + "epoch": 0.22, + "learning_rate": 0.0002668098818474758, + "loss": 0.116, + "step": 2060 + }, + { + "epoch": 0.22, + "learning_rate": 0.0002666487647690655, + "loss": 0.1197, + "step": 2070 + }, + { + "epoch": 0.22, + "learning_rate": 0.0002664876476906552, + "loss": 0.1138, + "step": 2080 + }, + { + "epoch": 0.22, + "learning_rate": 0.0002663265306122449, + "loss": 0.1079, + "step": 2090 + }, + { + "epoch": 0.23, + "learning_rate": 0.0002661654135338346, + "loss": 0.094, + "step": 2100 + }, + { + "epoch": 0.23, + "learning_rate": 0.00026600429645542425, + "loss": 0.0961, + "step": 2110 + }, + { + "epoch": 0.23, + "learning_rate": 0.00026584317937701393, + "loss": 0.1026, + "step": 2120 + }, + { + "epoch": 0.23, + "learning_rate": 0.0002656820622986036, + "loss": 0.1099, + "step": 2130 + }, + { + "epoch": 0.23, + "learning_rate": 0.00026552094522019334, + "loss": 0.1093, + "step": 2140 + }, + { + "epoch": 0.23, + "learning_rate": 0.000265359828141783, + "loss": 0.1048, + "step": 2150 + }, + { + "epoch": 0.23, + "learning_rate": 0.0002651987110633727, + "loss": 0.125, + "step": 2160 + }, + { + "epoch": 0.23, + "learning_rate": 0.0002650375939849624, + "loss": 0.099, + "step": 2170 + }, + { + "epoch": 0.23, + "learning_rate": 0.00026487647690655205, + "loss": 0.1033, + "step": 2180 + }, + { + "epoch": 0.24, + "learning_rate": 0.00026471535982814173, + "loss": 0.0946, + "step": 2190 + }, + { + "epoch": 0.24, + "learning_rate": 0.00026455424274973146, + "loss": 0.1169, + "step": 2200 + }, + { + "epoch": 0.24, + "learning_rate": 0.00026439312567132114, + "loss": 0.0982, + "step": 2210 + }, + { + "epoch": 0.24, + "learning_rate": 0.0002642320085929108, + "loss": 0.1095, + "step": 2220 + }, + { + "epoch": 0.24, + "learning_rate": 0.00026407089151450055, + "loss": 0.1036, + "step": 2230 + }, + { + "epoch": 0.24, + "learning_rate": 0.00026390977443609023, + "loss": 0.0981, + "step": 2240 + }, + { + "epoch": 0.24, + "learning_rate": 0.00026374865735767986, + "loss": 0.1017, + "step": 2250 + }, + { + "epoch": 0.24, + "learning_rate": 0.0002635875402792696, + "loss": 0.1, + "step": 2260 + }, + { + "epoch": 0.24, + "learning_rate": 0.00026342642320085927, + "loss": 0.1098, + "step": 2270 + }, + { + "epoch": 0.24, + "learning_rate": 0.00026326530612244894, + "loss": 0.1094, + "step": 2280 + }, + { + "epoch": 0.25, + "learning_rate": 0.0002631041890440387, + "loss": 0.1027, + "step": 2290 + }, + { + "epoch": 0.25, + "learning_rate": 0.00026294307196562836, + "loss": 0.0903, + "step": 2300 + }, + { + "epoch": 0.25, + "learning_rate": 0.00026278195488721803, + "loss": 0.1128, + "step": 2310 + }, + { + "epoch": 0.25, + "learning_rate": 0.0002626208378088077, + "loss": 0.0984, + "step": 2320 + }, + { + "epoch": 0.25, + "learning_rate": 0.0002624597207303974, + "loss": 0.1024, + "step": 2330 + }, + { + "epoch": 0.25, + "learning_rate": 0.00026229860365198707, + "loss": 0.0926, + "step": 2340 + }, + { + "epoch": 0.25, + "learning_rate": 0.00026213748657357675, + "loss": 0.1096, + "step": 2350 + }, + { + "epoch": 0.25, + "learning_rate": 0.0002619763694951665, + "loss": 0.0965, + "step": 2360 + }, + { + "epoch": 0.25, + "learning_rate": 0.00026181525241675616, + "loss": 0.0932, + "step": 2370 + }, + { + "epoch": 0.26, + "learning_rate": 0.00026165413533834584, + "loss": 0.1009, + "step": 2380 + }, + { + "epoch": 0.26, + "learning_rate": 0.0002614930182599355, + "loss": 0.0993, + "step": 2390 + }, + { + "epoch": 0.26, + "learning_rate": 0.0002613319011815252, + "loss": 0.1002, + "step": 2400 + }, + { + "epoch": 0.26, + "learning_rate": 0.00026117078410311487, + "loss": 0.0904, + "step": 2410 + }, + { + "epoch": 0.26, + "learning_rate": 0.0002610096670247046, + "loss": 0.0951, + "step": 2420 + }, + { + "epoch": 0.26, + "learning_rate": 0.0002608485499462943, + "loss": 0.0888, + "step": 2430 + }, + { + "epoch": 0.26, + "learning_rate": 0.00026068743286788396, + "loss": 0.1052, + "step": 2440 + }, + { + "epoch": 0.26, + "learning_rate": 0.0002605263157894737, + "loss": 0.0986, + "step": 2450 + }, + { + "epoch": 0.26, + "learning_rate": 0.00026036519871106337, + "loss": 0.0931, + "step": 2460 + }, + { + "epoch": 0.27, + "learning_rate": 0.00026020408163265305, + "loss": 0.0978, + "step": 2470 + }, + { + "epoch": 0.27, + "learning_rate": 0.0002600429645542427, + "loss": 0.1049, + "step": 2480 + }, + { + "epoch": 0.27, + "learning_rate": 0.0002598818474758324, + "loss": 0.1014, + "step": 2490 + }, + { + "epoch": 0.27, + "learning_rate": 0.0002597207303974221, + "loss": 0.0893, + "step": 2500 + }, + { + "epoch": 0.27, + "learning_rate": 0.0002595596133190118, + "loss": 0.114, + "step": 2510 + }, + { + "epoch": 0.27, + "learning_rate": 0.0002593984962406015, + "loss": 0.0946, + "step": 2520 + }, + { + "epoch": 0.27, + "learning_rate": 0.00025923737916219117, + "loss": 0.0894, + "step": 2530 + }, + { + "epoch": 0.27, + "learning_rate": 0.00025907626208378085, + "loss": 0.1111, + "step": 2540 + }, + { + "epoch": 0.27, + "learning_rate": 0.00025891514500537053, + "loss": 0.1149, + "step": 2550 + }, + { + "epoch": 0.27, + "learning_rate": 0.0002587540279269602, + "loss": 0.0927, + "step": 2560 + }, + { + "epoch": 0.28, + "learning_rate": 0.00025859291084854994, + "loss": 0.0956, + "step": 2570 + }, + { + "epoch": 0.28, + "learning_rate": 0.0002584317937701396, + "loss": 0.0836, + "step": 2580 + }, + { + "epoch": 0.28, + "learning_rate": 0.0002582706766917293, + "loss": 0.0973, + "step": 2590 + }, + { + "epoch": 0.28, + "learning_rate": 0.00025810955961331903, + "loss": 0.0933, + "step": 2600 + }, + { + "epoch": 0.28, + "learning_rate": 0.0002579484425349087, + "loss": 0.0897, + "step": 2610 + }, + { + "epoch": 0.28, + "learning_rate": 0.0002577873254564984, + "loss": 0.1161, + "step": 2620 + }, + { + "epoch": 0.28, + "learning_rate": 0.00025762620837808806, + "loss": 0.0974, + "step": 2630 + }, + { + "epoch": 0.28, + "learning_rate": 0.00025746509129967774, + "loss": 0.0935, + "step": 2640 + }, + { + "epoch": 0.28, + "learning_rate": 0.0002573039742212674, + "loss": 0.0919, + "step": 2650 + }, + { + "epoch": 0.29, + "learning_rate": 0.0002571428571428571, + "loss": 0.0948, + "step": 2660 + }, + { + "epoch": 0.29, + "learning_rate": 0.00025698174006444683, + "loss": 0.086, + "step": 2670 + }, + { + "epoch": 0.29, + "learning_rate": 0.0002568206229860365, + "loss": 0.0895, + "step": 2680 + }, + { + "epoch": 0.29, + "learning_rate": 0.0002566595059076262, + "loss": 0.0983, + "step": 2690 + }, + { + "epoch": 0.29, + "learning_rate": 0.00025649838882921586, + "loss": 0.0843, + "step": 2700 + }, + { + "epoch": 0.29, + "learning_rate": 0.00025633727175080554, + "loss": 0.0887, + "step": 2710 + }, + { + "epoch": 0.29, + "learning_rate": 0.0002561761546723952, + "loss": 0.1018, + "step": 2720 + }, + { + "epoch": 0.29, + "learning_rate": 0.00025601503759398495, + "loss": 0.0956, + "step": 2730 + }, + { + "epoch": 0.29, + "learning_rate": 0.00025585392051557463, + "loss": 0.0863, + "step": 2740 + }, + { + "epoch": 0.3, + "learning_rate": 0.0002556928034371643, + "loss": 0.0809, + "step": 2750 + }, + { + "epoch": 0.3, + "learning_rate": 0.00025553168635875404, + "loss": 0.0886, + "step": 2760 + }, + { + "epoch": 0.3, + "learning_rate": 0.0002553705692803437, + "loss": 0.0978, + "step": 2770 + }, + { + "epoch": 0.3, + "learning_rate": 0.0002552094522019334, + "loss": 0.1044, + "step": 2780 + }, + { + "epoch": 0.3, + "learning_rate": 0.0002550483351235231, + "loss": 0.0914, + "step": 2790 + }, + { + "epoch": 0.3, + "learning_rate": 0.00025488721804511276, + "loss": 0.0847, + "step": 2800 + }, + { + "epoch": 0.3, + "learning_rate": 0.00025472610096670243, + "loss": 0.0919, + "step": 2810 + }, + { + "epoch": 0.3, + "learning_rate": 0.00025456498388829217, + "loss": 0.108, + "step": 2820 + }, + { + "epoch": 0.3, + "learning_rate": 0.00025440386680988184, + "loss": 0.0873, + "step": 2830 + }, + { + "epoch": 0.31, + "learning_rate": 0.0002542427497314715, + "loss": 0.0965, + "step": 2840 + }, + { + "epoch": 0.31, + "learning_rate": 0.0002540816326530612, + "loss": 0.0886, + "step": 2850 + }, + { + "epoch": 0.31, + "learning_rate": 0.0002539205155746509, + "loss": 0.0848, + "step": 2860 + }, + { + "epoch": 0.31, + "learning_rate": 0.00025375939849624056, + "loss": 0.0786, + "step": 2870 + }, + { + "epoch": 0.31, + "learning_rate": 0.00025359828141783024, + "loss": 0.083, + "step": 2880 + }, + { + "epoch": 0.31, + "learning_rate": 0.00025343716433941997, + "loss": 0.0759, + "step": 2890 + }, + { + "epoch": 0.31, + "learning_rate": 0.00025327604726100965, + "loss": 0.0887, + "step": 2900 + }, + { + "epoch": 0.31, + "learning_rate": 0.0002531149301825993, + "loss": 0.0919, + "step": 2910 + }, + { + "epoch": 0.31, + "learning_rate": 0.000252953813104189, + "loss": 0.0965, + "step": 2920 + }, + { + "epoch": 0.31, + "learning_rate": 0.0002527926960257787, + "loss": 0.0886, + "step": 2930 + }, + { + "epoch": 0.32, + "learning_rate": 0.00025263157894736836, + "loss": 0.0839, + "step": 2940 + }, + { + "epoch": 0.32, + "learning_rate": 0.0002524704618689581, + "loss": 0.0949, + "step": 2950 + }, + { + "epoch": 0.32, + "learning_rate": 0.00025230934479054777, + "loss": 0.093, + "step": 2960 + }, + { + "epoch": 0.32, + "learning_rate": 0.00025214822771213745, + "loss": 0.0813, + "step": 2970 + }, + { + "epoch": 0.32, + "learning_rate": 0.0002519871106337272, + "loss": 0.0924, + "step": 2980 + }, + { + "epoch": 0.32, + "learning_rate": 0.00025182599355531686, + "loss": 0.0928, + "step": 2990 + }, + { + "epoch": 0.32, + "learning_rate": 0.00025166487647690654, + "loss": 0.09, + "step": 3000 + }, + { + "epoch": 0.32, + "eval_cer": 0.9240704731903279, + "eval_loss": 0.05863150954246521, + "eval_runtime": 118.8661, + "eval_samples_per_second": 16.809, + "eval_steps_per_second": 4.206, + "step": 3000 + }, + { + "epoch": 0.32, + "learning_rate": 0.0002515037593984962, + "loss": 0.094, + "step": 3010 + }, + { + "epoch": 0.32, + "learning_rate": 0.0002513426423200859, + "loss": 0.0981, + "step": 3020 + }, + { + "epoch": 0.33, + "learning_rate": 0.00025118152524167557, + "loss": 0.0854, + "step": 3030 + }, + { + "epoch": 0.33, + "learning_rate": 0.0002510204081632653, + "loss": 0.0993, + "step": 3040 + }, + { + "epoch": 0.33, + "learning_rate": 0.000250859291084855, + "loss": 0.0928, + "step": 3050 + }, + { + "epoch": 0.33, + "learning_rate": 0.00025069817400644466, + "loss": 0.0853, + "step": 3060 + }, + { + "epoch": 0.33, + "learning_rate": 0.00025053705692803434, + "loss": 0.0901, + "step": 3070 + }, + { + "epoch": 0.33, + "learning_rate": 0.000250375939849624, + "loss": 0.0957, + "step": 3080 + }, + { + "epoch": 0.33, + "learning_rate": 0.0002502148227712137, + "loss": 0.0904, + "step": 3090 + }, + { + "epoch": 0.33, + "learning_rate": 0.00025005370569280343, + "loss": 0.0836, + "step": 3100 + }, + { + "epoch": 0.33, + "learning_rate": 0.0002498925886143931, + "loss": 0.0792, + "step": 3110 + }, + { + "epoch": 0.34, + "learning_rate": 0.0002497314715359828, + "loss": 0.0761, + "step": 3120 + }, + { + "epoch": 0.34, + "learning_rate": 0.0002495703544575725, + "loss": 0.101, + "step": 3130 + }, + { + "epoch": 0.34, + "learning_rate": 0.0002494092373791622, + "loss": 0.0777, + "step": 3140 + }, + { + "epoch": 0.34, + "learning_rate": 0.0002492481203007519, + "loss": 0.0874, + "step": 3150 + }, + { + "epoch": 0.34, + "learning_rate": 0.00024908700322234155, + "loss": 0.0879, + "step": 3160 + }, + { + "epoch": 0.34, + "learning_rate": 0.00024892588614393123, + "loss": 0.0727, + "step": 3170 + }, + { + "epoch": 0.34, + "learning_rate": 0.0002487647690655209, + "loss": 0.0794, + "step": 3180 + }, + { + "epoch": 0.34, + "learning_rate": 0.0002486036519871106, + "loss": 0.0816, + "step": 3190 + }, + { + "epoch": 0.34, + "learning_rate": 0.0002484425349087003, + "loss": 0.0848, + "step": 3200 + }, + { + "epoch": 0.34, + "learning_rate": 0.00024828141783029, + "loss": 0.084, + "step": 3210 + }, + { + "epoch": 0.35, + "learning_rate": 0.0002481203007518797, + "loss": 0.0887, + "step": 3220 + }, + { + "epoch": 0.35, + "learning_rate": 0.00024795918367346935, + "loss": 0.0887, + "step": 3230 + }, + { + "epoch": 0.35, + "learning_rate": 0.00024779806659505903, + "loss": 0.0836, + "step": 3240 + }, + { + "epoch": 0.35, + "learning_rate": 0.0002476369495166487, + "loss": 0.097, + "step": 3250 + }, + { + "epoch": 0.35, + "learning_rate": 0.00024747583243823844, + "loss": 0.0915, + "step": 3260 + }, + { + "epoch": 0.35, + "learning_rate": 0.0002473147153598281, + "loss": 0.0854, + "step": 3270 + }, + { + "epoch": 0.35, + "learning_rate": 0.0002471535982814178, + "loss": 0.0987, + "step": 3280 + }, + { + "epoch": 0.35, + "learning_rate": 0.00024699248120300753, + "loss": 0.088, + "step": 3290 + }, + { + "epoch": 0.35, + "learning_rate": 0.0002468313641245972, + "loss": 0.0929, + "step": 3300 + }, + { + "epoch": 0.36, + "learning_rate": 0.0002466702470461869, + "loss": 0.09, + "step": 3310 + }, + { + "epoch": 0.36, + "learning_rate": 0.00024650912996777657, + "loss": 0.0871, + "step": 3320 + }, + { + "epoch": 0.36, + "learning_rate": 0.00024634801288936624, + "loss": 0.0821, + "step": 3330 + }, + { + "epoch": 0.36, + "learning_rate": 0.0002461868958109559, + "loss": 0.0961, + "step": 3340 + }, + { + "epoch": 0.36, + "learning_rate": 0.00024602577873254565, + "loss": 0.087, + "step": 3350 + }, + { + "epoch": 0.36, + "learning_rate": 0.00024586466165413533, + "loss": 0.0826, + "step": 3360 + }, + { + "epoch": 0.36, + "learning_rate": 0.000245703544575725, + "loss": 0.0837, + "step": 3370 + }, + { + "epoch": 0.36, + "learning_rate": 0.0002455424274973147, + "loss": 0.0788, + "step": 3380 + }, + { + "epoch": 0.36, + "learning_rate": 0.00024538131041890437, + "loss": 0.0833, + "step": 3390 + }, + { + "epoch": 0.37, + "learning_rate": 0.00024522019334049405, + "loss": 0.0833, + "step": 3400 + }, + { + "epoch": 0.37, + "learning_rate": 0.0002450590762620838, + "loss": 0.091, + "step": 3410 + }, + { + "epoch": 0.37, + "learning_rate": 0.00024489795918367346, + "loss": 0.0968, + "step": 3420 + }, + { + "epoch": 0.37, + "learning_rate": 0.00024473684210526314, + "loss": 0.0886, + "step": 3430 + }, + { + "epoch": 0.37, + "learning_rate": 0.0002445757250268528, + "loss": 0.1, + "step": 3440 + }, + { + "epoch": 0.37, + "learning_rate": 0.00024441460794844255, + "loss": 0.0917, + "step": 3450 + }, + { + "epoch": 0.37, + "learning_rate": 0.0002442534908700322, + "loss": 0.0764, + "step": 3460 + }, + { + "epoch": 0.37, + "learning_rate": 0.00024409237379162188, + "loss": 0.0802, + "step": 3470 + }, + { + "epoch": 0.37, + "learning_rate": 0.00024393125671321158, + "loss": 0.0901, + "step": 3480 + }, + { + "epoch": 0.37, + "learning_rate": 0.00024377013963480126, + "loss": 0.0953, + "step": 3490 + }, + { + "epoch": 0.38, + "learning_rate": 0.00024360902255639094, + "loss": 0.0778, + "step": 3500 + }, + { + "epoch": 0.38, + "learning_rate": 0.00024344790547798064, + "loss": 0.0828, + "step": 3510 + }, + { + "epoch": 0.38, + "learning_rate": 0.00024328678839957032, + "loss": 0.1002, + "step": 3520 + }, + { + "epoch": 0.38, + "learning_rate": 0.00024312567132116, + "loss": 0.0888, + "step": 3530 + }, + { + "epoch": 0.38, + "learning_rate": 0.00024296455424274973, + "loss": 0.0779, + "step": 3540 + }, + { + "epoch": 0.38, + "learning_rate": 0.0002428034371643394, + "loss": 0.0789, + "step": 3550 + }, + { + "epoch": 0.38, + "learning_rate": 0.0002426423200859291, + "loss": 0.0947, + "step": 3560 + }, + { + "epoch": 0.38, + "learning_rate": 0.0002424812030075188, + "loss": 0.078, + "step": 3570 + }, + { + "epoch": 0.38, + "learning_rate": 0.00024232008592910847, + "loss": 0.0945, + "step": 3580 + }, + { + "epoch": 0.39, + "learning_rate": 0.00024215896885069815, + "loss": 0.0784, + "step": 3590 + }, + { + "epoch": 0.39, + "learning_rate": 0.00024199785177228785, + "loss": 0.089, + "step": 3600 + }, + { + "epoch": 0.39, + "learning_rate": 0.00024183673469387753, + "loss": 0.0757, + "step": 3610 + }, + { + "epoch": 0.39, + "learning_rate": 0.0002416756176154672, + "loss": 0.0883, + "step": 3620 + }, + { + "epoch": 0.39, + "learning_rate": 0.00024151450053705692, + "loss": 0.0864, + "step": 3630 + }, + { + "epoch": 0.39, + "learning_rate": 0.0002413533834586466, + "loss": 0.0858, + "step": 3640 + }, + { + "epoch": 0.39, + "learning_rate": 0.00024119226638023627, + "loss": 0.0876, + "step": 3650 + }, + { + "epoch": 0.39, + "learning_rate": 0.00024103114930182598, + "loss": 0.0905, + "step": 3660 + }, + { + "epoch": 0.39, + "learning_rate": 0.00024087003222341566, + "loss": 0.0824, + "step": 3670 + }, + { + "epoch": 0.4, + "learning_rate": 0.00024070891514500534, + "loss": 0.0809, + "step": 3680 + }, + { + "epoch": 0.4, + "learning_rate": 0.00024054779806659507, + "loss": 0.079, + "step": 3690 + }, + { + "epoch": 0.4, + "learning_rate": 0.00024038668098818472, + "loss": 0.0903, + "step": 3700 + }, + { + "epoch": 0.4, + "learning_rate": 0.0002402255639097744, + "loss": 0.0819, + "step": 3710 + }, + { + "epoch": 0.4, + "learning_rate": 0.00024006444683136408, + "loss": 0.1012, + "step": 3720 + }, + { + "epoch": 0.4, + "learning_rate": 0.0002399033297529538, + "loss": 0.0848, + "step": 3730 + }, + { + "epoch": 0.4, + "learning_rate": 0.00023974221267454349, + "loss": 0.1144, + "step": 3740 + }, + { + "epoch": 0.4, + "learning_rate": 0.00023958109559613316, + "loss": 0.0733, + "step": 3750 + }, + { + "epoch": 0.4, + "learning_rate": 0.00023941997851772287, + "loss": 0.0824, + "step": 3760 + }, + { + "epoch": 0.4, + "learning_rate": 0.00023925886143931255, + "loss": 0.0918, + "step": 3770 + }, + { + "epoch": 0.41, + "learning_rate": 0.00023909774436090223, + "loss": 0.0959, + "step": 3780 + }, + { + "epoch": 0.41, + "learning_rate": 0.00023893662728249193, + "loss": 0.0852, + "step": 3790 + }, + { + "epoch": 0.41, + "learning_rate": 0.0002387755102040816, + "loss": 0.073, + "step": 3800 + }, + { + "epoch": 0.41, + "learning_rate": 0.0002386143931256713, + "loss": 0.0795, + "step": 3810 + }, + { + "epoch": 0.41, + "learning_rate": 0.000238453276047261, + "loss": 0.0767, + "step": 3820 + }, + { + "epoch": 0.41, + "learning_rate": 0.00023829215896885067, + "loss": 0.0829, + "step": 3830 + }, + { + "epoch": 0.41, + "learning_rate": 0.00023813104189044035, + "loss": 0.0818, + "step": 3840 + }, + { + "epoch": 0.41, + "learning_rate": 0.00023796992481203005, + "loss": 0.0874, + "step": 3850 + }, + { + "epoch": 0.41, + "learning_rate": 0.00023780880773361973, + "loss": 0.0789, + "step": 3860 + }, + { + "epoch": 0.42, + "learning_rate": 0.0002376476906552094, + "loss": 0.0872, + "step": 3870 + }, + { + "epoch": 0.42, + "learning_rate": 0.00023748657357679914, + "loss": 0.0862, + "step": 3880 + }, + { + "epoch": 0.42, + "learning_rate": 0.00023732545649838882, + "loss": 0.082, + "step": 3890 + }, + { + "epoch": 0.42, + "learning_rate": 0.0002371643394199785, + "loss": 0.0836, + "step": 3900 + }, + { + "epoch": 0.42, + "learning_rate": 0.0002370032223415682, + "loss": 0.0819, + "step": 3910 + }, + { + "epoch": 0.42, + "learning_rate": 0.00023684210526315788, + "loss": 0.0763, + "step": 3920 + }, + { + "epoch": 0.42, + "learning_rate": 0.00023668098818474756, + "loss": 0.0898, + "step": 3930 + }, + { + "epoch": 0.42, + "learning_rate": 0.00023651987110633727, + "loss": 0.0739, + "step": 3940 + }, + { + "epoch": 0.42, + "learning_rate": 0.00023635875402792695, + "loss": 0.0885, + "step": 3950 + }, + { + "epoch": 0.43, + "learning_rate": 0.00023619763694951662, + "loss": 0.0846, + "step": 3960 + }, + { + "epoch": 0.43, + "learning_rate": 0.00023603651987110633, + "loss": 0.0764, + "step": 3970 + }, + { + "epoch": 0.43, + "learning_rate": 0.000235875402792696, + "loss": 0.0748, + "step": 3980 + }, + { + "epoch": 0.43, + "learning_rate": 0.00023571428571428569, + "loss": 0.0763, + "step": 3990 + }, + { + "epoch": 0.43, + "learning_rate": 0.00023555316863587536, + "loss": 0.0891, + "step": 4000 + }, + { + "epoch": 0.43, + "eval_cer": 0.9239704275284617, + "eval_loss": 0.05304262414574623, + "eval_runtime": 118.902, + "eval_samples_per_second": 16.804, + "eval_steps_per_second": 4.205, + "step": 4000 + }, + { + "epoch": 0.43, + "learning_rate": 0.00023539205155746507, + "loss": 0.0696, + "step": 4010 + }, + { + "epoch": 0.43, + "learning_rate": 0.00023523093447905475, + "loss": 0.0851, + "step": 4020 + }, + { + "epoch": 0.43, + "learning_rate": 0.00023506981740064443, + "loss": 0.0877, + "step": 4030 + }, + { + "epoch": 0.43, + "learning_rate": 0.00023490870032223413, + "loss": 0.0833, + "step": 4040 + }, + { + "epoch": 0.43, + "learning_rate": 0.0002347475832438238, + "loss": 0.0859, + "step": 4050 + }, + { + "epoch": 0.44, + "learning_rate": 0.0002345864661654135, + "loss": 0.0736, + "step": 4060 + }, + { + "epoch": 0.44, + "learning_rate": 0.00023442534908700322, + "loss": 0.0787, + "step": 4070 + }, + { + "epoch": 0.44, + "learning_rate": 0.0002342642320085929, + "loss": 0.0854, + "step": 4080 + }, + { + "epoch": 0.44, + "learning_rate": 0.00023410311493018258, + "loss": 0.0832, + "step": 4090 + }, + { + "epoch": 0.44, + "learning_rate": 0.00023394199785177228, + "loss": 0.0869, + "step": 4100 + }, + { + "epoch": 0.44, + "learning_rate": 0.00023378088077336196, + "loss": 0.0737, + "step": 4110 + }, + { + "epoch": 0.44, + "learning_rate": 0.00023361976369495164, + "loss": 0.0882, + "step": 4120 + }, + { + "epoch": 0.44, + "learning_rate": 0.00023345864661654134, + "loss": 0.0832, + "step": 4130 + }, + { + "epoch": 0.44, + "learning_rate": 0.00023329752953813102, + "loss": 0.0879, + "step": 4140 + }, + { + "epoch": 0.45, + "learning_rate": 0.0002331364124597207, + "loss": 0.0808, + "step": 4150 + }, + { + "epoch": 0.45, + "learning_rate": 0.0002329752953813104, + "loss": 0.081, + "step": 4160 + }, + { + "epoch": 0.45, + "learning_rate": 0.00023281417830290008, + "loss": 0.0921, + "step": 4170 + }, + { + "epoch": 0.45, + "learning_rate": 0.00023265306122448976, + "loss": 0.0706, + "step": 4180 + }, + { + "epoch": 0.45, + "learning_rate": 0.00023249194414607947, + "loss": 0.0712, + "step": 4190 + }, + { + "epoch": 0.45, + "learning_rate": 0.00023233082706766915, + "loss": 0.0809, + "step": 4200 + }, + { + "epoch": 0.45, + "learning_rate": 0.00023216970998925882, + "loss": 0.0641, + "step": 4210 + }, + { + "epoch": 0.45, + "learning_rate": 0.00023200859291084856, + "loss": 0.0739, + "step": 4220 + }, + { + "epoch": 0.45, + "learning_rate": 0.00023184747583243823, + "loss": 0.0717, + "step": 4230 + }, + { + "epoch": 0.46, + "learning_rate": 0.0002316863587540279, + "loss": 0.0815, + "step": 4240 + }, + { + "epoch": 0.46, + "learning_rate": 0.0002315252416756176, + "loss": 0.0774, + "step": 4250 + }, + { + "epoch": 0.46, + "learning_rate": 0.0002313641245972073, + "loss": 0.0825, + "step": 4260 + }, + { + "epoch": 0.46, + "learning_rate": 0.00023120300751879697, + "loss": 0.08, + "step": 4270 + }, + { + "epoch": 0.46, + "learning_rate": 0.00023104189044038665, + "loss": 0.0762, + "step": 4280 + }, + { + "epoch": 0.46, + "learning_rate": 0.00023088077336197636, + "loss": 0.072, + "step": 4290 + }, + { + "epoch": 0.46, + "learning_rate": 0.00023071965628356604, + "loss": 0.0966, + "step": 4300 + }, + { + "epoch": 0.46, + "learning_rate": 0.00023055853920515571, + "loss": 0.1086, + "step": 4310 + }, + { + "epoch": 0.46, + "learning_rate": 0.00023039742212674542, + "loss": 0.0917, + "step": 4320 + }, + { + "epoch": 0.47, + "learning_rate": 0.0002302363050483351, + "loss": 0.0904, + "step": 4330 + }, + { + "epoch": 0.47, + "learning_rate": 0.00023007518796992478, + "loss": 0.0793, + "step": 4340 + }, + { + "epoch": 0.47, + "learning_rate": 0.00022991407089151448, + "loss": 0.0804, + "step": 4350 + }, + { + "epoch": 0.47, + "learning_rate": 0.00022975295381310416, + "loss": 0.0774, + "step": 4360 + }, + { + "epoch": 0.47, + "learning_rate": 0.00022959183673469384, + "loss": 0.0797, + "step": 4370 + }, + { + "epoch": 0.47, + "learning_rate": 0.00022943071965628354, + "loss": 0.0821, + "step": 4380 + }, + { + "epoch": 0.47, + "learning_rate": 0.00022926960257787322, + "loss": 0.0712, + "step": 4390 + }, + { + "epoch": 0.47, + "learning_rate": 0.0002291084854994629, + "loss": 0.0787, + "step": 4400 + }, + { + "epoch": 0.47, + "learning_rate": 0.00022894736842105263, + "loss": 0.0774, + "step": 4410 + }, + { + "epoch": 0.47, + "learning_rate": 0.0002287862513426423, + "loss": 0.0749, + "step": 4420 + }, + { + "epoch": 0.48, + "learning_rate": 0.000228625134264232, + "loss": 0.0822, + "step": 4430 + }, + { + "epoch": 0.48, + "learning_rate": 0.0002284640171858217, + "loss": 0.085, + "step": 4440 + }, + { + "epoch": 0.48, + "learning_rate": 0.00022830290010741137, + "loss": 0.0806, + "step": 4450 + }, + { + "epoch": 0.48, + "learning_rate": 0.00022814178302900105, + "loss": 0.0895, + "step": 4460 + }, + { + "epoch": 0.48, + "learning_rate": 0.00022798066595059076, + "loss": 0.0787, + "step": 4470 + }, + { + "epoch": 0.48, + "learning_rate": 0.00022781954887218043, + "loss": 0.0793, + "step": 4480 + }, + { + "epoch": 0.48, + "learning_rate": 0.0002276584317937701, + "loss": 0.0831, + "step": 4490 + }, + { + "epoch": 0.48, + "learning_rate": 0.00022749731471535982, + "loss": 0.0685, + "step": 4500 + }, + { + "epoch": 0.48, + "learning_rate": 0.0002273361976369495, + "loss": 0.0702, + "step": 4510 + }, + { + "epoch": 0.49, + "learning_rate": 0.00022717508055853917, + "loss": 0.0758, + "step": 4520 + }, + { + "epoch": 0.49, + "learning_rate": 0.00022701396348012885, + "loss": 0.0817, + "step": 4530 + }, + { + "epoch": 0.49, + "learning_rate": 0.00022685284640171856, + "loss": 0.0775, + "step": 4540 + }, + { + "epoch": 0.49, + "learning_rate": 0.00022669172932330824, + "loss": 0.0791, + "step": 4550 + }, + { + "epoch": 0.49, + "learning_rate": 0.00022653061224489791, + "loss": 0.0813, + "step": 4560 + }, + { + "epoch": 0.49, + "learning_rate": 0.00022636949516648765, + "loss": 0.0777, + "step": 4570 + }, + { + "epoch": 0.49, + "learning_rate": 0.00022620837808807733, + "loss": 0.0747, + "step": 4580 + }, + { + "epoch": 0.49, + "learning_rate": 0.000226047261009667, + "loss": 0.0696, + "step": 4590 + }, + { + "epoch": 0.49, + "learning_rate": 0.0002258861439312567, + "loss": 0.0847, + "step": 4600 + }, + { + "epoch": 0.5, + "learning_rate": 0.0002257250268528464, + "loss": 0.0699, + "step": 4610 + }, + { + "epoch": 0.5, + "learning_rate": 0.00022556390977443607, + "loss": 0.0784, + "step": 4620 + }, + { + "epoch": 0.5, + "learning_rate": 0.00022540279269602577, + "loss": 0.078, + "step": 4630 + }, + { + "epoch": 0.5, + "learning_rate": 0.00022524167561761545, + "loss": 0.0757, + "step": 4640 + }, + { + "epoch": 0.5, + "learning_rate": 0.00022508055853920513, + "loss": 0.0681, + "step": 4650 + }, + { + "epoch": 0.5, + "learning_rate": 0.00022491944146079483, + "loss": 0.0733, + "step": 4660 + }, + { + "epoch": 0.5, + "learning_rate": 0.0002247583243823845, + "loss": 0.0661, + "step": 4670 + }, + { + "epoch": 0.5, + "learning_rate": 0.0002245972073039742, + "loss": 0.089, + "step": 4680 + }, + { + "epoch": 0.5, + "learning_rate": 0.0002244360902255639, + "loss": 0.0748, + "step": 4690 + }, + { + "epoch": 0.5, + "learning_rate": 0.00022427497314715357, + "loss": 0.0693, + "step": 4700 + }, + { + "epoch": 0.51, + "learning_rate": 0.00022411385606874325, + "loss": 0.0818, + "step": 4710 + }, + { + "epoch": 0.51, + "learning_rate": 0.00022395273899033296, + "loss": 0.0689, + "step": 4720 + }, + { + "epoch": 0.51, + "learning_rate": 0.00022379162191192263, + "loss": 0.0716, + "step": 4730 + }, + { + "epoch": 0.51, + "learning_rate": 0.0002236305048335123, + "loss": 0.0676, + "step": 4740 + }, + { + "epoch": 0.51, + "learning_rate": 0.00022346938775510205, + "loss": 0.0818, + "step": 4750 + }, + { + "epoch": 0.51, + "learning_rate": 0.00022330827067669172, + "loss": 0.0714, + "step": 4760 + }, + { + "epoch": 0.51, + "learning_rate": 0.0002231471535982814, + "loss": 0.0734, + "step": 4770 + }, + { + "epoch": 0.51, + "learning_rate": 0.0002229860365198711, + "loss": 0.0759, + "step": 4780 + }, + { + "epoch": 0.51, + "learning_rate": 0.00022282491944146079, + "loss": 0.0758, + "step": 4790 + }, + { + "epoch": 0.52, + "learning_rate": 0.00022266380236305046, + "loss": 0.0679, + "step": 4800 + }, + { + "epoch": 0.52, + "learning_rate": 0.00022250268528464014, + "loss": 0.0769, + "step": 4810 + }, + { + "epoch": 0.52, + "learning_rate": 0.00022234156820622985, + "loss": 0.0761, + "step": 4820 + }, + { + "epoch": 0.52, + "learning_rate": 0.00022218045112781953, + "loss": 0.0687, + "step": 4830 + }, + { + "epoch": 0.52, + "learning_rate": 0.0002220193340494092, + "loss": 0.0702, + "step": 4840 + }, + { + "epoch": 0.52, + "learning_rate": 0.0002218582169709989, + "loss": 0.0799, + "step": 4850 + }, + { + "epoch": 0.52, + "learning_rate": 0.0002216970998925886, + "loss": 0.0724, + "step": 4860 + }, + { + "epoch": 0.52, + "learning_rate": 0.00022153598281417827, + "loss": 0.07, + "step": 4870 + }, + { + "epoch": 0.52, + "learning_rate": 0.00022137486573576797, + "loss": 0.0616, + "step": 4880 + }, + { + "epoch": 0.53, + "learning_rate": 0.00022121374865735765, + "loss": 0.0659, + "step": 4890 + }, + { + "epoch": 0.53, + "learning_rate": 0.00022105263157894733, + "loss": 0.0778, + "step": 4900 + }, + { + "epoch": 0.53, + "learning_rate": 0.00022089151450053706, + "loss": 0.0668, + "step": 4910 + }, + { + "epoch": 0.53, + "learning_rate": 0.00022073039742212674, + "loss": 0.0748, + "step": 4920 + }, + { + "epoch": 0.53, + "learning_rate": 0.00022056928034371642, + "loss": 0.082, + "step": 4930 + }, + { + "epoch": 0.53, + "learning_rate": 0.00022040816326530612, + "loss": 0.0782, + "step": 4940 + }, + { + "epoch": 0.53, + "learning_rate": 0.0002202470461868958, + "loss": 0.0759, + "step": 4950 + }, + { + "epoch": 0.53, + "learning_rate": 0.00022008592910848548, + "loss": 0.079, + "step": 4960 + }, + { + "epoch": 0.53, + "learning_rate": 0.00021992481203007518, + "loss": 0.0758, + "step": 4970 + }, + { + "epoch": 0.53, + "learning_rate": 0.00021976369495166486, + "loss": 0.0857, + "step": 4980 + }, + { + "epoch": 0.54, + "learning_rate": 0.00021960257787325454, + "loss": 0.079, + "step": 4990 + }, + { + "epoch": 0.54, + "learning_rate": 0.00021944146079484425, + "loss": 0.0753, + "step": 5000 + }, + { + "epoch": 0.54, + "eval_cer": 0.9239088609673133, + "eval_loss": 0.04911934211850166, + "eval_runtime": 119.4674, + "eval_samples_per_second": 16.724, + "eval_steps_per_second": 4.185, + "step": 5000 + }, + { + "epoch": 0.54, + "learning_rate": 0.00021928034371643392, + "loss": 0.0798, + "step": 5010 + }, + { + "epoch": 0.54, + "learning_rate": 0.0002191192266380236, + "loss": 0.0665, + "step": 5020 + }, + { + "epoch": 0.54, + "learning_rate": 0.0002189581095596133, + "loss": 0.0781, + "step": 5030 + }, + { + "epoch": 0.54, + "learning_rate": 0.00021879699248120299, + "loss": 0.0892, + "step": 5040 + }, + { + "epoch": 0.54, + "learning_rate": 0.00021863587540279266, + "loss": 0.0668, + "step": 5050 + }, + { + "epoch": 0.54, + "learning_rate": 0.00021847475832438237, + "loss": 0.0731, + "step": 5060 + }, + { + "epoch": 0.54, + "learning_rate": 0.00021831364124597205, + "loss": 0.0692, + "step": 5070 + }, + { + "epoch": 0.55, + "learning_rate": 0.00021815252416756173, + "loss": 0.072, + "step": 5080 + }, + { + "epoch": 0.55, + "learning_rate": 0.0002179914070891514, + "loss": 0.0769, + "step": 5090 + }, + { + "epoch": 0.55, + "learning_rate": 0.00021783029001074114, + "loss": 0.0631, + "step": 5100 + }, + { + "epoch": 0.55, + "learning_rate": 0.00021766917293233081, + "loss": 0.0684, + "step": 5110 + }, + { + "epoch": 0.55, + "learning_rate": 0.0002175080558539205, + "loss": 0.065, + "step": 5120 + }, + { + "epoch": 0.55, + "learning_rate": 0.0002173469387755102, + "loss": 0.0681, + "step": 5130 + }, + { + "epoch": 0.55, + "learning_rate": 0.00021718582169709988, + "loss": 0.0739, + "step": 5140 + }, + { + "epoch": 0.55, + "learning_rate": 0.00021702470461868955, + "loss": 0.0685, + "step": 5150 + }, + { + "epoch": 0.55, + "learning_rate": 0.00021686358754027926, + "loss": 0.0694, + "step": 5160 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021670247046186894, + "loss": 0.0716, + "step": 5170 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021654135338345862, + "loss": 0.0666, + "step": 5180 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021638023630504832, + "loss": 0.0693, + "step": 5190 + }, + { + "epoch": 0.56, + "learning_rate": 0.000216219119226638, + "loss": 0.0743, + "step": 5200 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021605800214822768, + "loss": 0.0775, + "step": 5210 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021589688506981738, + "loss": 0.0698, + "step": 5220 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021573576799140706, + "loss": 0.0736, + "step": 5230 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021557465091299674, + "loss": 0.0659, + "step": 5240 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021541353383458647, + "loss": 0.0702, + "step": 5250 + }, + { + "epoch": 0.56, + "learning_rate": 0.00021525241675617615, + "loss": 0.0759, + "step": 5260 + }, + { + "epoch": 0.57, + "learning_rate": 0.00021509129967776583, + "loss": 0.071, + "step": 5270 + }, + { + "epoch": 0.57, + "learning_rate": 0.00021493018259935553, + "loss": 0.0672, + "step": 5280 + }, + { + "epoch": 0.57, + "learning_rate": 0.0002147690655209452, + "loss": 0.0694, + "step": 5290 + }, + { + "epoch": 0.57, + "learning_rate": 0.0002146079484425349, + "loss": 0.0739, + "step": 5300 + }, + { + "epoch": 0.57, + "learning_rate": 0.0002144468313641246, + "loss": 0.0736, + "step": 5310 + }, + { + "epoch": 0.57, + "learning_rate": 0.00021428571428571427, + "loss": 0.0641, + "step": 5320 + }, + { + "epoch": 0.57, + "learning_rate": 0.00021412459720730395, + "loss": 0.0691, + "step": 5330 + }, + { + "epoch": 0.57, + "learning_rate": 0.00021396348012889363, + "loss": 0.067, + "step": 5340 + }, + { + "epoch": 0.57, + "learning_rate": 0.00021380236305048334, + "loss": 0.0758, + "step": 5350 + }, + { + "epoch": 0.58, + "learning_rate": 0.00021364124597207301, + "loss": 0.0726, + "step": 5360 + }, + { + "epoch": 0.58, + "learning_rate": 0.0002134801288936627, + "loss": 0.0804, + "step": 5370 + }, + { + "epoch": 0.58, + "learning_rate": 0.0002133190118152524, + "loss": 0.073, + "step": 5380 + }, + { + "epoch": 0.58, + "learning_rate": 0.00021315789473684208, + "loss": 0.0797, + "step": 5390 + }, + { + "epoch": 0.58, + "learning_rate": 0.00021299677765843175, + "loss": 0.08, + "step": 5400 + }, + { + "epoch": 0.58, + "learning_rate": 0.00021283566058002146, + "loss": 0.0738, + "step": 5410 + }, + { + "epoch": 0.58, + "learning_rate": 0.00021267454350161114, + "loss": 0.0767, + "step": 5420 + }, + { + "epoch": 0.58, + "learning_rate": 0.00021251342642320082, + "loss": 0.0666, + "step": 5430 + }, + { + "epoch": 0.58, + "learning_rate": 0.00021235230934479055, + "loss": 0.0631, + "step": 5440 + }, + { + "epoch": 0.59, + "learning_rate": 0.00021219119226638023, + "loss": 0.0751, + "step": 5450 + }, + { + "epoch": 0.59, + "learning_rate": 0.0002120300751879699, + "loss": 0.0733, + "step": 5460 + }, + { + "epoch": 0.59, + "learning_rate": 0.0002118689581095596, + "loss": 0.0674, + "step": 5470 + }, + { + "epoch": 0.59, + "learning_rate": 0.0002117078410311493, + "loss": 0.0725, + "step": 5480 + }, + { + "epoch": 0.59, + "learning_rate": 0.00021154672395273897, + "loss": 0.0815, + "step": 5490 + }, + { + "epoch": 0.59, + "learning_rate": 0.00021138560687432867, + "loss": 0.0705, + "step": 5500 + }, + { + "epoch": 0.59, + "learning_rate": 0.00021122448979591835, + "loss": 0.068, + "step": 5510 + }, + { + "epoch": 0.59, + "learning_rate": 0.00021106337271750803, + "loss": 0.0642, + "step": 5520 + }, + { + "epoch": 0.59, + "learning_rate": 0.00021090225563909773, + "loss": 0.0676, + "step": 5530 + }, + { + "epoch": 0.6, + "learning_rate": 0.0002107411385606874, + "loss": 0.0669, + "step": 5540 + }, + { + "epoch": 0.6, + "learning_rate": 0.0002105800214822771, + "loss": 0.068, + "step": 5550 + }, + { + "epoch": 0.6, + "learning_rate": 0.0002104189044038668, + "loss": 0.0695, + "step": 5560 + }, + { + "epoch": 0.6, + "learning_rate": 0.00021025778732545647, + "loss": 0.0711, + "step": 5570 + }, + { + "epoch": 0.6, + "learning_rate": 0.00021009667024704615, + "loss": 0.0577, + "step": 5580 + }, + { + "epoch": 0.6, + "learning_rate": 0.00020993555316863589, + "loss": 0.071, + "step": 5590 + }, + { + "epoch": 0.6, + "learning_rate": 0.00020977443609022556, + "loss": 0.0647, + "step": 5600 + }, + { + "epoch": 0.6, + "learning_rate": 0.00020961331901181524, + "loss": 0.0673, + "step": 5610 + }, + { + "epoch": 0.6, + "learning_rate": 0.0002094522019334049, + "loss": 0.0809, + "step": 5620 + }, + { + "epoch": 0.6, + "learning_rate": 0.00020929108485499463, + "loss": 0.0692, + "step": 5630 + }, + { + "epoch": 0.61, + "learning_rate": 0.0002091299677765843, + "loss": 0.0725, + "step": 5640 + }, + { + "epoch": 0.61, + "learning_rate": 0.00020896885069817398, + "loss": 0.0796, + "step": 5650 + }, + { + "epoch": 0.61, + "learning_rate": 0.0002088077336197637, + "loss": 0.0634, + "step": 5660 + }, + { + "epoch": 0.61, + "learning_rate": 0.00020864661654135337, + "loss": 0.0699, + "step": 5670 + }, + { + "epoch": 0.61, + "learning_rate": 0.00020848549946294304, + "loss": 0.0577, + "step": 5680 + }, + { + "epoch": 0.61, + "learning_rate": 0.00020832438238453275, + "loss": 0.0715, + "step": 5690 + }, + { + "epoch": 0.61, + "learning_rate": 0.00020816326530612243, + "loss": 0.0611, + "step": 5700 + }, + { + "epoch": 0.61, + "learning_rate": 0.0002080021482277121, + "loss": 0.0716, + "step": 5710 + }, + { + "epoch": 0.61, + "learning_rate": 0.0002078410311493018, + "loss": 0.0671, + "step": 5720 + }, + { + "epoch": 0.62, + "learning_rate": 0.0002076799140708915, + "loss": 0.0639, + "step": 5730 + }, + { + "epoch": 0.62, + "learning_rate": 0.00020751879699248117, + "loss": 0.0692, + "step": 5740 + }, + { + "epoch": 0.62, + "learning_rate": 0.00020735767991407087, + "loss": 0.0876, + "step": 5750 + }, + { + "epoch": 0.62, + "learning_rate": 0.00020719656283566055, + "loss": 0.0926, + "step": 5760 + }, + { + "epoch": 0.62, + "learning_rate": 0.00020703544575725023, + "loss": 0.0683, + "step": 5770 + }, + { + "epoch": 0.62, + "learning_rate": 0.00020687432867883996, + "loss": 0.0744, + "step": 5780 + }, + { + "epoch": 0.62, + "learning_rate": 0.00020671321160042964, + "loss": 0.0629, + "step": 5790 + }, + { + "epoch": 0.62, + "learning_rate": 0.00020655209452201932, + "loss": 0.0656, + "step": 5800 + }, + { + "epoch": 0.62, + "learning_rate": 0.00020639097744360902, + "loss": 0.0739, + "step": 5810 + }, + { + "epoch": 0.63, + "learning_rate": 0.0002062298603651987, + "loss": 0.0812, + "step": 5820 + }, + { + "epoch": 0.63, + "learning_rate": 0.00020606874328678838, + "loss": 0.0698, + "step": 5830 + }, + { + "epoch": 0.63, + "learning_rate": 0.00020590762620837809, + "loss": 0.0693, + "step": 5840 + }, + { + "epoch": 0.63, + "learning_rate": 0.00020574650912996776, + "loss": 0.0673, + "step": 5850 + }, + { + "epoch": 0.63, + "learning_rate": 0.00020558539205155744, + "loss": 0.0641, + "step": 5860 + }, + { + "epoch": 0.63, + "learning_rate": 0.00020542427497314715, + "loss": 0.0763, + "step": 5870 + }, + { + "epoch": 0.63, + "learning_rate": 0.00020526315789473683, + "loss": 0.0663, + "step": 5880 + }, + { + "epoch": 0.63, + "learning_rate": 0.0002051020408163265, + "loss": 0.0773, + "step": 5890 + }, + { + "epoch": 0.63, + "learning_rate": 0.00020494092373791618, + "loss": 0.0739, + "step": 5900 + }, + { + "epoch": 0.63, + "learning_rate": 0.0002047798066595059, + "loss": 0.0732, + "step": 5910 + }, + { + "epoch": 0.64, + "learning_rate": 0.00020461868958109557, + "loss": 0.0707, + "step": 5920 + }, + { + "epoch": 0.64, + "learning_rate": 0.00020445757250268524, + "loss": 0.0734, + "step": 5930 + }, + { + "epoch": 0.64, + "learning_rate": 0.00020429645542427498, + "loss": 0.0725, + "step": 5940 + }, + { + "epoch": 0.64, + "learning_rate": 0.00020413533834586463, + "loss": 0.0648, + "step": 5950 + }, + { + "epoch": 0.64, + "learning_rate": 0.0002039742212674543, + "loss": 0.0767, + "step": 5960 + }, + { + "epoch": 0.64, + "learning_rate": 0.00020381310418904404, + "loss": 0.072, + "step": 5970 + }, + { + "epoch": 0.64, + "learning_rate": 0.00020365198711063372, + "loss": 0.077, + "step": 5980 + }, + { + "epoch": 0.64, + "learning_rate": 0.0002034908700322234, + "loss": 0.0737, + "step": 5990 + }, + { + "epoch": 0.64, + "learning_rate": 0.0002033297529538131, + "loss": 0.069, + "step": 6000 + }, + { + "epoch": 0.64, + "eval_cer": 0.9238361782215131, + "eval_loss": 0.04587383195757866, + "eval_runtime": 119.1941, + "eval_samples_per_second": 16.763, + "eval_steps_per_second": 4.195, + "step": 6000 + }, + { + "epoch": 0.65, + "learning_rate": 0.00020316863587540278, + "loss": 0.0594, + "step": 6010 + }, + { + "epoch": 0.65, + "learning_rate": 0.00020300751879699246, + "loss": 0.0668, + "step": 6020 + }, + { + "epoch": 0.65, + "learning_rate": 0.00020284640171858216, + "loss": 0.0675, + "step": 6030 + }, + { + "epoch": 0.65, + "learning_rate": 0.00020268528464017184, + "loss": 0.0696, + "step": 6040 + }, + { + "epoch": 0.65, + "learning_rate": 0.00020252416756176152, + "loss": 0.0709, + "step": 6050 + }, + { + "epoch": 0.65, + "learning_rate": 0.00020236305048335122, + "loss": 0.0749, + "step": 6060 + }, + { + "epoch": 0.65, + "learning_rate": 0.0002022019334049409, + "loss": 0.0713, + "step": 6070 + }, + { + "epoch": 0.65, + "learning_rate": 0.00020204081632653058, + "loss": 0.0605, + "step": 6080 + }, + { + "epoch": 0.65, + "learning_rate": 0.00020187969924812029, + "loss": 0.0632, + "step": 6090 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020171858216970996, + "loss": 0.0737, + "step": 6100 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020155746509129964, + "loss": 0.0692, + "step": 6110 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020139634801288937, + "loss": 0.0702, + "step": 6120 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020123523093447905, + "loss": 0.0672, + "step": 6130 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020107411385606873, + "loss": 0.0776, + "step": 6140 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020091299677765844, + "loss": 0.0666, + "step": 6150 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020075187969924811, + "loss": 0.0642, + "step": 6160 + }, + { + "epoch": 0.66, + "learning_rate": 0.0002005907626208378, + "loss": 0.0601, + "step": 6170 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020042964554242747, + "loss": 0.0718, + "step": 6180 + }, + { + "epoch": 0.66, + "learning_rate": 0.00020026852846401718, + "loss": 0.0839, + "step": 6190 + }, + { + "epoch": 0.67, + "learning_rate": 0.00020010741138560685, + "loss": 0.0755, + "step": 6200 + }, + { + "epoch": 0.67, + "learning_rate": 0.00019994629430719653, + "loss": 0.064, + "step": 6210 + }, + { + "epoch": 0.67, + "learning_rate": 0.00019978517722878624, + "loss": 0.0763, + "step": 6220 + }, + { + "epoch": 0.67, + "learning_rate": 0.00019962406015037592, + "loss": 0.0658, + "step": 6230 + }, + { + "epoch": 0.67, + "learning_rate": 0.0001994629430719656, + "loss": 0.0684, + "step": 6240 + }, + { + "epoch": 0.67, + "learning_rate": 0.0001993018259935553, + "loss": 0.0777, + "step": 6250 + }, + { + "epoch": 0.67, + "learning_rate": 0.00019914070891514498, + "loss": 0.0739, + "step": 6260 + }, + { + "epoch": 0.67, + "learning_rate": 0.00019897959183673466, + "loss": 0.0701, + "step": 6270 + }, + { + "epoch": 0.67, + "learning_rate": 0.00019881847475832436, + "loss": 0.0799, + "step": 6280 + }, + { + "epoch": 0.68, + "learning_rate": 0.00019865735767991404, + "loss": 0.0729, + "step": 6290 + }, + { + "epoch": 0.68, + "learning_rate": 0.00019849624060150372, + "loss": 0.0753, + "step": 6300 + }, + { + "epoch": 0.68, + "learning_rate": 0.00019833512352309345, + "loss": 0.0754, + "step": 6310 + }, + { + "epoch": 0.68, + "learning_rate": 0.00019817400644468313, + "loss": 0.0633, + "step": 6320 + }, + { + "epoch": 0.68, + "learning_rate": 0.0001980128893662728, + "loss": 0.0638, + "step": 6330 + }, + { + "epoch": 0.68, + "learning_rate": 0.0001978517722878625, + "loss": 0.0584, + "step": 6340 + }, + { + "epoch": 0.68, + "learning_rate": 0.0001976906552094522, + "loss": 0.0611, + "step": 6350 + }, + { + "epoch": 0.68, + "learning_rate": 0.00019752953813104187, + "loss": 0.0659, + "step": 6360 + }, + { + "epoch": 0.68, + "learning_rate": 0.00019736842105263157, + "loss": 0.0697, + "step": 6370 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019720730397422125, + "loss": 0.0647, + "step": 6380 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019704618689581093, + "loss": 0.0754, + "step": 6390 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019688506981740064, + "loss": 0.0703, + "step": 6400 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019672395273899031, + "loss": 0.0691, + "step": 6410 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019656283566058, + "loss": 0.0663, + "step": 6420 + }, + { + "epoch": 0.69, + "learning_rate": 0.0001964017185821697, + "loss": 0.0717, + "step": 6430 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019624060150375938, + "loss": 0.0596, + "step": 6440 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019607948442534905, + "loss": 0.0688, + "step": 6450 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019591836734693873, + "loss": 0.0619, + "step": 6460 + }, + { + "epoch": 0.69, + "learning_rate": 0.00019575725026852847, + "loss": 0.068, + "step": 6470 + }, + { + "epoch": 0.7, + "learning_rate": 0.00019559613319011814, + "loss": 0.0724, + "step": 6480 + }, + { + "epoch": 0.7, + "learning_rate": 0.00019543501611170782, + "loss": 0.0662, + "step": 6490 + }, + { + "epoch": 0.7, + "learning_rate": 0.00019527389903329753, + "loss": 0.0595, + "step": 6500 + }, + { + "epoch": 0.7, + "learning_rate": 0.0001951127819548872, + "loss": 0.0648, + "step": 6510 + }, + { + "epoch": 0.7, + "learning_rate": 0.00019495166487647688, + "loss": 0.06, + "step": 6520 + }, + { + "epoch": 0.7, + "learning_rate": 0.0001947905477980666, + "loss": 0.0612, + "step": 6530 + }, + { + "epoch": 0.7, + "learning_rate": 0.00019462943071965627, + "loss": 0.0615, + "step": 6540 + }, + { + "epoch": 0.7, + "learning_rate": 0.00019446831364124595, + "loss": 0.0664, + "step": 6550 + }, + { + "epoch": 0.7, + "learning_rate": 0.00019430719656283565, + "loss": 0.0692, + "step": 6560 + }, + { + "epoch": 0.71, + "learning_rate": 0.00019414607948442533, + "loss": 0.0748, + "step": 6570 + }, + { + "epoch": 0.71, + "learning_rate": 0.000193984962406015, + "loss": 0.0641, + "step": 6580 + }, + { + "epoch": 0.71, + "learning_rate": 0.0001938238453276047, + "loss": 0.0586, + "step": 6590 + }, + { + "epoch": 0.71, + "learning_rate": 0.0001936627282491944, + "loss": 0.0663, + "step": 6600 + }, + { + "epoch": 0.71, + "learning_rate": 0.00019350161117078407, + "loss": 0.0565, + "step": 6610 + }, + { + "epoch": 0.71, + "learning_rate": 0.00019334049409237377, + "loss": 0.0566, + "step": 6620 + }, + { + "epoch": 0.71, + "learning_rate": 0.00019317937701396345, + "loss": 0.0628, + "step": 6630 + }, + { + "epoch": 0.71, + "learning_rate": 0.00019301825993555313, + "loss": 0.0674, + "step": 6640 + }, + { + "epoch": 0.71, + "learning_rate": 0.00019285714285714286, + "loss": 0.0626, + "step": 6650 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019269602577873254, + "loss": 0.0679, + "step": 6660 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019253490870032222, + "loss": 0.0703, + "step": 6670 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019237379162191193, + "loss": 0.0675, + "step": 6680 + }, + { + "epoch": 0.72, + "learning_rate": 0.0001922126745435016, + "loss": 0.0598, + "step": 6690 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019205155746509128, + "loss": 0.076, + "step": 6700 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019189044038668096, + "loss": 0.058, + "step": 6710 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019172932330827067, + "loss": 0.0538, + "step": 6720 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019156820622986034, + "loss": 0.0565, + "step": 6730 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019140708915145002, + "loss": 0.0696, + "step": 6740 + }, + { + "epoch": 0.72, + "learning_rate": 0.00019124597207303973, + "loss": 0.0678, + "step": 6750 + }, + { + "epoch": 0.73, + "learning_rate": 0.0001910848549946294, + "loss": 0.0692, + "step": 6760 + }, + { + "epoch": 0.73, + "learning_rate": 0.00019092373791621908, + "loss": 0.0622, + "step": 6770 + }, + { + "epoch": 0.73, + "learning_rate": 0.0001907626208378088, + "loss": 0.0704, + "step": 6780 + }, + { + "epoch": 0.73, + "learning_rate": 0.00019060150375939847, + "loss": 0.0594, + "step": 6790 + }, + { + "epoch": 0.73, + "learning_rate": 0.00019044038668098815, + "loss": 0.0646, + "step": 6800 + }, + { + "epoch": 0.73, + "learning_rate": 0.00019027926960257788, + "loss": 0.0608, + "step": 6810 + }, + { + "epoch": 0.73, + "learning_rate": 0.00019011815252416756, + "loss": 0.0712, + "step": 6820 + }, + { + "epoch": 0.73, + "learning_rate": 0.00018995703544575723, + "loss": 0.0649, + "step": 6830 + }, + { + "epoch": 0.73, + "learning_rate": 0.00018979591836734694, + "loss": 0.0627, + "step": 6840 + }, + { + "epoch": 0.74, + "learning_rate": 0.00018963480128893662, + "loss": 0.0656, + "step": 6850 + }, + { + "epoch": 0.74, + "learning_rate": 0.0001894736842105263, + "loss": 0.0572, + "step": 6860 + }, + { + "epoch": 0.74, + "learning_rate": 0.000189312567132116, + "loss": 0.0645, + "step": 6870 + }, + { + "epoch": 0.74, + "learning_rate": 0.00018915145005370568, + "loss": 0.0594, + "step": 6880 + }, + { + "epoch": 0.74, + "learning_rate": 0.00018899033297529536, + "loss": 0.0722, + "step": 6890 + }, + { + "epoch": 0.74, + "learning_rate": 0.00018882921589688506, + "loss": 0.0698, + "step": 6900 + }, + { + "epoch": 0.74, + "learning_rate": 0.00018866809881847474, + "loss": 0.0709, + "step": 6910 + }, + { + "epoch": 0.74, + "learning_rate": 0.00018850698174006442, + "loss": 0.0753, + "step": 6920 + }, + { + "epoch": 0.74, + "learning_rate": 0.00018834586466165413, + "loss": 0.071, + "step": 6930 + }, + { + "epoch": 0.75, + "learning_rate": 0.0001881847475832438, + "loss": 0.0711, + "step": 6940 + }, + { + "epoch": 0.75, + "learning_rate": 0.00018802363050483348, + "loss": 0.0669, + "step": 6950 + }, + { + "epoch": 0.75, + "learning_rate": 0.0001878625134264232, + "loss": 0.0684, + "step": 6960 + }, + { + "epoch": 0.75, + "learning_rate": 0.00018770139634801287, + "loss": 0.0721, + "step": 6970 + }, + { + "epoch": 0.75, + "learning_rate": 0.00018754027926960254, + "loss": 0.0598, + "step": 6980 + }, + { + "epoch": 0.75, + "learning_rate": 0.00018737916219119222, + "loss": 0.0676, + "step": 6990 + }, + { + "epoch": 0.75, + "learning_rate": 0.00018721804511278195, + "loss": 0.0615, + "step": 7000 + }, + { + "epoch": 0.75, + "eval_cer": 0.9238447291327837, + "eval_loss": 0.043542634695768356, + "eval_runtime": 119.2501, + "eval_samples_per_second": 16.755, + "eval_steps_per_second": 4.193, + "step": 7000 + }, + { + "epoch": 0.75, + "learning_rate": 0.00018705692803437163, + "loss": 0.084, + "step": 7010 + }, + { + "epoch": 0.75, + "learning_rate": 0.0001868958109559613, + "loss": 0.0532, + "step": 7020 + }, + { + "epoch": 0.76, + "learning_rate": 0.00018673469387755102, + "loss": 0.0624, + "step": 7030 + }, + { + "epoch": 0.76, + "learning_rate": 0.0001865735767991407, + "loss": 0.0632, + "step": 7040 + }, + { + "epoch": 0.76, + "learning_rate": 0.00018641245972073037, + "loss": 0.0636, + "step": 7050 + }, + { + "epoch": 0.76, + "learning_rate": 0.00018625134264232008, + "loss": 0.0637, + "step": 7060 + }, + { + "epoch": 0.76, + "learning_rate": 0.00018609022556390976, + "loss": 0.0669, + "step": 7070 + }, + { + "epoch": 0.76, + "learning_rate": 0.00018592910848549943, + "loss": 0.0693, + "step": 7080 + }, + { + "epoch": 0.76, + "learning_rate": 0.00018576799140708914, + "loss": 0.0612, + "step": 7090 + }, + { + "epoch": 0.76, + "learning_rate": 0.00018560687432867882, + "loss": 0.0685, + "step": 7100 + }, + { + "epoch": 0.76, + "learning_rate": 0.0001854457572502685, + "loss": 0.0758, + "step": 7110 + }, + { + "epoch": 0.76, + "learning_rate": 0.0001852846401718582, + "loss": 0.0693, + "step": 7120 + }, + { + "epoch": 0.77, + "learning_rate": 0.00018512352309344788, + "loss": 0.071, + "step": 7130 + }, + { + "epoch": 0.77, + "learning_rate": 0.00018496240601503756, + "loss": 0.0552, + "step": 7140 + }, + { + "epoch": 0.77, + "learning_rate": 0.0001848012889366273, + "loss": 0.0521, + "step": 7150 + }, + { + "epoch": 0.77, + "learning_rate": 0.00018464017185821697, + "loss": 0.0586, + "step": 7160 + }, + { + "epoch": 0.77, + "learning_rate": 0.00018447905477980665, + "loss": 0.0903, + "step": 7170 + }, + { + "epoch": 0.77, + "learning_rate": 0.00018431793770139635, + "loss": 0.0615, + "step": 7180 + }, + { + "epoch": 0.77, + "learning_rate": 0.00018415682062298603, + "loss": 0.0578, + "step": 7190 + }, + { + "epoch": 0.77, + "learning_rate": 0.0001839957035445757, + "loss": 0.0659, + "step": 7200 + }, + { + "epoch": 0.77, + "learning_rate": 0.00018383458646616541, + "loss": 0.0634, + "step": 7210 + }, + { + "epoch": 0.78, + "learning_rate": 0.0001836734693877551, + "loss": 0.0582, + "step": 7220 + }, + { + "epoch": 0.78, + "learning_rate": 0.00018351235230934477, + "loss": 0.0529, + "step": 7230 + }, + { + "epoch": 0.78, + "learning_rate": 0.00018335123523093448, + "loss": 0.0654, + "step": 7240 + }, + { + "epoch": 0.78, + "learning_rate": 0.00018319011815252415, + "loss": 0.0543, + "step": 7250 + }, + { + "epoch": 0.78, + "learning_rate": 0.00018302900107411383, + "loss": 0.0649, + "step": 7260 + }, + { + "epoch": 0.78, + "learning_rate": 0.0001828678839957035, + "loss": 0.0613, + "step": 7270 + }, + { + "epoch": 0.78, + "learning_rate": 0.00018270676691729322, + "loss": 0.065, + "step": 7280 + }, + { + "epoch": 0.78, + "learning_rate": 0.0001825456498388829, + "loss": 0.0615, + "step": 7290 + }, + { + "epoch": 0.78, + "learning_rate": 0.00018238453276047257, + "loss": 0.0775, + "step": 7300 + }, + { + "epoch": 0.79, + "learning_rate": 0.00018222341568206228, + "loss": 0.0544, + "step": 7310 + }, + { + "epoch": 0.79, + "learning_rate": 0.00018206229860365196, + "loss": 0.057, + "step": 7320 + }, + { + "epoch": 0.79, + "learning_rate": 0.00018190118152524163, + "loss": 0.0627, + "step": 7330 + }, + { + "epoch": 0.79, + "learning_rate": 0.00018174006444683137, + "loss": 0.0639, + "step": 7340 + }, + { + "epoch": 0.79, + "learning_rate": 0.00018157894736842105, + "loss": 0.0582, + "step": 7350 + }, + { + "epoch": 0.79, + "learning_rate": 0.00018141783029001072, + "loss": 0.0626, + "step": 7360 + }, + { + "epoch": 0.79, + "learning_rate": 0.00018125671321160043, + "loss": 0.0568, + "step": 7370 + }, + { + "epoch": 0.79, + "learning_rate": 0.0001810955961331901, + "loss": 0.0599, + "step": 7380 + }, + { + "epoch": 0.79, + "learning_rate": 0.00018093447905477979, + "loss": 0.0633, + "step": 7390 + }, + { + "epoch": 0.79, + "learning_rate": 0.0001807733619763695, + "loss": 0.0572, + "step": 7400 + }, + { + "epoch": 0.8, + "learning_rate": 0.00018061224489795917, + "loss": 0.0551, + "step": 7410 + }, + { + "epoch": 0.8, + "learning_rate": 0.00018045112781954885, + "loss": 0.0736, + "step": 7420 + }, + { + "epoch": 0.8, + "learning_rate": 0.00018029001074113855, + "loss": 0.0641, + "step": 7430 + }, + { + "epoch": 0.8, + "learning_rate": 0.00018012889366272823, + "loss": 0.0719, + "step": 7440 + }, + { + "epoch": 0.8, + "learning_rate": 0.0001799677765843179, + "loss": 0.0751, + "step": 7450 + }, + { + "epoch": 0.8, + "learning_rate": 0.00017980665950590761, + "loss": 0.0577, + "step": 7460 + }, + { + "epoch": 0.8, + "learning_rate": 0.0001796455424274973, + "loss": 0.0619, + "step": 7470 + }, + { + "epoch": 0.8, + "learning_rate": 0.00017948442534908697, + "loss": 0.0865, + "step": 7480 + }, + { + "epoch": 0.8, + "learning_rate": 0.0001793233082706767, + "loss": 0.0743, + "step": 7490 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017916219119226638, + "loss": 0.0654, + "step": 7500 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017900107411385606, + "loss": 0.0647, + "step": 7510 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017883995703544577, + "loss": 0.066, + "step": 7520 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017867883995703544, + "loss": 0.0615, + "step": 7530 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017851772287862512, + "loss": 0.0651, + "step": 7540 + }, + { + "epoch": 0.81, + "learning_rate": 0.0001783566058002148, + "loss": 0.0611, + "step": 7550 + }, + { + "epoch": 0.81, + "learning_rate": 0.0001781954887218045, + "loss": 0.0622, + "step": 7560 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017803437164339418, + "loss": 0.0643, + "step": 7570 + }, + { + "epoch": 0.81, + "learning_rate": 0.00017787325456498386, + "loss": 0.058, + "step": 7580 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017771213748657357, + "loss": 0.0661, + "step": 7590 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017755102040816325, + "loss": 0.0752, + "step": 7600 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017738990332975292, + "loss": 0.0587, + "step": 7610 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017722878625134263, + "loss": 0.0665, + "step": 7620 + }, + { + "epoch": 0.82, + "learning_rate": 0.0001770676691729323, + "loss": 0.0675, + "step": 7630 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017690655209452199, + "loss": 0.065, + "step": 7640 + }, + { + "epoch": 0.82, + "learning_rate": 0.0001767454350161117, + "loss": 0.0597, + "step": 7650 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017658431793770137, + "loss": 0.0698, + "step": 7660 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017642320085929105, + "loss": 0.0581, + "step": 7670 + }, + { + "epoch": 0.82, + "learning_rate": 0.00017626208378088078, + "loss": 0.0702, + "step": 7680 + }, + { + "epoch": 0.83, + "learning_rate": 0.00017610096670247046, + "loss": 0.0552, + "step": 7690 + }, + { + "epoch": 0.83, + "learning_rate": 0.00017593984962406014, + "loss": 0.0627, + "step": 7700 + }, + { + "epoch": 0.83, + "learning_rate": 0.00017577873254564984, + "loss": 0.0552, + "step": 7710 + }, + { + "epoch": 0.83, + "learning_rate": 0.00017561761546723952, + "loss": 0.0721, + "step": 7720 + }, + { + "epoch": 0.83, + "learning_rate": 0.0001754564983888292, + "loss": 0.0549, + "step": 7730 + }, + { + "epoch": 0.83, + "learning_rate": 0.0001752953813104189, + "loss": 0.0628, + "step": 7740 + }, + { + "epoch": 0.83, + "learning_rate": 0.00017513426423200858, + "loss": 0.0505, + "step": 7750 + }, + { + "epoch": 0.83, + "learning_rate": 0.00017497314715359826, + "loss": 0.062, + "step": 7760 + }, + { + "epoch": 0.83, + "learning_rate": 0.00017481203007518797, + "loss": 0.0651, + "step": 7770 + }, + { + "epoch": 0.84, + "learning_rate": 0.00017465091299677764, + "loss": 0.0576, + "step": 7780 + }, + { + "epoch": 0.84, + "learning_rate": 0.00017448979591836732, + "loss": 0.0627, + "step": 7790 + }, + { + "epoch": 0.84, + "learning_rate": 0.00017432867883995703, + "loss": 0.0637, + "step": 7800 + }, + { + "epoch": 0.84, + "learning_rate": 0.0001741675617615467, + "loss": 0.0641, + "step": 7810 + }, + { + "epoch": 0.84, + "learning_rate": 0.00017400644468313638, + "loss": 0.0579, + "step": 7820 + }, + { + "epoch": 0.84, + "learning_rate": 0.00017384532760472606, + "loss": 0.0607, + "step": 7830 + }, + { + "epoch": 0.84, + "learning_rate": 0.0001736842105263158, + "loss": 0.0641, + "step": 7840 + }, + { + "epoch": 0.84, + "learning_rate": 0.00017352309344790547, + "loss": 0.0648, + "step": 7850 + }, + { + "epoch": 0.84, + "learning_rate": 0.00017336197636949512, + "loss": 0.0642, + "step": 7860 + }, + { + "epoch": 0.85, + "learning_rate": 0.00017320085929108486, + "loss": 0.0673, + "step": 7870 + }, + { + "epoch": 0.85, + "learning_rate": 0.00017303974221267453, + "loss": 0.0545, + "step": 7880 + }, + { + "epoch": 0.85, + "learning_rate": 0.0001728786251342642, + "loss": 0.0566, + "step": 7890 + }, + { + "epoch": 0.85, + "learning_rate": 0.00017271750805585392, + "loss": 0.0551, + "step": 7900 + }, + { + "epoch": 0.85, + "learning_rate": 0.0001725563909774436, + "loss": 0.0711, + "step": 7910 + }, + { + "epoch": 0.85, + "learning_rate": 0.00017239527389903327, + "loss": 0.0576, + "step": 7920 + }, + { + "epoch": 0.85, + "learning_rate": 0.00017223415682062298, + "loss": 0.0537, + "step": 7930 + }, + { + "epoch": 0.85, + "learning_rate": 0.00017207303974221266, + "loss": 0.0685, + "step": 7940 + }, + { + "epoch": 0.85, + "learning_rate": 0.00017191192266380234, + "loss": 0.0613, + "step": 7950 + }, + { + "epoch": 0.85, + "learning_rate": 0.00017175080558539204, + "loss": 0.0591, + "step": 7960 + }, + { + "epoch": 0.86, + "learning_rate": 0.00017158968850698172, + "loss": 0.0568, + "step": 7970 + }, + { + "epoch": 0.86, + "learning_rate": 0.0001714285714285714, + "loss": 0.0465, + "step": 7980 + }, + { + "epoch": 0.86, + "learning_rate": 0.0001712674543501611, + "loss": 0.0656, + "step": 7990 + }, + { + "epoch": 0.86, + "learning_rate": 0.00017110633727175078, + "loss": 0.0494, + "step": 8000 + }, + { + "epoch": 0.86, + "eval_cer": 0.9236754210896255, + "eval_loss": 0.04086674004793167, + "eval_runtime": 119.0895, + "eval_samples_per_second": 16.777, + "eval_steps_per_second": 4.199, + "step": 8000 + }, + { + "epoch": 0.86, + "learning_rate": 0.00017094522019334046, + "loss": 0.057, + "step": 8010 + }, + { + "epoch": 0.86, + "learning_rate": 0.0001707841031149302, + "loss": 0.071, + "step": 8020 + }, + { + "epoch": 0.86, + "learning_rate": 0.00017062298603651987, + "loss": 0.0597, + "step": 8030 + }, + { + "epoch": 0.86, + "learning_rate": 0.00017046186895810955, + "loss": 0.0537, + "step": 8040 + }, + { + "epoch": 0.86, + "learning_rate": 0.00017030075187969925, + "loss": 0.0595, + "step": 8050 + }, + { + "epoch": 0.87, + "learning_rate": 0.00017013963480128893, + "loss": 0.0527, + "step": 8060 + }, + { + "epoch": 0.87, + "learning_rate": 0.0001699785177228786, + "loss": 0.0608, + "step": 8070 + }, + { + "epoch": 0.87, + "learning_rate": 0.0001698174006444683, + "loss": 0.0595, + "step": 8080 + }, + { + "epoch": 0.87, + "learning_rate": 0.000169656283566058, + "loss": 0.0583, + "step": 8090 + }, + { + "epoch": 0.87, + "learning_rate": 0.00016949516648764767, + "loss": 0.0655, + "step": 8100 + }, + { + "epoch": 0.87, + "learning_rate": 0.00016933404940923735, + "loss": 0.0479, + "step": 8110 + }, + { + "epoch": 0.87, + "learning_rate": 0.00016917293233082706, + "loss": 0.0607, + "step": 8120 + }, + { + "epoch": 0.87, + "learning_rate": 0.00016901181525241673, + "loss": 0.0582, + "step": 8130 + }, + { + "epoch": 0.87, + "learning_rate": 0.0001688506981740064, + "loss": 0.0613, + "step": 8140 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016868958109559612, + "loss": 0.0598, + "step": 8150 + }, + { + "epoch": 0.88, + "learning_rate": 0.0001685284640171858, + "loss": 0.0557, + "step": 8160 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016836734693877547, + "loss": 0.0597, + "step": 8170 + }, + { + "epoch": 0.88, + "learning_rate": 0.0001682062298603652, + "loss": 0.05, + "step": 8180 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016804511278195486, + "loss": 0.0541, + "step": 8190 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016788399570354454, + "loss": 0.0603, + "step": 8200 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016772287862513427, + "loss": 0.0555, + "step": 8210 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016756176154672395, + "loss": 0.0573, + "step": 8220 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016740064446831363, + "loss": 0.0578, + "step": 8230 + }, + { + "epoch": 0.88, + "learning_rate": 0.00016723952738990333, + "loss": 0.06, + "step": 8240 + }, + { + "epoch": 0.89, + "learning_rate": 0.000167078410311493, + "loss": 0.0568, + "step": 8250 + }, + { + "epoch": 0.89, + "learning_rate": 0.0001669172932330827, + "loss": 0.0522, + "step": 8260 + }, + { + "epoch": 0.89, + "learning_rate": 0.0001667561761546724, + "loss": 0.0647, + "step": 8270 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016659505907626207, + "loss": 0.0608, + "step": 8280 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016643394199785175, + "loss": 0.0692, + "step": 8290 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016627282491944145, + "loss": 0.0553, + "step": 8300 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016611170784103113, + "loss": 0.0615, + "step": 8310 + }, + { + "epoch": 0.89, + "learning_rate": 0.0001659505907626208, + "loss": 0.0574, + "step": 8320 + }, + { + "epoch": 0.89, + "learning_rate": 0.00016578947368421052, + "loss": 0.0602, + "step": 8330 + }, + { + "epoch": 0.9, + "learning_rate": 0.0001656283566058002, + "loss": 0.0549, + "step": 8340 + }, + { + "epoch": 0.9, + "learning_rate": 0.00016546723952738987, + "loss": 0.0503, + "step": 8350 + }, + { + "epoch": 0.9, + "learning_rate": 0.00016530612244897955, + "loss": 0.0612, + "step": 8360 + }, + { + "epoch": 0.9, + "learning_rate": 0.00016514500537056928, + "loss": 0.0592, + "step": 8370 + }, + { + "epoch": 0.9, + "learning_rate": 0.00016498388829215896, + "loss": 0.0584, + "step": 8380 + }, + { + "epoch": 0.9, + "learning_rate": 0.00016482277121374864, + "loss": 0.0628, + "step": 8390 + }, + { + "epoch": 0.9, + "learning_rate": 0.00016466165413533835, + "loss": 0.0504, + "step": 8400 + }, + { + "epoch": 0.9, + "learning_rate": 0.00016450053705692802, + "loss": 0.05, + "step": 8410 + }, + { + "epoch": 0.9, + "learning_rate": 0.0001643394199785177, + "loss": 0.0589, + "step": 8420 + }, + { + "epoch": 0.91, + "learning_rate": 0.0001641783029001074, + "loss": 0.0604, + "step": 8430 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016401718582169709, + "loss": 0.0508, + "step": 8440 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016385606874328676, + "loss": 0.0617, + "step": 8450 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016369495166487647, + "loss": 0.0768, + "step": 8460 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016353383458646615, + "loss": 0.0476, + "step": 8470 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016337271750805583, + "loss": 0.0574, + "step": 8480 + }, + { + "epoch": 0.91, + "learning_rate": 0.00016321160042964553, + "loss": 0.0527, + "step": 8490 + }, + { + "epoch": 0.91, + "learning_rate": 0.0001630504833512352, + "loss": 0.0509, + "step": 8500 + }, + { + "epoch": 0.91, + "learning_rate": 0.0001628893662728249, + "loss": 0.0659, + "step": 8510 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016272824919441462, + "loss": 0.0577, + "step": 8520 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016256713211600427, + "loss": 0.0569, + "step": 8530 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016240601503759395, + "loss": 0.0577, + "step": 8540 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016224489795918368, + "loss": 0.0534, + "step": 8550 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016208378088077336, + "loss": 0.051, + "step": 8560 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016192266380236304, + "loss": 0.0575, + "step": 8570 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016176154672395274, + "loss": 0.0583, + "step": 8580 + }, + { + "epoch": 0.92, + "learning_rate": 0.00016160042964554242, + "loss": 0.0645, + "step": 8590 + }, + { + "epoch": 0.92, + "learning_rate": 0.0001614393125671321, + "loss": 0.0641, + "step": 8600 + }, + { + "epoch": 0.92, + "learning_rate": 0.0001612781954887218, + "loss": 0.0602, + "step": 8610 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016111707841031148, + "loss": 0.0552, + "step": 8620 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016095596133190116, + "loss": 0.0612, + "step": 8630 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016079484425349084, + "loss": 0.0587, + "step": 8640 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016063372717508055, + "loss": 0.0587, + "step": 8650 + }, + { + "epoch": 0.93, + "learning_rate": 0.00016047261009667022, + "loss": 0.0553, + "step": 8660 + }, + { + "epoch": 0.93, + "learning_rate": 0.0001603114930182599, + "loss": 0.0612, + "step": 8670 + }, + { + "epoch": 0.93, + "learning_rate": 0.0001601503759398496, + "loss": 0.0552, + "step": 8680 + }, + { + "epoch": 0.93, + "learning_rate": 0.00015998925886143929, + "loss": 0.0551, + "step": 8690 + }, + { + "epoch": 0.93, + "learning_rate": 0.00015982814178302896, + "loss": 0.0496, + "step": 8700 + }, + { + "epoch": 0.94, + "learning_rate": 0.0001596670247046187, + "loss": 0.0707, + "step": 8710 + }, + { + "epoch": 0.94, + "learning_rate": 0.00015950590762620837, + "loss": 0.0581, + "step": 8720 + }, + { + "epoch": 0.94, + "learning_rate": 0.00015934479054779805, + "loss": 0.0653, + "step": 8730 + }, + { + "epoch": 0.94, + "learning_rate": 0.00015918367346938776, + "loss": 0.0486, + "step": 8740 + }, + { + "epoch": 0.94, + "learning_rate": 0.00015902255639097744, + "loss": 0.0562, + "step": 8750 + }, + { + "epoch": 0.94, + "learning_rate": 0.00015886143931256711, + "loss": 0.0572, + "step": 8760 + }, + { + "epoch": 0.94, + "learning_rate": 0.00015870032223415682, + "loss": 0.0482, + "step": 8770 + }, + { + "epoch": 0.94, + "learning_rate": 0.0001585392051557465, + "loss": 0.0588, + "step": 8780 + }, + { + "epoch": 0.94, + "learning_rate": 0.00015837808807733618, + "loss": 0.0464, + "step": 8790 + }, + { + "epoch": 0.95, + "learning_rate": 0.00015821697099892588, + "loss": 0.0558, + "step": 8800 + }, + { + "epoch": 0.95, + "learning_rate": 0.00015805585392051556, + "loss": 0.0518, + "step": 8810 + }, + { + "epoch": 0.95, + "learning_rate": 0.00015789473684210524, + "loss": 0.0533, + "step": 8820 + }, + { + "epoch": 0.95, + "learning_rate": 0.00015773361976369494, + "loss": 0.0529, + "step": 8830 + }, + { + "epoch": 0.95, + "learning_rate": 0.00015757250268528462, + "loss": 0.0536, + "step": 8840 + }, + { + "epoch": 0.95, + "learning_rate": 0.0001574113856068743, + "loss": 0.0642, + "step": 8850 + }, + { + "epoch": 0.95, + "learning_rate": 0.000157250268528464, + "loss": 0.0477, + "step": 8860 + }, + { + "epoch": 0.95, + "learning_rate": 0.00015708915145005368, + "loss": 0.0583, + "step": 8870 + }, + { + "epoch": 0.95, + "learning_rate": 0.00015692803437164336, + "loss": 0.0589, + "step": 8880 + }, + { + "epoch": 0.95, + "learning_rate": 0.0001567669172932331, + "loss": 0.0718, + "step": 8890 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015660580021482277, + "loss": 0.0594, + "step": 8900 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015644468313641245, + "loss": 0.0542, + "step": 8910 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015628356605800213, + "loss": 0.074, + "step": 8920 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015612244897959183, + "loss": 0.0675, + "step": 8930 + }, + { + "epoch": 0.96, + "learning_rate": 0.0001559613319011815, + "loss": 0.0544, + "step": 8940 + }, + { + "epoch": 0.96, + "learning_rate": 0.0001558002148227712, + "loss": 0.0541, + "step": 8950 + }, + { + "epoch": 0.96, + "learning_rate": 0.0001556390977443609, + "loss": 0.0602, + "step": 8960 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015547798066595057, + "loss": 0.0663, + "step": 8970 + }, + { + "epoch": 0.96, + "learning_rate": 0.00015531686358754025, + "loss": 0.0585, + "step": 8980 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015515574650912996, + "loss": 0.0559, + "step": 8990 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015499462943071964, + "loss": 0.0671, + "step": 9000 + }, + { + "epoch": 0.97, + "eval_cer": 0.9238327578570048, + "eval_loss": 0.038805074989795685, + "eval_runtime": 118.9099, + "eval_samples_per_second": 16.803, + "eval_steps_per_second": 4.205, + "step": 9000 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015483351235230931, + "loss": 0.0735, + "step": 9010 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015467239527389902, + "loss": 0.058, + "step": 9020 + }, + { + "epoch": 0.97, + "learning_rate": 0.0001545112781954887, + "loss": 0.0511, + "step": 9030 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015435016111707838, + "loss": 0.0639, + "step": 9040 + }, + { + "epoch": 0.97, + "learning_rate": 0.0001541890440386681, + "loss": 0.0611, + "step": 9050 + }, + { + "epoch": 0.97, + "learning_rate": 0.0001540279269602578, + "loss": 0.0549, + "step": 9060 + }, + { + "epoch": 0.97, + "learning_rate": 0.00015386680988184747, + "loss": 0.05, + "step": 9070 + }, + { + "epoch": 0.98, + "learning_rate": 0.00015370569280343717, + "loss": 0.0644, + "step": 9080 + }, + { + "epoch": 0.98, + "learning_rate": 0.00015354457572502685, + "loss": 0.0765, + "step": 9090 + }, + { + "epoch": 0.98, + "learning_rate": 0.00015338345864661653, + "loss": 0.0617, + "step": 9100 + }, + { + "epoch": 0.98, + "learning_rate": 0.00015322234156820623, + "loss": 0.0579, + "step": 9110 + }, + { + "epoch": 0.98, + "learning_rate": 0.0001530612244897959, + "loss": 0.0661, + "step": 9120 + }, + { + "epoch": 0.98, + "learning_rate": 0.0001529001074113856, + "loss": 0.0686, + "step": 9130 + }, + { + "epoch": 0.98, + "learning_rate": 0.0001527389903329753, + "loss": 0.0655, + "step": 9140 + }, + { + "epoch": 0.98, + "learning_rate": 0.00015257787325456497, + "loss": 0.0647, + "step": 9150 + }, + { + "epoch": 0.98, + "learning_rate": 0.00015241675617615465, + "loss": 0.0638, + "step": 9160 + }, + { + "epoch": 0.98, + "learning_rate": 0.00015225563909774436, + "loss": 0.0571, + "step": 9170 + }, + { + "epoch": 0.99, + "learning_rate": 0.00015209452201933403, + "loss": 0.0625, + "step": 9180 + }, + { + "epoch": 0.99, + "learning_rate": 0.0001519334049409237, + "loss": 0.057, + "step": 9190 + }, + { + "epoch": 0.99, + "learning_rate": 0.0001517722878625134, + "loss": 0.0578, + "step": 9200 + }, + { + "epoch": 0.99, + "learning_rate": 0.0001516111707841031, + "loss": 0.0659, + "step": 9210 + }, + { + "epoch": 0.99, + "learning_rate": 0.00015145005370569277, + "loss": 0.0506, + "step": 9220 + }, + { + "epoch": 0.99, + "learning_rate": 0.00015128893662728245, + "loss": 0.0582, + "step": 9230 + }, + { + "epoch": 0.99, + "learning_rate": 0.00015112781954887218, + "loss": 0.0488, + "step": 9240 + }, + { + "epoch": 0.99, + "learning_rate": 0.00015096670247046186, + "loss": 0.0538, + "step": 9250 + }, + { + "epoch": 0.99, + "learning_rate": 0.00015080558539205154, + "loss": 0.0495, + "step": 9260 + }, + { + "epoch": 1.0, + "learning_rate": 0.00015064446831364125, + "loss": 0.0533, + "step": 9270 + }, + { + "epoch": 1.0, + "learning_rate": 0.00015048335123523093, + "loss": 0.0526, + "step": 9280 + }, + { + "epoch": 1.0, + "learning_rate": 0.0001503222341568206, + "loss": 0.0621, + "step": 9290 + }, + { + "epoch": 1.0, + "learning_rate": 0.0001501611170784103, + "loss": 0.0648, + "step": 9300 + }, + { + "epoch": 1.0, + "learning_rate": 0.00015, + "loss": 0.0551, + "step": 9310 + }, + { + "epoch": 1.0, + "learning_rate": 0.0001498388829215897, + "loss": 0.0481, + "step": 9320 + }, + { + "epoch": 1.0, + "learning_rate": 0.00014967776584317937, + "loss": 0.0467, + "step": 9330 + }, + { + "epoch": 1.0, + "learning_rate": 0.00014951664876476905, + "loss": 0.0459, + "step": 9340 + }, + { + "epoch": 1.0, + "learning_rate": 0.00014935553168635875, + "loss": 0.047, + "step": 9350 + }, + { + "epoch": 1.01, + "learning_rate": 0.00014919441460794843, + "loss": 0.0502, + "step": 9360 + }, + { + "epoch": 1.01, + "learning_rate": 0.0001490332975295381, + "loss": 0.0493, + "step": 9370 + }, + { + "epoch": 1.01, + "learning_rate": 0.0001488721804511278, + "loss": 0.0415, + "step": 9380 + }, + { + "epoch": 1.01, + "learning_rate": 0.0001487110633727175, + "loss": 0.0476, + "step": 9390 + }, + { + "epoch": 1.01, + "learning_rate": 0.0001485499462943072, + "loss": 0.0457, + "step": 9400 + }, + { + "epoch": 1.01, + "learning_rate": 0.00014838882921589688, + "loss": 0.0409, + "step": 9410 + }, + { + "epoch": 1.01, + "learning_rate": 0.00014822771213748656, + "loss": 0.0473, + "step": 9420 + }, + { + "epoch": 1.01, + "learning_rate": 0.00014806659505907626, + "loss": 0.043, + "step": 9430 + }, + { + "epoch": 1.01, + "learning_rate": 0.00014790547798066594, + "loss": 0.0521, + "step": 9440 + }, + { + "epoch": 1.01, + "learning_rate": 0.00014774436090225562, + "loss": 0.0552, + "step": 9450 + }, + { + "epoch": 1.02, + "learning_rate": 0.00014758324382384532, + "loss": 0.0452, + "step": 9460 + }, + { + "epoch": 1.02, + "learning_rate": 0.000147422126745435, + "loss": 0.052, + "step": 9470 + }, + { + "epoch": 1.02, + "learning_rate": 0.00014726100966702468, + "loss": 0.0436, + "step": 9480 + }, + { + "epoch": 1.02, + "learning_rate": 0.00014709989258861438, + "loss": 0.0439, + "step": 9490 + }, + { + "epoch": 1.02, + "learning_rate": 0.00014693877551020406, + "loss": 0.0441, + "step": 9500 + }, + { + "epoch": 1.02, + "learning_rate": 0.00014677765843179377, + "loss": 0.0434, + "step": 9510 + }, + { + "epoch": 1.02, + "learning_rate": 0.00014661654135338345, + "loss": 0.0562, + "step": 9520 + }, + { + "epoch": 1.02, + "learning_rate": 0.00014645542427497313, + "loss": 0.0443, + "step": 9530 + }, + { + "epoch": 1.02, + "learning_rate": 0.00014629430719656283, + "loss": 0.0465, + "step": 9540 + }, + { + "epoch": 1.03, + "learning_rate": 0.0001461331901181525, + "loss": 0.0409, + "step": 9550 + }, + { + "epoch": 1.03, + "learning_rate": 0.0001459720730397422, + "loss": 0.0413, + "step": 9560 + }, + { + "epoch": 1.03, + "learning_rate": 0.0001458109559613319, + "loss": 0.0397, + "step": 9570 + }, + { + "epoch": 1.03, + "learning_rate": 0.00014564983888292157, + "loss": 0.0455, + "step": 9580 + }, + { + "epoch": 1.03, + "learning_rate": 0.00014548872180451128, + "loss": 0.0434, + "step": 9590 + }, + { + "epoch": 1.03, + "learning_rate": 0.00014532760472610095, + "loss": 0.0418, + "step": 9600 + }, + { + "epoch": 1.03, + "learning_rate": 0.00014516648764769063, + "loss": 0.0464, + "step": 9610 + }, + { + "epoch": 1.03, + "learning_rate": 0.00014500537056928034, + "loss": 0.0427, + "step": 9620 + }, + { + "epoch": 1.03, + "learning_rate": 0.00014484425349087002, + "loss": 0.0441, + "step": 9630 + }, + { + "epoch": 1.04, + "learning_rate": 0.0001446831364124597, + "loss": 0.0558, + "step": 9640 + }, + { + "epoch": 1.04, + "learning_rate": 0.0001445220193340494, + "loss": 0.0432, + "step": 9650 + }, + { + "epoch": 1.04, + "learning_rate": 0.00014436090225563908, + "loss": 0.0506, + "step": 9660 + }, + { + "epoch": 1.04, + "learning_rate": 0.00014419978517722878, + "loss": 0.0416, + "step": 9670 + }, + { + "epoch": 1.04, + "learning_rate": 0.00014403866809881846, + "loss": 0.0456, + "step": 9680 + }, + { + "epoch": 1.04, + "learning_rate": 0.00014387755102040814, + "loss": 0.0463, + "step": 9690 + }, + { + "epoch": 1.04, + "learning_rate": 0.00014371643394199784, + "loss": 0.0589, + "step": 9700 + }, + { + "epoch": 1.04, + "learning_rate": 0.00014355531686358752, + "loss": 0.0413, + "step": 9710 + }, + { + "epoch": 1.04, + "learning_rate": 0.0001433941997851772, + "loss": 0.044, + "step": 9720 + }, + { + "epoch": 1.05, + "learning_rate": 0.0001432330827067669, + "loss": 0.0465, + "step": 9730 + }, + { + "epoch": 1.05, + "learning_rate": 0.0001430719656283566, + "loss": 0.0423, + "step": 9740 + }, + { + "epoch": 1.05, + "learning_rate": 0.0001429108485499463, + "loss": 0.0369, + "step": 9750 + }, + { + "epoch": 1.05, + "learning_rate": 0.00014274973147153597, + "loss": 0.0428, + "step": 9760 + }, + { + "epoch": 1.05, + "learning_rate": 0.00014258861439312567, + "loss": 0.0458, + "step": 9770 + }, + { + "epoch": 1.05, + "learning_rate": 0.00014242749731471535, + "loss": 0.0489, + "step": 9780 + }, + { + "epoch": 1.05, + "learning_rate": 0.00014226638023630503, + "loss": 0.0435, + "step": 9790 + }, + { + "epoch": 1.05, + "learning_rate": 0.0001421052631578947, + "loss": 0.0407, + "step": 9800 + }, + { + "epoch": 1.05, + "learning_rate": 0.00014194414607948441, + "loss": 0.0439, + "step": 9810 + }, + { + "epoch": 1.05, + "learning_rate": 0.0001417830290010741, + "loss": 0.0446, + "step": 9820 + }, + { + "epoch": 1.06, + "learning_rate": 0.00014162191192266377, + "loss": 0.0437, + "step": 9830 + }, + { + "epoch": 1.06, + "learning_rate": 0.00014146079484425348, + "loss": 0.0416, + "step": 9840 + }, + { + "epoch": 1.06, + "learning_rate": 0.00014129967776584318, + "loss": 0.0448, + "step": 9850 + }, + { + "epoch": 1.06, + "learning_rate": 0.00014113856068743286, + "loss": 0.0457, + "step": 9860 + }, + { + "epoch": 1.06, + "learning_rate": 0.00014097744360902254, + "loss": 0.036, + "step": 9870 + }, + { + "epoch": 1.06, + "learning_rate": 0.00014081632653061224, + "loss": 0.0409, + "step": 9880 + }, + { + "epoch": 1.06, + "learning_rate": 0.00014065520945220192, + "loss": 0.0505, + "step": 9890 + }, + { + "epoch": 1.06, + "learning_rate": 0.0001404940923737916, + "loss": 0.0417, + "step": 9900 + }, + { + "epoch": 1.06, + "learning_rate": 0.0001403329752953813, + "loss": 0.0422, + "step": 9910 + }, + { + "epoch": 1.07, + "learning_rate": 0.00014017185821697098, + "loss": 0.0524, + "step": 9920 + }, + { + "epoch": 1.07, + "learning_rate": 0.0001400107411385607, + "loss": 0.0435, + "step": 9930 + }, + { + "epoch": 1.07, + "learning_rate": 0.00013984962406015037, + "loss": 0.0434, + "step": 9940 + }, + { + "epoch": 1.07, + "learning_rate": 0.00013968850698174004, + "loss": 0.0447, + "step": 9950 + }, + { + "epoch": 1.07, + "learning_rate": 0.00013952738990332975, + "loss": 0.0499, + "step": 9960 + }, + { + "epoch": 1.07, + "learning_rate": 0.00013936627282491943, + "loss": 0.0446, + "step": 9970 + }, + { + "epoch": 1.07, + "learning_rate": 0.0001392051557465091, + "loss": 0.0402, + "step": 9980 + }, + { + "epoch": 1.07, + "learning_rate": 0.0001390440386680988, + "loss": 0.0457, + "step": 9990 + }, + { + "epoch": 1.07, + "learning_rate": 0.0001388829215896885, + "loss": 0.0425, + "step": 10000 + }, + { + "epoch": 1.07, + "eval_cer": 0.9237267265572492, + "eval_loss": 0.036706846207380295, + "eval_runtime": 118.777, + "eval_samples_per_second": 16.821, + "eval_steps_per_second": 4.21, + "step": 10000 + }, + { + "epoch": 1.08, + "learning_rate": 0.0001387218045112782, + "loss": 0.046, + "step": 10010 + }, + { + "epoch": 1.08, + "learning_rate": 0.00013856068743286787, + "loss": 0.0437, + "step": 10020 + }, + { + "epoch": 1.08, + "learning_rate": 0.00013839957035445755, + "loss": 0.0458, + "step": 10030 + }, + { + "epoch": 1.08, + "learning_rate": 0.00013823845327604726, + "loss": 0.0469, + "step": 10040 + }, + { + "epoch": 1.08, + "learning_rate": 0.00013807733619763694, + "loss": 0.0527, + "step": 10050 + }, + { + "epoch": 1.08, + "learning_rate": 0.00013791621911922661, + "loss": 0.047, + "step": 10060 + }, + { + "epoch": 1.08, + "learning_rate": 0.00013775510204081632, + "loss": 0.0392, + "step": 10070 + }, + { + "epoch": 1.08, + "learning_rate": 0.000137593984962406, + "loss": 0.0462, + "step": 10080 + }, + { + "epoch": 1.08, + "learning_rate": 0.0001374328678839957, + "loss": 0.0441, + "step": 10090 + }, + { + "epoch": 1.08, + "learning_rate": 0.00013727175080558538, + "loss": 0.0437, + "step": 10100 + }, + { + "epoch": 1.09, + "learning_rate": 0.00013711063372717506, + "loss": 0.0445, + "step": 10110 + }, + { + "epoch": 1.09, + "learning_rate": 0.00013694951664876476, + "loss": 0.0391, + "step": 10120 + }, + { + "epoch": 1.09, + "learning_rate": 0.00013678839957035444, + "loss": 0.0443, + "step": 10130 + }, + { + "epoch": 1.09, + "learning_rate": 0.00013662728249194412, + "loss": 0.0444, + "step": 10140 + }, + { + "epoch": 1.09, + "learning_rate": 0.00013646616541353383, + "loss": 0.0411, + "step": 10150 + }, + { + "epoch": 1.09, + "learning_rate": 0.0001363050483351235, + "loss": 0.057, + "step": 10160 + }, + { + "epoch": 1.09, + "learning_rate": 0.00013614393125671318, + "loss": 0.0425, + "step": 10170 + }, + { + "epoch": 1.09, + "learning_rate": 0.0001359828141783029, + "loss": 0.0452, + "step": 10180 + }, + { + "epoch": 1.09, + "learning_rate": 0.0001358216970998926, + "loss": 0.0472, + "step": 10190 + }, + { + "epoch": 1.1, + "learning_rate": 0.00013566058002148227, + "loss": 0.0429, + "step": 10200 + }, + { + "epoch": 1.1, + "learning_rate": 0.00013549946294307195, + "loss": 0.0446, + "step": 10210 + }, + { + "epoch": 1.1, + "learning_rate": 0.00013533834586466163, + "loss": 0.047, + "step": 10220 + }, + { + "epoch": 1.1, + "learning_rate": 0.00013517722878625133, + "loss": 0.0436, + "step": 10230 + }, + { + "epoch": 1.1, + "learning_rate": 0.000135016111707841, + "loss": 0.0391, + "step": 10240 + }, + { + "epoch": 1.1, + "learning_rate": 0.0001348549946294307, + "loss": 0.0418, + "step": 10250 + }, + { + "epoch": 1.1, + "learning_rate": 0.0001346938775510204, + "loss": 0.043, + "step": 10260 + }, + { + "epoch": 1.1, + "learning_rate": 0.0001345327604726101, + "loss": 0.0409, + "step": 10270 + }, + { + "epoch": 1.1, + "learning_rate": 0.00013437164339419978, + "loss": 0.0393, + "step": 10280 + }, + { + "epoch": 1.11, + "learning_rate": 0.00013421052631578946, + "loss": 0.0523, + "step": 10290 + }, + { + "epoch": 1.11, + "learning_rate": 0.00013404940923737916, + "loss": 0.0422, + "step": 10300 + }, + { + "epoch": 1.11, + "learning_rate": 0.00013388829215896884, + "loss": 0.0445, + "step": 10310 + }, + { + "epoch": 1.11, + "learning_rate": 0.00013372717508055852, + "loss": 0.0409, + "step": 10320 + }, + { + "epoch": 1.11, + "learning_rate": 0.00013356605800214822, + "loss": 0.0406, + "step": 10330 + }, + { + "epoch": 1.11, + "learning_rate": 0.0001334049409237379, + "loss": 0.0423, + "step": 10340 + }, + { + "epoch": 1.11, + "learning_rate": 0.0001332438238453276, + "loss": 0.0468, + "step": 10350 + }, + { + "epoch": 1.11, + "learning_rate": 0.0001330827067669173, + "loss": 0.0504, + "step": 10360 + }, + { + "epoch": 1.11, + "learning_rate": 0.00013292158968850696, + "loss": 0.0419, + "step": 10370 + }, + { + "epoch": 1.11, + "learning_rate": 0.00013276047261009667, + "loss": 0.048, + "step": 10380 + }, + { + "epoch": 1.12, + "learning_rate": 0.00013259935553168635, + "loss": 0.0456, + "step": 10390 + }, + { + "epoch": 1.12, + "learning_rate": 0.00013243823845327603, + "loss": 0.0427, + "step": 10400 + }, + { + "epoch": 1.12, + "learning_rate": 0.00013227712137486573, + "loss": 0.0425, + "step": 10410 + }, + { + "epoch": 1.12, + "learning_rate": 0.0001321160042964554, + "loss": 0.043, + "step": 10420 + }, + { + "epoch": 1.12, + "learning_rate": 0.00013195488721804512, + "loss": 0.0444, + "step": 10430 + }, + { + "epoch": 1.12, + "learning_rate": 0.0001317937701396348, + "loss": 0.0398, + "step": 10440 + }, + { + "epoch": 1.12, + "learning_rate": 0.00013163265306122447, + "loss": 0.042, + "step": 10450 + }, + { + "epoch": 1.12, + "learning_rate": 0.00013147153598281418, + "loss": 0.05, + "step": 10460 + }, + { + "epoch": 1.12, + "learning_rate": 0.00013131041890440386, + "loss": 0.0375, + "step": 10470 + }, + { + "epoch": 1.13, + "learning_rate": 0.00013114930182599353, + "loss": 0.0413, + "step": 10480 + }, + { + "epoch": 1.13, + "learning_rate": 0.00013098818474758324, + "loss": 0.0427, + "step": 10490 + }, + { + "epoch": 1.13, + "learning_rate": 0.00013082706766917292, + "loss": 0.0425, + "step": 10500 + }, + { + "epoch": 1.13, + "learning_rate": 0.0001306659505907626, + "loss": 0.0486, + "step": 10510 + }, + { + "epoch": 1.13, + "learning_rate": 0.0001305048335123523, + "loss": 0.053, + "step": 10520 + }, + { + "epoch": 1.13, + "learning_rate": 0.00013034371643394198, + "loss": 0.0465, + "step": 10530 + }, + { + "epoch": 1.13, + "learning_rate": 0.00013018259935553168, + "loss": 0.0422, + "step": 10540 + }, + { + "epoch": 1.13, + "learning_rate": 0.00013002148227712136, + "loss": 0.0455, + "step": 10550 + }, + { + "epoch": 1.13, + "learning_rate": 0.00012986036519871104, + "loss": 0.0381, + "step": 10560 + }, + { + "epoch": 1.14, + "learning_rate": 0.00012969924812030075, + "loss": 0.0447, + "step": 10570 + }, + { + "epoch": 1.14, + "learning_rate": 0.00012953813104189042, + "loss": 0.0398, + "step": 10580 + }, + { + "epoch": 1.14, + "learning_rate": 0.0001293770139634801, + "loss": 0.0372, + "step": 10590 + }, + { + "epoch": 1.14, + "learning_rate": 0.0001292158968850698, + "loss": 0.0425, + "step": 10600 + }, + { + "epoch": 1.14, + "learning_rate": 0.00012905477980665951, + "loss": 0.047, + "step": 10610 + }, + { + "epoch": 1.14, + "learning_rate": 0.0001288936627282492, + "loss": 0.0399, + "step": 10620 + }, + { + "epoch": 1.14, + "learning_rate": 0.00012873254564983887, + "loss": 0.0391, + "step": 10630 + }, + { + "epoch": 1.14, + "learning_rate": 0.00012857142857142855, + "loss": 0.0393, + "step": 10640 + }, + { + "epoch": 1.14, + "learning_rate": 0.00012841031149301825, + "loss": 0.049, + "step": 10650 + }, + { + "epoch": 1.14, + "learning_rate": 0.00012824919441460793, + "loss": 0.0391, + "step": 10660 + }, + { + "epoch": 1.15, + "learning_rate": 0.0001280880773361976, + "loss": 0.0398, + "step": 10670 + }, + { + "epoch": 1.15, + "learning_rate": 0.00012792696025778732, + "loss": 0.0388, + "step": 10680 + }, + { + "epoch": 1.15, + "learning_rate": 0.00012776584317937702, + "loss": 0.0453, + "step": 10690 + }, + { + "epoch": 1.15, + "learning_rate": 0.0001276047261009667, + "loss": 0.044, + "step": 10700 + }, + { + "epoch": 1.15, + "learning_rate": 0.00012744360902255638, + "loss": 0.0426, + "step": 10710 + }, + { + "epoch": 1.15, + "learning_rate": 0.00012728249194414608, + "loss": 0.046, + "step": 10720 + }, + { + "epoch": 1.15, + "learning_rate": 0.00012712137486573576, + "loss": 0.041, + "step": 10730 + }, + { + "epoch": 1.15, + "learning_rate": 0.00012696025778732544, + "loss": 0.0447, + "step": 10740 + }, + { + "epoch": 1.15, + "learning_rate": 0.00012679914070891512, + "loss": 0.0446, + "step": 10750 + }, + { + "epoch": 1.16, + "learning_rate": 0.00012663802363050482, + "loss": 0.0458, + "step": 10760 + }, + { + "epoch": 1.16, + "learning_rate": 0.0001264769065520945, + "loss": 0.0409, + "step": 10770 + }, + { + "epoch": 1.16, + "learning_rate": 0.00012631578947368418, + "loss": 0.0416, + "step": 10780 + }, + { + "epoch": 1.16, + "learning_rate": 0.00012615467239527388, + "loss": 0.0414, + "step": 10790 + }, + { + "epoch": 1.16, + "learning_rate": 0.0001259935553168636, + "loss": 0.0433, + "step": 10800 + }, + { + "epoch": 1.16, + "learning_rate": 0.00012583243823845327, + "loss": 0.0404, + "step": 10810 + }, + { + "epoch": 1.16, + "learning_rate": 0.00012567132116004295, + "loss": 0.0473, + "step": 10820 + }, + { + "epoch": 1.16, + "learning_rate": 0.00012551020408163265, + "loss": 0.0372, + "step": 10830 + }, + { + "epoch": 1.16, + "learning_rate": 0.00012534908700322233, + "loss": 0.0427, + "step": 10840 + }, + { + "epoch": 1.17, + "learning_rate": 0.000125187969924812, + "loss": 0.0423, + "step": 10850 + }, + { + "epoch": 1.17, + "learning_rate": 0.00012502685284640171, + "loss": 0.0463, + "step": 10860 + }, + { + "epoch": 1.17, + "learning_rate": 0.0001248657357679914, + "loss": 0.0466, + "step": 10870 + }, + { + "epoch": 1.17, + "learning_rate": 0.0001247046186895811, + "loss": 0.046, + "step": 10880 + }, + { + "epoch": 1.17, + "learning_rate": 0.00012454350161117078, + "loss": 0.0358, + "step": 10890 + }, + { + "epoch": 1.17, + "learning_rate": 0.00012438238453276045, + "loss": 0.0424, + "step": 10900 + }, + { + "epoch": 1.17, + "learning_rate": 0.00012422126745435016, + "loss": 0.041, + "step": 10910 + }, + { + "epoch": 1.17, + "learning_rate": 0.00012406015037593984, + "loss": 0.0425, + "step": 10920 + }, + { + "epoch": 1.17, + "learning_rate": 0.00012389903329752952, + "loss": 0.0431, + "step": 10930 + }, + { + "epoch": 1.17, + "learning_rate": 0.00012373791621911922, + "loss": 0.0453, + "step": 10940 + }, + { + "epoch": 1.18, + "learning_rate": 0.0001235767991407089, + "loss": 0.0473, + "step": 10950 + }, + { + "epoch": 1.18, + "learning_rate": 0.0001234156820622986, + "loss": 0.0426, + "step": 10960 + }, + { + "epoch": 1.18, + "learning_rate": 0.00012325456498388828, + "loss": 0.0435, + "step": 10970 + }, + { + "epoch": 1.18, + "learning_rate": 0.00012309344790547796, + "loss": 0.0441, + "step": 10980 + }, + { + "epoch": 1.18, + "learning_rate": 0.00012293233082706767, + "loss": 0.0423, + "step": 10990 + }, + { + "epoch": 1.18, + "learning_rate": 0.00012277121374865734, + "loss": 0.0394, + "step": 11000 + }, + { + "epoch": 1.18, + "eval_cer": 0.9236702905428632, + "eval_loss": 0.03555446118116379, + "eval_runtime": 119.0714, + "eval_samples_per_second": 16.78, + "eval_steps_per_second": 4.199, + "step": 11000 + }, + { + "epoch": 1.18, + "learning_rate": 0.00012261009667024702, + "loss": 0.0425, + "step": 11010 + }, + { + "epoch": 1.18, + "learning_rate": 0.00012244897959183673, + "loss": 0.0404, + "step": 11020 + }, + { + "epoch": 1.18, + "learning_rate": 0.0001222878625134264, + "loss": 0.0404, + "step": 11030 + }, + { + "epoch": 1.19, + "learning_rate": 0.0001221267454350161, + "loss": 0.0464, + "step": 11040 + }, + { + "epoch": 1.19, + "learning_rate": 0.00012196562835660579, + "loss": 0.0398, + "step": 11050 + }, + { + "epoch": 1.19, + "learning_rate": 0.00012180451127819547, + "loss": 0.0373, + "step": 11060 + }, + { + "epoch": 1.19, + "learning_rate": 0.00012164339419978516, + "loss": 0.0465, + "step": 11070 + }, + { + "epoch": 1.19, + "learning_rate": 0.00012148227712137487, + "loss": 0.0399, + "step": 11080 + }, + { + "epoch": 1.19, + "learning_rate": 0.00012132116004296454, + "loss": 0.0428, + "step": 11090 + }, + { + "epoch": 1.19, + "learning_rate": 0.00012116004296455424, + "loss": 0.0422, + "step": 11100 + }, + { + "epoch": 1.19, + "learning_rate": 0.00012099892588614393, + "loss": 0.0461, + "step": 11110 + }, + { + "epoch": 1.19, + "learning_rate": 0.0001208378088077336, + "loss": 0.0413, + "step": 11120 + }, + { + "epoch": 1.2, + "learning_rate": 0.0001206766917293233, + "loss": 0.0433, + "step": 11130 + }, + { + "epoch": 1.2, + "learning_rate": 0.00012051557465091299, + "loss": 0.0476, + "step": 11140 + }, + { + "epoch": 1.2, + "learning_rate": 0.00012035445757250267, + "loss": 0.0466, + "step": 11150 + }, + { + "epoch": 1.2, + "learning_rate": 0.00012019334049409236, + "loss": 0.0462, + "step": 11160 + }, + { + "epoch": 1.2, + "learning_rate": 0.00012003222341568204, + "loss": 0.0412, + "step": 11170 + }, + { + "epoch": 1.2, + "learning_rate": 0.00011987110633727174, + "loss": 0.0389, + "step": 11180 + }, + { + "epoch": 1.2, + "learning_rate": 0.00011970998925886143, + "loss": 0.0475, + "step": 11190 + }, + { + "epoch": 1.2, + "learning_rate": 0.00011954887218045111, + "loss": 0.0382, + "step": 11200 + }, + { + "epoch": 1.2, + "learning_rate": 0.0001193877551020408, + "loss": 0.0431, + "step": 11210 + }, + { + "epoch": 1.21, + "learning_rate": 0.0001192266380236305, + "loss": 0.0388, + "step": 11220 + }, + { + "epoch": 1.21, + "learning_rate": 0.00011906552094522017, + "loss": 0.0445, + "step": 11230 + }, + { + "epoch": 1.21, + "learning_rate": 0.00011890440386680987, + "loss": 0.0425, + "step": 11240 + }, + { + "epoch": 1.21, + "learning_rate": 0.00011874328678839957, + "loss": 0.0415, + "step": 11250 + }, + { + "epoch": 1.21, + "learning_rate": 0.00011858216970998925, + "loss": 0.044, + "step": 11260 + }, + { + "epoch": 1.21, + "learning_rate": 0.00011842105263157894, + "loss": 0.0454, + "step": 11270 + }, + { + "epoch": 1.21, + "learning_rate": 0.00011825993555316863, + "loss": 0.0456, + "step": 11280 + }, + { + "epoch": 1.21, + "learning_rate": 0.00011809881847475831, + "loss": 0.0412, + "step": 11290 + }, + { + "epoch": 1.21, + "learning_rate": 0.000117937701396348, + "loss": 0.0475, + "step": 11300 + }, + { + "epoch": 1.21, + "learning_rate": 0.00011777658431793768, + "loss": 0.0407, + "step": 11310 + }, + { + "epoch": 1.22, + "learning_rate": 0.00011761546723952737, + "loss": 0.0412, + "step": 11320 + }, + { + "epoch": 1.22, + "learning_rate": 0.00011745435016111707, + "loss": 0.0402, + "step": 11330 + }, + { + "epoch": 1.22, + "learning_rate": 0.00011729323308270674, + "loss": 0.037, + "step": 11340 + }, + { + "epoch": 1.22, + "learning_rate": 0.00011713211600429645, + "loss": 0.0501, + "step": 11350 + }, + { + "epoch": 1.22, + "learning_rate": 0.00011697099892588614, + "loss": 0.0472, + "step": 11360 + }, + { + "epoch": 1.22, + "learning_rate": 0.00011680988184747582, + "loss": 0.0469, + "step": 11370 + }, + { + "epoch": 1.22, + "learning_rate": 0.00011664876476906551, + "loss": 0.0396, + "step": 11380 + }, + { + "epoch": 1.22, + "learning_rate": 0.0001164876476906552, + "loss": 0.0461, + "step": 11390 + }, + { + "epoch": 1.22, + "learning_rate": 0.00011632653061224488, + "loss": 0.0364, + "step": 11400 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011616541353383457, + "loss": 0.0426, + "step": 11410 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011600429645542428, + "loss": 0.0399, + "step": 11420 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011584317937701396, + "loss": 0.0403, + "step": 11430 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011568206229860365, + "loss": 0.0406, + "step": 11440 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011552094522019333, + "loss": 0.0421, + "step": 11450 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011535982814178302, + "loss": 0.0382, + "step": 11460 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011519871106337271, + "loss": 0.0389, + "step": 11470 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011503759398496239, + "loss": 0.0395, + "step": 11480 + }, + { + "epoch": 1.23, + "learning_rate": 0.00011487647690655208, + "loss": 0.0446, + "step": 11490 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011471535982814177, + "loss": 0.0361, + "step": 11500 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011455424274973145, + "loss": 0.0445, + "step": 11510 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011439312567132116, + "loss": 0.0406, + "step": 11520 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011423200859291085, + "loss": 0.0387, + "step": 11530 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011407089151450053, + "loss": 0.04, + "step": 11540 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011390977443609022, + "loss": 0.0538, + "step": 11550 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011374865735767991, + "loss": 0.0445, + "step": 11560 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011358754027926959, + "loss": 0.0417, + "step": 11570 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011342642320085928, + "loss": 0.0413, + "step": 11580 + }, + { + "epoch": 1.24, + "learning_rate": 0.00011326530612244896, + "loss": 0.0415, + "step": 11590 + }, + { + "epoch": 1.25, + "learning_rate": 0.00011310418904403866, + "loss": 0.041, + "step": 11600 + }, + { + "epoch": 1.25, + "learning_rate": 0.00011294307196562835, + "loss": 0.041, + "step": 11610 + }, + { + "epoch": 1.25, + "learning_rate": 0.00011278195488721803, + "loss": 0.0417, + "step": 11620 + }, + { + "epoch": 1.25, + "learning_rate": 0.00011262083780880772, + "loss": 0.0371, + "step": 11630 + }, + { + "epoch": 1.25, + "learning_rate": 0.00011245972073039742, + "loss": 0.0393, + "step": 11640 + }, + { + "epoch": 1.25, + "learning_rate": 0.0001122986036519871, + "loss": 0.042, + "step": 11650 + }, + { + "epoch": 1.25, + "learning_rate": 0.00011213748657357679, + "loss": 0.0374, + "step": 11660 + }, + { + "epoch": 1.25, + "learning_rate": 0.00011197636949516648, + "loss": 0.0355, + "step": 11670 + }, + { + "epoch": 1.25, + "learning_rate": 0.00011181525241675616, + "loss": 0.0419, + "step": 11680 + }, + { + "epoch": 1.26, + "learning_rate": 0.00011165413533834586, + "loss": 0.0377, + "step": 11690 + }, + { + "epoch": 1.26, + "learning_rate": 0.00011149301825993555, + "loss": 0.0422, + "step": 11700 + }, + { + "epoch": 1.26, + "learning_rate": 0.00011133190118152523, + "loss": 0.0411, + "step": 11710 + }, + { + "epoch": 1.26, + "learning_rate": 0.00011117078410311492, + "loss": 0.0449, + "step": 11720 + }, + { + "epoch": 1.26, + "learning_rate": 0.0001110096670247046, + "loss": 0.0465, + "step": 11730 + }, + { + "epoch": 1.26, + "learning_rate": 0.0001108485499462943, + "loss": 0.041, + "step": 11740 + }, + { + "epoch": 1.26, + "learning_rate": 0.00011068743286788399, + "loss": 0.0464, + "step": 11750 + }, + { + "epoch": 1.26, + "learning_rate": 0.00011052631578947366, + "loss": 0.039, + "step": 11760 + }, + { + "epoch": 1.26, + "learning_rate": 0.00011036519871106337, + "loss": 0.0482, + "step": 11770 + }, + { + "epoch": 1.27, + "learning_rate": 0.00011020408163265306, + "loss": 0.0391, + "step": 11780 + }, + { + "epoch": 1.27, + "learning_rate": 0.00011004296455424274, + "loss": 0.0336, + "step": 11790 + }, + { + "epoch": 1.27, + "learning_rate": 0.00010988184747583243, + "loss": 0.0367, + "step": 11800 + }, + { + "epoch": 1.27, + "learning_rate": 0.00010972073039742212, + "loss": 0.0318, + "step": 11810 + }, + { + "epoch": 1.27, + "learning_rate": 0.0001095596133190118, + "loss": 0.038, + "step": 11820 + }, + { + "epoch": 1.27, + "learning_rate": 0.00010939849624060149, + "loss": 0.0426, + "step": 11830 + }, + { + "epoch": 1.27, + "learning_rate": 0.00010923737916219118, + "loss": 0.0373, + "step": 11840 + }, + { + "epoch": 1.27, + "learning_rate": 0.00010907626208378086, + "loss": 0.042, + "step": 11850 + }, + { + "epoch": 1.27, + "learning_rate": 0.00010891514500537057, + "loss": 0.0376, + "step": 11860 + }, + { + "epoch": 1.27, + "learning_rate": 0.00010875402792696025, + "loss": 0.0384, + "step": 11870 + }, + { + "epoch": 1.28, + "learning_rate": 0.00010859291084854994, + "loss": 0.0408, + "step": 11880 + }, + { + "epoch": 1.28, + "learning_rate": 0.00010843179377013963, + "loss": 0.0466, + "step": 11890 + }, + { + "epoch": 1.28, + "learning_rate": 0.00010827067669172931, + "loss": 0.043, + "step": 11900 + }, + { + "epoch": 1.28, + "learning_rate": 0.000108109559613319, + "loss": 0.038, + "step": 11910 + }, + { + "epoch": 1.28, + "learning_rate": 0.00010794844253490869, + "loss": 0.0415, + "step": 11920 + }, + { + "epoch": 1.28, + "learning_rate": 0.00010778732545649837, + "loss": 0.044, + "step": 11930 + }, + { + "epoch": 1.28, + "learning_rate": 0.00010762620837808808, + "loss": 0.046, + "step": 11940 + }, + { + "epoch": 1.28, + "learning_rate": 0.00010746509129967777, + "loss": 0.0414, + "step": 11950 + }, + { + "epoch": 1.28, + "learning_rate": 0.00010730397422126745, + "loss": 0.0445, + "step": 11960 + }, + { + "epoch": 1.29, + "learning_rate": 0.00010714285714285714, + "loss": 0.0412, + "step": 11970 + }, + { + "epoch": 1.29, + "learning_rate": 0.00010698174006444682, + "loss": 0.0449, + "step": 11980 + }, + { + "epoch": 1.29, + "learning_rate": 0.00010682062298603651, + "loss": 0.0468, + "step": 11990 + }, + { + "epoch": 1.29, + "learning_rate": 0.0001066595059076262, + "loss": 0.0399, + "step": 12000 + }, + { + "epoch": 1.29, + "eval_cer": 0.9236283910776372, + "eval_loss": 0.034416958689689636, + "eval_runtime": 118.9502, + "eval_samples_per_second": 16.797, + "eval_steps_per_second": 4.203, + "step": 12000 + }, + { + "epoch": 1.29, + "learning_rate": 0.00010649838882921588, + "loss": 0.0433, + "step": 12010 + }, + { + "epoch": 1.29, + "learning_rate": 0.00010633727175080557, + "loss": 0.0408, + "step": 12020 + }, + { + "epoch": 1.29, + "learning_rate": 0.00010617615467239527, + "loss": 0.0426, + "step": 12030 + }, + { + "epoch": 1.29, + "learning_rate": 0.00010601503759398495, + "loss": 0.0491, + "step": 12040 + }, + { + "epoch": 1.29, + "learning_rate": 0.00010585392051557464, + "loss": 0.0486, + "step": 12050 + }, + { + "epoch": 1.3, + "learning_rate": 0.00010569280343716434, + "loss": 0.0426, + "step": 12060 + }, + { + "epoch": 1.3, + "learning_rate": 0.00010553168635875401, + "loss": 0.0452, + "step": 12070 + }, + { + "epoch": 1.3, + "learning_rate": 0.0001053705692803437, + "loss": 0.0404, + "step": 12080 + }, + { + "epoch": 1.3, + "learning_rate": 0.0001052094522019334, + "loss": 0.042, + "step": 12090 + }, + { + "epoch": 1.3, + "learning_rate": 0.00010504833512352308, + "loss": 0.0413, + "step": 12100 + }, + { + "epoch": 1.3, + "learning_rate": 0.00010488721804511278, + "loss": 0.0405, + "step": 12110 + }, + { + "epoch": 1.3, + "learning_rate": 0.00010472610096670245, + "loss": 0.0434, + "step": 12120 + }, + { + "epoch": 1.3, + "learning_rate": 0.00010456498388829215, + "loss": 0.0385, + "step": 12130 + }, + { + "epoch": 1.3, + "learning_rate": 0.00010440386680988184, + "loss": 0.0384, + "step": 12140 + }, + { + "epoch": 1.3, + "learning_rate": 0.00010424274973147152, + "loss": 0.0327, + "step": 12150 + }, + { + "epoch": 1.31, + "learning_rate": 0.00010408163265306121, + "loss": 0.0387, + "step": 12160 + }, + { + "epoch": 1.31, + "learning_rate": 0.0001039205155746509, + "loss": 0.0433, + "step": 12170 + }, + { + "epoch": 1.31, + "learning_rate": 0.00010375939849624058, + "loss": 0.0391, + "step": 12180 + }, + { + "epoch": 1.31, + "learning_rate": 0.00010359828141783028, + "loss": 0.046, + "step": 12190 + }, + { + "epoch": 1.31, + "learning_rate": 0.00010343716433941998, + "loss": 0.0407, + "step": 12200 + }, + { + "epoch": 1.31, + "learning_rate": 0.00010327604726100966, + "loss": 0.0372, + "step": 12210 + }, + { + "epoch": 1.31, + "learning_rate": 0.00010311493018259935, + "loss": 0.043, + "step": 12220 + }, + { + "epoch": 1.31, + "learning_rate": 0.00010295381310418904, + "loss": 0.0408, + "step": 12230 + }, + { + "epoch": 1.31, + "learning_rate": 0.00010279269602577872, + "loss": 0.0415, + "step": 12240 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010263157894736841, + "loss": 0.037, + "step": 12250 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010247046186895809, + "loss": 0.0391, + "step": 12260 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010230934479054778, + "loss": 0.0408, + "step": 12270 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010214822771213749, + "loss": 0.0403, + "step": 12280 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010198711063372715, + "loss": 0.0448, + "step": 12290 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010182599355531686, + "loss": 0.0391, + "step": 12300 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010166487647690655, + "loss": 0.0377, + "step": 12310 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010150375939849623, + "loss": 0.042, + "step": 12320 + }, + { + "epoch": 1.32, + "learning_rate": 0.00010134264232008592, + "loss": 0.0384, + "step": 12330 + }, + { + "epoch": 1.33, + "learning_rate": 0.00010118152524167561, + "loss": 0.0463, + "step": 12340 + }, + { + "epoch": 1.33, + "learning_rate": 0.00010102040816326529, + "loss": 0.0424, + "step": 12350 + }, + { + "epoch": 1.33, + "learning_rate": 0.00010085929108485498, + "loss": 0.0488, + "step": 12360 + }, + { + "epoch": 1.33, + "learning_rate": 0.00010069817400644469, + "loss": 0.0483, + "step": 12370 + }, + { + "epoch": 1.33, + "learning_rate": 0.00010053705692803437, + "loss": 0.0406, + "step": 12380 + }, + { + "epoch": 1.33, + "learning_rate": 0.00010037593984962406, + "loss": 0.0433, + "step": 12390 + }, + { + "epoch": 1.33, + "learning_rate": 0.00010021482277121374, + "loss": 0.0363, + "step": 12400 + }, + { + "epoch": 1.33, + "learning_rate": 0.00010005370569280343, + "loss": 0.0397, + "step": 12410 + }, + { + "epoch": 1.33, + "learning_rate": 9.989258861439312e-05, + "loss": 0.0349, + "step": 12420 + }, + { + "epoch": 1.34, + "learning_rate": 9.97314715359828e-05, + "loss": 0.0372, + "step": 12430 + }, + { + "epoch": 1.34, + "learning_rate": 9.957035445757249e-05, + "loss": 0.0354, + "step": 12440 + }, + { + "epoch": 1.34, + "learning_rate": 9.940923737916218e-05, + "loss": 0.0353, + "step": 12450 + }, + { + "epoch": 1.34, + "learning_rate": 9.924812030075186e-05, + "loss": 0.0339, + "step": 12460 + }, + { + "epoch": 1.34, + "learning_rate": 9.908700322234156e-05, + "loss": 0.0398, + "step": 12470 + }, + { + "epoch": 1.34, + "learning_rate": 9.892588614393126e-05, + "loss": 0.0433, + "step": 12480 + }, + { + "epoch": 1.34, + "learning_rate": 9.876476906552093e-05, + "loss": 0.0448, + "step": 12490 + }, + { + "epoch": 1.34, + "learning_rate": 9.860365198711063e-05, + "loss": 0.0317, + "step": 12500 + }, + { + "epoch": 1.34, + "learning_rate": 9.844253490870032e-05, + "loss": 0.0422, + "step": 12510 + }, + { + "epoch": 1.34, + "learning_rate": 9.828141783029e-05, + "loss": 0.0364, + "step": 12520 + }, + { + "epoch": 1.35, + "learning_rate": 9.812030075187969e-05, + "loss": 0.0415, + "step": 12530 + }, + { + "epoch": 1.35, + "learning_rate": 9.795918367346937e-05, + "loss": 0.0372, + "step": 12540 + }, + { + "epoch": 1.35, + "learning_rate": 9.779806659505907e-05, + "loss": 0.041, + "step": 12550 + }, + { + "epoch": 1.35, + "learning_rate": 9.763694951664876e-05, + "loss": 0.0351, + "step": 12560 + }, + { + "epoch": 1.35, + "learning_rate": 9.747583243823844e-05, + "loss": 0.0353, + "step": 12570 + }, + { + "epoch": 1.35, + "learning_rate": 9.731471535982813e-05, + "loss": 0.0402, + "step": 12580 + }, + { + "epoch": 1.35, + "learning_rate": 9.715359828141783e-05, + "loss": 0.0429, + "step": 12590 + }, + { + "epoch": 1.35, + "learning_rate": 9.69924812030075e-05, + "loss": 0.0349, + "step": 12600 + }, + { + "epoch": 1.35, + "learning_rate": 9.68313641245972e-05, + "loss": 0.0429, + "step": 12610 + }, + { + "epoch": 1.36, + "learning_rate": 9.667024704618689e-05, + "loss": 0.0401, + "step": 12620 + }, + { + "epoch": 1.36, + "learning_rate": 9.650912996777657e-05, + "loss": 0.0374, + "step": 12630 + }, + { + "epoch": 1.36, + "learning_rate": 9.634801288936627e-05, + "loss": 0.0421, + "step": 12640 + }, + { + "epoch": 1.36, + "learning_rate": 9.618689581095596e-05, + "loss": 0.0394, + "step": 12650 + }, + { + "epoch": 1.36, + "learning_rate": 9.602577873254564e-05, + "loss": 0.0379, + "step": 12660 + }, + { + "epoch": 1.36, + "learning_rate": 9.586466165413533e-05, + "loss": 0.0365, + "step": 12670 + }, + { + "epoch": 1.36, + "learning_rate": 9.570354457572501e-05, + "loss": 0.0375, + "step": 12680 + }, + { + "epoch": 1.36, + "learning_rate": 9.55424274973147e-05, + "loss": 0.0432, + "step": 12690 + }, + { + "epoch": 1.36, + "learning_rate": 9.53813104189044e-05, + "loss": 0.046, + "step": 12700 + }, + { + "epoch": 1.37, + "learning_rate": 9.522019334049407e-05, + "loss": 0.0366, + "step": 12710 + }, + { + "epoch": 1.37, + "learning_rate": 9.505907626208378e-05, + "loss": 0.045, + "step": 12720 + }, + { + "epoch": 1.37, + "learning_rate": 9.489795918367347e-05, + "loss": 0.0391, + "step": 12730 + }, + { + "epoch": 1.37, + "learning_rate": 9.473684210526315e-05, + "loss": 0.0353, + "step": 12740 + }, + { + "epoch": 1.37, + "learning_rate": 9.457572502685284e-05, + "loss": 0.0345, + "step": 12750 + }, + { + "epoch": 1.37, + "learning_rate": 9.441460794844253e-05, + "loss": 0.0518, + "step": 12760 + }, + { + "epoch": 1.37, + "learning_rate": 9.425349087003221e-05, + "loss": 0.0411, + "step": 12770 + }, + { + "epoch": 1.37, + "learning_rate": 9.40923737916219e-05, + "loss": 0.0381, + "step": 12780 + }, + { + "epoch": 1.37, + "learning_rate": 9.39312567132116e-05, + "loss": 0.0393, + "step": 12790 + }, + { + "epoch": 1.37, + "learning_rate": 9.377013963480127e-05, + "loss": 0.0398, + "step": 12800 + }, + { + "epoch": 1.38, + "learning_rate": 9.360902255639098e-05, + "loss": 0.0432, + "step": 12810 + }, + { + "epoch": 1.38, + "learning_rate": 9.344790547798066e-05, + "loss": 0.0364, + "step": 12820 + }, + { + "epoch": 1.38, + "learning_rate": 9.328678839957035e-05, + "loss": 0.0434, + "step": 12830 + }, + { + "epoch": 1.38, + "learning_rate": 9.312567132116004e-05, + "loss": 0.0361, + "step": 12840 + }, + { + "epoch": 1.38, + "learning_rate": 9.296455424274972e-05, + "loss": 0.0353, + "step": 12850 + }, + { + "epoch": 1.38, + "learning_rate": 9.280343716433941e-05, + "loss": 0.0472, + "step": 12860 + }, + { + "epoch": 1.38, + "learning_rate": 9.26423200859291e-05, + "loss": 0.0398, + "step": 12870 + }, + { + "epoch": 1.38, + "learning_rate": 9.248120300751878e-05, + "loss": 0.0374, + "step": 12880 + }, + { + "epoch": 1.38, + "learning_rate": 9.232008592910848e-05, + "loss": 0.0359, + "step": 12890 + }, + { + "epoch": 1.39, + "learning_rate": 9.215896885069818e-05, + "loss": 0.0363, + "step": 12900 + }, + { + "epoch": 1.39, + "learning_rate": 9.199785177228785e-05, + "loss": 0.0386, + "step": 12910 + }, + { + "epoch": 1.39, + "learning_rate": 9.183673469387755e-05, + "loss": 0.0384, + "step": 12920 + }, + { + "epoch": 1.39, + "learning_rate": 9.167561761546724e-05, + "loss": 0.0381, + "step": 12930 + }, + { + "epoch": 1.39, + "learning_rate": 9.151450053705692e-05, + "loss": 0.039, + "step": 12940 + }, + { + "epoch": 1.39, + "learning_rate": 9.135338345864661e-05, + "loss": 0.0404, + "step": 12950 + }, + { + "epoch": 1.39, + "learning_rate": 9.119226638023629e-05, + "loss": 0.0436, + "step": 12960 + }, + { + "epoch": 1.39, + "learning_rate": 9.103114930182598e-05, + "loss": 0.0412, + "step": 12970 + }, + { + "epoch": 1.39, + "learning_rate": 9.087003222341568e-05, + "loss": 0.0328, + "step": 12980 + }, + { + "epoch": 1.4, + "learning_rate": 9.070891514500536e-05, + "loss": 0.0412, + "step": 12990 + }, + { + "epoch": 1.4, + "learning_rate": 9.054779806659505e-05, + "loss": 0.0375, + "step": 13000 + }, + { + "epoch": 1.4, + "eval_cer": 0.923528345415771, + "eval_loss": 0.03334718942642212, + "eval_runtime": 118.8952, + "eval_samples_per_second": 16.805, + "eval_steps_per_second": 4.205, + "step": 13000 + }, + { + "epoch": 1.4, + "learning_rate": 9.038668098818475e-05, + "loss": 0.0373, + "step": 13010 + }, + { + "epoch": 1.4, + "learning_rate": 9.022556390977442e-05, + "loss": 0.0441, + "step": 13020 + }, + { + "epoch": 1.4, + "learning_rate": 9.006444683136412e-05, + "loss": 0.039, + "step": 13030 + }, + { + "epoch": 1.4, + "learning_rate": 8.990332975295381e-05, + "loss": 0.0403, + "step": 13040 + }, + { + "epoch": 1.4, + "learning_rate": 8.974221267454349e-05, + "loss": 0.0423, + "step": 13050 + }, + { + "epoch": 1.4, + "learning_rate": 8.958109559613319e-05, + "loss": 0.0438, + "step": 13060 + }, + { + "epoch": 1.4, + "learning_rate": 8.941997851772288e-05, + "loss": 0.0409, + "step": 13070 + }, + { + "epoch": 1.4, + "learning_rate": 8.925886143931256e-05, + "loss": 0.036, + "step": 13080 + }, + { + "epoch": 1.41, + "learning_rate": 8.909774436090225e-05, + "loss": 0.0356, + "step": 13090 + }, + { + "epoch": 1.41, + "learning_rate": 8.893662728249193e-05, + "loss": 0.035, + "step": 13100 + }, + { + "epoch": 1.41, + "learning_rate": 8.877551020408162e-05, + "loss": 0.0323, + "step": 13110 + }, + { + "epoch": 1.41, + "learning_rate": 8.861439312567131e-05, + "loss": 0.0433, + "step": 13120 + }, + { + "epoch": 1.41, + "learning_rate": 8.845327604726099e-05, + "loss": 0.0357, + "step": 13130 + }, + { + "epoch": 1.41, + "learning_rate": 8.829215896885068e-05, + "loss": 0.0381, + "step": 13140 + }, + { + "epoch": 1.41, + "learning_rate": 8.813104189044039e-05, + "loss": 0.0488, + "step": 13150 + }, + { + "epoch": 1.41, + "learning_rate": 8.796992481203007e-05, + "loss": 0.0425, + "step": 13160 + }, + { + "epoch": 1.41, + "learning_rate": 8.780880773361976e-05, + "loss": 0.0525, + "step": 13170 + }, + { + "epoch": 1.42, + "learning_rate": 8.764769065520945e-05, + "loss": 0.0462, + "step": 13180 + }, + { + "epoch": 1.42, + "learning_rate": 8.748657357679913e-05, + "loss": 0.0389, + "step": 13190 + }, + { + "epoch": 1.42, + "learning_rate": 8.732545649838882e-05, + "loss": 0.0368, + "step": 13200 + }, + { + "epoch": 1.42, + "learning_rate": 8.716433941997851e-05, + "loss": 0.034, + "step": 13210 + }, + { + "epoch": 1.42, + "learning_rate": 8.700322234156819e-05, + "loss": 0.0398, + "step": 13220 + }, + { + "epoch": 1.42, + "learning_rate": 8.68421052631579e-05, + "loss": 0.0404, + "step": 13230 + }, + { + "epoch": 1.42, + "learning_rate": 8.668098818474756e-05, + "loss": 0.0402, + "step": 13240 + }, + { + "epoch": 1.42, + "learning_rate": 8.651987110633727e-05, + "loss": 0.0429, + "step": 13250 + }, + { + "epoch": 1.42, + "learning_rate": 8.635875402792696e-05, + "loss": 0.0371, + "step": 13260 + }, + { + "epoch": 1.43, + "learning_rate": 8.619763694951664e-05, + "loss": 0.0377, + "step": 13270 + }, + { + "epoch": 1.43, + "learning_rate": 8.603651987110633e-05, + "loss": 0.0381, + "step": 13280 + }, + { + "epoch": 1.43, + "learning_rate": 8.587540279269602e-05, + "loss": 0.035, + "step": 13290 + }, + { + "epoch": 1.43, + "learning_rate": 8.57142857142857e-05, + "loss": 0.0366, + "step": 13300 + }, + { + "epoch": 1.43, + "learning_rate": 8.555316863587539e-05, + "loss": 0.0395, + "step": 13310 + }, + { + "epoch": 1.43, + "learning_rate": 8.53920515574651e-05, + "loss": 0.0377, + "step": 13320 + }, + { + "epoch": 1.43, + "learning_rate": 8.523093447905477e-05, + "loss": 0.0404, + "step": 13330 + }, + { + "epoch": 1.43, + "learning_rate": 8.506981740064447e-05, + "loss": 0.0339, + "step": 13340 + }, + { + "epoch": 1.43, + "learning_rate": 8.490870032223414e-05, + "loss": 0.0348, + "step": 13350 + }, + { + "epoch": 1.43, + "learning_rate": 8.474758324382384e-05, + "loss": 0.0393, + "step": 13360 + }, + { + "epoch": 1.44, + "learning_rate": 8.458646616541353e-05, + "loss": 0.0353, + "step": 13370 + }, + { + "epoch": 1.44, + "learning_rate": 8.44253490870032e-05, + "loss": 0.0342, + "step": 13380 + }, + { + "epoch": 1.44, + "learning_rate": 8.42642320085929e-05, + "loss": 0.0366, + "step": 13390 + }, + { + "epoch": 1.44, + "learning_rate": 8.41031149301826e-05, + "loss": 0.0375, + "step": 13400 + }, + { + "epoch": 1.44, + "learning_rate": 8.394199785177227e-05, + "loss": 0.0372, + "step": 13410 + }, + { + "epoch": 1.44, + "learning_rate": 8.378088077336197e-05, + "loss": 0.0376, + "step": 13420 + }, + { + "epoch": 1.44, + "learning_rate": 8.361976369495167e-05, + "loss": 0.0361, + "step": 13430 + }, + { + "epoch": 1.44, + "learning_rate": 8.345864661654134e-05, + "loss": 0.0396, + "step": 13440 + }, + { + "epoch": 1.44, + "learning_rate": 8.329752953813104e-05, + "loss": 0.0387, + "step": 13450 + }, + { + "epoch": 1.45, + "learning_rate": 8.313641245972073e-05, + "loss": 0.0361, + "step": 13460 + }, + { + "epoch": 1.45, + "learning_rate": 8.29752953813104e-05, + "loss": 0.0376, + "step": 13470 + }, + { + "epoch": 1.45, + "learning_rate": 8.28141783029001e-05, + "loss": 0.0354, + "step": 13480 + }, + { + "epoch": 1.45, + "learning_rate": 8.265306122448978e-05, + "loss": 0.0392, + "step": 13490 + }, + { + "epoch": 1.45, + "learning_rate": 8.249194414607948e-05, + "loss": 0.0344, + "step": 13500 + }, + { + "epoch": 1.45, + "learning_rate": 8.233082706766917e-05, + "loss": 0.0325, + "step": 13510 + }, + { + "epoch": 1.45, + "learning_rate": 8.216970998925885e-05, + "loss": 0.0354, + "step": 13520 + }, + { + "epoch": 1.45, + "learning_rate": 8.200859291084854e-05, + "loss": 0.0376, + "step": 13530 + }, + { + "epoch": 1.45, + "learning_rate": 8.184747583243823e-05, + "loss": 0.04, + "step": 13540 + }, + { + "epoch": 1.46, + "learning_rate": 8.168635875402791e-05, + "loss": 0.0411, + "step": 13550 + }, + { + "epoch": 1.46, + "learning_rate": 8.15252416756176e-05, + "loss": 0.04, + "step": 13560 + }, + { + "epoch": 1.46, + "learning_rate": 8.136412459720731e-05, + "loss": 0.0349, + "step": 13570 + }, + { + "epoch": 1.46, + "learning_rate": 8.120300751879697e-05, + "loss": 0.0349, + "step": 13580 + }, + { + "epoch": 1.46, + "learning_rate": 8.104189044038668e-05, + "loss": 0.0458, + "step": 13590 + }, + { + "epoch": 1.46, + "learning_rate": 8.088077336197637e-05, + "loss": 0.0348, + "step": 13600 + }, + { + "epoch": 1.46, + "learning_rate": 8.071965628356605e-05, + "loss": 0.038, + "step": 13610 + }, + { + "epoch": 1.46, + "learning_rate": 8.055853920515574e-05, + "loss": 0.0375, + "step": 13620 + }, + { + "epoch": 1.46, + "learning_rate": 8.039742212674542e-05, + "loss": 0.0401, + "step": 13630 + }, + { + "epoch": 1.46, + "learning_rate": 8.023630504833511e-05, + "loss": 0.0326, + "step": 13640 + }, + { + "epoch": 1.47, + "learning_rate": 8.00751879699248e-05, + "loss": 0.0374, + "step": 13650 + }, + { + "epoch": 1.47, + "learning_rate": 7.991407089151448e-05, + "loss": 0.0372, + "step": 13660 + }, + { + "epoch": 1.47, + "learning_rate": 7.975295381310419e-05, + "loss": 0.0384, + "step": 13670 + }, + { + "epoch": 1.47, + "learning_rate": 7.959183673469388e-05, + "loss": 0.0352, + "step": 13680 + }, + { + "epoch": 1.47, + "learning_rate": 7.943071965628356e-05, + "loss": 0.0341, + "step": 13690 + }, + { + "epoch": 1.47, + "learning_rate": 7.926960257787325e-05, + "loss": 0.0377, + "step": 13700 + }, + { + "epoch": 1.47, + "learning_rate": 7.910848549946294e-05, + "loss": 0.0386, + "step": 13710 + }, + { + "epoch": 1.47, + "learning_rate": 7.894736842105262e-05, + "loss": 0.0347, + "step": 13720 + }, + { + "epoch": 1.47, + "learning_rate": 7.878625134264231e-05, + "loss": 0.0342, + "step": 13730 + }, + { + "epoch": 1.48, + "learning_rate": 7.8625134264232e-05, + "loss": 0.0379, + "step": 13740 + }, + { + "epoch": 1.48, + "learning_rate": 7.846401718582168e-05, + "loss": 0.0414, + "step": 13750 + }, + { + "epoch": 1.48, + "learning_rate": 7.830290010741139e-05, + "loss": 0.0371, + "step": 13760 + }, + { + "epoch": 1.48, + "learning_rate": 7.814178302900106e-05, + "loss": 0.0301, + "step": 13770 + }, + { + "epoch": 1.48, + "learning_rate": 7.798066595059076e-05, + "loss": 0.0388, + "step": 13780 + }, + { + "epoch": 1.48, + "learning_rate": 7.781954887218045e-05, + "loss": 0.0316, + "step": 13790 + }, + { + "epoch": 1.48, + "learning_rate": 7.765843179377013e-05, + "loss": 0.035, + "step": 13800 + }, + { + "epoch": 1.48, + "learning_rate": 7.749731471535982e-05, + "loss": 0.0402, + "step": 13810 + }, + { + "epoch": 1.48, + "learning_rate": 7.733619763694951e-05, + "loss": 0.0414, + "step": 13820 + }, + { + "epoch": 1.49, + "learning_rate": 7.717508055853919e-05, + "loss": 0.0357, + "step": 13830 + }, + { + "epoch": 1.49, + "learning_rate": 7.70139634801289e-05, + "loss": 0.0342, + "step": 13840 + }, + { + "epoch": 1.49, + "learning_rate": 7.685284640171859e-05, + "loss": 0.0388, + "step": 13850 + }, + { + "epoch": 1.49, + "learning_rate": 7.669172932330826e-05, + "loss": 0.0356, + "step": 13860 + }, + { + "epoch": 1.49, + "learning_rate": 7.653061224489796e-05, + "loss": 0.0417, + "step": 13870 + }, + { + "epoch": 1.49, + "learning_rate": 7.636949516648765e-05, + "loss": 0.0351, + "step": 13880 + }, + { + "epoch": 1.49, + "learning_rate": 7.620837808807733e-05, + "loss": 0.0347, + "step": 13890 + }, + { + "epoch": 1.49, + "learning_rate": 7.604726100966702e-05, + "loss": 0.0404, + "step": 13900 + }, + { + "epoch": 1.49, + "learning_rate": 7.58861439312567e-05, + "loss": 0.0329, + "step": 13910 + }, + { + "epoch": 1.5, + "learning_rate": 7.572502685284639e-05, + "loss": 0.0325, + "step": 13920 + }, + { + "epoch": 1.5, + "learning_rate": 7.556390977443609e-05, + "loss": 0.0432, + "step": 13930 + }, + { + "epoch": 1.5, + "learning_rate": 7.540279269602577e-05, + "loss": 0.0408, + "step": 13940 + }, + { + "epoch": 1.5, + "learning_rate": 7.524167561761546e-05, + "loss": 0.0385, + "step": 13950 + }, + { + "epoch": 1.5, + "learning_rate": 7.508055853920515e-05, + "loss": 0.0358, + "step": 13960 + }, + { + "epoch": 1.5, + "learning_rate": 7.491944146079485e-05, + "loss": 0.0452, + "step": 13970 + }, + { + "epoch": 1.5, + "learning_rate": 7.475832438238452e-05, + "loss": 0.0395, + "step": 13980 + }, + { + "epoch": 1.5, + "learning_rate": 7.459720730397422e-05, + "loss": 0.0374, + "step": 13990 + }, + { + "epoch": 1.5, + "learning_rate": 7.44360902255639e-05, + "loss": 0.0409, + "step": 14000 + }, + { + "epoch": 1.5, + "eval_cer": 0.9236660150872279, + "eval_loss": 0.031547173857688904, + "eval_runtime": 119.1295, + "eval_samples_per_second": 16.772, + "eval_steps_per_second": 4.197, + "step": 14000 + }, + { + "epoch": 1.5, + "learning_rate": 7.42749731471536e-05, + "loss": 0.0433, + "step": 14010 + }, + { + "epoch": 1.51, + "learning_rate": 7.411385606874328e-05, + "loss": 0.0365, + "step": 14020 + }, + { + "epoch": 1.51, + "learning_rate": 7.395273899033297e-05, + "loss": 0.0374, + "step": 14030 + }, + { + "epoch": 1.51, + "learning_rate": 7.379162191192266e-05, + "loss": 0.0415, + "step": 14040 + }, + { + "epoch": 1.51, + "learning_rate": 7.363050483351234e-05, + "loss": 0.0334, + "step": 14050 + }, + { + "epoch": 1.51, + "learning_rate": 7.346938775510203e-05, + "loss": 0.0345, + "step": 14060 + }, + { + "epoch": 1.51, + "learning_rate": 7.330827067669172e-05, + "loss": 0.0313, + "step": 14070 + }, + { + "epoch": 1.51, + "learning_rate": 7.314715359828142e-05, + "loss": 0.0329, + "step": 14080 + }, + { + "epoch": 1.51, + "learning_rate": 7.29860365198711e-05, + "loss": 0.0413, + "step": 14090 + }, + { + "epoch": 1.51, + "learning_rate": 7.282491944146079e-05, + "loss": 0.0356, + "step": 14100 + }, + { + "epoch": 1.52, + "learning_rate": 7.266380236305048e-05, + "loss": 0.0361, + "step": 14110 + }, + { + "epoch": 1.52, + "learning_rate": 7.250268528464017e-05, + "loss": 0.0309, + "step": 14120 + }, + { + "epoch": 1.52, + "learning_rate": 7.234156820622985e-05, + "loss": 0.0365, + "step": 14130 + }, + { + "epoch": 1.52, + "learning_rate": 7.218045112781954e-05, + "loss": 0.037, + "step": 14140 + }, + { + "epoch": 1.52, + "learning_rate": 7.201933404940923e-05, + "loss": 0.0392, + "step": 14150 + }, + { + "epoch": 1.52, + "learning_rate": 7.185821697099892e-05, + "loss": 0.0334, + "step": 14160 + }, + { + "epoch": 1.52, + "learning_rate": 7.16970998925886e-05, + "loss": 0.0451, + "step": 14170 + }, + { + "epoch": 1.52, + "learning_rate": 7.15359828141783e-05, + "loss": 0.0366, + "step": 14180 + }, + { + "epoch": 1.52, + "learning_rate": 7.137486573576798e-05, + "loss": 0.0374, + "step": 14190 + }, + { + "epoch": 1.53, + "learning_rate": 7.121374865735768e-05, + "loss": 0.0399, + "step": 14200 + }, + { + "epoch": 1.53, + "learning_rate": 7.105263157894735e-05, + "loss": 0.0331, + "step": 14210 + }, + { + "epoch": 1.53, + "learning_rate": 7.089151450053705e-05, + "loss": 0.0415, + "step": 14220 + }, + { + "epoch": 1.53, + "learning_rate": 7.073039742212674e-05, + "loss": 0.0353, + "step": 14230 + }, + { + "epoch": 1.53, + "learning_rate": 7.056928034371643e-05, + "loss": 0.0345, + "step": 14240 + }, + { + "epoch": 1.53, + "learning_rate": 7.040816326530612e-05, + "loss": 0.039, + "step": 14250 + }, + { + "epoch": 1.53, + "learning_rate": 7.02470461868958e-05, + "loss": 0.0355, + "step": 14260 + }, + { + "epoch": 1.53, + "learning_rate": 7.008592910848549e-05, + "loss": 0.0362, + "step": 14270 + }, + { + "epoch": 1.53, + "learning_rate": 6.992481203007518e-05, + "loss": 0.0379, + "step": 14280 + }, + { + "epoch": 1.53, + "learning_rate": 6.976369495166488e-05, + "loss": 0.0359, + "step": 14290 + }, + { + "epoch": 1.54, + "learning_rate": 6.960257787325455e-05, + "loss": 0.041, + "step": 14300 + }, + { + "epoch": 1.54, + "learning_rate": 6.944146079484425e-05, + "loss": 0.0337, + "step": 14310 + }, + { + "epoch": 1.54, + "learning_rate": 6.928034371643394e-05, + "loss": 0.0365, + "step": 14320 + }, + { + "epoch": 1.54, + "learning_rate": 6.911922663802363e-05, + "loss": 0.033, + "step": 14330 + }, + { + "epoch": 1.54, + "learning_rate": 6.895810955961331e-05, + "loss": 0.0353, + "step": 14340 + }, + { + "epoch": 1.54, + "learning_rate": 6.8796992481203e-05, + "loss": 0.0325, + "step": 14350 + }, + { + "epoch": 1.54, + "learning_rate": 6.863587540279269e-05, + "loss": 0.0379, + "step": 14360 + }, + { + "epoch": 1.54, + "learning_rate": 6.847475832438238e-05, + "loss": 0.0382, + "step": 14370 + }, + { + "epoch": 1.54, + "learning_rate": 6.831364124597206e-05, + "loss": 0.0347, + "step": 14380 + }, + { + "epoch": 1.55, + "learning_rate": 6.815252416756175e-05, + "loss": 0.0411, + "step": 14390 + }, + { + "epoch": 1.55, + "learning_rate": 6.799140708915144e-05, + "loss": 0.0307, + "step": 14400 + }, + { + "epoch": 1.55, + "learning_rate": 6.783029001074114e-05, + "loss": 0.0397, + "step": 14410 + }, + { + "epoch": 1.55, + "learning_rate": 6.766917293233081e-05, + "loss": 0.0396, + "step": 14420 + }, + { + "epoch": 1.55, + "learning_rate": 6.75080558539205e-05, + "loss": 0.0362, + "step": 14430 + }, + { + "epoch": 1.55, + "learning_rate": 6.73469387755102e-05, + "loss": 0.0425, + "step": 14440 + }, + { + "epoch": 1.55, + "learning_rate": 6.718582169709989e-05, + "loss": 0.0349, + "step": 14450 + }, + { + "epoch": 1.55, + "learning_rate": 6.702470461868958e-05, + "loss": 0.0343, + "step": 14460 + }, + { + "epoch": 1.55, + "learning_rate": 6.686358754027926e-05, + "loss": 0.0356, + "step": 14470 + }, + { + "epoch": 1.56, + "learning_rate": 6.670247046186895e-05, + "loss": 0.0335, + "step": 14480 + }, + { + "epoch": 1.56, + "learning_rate": 6.654135338345864e-05, + "loss": 0.0278, + "step": 14490 + }, + { + "epoch": 1.56, + "learning_rate": 6.638023630504834e-05, + "loss": 0.0358, + "step": 14500 + }, + { + "epoch": 1.56, + "learning_rate": 6.621911922663801e-05, + "loss": 0.0318, + "step": 14510 + }, + { + "epoch": 1.56, + "learning_rate": 6.60580021482277e-05, + "loss": 0.031, + "step": 14520 + }, + { + "epoch": 1.56, + "learning_rate": 6.58968850698174e-05, + "loss": 0.0364, + "step": 14530 + }, + { + "epoch": 1.56, + "learning_rate": 6.573576799140709e-05, + "loss": 0.0335, + "step": 14540 + }, + { + "epoch": 1.56, + "learning_rate": 6.557465091299677e-05, + "loss": 0.0413, + "step": 14550 + }, + { + "epoch": 1.56, + "learning_rate": 6.541353383458646e-05, + "loss": 0.0382, + "step": 14560 + }, + { + "epoch": 1.56, + "learning_rate": 6.525241675617615e-05, + "loss": 0.0337, + "step": 14570 + }, + { + "epoch": 1.57, + "learning_rate": 6.509129967776584e-05, + "loss": 0.0348, + "step": 14580 + }, + { + "epoch": 1.57, + "learning_rate": 6.493018259935552e-05, + "loss": 0.0369, + "step": 14590 + }, + { + "epoch": 1.57, + "learning_rate": 6.476906552094521e-05, + "loss": 0.0347, + "step": 14600 + }, + { + "epoch": 1.57, + "learning_rate": 6.46079484425349e-05, + "loss": 0.0364, + "step": 14610 + }, + { + "epoch": 1.57, + "learning_rate": 6.44468313641246e-05, + "loss": 0.0363, + "step": 14620 + }, + { + "epoch": 1.57, + "learning_rate": 6.428571428571427e-05, + "loss": 0.04, + "step": 14630 + }, + { + "epoch": 1.57, + "learning_rate": 6.412459720730397e-05, + "loss": 0.0414, + "step": 14640 + }, + { + "epoch": 1.57, + "learning_rate": 6.396348012889366e-05, + "loss": 0.0348, + "step": 14650 + }, + { + "epoch": 1.57, + "learning_rate": 6.380236305048335e-05, + "loss": 0.0373, + "step": 14660 + }, + { + "epoch": 1.58, + "learning_rate": 6.364124597207304e-05, + "loss": 0.0311, + "step": 14670 + }, + { + "epoch": 1.58, + "learning_rate": 6.348012889366272e-05, + "loss": 0.0393, + "step": 14680 + }, + { + "epoch": 1.58, + "learning_rate": 6.331901181525241e-05, + "loss": 0.0329, + "step": 14690 + }, + { + "epoch": 1.58, + "learning_rate": 6.315789473684209e-05, + "loss": 0.0292, + "step": 14700 + }, + { + "epoch": 1.58, + "learning_rate": 6.29967776584318e-05, + "loss": 0.0406, + "step": 14710 + }, + { + "epoch": 1.58, + "learning_rate": 6.283566058002147e-05, + "loss": 0.0336, + "step": 14720 + }, + { + "epoch": 1.58, + "learning_rate": 6.267454350161117e-05, + "loss": 0.0377, + "step": 14730 + }, + { + "epoch": 1.58, + "learning_rate": 6.251342642320086e-05, + "loss": 0.0395, + "step": 14740 + }, + { + "epoch": 1.58, + "learning_rate": 6.235230934479055e-05, + "loss": 0.0326, + "step": 14750 + }, + { + "epoch": 1.59, + "learning_rate": 6.219119226638023e-05, + "loss": 0.0394, + "step": 14760 + }, + { + "epoch": 1.59, + "learning_rate": 6.203007518796992e-05, + "loss": 0.039, + "step": 14770 + }, + { + "epoch": 1.59, + "learning_rate": 6.186895810955961e-05, + "loss": 0.033, + "step": 14780 + }, + { + "epoch": 1.59, + "learning_rate": 6.17078410311493e-05, + "loss": 0.0425, + "step": 14790 + }, + { + "epoch": 1.59, + "learning_rate": 6.154672395273898e-05, + "loss": 0.0371, + "step": 14800 + }, + { + "epoch": 1.59, + "learning_rate": 6.138560687432867e-05, + "loss": 0.0308, + "step": 14810 + }, + { + "epoch": 1.59, + "learning_rate": 6.122448979591836e-05, + "loss": 0.0333, + "step": 14820 + }, + { + "epoch": 1.59, + "learning_rate": 6.106337271750806e-05, + "loss": 0.0325, + "step": 14830 + }, + { + "epoch": 1.59, + "learning_rate": 6.0902255639097734e-05, + "loss": 0.0372, + "step": 14840 + }, + { + "epoch": 1.59, + "learning_rate": 6.074113856068743e-05, + "loss": 0.0375, + "step": 14850 + }, + { + "epoch": 1.6, + "learning_rate": 6.058002148227712e-05, + "loss": 0.0298, + "step": 14860 + }, + { + "epoch": 1.6, + "learning_rate": 6.04189044038668e-05, + "loss": 0.0323, + "step": 14870 + }, + { + "epoch": 1.6, + "learning_rate": 6.0257787325456495e-05, + "loss": 0.034, + "step": 14880 + }, + { + "epoch": 1.6, + "learning_rate": 6.009667024704618e-05, + "loss": 0.0324, + "step": 14890 + }, + { + "epoch": 1.6, + "learning_rate": 5.993555316863587e-05, + "loss": 0.0408, + "step": 14900 + }, + { + "epoch": 1.6, + "learning_rate": 5.9774436090225556e-05, + "loss": 0.0378, + "step": 14910 + }, + { + "epoch": 1.6, + "learning_rate": 5.961331901181525e-05, + "loss": 0.04, + "step": 14920 + }, + { + "epoch": 1.6, + "learning_rate": 5.945220193340493e-05, + "loss": 0.0319, + "step": 14930 + }, + { + "epoch": 1.6, + "learning_rate": 5.9291084854994625e-05, + "loss": 0.0309, + "step": 14940 + }, + { + "epoch": 1.61, + "learning_rate": 5.912996777658432e-05, + "loss": 0.0322, + "step": 14950 + }, + { + "epoch": 1.61, + "learning_rate": 5.8968850698174e-05, + "loss": 0.0334, + "step": 14960 + }, + { + "epoch": 1.61, + "learning_rate": 5.880773361976369e-05, + "loss": 0.0396, + "step": 14970 + }, + { + "epoch": 1.61, + "learning_rate": 5.864661654135337e-05, + "loss": 0.0387, + "step": 14980 + }, + { + "epoch": 1.61, + "learning_rate": 5.848549946294307e-05, + "loss": 0.0377, + "step": 14990 + }, + { + "epoch": 1.61, + "learning_rate": 5.8324382384532756e-05, + "loss": 0.0291, + "step": 15000 + }, + { + "epoch": 1.61, + "eval_cer": 0.9235796508833947, + "eval_loss": 0.03040146268904209, + "eval_runtime": 119.4109, + "eval_samples_per_second": 16.732, + "eval_steps_per_second": 4.187, + "step": 15000 + }, + { + "epoch": 1.61, + "learning_rate": 5.816326530612244e-05, + "loss": 0.0392, + "step": 15010 + }, + { + "epoch": 1.61, + "learning_rate": 5.800214822771214e-05, + "loss": 0.04, + "step": 15020 + }, + { + "epoch": 1.61, + "learning_rate": 5.7841031149301824e-05, + "loss": 0.038, + "step": 15030 + }, + { + "epoch": 1.62, + "learning_rate": 5.767991407089151e-05, + "loss": 0.0372, + "step": 15040 + }, + { + "epoch": 1.62, + "learning_rate": 5.7518796992481194e-05, + "loss": 0.0338, + "step": 15050 + }, + { + "epoch": 1.62, + "learning_rate": 5.7357679914070886e-05, + "loss": 0.0307, + "step": 15060 + }, + { + "epoch": 1.62, + "learning_rate": 5.719656283566058e-05, + "loss": 0.0295, + "step": 15070 + }, + { + "epoch": 1.62, + "learning_rate": 5.703544575725026e-05, + "loss": 0.0403, + "step": 15080 + }, + { + "epoch": 1.62, + "learning_rate": 5.6874328678839955e-05, + "loss": 0.0318, + "step": 15090 + }, + { + "epoch": 1.62, + "learning_rate": 5.671321160042964e-05, + "loss": 0.0303, + "step": 15100 + }, + { + "epoch": 1.62, + "learning_rate": 5.655209452201933e-05, + "loss": 0.0308, + "step": 15110 + }, + { + "epoch": 1.62, + "learning_rate": 5.6390977443609016e-05, + "loss": 0.0366, + "step": 15120 + }, + { + "epoch": 1.62, + "learning_rate": 5.622986036519871e-05, + "loss": 0.0392, + "step": 15130 + }, + { + "epoch": 1.63, + "learning_rate": 5.606874328678839e-05, + "loss": 0.0411, + "step": 15140 + }, + { + "epoch": 1.63, + "learning_rate": 5.590762620837808e-05, + "loss": 0.04, + "step": 15150 + }, + { + "epoch": 1.63, + "learning_rate": 5.574650912996778e-05, + "loss": 0.0375, + "step": 15160 + }, + { + "epoch": 1.63, + "learning_rate": 5.558539205155746e-05, + "loss": 0.0455, + "step": 15170 + }, + { + "epoch": 1.63, + "learning_rate": 5.542427497314715e-05, + "loss": 0.0352, + "step": 15180 + }, + { + "epoch": 1.63, + "learning_rate": 5.526315789473683e-05, + "loss": 0.0332, + "step": 15190 + }, + { + "epoch": 1.63, + "learning_rate": 5.510204081632653e-05, + "loss": 0.0422, + "step": 15200 + }, + { + "epoch": 1.63, + "learning_rate": 5.4940923737916216e-05, + "loss": 0.0474, + "step": 15210 + }, + { + "epoch": 1.63, + "learning_rate": 5.47798066595059e-05, + "loss": 0.0354, + "step": 15220 + }, + { + "epoch": 1.64, + "learning_rate": 5.461868958109559e-05, + "loss": 0.0356, + "step": 15230 + }, + { + "epoch": 1.64, + "learning_rate": 5.4457572502685284e-05, + "loss": 0.0321, + "step": 15240 + }, + { + "epoch": 1.64, + "learning_rate": 5.429645542427497e-05, + "loss": 0.0345, + "step": 15250 + }, + { + "epoch": 1.64, + "learning_rate": 5.4135338345864654e-05, + "loss": 0.0327, + "step": 15260 + }, + { + "epoch": 1.64, + "learning_rate": 5.3974221267454346e-05, + "loss": 0.0359, + "step": 15270 + }, + { + "epoch": 1.64, + "learning_rate": 5.381310418904404e-05, + "loss": 0.0477, + "step": 15280 + }, + { + "epoch": 1.64, + "learning_rate": 5.365198711063372e-05, + "loss": 0.0417, + "step": 15290 + }, + { + "epoch": 1.64, + "learning_rate": 5.349087003222341e-05, + "loss": 0.0349, + "step": 15300 + }, + { + "epoch": 1.64, + "learning_rate": 5.33297529538131e-05, + "loss": 0.0372, + "step": 15310 + }, + { + "epoch": 1.65, + "learning_rate": 5.3168635875402785e-05, + "loss": 0.0406, + "step": 15320 + }, + { + "epoch": 1.65, + "learning_rate": 5.3007518796992476e-05, + "loss": 0.0341, + "step": 15330 + }, + { + "epoch": 1.65, + "learning_rate": 5.284640171858217e-05, + "loss": 0.0335, + "step": 15340 + }, + { + "epoch": 1.65, + "learning_rate": 5.268528464017185e-05, + "loss": 0.0366, + "step": 15350 + }, + { + "epoch": 1.65, + "learning_rate": 5.252416756176154e-05, + "loss": 0.0402, + "step": 15360 + }, + { + "epoch": 1.65, + "learning_rate": 5.236305048335122e-05, + "loss": 0.0304, + "step": 15370 + }, + { + "epoch": 1.65, + "learning_rate": 5.220193340494092e-05, + "loss": 0.0329, + "step": 15380 + }, + { + "epoch": 1.65, + "learning_rate": 5.204081632653061e-05, + "loss": 0.0377, + "step": 15390 + }, + { + "epoch": 1.65, + "learning_rate": 5.187969924812029e-05, + "loss": 0.0354, + "step": 15400 + }, + { + "epoch": 1.66, + "learning_rate": 5.171858216970999e-05, + "loss": 0.0289, + "step": 15410 + }, + { + "epoch": 1.66, + "learning_rate": 5.1557465091299675e-05, + "loss": 0.0291, + "step": 15420 + }, + { + "epoch": 1.66, + "learning_rate": 5.139634801288936e-05, + "loss": 0.0328, + "step": 15430 + }, + { + "epoch": 1.66, + "learning_rate": 5.1235230934479046e-05, + "loss": 0.0265, + "step": 15440 + }, + { + "epoch": 1.66, + "learning_rate": 5.1074113856068744e-05, + "loss": 0.0363, + "step": 15450 + }, + { + "epoch": 1.66, + "learning_rate": 5.091299677765843e-05, + "loss": 0.0294, + "step": 15460 + }, + { + "epoch": 1.66, + "learning_rate": 5.0751879699248114e-05, + "loss": 0.0355, + "step": 15470 + }, + { + "epoch": 1.66, + "learning_rate": 5.0590762620837806e-05, + "loss": 0.0344, + "step": 15480 + }, + { + "epoch": 1.66, + "learning_rate": 5.042964554242749e-05, + "loss": 0.0334, + "step": 15490 + }, + { + "epoch": 1.66, + "learning_rate": 5.026852846401718e-05, + "loss": 0.0349, + "step": 15500 + }, + { + "epoch": 1.67, + "learning_rate": 5.010741138560687e-05, + "loss": 0.029, + "step": 15510 + }, + { + "epoch": 1.67, + "learning_rate": 4.994629430719656e-05, + "loss": 0.0372, + "step": 15520 + }, + { + "epoch": 1.67, + "learning_rate": 4.9785177228786245e-05, + "loss": 0.0274, + "step": 15530 + }, + { + "epoch": 1.67, + "learning_rate": 4.962406015037593e-05, + "loss": 0.0404, + "step": 15540 + }, + { + "epoch": 1.67, + "learning_rate": 4.946294307196563e-05, + "loss": 0.0293, + "step": 15550 + }, + { + "epoch": 1.67, + "learning_rate": 4.930182599355531e-05, + "loss": 0.0383, + "step": 15560 + }, + { + "epoch": 1.67, + "learning_rate": 4.9140708915145e-05, + "loss": 0.0362, + "step": 15570 + }, + { + "epoch": 1.67, + "learning_rate": 4.897959183673468e-05, + "loss": 0.0312, + "step": 15580 + }, + { + "epoch": 1.67, + "learning_rate": 4.881847475832438e-05, + "loss": 0.0336, + "step": 15590 + }, + { + "epoch": 1.68, + "learning_rate": 4.865735767991407e-05, + "loss": 0.0305, + "step": 15600 + }, + { + "epoch": 1.68, + "learning_rate": 4.849624060150375e-05, + "loss": 0.0378, + "step": 15610 + }, + { + "epoch": 1.68, + "learning_rate": 4.8335123523093444e-05, + "loss": 0.0348, + "step": 15620 + }, + { + "epoch": 1.68, + "learning_rate": 4.8174006444683135e-05, + "loss": 0.0366, + "step": 15630 + }, + { + "epoch": 1.68, + "learning_rate": 4.801288936627282e-05, + "loss": 0.0385, + "step": 15640 + }, + { + "epoch": 1.68, + "learning_rate": 4.7851772287862505e-05, + "loss": 0.035, + "step": 15650 + }, + { + "epoch": 1.68, + "learning_rate": 4.76906552094522e-05, + "loss": 0.0337, + "step": 15660 + }, + { + "epoch": 1.68, + "learning_rate": 4.752953813104189e-05, + "loss": 0.0363, + "step": 15670 + }, + { + "epoch": 1.68, + "learning_rate": 4.7368421052631574e-05, + "loss": 0.0359, + "step": 15680 + }, + { + "epoch": 1.69, + "learning_rate": 4.7207303974221266e-05, + "loss": 0.0333, + "step": 15690 + }, + { + "epoch": 1.69, + "learning_rate": 4.704618689581095e-05, + "loss": 0.0347, + "step": 15700 + }, + { + "epoch": 1.69, + "learning_rate": 4.6885069817400636e-05, + "loss": 0.0334, + "step": 15710 + }, + { + "epoch": 1.69, + "learning_rate": 4.672395273899033e-05, + "loss": 0.039, + "step": 15720 + }, + { + "epoch": 1.69, + "learning_rate": 4.656283566058002e-05, + "loss": 0.0347, + "step": 15730 + }, + { + "epoch": 1.69, + "learning_rate": 4.6401718582169705e-05, + "loss": 0.033, + "step": 15740 + }, + { + "epoch": 1.69, + "learning_rate": 4.624060150375939e-05, + "loss": 0.0319, + "step": 15750 + }, + { + "epoch": 1.69, + "learning_rate": 4.607948442534909e-05, + "loss": 0.0339, + "step": 15760 + }, + { + "epoch": 1.69, + "learning_rate": 4.591836734693877e-05, + "loss": 0.0383, + "step": 15770 + }, + { + "epoch": 1.69, + "learning_rate": 4.575725026852846e-05, + "loss": 0.0419, + "step": 15780 + }, + { + "epoch": 1.7, + "learning_rate": 4.559613319011814e-05, + "loss": 0.0281, + "step": 15790 + }, + { + "epoch": 1.7, + "learning_rate": 4.543501611170784e-05, + "loss": 0.0376, + "step": 15800 + }, + { + "epoch": 1.7, + "learning_rate": 4.527389903329753e-05, + "loss": 0.0281, + "step": 15810 + }, + { + "epoch": 1.7, + "learning_rate": 4.511278195488721e-05, + "loss": 0.0351, + "step": 15820 + }, + { + "epoch": 1.7, + "learning_rate": 4.4951664876476904e-05, + "loss": 0.0366, + "step": 15830 + }, + { + "epoch": 1.7, + "learning_rate": 4.4790547798066595e-05, + "loss": 0.034, + "step": 15840 + }, + { + "epoch": 1.7, + "learning_rate": 4.462943071965628e-05, + "loss": 0.0357, + "step": 15850 + }, + { + "epoch": 1.7, + "learning_rate": 4.4468313641245965e-05, + "loss": 0.0315, + "step": 15860 + }, + { + "epoch": 1.7, + "learning_rate": 4.430719656283566e-05, + "loss": 0.0364, + "step": 15870 + }, + { + "epoch": 1.71, + "learning_rate": 4.414607948442534e-05, + "loss": 0.0411, + "step": 15880 + }, + { + "epoch": 1.71, + "learning_rate": 4.3984962406015034e-05, + "loss": 0.0369, + "step": 15890 + }, + { + "epoch": 1.71, + "learning_rate": 4.3823845327604726e-05, + "loss": 0.0362, + "step": 15900 + }, + { + "epoch": 1.71, + "learning_rate": 4.366272824919441e-05, + "loss": 0.0348, + "step": 15910 + }, + { + "epoch": 1.71, + "learning_rate": 4.3501611170784096e-05, + "loss": 0.0362, + "step": 15920 + }, + { + "epoch": 1.71, + "learning_rate": 4.334049409237378e-05, + "loss": 0.0274, + "step": 15930 + }, + { + "epoch": 1.71, + "learning_rate": 4.317937701396348e-05, + "loss": 0.0329, + "step": 15940 + }, + { + "epoch": 1.71, + "learning_rate": 4.3018259935553164e-05, + "loss": 0.0346, + "step": 15950 + }, + { + "epoch": 1.71, + "learning_rate": 4.285714285714285e-05, + "loss": 0.0377, + "step": 15960 + }, + { + "epoch": 1.72, + "learning_rate": 4.269602577873255e-05, + "loss": 0.0325, + "step": 15970 + }, + { + "epoch": 1.72, + "learning_rate": 4.253490870032223e-05, + "loss": 0.0327, + "step": 15980 + }, + { + "epoch": 1.72, + "learning_rate": 4.237379162191192e-05, + "loss": 0.0307, + "step": 15990 + }, + { + "epoch": 1.72, + "learning_rate": 4.22126745435016e-05, + "loss": 0.0268, + "step": 16000 + }, + { + "epoch": 1.72, + "eval_cer": 0.92358392633903, + "eval_loss": 0.029257148504257202, + "eval_runtime": 118.9085, + "eval_samples_per_second": 16.803, + "eval_steps_per_second": 4.205, + "step": 16000 + }, + { + "epoch": 1.72, + "learning_rate": 4.20515574650913e-05, + "loss": 0.0328, + "step": 16010 + }, + { + "epoch": 1.72, + "learning_rate": 4.189044038668099e-05, + "loss": 0.0355, + "step": 16020 + }, + { + "epoch": 1.72, + "learning_rate": 4.172932330827067e-05, + "loss": 0.0416, + "step": 16030 + }, + { + "epoch": 1.72, + "learning_rate": 4.1568206229860364e-05, + "loss": 0.0289, + "step": 16040 + }, + { + "epoch": 1.72, + "learning_rate": 4.140708915145005e-05, + "loss": 0.0344, + "step": 16050 + }, + { + "epoch": 1.72, + "learning_rate": 4.124597207303974e-05, + "loss": 0.0318, + "step": 16060 + }, + { + "epoch": 1.73, + "learning_rate": 4.1084854994629425e-05, + "loss": 0.0351, + "step": 16070 + }, + { + "epoch": 1.73, + "learning_rate": 4.092373791621912e-05, + "loss": 0.0359, + "step": 16080 + }, + { + "epoch": 1.73, + "learning_rate": 4.07626208378088e-05, + "loss": 0.0336, + "step": 16090 + }, + { + "epoch": 1.73, + "learning_rate": 4.060150375939849e-05, + "loss": 0.0319, + "step": 16100 + }, + { + "epoch": 1.73, + "learning_rate": 4.0440386680988186e-05, + "loss": 0.0394, + "step": 16110 + }, + { + "epoch": 1.73, + "learning_rate": 4.027926960257787e-05, + "loss": 0.0351, + "step": 16120 + }, + { + "epoch": 1.73, + "learning_rate": 4.0118152524167556e-05, + "loss": 0.0269, + "step": 16130 + }, + { + "epoch": 1.73, + "learning_rate": 3.995703544575724e-05, + "loss": 0.0354, + "step": 16140 + }, + { + "epoch": 1.73, + "learning_rate": 3.979591836734694e-05, + "loss": 0.0417, + "step": 16150 + }, + { + "epoch": 1.74, + "learning_rate": 3.9634801288936624e-05, + "loss": 0.0364, + "step": 16160 + }, + { + "epoch": 1.74, + "learning_rate": 3.947368421052631e-05, + "loss": 0.0378, + "step": 16170 + }, + { + "epoch": 1.74, + "learning_rate": 3.9312567132116e-05, + "loss": 0.0392, + "step": 16180 + }, + { + "epoch": 1.74, + "learning_rate": 3.915145005370569e-05, + "loss": 0.0299, + "step": 16190 + }, + { + "epoch": 1.74, + "learning_rate": 3.899033297529538e-05, + "loss": 0.0311, + "step": 16200 + }, + { + "epoch": 1.74, + "learning_rate": 3.882921589688506e-05, + "loss": 0.0374, + "step": 16210 + }, + { + "epoch": 1.74, + "learning_rate": 3.8668098818474755e-05, + "loss": 0.0247, + "step": 16220 + }, + { + "epoch": 1.74, + "learning_rate": 3.850698174006445e-05, + "loss": 0.0331, + "step": 16230 + }, + { + "epoch": 1.74, + "learning_rate": 3.834586466165413e-05, + "loss": 0.0302, + "step": 16240 + }, + { + "epoch": 1.75, + "learning_rate": 3.8184747583243824e-05, + "loss": 0.0295, + "step": 16250 + }, + { + "epoch": 1.75, + "learning_rate": 3.802363050483351e-05, + "loss": 0.0354, + "step": 16260 + }, + { + "epoch": 1.75, + "learning_rate": 3.7862513426423194e-05, + "loss": 0.0314, + "step": 16270 + }, + { + "epoch": 1.75, + "learning_rate": 3.7701396348012885e-05, + "loss": 0.0285, + "step": 16280 + }, + { + "epoch": 1.75, + "learning_rate": 3.754027926960258e-05, + "loss": 0.0368, + "step": 16290 + }, + { + "epoch": 1.75, + "learning_rate": 3.737916219119226e-05, + "loss": 0.0394, + "step": 16300 + }, + { + "epoch": 1.75, + "learning_rate": 3.721804511278195e-05, + "loss": 0.0314, + "step": 16310 + }, + { + "epoch": 1.75, + "learning_rate": 3.705692803437164e-05, + "loss": 0.034, + "step": 16320 + }, + { + "epoch": 1.75, + "learning_rate": 3.689581095596133e-05, + "loss": 0.0287, + "step": 16330 + }, + { + "epoch": 1.75, + "learning_rate": 3.6734693877551016e-05, + "loss": 0.0346, + "step": 16340 + }, + { + "epoch": 1.76, + "learning_rate": 3.657357679914071e-05, + "loss": 0.0346, + "step": 16350 + }, + { + "epoch": 1.76, + "learning_rate": 3.641245972073039e-05, + "loss": 0.0389, + "step": 16360 + }, + { + "epoch": 1.76, + "learning_rate": 3.6251342642320084e-05, + "loss": 0.0307, + "step": 16370 + }, + { + "epoch": 1.76, + "learning_rate": 3.609022556390977e-05, + "loss": 0.033, + "step": 16380 + }, + { + "epoch": 1.76, + "learning_rate": 3.592910848549946e-05, + "loss": 0.0369, + "step": 16390 + }, + { + "epoch": 1.76, + "learning_rate": 3.576799140708915e-05, + "loss": 0.0322, + "step": 16400 + }, + { + "epoch": 1.76, + "learning_rate": 3.560687432867884e-05, + "loss": 0.0328, + "step": 16410 + }, + { + "epoch": 1.76, + "learning_rate": 3.544575725026852e-05, + "loss": 0.0288, + "step": 16420 + }, + { + "epoch": 1.76, + "learning_rate": 3.5284640171858215e-05, + "loss": 0.0356, + "step": 16430 + }, + { + "epoch": 1.77, + "learning_rate": 3.51235230934479e-05, + "loss": 0.0338, + "step": 16440 + }, + { + "epoch": 1.77, + "learning_rate": 3.496240601503759e-05, + "loss": 0.0346, + "step": 16450 + }, + { + "epoch": 1.77, + "learning_rate": 3.480128893662728e-05, + "loss": 0.0364, + "step": 16460 + }, + { + "epoch": 1.77, + "learning_rate": 3.464017185821697e-05, + "loss": 0.0294, + "step": 16470 + }, + { + "epoch": 1.77, + "learning_rate": 3.4479054779806654e-05, + "loss": 0.0324, + "step": 16480 + }, + { + "epoch": 1.77, + "learning_rate": 3.4317937701396345e-05, + "loss": 0.0338, + "step": 16490 + }, + { + "epoch": 1.77, + "learning_rate": 3.415682062298603e-05, + "loss": 0.0348, + "step": 16500 + }, + { + "epoch": 1.77, + "learning_rate": 3.399570354457572e-05, + "loss": 0.0337, + "step": 16510 + }, + { + "epoch": 1.77, + "learning_rate": 3.383458646616541e-05, + "loss": 0.0321, + "step": 16520 + }, + { + "epoch": 1.78, + "learning_rate": 3.36734693877551e-05, + "loss": 0.0266, + "step": 16530 + }, + { + "epoch": 1.78, + "learning_rate": 3.351235230934479e-05, + "loss": 0.0332, + "step": 16540 + }, + { + "epoch": 1.78, + "learning_rate": 3.3351235230934476e-05, + "loss": 0.0419, + "step": 16550 + }, + { + "epoch": 1.78, + "learning_rate": 3.319011815252417e-05, + "loss": 0.0323, + "step": 16560 + }, + { + "epoch": 1.78, + "learning_rate": 3.302900107411385e-05, + "loss": 0.0332, + "step": 16570 + }, + { + "epoch": 1.78, + "learning_rate": 3.2867883995703544e-05, + "loss": 0.0317, + "step": 16580 + }, + { + "epoch": 1.78, + "learning_rate": 3.270676691729323e-05, + "loss": 0.029, + "step": 16590 + }, + { + "epoch": 1.78, + "learning_rate": 3.254564983888292e-05, + "loss": 0.0284, + "step": 16600 + }, + { + "epoch": 1.78, + "learning_rate": 3.2384532760472606e-05, + "loss": 0.0275, + "step": 16610 + }, + { + "epoch": 1.79, + "learning_rate": 3.22234156820623e-05, + "loss": 0.0274, + "step": 16620 + }, + { + "epoch": 1.79, + "learning_rate": 3.206229860365198e-05, + "loss": 0.0321, + "step": 16630 + }, + { + "epoch": 1.79, + "learning_rate": 3.1901181525241675e-05, + "loss": 0.0398, + "step": 16640 + }, + { + "epoch": 1.79, + "learning_rate": 3.174006444683136e-05, + "loss": 0.0304, + "step": 16650 + }, + { + "epoch": 1.79, + "learning_rate": 3.1578947368421045e-05, + "loss": 0.0372, + "step": 16660 + }, + { + "epoch": 1.79, + "learning_rate": 3.141783029001074e-05, + "loss": 0.0329, + "step": 16670 + }, + { + "epoch": 1.79, + "learning_rate": 3.125671321160043e-05, + "loss": 0.0281, + "step": 16680 + }, + { + "epoch": 1.79, + "learning_rate": 3.1095596133190113e-05, + "loss": 0.0365, + "step": 16690 + }, + { + "epoch": 1.79, + "learning_rate": 3.0934479054779805e-05, + "loss": 0.0466, + "step": 16700 + }, + { + "epoch": 1.79, + "learning_rate": 3.077336197636949e-05, + "loss": 0.0318, + "step": 16710 + }, + { + "epoch": 1.8, + "learning_rate": 3.061224489795918e-05, + "loss": 0.0359, + "step": 16720 + }, + { + "epoch": 1.8, + "learning_rate": 3.0451127819548867e-05, + "loss": 0.0412, + "step": 16730 + }, + { + "epoch": 1.8, + "learning_rate": 3.029001074113856e-05, + "loss": 0.0319, + "step": 16740 + }, + { + "epoch": 1.8, + "learning_rate": 3.0128893662728247e-05, + "loss": 0.0355, + "step": 16750 + }, + { + "epoch": 1.8, + "learning_rate": 2.9967776584317936e-05, + "loss": 0.0325, + "step": 16760 + }, + { + "epoch": 1.8, + "learning_rate": 2.9806659505907624e-05, + "loss": 0.0289, + "step": 16770 + }, + { + "epoch": 1.8, + "learning_rate": 2.9645542427497313e-05, + "loss": 0.0263, + "step": 16780 + }, + { + "epoch": 1.8, + "learning_rate": 2.9484425349087e-05, + "loss": 0.0336, + "step": 16790 + }, + { + "epoch": 1.8, + "learning_rate": 2.9323308270676686e-05, + "loss": 0.031, + "step": 16800 + }, + { + "epoch": 1.81, + "learning_rate": 2.9162191192266378e-05, + "loss": 0.0281, + "step": 16810 + }, + { + "epoch": 1.81, + "learning_rate": 2.900107411385607e-05, + "loss": 0.0346, + "step": 16820 + }, + { + "epoch": 1.81, + "learning_rate": 2.8839957035445755e-05, + "loss": 0.031, + "step": 16830 + }, + { + "epoch": 1.81, + "learning_rate": 2.8678839957035443e-05, + "loss": 0.0293, + "step": 16840 + }, + { + "epoch": 1.81, + "learning_rate": 2.851772287862513e-05, + "loss": 0.0308, + "step": 16850 + }, + { + "epoch": 1.81, + "learning_rate": 2.835660580021482e-05, + "loss": 0.0274, + "step": 16860 + }, + { + "epoch": 1.81, + "learning_rate": 2.8195488721804508e-05, + "loss": 0.0291, + "step": 16870 + }, + { + "epoch": 1.81, + "learning_rate": 2.8034371643394197e-05, + "loss": 0.0311, + "step": 16880 + }, + { + "epoch": 1.81, + "learning_rate": 2.787325456498389e-05, + "loss": 0.0314, + "step": 16890 + }, + { + "epoch": 1.82, + "learning_rate": 2.7712137486573573e-05, + "loss": 0.0358, + "step": 16900 + }, + { + "epoch": 1.82, + "learning_rate": 2.7551020408163265e-05, + "loss": 0.0446, + "step": 16910 + }, + { + "epoch": 1.82, + "learning_rate": 2.738990332975295e-05, + "loss": 0.0297, + "step": 16920 + }, + { + "epoch": 1.82, + "learning_rate": 2.7228786251342642e-05, + "loss": 0.03, + "step": 16930 + }, + { + "epoch": 1.82, + "learning_rate": 2.7067669172932327e-05, + "loss": 0.0343, + "step": 16940 + }, + { + "epoch": 1.82, + "learning_rate": 2.690655209452202e-05, + "loss": 0.0313, + "step": 16950 + }, + { + "epoch": 1.82, + "learning_rate": 2.6745435016111704e-05, + "loss": 0.0299, + "step": 16960 + }, + { + "epoch": 1.82, + "learning_rate": 2.6584317937701392e-05, + "loss": 0.0338, + "step": 16970 + }, + { + "epoch": 1.82, + "learning_rate": 2.6423200859291084e-05, + "loss": 0.0327, + "step": 16980 + }, + { + "epoch": 1.82, + "learning_rate": 2.626208378088077e-05, + "loss": 0.0362, + "step": 16990 + }, + { + "epoch": 1.83, + "learning_rate": 2.610096670247046e-05, + "loss": 0.0309, + "step": 17000 + }, + { + "epoch": 1.83, + "eval_cer": 0.9235215046867544, + "eval_loss": 0.02837272547185421, + "eval_runtime": 118.9868, + "eval_samples_per_second": 16.792, + "eval_steps_per_second": 4.202, + "step": 17000 + }, + { + "epoch": 1.83, + "learning_rate": 2.5939849624060146e-05, + "loss": 0.0315, + "step": 17010 + }, + { + "epoch": 1.83, + "learning_rate": 2.5778732545649838e-05, + "loss": 0.0324, + "step": 17020 + }, + { + "epoch": 1.83, + "learning_rate": 2.5617615467239523e-05, + "loss": 0.0316, + "step": 17030 + }, + { + "epoch": 1.83, + "learning_rate": 2.5456498388829215e-05, + "loss": 0.0273, + "step": 17040 + }, + { + "epoch": 1.83, + "learning_rate": 2.5295381310418903e-05, + "loss": 0.0307, + "step": 17050 + }, + { + "epoch": 1.83, + "learning_rate": 2.513426423200859e-05, + "loss": 0.0338, + "step": 17060 + }, + { + "epoch": 1.83, + "learning_rate": 2.497314715359828e-05, + "loss": 0.0311, + "step": 17070 + }, + { + "epoch": 1.83, + "learning_rate": 2.4812030075187965e-05, + "loss": 0.0301, + "step": 17080 + }, + { + "epoch": 1.84, + "learning_rate": 2.4650912996777657e-05, + "loss": 0.0361, + "step": 17090 + }, + { + "epoch": 1.84, + "learning_rate": 2.448979591836734e-05, + "loss": 0.0297, + "step": 17100 + }, + { + "epoch": 1.84, + "learning_rate": 2.4328678839957033e-05, + "loss": 0.0421, + "step": 17110 + }, + { + "epoch": 1.84, + "learning_rate": 2.4167561761546722e-05, + "loss": 0.0319, + "step": 17120 + }, + { + "epoch": 1.84, + "learning_rate": 2.400644468313641e-05, + "loss": 0.037, + "step": 17130 + }, + { + "epoch": 1.84, + "learning_rate": 2.38453276047261e-05, + "loss": 0.0339, + "step": 17140 + }, + { + "epoch": 1.84, + "learning_rate": 2.3684210526315787e-05, + "loss": 0.0323, + "step": 17150 + }, + { + "epoch": 1.84, + "learning_rate": 2.3523093447905475e-05, + "loss": 0.0329, + "step": 17160 + }, + { + "epoch": 1.84, + "learning_rate": 2.3361976369495164e-05, + "loss": 0.0326, + "step": 17170 + }, + { + "epoch": 1.85, + "learning_rate": 2.3200859291084852e-05, + "loss": 0.0351, + "step": 17180 + }, + { + "epoch": 1.85, + "learning_rate": 2.3039742212674544e-05, + "loss": 0.0291, + "step": 17190 + }, + { + "epoch": 1.85, + "learning_rate": 2.287862513426423e-05, + "loss": 0.0283, + "step": 17200 + }, + { + "epoch": 1.85, + "learning_rate": 2.271750805585392e-05, + "loss": 0.0335, + "step": 17210 + }, + { + "epoch": 1.85, + "learning_rate": 2.2556390977443606e-05, + "loss": 0.0327, + "step": 17220 + }, + { + "epoch": 1.85, + "learning_rate": 2.2395273899033298e-05, + "loss": 0.0328, + "step": 17230 + }, + { + "epoch": 1.85, + "learning_rate": 2.2234156820622983e-05, + "loss": 0.0329, + "step": 17240 + }, + { + "epoch": 1.85, + "learning_rate": 2.207303974221267e-05, + "loss": 0.0315, + "step": 17250 + }, + { + "epoch": 1.85, + "learning_rate": 2.1911922663802363e-05, + "loss": 0.0302, + "step": 17260 + }, + { + "epoch": 1.85, + "learning_rate": 2.1750805585392048e-05, + "loss": 0.0368, + "step": 17270 + }, + { + "epoch": 1.86, + "learning_rate": 2.158968850698174e-05, + "loss": 0.0353, + "step": 17280 + }, + { + "epoch": 1.86, + "learning_rate": 2.1428571428571425e-05, + "loss": 0.0299, + "step": 17290 + }, + { + "epoch": 1.86, + "learning_rate": 2.1267454350161117e-05, + "loss": 0.032, + "step": 17300 + }, + { + "epoch": 1.86, + "learning_rate": 2.11063372717508e-05, + "loss": 0.0305, + "step": 17310 + }, + { + "epoch": 1.86, + "learning_rate": 2.0945220193340493e-05, + "loss": 0.0327, + "step": 17320 + }, + { + "epoch": 1.86, + "learning_rate": 2.0784103114930182e-05, + "loss": 0.0324, + "step": 17330 + }, + { + "epoch": 1.86, + "learning_rate": 2.062298603651987e-05, + "loss": 0.0277, + "step": 17340 + }, + { + "epoch": 1.86, + "learning_rate": 2.046186895810956e-05, + "loss": 0.0332, + "step": 17350 + }, + { + "epoch": 1.86, + "learning_rate": 2.0300751879699244e-05, + "loss": 0.0325, + "step": 17360 + }, + { + "epoch": 1.87, + "learning_rate": 2.0139634801288935e-05, + "loss": 0.0337, + "step": 17370 + }, + { + "epoch": 1.87, + "learning_rate": 1.997851772287862e-05, + "loss": 0.0303, + "step": 17380 + }, + { + "epoch": 1.87, + "learning_rate": 1.9817400644468312e-05, + "loss": 0.0356, + "step": 17390 + }, + { + "epoch": 1.87, + "learning_rate": 1.9656283566058e-05, + "loss": 0.0292, + "step": 17400 + }, + { + "epoch": 1.87, + "learning_rate": 1.949516648764769e-05, + "loss": 0.0384, + "step": 17410 + }, + { + "epoch": 1.87, + "learning_rate": 1.9334049409237377e-05, + "loss": 0.031, + "step": 17420 + }, + { + "epoch": 1.87, + "learning_rate": 1.9172932330827066e-05, + "loss": 0.0336, + "step": 17430 + }, + { + "epoch": 1.87, + "learning_rate": 1.9011815252416754e-05, + "loss": 0.0335, + "step": 17440 + }, + { + "epoch": 1.87, + "learning_rate": 1.8850698174006443e-05, + "loss": 0.0323, + "step": 17450 + }, + { + "epoch": 1.88, + "learning_rate": 1.868958109559613e-05, + "loss": 0.0322, + "step": 17460 + }, + { + "epoch": 1.88, + "learning_rate": 1.852846401718582e-05, + "loss": 0.0293, + "step": 17470 + }, + { + "epoch": 1.88, + "learning_rate": 1.8367346938775508e-05, + "loss": 0.0352, + "step": 17480 + }, + { + "epoch": 1.88, + "learning_rate": 1.8206229860365196e-05, + "loss": 0.0292, + "step": 17490 + }, + { + "epoch": 1.88, + "learning_rate": 1.8045112781954885e-05, + "loss": 0.0304, + "step": 17500 + }, + { + "epoch": 1.88, + "learning_rate": 1.7883995703544577e-05, + "loss": 0.0291, + "step": 17510 + }, + { + "epoch": 1.88, + "learning_rate": 1.772287862513426e-05, + "loss": 0.035, + "step": 17520 + }, + { + "epoch": 1.88, + "learning_rate": 1.756176154672395e-05, + "loss": 0.0289, + "step": 17530 + }, + { + "epoch": 1.88, + "learning_rate": 1.740064446831364e-05, + "loss": 0.03, + "step": 17540 + }, + { + "epoch": 1.88, + "learning_rate": 1.7239527389903327e-05, + "loss": 0.0344, + "step": 17550 + }, + { + "epoch": 1.89, + "learning_rate": 1.7078410311493015e-05, + "loss": 0.0277, + "step": 17560 + }, + { + "epoch": 1.89, + "learning_rate": 1.6917293233082704e-05, + "loss": 0.029, + "step": 17570 + }, + { + "epoch": 1.89, + "learning_rate": 1.6756176154672395e-05, + "loss": 0.0303, + "step": 17580 + }, + { + "epoch": 1.89, + "learning_rate": 1.6595059076262084e-05, + "loss": 0.0309, + "step": 17590 + }, + { + "epoch": 1.89, + "learning_rate": 1.6433941997851772e-05, + "loss": 0.0313, + "step": 17600 + }, + { + "epoch": 1.89, + "learning_rate": 1.627282491944146e-05, + "loss": 0.0297, + "step": 17610 + }, + { + "epoch": 1.89, + "learning_rate": 1.611170784103115e-05, + "loss": 0.0324, + "step": 17620 + }, + { + "epoch": 1.89, + "learning_rate": 1.5950590762620837e-05, + "loss": 0.0317, + "step": 17630 + }, + { + "epoch": 1.89, + "learning_rate": 1.5789473684210522e-05, + "loss": 0.033, + "step": 17640 + }, + { + "epoch": 1.9, + "learning_rate": 1.5628356605800214e-05, + "loss": 0.0293, + "step": 17650 + }, + { + "epoch": 1.9, + "learning_rate": 1.5467239527389903e-05, + "loss": 0.0364, + "step": 17660 + }, + { + "epoch": 1.9, + "learning_rate": 1.530612244897959e-05, + "loss": 0.0312, + "step": 17670 + }, + { + "epoch": 1.9, + "learning_rate": 1.514500537056928e-05, + "loss": 0.0301, + "step": 17680 + }, + { + "epoch": 1.9, + "learning_rate": 1.4983888292158968e-05, + "loss": 0.0317, + "step": 17690 + }, + { + "epoch": 1.9, + "learning_rate": 1.4822771213748656e-05, + "loss": 0.0368, + "step": 17700 + }, + { + "epoch": 1.9, + "learning_rate": 1.4661654135338343e-05, + "loss": 0.0338, + "step": 17710 + }, + { + "epoch": 1.9, + "learning_rate": 1.4500537056928035e-05, + "loss": 0.0245, + "step": 17720 + }, + { + "epoch": 1.9, + "learning_rate": 1.4339419978517721e-05, + "loss": 0.0319, + "step": 17730 + }, + { + "epoch": 1.91, + "learning_rate": 1.417830290010741e-05, + "loss": 0.0389, + "step": 17740 + }, + { + "epoch": 1.91, + "learning_rate": 1.4017185821697098e-05, + "loss": 0.0323, + "step": 17750 + }, + { + "epoch": 1.91, + "learning_rate": 1.3856068743286787e-05, + "loss": 0.0355, + "step": 17760 + }, + { + "epoch": 1.91, + "learning_rate": 1.3694951664876475e-05, + "loss": 0.0296, + "step": 17770 + }, + { + "epoch": 1.91, + "learning_rate": 1.3533834586466164e-05, + "loss": 0.0336, + "step": 17780 + }, + { + "epoch": 1.91, + "learning_rate": 1.3372717508055852e-05, + "loss": 0.0313, + "step": 17790 + }, + { + "epoch": 1.91, + "learning_rate": 1.3211600429645542e-05, + "loss": 0.0363, + "step": 17800 + }, + { + "epoch": 1.91, + "learning_rate": 1.305048335123523e-05, + "loss": 0.0307, + "step": 17810 + }, + { + "epoch": 1.91, + "learning_rate": 1.2889366272824919e-05, + "loss": 0.0333, + "step": 17820 + }, + { + "epoch": 1.91, + "learning_rate": 1.2728249194414607e-05, + "loss": 0.0294, + "step": 17830 + }, + { + "epoch": 1.92, + "learning_rate": 1.2567132116004296e-05, + "loss": 0.0309, + "step": 17840 + }, + { + "epoch": 1.92, + "learning_rate": 1.2406015037593982e-05, + "loss": 0.0319, + "step": 17850 + }, + { + "epoch": 1.92, + "learning_rate": 1.224489795918367e-05, + "loss": 0.0257, + "step": 17860 + }, + { + "epoch": 1.92, + "learning_rate": 1.2083780880773361e-05, + "loss": 0.0316, + "step": 17870 + }, + { + "epoch": 1.92, + "learning_rate": 1.192266380236305e-05, + "loss": 0.0284, + "step": 17880 + }, + { + "epoch": 1.92, + "learning_rate": 1.1761546723952738e-05, + "loss": 0.0318, + "step": 17890 + }, + { + "epoch": 1.92, + "learning_rate": 1.1600429645542426e-05, + "loss": 0.0265, + "step": 17900 + }, + { + "epoch": 1.92, + "learning_rate": 1.1439312567132115e-05, + "loss": 0.0282, + "step": 17910 + }, + { + "epoch": 1.92, + "learning_rate": 1.1278195488721803e-05, + "loss": 0.0331, + "step": 17920 + }, + { + "epoch": 1.93, + "learning_rate": 1.1117078410311491e-05, + "loss": 0.0395, + "step": 17930 + }, + { + "epoch": 1.93, + "learning_rate": 1.0955961331901181e-05, + "loss": 0.0259, + "step": 17940 + }, + { + "epoch": 1.93, + "learning_rate": 1.079484425349087e-05, + "loss": 0.0314, + "step": 17950 + }, + { + "epoch": 1.93, + "learning_rate": 1.0633727175080558e-05, + "loss": 0.0337, + "step": 17960 + }, + { + "epoch": 1.93, + "learning_rate": 1.0472610096670247e-05, + "loss": 0.034, + "step": 17970 + }, + { + "epoch": 1.93, + "learning_rate": 1.0311493018259935e-05, + "loss": 0.0325, + "step": 17980 + }, + { + "epoch": 1.93, + "learning_rate": 1.0150375939849622e-05, + "loss": 0.0287, + "step": 17990 + }, + { + "epoch": 1.93, + "learning_rate": 9.98925886143931e-06, + "loss": 0.0362, + "step": 18000 + }, + { + "epoch": 1.93, + "eval_cer": 0.9235026926819592, + "eval_loss": 0.02764066681265831, + "eval_runtime": 119.0918, + "eval_samples_per_second": 16.777, + "eval_steps_per_second": 4.198, + "step": 18000 + }, + { + "epoch": 1.93, + "learning_rate": 9.828141783029e-06, + "loss": 0.0338, + "step": 18010 + }, + { + "epoch": 1.94, + "learning_rate": 9.667024704618689e-06, + "loss": 0.0331, + "step": 18020 + }, + { + "epoch": 1.94, + "learning_rate": 9.505907626208377e-06, + "loss": 0.0269, + "step": 18030 + }, + { + "epoch": 1.94, + "learning_rate": 9.344790547798066e-06, + "loss": 0.029, + "step": 18040 + }, + { + "epoch": 1.94, + "learning_rate": 9.183673469387754e-06, + "loss": 0.0307, + "step": 18050 + }, + { + "epoch": 1.94, + "learning_rate": 9.022556390977442e-06, + "loss": 0.032, + "step": 18060 + }, + { + "epoch": 1.94, + "learning_rate": 8.86143931256713e-06, + "loss": 0.0255, + "step": 18070 + }, + { + "epoch": 1.94, + "learning_rate": 8.70032223415682e-06, + "loss": 0.0265, + "step": 18080 + }, + { + "epoch": 1.94, + "learning_rate": 8.539205155746508e-06, + "loss": 0.0286, + "step": 18090 + }, + { + "epoch": 1.94, + "learning_rate": 8.378088077336198e-06, + "loss": 0.0342, + "step": 18100 + }, + { + "epoch": 1.95, + "learning_rate": 8.216970998925886e-06, + "loss": 0.0289, + "step": 18110 + }, + { + "epoch": 1.95, + "learning_rate": 8.055853920515575e-06, + "loss": 0.0375, + "step": 18120 + }, + { + "epoch": 1.95, + "learning_rate": 7.894736842105261e-06, + "loss": 0.0245, + "step": 18130 + }, + { + "epoch": 1.95, + "learning_rate": 7.733619763694951e-06, + "loss": 0.0273, + "step": 18140 + }, + { + "epoch": 1.95, + "learning_rate": 7.57250268528464e-06, + "loss": 0.0357, + "step": 18150 + }, + { + "epoch": 1.95, + "learning_rate": 7.411385606874328e-06, + "loss": 0.0328, + "step": 18160 + }, + { + "epoch": 1.95, + "learning_rate": 7.250268528464017e-06, + "loss": 0.0313, + "step": 18170 + }, + { + "epoch": 1.95, + "learning_rate": 7.089151450053705e-06, + "loss": 0.026, + "step": 18180 + }, + { + "epoch": 1.95, + "learning_rate": 6.928034371643393e-06, + "loss": 0.0335, + "step": 18190 + }, + { + "epoch": 1.95, + "learning_rate": 6.766917293233082e-06, + "loss": 0.025, + "step": 18200 + }, + { + "epoch": 1.96, + "learning_rate": 6.605800214822771e-06, + "loss": 0.0315, + "step": 18210 + }, + { + "epoch": 1.96, + "learning_rate": 6.4446831364124594e-06, + "loss": 0.0305, + "step": 18220 + }, + { + "epoch": 1.96, + "learning_rate": 6.283566058002148e-06, + "loss": 0.0294, + "step": 18230 + }, + { + "epoch": 1.96, + "learning_rate": 6.122448979591835e-06, + "loss": 0.0273, + "step": 18240 + }, + { + "epoch": 1.96, + "learning_rate": 5.961331901181525e-06, + "loss": 0.0325, + "step": 18250 + }, + { + "epoch": 1.96, + "learning_rate": 5.800214822771213e-06, + "loss": 0.0355, + "step": 18260 + }, + { + "epoch": 1.96, + "learning_rate": 5.6390977443609015e-06, + "loss": 0.029, + "step": 18270 + }, + { + "epoch": 1.96, + "learning_rate": 5.477980665950591e-06, + "loss": 0.0302, + "step": 18280 + }, + { + "epoch": 1.96, + "learning_rate": 5.316863587540279e-06, + "loss": 0.0324, + "step": 18290 + }, + { + "epoch": 1.97, + "learning_rate": 5.1557465091299675e-06, + "loss": 0.0344, + "step": 18300 + }, + { + "epoch": 1.97, + "learning_rate": 4.994629430719655e-06, + "loss": 0.0307, + "step": 18310 + }, + { + "epoch": 1.97, + "learning_rate": 4.833512352309344e-06, + "loss": 0.0268, + "step": 18320 + }, + { + "epoch": 1.97, + "learning_rate": 4.672395273899033e-06, + "loss": 0.0303, + "step": 18330 + }, + { + "epoch": 1.97, + "learning_rate": 4.511278195488721e-06, + "loss": 0.0308, + "step": 18340 + }, + { + "epoch": 1.97, + "learning_rate": 4.35016111707841e-06, + "loss": 0.031, + "step": 18350 + }, + { + "epoch": 1.97, + "learning_rate": 4.189044038668099e-06, + "loss": 0.032, + "step": 18360 + }, + { + "epoch": 1.97, + "learning_rate": 4.027926960257787e-06, + "loss": 0.0326, + "step": 18370 + }, + { + "epoch": 1.97, + "learning_rate": 3.866809881847476e-06, + "loss": 0.0317, + "step": 18380 + }, + { + "epoch": 1.98, + "learning_rate": 3.705692803437164e-06, + "loss": 0.0343, + "step": 18390 + }, + { + "epoch": 1.98, + "learning_rate": 3.5445757250268525e-06, + "loss": 0.0291, + "step": 18400 + }, + { + "epoch": 1.98, + "learning_rate": 3.383458646616541e-06, + "loss": 0.0268, + "step": 18410 + }, + { + "epoch": 1.98, + "learning_rate": 3.2223415682062297e-06, + "loss": 0.0312, + "step": 18420 + }, + { + "epoch": 1.98, + "learning_rate": 3.0612244897959177e-06, + "loss": 0.0429, + "step": 18430 + }, + { + "epoch": 1.98, + "learning_rate": 2.9001074113856065e-06, + "loss": 0.0418, + "step": 18440 + }, + { + "epoch": 1.98, + "learning_rate": 2.7389903329752954e-06, + "loss": 0.0355, + "step": 18450 + }, + { + "epoch": 1.98, + "learning_rate": 2.5778732545649838e-06, + "loss": 0.0329, + "step": 18460 + }, + { + "epoch": 1.98, + "learning_rate": 2.416756176154672e-06, + "loss": 0.0304, + "step": 18470 + }, + { + "epoch": 1.98, + "learning_rate": 2.2556390977443606e-06, + "loss": 0.0281, + "step": 18480 + }, + { + "epoch": 1.99, + "learning_rate": 2.0945220193340494e-06, + "loss": 0.035, + "step": 18490 + }, + { + "epoch": 1.99, + "learning_rate": 1.933404940923738e-06, + "loss": 0.0377, + "step": 18500 + }, + { + "epoch": 1.99, + "learning_rate": 1.7722878625134262e-06, + "loss": 0.0297, + "step": 18510 + }, + { + "epoch": 1.99, + "learning_rate": 1.6111707841031149e-06, + "loss": 0.0373, + "step": 18520 + }, + { + "epoch": 1.99, + "learning_rate": 1.4500537056928033e-06, + "loss": 0.0359, + "step": 18530 + }, + { + "epoch": 1.99, + "learning_rate": 1.2889366272824919e-06, + "loss": 0.0273, + "step": 18540 + }, + { + "epoch": 1.99, + "learning_rate": 1.1278195488721803e-06, + "loss": 0.0313, + "step": 18550 + }, + { + "epoch": 1.99, + "learning_rate": 9.66702470461869e-07, + "loss": 0.0325, + "step": 18560 + }, + { + "epoch": 1.99, + "learning_rate": 8.055853920515574e-07, + "loss": 0.0251, + "step": 18570 + }, + { + "epoch": 2.0, + "learning_rate": 6.444683136412459e-07, + "loss": 0.0324, + "step": 18580 + }, + { + "epoch": 2.0, + "learning_rate": 4.833512352309345e-07, + "loss": 0.032, + "step": 18590 + }, + { + "epoch": 2.0, + "learning_rate": 3.2223415682062297e-07, + "loss": 0.0286, + "step": 18600 + }, + { + "epoch": 2.0, + "learning_rate": 1.6111707841031149e-07, + "loss": 0.0333, + "step": 18610 + }, + { + "epoch": 2.0, + "learning_rate": 0.0, + "loss": 0.0284, + "step": 18620 + }, + { + "epoch": 2.0, + "step": 18620, + "total_flos": 1.0037352244727808e+17, + "train_loss": 0.07157236041639212, + "train_runtime": 20459.2275, + "train_samples_per_second": 29.126, + "train_steps_per_second": 0.91 + } + ], + "max_steps": 18620, + "num_train_epochs": 2, + "total_flos": 1.0037352244727808e+17, + "trial_name": null, + "trial_params": null +}