{ "best_metric": 29.205723913714138, "best_model_checkpoint": "./hamsa-pretrained/checkpoint-35000", "epoch": 0.17544536787336215, "eval_steps": 1000, "global_step": 35001, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 1.0500000000000001e-05, "loss": 4.6001, "step": 25 }, { "epoch": 0.0, "learning_rate": 2.3e-05, "loss": 2.9857, "step": 50 }, { "epoch": 0.0, "learning_rate": 3.5499999999999996e-05, "loss": 2.8432, "step": 75 }, { "epoch": 0.0, "learning_rate": 4.7000000000000004e-05, "loss": 3.0861, "step": 100 }, { "epoch": 0.0, "learning_rate": 5.9499999999999996e-05, "loss": 1.3867, "step": 125 }, { "epoch": 0.0, "learning_rate": 7.2e-05, "loss": 1.307, "step": 150 }, { "epoch": 0.0, "learning_rate": 8.450000000000001e-05, "loss": 1.3456, "step": 175 }, { "epoch": 0.0, "learning_rate": 9.7e-05, "loss": 1.3625, "step": 200 }, { "epoch": 0.0, "learning_rate": 0.0001095, "loss": 1.4828, "step": 225 }, { "epoch": 0.0, "learning_rate": 0.000122, "loss": 1.653, "step": 250 }, { "epoch": 0.0, "learning_rate": 0.00013450000000000002, "loss": 1.527, "step": 275 }, { "epoch": 0.0, "learning_rate": 0.000147, "loss": 1.6017, "step": 300 }, { "epoch": 0.0, "learning_rate": 0.0001595, "loss": 1.7962, "step": 325 }, { "epoch": 0.0, "learning_rate": 0.00017199999999999998, "loss": 1.7863, "step": 350 }, { "epoch": 0.0, "learning_rate": 0.0001845, "loss": 1.751, "step": 375 }, { "epoch": 0.0, "learning_rate": 0.00019700000000000002, "loss": 1.767, "step": 400 }, { "epoch": 0.0, "learning_rate": 0.0002095, "loss": 1.8878, "step": 425 }, { "epoch": 0.0, "learning_rate": 0.000222, "loss": 2.1027, "step": 450 }, { "epoch": 0.0, "learning_rate": 0.00023449999999999998, "loss": 2.6898, "step": 475 }, { "epoch": 0.0, "learning_rate": 0.000247, "loss": 2.2361, "step": 500 }, { "epoch": 0.0, "learning_rate": 0.0002498623188405797, "loss": 2.1121, "step": 525 }, { "epoch": 0.0, "learning_rate": 0.0002496811594202899, "loss": 1.9875, "step": 550 }, { "epoch": 0.0, "learning_rate": 0.0002495, "loss": 2.0148, "step": 575 }, { "epoch": 0.0, "learning_rate": 0.00024931884057971017, "loss": 1.8997, "step": 600 }, { "epoch": 0.0, "learning_rate": 0.0002491376811594203, "loss": 1.9364, "step": 625 }, { "epoch": 0.0, "learning_rate": 0.00024895652173913046, "loss": 1.9569, "step": 650 }, { "epoch": 0.0, "learning_rate": 0.0002487753623188406, "loss": 1.9624, "step": 675 }, { "epoch": 0.0, "learning_rate": 0.0002485942028985507, "loss": 1.9502, "step": 700 }, { "epoch": 0.0, "learning_rate": 0.0002484130434782609, "loss": 1.9584, "step": 725 }, { "epoch": 0.0, "learning_rate": 0.000248231884057971, "loss": 2.0121, "step": 750 }, { "epoch": 0.0, "learning_rate": 0.00024805072463768117, "loss": 1.8849, "step": 775 }, { "epoch": 0.0, "learning_rate": 0.00024786956521739134, "loss": 1.9201, "step": 800 }, { "epoch": 0.0, "learning_rate": 0.00024768840579710147, "loss": 1.9334, "step": 825 }, { "epoch": 0.0, "learning_rate": 0.0002475072463768116, "loss": 1.9636, "step": 850 }, { "epoch": 0.0, "learning_rate": 0.00024732608695652176, "loss": 1.8136, "step": 875 }, { "epoch": 0.0, "learning_rate": 0.0002471449275362319, "loss": 1.8495, "step": 900 }, { "epoch": 0.0, "learning_rate": 0.00024696376811594205, "loss": 1.8316, "step": 925 }, { "epoch": 0.0, "learning_rate": 0.0002467826086956522, "loss": 1.7647, "step": 950 }, { "epoch": 0.0, "learning_rate": 0.00024660144927536235, "loss": 1.7411, "step": 975 }, { "epoch": 0.01, "learning_rate": 0.00024642028985507247, "loss": 1.895, "step": 1000 }, { "epoch": 0.01, "eval_loss": 1.876536250114441, "eval_runtime": 5942.4913, "eval_samples_per_second": 0.842, "eval_steps_per_second": 0.211, "eval_wer": 86.70115569899617, "step": 1000 }, { "epoch": 0.01, "learning_rate": 0.0002462391304347826, "loss": 1.7266, "step": 1025 }, { "epoch": 0.01, "learning_rate": 0.00024605797101449276, "loss": 1.7269, "step": 1050 }, { "epoch": 0.01, "learning_rate": 0.00024587681159420294, "loss": 1.8598, "step": 1075 }, { "epoch": 0.01, "learning_rate": 0.00024569565217391306, "loss": 1.6687, "step": 1100 }, { "epoch": 0.01, "learning_rate": 0.0002455144927536232, "loss": 1.8505, "step": 1125 }, { "epoch": 0.01, "learning_rate": 0.00024533333333333335, "loss": 1.7166, "step": 1150 }, { "epoch": 0.01, "learning_rate": 0.00024515217391304347, "loss": 1.6826, "step": 1175 }, { "epoch": 0.01, "learning_rate": 0.00024497101449275365, "loss": 1.8421, "step": 1200 }, { "epoch": 0.01, "learning_rate": 0.00024478985507246377, "loss": 1.7725, "step": 1225 }, { "epoch": 0.01, "learning_rate": 0.00024460869565217394, "loss": 1.7464, "step": 1250 }, { "epoch": 0.01, "learning_rate": 0.00024442753623188406, "loss": 1.7435, "step": 1275 }, { "epoch": 0.01, "learning_rate": 0.0002442463768115942, "loss": 1.7296, "step": 1300 }, { "epoch": 0.01, "learning_rate": 0.00024406521739130435, "loss": 1.8299, "step": 1325 }, { "epoch": 0.01, "learning_rate": 0.0002438840579710145, "loss": 1.7485, "step": 1350 }, { "epoch": 0.01, "learning_rate": 0.00024370289855072465, "loss": 1.7177, "step": 1375 }, { "epoch": 0.01, "learning_rate": 0.0002435217391304348, "loss": 1.7324, "step": 1400 }, { "epoch": 0.01, "learning_rate": 0.00024334057971014492, "loss": 1.7653, "step": 1425 }, { "epoch": 0.01, "learning_rate": 0.00024315942028985506, "loss": 1.7578, "step": 1450 }, { "epoch": 0.01, "learning_rate": 0.00024297826086956524, "loss": 1.7259, "step": 1475 }, { "epoch": 0.01, "learning_rate": 0.00024279710144927536, "loss": 1.7141, "step": 1500 }, { "epoch": 0.01, "learning_rate": 0.0002426159420289855, "loss": 1.7828, "step": 1525 }, { "epoch": 0.01, "learning_rate": 0.00024243478260869568, "loss": 1.6833, "step": 1550 }, { "epoch": 0.01, "learning_rate": 0.0002422536231884058, "loss": 1.7581, "step": 1575 }, { "epoch": 0.01, "learning_rate": 0.00024207246376811595, "loss": 1.6707, "step": 1600 }, { "epoch": 0.01, "learning_rate": 0.00024189130434782607, "loss": 1.6099, "step": 1625 }, { "epoch": 0.01, "learning_rate": 0.00024171014492753624, "loss": 1.6632, "step": 1650 }, { "epoch": 0.01, "learning_rate": 0.0002415289855072464, "loss": 1.7326, "step": 1675 }, { "epoch": 0.01, "learning_rate": 0.0002413478260869565, "loss": 1.7251, "step": 1700 }, { "epoch": 0.01, "learning_rate": 0.00024116666666666668, "loss": 1.655, "step": 1725 }, { "epoch": 0.01, "learning_rate": 0.00024098550724637683, "loss": 1.7368, "step": 1750 }, { "epoch": 0.01, "learning_rate": 0.00024080434782608695, "loss": 1.7527, "step": 1775 }, { "epoch": 0.01, "learning_rate": 0.0002406231884057971, "loss": 1.6496, "step": 1800 }, { "epoch": 0.01, "learning_rate": 0.00024044202898550724, "loss": 1.7439, "step": 1825 }, { "epoch": 0.01, "learning_rate": 0.0002402608695652174, "loss": 1.6974, "step": 1850 }, { "epoch": 0.01, "learning_rate": 0.00024007971014492754, "loss": 1.5964, "step": 1875 }, { "epoch": 0.01, "learning_rate": 0.00023989855072463769, "loss": 1.5731, "step": 1900 }, { "epoch": 0.01, "learning_rate": 0.00023971739130434783, "loss": 1.5764, "step": 1925 }, { "epoch": 0.01, "learning_rate": 0.00023953623188405798, "loss": 1.7154, "step": 1950 }, { "epoch": 0.01, "learning_rate": 0.0002393550724637681, "loss": 1.7839, "step": 1975 }, { "epoch": 0.01, "learning_rate": 0.00023917391304347828, "loss": 1.6569, "step": 2000 }, { "epoch": 0.01, "eval_loss": 1.5809087753295898, "eval_runtime": 6451.2844, "eval_samples_per_second": 0.776, "eval_steps_per_second": 0.194, "eval_wer": 84.0907472887349, "step": 2000 }, { "epoch": 0.01, "learning_rate": 0.0002389927536231884, "loss": 1.6677, "step": 2025 }, { "epoch": 0.01, "learning_rate": 0.00023881159420289854, "loss": 1.6182, "step": 2050 }, { "epoch": 0.01, "learning_rate": 0.00023863043478260872, "loss": 1.6538, "step": 2075 }, { "epoch": 0.01, "learning_rate": 0.00023844927536231884, "loss": 1.6372, "step": 2100 }, { "epoch": 0.01, "learning_rate": 0.00023826811594202898, "loss": 1.5046, "step": 2125 }, { "epoch": 0.01, "learning_rate": 0.00023808695652173916, "loss": 1.5537, "step": 2150 }, { "epoch": 0.01, "learning_rate": 0.00023790579710144928, "loss": 1.6118, "step": 2175 }, { "epoch": 0.01, "learning_rate": 0.00023772463768115943, "loss": 1.5437, "step": 2200 }, { "epoch": 0.01, "learning_rate": 0.00023754347826086957, "loss": 1.6297, "step": 2225 }, { "epoch": 0.01, "learning_rate": 0.00023736231884057972, "loss": 1.5116, "step": 2250 }, { "epoch": 0.01, "learning_rate": 0.00023718115942028987, "loss": 1.5196, "step": 2275 }, { "epoch": 0.01, "learning_rate": 0.000237, "loss": 1.5372, "step": 2300 }, { "epoch": 0.01, "learning_rate": 0.00023681884057971016, "loss": 1.5344, "step": 2325 }, { "epoch": 0.01, "learning_rate": 0.0002366376811594203, "loss": 1.5114, "step": 2350 }, { "epoch": 0.01, "learning_rate": 0.00023645652173913043, "loss": 1.5755, "step": 2375 }, { "epoch": 0.01, "learning_rate": 0.00023627536231884058, "loss": 1.4815, "step": 2400 }, { "epoch": 0.01, "learning_rate": 0.00023609420289855075, "loss": 1.4123, "step": 2425 }, { "epoch": 0.01, "learning_rate": 0.00023591304347826087, "loss": 1.5506, "step": 2450 }, { "epoch": 0.01, "learning_rate": 0.00023573188405797102, "loss": 1.5217, "step": 2475 }, { "epoch": 0.01, "learning_rate": 0.00023555072463768116, "loss": 1.5644, "step": 2500 }, { "epoch": 0.01, "learning_rate": 0.0002353695652173913, "loss": 1.5734, "step": 2525 }, { "epoch": 0.01, "learning_rate": 0.00023518840579710146, "loss": 1.5853, "step": 2550 }, { "epoch": 0.01, "learning_rate": 0.00023500724637681158, "loss": 1.4974, "step": 2575 }, { "epoch": 0.01, "learning_rate": 0.00023482608695652175, "loss": 1.4737, "step": 2600 }, { "epoch": 0.01, "learning_rate": 0.0002346449275362319, "loss": 1.4675, "step": 2625 }, { "epoch": 0.01, "learning_rate": 0.00023446376811594202, "loss": 1.5251, "step": 2650 }, { "epoch": 0.01, "learning_rate": 0.0002342826086956522, "loss": 1.5386, "step": 2675 }, { "epoch": 0.01, "learning_rate": 0.00023410144927536232, "loss": 1.5002, "step": 2700 }, { "epoch": 0.01, "learning_rate": 0.00023392028985507246, "loss": 1.375, "step": 2725 }, { "epoch": 0.01, "learning_rate": 0.0002337391304347826, "loss": 1.519, "step": 2750 }, { "epoch": 0.01, "learning_rate": 0.00023355797101449276, "loss": 1.4296, "step": 2775 }, { "epoch": 0.01, "learning_rate": 0.0002333768115942029, "loss": 1.4506, "step": 2800 }, { "epoch": 0.01, "learning_rate": 0.00023319565217391305, "loss": 1.4481, "step": 2825 }, { "epoch": 0.01, "learning_rate": 0.0002330144927536232, "loss": 1.4097, "step": 2850 }, { "epoch": 0.01, "learning_rate": 0.00023283333333333335, "loss": 1.5138, "step": 2875 }, { "epoch": 0.01, "learning_rate": 0.0002326521739130435, "loss": 1.5278, "step": 2900 }, { "epoch": 0.01, "learning_rate": 0.0002324710144927536, "loss": 1.3968, "step": 2925 }, { "epoch": 0.01, "learning_rate": 0.0002322898550724638, "loss": 1.3648, "step": 2950 }, { "epoch": 0.01, "learning_rate": 0.0002321086956521739, "loss": 1.4226, "step": 2975 }, { "epoch": 0.02, "learning_rate": 0.00023192753623188405, "loss": 1.3312, "step": 3000 }, { "epoch": 0.02, "eval_loss": 1.3458073139190674, "eval_runtime": 6200.517, "eval_samples_per_second": 0.807, "eval_steps_per_second": 0.202, "eval_wer": 75.70896319324142, "step": 3000 }, { "epoch": 0.02, "learning_rate": 0.00023174637681159423, "loss": 1.3902, "step": 3025 }, { "epoch": 0.02, "learning_rate": 0.00023156521739130435, "loss": 1.4262, "step": 3050 }, { "epoch": 0.02, "learning_rate": 0.0002313840579710145, "loss": 1.4074, "step": 3075 }, { "epoch": 0.02, "learning_rate": 0.00023120289855072464, "loss": 1.4074, "step": 3100 }, { "epoch": 0.02, "learning_rate": 0.0002310217391304348, "loss": 1.3735, "step": 3125 }, { "epoch": 0.02, "learning_rate": 0.00023084057971014494, "loss": 1.4296, "step": 3150 }, { "epoch": 0.02, "learning_rate": 0.00023065942028985506, "loss": 1.3597, "step": 3175 }, { "epoch": 0.02, "learning_rate": 0.00023047826086956523, "loss": 1.4012, "step": 3200 }, { "epoch": 0.02, "learning_rate": 0.00023029710144927538, "loss": 1.3194, "step": 3225 }, { "epoch": 0.02, "learning_rate": 0.0002301159420289855, "loss": 1.3787, "step": 3250 }, { "epoch": 0.02, "learning_rate": 0.00022993478260869565, "loss": 1.334, "step": 3275 }, { "epoch": 0.02, "learning_rate": 0.00022975362318840582, "loss": 1.3462, "step": 3300 }, { "epoch": 0.02, "learning_rate": 0.00022957246376811594, "loss": 1.3108, "step": 3325 }, { "epoch": 0.02, "learning_rate": 0.0002293913043478261, "loss": 1.3052, "step": 3350 }, { "epoch": 0.02, "learning_rate": 0.00022921014492753624, "loss": 1.3761, "step": 3375 }, { "epoch": 0.02, "learning_rate": 0.00022902898550724638, "loss": 1.2893, "step": 3400 }, { "epoch": 0.02, "learning_rate": 0.00022884782608695653, "loss": 1.345, "step": 3425 }, { "epoch": 0.02, "learning_rate": 0.00022866666666666665, "loss": 1.2041, "step": 3450 }, { "epoch": 0.02, "learning_rate": 0.00022848550724637682, "loss": 1.3416, "step": 3475 }, { "epoch": 0.02, "learning_rate": 0.00022830434782608697, "loss": 1.2589, "step": 3500 }, { "epoch": 0.02, "learning_rate": 0.0002281231884057971, "loss": 1.4042, "step": 3525 }, { "epoch": 0.02, "learning_rate": 0.00022794202898550727, "loss": 1.3195, "step": 3550 }, { "epoch": 0.02, "learning_rate": 0.00022776086956521739, "loss": 1.2601, "step": 3575 }, { "epoch": 0.02, "learning_rate": 0.00022757971014492753, "loss": 1.3014, "step": 3600 }, { "epoch": 0.02, "learning_rate": 0.00022739855072463768, "loss": 1.2728, "step": 3625 }, { "epoch": 0.02, "learning_rate": 0.00022721739130434783, "loss": 1.2994, "step": 3650 }, { "epoch": 0.02, "learning_rate": 0.00022703623188405798, "loss": 1.2771, "step": 3675 }, { "epoch": 0.02, "learning_rate": 0.00022685507246376812, "loss": 1.2256, "step": 3700 }, { "epoch": 0.02, "learning_rate": 0.00022667391304347827, "loss": 1.4345, "step": 3725 }, { "epoch": 0.02, "learning_rate": 0.00022649275362318842, "loss": 1.293, "step": 3750 }, { "epoch": 0.02, "learning_rate": 0.00022631159420289856, "loss": 1.2223, "step": 3775 }, { "epoch": 0.02, "learning_rate": 0.00022613043478260868, "loss": 1.2575, "step": 3800 }, { "epoch": 0.02, "learning_rate": 0.00022594927536231886, "loss": 1.3047, "step": 3825 }, { "epoch": 0.02, "learning_rate": 0.00022576811594202898, "loss": 1.2402, "step": 3850 }, { "epoch": 0.02, "learning_rate": 0.00022558695652173913, "loss": 1.203, "step": 3875 }, { "epoch": 0.02, "learning_rate": 0.0002254057971014493, "loss": 1.1681, "step": 3900 }, { "epoch": 0.02, "learning_rate": 0.00022522463768115942, "loss": 1.1496, "step": 3925 }, { "epoch": 0.02, "learning_rate": 0.00022504347826086957, "loss": 1.258, "step": 3950 }, { "epoch": 0.02, "learning_rate": 0.00022486231884057971, "loss": 1.1696, "step": 3975 }, { "epoch": 0.02, "learning_rate": 0.00022468115942028986, "loss": 1.2369, "step": 4000 }, { "epoch": 0.02, "eval_loss": 1.2389202117919922, "eval_runtime": 6171.9451, "eval_samples_per_second": 0.811, "eval_steps_per_second": 0.203, "eval_wer": 73.13652435985667, "step": 4000 }, { "epoch": 0.02, "learning_rate": 0.0002245, "loss": 1.2727, "step": 4025 }, { "epoch": 0.02, "learning_rate": 0.00022431884057971013, "loss": 1.2076, "step": 4050 }, { "epoch": 0.02, "learning_rate": 0.0002241376811594203, "loss": 1.3128, "step": 4075 }, { "epoch": 0.02, "learning_rate": 0.00022395652173913045, "loss": 1.2745, "step": 4100 }, { "epoch": 0.02, "learning_rate": 0.00022377536231884057, "loss": 1.2999, "step": 4125 }, { "epoch": 0.02, "learning_rate": 0.00022359420289855072, "loss": 1.3311, "step": 4150 }, { "epoch": 0.02, "learning_rate": 0.0002234130434782609, "loss": 1.252, "step": 4175 }, { "epoch": 0.02, "learning_rate": 0.000223231884057971, "loss": 1.2794, "step": 4200 }, { "epoch": 0.02, "learning_rate": 0.00022305072463768116, "loss": 1.3587, "step": 4225 }, { "epoch": 0.02, "learning_rate": 0.0002228695652173913, "loss": 1.2422, "step": 4250 }, { "epoch": 0.02, "learning_rate": 0.00022268840579710145, "loss": 1.2734, "step": 4275 }, { "epoch": 0.02, "learning_rate": 0.0002225072463768116, "loss": 1.1935, "step": 4300 }, { "epoch": 0.02, "learning_rate": 0.00022232608695652172, "loss": 1.1646, "step": 4325 }, { "epoch": 0.02, "learning_rate": 0.0002221449275362319, "loss": 1.2976, "step": 4350 }, { "epoch": 0.02, "learning_rate": 0.00022196376811594204, "loss": 1.2087, "step": 4375 }, { "epoch": 0.02, "learning_rate": 0.00022178260869565216, "loss": 1.2437, "step": 4400 }, { "epoch": 0.02, "learning_rate": 0.00022160144927536234, "loss": 1.2036, "step": 4425 }, { "epoch": 0.02, "learning_rate": 0.00022142028985507248, "loss": 1.2467, "step": 4450 }, { "epoch": 0.02, "learning_rate": 0.0002212391304347826, "loss": 1.2779, "step": 4475 }, { "epoch": 0.02, "learning_rate": 0.00022105797101449275, "loss": 1.2326, "step": 4500 }, { "epoch": 0.02, "learning_rate": 0.0002208768115942029, "loss": 1.1391, "step": 4525 }, { "epoch": 0.02, "learning_rate": 0.00022069565217391305, "loss": 1.1322, "step": 4550 }, { "epoch": 0.02, "learning_rate": 0.0002205144927536232, "loss": 1.257, "step": 4575 }, { "epoch": 0.02, "learning_rate": 0.00022033333333333334, "loss": 1.1633, "step": 4600 }, { "epoch": 0.02, "learning_rate": 0.0002201521739130435, "loss": 1.174, "step": 4625 }, { "epoch": 0.02, "learning_rate": 0.00021997101449275363, "loss": 1.0957, "step": 4650 }, { "epoch": 0.02, "learning_rate": 0.00021978985507246375, "loss": 1.1833, "step": 4675 }, { "epoch": 0.02, "learning_rate": 0.00021960869565217393, "loss": 1.199, "step": 4700 }, { "epoch": 0.02, "learning_rate": 0.00021942753623188405, "loss": 1.2279, "step": 4725 }, { "epoch": 0.02, "learning_rate": 0.0002192463768115942, "loss": 1.1736, "step": 4750 }, { "epoch": 0.02, "learning_rate": 0.00021906521739130437, "loss": 1.1281, "step": 4775 }, { "epoch": 0.02, "learning_rate": 0.0002188840579710145, "loss": 1.1309, "step": 4800 }, { "epoch": 0.02, "learning_rate": 0.00021870289855072464, "loss": 1.2036, "step": 4825 }, { "epoch": 0.02, "learning_rate": 0.0002185217391304348, "loss": 1.1075, "step": 4850 }, { "epoch": 0.02, "learning_rate": 0.00021834057971014493, "loss": 1.099, "step": 4875 }, { "epoch": 0.02, "learning_rate": 0.00021815942028985508, "loss": 1.1406, "step": 4900 }, { "epoch": 0.02, "learning_rate": 0.0002179782608695652, "loss": 1.1683, "step": 4925 }, { "epoch": 0.02, "learning_rate": 0.00021779710144927537, "loss": 1.1565, "step": 4950 }, { "epoch": 0.02, "learning_rate": 0.00021761594202898552, "loss": 1.1336, "step": 4975 }, { "epoch": 0.03, "learning_rate": 0.00021743478260869564, "loss": 1.1518, "step": 5000 }, { "epoch": 0.03, "eval_loss": 1.1097463369369507, "eval_runtime": 6186.9062, "eval_samples_per_second": 0.809, "eval_steps_per_second": 0.202, "eval_wer": 66.81696290846959, "step": 5000 }, { "epoch": 0.03, "learning_rate": 0.00021725362318840582, "loss": 1.0662, "step": 5025 }, { "epoch": 0.03, "learning_rate": 0.00021707246376811596, "loss": 1.0916, "step": 5050 }, { "epoch": 0.03, "learning_rate": 0.00021689130434782608, "loss": 1.1352, "step": 5075 }, { "epoch": 0.03, "learning_rate": 0.00021671014492753623, "loss": 1.2055, "step": 5100 }, { "epoch": 0.03, "learning_rate": 0.00021652898550724638, "loss": 1.1505, "step": 5125 }, { "epoch": 0.03, "learning_rate": 0.00021634782608695652, "loss": 1.1587, "step": 5150 }, { "epoch": 0.03, "learning_rate": 0.00021616666666666667, "loss": 1.2408, "step": 5175 }, { "epoch": 0.03, "learning_rate": 0.00021598550724637682, "loss": 1.134, "step": 5200 }, { "epoch": 0.03, "learning_rate": 0.00021580434782608697, "loss": 1.0902, "step": 5225 }, { "epoch": 0.03, "learning_rate": 0.0002156231884057971, "loss": 1.1215, "step": 5250 }, { "epoch": 0.03, "learning_rate": 0.00021544202898550723, "loss": 1.0944, "step": 5275 }, { "epoch": 0.03, "learning_rate": 0.0002152608695652174, "loss": 1.0846, "step": 5300 }, { "epoch": 0.03, "learning_rate": 0.00021507971014492755, "loss": 1.1319, "step": 5325 }, { "epoch": 0.03, "learning_rate": 0.00021489855072463767, "loss": 1.0893, "step": 5350 }, { "epoch": 0.03, "learning_rate": 0.00021471739130434785, "loss": 1.1424, "step": 5375 }, { "epoch": 0.03, "learning_rate": 0.00021453623188405797, "loss": 1.0745, "step": 5400 }, { "epoch": 0.03, "learning_rate": 0.00021435507246376812, "loss": 1.1504, "step": 5425 }, { "epoch": 0.03, "learning_rate": 0.00021417391304347826, "loss": 1.0584, "step": 5450 }, { "epoch": 0.03, "learning_rate": 0.0002139927536231884, "loss": 1.1529, "step": 5475 }, { "epoch": 0.03, "learning_rate": 0.00021381159420289856, "loss": 1.1315, "step": 5500 }, { "epoch": 0.03, "learning_rate": 0.0002136304347826087, "loss": 1.1588, "step": 5525 }, { "epoch": 0.03, "learning_rate": 0.00021344927536231885, "loss": 1.0836, "step": 5550 }, { "epoch": 0.03, "learning_rate": 0.000213268115942029, "loss": 1.1677, "step": 5575 }, { "epoch": 0.03, "learning_rate": 0.00021308695652173912, "loss": 1.0709, "step": 5600 }, { "epoch": 0.03, "learning_rate": 0.00021290579710144927, "loss": 1.1992, "step": 5625 }, { "epoch": 0.03, "learning_rate": 0.00021272463768115944, "loss": 1.129, "step": 5650 }, { "epoch": 0.03, "learning_rate": 0.00021254347826086956, "loss": 1.0316, "step": 5675 }, { "epoch": 0.03, "learning_rate": 0.0002123623188405797, "loss": 1.1119, "step": 5700 }, { "epoch": 0.03, "learning_rate": 0.00021218115942028988, "loss": 1.0713, "step": 5725 }, { "epoch": 0.03, "learning_rate": 0.000212, "loss": 1.0471, "step": 5750 }, { "epoch": 0.03, "learning_rate": 0.00021181884057971015, "loss": 1.03, "step": 5775 }, { "epoch": 0.03, "learning_rate": 0.00021163768115942027, "loss": 0.9324, "step": 5800 }, { "epoch": 0.03, "learning_rate": 0.00021145652173913044, "loss": 1.1022, "step": 5825 }, { "epoch": 0.03, "learning_rate": 0.0002112753623188406, "loss": 1.0703, "step": 5850 }, { "epoch": 0.03, "learning_rate": 0.0002110942028985507, "loss": 0.9873, "step": 5875 }, { "epoch": 0.03, "learning_rate": 0.00021091304347826089, "loss": 1.1181, "step": 5900 }, { "epoch": 0.03, "learning_rate": 0.00021073188405797103, "loss": 1.0425, "step": 5925 }, { "epoch": 0.03, "learning_rate": 0.00021055072463768115, "loss": 1.1457, "step": 5950 }, { "epoch": 0.03, "learning_rate": 0.0002103695652173913, "loss": 1.1561, "step": 5975 }, { "epoch": 0.03, "learning_rate": 0.00021018840579710148, "loss": 1.0135, "step": 6000 }, { "epoch": 0.03, "eval_loss": 1.0616226196289062, "eval_runtime": 6268.4143, "eval_samples_per_second": 0.798, "eval_steps_per_second": 0.2, "eval_wer": 65.18427110277891, "step": 6000 }, { "epoch": 0.03, "learning_rate": 0.0002100072463768116, "loss": 0.9881, "step": 6025 }, { "epoch": 0.03, "learning_rate": 0.00020982608695652174, "loss": 1.0689, "step": 6050 }, { "epoch": 0.03, "learning_rate": 0.0002096449275362319, "loss": 1.0569, "step": 6075 }, { "epoch": 0.03, "learning_rate": 0.00020946376811594204, "loss": 1.1212, "step": 6100 }, { "epoch": 0.03, "learning_rate": 0.00020928260869565218, "loss": 1.0133, "step": 6125 }, { "epoch": 0.03, "learning_rate": 0.0002091014492753623, "loss": 1.0957, "step": 6150 }, { "epoch": 0.03, "learning_rate": 0.00020892028985507248, "loss": 1.0328, "step": 6175 }, { "epoch": 0.03, "learning_rate": 0.00020873913043478263, "loss": 1.0581, "step": 6200 }, { "epoch": 0.03, "learning_rate": 0.00020855797101449275, "loss": 0.9603, "step": 6225 }, { "epoch": 0.03, "learning_rate": 0.00020837681159420292, "loss": 1.0873, "step": 6250 }, { "epoch": 0.03, "learning_rate": 0.00020819565217391304, "loss": 0.9497, "step": 6275 }, { "epoch": 0.03, "learning_rate": 0.0002080144927536232, "loss": 0.997, "step": 6300 }, { "epoch": 0.03, "learning_rate": 0.00020783333333333333, "loss": 1.0641, "step": 6325 }, { "epoch": 0.03, "learning_rate": 0.00020765217391304348, "loss": 1.0911, "step": 6350 }, { "epoch": 0.03, "learning_rate": 0.00020747101449275363, "loss": 1.1041, "step": 6375 }, { "epoch": 0.03, "learning_rate": 0.00020728985507246378, "loss": 1.0997, "step": 6400 }, { "epoch": 0.03, "learning_rate": 0.00020710869565217392, "loss": 1.0995, "step": 6425 }, { "epoch": 0.03, "learning_rate": 0.00020692753623188407, "loss": 1.0015, "step": 6450 }, { "epoch": 0.03, "learning_rate": 0.0002067463768115942, "loss": 1.0448, "step": 6475 }, { "epoch": 0.03, "learning_rate": 0.00020656521739130434, "loss": 1.0744, "step": 6500 }, { "epoch": 0.03, "learning_rate": 0.0002063840579710145, "loss": 0.9513, "step": 6525 }, { "epoch": 0.03, "learning_rate": 0.00020620289855072463, "loss": 1.0539, "step": 6550 }, { "epoch": 0.03, "learning_rate": 0.00020602173913043478, "loss": 1.058, "step": 6575 }, { "epoch": 0.03, "learning_rate": 0.00020584057971014495, "loss": 0.9622, "step": 6600 }, { "epoch": 0.03, "learning_rate": 0.00020565942028985507, "loss": 1.0784, "step": 6625 }, { "epoch": 0.03, "learning_rate": 0.00020547826086956522, "loss": 1.0293, "step": 6650 }, { "epoch": 0.03, "learning_rate": 0.00020529710144927534, "loss": 1.1025, "step": 6675 }, { "epoch": 0.03, "learning_rate": 0.00020511594202898552, "loss": 1.0524, "step": 6700 }, { "epoch": 0.03, "learning_rate": 0.00020493478260869566, "loss": 1.0089, "step": 6725 }, { "epoch": 0.03, "learning_rate": 0.00020475362318840578, "loss": 1.0969, "step": 6750 }, { "epoch": 0.03, "learning_rate": 0.00020457246376811596, "loss": 0.949, "step": 6775 }, { "epoch": 0.03, "learning_rate": 0.0002043913043478261, "loss": 1.0159, "step": 6800 }, { "epoch": 0.03, "learning_rate": 0.00020421014492753622, "loss": 0.9893, "step": 6825 }, { "epoch": 0.03, "learning_rate": 0.00020402898550724637, "loss": 1.0183, "step": 6850 }, { "epoch": 0.03, "learning_rate": 0.0002038550724637681, "loss": 1.1104, "step": 6875 }, { "epoch": 0.03, "learning_rate": 0.00020367391304347825, "loss": 0.9909, "step": 6900 }, { "epoch": 0.03, "learning_rate": 0.00020349275362318843, "loss": 0.9662, "step": 6925 }, { "epoch": 0.03, "learning_rate": 0.00020331159420289855, "loss": 0.9838, "step": 6950 }, { "epoch": 0.03, "learning_rate": 0.0002031304347826087, "loss": 1.018, "step": 6975 }, { "epoch": 0.04, "learning_rate": 0.00020294927536231887, "loss": 1.0965, "step": 7000 }, { "epoch": 0.04, "eval_loss": 1.008375883102417, "eval_runtime": 6473.3745, "eval_samples_per_second": 0.773, "eval_steps_per_second": 0.193, "eval_wer": 65.85823109233726, "step": 7000 }, { "epoch": 0.04, "learning_rate": 0.000202768115942029, "loss": 1.0087, "step": 7025 }, { "epoch": 0.04, "learning_rate": 0.00020258695652173914, "loss": 1.0846, "step": 7050 }, { "epoch": 0.04, "learning_rate": 0.00020240579710144926, "loss": 1.0778, "step": 7075 }, { "epoch": 0.04, "learning_rate": 0.00020222463768115943, "loss": 0.9426, "step": 7100 }, { "epoch": 0.04, "learning_rate": 0.00020204347826086958, "loss": 0.9758, "step": 7125 }, { "epoch": 0.04, "learning_rate": 0.0002018623188405797, "loss": 1.0578, "step": 7150 }, { "epoch": 0.04, "learning_rate": 0.00020168115942028987, "loss": 0.9529, "step": 7175 }, { "epoch": 0.04, "learning_rate": 0.00020150000000000002, "loss": 1.0048, "step": 7200 }, { "epoch": 0.04, "learning_rate": 0.00020131884057971014, "loss": 0.9114, "step": 7225 }, { "epoch": 0.04, "learning_rate": 0.00020113768115942029, "loss": 1.0504, "step": 7250 }, { "epoch": 0.04, "learning_rate": 0.00020095652173913043, "loss": 0.9835, "step": 7275 }, { "epoch": 0.04, "learning_rate": 0.00020077536231884058, "loss": 0.9254, "step": 7300 }, { "epoch": 0.04, "learning_rate": 0.00020059420289855073, "loss": 0.9491, "step": 7325 }, { "epoch": 0.04, "learning_rate": 0.00020041304347826088, "loss": 0.9581, "step": 7350 }, { "epoch": 0.04, "learning_rate": 0.00020023188405797102, "loss": 0.9441, "step": 7375 }, { "epoch": 0.04, "learning_rate": 0.00020005072463768117, "loss": 0.9157, "step": 7400 }, { "epoch": 0.04, "learning_rate": 0.0001998695652173913, "loss": 0.8822, "step": 7425 }, { "epoch": 0.04, "learning_rate": 0.00019968840579710146, "loss": 1.0002, "step": 7450 }, { "epoch": 0.04, "learning_rate": 0.00019950724637681158, "loss": 0.96, "step": 7475 }, { "epoch": 0.04, "learning_rate": 0.00019932608695652173, "loss": 0.9303, "step": 7500 }, { "epoch": 0.04, "learning_rate": 0.0001991449275362319, "loss": 0.9066, "step": 7525 }, { "epoch": 0.04, "learning_rate": 0.00019896376811594203, "loss": 0.9793, "step": 7550 }, { "epoch": 0.04, "learning_rate": 0.00019878260869565217, "loss": 1.0105, "step": 7575 }, { "epoch": 0.04, "learning_rate": 0.00019860144927536235, "loss": 0.9897, "step": 7600 }, { "epoch": 0.04, "learning_rate": 0.00019842028985507247, "loss": 0.9949, "step": 7625 }, { "epoch": 0.04, "learning_rate": 0.00019823913043478261, "loss": 0.9493, "step": 7650 }, { "epoch": 0.04, "learning_rate": 0.00019805797101449276, "loss": 0.8889, "step": 7675 }, { "epoch": 0.04, "learning_rate": 0.0001978768115942029, "loss": 0.9088, "step": 7700 }, { "epoch": 0.04, "learning_rate": 0.00019769565217391306, "loss": 0.9772, "step": 7725 }, { "epoch": 0.04, "learning_rate": 0.00019751449275362318, "loss": 0.9503, "step": 7750 }, { "epoch": 0.04, "learning_rate": 0.00019733333333333335, "loss": 0.8893, "step": 7775 }, { "epoch": 0.04, "learning_rate": 0.0001971521739130435, "loss": 1.0032, "step": 7800 }, { "epoch": 0.04, "learning_rate": 0.00019697101449275362, "loss": 0.9792, "step": 7825 }, { "epoch": 0.04, "learning_rate": 0.00019678985507246376, "loss": 0.965, "step": 7850 }, { "epoch": 0.04, "learning_rate": 0.00019660869565217394, "loss": 0.9428, "step": 7875 }, { "epoch": 0.04, "learning_rate": 0.00019642753623188406, "loss": 1.0196, "step": 7900 }, { "epoch": 0.04, "learning_rate": 0.0001962463768115942, "loss": 0.9396, "step": 7925 }, { "epoch": 0.04, "learning_rate": 0.00019606521739130435, "loss": 0.9787, "step": 7950 }, { "epoch": 0.04, "learning_rate": 0.0001958840579710145, "loss": 0.9023, "step": 7975 }, { "epoch": 0.04, "learning_rate": 0.00019570289855072465, "loss": 0.867, "step": 8000 }, { "epoch": 0.04, "eval_loss": 0.9304548501968384, "eval_runtime": 6180.6024, "eval_samples_per_second": 0.81, "eval_steps_per_second": 0.203, "eval_wer": 57.609340515911626, "step": 8000 }, { "epoch": 0.04, "learning_rate": 0.00019552173913043477, "loss": 0.9631, "step": 8025 }, { "epoch": 0.04, "learning_rate": 0.00019534057971014494, "loss": 0.9408, "step": 8050 }, { "epoch": 0.04, "learning_rate": 0.0001951594202898551, "loss": 1.0169, "step": 8075 }, { "epoch": 0.04, "learning_rate": 0.0001949782608695652, "loss": 1.0044, "step": 8100 }, { "epoch": 0.04, "learning_rate": 0.00019479710144927538, "loss": 1.0462, "step": 8125 }, { "epoch": 0.04, "learning_rate": 0.0001946159420289855, "loss": 0.9778, "step": 8150 }, { "epoch": 0.04, "learning_rate": 0.00019443478260869565, "loss": 0.9329, "step": 8175 }, { "epoch": 0.04, "learning_rate": 0.0001942536231884058, "loss": 0.9757, "step": 8200 }, { "epoch": 0.04, "learning_rate": 0.00019407246376811595, "loss": 0.9153, "step": 8225 }, { "epoch": 0.04, "learning_rate": 0.0001938913043478261, "loss": 0.9201, "step": 8250 }, { "epoch": 0.04, "learning_rate": 0.00019371014492753624, "loss": 0.8937, "step": 8275 }, { "epoch": 0.04, "learning_rate": 0.0001935289855072464, "loss": 0.91, "step": 8300 }, { "epoch": 0.04, "learning_rate": 0.00019334782608695653, "loss": 0.884, "step": 8325 }, { "epoch": 0.04, "learning_rate": 0.00019316666666666665, "loss": 0.9819, "step": 8350 }, { "epoch": 0.04, "learning_rate": 0.0001929855072463768, "loss": 0.9816, "step": 8375 }, { "epoch": 0.04, "learning_rate": 0.00019280434782608698, "loss": 0.9698, "step": 8400 }, { "epoch": 0.04, "learning_rate": 0.0001926231884057971, "loss": 0.9072, "step": 8425 }, { "epoch": 0.04, "learning_rate": 0.00019244202898550724, "loss": 1.0583, "step": 8450 }, { "epoch": 0.04, "learning_rate": 0.00019226086956521742, "loss": 0.8779, "step": 8475 }, { "epoch": 0.04, "learning_rate": 0.00019207971014492754, "loss": 0.8537, "step": 8500 }, { "epoch": 0.04, "learning_rate": 0.00019189855072463769, "loss": 0.9183, "step": 8525 }, { "epoch": 0.04, "learning_rate": 0.00019171739130434783, "loss": 0.829, "step": 8550 }, { "epoch": 0.04, "learning_rate": 0.00019153623188405798, "loss": 0.9313, "step": 8575 }, { "epoch": 0.04, "learning_rate": 0.00019135507246376813, "loss": 0.8861, "step": 8600 }, { "epoch": 0.04, "learning_rate": 0.00019117391304347825, "loss": 0.9286, "step": 8625 }, { "epoch": 0.04, "learning_rate": 0.00019099275362318842, "loss": 0.9069, "step": 8650 }, { "epoch": 0.04, "learning_rate": 0.00019081159420289857, "loss": 0.9949, "step": 8675 }, { "epoch": 0.04, "learning_rate": 0.0001906304347826087, "loss": 0.9647, "step": 8700 }, { "epoch": 0.04, "learning_rate": 0.00019044927536231884, "loss": 0.9581, "step": 8725 }, { "epoch": 0.04, "learning_rate": 0.000190268115942029, "loss": 0.9482, "step": 8750 }, { "epoch": 0.04, "learning_rate": 0.00019008695652173913, "loss": 0.8733, "step": 8775 }, { "epoch": 0.04, "learning_rate": 0.00018990579710144928, "loss": 0.8457, "step": 8800 }, { "epoch": 0.04, "learning_rate": 0.00018972463768115942, "loss": 0.8804, "step": 8825 }, { "epoch": 0.04, "learning_rate": 0.00018954347826086957, "loss": 0.9482, "step": 8850 }, { "epoch": 0.04, "learning_rate": 0.00018936231884057972, "loss": 0.922, "step": 8875 }, { "epoch": 0.04, "learning_rate": 0.00018918115942028984, "loss": 0.8846, "step": 8900 }, { "epoch": 0.04, "learning_rate": 0.000189, "loss": 0.9022, "step": 8925 }, { "epoch": 0.04, "learning_rate": 0.00018881884057971016, "loss": 0.8839, "step": 8950 }, { "epoch": 0.04, "learning_rate": 0.00018863768115942028, "loss": 0.9456, "step": 8975 }, { "epoch": 0.05, "learning_rate": 0.00018845652173913045, "loss": 0.9425, "step": 9000 }, { "epoch": 0.05, "eval_loss": 0.8907226920127869, "eval_runtime": 6145.942, "eval_samples_per_second": 0.814, "eval_steps_per_second": 0.204, "eval_wer": 55.48541730938086, "step": 9000 }, { "epoch": 0.05, "learning_rate": 0.00018827536231884057, "loss": 0.8422, "step": 9025 }, { "epoch": 0.05, "learning_rate": 0.00018809420289855072, "loss": 0.8493, "step": 9050 }, { "epoch": 0.05, "learning_rate": 0.00018791304347826087, "loss": 0.844, "step": 9075 }, { "epoch": 0.05, "learning_rate": 0.00018773188405797102, "loss": 0.9008, "step": 9100 }, { "epoch": 0.05, "learning_rate": 0.00018755072463768116, "loss": 0.8489, "step": 9125 }, { "epoch": 0.05, "learning_rate": 0.0001873695652173913, "loss": 0.9133, "step": 9150 }, { "epoch": 0.05, "learning_rate": 0.00018718840579710146, "loss": 0.8765, "step": 9175 }, { "epoch": 0.05, "learning_rate": 0.0001870072463768116, "loss": 0.8892, "step": 9200 }, { "epoch": 0.05, "learning_rate": 0.00018682608695652173, "loss": 0.9426, "step": 9225 }, { "epoch": 0.05, "learning_rate": 0.00018664492753623187, "loss": 0.9292, "step": 9250 }, { "epoch": 0.05, "learning_rate": 0.00018646376811594205, "loss": 0.9203, "step": 9275 }, { "epoch": 0.05, "learning_rate": 0.00018628260869565217, "loss": 0.881, "step": 9300 }, { "epoch": 0.05, "learning_rate": 0.00018610144927536231, "loss": 0.9301, "step": 9325 }, { "epoch": 0.05, "learning_rate": 0.0001859202898550725, "loss": 0.8434, "step": 9350 }, { "epoch": 0.05, "learning_rate": 0.0001857391304347826, "loss": 0.9381, "step": 9375 }, { "epoch": 0.05, "learning_rate": 0.00018555797101449276, "loss": 0.891, "step": 9400 }, { "epoch": 0.05, "learning_rate": 0.0001853768115942029, "loss": 0.8266, "step": 9425 }, { "epoch": 0.05, "learning_rate": 0.00018519565217391305, "loss": 0.9595, "step": 9450 }, { "epoch": 0.05, "learning_rate": 0.0001850144927536232, "loss": 0.8791, "step": 9475 }, { "epoch": 0.05, "learning_rate": 0.00018483333333333332, "loss": 0.8108, "step": 9500 }, { "epoch": 0.05, "learning_rate": 0.0001846521739130435, "loss": 0.845, "step": 9525 }, { "epoch": 0.05, "learning_rate": 0.00018447101449275364, "loss": 0.8476, "step": 9550 }, { "epoch": 0.05, "learning_rate": 0.00018428985507246376, "loss": 0.8943, "step": 9575 }, { "epoch": 0.05, "learning_rate": 0.0001841086956521739, "loss": 0.8937, "step": 9600 }, { "epoch": 0.05, "learning_rate": 0.00018392753623188408, "loss": 0.8666, "step": 9625 }, { "epoch": 0.05, "learning_rate": 0.0001837463768115942, "loss": 0.8304, "step": 9650 }, { "epoch": 0.05, "learning_rate": 0.00018356521739130435, "loss": 0.8912, "step": 9675 }, { "epoch": 0.05, "learning_rate": 0.0001833840579710145, "loss": 0.9049, "step": 9700 }, { "epoch": 0.05, "learning_rate": 0.00018320289855072464, "loss": 0.8729, "step": 9725 }, { "epoch": 0.05, "learning_rate": 0.0001830217391304348, "loss": 0.8693, "step": 9750 }, { "epoch": 0.05, "learning_rate": 0.0001828405797101449, "loss": 0.8315, "step": 9775 }, { "epoch": 0.05, "learning_rate": 0.00018265942028985508, "loss": 0.8364, "step": 9800 }, { "epoch": 0.05, "learning_rate": 0.00018247826086956523, "loss": 0.8924, "step": 9825 }, { "epoch": 0.05, "learning_rate": 0.00018229710144927535, "loss": 0.8813, "step": 9850 }, { "epoch": 0.05, "learning_rate": 0.00018211594202898553, "loss": 0.8711, "step": 9875 }, { "epoch": 0.05, "learning_rate": 0.00018193478260869565, "loss": 0.8155, "step": 9900 }, { "epoch": 0.05, "learning_rate": 0.0001817536231884058, "loss": 0.8729, "step": 9925 }, { "epoch": 0.05, "learning_rate": 0.00018157246376811594, "loss": 0.9039, "step": 9950 }, { "epoch": 0.05, "learning_rate": 0.0001813913043478261, "loss": 0.7705, "step": 9975 }, { "epoch": 0.05, "learning_rate": 0.00018121014492753623, "loss": 0.9501, "step": 10000 }, { "epoch": 0.05, "eval_loss": 0.8392999172210693, "eval_runtime": 6230.858, "eval_samples_per_second": 0.803, "eval_steps_per_second": 0.201, "eval_wer": 54.02121550107976, "step": 10000 }, { "epoch": 0.05, "learning_rate": 0.00018102898550724638, "loss": 0.8114, "step": 10025 }, { "epoch": 0.05, "learning_rate": 0.00018084782608695653, "loss": 0.8974, "step": 10050 }, { "epoch": 0.05, "learning_rate": 0.00018066666666666668, "loss": 0.7821, "step": 10075 }, { "epoch": 0.05, "learning_rate": 0.00018048550724637682, "loss": 0.8489, "step": 10100 }, { "epoch": 0.05, "learning_rate": 0.00018030434782608694, "loss": 0.8337, "step": 10125 }, { "epoch": 0.05, "learning_rate": 0.00018012318840579712, "loss": 0.9795, "step": 10150 }, { "epoch": 0.05, "learning_rate": 0.00017994202898550724, "loss": 0.9552, "step": 10175 }, { "epoch": 0.05, "learning_rate": 0.00017976086956521739, "loss": 0.8544, "step": 10200 }, { "epoch": 0.05, "learning_rate": 0.00017957971014492756, "loss": 0.8426, "step": 10225 }, { "epoch": 0.05, "learning_rate": 0.00017939855072463768, "loss": 0.8073, "step": 10250 }, { "epoch": 0.05, "learning_rate": 0.00017921739130434783, "loss": 0.8112, "step": 10275 }, { "epoch": 0.05, "learning_rate": 0.00017903623188405797, "loss": 0.8495, "step": 10300 }, { "epoch": 0.05, "learning_rate": 0.00017885507246376812, "loss": 0.9735, "step": 10325 }, { "epoch": 0.05, "learning_rate": 0.00017867391304347827, "loss": 0.8394, "step": 10350 }, { "epoch": 0.05, "learning_rate": 0.0001784927536231884, "loss": 0.7613, "step": 10375 }, { "epoch": 0.05, "learning_rate": 0.00017831159420289856, "loss": 0.8697, "step": 10400 }, { "epoch": 0.05, "learning_rate": 0.0001781304347826087, "loss": 0.8087, "step": 10425 }, { "epoch": 0.05, "learning_rate": 0.00017794927536231883, "loss": 0.8902, "step": 10450 }, { "epoch": 0.05, "learning_rate": 0.00017776811594202898, "loss": 0.787, "step": 10475 }, { "epoch": 0.05, "learning_rate": 0.00017758695652173915, "loss": 0.8174, "step": 10500 }, { "epoch": 0.05, "learning_rate": 0.00017740579710144927, "loss": 0.7975, "step": 10525 }, { "epoch": 0.05, "learning_rate": 0.00017722463768115942, "loss": 0.7892, "step": 10550 }, { "epoch": 0.05, "learning_rate": 0.00017704347826086957, "loss": 0.7133, "step": 10575 }, { "epoch": 0.05, "learning_rate": 0.0001768623188405797, "loss": 0.8419, "step": 10600 }, { "epoch": 0.05, "learning_rate": 0.00017668115942028986, "loss": 0.7968, "step": 10625 }, { "epoch": 0.05, "learning_rate": 0.00017649999999999998, "loss": 0.8893, "step": 10650 }, { "epoch": 0.05, "learning_rate": 0.00017631884057971015, "loss": 0.8464, "step": 10675 }, { "epoch": 0.05, "learning_rate": 0.0001761376811594203, "loss": 0.8745, "step": 10700 }, { "epoch": 0.05, "learning_rate": 0.00017595652173913042, "loss": 0.853, "step": 10725 }, { "epoch": 0.05, "learning_rate": 0.0001757753623188406, "loss": 0.8488, "step": 10750 }, { "epoch": 0.05, "learning_rate": 0.00017559420289855072, "loss": 0.8435, "step": 10775 }, { "epoch": 0.05, "learning_rate": 0.00017541304347826086, "loss": 0.7949, "step": 10800 }, { "epoch": 0.05, "learning_rate": 0.00017523188405797104, "loss": 0.7916, "step": 10825 }, { "epoch": 0.05, "learning_rate": 0.00017505072463768116, "loss": 0.7734, "step": 10850 }, { "epoch": 0.05, "learning_rate": 0.0001748695652173913, "loss": 0.8258, "step": 10875 }, { "epoch": 0.05, "learning_rate": 0.00017468840579710145, "loss": 0.8101, "step": 10900 }, { "epoch": 0.05, "learning_rate": 0.0001745072463768116, "loss": 0.8221, "step": 10925 }, { "epoch": 0.05, "learning_rate": 0.00017432608695652175, "loss": 0.8009, "step": 10950 }, { "epoch": 0.06, "learning_rate": 0.0001741449275362319, "loss": 0.7875, "step": 10975 }, { "epoch": 0.06, "learning_rate": 0.00017396376811594204, "loss": 0.8602, "step": 11000 }, { "epoch": 0.06, "eval_loss": 0.8095631003379822, "eval_runtime": 6209.5684, "eval_samples_per_second": 0.806, "eval_steps_per_second": 0.202, "eval_wer": 53.49676072047272, "step": 11000 }, { "epoch": 0.06, "learning_rate": 0.0001737826086956522, "loss": 0.8403, "step": 11025 }, { "epoch": 0.06, "learning_rate": 0.0001736014492753623, "loss": 0.756, "step": 11050 }, { "epoch": 0.06, "learning_rate": 0.00017342028985507246, "loss": 0.8086, "step": 11075 }, { "epoch": 0.06, "learning_rate": 0.00017323913043478263, "loss": 0.7628, "step": 11100 }, { "epoch": 0.06, "learning_rate": 0.00017305797101449275, "loss": 0.8002, "step": 11125 }, { "epoch": 0.06, "learning_rate": 0.0001728768115942029, "loss": 0.8258, "step": 11150 }, { "epoch": 0.06, "learning_rate": 0.00017269565217391307, "loss": 0.7933, "step": 11175 }, { "epoch": 0.06, "learning_rate": 0.0001725144927536232, "loss": 0.7734, "step": 11200 }, { "epoch": 0.06, "learning_rate": 0.00017233333333333334, "loss": 0.8077, "step": 11225 }, { "epoch": 0.06, "learning_rate": 0.00017215217391304346, "loss": 0.8157, "step": 11250 }, { "epoch": 0.06, "learning_rate": 0.00017197101449275363, "loss": 0.822, "step": 11275 }, { "epoch": 0.06, "learning_rate": 0.00017178985507246378, "loss": 0.7343, "step": 11300 }, { "epoch": 0.06, "learning_rate": 0.0001716086956521739, "loss": 0.8093, "step": 11325 }, { "epoch": 0.06, "learning_rate": 0.00017142753623188408, "loss": 0.7928, "step": 11350 }, { "epoch": 0.06, "learning_rate": 0.00017124637681159422, "loss": 0.8318, "step": 11375 }, { "epoch": 0.06, "learning_rate": 0.00017106521739130434, "loss": 0.8163, "step": 11400 }, { "epoch": 0.06, "learning_rate": 0.0001708840579710145, "loss": 0.765, "step": 11425 }, { "epoch": 0.06, "learning_rate": 0.00017070289855072464, "loss": 0.8056, "step": 11450 }, { "epoch": 0.06, "learning_rate": 0.00017052173913043478, "loss": 0.7481, "step": 11475 }, { "epoch": 0.06, "learning_rate": 0.00017034057971014493, "loss": 0.7827, "step": 11500 }, { "epoch": 0.06, "learning_rate": 0.00017015942028985508, "loss": 0.7476, "step": 11525 }, { "epoch": 0.06, "learning_rate": 0.00016997826086956523, "loss": 0.7672, "step": 11550 }, { "epoch": 0.06, "learning_rate": 0.00016979710144927537, "loss": 0.9035, "step": 11575 }, { "epoch": 0.06, "learning_rate": 0.0001696159420289855, "loss": 0.8095, "step": 11600 }, { "epoch": 0.06, "learning_rate": 0.00016943478260869567, "loss": 0.8071, "step": 11625 }, { "epoch": 0.06, "learning_rate": 0.00016925362318840581, "loss": 0.8259, "step": 11650 }, { "epoch": 0.06, "learning_rate": 0.00016907246376811593, "loss": 0.7824, "step": 11675 }, { "epoch": 0.06, "learning_rate": 0.0001688913043478261, "loss": 0.6768, "step": 11700 }, { "epoch": 0.06, "learning_rate": 0.00016871014492753623, "loss": 0.7784, "step": 11725 }, { "epoch": 0.06, "learning_rate": 0.00016852898550724638, "loss": 0.7946, "step": 11750 }, { "epoch": 0.06, "learning_rate": 0.00016834782608695652, "loss": 0.7938, "step": 11775 }, { "epoch": 0.06, "learning_rate": 0.00016816666666666667, "loss": 0.7757, "step": 11800 }, { "epoch": 0.06, "learning_rate": 0.00016798550724637682, "loss": 0.784, "step": 11825 }, { "epoch": 0.06, "learning_rate": 0.00016780434782608696, "loss": 0.8067, "step": 11850 }, { "epoch": 0.06, "learning_rate": 0.0001676231884057971, "loss": 0.8073, "step": 11875 }, { "epoch": 0.06, "learning_rate": 0.00016744202898550726, "loss": 0.7267, "step": 11900 }, { "epoch": 0.06, "learning_rate": 0.00016726086956521738, "loss": 0.8355, "step": 11925 }, { "epoch": 0.06, "learning_rate": 0.00016707971014492753, "loss": 0.8044, "step": 11950 }, { "epoch": 0.06, "learning_rate": 0.0001668985507246377, "loss": 0.8151, "step": 11975 }, { "epoch": 0.06, "learning_rate": 0.00016671739130434782, "loss": 0.7596, "step": 12000 }, { "epoch": 0.06, "eval_loss": 0.7761228680610657, "eval_runtime": 6256.6305, "eval_samples_per_second": 0.8, "eval_steps_per_second": 0.2, "eval_wer": 51.930515674315956, "step": 12000 }, { "epoch": 0.06, "learning_rate": 0.00016653623188405797, "loss": 0.7608, "step": 12025 }, { "epoch": 0.06, "learning_rate": 0.00016635507246376814, "loss": 0.6833, "step": 12050 }, { "epoch": 0.06, "learning_rate": 0.00016617391304347826, "loss": 0.8248, "step": 12075 }, { "epoch": 0.06, "learning_rate": 0.0001659927536231884, "loss": 0.869, "step": 12100 }, { "epoch": 0.06, "learning_rate": 0.00016581159420289853, "loss": 0.7692, "step": 12125 }, { "epoch": 0.06, "learning_rate": 0.0001656304347826087, "loss": 0.8126, "step": 12150 }, { "epoch": 0.06, "learning_rate": 0.00016544927536231885, "loss": 0.7638, "step": 12175 }, { "epoch": 0.06, "learning_rate": 0.00016526811594202897, "loss": 0.755, "step": 12200 }, { "epoch": 0.06, "learning_rate": 0.00016508695652173915, "loss": 0.798, "step": 12225 }, { "epoch": 0.06, "learning_rate": 0.0001649057971014493, "loss": 0.8398, "step": 12250 }, { "epoch": 0.06, "learning_rate": 0.0001647246376811594, "loss": 0.7998, "step": 12275 }, { "epoch": 0.06, "learning_rate": 0.00016454347826086956, "loss": 0.7801, "step": 12300 }, { "epoch": 0.06, "learning_rate": 0.0001643623188405797, "loss": 0.7226, "step": 12325 }, { "epoch": 0.06, "learning_rate": 0.00016418115942028985, "loss": 0.8209, "step": 12350 }, { "epoch": 0.06, "learning_rate": 0.000164, "loss": 0.7869, "step": 12375 }, { "epoch": 0.06, "learning_rate": 0.00016381884057971015, "loss": 0.7806, "step": 12400 }, { "epoch": 0.06, "learning_rate": 0.0001636376811594203, "loss": 0.6899, "step": 12425 }, { "epoch": 0.06, "learning_rate": 0.00016345652173913044, "loss": 0.7796, "step": 12450 }, { "epoch": 0.06, "learning_rate": 0.00016327536231884056, "loss": 0.7448, "step": 12475 }, { "epoch": 0.06, "learning_rate": 0.00016309420289855074, "loss": 0.702, "step": 12500 }, { "epoch": 0.06, "learning_rate": 0.00016291304347826089, "loss": 0.8131, "step": 12525 }, { "epoch": 0.06, "learning_rate": 0.000162731884057971, "loss": 0.8154, "step": 12550 }, { "epoch": 0.06, "learning_rate": 0.00016255072463768118, "loss": 0.7873, "step": 12575 }, { "epoch": 0.06, "learning_rate": 0.0001623695652173913, "loss": 0.8329, "step": 12600 }, { "epoch": 0.06, "learning_rate": 0.00016218840579710145, "loss": 0.7596, "step": 12625 }, { "epoch": 0.06, "learning_rate": 0.0001620072463768116, "loss": 0.6257, "step": 12650 }, { "epoch": 0.06, "learning_rate": 0.00016182608695652174, "loss": 0.766, "step": 12675 }, { "epoch": 0.06, "learning_rate": 0.0001616449275362319, "loss": 0.7594, "step": 12700 }, { "epoch": 0.06, "learning_rate": 0.00016146376811594204, "loss": 0.776, "step": 12725 }, { "epoch": 0.06, "learning_rate": 0.00016128260869565218, "loss": 0.8994, "step": 12750 }, { "epoch": 0.06, "learning_rate": 0.00016110144927536233, "loss": 0.7237, "step": 12775 }, { "epoch": 0.06, "learning_rate": 0.00016092028985507245, "loss": 0.7658, "step": 12800 }, { "epoch": 0.06, "learning_rate": 0.0001607391304347826, "loss": 0.7995, "step": 12825 }, { "epoch": 0.06, "learning_rate": 0.00016055797101449277, "loss": 0.8313, "step": 12850 }, { "epoch": 0.06, "learning_rate": 0.0001603768115942029, "loss": 0.7675, "step": 12875 }, { "epoch": 0.06, "learning_rate": 0.00016019565217391304, "loss": 0.7119, "step": 12900 }, { "epoch": 0.06, "learning_rate": 0.0001600144927536232, "loss": 0.7035, "step": 12925 }, { "epoch": 0.06, "learning_rate": 0.00015983333333333333, "loss": 0.737, "step": 12950 }, { "epoch": 0.07, "learning_rate": 0.00015965217391304348, "loss": 0.694, "step": 12975 }, { "epoch": 0.07, "learning_rate": 0.0001594710144927536, "loss": 0.7334, "step": 13000 }, { "epoch": 0.07, "eval_loss": 0.7694315314292908, "eval_runtime": 6125.7667, "eval_samples_per_second": 0.817, "eval_steps_per_second": 0.204, "eval_wer": 49.44113529034861, "step": 13000 }, { "epoch": 0.07, "learning_rate": 0.00015928985507246377, "loss": 0.6969, "step": 13025 }, { "epoch": 0.07, "learning_rate": 0.00015910869565217392, "loss": 0.69, "step": 13050 }, { "epoch": 0.07, "learning_rate": 0.00015892753623188404, "loss": 0.8317, "step": 13075 }, { "epoch": 0.07, "learning_rate": 0.00015874637681159422, "loss": 0.8752, "step": 13100 }, { "epoch": 0.07, "learning_rate": 0.00015856521739130436, "loss": 0.7777, "step": 13125 }, { "epoch": 0.07, "learning_rate": 0.00015838405797101448, "loss": 0.7003, "step": 13150 }, { "epoch": 0.07, "learning_rate": 0.00015820289855072463, "loss": 0.745, "step": 13175 }, { "epoch": 0.07, "learning_rate": 0.0001580217391304348, "loss": 0.6816, "step": 13200 }, { "epoch": 0.07, "learning_rate": 0.00015784057971014493, "loss": 0.7321, "step": 13225 }, { "epoch": 0.07, "learning_rate": 0.00015765942028985507, "loss": 0.8165, "step": 13250 }, { "epoch": 0.07, "learning_rate": 0.00015747826086956522, "loss": 0.8318, "step": 13275 }, { "epoch": 0.07, "learning_rate": 0.00015729710144927537, "loss": 0.7568, "step": 13300 }, { "epoch": 0.07, "learning_rate": 0.00015711594202898551, "loss": 0.7191, "step": 13325 }, { "epoch": 0.07, "learning_rate": 0.00015693478260869563, "loss": 0.7375, "step": 13350 }, { "epoch": 0.07, "learning_rate": 0.0001567536231884058, "loss": 0.6437, "step": 13375 }, { "epoch": 0.07, "learning_rate": 0.00015657246376811596, "loss": 0.8342, "step": 13400 }, { "epoch": 0.07, "learning_rate": 0.00015639130434782608, "loss": 0.7003, "step": 13425 }, { "epoch": 0.07, "learning_rate": 0.00015621014492753625, "loss": 0.6729, "step": 13450 }, { "epoch": 0.07, "learning_rate": 0.00015603623188405796, "loss": 0.7319, "step": 13475 }, { "epoch": 0.07, "learning_rate": 0.00015585507246376813, "loss": 0.6888, "step": 13500 }, { "epoch": 0.07, "learning_rate": 0.00015567391304347828, "loss": 0.7646, "step": 13525 }, { "epoch": 0.07, "learning_rate": 0.0001554927536231884, "loss": 0.7568, "step": 13550 }, { "epoch": 0.07, "learning_rate": 0.00015531159420289857, "loss": 0.7149, "step": 13575 }, { "epoch": 0.07, "learning_rate": 0.0001551304347826087, "loss": 0.776, "step": 13600 }, { "epoch": 0.07, "learning_rate": 0.00015494927536231884, "loss": 0.735, "step": 13625 }, { "epoch": 0.07, "learning_rate": 0.000154768115942029, "loss": 0.7454, "step": 13650 }, { "epoch": 0.07, "learning_rate": 0.00015458695652173913, "loss": 0.7072, "step": 13675 }, { "epoch": 0.07, "learning_rate": 0.00015440579710144928, "loss": 0.7504, "step": 13700 }, { "epoch": 0.07, "learning_rate": 0.00015422463768115943, "loss": 0.6737, "step": 13725 }, { "epoch": 0.07, "learning_rate": 0.00015404347826086958, "loss": 0.7757, "step": 13750 }, { "epoch": 0.07, "learning_rate": 0.00015386231884057972, "loss": 0.7659, "step": 13775 }, { "epoch": 0.07, "learning_rate": 0.00015368115942028984, "loss": 0.6894, "step": 13800 }, { "epoch": 0.07, "learning_rate": 0.0001535, "loss": 0.6959, "step": 13825 }, { "epoch": 0.07, "learning_rate": 0.00015331884057971016, "loss": 0.7326, "step": 13850 }, { "epoch": 0.07, "learning_rate": 0.00015313768115942029, "loss": 0.7797, "step": 13875 }, { "epoch": 0.07, "learning_rate": 0.00015295652173913043, "loss": 0.6998, "step": 13900 }, { "epoch": 0.07, "learning_rate": 0.0001527753623188406, "loss": 0.7806, "step": 13925 }, { "epoch": 0.07, "learning_rate": 0.00015259420289855073, "loss": 0.7847, "step": 13950 }, { "epoch": 0.07, "learning_rate": 0.00015241304347826087, "loss": 0.7633, "step": 13975 }, { "epoch": 0.07, "learning_rate": 0.000152231884057971, "loss": 0.708, "step": 14000 }, { "epoch": 0.07, "eval_loss": 0.7335543632507324, "eval_runtime": 6161.4493, "eval_samples_per_second": 0.812, "eval_steps_per_second": 0.203, "eval_wer": 47.00396307458649, "step": 14000 }, { "epoch": 0.07, "learning_rate": 0.00015205072463768117, "loss": 0.7598, "step": 14025 }, { "epoch": 0.07, "learning_rate": 0.00015186956521739132, "loss": 0.7354, "step": 14050 }, { "epoch": 0.07, "learning_rate": 0.00015168840579710144, "loss": 0.7034, "step": 14075 }, { "epoch": 0.07, "learning_rate": 0.0001515072463768116, "loss": 0.6852, "step": 14100 }, { "epoch": 0.07, "learning_rate": 0.00015132608695652176, "loss": 0.7925, "step": 14125 }, { "epoch": 0.07, "learning_rate": 0.00015114492753623188, "loss": 0.7146, "step": 14150 }, { "epoch": 0.07, "learning_rate": 0.00015096376811594202, "loss": 0.714, "step": 14175 }, { "epoch": 0.07, "learning_rate": 0.0001507826086956522, "loss": 0.7124, "step": 14200 }, { "epoch": 0.07, "learning_rate": 0.00015060144927536232, "loss": 0.7223, "step": 14225 }, { "epoch": 0.07, "learning_rate": 0.00015042028985507247, "loss": 0.8055, "step": 14250 }, { "epoch": 0.07, "learning_rate": 0.0001502391304347826, "loss": 0.651, "step": 14275 }, { "epoch": 0.07, "learning_rate": 0.00015005797101449276, "loss": 0.6731, "step": 14300 }, { "epoch": 0.07, "learning_rate": 0.0001498768115942029, "loss": 0.7672, "step": 14325 }, { "epoch": 0.07, "learning_rate": 0.00014969565217391303, "loss": 0.706, "step": 14350 }, { "epoch": 0.07, "learning_rate": 0.0001495144927536232, "loss": 0.6925, "step": 14375 }, { "epoch": 0.07, "learning_rate": 0.00014933333333333335, "loss": 0.8309, "step": 14400 }, { "epoch": 0.07, "learning_rate": 0.00014915217391304347, "loss": 0.7526, "step": 14425 }, { "epoch": 0.07, "learning_rate": 0.00014897101449275364, "loss": 0.7555, "step": 14450 }, { "epoch": 0.07, "learning_rate": 0.00014878985507246376, "loss": 0.7689, "step": 14475 }, { "epoch": 0.07, "learning_rate": 0.0001486086956521739, "loss": 0.6583, "step": 14500 }, { "epoch": 0.07, "learning_rate": 0.00014842753623188406, "loss": 0.7107, "step": 14525 }, { "epoch": 0.07, "learning_rate": 0.0001482463768115942, "loss": 0.7068, "step": 14550 }, { "epoch": 0.07, "learning_rate": 0.00014806521739130435, "loss": 0.795, "step": 14575 }, { "epoch": 0.07, "learning_rate": 0.0001478840579710145, "loss": 0.679, "step": 14600 }, { "epoch": 0.07, "learning_rate": 0.00014770289855072465, "loss": 0.7713, "step": 14625 }, { "epoch": 0.07, "learning_rate": 0.0001475217391304348, "loss": 0.6923, "step": 14650 }, { "epoch": 0.07, "learning_rate": 0.00014734057971014491, "loss": 0.7371, "step": 14675 }, { "epoch": 0.07, "learning_rate": 0.00014715942028985506, "loss": 0.7118, "step": 14700 }, { "epoch": 0.07, "learning_rate": 0.00014697826086956524, "loss": 0.724, "step": 14725 }, { "epoch": 0.07, "learning_rate": 0.00014679710144927536, "loss": 0.7121, "step": 14750 }, { "epoch": 0.07, "learning_rate": 0.0001466159420289855, "loss": 0.6898, "step": 14775 }, { "epoch": 0.07, "learning_rate": 0.00014643478260869568, "loss": 0.6558, "step": 14800 }, { "epoch": 0.07, "learning_rate": 0.0001462536231884058, "loss": 0.7129, "step": 14825 }, { "epoch": 0.07, "learning_rate": 0.00014607246376811594, "loss": 0.6833, "step": 14850 }, { "epoch": 0.07, "learning_rate": 0.0001458913043478261, "loss": 0.6877, "step": 14875 }, { "epoch": 0.07, "learning_rate": 0.00014571014492753624, "loss": 0.6646, "step": 14900 }, { "epoch": 0.07, "learning_rate": 0.00014552898550724639, "loss": 0.6781, "step": 14925 }, { "epoch": 0.07, "learning_rate": 0.0001453478260869565, "loss": 0.7139, "step": 14950 }, { "epoch": 0.08, "learning_rate": 0.00014516666666666668, "loss": 0.7152, "step": 14975 }, { "epoch": 0.08, "learning_rate": 0.00014498550724637683, "loss": 0.7112, "step": 15000 }, { "epoch": 0.08, "eval_loss": 0.7149045467376709, "eval_runtime": 6211.0734, "eval_samples_per_second": 0.806, "eval_steps_per_second": 0.202, "eval_wer": 47.578252924843966, "step": 15000 }, { "epoch": 0.08, "learning_rate": 0.00014480434782608695, "loss": 0.7359, "step": 15025 }, { "epoch": 0.08, "learning_rate": 0.0001446231884057971, "loss": 0.7161, "step": 15050 }, { "epoch": 0.08, "learning_rate": 0.00014444202898550727, "loss": 0.7225, "step": 15075 }, { "epoch": 0.08, "learning_rate": 0.0001442608695652174, "loss": 0.6912, "step": 15100 }, { "epoch": 0.08, "learning_rate": 0.00014407971014492754, "loss": 0.7643, "step": 15125 }, { "epoch": 0.08, "learning_rate": 0.00014389855072463768, "loss": 0.7172, "step": 15150 }, { "epoch": 0.08, "learning_rate": 0.00014371739130434783, "loss": 0.6586, "step": 15175 }, { "epoch": 0.08, "learning_rate": 0.00014353623188405798, "loss": 0.7689, "step": 15200 }, { "epoch": 0.08, "learning_rate": 0.0001433550724637681, "loss": 0.6689, "step": 15225 }, { "epoch": 0.08, "learning_rate": 0.00014317391304347827, "loss": 0.6828, "step": 15250 }, { "epoch": 0.08, "learning_rate": 0.00014299275362318842, "loss": 0.626, "step": 15275 }, { "epoch": 0.08, "learning_rate": 0.00014281159420289854, "loss": 0.6763, "step": 15300 }, { "epoch": 0.08, "learning_rate": 0.00014263043478260871, "loss": 0.733, "step": 15325 }, { "epoch": 0.08, "learning_rate": 0.00014244927536231883, "loss": 0.712, "step": 15350 }, { "epoch": 0.08, "learning_rate": 0.00014226811594202898, "loss": 0.7525, "step": 15375 }, { "epoch": 0.08, "learning_rate": 0.00014208695652173913, "loss": 0.6006, "step": 15400 }, { "epoch": 0.08, "learning_rate": 0.00014190579710144928, "loss": 0.66, "step": 15425 }, { "epoch": 0.08, "learning_rate": 0.00014172463768115942, "loss": 0.6597, "step": 15450 }, { "epoch": 0.08, "learning_rate": 0.00014154347826086957, "loss": 0.7359, "step": 15475 }, { "epoch": 0.08, "learning_rate": 0.00014136231884057972, "loss": 0.7218, "step": 15500 }, { "epoch": 0.08, "learning_rate": 0.00014118115942028986, "loss": 0.6768, "step": 15525 }, { "epoch": 0.08, "learning_rate": 0.00014099999999999998, "loss": 0.7398, "step": 15550 }, { "epoch": 0.08, "learning_rate": 0.00014081884057971013, "loss": 0.6622, "step": 15575 }, { "epoch": 0.08, "learning_rate": 0.0001406376811594203, "loss": 0.6799, "step": 15600 }, { "epoch": 0.08, "learning_rate": 0.00014045652173913043, "loss": 0.6402, "step": 15625 }, { "epoch": 0.08, "learning_rate": 0.00014027536231884057, "loss": 0.6506, "step": 15650 }, { "epoch": 0.08, "learning_rate": 0.00014009420289855075, "loss": 0.6803, "step": 15675 }, { "epoch": 0.08, "learning_rate": 0.00013991304347826087, "loss": 0.6568, "step": 15700 }, { "epoch": 0.08, "learning_rate": 0.00013973188405797102, "loss": 0.6378, "step": 15725 }, { "epoch": 0.08, "learning_rate": 0.00013955072463768116, "loss": 0.6953, "step": 15750 }, { "epoch": 0.08, "learning_rate": 0.0001393695652173913, "loss": 0.6846, "step": 15775 }, { "epoch": 0.08, "learning_rate": 0.00013918840579710146, "loss": 0.6657, "step": 15800 }, { "epoch": 0.08, "learning_rate": 0.00013900724637681158, "loss": 0.639, "step": 15825 }, { "epoch": 0.08, "learning_rate": 0.00013883333333333334, "loss": 0.6946, "step": 15850 }, { "epoch": 0.08, "learning_rate": 0.00013865217391304349, "loss": 0.7005, "step": 15875 }, { "epoch": 0.08, "learning_rate": 0.00013847101449275363, "loss": 0.7449, "step": 15900 }, { "epoch": 0.08, "learning_rate": 0.00013828985507246378, "loss": 0.6588, "step": 15925 }, { "epoch": 0.08, "learning_rate": 0.0001381086956521739, "loss": 0.6473, "step": 15950 }, { "epoch": 0.08, "learning_rate": 0.00013792753623188405, "loss": 0.6854, "step": 15975 }, { "epoch": 0.08, "learning_rate": 0.00013774637681159422, "loss": 0.6989, "step": 16000 }, { "epoch": 0.08, "eval_loss": 0.6713331937789917, "eval_runtime": 6174.3958, "eval_samples_per_second": 0.811, "eval_steps_per_second": 0.203, "eval_wer": 44.29863072213389, "step": 16000 }, { "epoch": 0.08, "learning_rate": 0.00013756521739130434, "loss": 0.6563, "step": 16025 }, { "epoch": 0.08, "learning_rate": 0.0001373840579710145, "loss": 0.6982, "step": 16050 }, { "epoch": 0.08, "learning_rate": 0.00013720289855072466, "loss": 0.6678, "step": 16075 }, { "epoch": 0.08, "learning_rate": 0.00013702173913043478, "loss": 0.6593, "step": 16100 }, { "epoch": 0.08, "learning_rate": 0.00013684057971014493, "loss": 0.6349, "step": 16125 }, { "epoch": 0.08, "learning_rate": 0.00013665942028985505, "loss": 0.7161, "step": 16150 }, { "epoch": 0.08, "learning_rate": 0.00013647826086956522, "loss": 0.6789, "step": 16175 }, { "epoch": 0.08, "learning_rate": 0.00013629710144927537, "loss": 0.6891, "step": 16200 }, { "epoch": 0.08, "learning_rate": 0.0001361159420289855, "loss": 0.752, "step": 16225 }, { "epoch": 0.08, "learning_rate": 0.00013593478260869567, "loss": 0.6384, "step": 16250 }, { "epoch": 0.08, "learning_rate": 0.0001357536231884058, "loss": 0.6149, "step": 16275 }, { "epoch": 0.08, "learning_rate": 0.00013557246376811593, "loss": 0.7397, "step": 16300 }, { "epoch": 0.08, "learning_rate": 0.0001353913043478261, "loss": 0.6256, "step": 16325 }, { "epoch": 0.08, "learning_rate": 0.00013521014492753623, "loss": 0.7775, "step": 16350 }, { "epoch": 0.08, "learning_rate": 0.00013502898550724637, "loss": 0.6211, "step": 16375 }, { "epoch": 0.08, "learning_rate": 0.00013484782608695652, "loss": 0.6559, "step": 16400 }, { "epoch": 0.08, "learning_rate": 0.00013466666666666667, "loss": 0.6802, "step": 16425 }, { "epoch": 0.08, "learning_rate": 0.00013448550724637682, "loss": 0.7037, "step": 16450 }, { "epoch": 0.08, "learning_rate": 0.00013430434782608696, "loss": 0.6405, "step": 16475 }, { "epoch": 0.08, "learning_rate": 0.0001341231884057971, "loss": 0.6445, "step": 16500 }, { "epoch": 0.08, "learning_rate": 0.00013394202898550726, "loss": 0.6343, "step": 16525 }, { "epoch": 0.08, "learning_rate": 0.0001337608695652174, "loss": 0.6078, "step": 16550 }, { "epoch": 0.08, "learning_rate": 0.00013357971014492753, "loss": 0.6994, "step": 16575 }, { "epoch": 0.08, "learning_rate": 0.0001333985507246377, "loss": 0.6863, "step": 16600 }, { "epoch": 0.08, "learning_rate": 0.00013321739130434782, "loss": 0.7175, "step": 16625 }, { "epoch": 0.08, "learning_rate": 0.00013303623188405797, "loss": 0.654, "step": 16650 }, { "epoch": 0.08, "learning_rate": 0.00013285507246376814, "loss": 0.7037, "step": 16675 }, { "epoch": 0.08, "learning_rate": 0.00013267391304347826, "loss": 0.6302, "step": 16700 }, { "epoch": 0.08, "learning_rate": 0.0001324927536231884, "loss": 0.6877, "step": 16725 }, { "epoch": 0.08, "learning_rate": 0.00013231159420289856, "loss": 0.698, "step": 16750 }, { "epoch": 0.08, "learning_rate": 0.0001321304347826087, "loss": 0.6587, "step": 16775 }, { "epoch": 0.08, "learning_rate": 0.00013194927536231885, "loss": 0.6436, "step": 16800 }, { "epoch": 0.08, "learning_rate": 0.00013176811594202897, "loss": 0.6632, "step": 16825 }, { "epoch": 0.08, "learning_rate": 0.00013158695652173914, "loss": 0.7216, "step": 16850 }, { "epoch": 0.08, "learning_rate": 0.0001314057971014493, "loss": 0.7142, "step": 16875 }, { "epoch": 0.08, "learning_rate": 0.0001312246376811594, "loss": 0.6373, "step": 16900 }, { "epoch": 0.08, "learning_rate": 0.00013104347826086956, "loss": 0.7663, "step": 16925 }, { "epoch": 0.08, "learning_rate": 0.00013086231884057973, "loss": 0.6699, "step": 16950 }, { "epoch": 0.09, "learning_rate": 0.00013068115942028985, "loss": 0.6204, "step": 16975 }, { "epoch": 0.09, "learning_rate": 0.0001305, "loss": 0.7025, "step": 17000 }, { "epoch": 0.09, "eval_loss": 0.6639269590377808, "eval_runtime": 6204.7668, "eval_samples_per_second": 0.807, "eval_steps_per_second": 0.202, "eval_wer": 43.748071857424236, "step": 17000 }, { "epoch": 0.09, "learning_rate": 0.00013031884057971015, "loss": 0.6292, "step": 17025 }, { "epoch": 0.09, "learning_rate": 0.0001301376811594203, "loss": 0.6168, "step": 17050 }, { "epoch": 0.09, "learning_rate": 0.00012995652173913044, "loss": 0.6671, "step": 17075 }, { "epoch": 0.09, "learning_rate": 0.00012977536231884056, "loss": 0.6828, "step": 17100 }, { "epoch": 0.09, "learning_rate": 0.00012959420289855074, "loss": 0.6362, "step": 17125 }, { "epoch": 0.09, "learning_rate": 0.00012941304347826088, "loss": 0.6586, "step": 17150 }, { "epoch": 0.09, "learning_rate": 0.000129231884057971, "loss": 0.6173, "step": 17175 }, { "epoch": 0.09, "learning_rate": 0.00012905072463768118, "loss": 0.6647, "step": 17200 }, { "epoch": 0.09, "learning_rate": 0.0001288695652173913, "loss": 0.6498, "step": 17225 }, { "epoch": 0.09, "learning_rate": 0.00012868840579710145, "loss": 0.6818, "step": 17250 }, { "epoch": 0.09, "learning_rate": 0.0001285072463768116, "loss": 0.6441, "step": 17275 }, { "epoch": 0.09, "learning_rate": 0.00012832608695652174, "loss": 0.6493, "step": 17300 }, { "epoch": 0.09, "learning_rate": 0.0001281449275362319, "loss": 0.5915, "step": 17325 }, { "epoch": 0.09, "learning_rate": 0.00012796376811594203, "loss": 0.6864, "step": 17350 }, { "epoch": 0.09, "learning_rate": 0.00012778260869565218, "loss": 0.6784, "step": 17375 }, { "epoch": 0.09, "learning_rate": 0.00012760144927536233, "loss": 0.6209, "step": 17400 }, { "epoch": 0.09, "learning_rate": 0.00012742028985507248, "loss": 0.681, "step": 17425 }, { "epoch": 0.09, "learning_rate": 0.0001272391304347826, "loss": 0.6087, "step": 17450 }, { "epoch": 0.09, "learning_rate": 0.00012705797101449277, "loss": 0.6084, "step": 17475 }, { "epoch": 0.09, "learning_rate": 0.0001268768115942029, "loss": 0.7131, "step": 17500 }, { "epoch": 0.09, "learning_rate": 0.00012669565217391304, "loss": 0.6433, "step": 17525 }, { "epoch": 0.09, "learning_rate": 0.0001265144927536232, "loss": 0.7002, "step": 17550 }, { "epoch": 0.09, "learning_rate": 0.00012633333333333333, "loss": 0.6467, "step": 17575 }, { "epoch": 0.09, "learning_rate": 0.00012615217391304348, "loss": 0.7296, "step": 17600 }, { "epoch": 0.09, "learning_rate": 0.00012597101449275363, "loss": 0.6358, "step": 17625 }, { "epoch": 0.09, "learning_rate": 0.00012578985507246377, "loss": 0.6227, "step": 17650 }, { "epoch": 0.09, "learning_rate": 0.00012560869565217392, "loss": 0.6872, "step": 17675 }, { "epoch": 0.09, "learning_rate": 0.00012542753623188404, "loss": 0.661, "step": 17700 }, { "epoch": 0.09, "learning_rate": 0.00012524637681159422, "loss": 0.661, "step": 17725 }, { "epoch": 0.09, "learning_rate": 0.00012506521739130436, "loss": 0.6795, "step": 17750 }, { "epoch": 0.09, "learning_rate": 0.0001248840579710145, "loss": 0.6197, "step": 17775 }, { "epoch": 0.09, "learning_rate": 0.00012470289855072463, "loss": 0.6697, "step": 17800 }, { "epoch": 0.09, "learning_rate": 0.00012452173913043478, "loss": 0.6658, "step": 17825 }, { "epoch": 0.09, "learning_rate": 0.00012434057971014492, "loss": 0.5313, "step": 17850 }, { "epoch": 0.09, "learning_rate": 0.00012415942028985507, "loss": 0.7019, "step": 17875 }, { "epoch": 0.09, "learning_rate": 0.00012397826086956522, "loss": 0.6018, "step": 17900 }, { "epoch": 0.09, "learning_rate": 0.00012379710144927537, "loss": 0.5935, "step": 17925 }, { "epoch": 0.09, "learning_rate": 0.0001236159420289855, "loss": 0.6657, "step": 17950 }, { "epoch": 0.09, "learning_rate": 0.00012343478260869566, "loss": 0.6591, "step": 17975 }, { "epoch": 0.09, "learning_rate": 0.0001232536231884058, "loss": 0.6127, "step": 18000 }, { "epoch": 0.09, "eval_loss": 0.6477032899856567, "eval_runtime": 6184.9416, "eval_samples_per_second": 0.809, "eval_steps_per_second": 0.202, "eval_wer": 42.91274116614063, "step": 18000 }, { "epoch": 0.09, "learning_rate": 0.00012307246376811595, "loss": 0.6471, "step": 18025 }, { "epoch": 0.09, "learning_rate": 0.00012289130434782607, "loss": 0.5746, "step": 18050 }, { "epoch": 0.09, "learning_rate": 0.00012271014492753625, "loss": 0.6262, "step": 18075 }, { "epoch": 0.09, "learning_rate": 0.00012252898550724637, "loss": 0.646, "step": 18100 }, { "epoch": 0.09, "learning_rate": 0.00012234782608695652, "loss": 0.5969, "step": 18125 }, { "epoch": 0.09, "learning_rate": 0.00012216666666666666, "loss": 0.6934, "step": 18150 }, { "epoch": 0.09, "learning_rate": 0.00012198550724637681, "loss": 0.6236, "step": 18175 }, { "epoch": 0.09, "learning_rate": 0.00012180434782608696, "loss": 0.6376, "step": 18200 }, { "epoch": 0.09, "learning_rate": 0.0001216231884057971, "loss": 0.6312, "step": 18225 }, { "epoch": 0.09, "learning_rate": 0.00012144202898550725, "loss": 0.5801, "step": 18250 }, { "epoch": 0.09, "learning_rate": 0.0001212608695652174, "loss": 0.6155, "step": 18275 }, { "epoch": 0.09, "learning_rate": 0.00012107971014492753, "loss": 0.6514, "step": 18300 }, { "epoch": 0.09, "learning_rate": 0.00012089855072463768, "loss": 0.627, "step": 18325 }, { "epoch": 0.09, "learning_rate": 0.00012071739130434783, "loss": 0.6617, "step": 18350 }, { "epoch": 0.09, "learning_rate": 0.00012054347826086957, "loss": 0.6565, "step": 18375 }, { "epoch": 0.09, "learning_rate": 0.00012036231884057971, "loss": 0.6107, "step": 18400 }, { "epoch": 0.09, "learning_rate": 0.00012018115942028986, "loss": 0.6027, "step": 18425 }, { "epoch": 0.09, "learning_rate": 0.00012, "loss": 0.6002, "step": 18450 }, { "epoch": 0.09, "learning_rate": 0.00011981884057971015, "loss": 0.5878, "step": 18475 }, { "epoch": 0.09, "learning_rate": 0.0001196376811594203, "loss": 0.6664, "step": 18500 }, { "epoch": 0.09, "learning_rate": 0.00011945652173913043, "loss": 0.6488, "step": 18525 }, { "epoch": 0.09, "learning_rate": 0.00011927536231884058, "loss": 0.5983, "step": 18550 }, { "epoch": 0.09, "learning_rate": 0.00011909420289855074, "loss": 0.6504, "step": 18575 }, { "epoch": 0.09, "learning_rate": 0.00011891304347826087, "loss": 0.6248, "step": 18600 }, { "epoch": 0.09, "learning_rate": 0.00011873188405797102, "loss": 0.5826, "step": 18625 }, { "epoch": 0.09, "learning_rate": 0.00011855072463768115, "loss": 0.6357, "step": 18650 }, { "epoch": 0.09, "learning_rate": 0.00011836956521739131, "loss": 0.6931, "step": 18675 }, { "epoch": 0.09, "learning_rate": 0.00011818840579710145, "loss": 0.5672, "step": 18700 }, { "epoch": 0.09, "learning_rate": 0.0001180072463768116, "loss": 0.6437, "step": 18725 }, { "epoch": 0.09, "learning_rate": 0.00011782608695652174, "loss": 0.6527, "step": 18750 }, { "epoch": 0.09, "learning_rate": 0.00011764492753623189, "loss": 0.6001, "step": 18775 }, { "epoch": 0.09, "learning_rate": 0.00011746376811594204, "loss": 0.6255, "step": 18800 }, { "epoch": 0.09, "learning_rate": 0.00011728260869565217, "loss": 0.6615, "step": 18825 }, { "epoch": 0.09, "learning_rate": 0.00011710144927536232, "loss": 0.6139, "step": 18850 }, { "epoch": 0.09, "learning_rate": 0.00011692028985507246, "loss": 0.6925, "step": 18875 }, { "epoch": 0.09, "learning_rate": 0.00011673913043478261, "loss": 0.6406, "step": 18900 }, { "epoch": 0.09, "learning_rate": 0.00011655797101449276, "loss": 0.6745, "step": 18925 }, { "epoch": 0.09, "learning_rate": 0.00011637681159420289, "loss": 0.6063, "step": 18950 }, { "epoch": 0.1, "learning_rate": 0.00011619565217391305, "loss": 0.617, "step": 18975 }, { "epoch": 0.1, "learning_rate": 0.00011601449275362319, "loss": 0.6342, "step": 19000 }, { "epoch": 0.1, "eval_loss": 0.6297908425331116, "eval_runtime": 6186.457, "eval_samples_per_second": 0.809, "eval_steps_per_second": 0.202, "eval_wer": 42.68255060632668, "step": 19000 }, { "epoch": 0.1, "learning_rate": 0.00011583333333333333, "loss": 0.6395, "step": 19025 }, { "epoch": 0.1, "learning_rate": 0.00011565217391304348, "loss": 0.6232, "step": 19050 }, { "epoch": 0.1, "learning_rate": 0.00011547101449275363, "loss": 0.6556, "step": 19075 }, { "epoch": 0.1, "learning_rate": 0.00011528985507246378, "loss": 0.5557, "step": 19100 }, { "epoch": 0.1, "learning_rate": 0.00011510869565217391, "loss": 0.6073, "step": 19125 }, { "epoch": 0.1, "learning_rate": 0.00011492753623188407, "loss": 0.6471, "step": 19150 }, { "epoch": 0.1, "learning_rate": 0.0001147463768115942, "loss": 0.5657, "step": 19175 }, { "epoch": 0.1, "learning_rate": 0.00011456521739130435, "loss": 0.6348, "step": 19200 }, { "epoch": 0.1, "learning_rate": 0.00011438405797101448, "loss": 0.6581, "step": 19225 }, { "epoch": 0.1, "learning_rate": 0.00011420289855072465, "loss": 0.5888, "step": 19250 }, { "epoch": 0.1, "learning_rate": 0.00011402173913043479, "loss": 0.648, "step": 19275 }, { "epoch": 0.1, "learning_rate": 0.00011384057971014493, "loss": 0.6632, "step": 19300 }, { "epoch": 0.1, "learning_rate": 0.00011365942028985507, "loss": 0.627, "step": 19325 }, { "epoch": 0.1, "learning_rate": 0.00011347826086956522, "loss": 0.6466, "step": 19350 }, { "epoch": 0.1, "learning_rate": 0.00011329710144927537, "loss": 0.6318, "step": 19375 }, { "epoch": 0.1, "learning_rate": 0.0001131159420289855, "loss": 0.6421, "step": 19400 }, { "epoch": 0.1, "learning_rate": 0.00011293478260869565, "loss": 0.6173, "step": 19425 }, { "epoch": 0.1, "learning_rate": 0.00011275362318840581, "loss": 0.6033, "step": 19450 }, { "epoch": 0.1, "learning_rate": 0.00011257246376811594, "loss": 0.5807, "step": 19475 }, { "epoch": 0.1, "learning_rate": 0.00011239130434782609, "loss": 0.5878, "step": 19500 }, { "epoch": 0.1, "learning_rate": 0.00011221014492753622, "loss": 0.5989, "step": 19525 }, { "epoch": 0.1, "learning_rate": 0.00011202898550724639, "loss": 0.5648, "step": 19550 }, { "epoch": 0.1, "learning_rate": 0.00011184782608695652, "loss": 0.6106, "step": 19575 }, { "epoch": 0.1, "learning_rate": 0.00011166666666666667, "loss": 0.6043, "step": 19600 }, { "epoch": 0.1, "learning_rate": 0.00011148550724637681, "loss": 0.6108, "step": 19625 }, { "epoch": 0.1, "learning_rate": 0.00011130434782608696, "loss": 0.6729, "step": 19650 }, { "epoch": 0.1, "learning_rate": 0.00011112318840579711, "loss": 0.6472, "step": 19675 }, { "epoch": 0.1, "learning_rate": 0.00011094202898550724, "loss": 0.6716, "step": 19700 }, { "epoch": 0.1, "learning_rate": 0.00011076086956521739, "loss": 0.6657, "step": 19725 }, { "epoch": 0.1, "learning_rate": 0.00011057971014492754, "loss": 0.6297, "step": 19750 }, { "epoch": 0.1, "learning_rate": 0.00011039855072463768, "loss": 0.6371, "step": 19775 }, { "epoch": 0.1, "learning_rate": 0.00011021739130434783, "loss": 0.6213, "step": 19800 }, { "epoch": 0.1, "learning_rate": 0.00011003623188405798, "loss": 0.6303, "step": 19825 }, { "epoch": 0.1, "learning_rate": 0.00010985507246376812, "loss": 0.665, "step": 19850 }, { "epoch": 0.1, "learning_rate": 0.00010967391304347826, "loss": 0.6263, "step": 19875 }, { "epoch": 0.1, "learning_rate": 0.0001094927536231884, "loss": 0.6424, "step": 19900 }, { "epoch": 0.1, "learning_rate": 0.00010931159420289857, "loss": 0.6549, "step": 19925 }, { "epoch": 0.1, "learning_rate": 0.0001091304347826087, "loss": 0.6327, "step": 19950 }, { "epoch": 0.1, "learning_rate": 0.00010894927536231885, "loss": 0.5384, "step": 19975 }, { "epoch": 0.1, "learning_rate": 0.00010876811594202898, "loss": 0.6174, "step": 20000 }, { "epoch": 0.1, "eval_loss": 0.6079863905906677, "eval_runtime": 6150.9443, "eval_samples_per_second": 0.814, "eval_steps_per_second": 0.204, "eval_wer": 40.11723106860628, "step": 20000 }, { "epoch": 0.1, "learning_rate": 0.00010858695652173914, "loss": 0.6349, "step": 20025 }, { "epoch": 0.1, "learning_rate": 0.00010840579710144927, "loss": 0.5833, "step": 20050 }, { "epoch": 0.1, "learning_rate": 0.00010822463768115942, "loss": 0.5931, "step": 20075 }, { "epoch": 0.1, "learning_rate": 0.00010804347826086957, "loss": 0.5622, "step": 20100 }, { "epoch": 0.1, "learning_rate": 0.00010786231884057972, "loss": 0.5825, "step": 20125 }, { "epoch": 0.1, "learning_rate": 0.00010768115942028986, "loss": 0.5919, "step": 20150 }, { "epoch": 0.1, "learning_rate": 0.0001075, "loss": 0.5773, "step": 20175 }, { "epoch": 0.1, "learning_rate": 0.00010731884057971014, "loss": 0.6487, "step": 20200 }, { "epoch": 0.1, "learning_rate": 0.00010713768115942029, "loss": 0.6037, "step": 20225 }, { "epoch": 0.1, "learning_rate": 0.00010695652173913044, "loss": 0.4728, "step": 20250 }, { "epoch": 0.1, "learning_rate": 0.00010677536231884059, "loss": 0.594, "step": 20275 }, { "epoch": 0.1, "learning_rate": 0.00010659420289855072, "loss": 0.5955, "step": 20300 }, { "epoch": 0.1, "learning_rate": 0.00010641304347826088, "loss": 0.607, "step": 20325 }, { "epoch": 0.1, "learning_rate": 0.00010623188405797101, "loss": 0.5924, "step": 20350 }, { "epoch": 0.1, "learning_rate": 0.00010605072463768116, "loss": 0.5852, "step": 20375 }, { "epoch": 0.1, "learning_rate": 0.0001058695652173913, "loss": 0.5136, "step": 20400 }, { "epoch": 0.1, "learning_rate": 0.00010568840579710146, "loss": 0.537, "step": 20425 }, { "epoch": 0.1, "learning_rate": 0.0001055072463768116, "loss": 0.6373, "step": 20450 }, { "epoch": 0.1, "learning_rate": 0.00010532608695652174, "loss": 0.5526, "step": 20475 }, { "epoch": 0.1, "learning_rate": 0.00010514492753623188, "loss": 0.5994, "step": 20500 }, { "epoch": 0.1, "learning_rate": 0.00010496376811594203, "loss": 0.583, "step": 20525 }, { "epoch": 0.1, "learning_rate": 0.00010478260869565218, "loss": 0.6394, "step": 20550 }, { "epoch": 0.1, "learning_rate": 0.00010460144927536231, "loss": 0.55, "step": 20575 }, { "epoch": 0.1, "learning_rate": 0.00010442028985507247, "loss": 0.5821, "step": 20600 }, { "epoch": 0.1, "learning_rate": 0.00010423913043478262, "loss": 0.6771, "step": 20625 }, { "epoch": 0.1, "learning_rate": 0.00010405797101449275, "loss": 0.6192, "step": 20650 }, { "epoch": 0.1, "learning_rate": 0.0001038768115942029, "loss": 0.6443, "step": 20675 }, { "epoch": 0.1, "learning_rate": 0.00010369565217391305, "loss": 0.6412, "step": 20700 }, { "epoch": 0.1, "learning_rate": 0.0001035144927536232, "loss": 0.6015, "step": 20725 }, { "epoch": 0.1, "learning_rate": 0.00010333333333333333, "loss": 0.6236, "step": 20750 }, { "epoch": 0.1, "learning_rate": 0.00010315217391304348, "loss": 0.5623, "step": 20775 }, { "epoch": 0.1, "learning_rate": 0.00010297101449275364, "loss": 0.615, "step": 20800 }, { "epoch": 0.1, "learning_rate": 0.00010278985507246377, "loss": 0.5963, "step": 20825 }, { "epoch": 0.1, "learning_rate": 0.00010260869565217392, "loss": 0.5786, "step": 20850 }, { "epoch": 0.1, "learning_rate": 0.00010242753623188405, "loss": 0.5758, "step": 20875 }, { "epoch": 0.1, "learning_rate": 0.0001022536231884058, "loss": 0.6087, "step": 20900 }, { "epoch": 0.1, "learning_rate": 0.00010207246376811595, "loss": 0.5862, "step": 20925 }, { "epoch": 0.11, "learning_rate": 0.00010189130434782609, "loss": 0.6518, "step": 20950 }, { "epoch": 0.11, "learning_rate": 0.00010171014492753623, "loss": 0.602, "step": 20975 }, { "epoch": 0.11, "learning_rate": 0.00010152898550724637, "loss": 0.5551, "step": 21000 }, { "epoch": 0.11, "eval_loss": 0.5896329283714294, "eval_runtime": 6188.8975, "eval_samples_per_second": 0.809, "eval_steps_per_second": 0.202, "eval_wer": 39.03984432473481, "step": 21000 }, { "epoch": 0.11, "learning_rate": 0.00010134782608695653, "loss": 0.565, "step": 21025 }, { "epoch": 0.11, "learning_rate": 0.00010116666666666667, "loss": 0.6315, "step": 21050 }, { "epoch": 0.11, "learning_rate": 0.00010098550724637682, "loss": 0.5814, "step": 21075 }, { "epoch": 0.11, "learning_rate": 0.00010080434782608695, "loss": 0.6221, "step": 21100 }, { "epoch": 0.11, "learning_rate": 0.00010062318840579711, "loss": 0.5563, "step": 21125 }, { "epoch": 0.11, "learning_rate": 0.00010044202898550724, "loss": 0.648, "step": 21150 }, { "epoch": 0.11, "learning_rate": 0.00010026086956521739, "loss": 0.6128, "step": 21175 }, { "epoch": 0.11, "learning_rate": 0.00010007971014492754, "loss": 0.533, "step": 21200 }, { "epoch": 0.11, "learning_rate": 9.989855072463769e-05, "loss": 0.6296, "step": 21225 }, { "epoch": 0.11, "learning_rate": 9.971739130434783e-05, "loss": 0.6338, "step": 21250 }, { "epoch": 0.11, "learning_rate": 9.953623188405797e-05, "loss": 0.579, "step": 21275 }, { "epoch": 0.11, "learning_rate": 9.935507246376811e-05, "loss": 0.5706, "step": 21300 }, { "epoch": 0.11, "learning_rate": 9.917391304347827e-05, "loss": 0.576, "step": 21325 }, { "epoch": 0.11, "learning_rate": 9.899275362318841e-05, "loss": 0.6029, "step": 21350 }, { "epoch": 0.11, "learning_rate": 9.881159420289855e-05, "loss": 0.585, "step": 21375 }, { "epoch": 0.11, "learning_rate": 9.863043478260869e-05, "loss": 0.5899, "step": 21400 }, { "epoch": 0.11, "learning_rate": 9.844927536231885e-05, "loss": 0.563, "step": 21425 }, { "epoch": 0.11, "learning_rate": 9.826811594202898e-05, "loss": 0.5222, "step": 21450 }, { "epoch": 0.11, "learning_rate": 9.808695652173913e-05, "loss": 0.611, "step": 21475 }, { "epoch": 0.11, "learning_rate": 9.790579710144929e-05, "loss": 0.5361, "step": 21500 }, { "epoch": 0.11, "learning_rate": 9.772463768115942e-05, "loss": 0.5106, "step": 21525 }, { "epoch": 0.11, "learning_rate": 9.754347826086957e-05, "loss": 0.5718, "step": 21550 }, { "epoch": 0.11, "learning_rate": 9.73623188405797e-05, "loss": 0.4927, "step": 21575 }, { "epoch": 0.11, "learning_rate": 9.718115942028987e-05, "loss": 0.6652, "step": 21600 }, { "epoch": 0.11, "learning_rate": 9.7e-05, "loss": 0.5688, "step": 21625 }, { "epoch": 0.11, "learning_rate": 9.681884057971015e-05, "loss": 0.5386, "step": 21650 }, { "epoch": 0.11, "learning_rate": 9.66376811594203e-05, "loss": 0.5522, "step": 21675 }, { "epoch": 0.11, "learning_rate": 9.645652173913044e-05, "loss": 0.5484, "step": 21700 }, { "epoch": 0.11, "learning_rate": 9.627536231884059e-05, "loss": 0.6428, "step": 21725 }, { "epoch": 0.11, "learning_rate": 9.609420289855072e-05, "loss": 0.6201, "step": 21750 }, { "epoch": 0.11, "learning_rate": 9.591304347826087e-05, "loss": 0.5935, "step": 21775 }, { "epoch": 0.11, "learning_rate": 9.573188405797102e-05, "loss": 0.5543, "step": 21800 }, { "epoch": 0.11, "learning_rate": 9.555072463768116e-05, "loss": 0.6147, "step": 21825 }, { "epoch": 0.11, "learning_rate": 9.536956521739131e-05, "loss": 0.6359, "step": 21850 }, { "epoch": 0.11, "learning_rate": 9.518840579710144e-05, "loss": 0.607, "step": 21875 }, { "epoch": 0.11, "learning_rate": 9.50072463768116e-05, "loss": 0.5923, "step": 21900 }, { "epoch": 0.11, "learning_rate": 9.482608695652174e-05, "loss": 0.5105, "step": 21925 }, { "epoch": 0.11, "learning_rate": 9.464492753623189e-05, "loss": 0.604, "step": 21950 }, { "epoch": 0.11, "learning_rate": 9.446376811594202e-05, "loss": 0.5682, "step": 21975 }, { "epoch": 0.11, "learning_rate": 9.428260869565218e-05, "loss": 0.5353, "step": 22000 }, { "epoch": 0.11, "eval_loss": 0.5752917528152466, "eval_runtime": 6209.9534, "eval_samples_per_second": 0.806, "eval_steps_per_second": 0.202, "eval_wer": 39.12527587270699, "step": 22000 }, { "epoch": 0.11, "learning_rate": 9.410144927536233e-05, "loss": 0.5695, "step": 22025 }, { "epoch": 0.11, "learning_rate": 9.392028985507246e-05, "loss": 0.6238, "step": 22050 }, { "epoch": 0.11, "learning_rate": 9.373913043478261e-05, "loss": 0.5802, "step": 22075 }, { "epoch": 0.11, "learning_rate": 9.355797101449276e-05, "loss": 0.5519, "step": 22100 }, { "epoch": 0.11, "learning_rate": 9.33768115942029e-05, "loss": 0.5285, "step": 22125 }, { "epoch": 0.11, "learning_rate": 9.319565217391304e-05, "loss": 0.6007, "step": 22150 }, { "epoch": 0.11, "learning_rate": 9.301449275362318e-05, "loss": 0.6266, "step": 22175 }, { "epoch": 0.11, "learning_rate": 9.283333333333334e-05, "loss": 0.4953, "step": 22200 }, { "epoch": 0.11, "learning_rate": 9.265217391304348e-05, "loss": 0.5415, "step": 22225 }, { "epoch": 0.11, "learning_rate": 9.247101449275363e-05, "loss": 0.5891, "step": 22250 }, { "epoch": 0.11, "learning_rate": 9.228985507246376e-05, "loss": 0.5733, "step": 22275 }, { "epoch": 0.11, "learning_rate": 9.210869565217392e-05, "loss": 0.5113, "step": 22300 }, { "epoch": 0.11, "learning_rate": 9.192753623188405e-05, "loss": 0.5733, "step": 22325 }, { "epoch": 0.11, "learning_rate": 9.17463768115942e-05, "loss": 0.6071, "step": 22350 }, { "epoch": 0.11, "learning_rate": 9.156521739130436e-05, "loss": 0.6016, "step": 22375 }, { "epoch": 0.11, "learning_rate": 9.13840579710145e-05, "loss": 0.583, "step": 22400 }, { "epoch": 0.11, "learning_rate": 9.120289855072464e-05, "loss": 0.5587, "step": 22425 }, { "epoch": 0.11, "learning_rate": 9.102173913043478e-05, "loss": 0.5064, "step": 22450 }, { "epoch": 0.11, "learning_rate": 9.084057971014494e-05, "loss": 0.5962, "step": 22475 }, { "epoch": 0.11, "learning_rate": 9.065942028985507e-05, "loss": 0.6031, "step": 22500 }, { "epoch": 0.11, "learning_rate": 9.047826086956522e-05, "loss": 0.6156, "step": 22525 }, { "epoch": 0.11, "learning_rate": 9.029710144927536e-05, "loss": 0.5686, "step": 22550 }, { "epoch": 0.11, "learning_rate": 9.011594202898551e-05, "loss": 0.5362, "step": 22575 }, { "epoch": 0.11, "learning_rate": 8.993478260869566e-05, "loss": 0.6052, "step": 22600 }, { "epoch": 0.11, "learning_rate": 8.975362318840579e-05, "loss": 0.562, "step": 22625 }, { "epoch": 0.11, "learning_rate": 8.957246376811594e-05, "loss": 0.5782, "step": 22650 }, { "epoch": 0.11, "learning_rate": 8.939130434782609e-05, "loss": 0.5337, "step": 22675 }, { "epoch": 0.11, "learning_rate": 8.921014492753623e-05, "loss": 0.5238, "step": 22700 }, { "epoch": 0.11, "learning_rate": 8.902898550724638e-05, "loss": 0.5321, "step": 22725 }, { "epoch": 0.11, "learning_rate": 8.884782608695652e-05, "loss": 0.5133, "step": 22750 }, { "epoch": 0.11, "learning_rate": 8.866666666666668e-05, "loss": 0.6038, "step": 22775 }, { "epoch": 0.11, "learning_rate": 8.848550724637681e-05, "loss": 0.6108, "step": 22800 }, { "epoch": 0.11, "learning_rate": 8.830434782608696e-05, "loss": 0.5916, "step": 22825 }, { "epoch": 0.11, "learning_rate": 8.812318840579709e-05, "loss": 0.5767, "step": 22850 }, { "epoch": 0.11, "learning_rate": 8.794202898550725e-05, "loss": 0.601, "step": 22875 }, { "epoch": 0.11, "learning_rate": 8.77608695652174e-05, "loss": 0.5565, "step": 22900 }, { "epoch": 0.11, "learning_rate": 8.757971014492753e-05, "loss": 0.6307, "step": 22925 }, { "epoch": 0.12, "learning_rate": 8.739855072463768e-05, "loss": 0.5092, "step": 22950 }, { "epoch": 0.12, "learning_rate": 8.721739130434783e-05, "loss": 0.5775, "step": 22975 }, { "epoch": 0.12, "learning_rate": 8.703623188405797e-05, "loss": 0.5528, "step": 23000 }, { "epoch": 0.12, "eval_loss": 0.558778703212738, "eval_runtime": 6279.7061, "eval_samples_per_second": 0.797, "eval_steps_per_second": 0.199, "eval_wer": 40.28809416455065, "step": 23000 }, { "epoch": 0.12, "learning_rate": 8.685507246376812e-05, "loss": 0.5581, "step": 23025 }, { "epoch": 0.12, "learning_rate": 8.667391304347825e-05, "loss": 0.5841, "step": 23050 }, { "epoch": 0.12, "learning_rate": 8.649275362318842e-05, "loss": 0.5456, "step": 23075 }, { "epoch": 0.12, "learning_rate": 8.631159420289855e-05, "loss": 0.603, "step": 23100 }, { "epoch": 0.12, "learning_rate": 8.61304347826087e-05, "loss": 0.6094, "step": 23125 }, { "epoch": 0.12, "learning_rate": 8.594927536231884e-05, "loss": 0.468, "step": 23150 }, { "epoch": 0.12, "learning_rate": 8.576811594202899e-05, "loss": 0.63, "step": 23175 }, { "epoch": 0.12, "learning_rate": 8.558695652173914e-05, "loss": 0.6095, "step": 23200 }, { "epoch": 0.12, "learning_rate": 8.540579710144927e-05, "loss": 0.5802, "step": 23225 }, { "epoch": 0.12, "learning_rate": 8.522463768115943e-05, "loss": 0.5892, "step": 23250 }, { "epoch": 0.12, "learning_rate": 8.504347826086957e-05, "loss": 0.5736, "step": 23275 }, { "epoch": 0.12, "learning_rate": 8.486231884057971e-05, "loss": 0.4859, "step": 23300 }, { "epoch": 0.12, "learning_rate": 8.468115942028985e-05, "loss": 0.5839, "step": 23325 }, { "epoch": 0.12, "learning_rate": 8.450000000000001e-05, "loss": 0.5858, "step": 23350 }, { "epoch": 0.12, "learning_rate": 8.431884057971015e-05, "loss": 0.5604, "step": 23375 }, { "epoch": 0.12, "learning_rate": 8.413768115942029e-05, "loss": 0.5168, "step": 23400 }, { "epoch": 0.12, "learning_rate": 8.395652173913044e-05, "loss": 0.5481, "step": 23425 }, { "epoch": 0.12, "learning_rate": 8.377536231884058e-05, "loss": 0.4701, "step": 23450 }, { "epoch": 0.12, "learning_rate": 8.359420289855073e-05, "loss": 0.5258, "step": 23475 }, { "epoch": 0.12, "learning_rate": 8.341304347826086e-05, "loss": 0.5736, "step": 23500 }, { "epoch": 0.12, "learning_rate": 8.323188405797101e-05, "loss": 0.5622, "step": 23525 }, { "epoch": 0.12, "learning_rate": 8.305072463768117e-05, "loss": 0.5255, "step": 23550 }, { "epoch": 0.12, "learning_rate": 8.28695652173913e-05, "loss": 0.6043, "step": 23575 }, { "epoch": 0.12, "learning_rate": 8.268840579710145e-05, "loss": 0.8682, "step": 23600 }, { "epoch": 0.12, "learning_rate": 8.250724637681159e-05, "loss": 0.4914, "step": 23625 }, { "epoch": 0.12, "learning_rate": 8.232608695652175e-05, "loss": 0.5564, "step": 23650 }, { "epoch": 0.12, "learning_rate": 8.214492753623188e-05, "loss": 0.5782, "step": 23675 }, { "epoch": 0.12, "learning_rate": 8.196376811594203e-05, "loss": 0.5916, "step": 23700 }, { "epoch": 0.12, "learning_rate": 8.178260869565217e-05, "loss": 0.5447, "step": 23725 }, { "epoch": 0.12, "learning_rate": 8.160144927536232e-05, "loss": 0.514, "step": 23750 }, { "epoch": 0.12, "learning_rate": 8.142028985507247e-05, "loss": 0.5487, "step": 23775 }, { "epoch": 0.12, "learning_rate": 8.12391304347826e-05, "loss": 0.5429, "step": 23800 }, { "epoch": 0.12, "learning_rate": 8.105797101449275e-05, "loss": 0.4828, "step": 23825 }, { "epoch": 0.12, "learning_rate": 8.08768115942029e-05, "loss": 0.5015, "step": 23850 }, { "epoch": 0.12, "learning_rate": 8.069565217391304e-05, "loss": 0.517, "step": 23875 }, { "epoch": 0.12, "learning_rate": 8.051449275362319e-05, "loss": 0.5219, "step": 23900 }, { "epoch": 0.12, "learning_rate": 8.033333333333334e-05, "loss": 0.5493, "step": 23925 }, { "epoch": 0.12, "learning_rate": 8.015217391304349e-05, "loss": 0.4885, "step": 23950 }, { "epoch": 0.12, "learning_rate": 7.997101449275362e-05, "loss": 0.5623, "step": 23975 }, { "epoch": 0.12, "learning_rate": 7.978985507246377e-05, "loss": 0.5423, "step": 24000 }, { "epoch": 0.12, "eval_loss": 0.5444626212120056, "eval_runtime": 6160.1135, "eval_samples_per_second": 0.812, "eval_steps_per_second": 0.203, "eval_wer": 35.660551982723845, "step": 24000 }, { "epoch": 0.12, "learning_rate": 7.960869565217391e-05, "loss": 0.557, "step": 24025 }, { "epoch": 0.12, "learning_rate": 7.942753623188406e-05, "loss": 0.5214, "step": 24050 }, { "epoch": 0.12, "learning_rate": 7.924637681159421e-05, "loss": 0.5122, "step": 24075 }, { "epoch": 0.12, "learning_rate": 7.906521739130434e-05, "loss": 0.5814, "step": 24100 }, { "epoch": 0.12, "learning_rate": 7.88840579710145e-05, "loss": 0.5421, "step": 24125 }, { "epoch": 0.12, "learning_rate": 7.870289855072464e-05, "loss": 0.526, "step": 24150 }, { "epoch": 0.12, "learning_rate": 7.852173913043478e-05, "loss": 0.5424, "step": 24175 }, { "epoch": 0.12, "learning_rate": 7.834057971014492e-05, "loss": 0.5408, "step": 24200 }, { "epoch": 0.12, "learning_rate": 7.815942028985508e-05, "loss": 0.5618, "step": 24225 }, { "epoch": 0.12, "learning_rate": 7.797826086956523e-05, "loss": 0.5007, "step": 24250 }, { "epoch": 0.12, "learning_rate": 7.779710144927536e-05, "loss": 0.4739, "step": 24275 }, { "epoch": 0.12, "learning_rate": 7.76159420289855e-05, "loss": 0.5238, "step": 24300 }, { "epoch": 0.12, "learning_rate": 7.743478260869565e-05, "loss": 0.524, "step": 24325 }, { "epoch": 0.12, "learning_rate": 7.72536231884058e-05, "loss": 0.5054, "step": 24350 }, { "epoch": 0.12, "learning_rate": 7.707246376811595e-05, "loss": 0.5595, "step": 24375 }, { "epoch": 0.12, "learning_rate": 7.689130434782608e-05, "loss": 0.5144, "step": 24400 }, { "epoch": 0.12, "learning_rate": 7.671014492753624e-05, "loss": 0.5069, "step": 24425 }, { "epoch": 0.12, "learning_rate": 7.652898550724638e-05, "loss": 0.5121, "step": 24450 }, { "epoch": 0.12, "learning_rate": 7.634782608695652e-05, "loss": 0.5281, "step": 24475 }, { "epoch": 0.12, "learning_rate": 7.616666666666666e-05, "loss": 0.5126, "step": 24500 }, { "epoch": 0.12, "learning_rate": 7.598550724637682e-05, "loss": 0.6006, "step": 24525 }, { "epoch": 0.12, "learning_rate": 7.580434782608696e-05, "loss": 0.5609, "step": 24550 }, { "epoch": 0.12, "learning_rate": 7.56231884057971e-05, "loss": 0.5247, "step": 24575 }, { "epoch": 0.12, "learning_rate": 7.544202898550725e-05, "loss": 0.5286, "step": 24600 }, { "epoch": 0.12, "learning_rate": 7.526086956521739e-05, "loss": 0.5367, "step": 24625 }, { "epoch": 0.12, "learning_rate": 7.507971014492754e-05, "loss": 0.6143, "step": 24650 }, { "epoch": 0.12, "learning_rate": 7.489855072463767e-05, "loss": 0.579, "step": 24675 }, { "epoch": 0.12, "learning_rate": 7.471739130434783e-05, "loss": 0.5103, "step": 24700 }, { "epoch": 0.12, "learning_rate": 7.453623188405798e-05, "loss": 0.5062, "step": 24725 }, { "epoch": 0.12, "learning_rate": 7.435507246376812e-05, "loss": 0.5085, "step": 24750 }, { "epoch": 0.12, "learning_rate": 7.417391304347826e-05, "loss": 0.6252, "step": 24775 }, { "epoch": 0.12, "learning_rate": 7.399275362318841e-05, "loss": 0.5466, "step": 24800 }, { "epoch": 0.12, "learning_rate": 7.381159420289856e-05, "loss": 0.5409, "step": 24825 }, { "epoch": 0.12, "learning_rate": 7.363043478260869e-05, "loss": 0.5708, "step": 24850 }, { "epoch": 0.12, "learning_rate": 7.344927536231884e-05, "loss": 0.4924, "step": 24875 }, { "epoch": 0.12, "learning_rate": 7.327536231884057e-05, "loss": 0.5049, "step": 24900 }, { "epoch": 0.12, "learning_rate": 7.309420289855073e-05, "loss": 0.5233, "step": 24925 }, { "epoch": 0.13, "learning_rate": 7.291304347826088e-05, "loss": 0.5151, "step": 24950 }, { "epoch": 0.13, "learning_rate": 7.273188405797101e-05, "loss": 0.4914, "step": 24975 }, { "epoch": 0.13, "learning_rate": 7.255072463768116e-05, "loss": 0.5069, "step": 25000 }, { "epoch": 0.13, "eval_loss": 0.5304020643234253, "eval_runtime": 6200.0725, "eval_samples_per_second": 0.807, "eval_steps_per_second": 0.202, "eval_wer": 35.93583141507867, "step": 25000 }, { "epoch": 0.13, "learning_rate": 7.236956521739131e-05, "loss": 0.5066, "step": 25025 }, { "epoch": 0.13, "learning_rate": 7.218840579710145e-05, "loss": 0.5338, "step": 25050 }, { "epoch": 0.13, "learning_rate": 7.200724637681159e-05, "loss": 0.4938, "step": 25075 }, { "epoch": 0.13, "learning_rate": 7.182608695652174e-05, "loss": 0.4742, "step": 25100 }, { "epoch": 0.13, "learning_rate": 7.16449275362319e-05, "loss": 0.439, "step": 25125 }, { "epoch": 0.13, "learning_rate": 7.146376811594203e-05, "loss": 0.5899, "step": 25150 }, { "epoch": 0.13, "learning_rate": 7.128260869565218e-05, "loss": 0.4483, "step": 25175 }, { "epoch": 0.13, "learning_rate": 7.110144927536231e-05, "loss": 0.5661, "step": 25200 }, { "epoch": 0.13, "learning_rate": 7.092028985507247e-05, "loss": 0.5919, "step": 25225 }, { "epoch": 0.13, "learning_rate": 7.07391304347826e-05, "loss": 0.4567, "step": 25250 }, { "epoch": 0.13, "learning_rate": 7.055797101449275e-05, "loss": 0.4964, "step": 25275 }, { "epoch": 0.13, "learning_rate": 7.03768115942029e-05, "loss": 0.5314, "step": 25300 }, { "epoch": 0.13, "learning_rate": 7.019565217391305e-05, "loss": 0.5589, "step": 25325 }, { "epoch": 0.13, "learning_rate": 7.00144927536232e-05, "loss": 0.5106, "step": 25350 }, { "epoch": 0.13, "learning_rate": 6.983333333333333e-05, "loss": 0.5249, "step": 25375 }, { "epoch": 0.13, "learning_rate": 6.965217391304347e-05, "loss": 0.5737, "step": 25400 }, { "epoch": 0.13, "learning_rate": 6.947101449275362e-05, "loss": 0.5539, "step": 25425 }, { "epoch": 0.13, "learning_rate": 6.928985507246377e-05, "loss": 0.5074, "step": 25450 }, { "epoch": 0.13, "learning_rate": 6.910869565217392e-05, "loss": 0.5577, "step": 25475 }, { "epoch": 0.13, "learning_rate": 6.892753623188405e-05, "loss": 0.4824, "step": 25500 }, { "epoch": 0.13, "learning_rate": 6.874637681159421e-05, "loss": 0.5336, "step": 25525 }, { "epoch": 0.13, "learning_rate": 6.856521739130434e-05, "loss": 0.5886, "step": 25550 }, { "epoch": 0.13, "learning_rate": 6.838405797101449e-05, "loss": 0.5536, "step": 25575 }, { "epoch": 0.13, "learning_rate": 6.820289855072465e-05, "loss": 0.5087, "step": 25600 }, { "epoch": 0.13, "learning_rate": 6.802173913043479e-05, "loss": 0.5165, "step": 25625 }, { "epoch": 0.13, "learning_rate": 6.784057971014493e-05, "loss": 0.5782, "step": 25650 }, { "epoch": 0.13, "learning_rate": 6.765942028985507e-05, "loss": 0.5582, "step": 25675 }, { "epoch": 0.13, "learning_rate": 6.747826086956523e-05, "loss": 0.5829, "step": 25700 }, { "epoch": 0.13, "learning_rate": 6.729710144927536e-05, "loss": 0.4834, "step": 25725 }, { "epoch": 0.13, "learning_rate": 6.711594202898551e-05, "loss": 0.5246, "step": 25750 }, { "epoch": 0.13, "learning_rate": 6.693478260869566e-05, "loss": 0.5443, "step": 25775 }, { "epoch": 0.13, "learning_rate": 6.67536231884058e-05, "loss": 0.5, "step": 25800 }, { "epoch": 0.13, "learning_rate": 6.657246376811595e-05, "loss": 0.5199, "step": 25825 }, { "epoch": 0.13, "learning_rate": 6.639130434782608e-05, "loss": 0.4944, "step": 25850 }, { "epoch": 0.13, "learning_rate": 6.621014492753623e-05, "loss": 0.5081, "step": 25875 }, { "epoch": 0.13, "learning_rate": 6.602898550724638e-05, "loss": 0.5663, "step": 25900 }, { "epoch": 0.13, "learning_rate": 6.584782608695653e-05, "loss": 0.4847, "step": 25925 }, { "epoch": 0.13, "learning_rate": 6.566666666666667e-05, "loss": 0.486, "step": 25950 }, { "epoch": 0.13, "learning_rate": 6.54855072463768e-05, "loss": 0.5484, "step": 25975 }, { "epoch": 0.13, "learning_rate": 6.530434782608697e-05, "loss": 0.4356, "step": 26000 }, { "epoch": 0.13, "eval_loss": 0.5187134742736816, "eval_runtime": 6155.9894, "eval_samples_per_second": 0.813, "eval_steps_per_second": 0.203, "eval_wer": 34.49298749377061, "step": 26000 }, { "epoch": 0.13, "learning_rate": 6.51231884057971e-05, "loss": 0.4864, "step": 26025 }, { "epoch": 0.13, "learning_rate": 6.494202898550725e-05, "loss": 0.5086, "step": 26050 }, { "epoch": 0.13, "learning_rate": 6.476086956521738e-05, "loss": 0.5245, "step": 26075 }, { "epoch": 0.13, "learning_rate": 6.457971014492754e-05, "loss": 0.4574, "step": 26100 }, { "epoch": 0.13, "learning_rate": 6.439855072463769e-05, "loss": 0.5071, "step": 26125 }, { "epoch": 0.13, "learning_rate": 6.421739130434782e-05, "loss": 0.5297, "step": 26150 }, { "epoch": 0.13, "learning_rate": 6.403623188405797e-05, "loss": 0.5426, "step": 26175 }, { "epoch": 0.13, "learning_rate": 6.385507246376812e-05, "loss": 0.4716, "step": 26200 }, { "epoch": 0.13, "learning_rate": 6.367391304347826e-05, "loss": 0.4649, "step": 26225 }, { "epoch": 0.13, "learning_rate": 6.34927536231884e-05, "loss": 0.4876, "step": 26250 }, { "epoch": 0.13, "learning_rate": 6.331159420289855e-05, "loss": 0.4955, "step": 26275 }, { "epoch": 0.13, "learning_rate": 6.31304347826087e-05, "loss": 0.4935, "step": 26300 }, { "epoch": 0.13, "learning_rate": 6.294927536231884e-05, "loss": 0.5393, "step": 26325 }, { "epoch": 0.13, "learning_rate": 6.276811594202899e-05, "loss": 0.5106, "step": 26350 }, { "epoch": 0.13, "learning_rate": 6.258695652173913e-05, "loss": 0.4791, "step": 26375 }, { "epoch": 0.13, "learning_rate": 6.240579710144928e-05, "loss": 0.4947, "step": 26400 }, { "epoch": 0.13, "learning_rate": 6.222463768115942e-05, "loss": 0.5272, "step": 26425 }, { "epoch": 0.13, "learning_rate": 6.204347826086958e-05, "loss": 0.5317, "step": 26450 }, { "epoch": 0.13, "learning_rate": 6.186231884057971e-05, "loss": 0.541, "step": 26475 }, { "epoch": 0.13, "learning_rate": 6.168115942028986e-05, "loss": 0.4706, "step": 26500 }, { "epoch": 0.13, "learning_rate": 6.15e-05, "loss": 0.4843, "step": 26525 }, { "epoch": 0.13, "learning_rate": 6.131884057971015e-05, "loss": 0.55, "step": 26550 }, { "epoch": 0.13, "learning_rate": 6.113768115942028e-05, "loss": 0.5025, "step": 26575 }, { "epoch": 0.13, "learning_rate": 6.095652173913044e-05, "loss": 0.5481, "step": 26600 }, { "epoch": 0.13, "learning_rate": 6.077536231884058e-05, "loss": 0.5372, "step": 26625 }, { "epoch": 0.13, "learning_rate": 6.0594202898550727e-05, "loss": 0.4827, "step": 26650 }, { "epoch": 0.13, "learning_rate": 6.041304347826087e-05, "loss": 0.5627, "step": 26675 }, { "epoch": 0.13, "learning_rate": 6.023913043478261e-05, "loss": 0.6139, "step": 26700 }, { "epoch": 0.13, "learning_rate": 6.0057971014492755e-05, "loss": 0.486, "step": 26725 }, { "epoch": 0.13, "learning_rate": 5.9876811594202895e-05, "loss": 0.4263, "step": 26750 }, { "epoch": 0.13, "learning_rate": 5.969565217391305e-05, "loss": 0.5085, "step": 26775 }, { "epoch": 0.13, "learning_rate": 5.951449275362319e-05, "loss": 0.5375, "step": 26800 }, { "epoch": 0.13, "learning_rate": 5.933333333333334e-05, "loss": 0.46, "step": 26825 }, { "epoch": 0.13, "learning_rate": 5.915217391304348e-05, "loss": 0.4551, "step": 26850 }, { "epoch": 0.13, "learning_rate": 5.8971014492753624e-05, "loss": 0.4416, "step": 26875 }, { "epoch": 0.13, "learning_rate": 5.8789855072463765e-05, "loss": 0.4306, "step": 26900 }, { "epoch": 0.13, "learning_rate": 5.860869565217392e-05, "loss": 0.4464, "step": 26925 }, { "epoch": 0.14, "learning_rate": 5.842753623188406e-05, "loss": 0.5108, "step": 26950 }, { "epoch": 0.14, "learning_rate": 5.82536231884058e-05, "loss": 0.5432, "step": 26975 }, { "epoch": 0.14, "learning_rate": 5.807246376811595e-05, "loss": 0.5111, "step": 27000 }, { "epoch": 0.14, "eval_loss": 0.5034857392311096, "eval_runtime": 6134.0304, "eval_samples_per_second": 0.816, "eval_steps_per_second": 0.204, "eval_wer": 33.422720045563494, "step": 27000 }, { "epoch": 0.14, "learning_rate": 5.789130434782609e-05, "loss": 0.4599, "step": 27025 }, { "epoch": 0.14, "learning_rate": 5.7710144927536235e-05, "loss": 0.4427, "step": 27050 }, { "epoch": 0.14, "learning_rate": 5.7528985507246375e-05, "loss": 0.5067, "step": 27075 }, { "epoch": 0.14, "learning_rate": 5.734782608695652e-05, "loss": 0.4341, "step": 27100 }, { "epoch": 0.14, "learning_rate": 5.716666666666666e-05, "loss": 0.4668, "step": 27125 }, { "epoch": 0.14, "learning_rate": 5.6985507246376817e-05, "loss": 0.4743, "step": 27150 }, { "epoch": 0.14, "learning_rate": 5.680434782608696e-05, "loss": 0.488, "step": 27175 }, { "epoch": 0.14, "learning_rate": 5.6623188405797104e-05, "loss": 0.4982, "step": 27200 }, { "epoch": 0.14, "learning_rate": 5.6442028985507245e-05, "loss": 0.5173, "step": 27225 }, { "epoch": 0.14, "learning_rate": 5.626086956521739e-05, "loss": 0.5259, "step": 27250 }, { "epoch": 0.14, "learning_rate": 5.607971014492753e-05, "loss": 0.4459, "step": 27275 }, { "epoch": 0.14, "learning_rate": 5.589855072463768e-05, "loss": 0.4874, "step": 27300 }, { "epoch": 0.14, "learning_rate": 5.571739130434783e-05, "loss": 0.4779, "step": 27325 }, { "epoch": 0.14, "learning_rate": 5.5536231884057974e-05, "loss": 0.4924, "step": 27350 }, { "epoch": 0.14, "learning_rate": 5.535507246376812e-05, "loss": 0.4998, "step": 27375 }, { "epoch": 0.14, "learning_rate": 5.517391304347826e-05, "loss": 0.4812, "step": 27400 }, { "epoch": 0.14, "learning_rate": 5.499275362318841e-05, "loss": 0.5353, "step": 27425 }, { "epoch": 0.14, "learning_rate": 5.481159420289855e-05, "loss": 0.4757, "step": 27450 }, { "epoch": 0.14, "learning_rate": 5.46304347826087e-05, "loss": 0.455, "step": 27475 }, { "epoch": 0.14, "learning_rate": 5.4449275362318843e-05, "loss": 0.4996, "step": 27500 }, { "epoch": 0.14, "learning_rate": 5.426811594202899e-05, "loss": 0.4718, "step": 27525 }, { "epoch": 0.14, "learning_rate": 5.408695652173913e-05, "loss": 0.4887, "step": 27550 }, { "epoch": 0.14, "learning_rate": 5.390579710144928e-05, "loss": 0.5006, "step": 27575 }, { "epoch": 0.14, "learning_rate": 5.372463768115942e-05, "loss": 0.5329, "step": 27600 }, { "epoch": 0.14, "learning_rate": 5.3543478260869566e-05, "loss": 0.4868, "step": 27625 }, { "epoch": 0.14, "learning_rate": 5.336231884057971e-05, "loss": 0.487, "step": 27650 }, { "epoch": 0.14, "learning_rate": 5.318115942028986e-05, "loss": 0.5835, "step": 27675 }, { "epoch": 0.14, "learning_rate": 5.3e-05, "loss": 0.4904, "step": 27700 }, { "epoch": 0.14, "learning_rate": 5.281884057971015e-05, "loss": 0.5261, "step": 27725 }, { "epoch": 0.14, "learning_rate": 5.263768115942029e-05, "loss": 0.4581, "step": 27750 }, { "epoch": 0.14, "learning_rate": 5.2456521739130436e-05, "loss": 0.4993, "step": 27775 }, { "epoch": 0.14, "learning_rate": 5.2275362318840576e-05, "loss": 0.5257, "step": 27800 }, { "epoch": 0.14, "learning_rate": 5.209420289855073e-05, "loss": 0.4443, "step": 27825 }, { "epoch": 0.14, "learning_rate": 5.191304347826087e-05, "loss": 0.5031, "step": 27850 }, { "epoch": 0.14, "learning_rate": 5.173188405797102e-05, "loss": 0.475, "step": 27875 }, { "epoch": 0.14, "learning_rate": 5.155072463768116e-05, "loss": 0.52, "step": 27900 }, { "epoch": 0.14, "learning_rate": 5.1369565217391305e-05, "loss": 0.5038, "step": 27925 }, { "epoch": 0.14, "learning_rate": 5.1188405797101446e-05, "loss": 0.4803, "step": 27950 }, { "epoch": 0.14, "learning_rate": 5.100724637681159e-05, "loss": 0.493, "step": 27975 }, { "epoch": 0.14, "learning_rate": 5.082608695652174e-05, "loss": 0.5613, "step": 28000 }, { "epoch": 0.14, "eval_loss": 0.4912257194519043, "eval_runtime": 6183.2879, "eval_samples_per_second": 0.809, "eval_steps_per_second": 0.202, "eval_wer": 33.09523244500344, "step": 28000 }, { "epoch": 0.14, "learning_rate": 5.064492753623189e-05, "loss": 0.4801, "step": 28025 }, { "epoch": 0.14, "learning_rate": 5.046376811594203e-05, "loss": 0.4915, "step": 28050 }, { "epoch": 0.14, "learning_rate": 5.0282608695652175e-05, "loss": 0.4282, "step": 28075 }, { "epoch": 0.14, "learning_rate": 5.0101449275362315e-05, "loss": 0.4381, "step": 28100 }, { "epoch": 0.14, "learning_rate": 4.992028985507246e-05, "loss": 0.4421, "step": 28125 }, { "epoch": 0.14, "learning_rate": 4.9739130434782617e-05, "loss": 0.4561, "step": 28150 }, { "epoch": 0.14, "learning_rate": 4.955797101449276e-05, "loss": 0.5123, "step": 28175 }, { "epoch": 0.14, "learning_rate": 4.9376811594202904e-05, "loss": 0.5428, "step": 28200 }, { "epoch": 0.14, "learning_rate": 4.9195652173913045e-05, "loss": 0.4695, "step": 28225 }, { "epoch": 0.14, "learning_rate": 4.901449275362319e-05, "loss": 0.4128, "step": 28250 }, { "epoch": 0.14, "learning_rate": 4.883333333333333e-05, "loss": 0.4851, "step": 28275 }, { "epoch": 0.14, "learning_rate": 4.865217391304348e-05, "loss": 0.4081, "step": 28300 }, { "epoch": 0.14, "learning_rate": 4.847101449275363e-05, "loss": 0.4893, "step": 28325 }, { "epoch": 0.14, "learning_rate": 4.8289855072463774e-05, "loss": 0.4521, "step": 28350 }, { "epoch": 0.14, "learning_rate": 4.8108695652173914e-05, "loss": 0.5489, "step": 28375 }, { "epoch": 0.14, "learning_rate": 4.792753623188406e-05, "loss": 0.5304, "step": 28400 }, { "epoch": 0.14, "learning_rate": 4.77463768115942e-05, "loss": 0.4512, "step": 28425 }, { "epoch": 0.14, "learning_rate": 4.756521739130435e-05, "loss": 0.4725, "step": 28450 }, { "epoch": 0.14, "learning_rate": 4.738405797101449e-05, "loss": 0.5036, "step": 28475 }, { "epoch": 0.14, "learning_rate": 4.7202898550724643e-05, "loss": 0.4832, "step": 28500 }, { "epoch": 0.14, "learning_rate": 4.7021739130434784e-05, "loss": 0.534, "step": 28525 }, { "epoch": 0.14, "learning_rate": 4.684057971014493e-05, "loss": 0.4533, "step": 28550 }, { "epoch": 0.14, "learning_rate": 4.665942028985507e-05, "loss": 0.5385, "step": 28575 }, { "epoch": 0.14, "learning_rate": 4.647826086956522e-05, "loss": 0.4935, "step": 28600 }, { "epoch": 0.14, "learning_rate": 4.629710144927536e-05, "loss": 0.443, "step": 28625 }, { "epoch": 0.14, "learning_rate": 4.6115942028985506e-05, "loss": 0.4324, "step": 28650 }, { "epoch": 0.14, "learning_rate": 4.5934782608695654e-05, "loss": 0.4784, "step": 28675 }, { "epoch": 0.14, "learning_rate": 4.57536231884058e-05, "loss": 0.5117, "step": 28700 }, { "epoch": 0.14, "learning_rate": 4.557246376811594e-05, "loss": 0.4398, "step": 28725 }, { "epoch": 0.14, "learning_rate": 4.539130434782609e-05, "loss": 0.4913, "step": 28750 }, { "epoch": 0.14, "learning_rate": 4.521014492753623e-05, "loss": 0.4358, "step": 28775 }, { "epoch": 0.14, "learning_rate": 4.5028985507246376e-05, "loss": 0.4604, "step": 28800 }, { "epoch": 0.14, "learning_rate": 4.4847826086956516e-05, "loss": 0.4732, "step": 28825 }, { "epoch": 0.14, "learning_rate": 4.466666666666667e-05, "loss": 0.4656, "step": 28850 }, { "epoch": 0.14, "learning_rate": 4.448550724637681e-05, "loss": 0.4768, "step": 28875 }, { "epoch": 0.14, "learning_rate": 4.430434782608696e-05, "loss": 0.4306, "step": 28900 }, { "epoch": 0.14, "learning_rate": 4.41231884057971e-05, "loss": 0.4821, "step": 28925 }, { "epoch": 0.15, "learning_rate": 4.3942028985507246e-05, "loss": 0.538, "step": 28950 }, { "epoch": 0.15, "learning_rate": 4.376086956521739e-05, "loss": 0.4845, "step": 28975 }, { "epoch": 0.15, "learning_rate": 4.357971014492754e-05, "loss": 0.4165, "step": 29000 }, { "epoch": 0.15, "eval_loss": 0.48251873254776, "eval_runtime": 6175.4519, "eval_samples_per_second": 0.81, "eval_steps_per_second": 0.203, "eval_wer": 32.015472602577184, "step": 29000 }, { "epoch": 0.15, "learning_rate": 4.339855072463769e-05, "loss": 0.48, "step": 29025 }, { "epoch": 0.15, "learning_rate": 4.321739130434783e-05, "loss": 0.5295, "step": 29050 }, { "epoch": 0.15, "learning_rate": 4.3036231884057975e-05, "loss": 0.4944, "step": 29075 }, { "epoch": 0.15, "learning_rate": 4.2855072463768115e-05, "loss": 0.4394, "step": 29100 }, { "epoch": 0.15, "learning_rate": 4.267391304347826e-05, "loss": 0.5057, "step": 29125 }, { "epoch": 0.15, "learning_rate": 4.24927536231884e-05, "loss": 0.5016, "step": 29150 }, { "epoch": 0.15, "learning_rate": 4.231159420289856e-05, "loss": 0.3946, "step": 29175 }, { "epoch": 0.15, "learning_rate": 4.21304347826087e-05, "loss": 0.5014, "step": 29200 }, { "epoch": 0.15, "learning_rate": 4.1949275362318845e-05, "loss": 0.5253, "step": 29225 }, { "epoch": 0.15, "learning_rate": 4.1768115942028985e-05, "loss": 0.4362, "step": 29250 }, { "epoch": 0.15, "learning_rate": 4.158695652173913e-05, "loss": 0.5088, "step": 29275 }, { "epoch": 0.15, "learning_rate": 4.140579710144927e-05, "loss": 0.4703, "step": 29300 }, { "epoch": 0.15, "learning_rate": 4.122463768115942e-05, "loss": 0.503, "step": 29325 }, { "epoch": 0.15, "learning_rate": 4.104347826086957e-05, "loss": 0.4592, "step": 29350 }, { "epoch": 0.15, "learning_rate": 4.0862318840579714e-05, "loss": 0.4553, "step": 29375 }, { "epoch": 0.15, "learning_rate": 4.0681159420289855e-05, "loss": 0.5688, "step": 29400 }, { "epoch": 0.15, "learning_rate": 4.05e-05, "loss": 0.4286, "step": 29425 }, { "epoch": 0.15, "learning_rate": 4.031884057971014e-05, "loss": 0.5024, "step": 29450 }, { "epoch": 0.15, "learning_rate": 4.013768115942029e-05, "loss": 0.5156, "step": 29475 }, { "epoch": 0.15, "learning_rate": 3.995652173913043e-05, "loss": 0.4684, "step": 29500 }, { "epoch": 0.15, "learning_rate": 3.9775362318840584e-05, "loss": 0.4606, "step": 29525 }, { "epoch": 0.15, "learning_rate": 3.9594202898550724e-05, "loss": 0.4924, "step": 29550 }, { "epoch": 0.15, "learning_rate": 3.941304347826087e-05, "loss": 0.4926, "step": 29575 }, { "epoch": 0.15, "learning_rate": 3.923188405797101e-05, "loss": 0.4257, "step": 29600 }, { "epoch": 0.15, "learning_rate": 3.905072463768116e-05, "loss": 0.5036, "step": 29625 }, { "epoch": 0.15, "learning_rate": 3.88695652173913e-05, "loss": 0.5142, "step": 29650 }, { "epoch": 0.15, "learning_rate": 3.8688405797101454e-05, "loss": 0.5144, "step": 29675 }, { "epoch": 0.15, "learning_rate": 3.8507246376811594e-05, "loss": 0.4234, "step": 29700 }, { "epoch": 0.15, "learning_rate": 3.832608695652174e-05, "loss": 0.4776, "step": 29725 }, { "epoch": 0.15, "learning_rate": 3.814492753623189e-05, "loss": 0.574, "step": 29750 }, { "epoch": 0.15, "learning_rate": 3.796376811594203e-05, "loss": 0.451, "step": 29775 }, { "epoch": 0.15, "learning_rate": 3.7782608695652176e-05, "loss": 0.478, "step": 29800 }, { "epoch": 0.15, "learning_rate": 3.7601449275362316e-05, "loss": 0.4856, "step": 29825 }, { "epoch": 0.15, "learning_rate": 3.742028985507247e-05, "loss": 0.5041, "step": 29850 }, { "epoch": 0.15, "learning_rate": 3.723913043478261e-05, "loss": 0.4875, "step": 29875 }, { "epoch": 0.15, "learning_rate": 3.705797101449276e-05, "loss": 0.4526, "step": 29900 }, { "epoch": 0.15, "learning_rate": 3.68768115942029e-05, "loss": 0.4811, "step": 29925 }, { "epoch": 0.15, "learning_rate": 3.6695652173913046e-05, "loss": 0.4831, "step": 29950 }, { "epoch": 0.15, "learning_rate": 3.6514492753623186e-05, "loss": 0.4735, "step": 29975 }, { "epoch": 0.15, "learning_rate": 3.633333333333333e-05, "loss": 0.4736, "step": 30000 }, { "epoch": 0.15, "eval_loss": 0.47158339619636536, "eval_runtime": 6176.7504, "eval_samples_per_second": 0.81, "eval_steps_per_second": 0.203, "eval_wer": 32.09141175633024, "step": 30000 }, { "epoch": 0.15, "learning_rate": 3.615217391304348e-05, "loss": 0.5359, "step": 30025 }, { "epoch": 0.15, "learning_rate": 3.597101449275363e-05, "loss": 0.4393, "step": 30050 }, { "epoch": 0.15, "learning_rate": 3.578985507246377e-05, "loss": 0.527, "step": 30075 }, { "epoch": 0.15, "learning_rate": 3.5608695652173915e-05, "loss": 0.4477, "step": 30100 }, { "epoch": 0.15, "learning_rate": 3.5427536231884056e-05, "loss": 0.4318, "step": 30125 }, { "epoch": 0.15, "learning_rate": 3.52463768115942e-05, "loss": 0.4479, "step": 30150 }, { "epoch": 0.15, "learning_rate": 3.5065217391304343e-05, "loss": 0.543, "step": 30175 }, { "epoch": 0.15, "learning_rate": 3.48840579710145e-05, "loss": 0.421, "step": 30200 }, { "epoch": 0.15, "learning_rate": 3.470289855072464e-05, "loss": 0.4694, "step": 30225 }, { "epoch": 0.15, "learning_rate": 3.4521739130434785e-05, "loss": 0.4164, "step": 30250 }, { "epoch": 0.15, "learning_rate": 3.4340579710144925e-05, "loss": 0.459, "step": 30275 }, { "epoch": 0.15, "learning_rate": 3.415942028985507e-05, "loss": 0.454, "step": 30300 }, { "epoch": 0.15, "learning_rate": 3.397826086956521e-05, "loss": 0.5195, "step": 30325 }, { "epoch": 0.15, "learning_rate": 3.379710144927537e-05, "loss": 0.445, "step": 30350 }, { "epoch": 0.15, "learning_rate": 3.361594202898551e-05, "loss": 0.5167, "step": 30375 }, { "epoch": 0.15, "learning_rate": 3.3434782608695655e-05, "loss": 0.5422, "step": 30400 }, { "epoch": 0.15, "learning_rate": 3.3253623188405795e-05, "loss": 0.5054, "step": 30425 }, { "epoch": 0.15, "learning_rate": 3.307246376811594e-05, "loss": 0.4505, "step": 30450 }, { "epoch": 0.15, "learning_rate": 3.289130434782608e-05, "loss": 0.3841, "step": 30475 }, { "epoch": 0.15, "learning_rate": 3.271014492753623e-05, "loss": 0.4823, "step": 30500 }, { "epoch": 0.15, "learning_rate": 3.2528985507246384e-05, "loss": 0.4393, "step": 30525 }, { "epoch": 0.15, "learning_rate": 3.2347826086956524e-05, "loss": 0.3793, "step": 30550 }, { "epoch": 0.15, "learning_rate": 3.216666666666667e-05, "loss": 0.4372, "step": 30575 }, { "epoch": 0.15, "learning_rate": 3.198550724637681e-05, "loss": 0.4921, "step": 30600 }, { "epoch": 0.15, "learning_rate": 3.180434782608696e-05, "loss": 0.4891, "step": 30625 }, { "epoch": 0.15, "learning_rate": 3.16231884057971e-05, "loss": 0.4375, "step": 30650 }, { "epoch": 0.15, "learning_rate": 3.144202898550725e-05, "loss": 0.4652, "step": 30675 }, { "epoch": 0.15, "learning_rate": 3.1260869565217394e-05, "loss": 0.4635, "step": 30700 }, { "epoch": 0.15, "learning_rate": 3.1079710144927534e-05, "loss": 0.4937, "step": 30725 }, { "epoch": 0.15, "learning_rate": 3.089855072463768e-05, "loss": 0.4578, "step": 30750 }, { "epoch": 0.15, "learning_rate": 3.071739130434783e-05, "loss": 0.4976, "step": 30775 }, { "epoch": 0.15, "learning_rate": 3.053623188405797e-05, "loss": 0.4287, "step": 30800 }, { "epoch": 0.15, "learning_rate": 3.0355072463768116e-05, "loss": 0.4332, "step": 30825 }, { "epoch": 0.15, "learning_rate": 3.017391304347826e-05, "loss": 0.4527, "step": 30850 }, { "epoch": 0.15, "learning_rate": 2.9992753623188404e-05, "loss": 0.5016, "step": 30875 }, { "epoch": 0.15, "learning_rate": 2.9811594202898555e-05, "loss": 0.425, "step": 30900 }, { "epoch": 0.16, "learning_rate": 2.96304347826087e-05, "loss": 0.419, "step": 30925 }, { "epoch": 0.16, "learning_rate": 2.9449275362318842e-05, "loss": 0.4652, "step": 30950 }, { "epoch": 0.16, "learning_rate": 2.9268115942028986e-05, "loss": 0.4079, "step": 30975 }, { "epoch": 0.16, "learning_rate": 2.9086956521739133e-05, "loss": 0.4213, "step": 31000 }, { "epoch": 0.16, "eval_loss": 0.4618246555328369, "eval_runtime": 6209.3649, "eval_samples_per_second": 0.806, "eval_steps_per_second": 0.202, "eval_wer": 31.602553454044944, "step": 31000 }, { "epoch": 0.16, "learning_rate": 2.8905797101449277e-05, "loss": 0.4598, "step": 31025 }, { "epoch": 0.16, "learning_rate": 2.872463768115942e-05, "loss": 0.4848, "step": 31050 }, { "epoch": 0.16, "learning_rate": 2.8543478260869568e-05, "loss": 0.4737, "step": 31075 }, { "epoch": 0.16, "learning_rate": 2.8362318840579712e-05, "loss": 0.41, "step": 31100 }, { "epoch": 0.16, "learning_rate": 2.8181159420289856e-05, "loss": 0.5066, "step": 31125 }, { "epoch": 0.16, "learning_rate": 2.8e-05, "loss": 0.4915, "step": 31150 }, { "epoch": 0.16, "learning_rate": 2.7818840579710147e-05, "loss": 0.4951, "step": 31175 }, { "epoch": 0.16, "learning_rate": 2.763768115942029e-05, "loss": 0.4934, "step": 31200 }, { "epoch": 0.16, "learning_rate": 2.7456521739130434e-05, "loss": 0.4841, "step": 31225 }, { "epoch": 0.16, "learning_rate": 2.727536231884058e-05, "loss": 0.4512, "step": 31250 }, { "epoch": 0.16, "learning_rate": 2.7094202898550725e-05, "loss": 0.4406, "step": 31275 }, { "epoch": 0.16, "learning_rate": 2.691304347826087e-05, "loss": 0.4719, "step": 31300 }, { "epoch": 0.16, "learning_rate": 2.6731884057971016e-05, "loss": 0.4947, "step": 31325 }, { "epoch": 0.16, "learning_rate": 2.655072463768116e-05, "loss": 0.4436, "step": 31350 }, { "epoch": 0.16, "learning_rate": 2.6369565217391304e-05, "loss": 0.4927, "step": 31375 }, { "epoch": 0.16, "learning_rate": 2.6188405797101448e-05, "loss": 0.44, "step": 31400 }, { "epoch": 0.16, "learning_rate": 2.6007246376811595e-05, "loss": 0.4541, "step": 31425 }, { "epoch": 0.16, "learning_rate": 2.582608695652174e-05, "loss": 0.4889, "step": 31450 }, { "epoch": 0.16, "learning_rate": 2.5644927536231883e-05, "loss": 0.4327, "step": 31475 }, { "epoch": 0.16, "learning_rate": 2.546376811594203e-05, "loss": 0.4534, "step": 31500 }, { "epoch": 0.16, "learning_rate": 2.5282608695652174e-05, "loss": 0.4646, "step": 31525 }, { "epoch": 0.16, "learning_rate": 2.5101449275362318e-05, "loss": 0.4756, "step": 31550 }, { "epoch": 0.16, "learning_rate": 2.492028985507246e-05, "loss": 0.4588, "step": 31575 }, { "epoch": 0.16, "learning_rate": 2.473913043478261e-05, "loss": 0.4061, "step": 31600 }, { "epoch": 0.16, "learning_rate": 2.4557971014492752e-05, "loss": 0.4451, "step": 31625 }, { "epoch": 0.16, "learning_rate": 2.4376811594202896e-05, "loss": 0.4569, "step": 31650 }, { "epoch": 0.16, "learning_rate": 2.4195652173913047e-05, "loss": 0.47, "step": 31675 }, { "epoch": 0.16, "learning_rate": 2.401449275362319e-05, "loss": 0.4827, "step": 31700 }, { "epoch": 0.16, "learning_rate": 2.3833333333333334e-05, "loss": 0.4325, "step": 31725 }, { "epoch": 0.16, "learning_rate": 2.365217391304348e-05, "loss": 0.5046, "step": 31750 }, { "epoch": 0.16, "learning_rate": 2.3471014492753625e-05, "loss": 0.4215, "step": 31775 }, { "epoch": 0.16, "learning_rate": 2.328985507246377e-05, "loss": 0.4677, "step": 31800 }, { "epoch": 0.16, "learning_rate": 2.3108695652173913e-05, "loss": 0.5024, "step": 31825 }, { "epoch": 0.16, "learning_rate": 2.292753623188406e-05, "loss": 0.4391, "step": 31850 }, { "epoch": 0.16, "learning_rate": 2.2746376811594204e-05, "loss": 0.4453, "step": 31875 }, { "epoch": 0.16, "learning_rate": 2.2565217391304348e-05, "loss": 0.4534, "step": 31900 }, { "epoch": 0.16, "learning_rate": 2.2384057971014495e-05, "loss": 0.4199, "step": 31925 }, { "epoch": 0.16, "learning_rate": 2.220289855072464e-05, "loss": 0.49, "step": 31950 }, { "epoch": 0.16, "learning_rate": 2.2021739130434783e-05, "loss": 0.4466, "step": 31975 }, { "epoch": 0.16, "learning_rate": 2.184057971014493e-05, "loss": 0.4242, "step": 32000 }, { "epoch": 0.16, "eval_loss": 0.45143240690231323, "eval_runtime": 6170.6869, "eval_samples_per_second": 0.811, "eval_steps_per_second": 0.203, "eval_wer": 30.375661501222144, "step": 32000 }, { "epoch": 0.16, "learning_rate": 2.1659420289855074e-05, "loss": 0.5117, "step": 32025 }, { "epoch": 0.16, "learning_rate": 2.1478260869565218e-05, "loss": 0.4279, "step": 32050 }, { "epoch": 0.16, "learning_rate": 2.129710144927536e-05, "loss": 0.3904, "step": 32075 }, { "epoch": 0.16, "learning_rate": 2.111594202898551e-05, "loss": 0.4152, "step": 32100 }, { "epoch": 0.16, "learning_rate": 2.0934782608695652e-05, "loss": 0.4275, "step": 32125 }, { "epoch": 0.16, "learning_rate": 2.0753623188405796e-05, "loss": 0.4093, "step": 32150 }, { "epoch": 0.16, "learning_rate": 2.0572463768115943e-05, "loss": 0.4287, "step": 32175 }, { "epoch": 0.16, "learning_rate": 2.0391304347826087e-05, "loss": 0.4427, "step": 32200 }, { "epoch": 0.16, "learning_rate": 2.021014492753623e-05, "loss": 0.4304, "step": 32225 }, { "epoch": 0.16, "learning_rate": 2.0028985507246375e-05, "loss": 0.4881, "step": 32250 }, { "epoch": 0.16, "learning_rate": 1.9847826086956522e-05, "loss": 0.4714, "step": 32275 }, { "epoch": 0.16, "learning_rate": 1.9666666666666666e-05, "loss": 0.4183, "step": 32300 }, { "epoch": 0.16, "learning_rate": 1.948550724637681e-05, "loss": 0.4294, "step": 32325 }, { "epoch": 0.16, "learning_rate": 1.9304347826086957e-05, "loss": 0.4769, "step": 32350 }, { "epoch": 0.16, "learning_rate": 1.91231884057971e-05, "loss": 0.451, "step": 32375 }, { "epoch": 0.16, "learning_rate": 1.8942028985507244e-05, "loss": 0.3997, "step": 32400 }, { "epoch": 0.16, "learning_rate": 1.876086956521739e-05, "loss": 0.3907, "step": 32425 }, { "epoch": 0.16, "learning_rate": 1.8579710144927536e-05, "loss": 0.469, "step": 32450 }, { "epoch": 0.16, "learning_rate": 1.8398550724637683e-05, "loss": 0.4591, "step": 32475 }, { "epoch": 0.16, "learning_rate": 1.8217391304347827e-05, "loss": 0.4348, "step": 32500 }, { "epoch": 0.16, "learning_rate": 1.8036231884057974e-05, "loss": 0.4522, "step": 32525 }, { "epoch": 0.16, "learning_rate": 1.7855072463768118e-05, "loss": 0.4456, "step": 32550 }, { "epoch": 0.16, "learning_rate": 1.767391304347826e-05, "loss": 0.396, "step": 32575 }, { "epoch": 0.16, "learning_rate": 1.749275362318841e-05, "loss": 0.3869, "step": 32600 }, { "epoch": 0.16, "learning_rate": 1.7311594202898552e-05, "loss": 0.4212, "step": 32625 }, { "epoch": 0.16, "learning_rate": 1.7130434782608696e-05, "loss": 0.3793, "step": 32650 }, { "epoch": 0.16, "learning_rate": 1.6949275362318843e-05, "loss": 0.4553, "step": 32675 }, { "epoch": 0.16, "learning_rate": 1.6768115942028987e-05, "loss": 0.4395, "step": 32700 }, { "epoch": 0.16, "learning_rate": 1.658695652173913e-05, "loss": 0.4249, "step": 32725 }, { "epoch": 0.16, "learning_rate": 1.6405797101449275e-05, "loss": 0.4313, "step": 32750 }, { "epoch": 0.16, "learning_rate": 1.6224637681159422e-05, "loss": 0.4577, "step": 32775 }, { "epoch": 0.16, "learning_rate": 1.6043478260869566e-05, "loss": 0.381, "step": 32800 }, { "epoch": 0.16, "learning_rate": 1.586231884057971e-05, "loss": 0.458, "step": 32825 }, { "epoch": 0.16, "learning_rate": 1.5681159420289857e-05, "loss": 0.4388, "step": 32850 }, { "epoch": 0.16, "learning_rate": 1.55e-05, "loss": 0.4619, "step": 32875 }, { "epoch": 0.16, "learning_rate": 1.5318840579710144e-05, "loss": 0.4418, "step": 32900 }, { "epoch": 0.17, "learning_rate": 1.513768115942029e-05, "loss": 0.388, "step": 32925 }, { "epoch": 0.17, "learning_rate": 1.4956521739130434e-05, "loss": 0.4835, "step": 32950 }, { "epoch": 0.17, "learning_rate": 1.477536231884058e-05, "loss": 0.4566, "step": 32975 }, { "epoch": 0.17, "learning_rate": 1.4594202898550725e-05, "loss": 0.3837, "step": 33000 }, { "epoch": 0.17, "eval_loss": 0.44476595520973206, "eval_runtime": 6195.5396, "eval_samples_per_second": 0.808, "eval_steps_per_second": 0.202, "eval_wer": 30.311587840243003, "step": 33000 }, { "epoch": 0.17, "learning_rate": 1.4420289855072464e-05, "loss": 0.439, "step": 33025 }, { "epoch": 0.17, "learning_rate": 1.4239130434782609e-05, "loss": 0.4526, "step": 33050 }, { "epoch": 0.17, "learning_rate": 1.4057971014492753e-05, "loss": 0.3809, "step": 33075 }, { "epoch": 0.17, "learning_rate": 1.3876811594202898e-05, "loss": 0.4424, "step": 33100 }, { "epoch": 0.17, "learning_rate": 1.3695652173913042e-05, "loss": 0.4592, "step": 33125 }, { "epoch": 0.17, "learning_rate": 1.3514492753623188e-05, "loss": 0.4471, "step": 33150 }, { "epoch": 0.17, "learning_rate": 1.3333333333333335e-05, "loss": 0.4752, "step": 33175 }, { "epoch": 0.17, "learning_rate": 1.3152173913043479e-05, "loss": 0.4864, "step": 33200 }, { "epoch": 0.17, "learning_rate": 1.2971014492753624e-05, "loss": 0.415, "step": 33225 }, { "epoch": 0.17, "learning_rate": 1.2789855072463768e-05, "loss": 0.5136, "step": 33250 }, { "epoch": 0.17, "learning_rate": 1.2608695652173914e-05, "loss": 0.4896, "step": 33275 }, { "epoch": 0.17, "learning_rate": 1.2427536231884059e-05, "loss": 0.4255, "step": 33300 }, { "epoch": 0.17, "learning_rate": 1.2246376811594203e-05, "loss": 0.4414, "step": 33325 }, { "epoch": 0.17, "learning_rate": 1.2065217391304348e-05, "loss": 0.4077, "step": 33350 }, { "epoch": 0.17, "learning_rate": 1.1884057971014492e-05, "loss": 0.3616, "step": 33375 }, { "epoch": 0.17, "learning_rate": 1.1702898550724638e-05, "loss": 0.406, "step": 33400 }, { "epoch": 0.17, "learning_rate": 1.1521739130434783e-05, "loss": 0.4286, "step": 33425 }, { "epoch": 0.17, "learning_rate": 1.1340579710144927e-05, "loss": 0.4421, "step": 33450 }, { "epoch": 0.17, "learning_rate": 1.1159420289855073e-05, "loss": 0.3742, "step": 33475 }, { "epoch": 0.17, "learning_rate": 1.0978260869565216e-05, "loss": 0.3674, "step": 33500 }, { "epoch": 0.17, "learning_rate": 1.0797101449275362e-05, "loss": 0.4166, "step": 33525 }, { "epoch": 0.17, "learning_rate": 1.0615942028985507e-05, "loss": 0.4147, "step": 33550 }, { "epoch": 0.17, "learning_rate": 1.0434782608695653e-05, "loss": 0.4894, "step": 33575 }, { "epoch": 0.17, "learning_rate": 1.0253623188405798e-05, "loss": 0.4049, "step": 33600 }, { "epoch": 0.17, "learning_rate": 1.0072463768115942e-05, "loss": 0.4257, "step": 33625 }, { "epoch": 0.17, "learning_rate": 9.891304347826088e-06, "loss": 0.472, "step": 33650 }, { "epoch": 0.17, "learning_rate": 9.710144927536233e-06, "loss": 0.404, "step": 33675 }, { "epoch": 0.17, "learning_rate": 9.528985507246377e-06, "loss": 0.4576, "step": 33700 }, { "epoch": 0.17, "learning_rate": 9.347826086956523e-06, "loss": 0.4128, "step": 33725 }, { "epoch": 0.17, "learning_rate": 9.166666666666666e-06, "loss": 0.4063, "step": 33750 }, { "epoch": 0.17, "learning_rate": 8.985507246376812e-06, "loss": 0.4406, "step": 33775 }, { "epoch": 0.17, "learning_rate": 8.804347826086956e-06, "loss": 0.4676, "step": 33800 }, { "epoch": 0.17, "learning_rate": 8.623188405797101e-06, "loss": 0.4796, "step": 33825 }, { "epoch": 0.17, "learning_rate": 8.442028985507247e-06, "loss": 0.4821, "step": 33850 }, { "epoch": 0.17, "learning_rate": 8.26086956521739e-06, "loss": 0.4106, "step": 33875 }, { "epoch": 0.17, "learning_rate": 8.079710144927536e-06, "loss": 0.4944, "step": 33900 }, { "epoch": 0.17, "learning_rate": 7.89855072463768e-06, "loss": 0.4475, "step": 33925 }, { "epoch": 0.17, "learning_rate": 7.717391304347827e-06, "loss": 0.4984, "step": 33950 }, { "epoch": 0.17, "learning_rate": 7.536231884057971e-06, "loss": 0.4352, "step": 33975 }, { "epoch": 0.17, "learning_rate": 7.355072463768116e-06, "loss": 0.4321, "step": 34000 }, { "epoch": 0.17, "eval_loss": 0.43773025274276733, "eval_runtime": 6174.6708, "eval_samples_per_second": 0.811, "eval_steps_per_second": 0.203, "eval_wer": 29.469137853295045, "step": 34000 }, { "epoch": 0.17, "learning_rate": 7.173913043478261e-06, "loss": 0.4081, "step": 34025 }, { "epoch": 0.17, "learning_rate": 6.992753623188406e-06, "loss": 0.4442, "step": 34050 }, { "epoch": 0.17, "learning_rate": 6.811594202898551e-06, "loss": 0.4163, "step": 34075 }, { "epoch": 0.17, "learning_rate": 6.630434782608696e-06, "loss": 0.4479, "step": 34100 }, { "epoch": 0.17, "learning_rate": 6.449275362318841e-06, "loss": 0.4042, "step": 34125 }, { "epoch": 0.17, "learning_rate": 6.268115942028985e-06, "loss": 0.4084, "step": 34150 }, { "epoch": 0.17, "learning_rate": 6.086956521739131e-06, "loss": 0.4172, "step": 34175 }, { "epoch": 0.17, "learning_rate": 5.905797101449275e-06, "loss": 0.4458, "step": 34200 }, { "epoch": 0.17, "learning_rate": 5.72463768115942e-06, "loss": 0.414, "step": 34225 }, { "epoch": 0.17, "learning_rate": 5.543478260869565e-06, "loss": 0.4291, "step": 34250 }, { "epoch": 0.17, "learning_rate": 5.36231884057971e-06, "loss": 0.4316, "step": 34275 }, { "epoch": 0.17, "learning_rate": 5.181159420289856e-06, "loss": 0.4382, "step": 34300 }, { "epoch": 0.17, "learning_rate": 5e-06, "loss": 0.4041, "step": 34325 }, { "epoch": 0.17, "learning_rate": 4.818840579710145e-06, "loss": 0.4503, "step": 34350 }, { "epoch": 0.17, "learning_rate": 4.63768115942029e-06, "loss": 0.4374, "step": 34375 }, { "epoch": 0.17, "learning_rate": 4.456521739130434e-06, "loss": 0.4563, "step": 34400 }, { "epoch": 0.17, "learning_rate": 4.275362318840579e-06, "loss": 0.4047, "step": 34425 }, { "epoch": 0.17, "learning_rate": 4.094202898550725e-06, "loss": 0.3997, "step": 34450 }, { "epoch": 0.17, "learning_rate": 3.91304347826087e-06, "loss": 0.4044, "step": 34475 }, { "epoch": 0.17, "learning_rate": 3.7318840579710147e-06, "loss": 0.3637, "step": 34500 }, { "epoch": 0.17, "learning_rate": 3.5507246376811594e-06, "loss": 0.4967, "step": 34525 }, { "epoch": 0.17, "learning_rate": 3.3695652173913045e-06, "loss": 0.4352, "step": 34550 }, { "epoch": 0.17, "learning_rate": 3.188405797101449e-06, "loss": 0.4373, "step": 34575 }, { "epoch": 0.17, "learning_rate": 3.0072463768115946e-06, "loss": 0.4713, "step": 34600 }, { "epoch": 0.17, "learning_rate": 2.8260869565217393e-06, "loss": 0.4667, "step": 34625 }, { "epoch": 0.17, "learning_rate": 2.644927536231884e-06, "loss": 0.4344, "step": 34650 }, { "epoch": 0.17, "learning_rate": 2.463768115942029e-06, "loss": 0.4458, "step": 34675 }, { "epoch": 0.17, "learning_rate": 2.282608695652174e-06, "loss": 0.3944, "step": 34700 }, { "epoch": 0.17, "learning_rate": 2.101449275362319e-06, "loss": 0.4295, "step": 34725 }, { "epoch": 0.17, "learning_rate": 1.920289855072464e-06, "loss": 0.4773, "step": 34750 }, { "epoch": 0.17, "learning_rate": 1.7391304347826088e-06, "loss": 0.4438, "step": 34775 }, { "epoch": 0.17, "learning_rate": 1.5579710144927536e-06, "loss": 0.3903, "step": 34800 }, { "epoch": 0.17, "learning_rate": 1.3768115942028987e-06, "loss": 0.4185, "step": 34825 }, { "epoch": 0.17, "learning_rate": 1.1956521739130434e-06, "loss": 0.4553, "step": 34850 }, { "epoch": 0.17, "learning_rate": 1.0144927536231885e-06, "loss": 0.3928, "step": 34875 }, { "epoch": 0.17, "learning_rate": 8.333333333333334e-07, "loss": 0.4137, "step": 34900 }, { "epoch": 0.18, "learning_rate": 6.521739130434783e-07, "loss": 0.4452, "step": 34925 }, { "epoch": 0.18, "learning_rate": 4.710144927536232e-07, "loss": 0.4904, "step": 34950 }, { "epoch": 0.18, "learning_rate": 2.898550724637681e-07, "loss": 0.4443, "step": 34975 }, { "epoch": 0.18, "learning_rate": 1.0869565217391305e-07, "loss": 0.4268, "step": 35000 }, { "epoch": 0.18, "eval_loss": 0.4343973398208618, "eval_runtime": 6168.1717, "eval_samples_per_second": 0.811, "eval_steps_per_second": 0.203, "eval_wer": 29.205723913714138, "step": 35000 }, { "epoch": 0.18, "step": 35001, "total_flos": 4.7538258866601984e+20, "train_loss": 1.5390448555129346e-05, "train_runtime": 6.8399, "train_samples_per_second": 40936.549, "train_steps_per_second": 5117.069 } ], "logging_steps": 25, "max_steps": 35000, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 1000, "total_flos": 4.7538258866601984e+20, "train_batch_size": 8, "trial_name": null, "trial_params": null }