diff --git "a/nohup.out" "b/nohup.out" --- "a/nohup.out" +++ "b/nohup.out" @@ -20591,3 +20591,11453 @@ huggingface/tokenizers: The current process just got forked, after parallelism h To disable this warning, you can either: - Avoid using `tokenizers` before the fork if possible - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) +huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks... +To disable this warning, you can either: + - Avoid using `tokenizers` before the fork if possible + - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false) + + Training...: 28% 1214/4393 [1:58:17<384:10:13, 435.05s/it]/home/sanchitgandhi/hf/lib/python3.8/site-packages/flax/training/common_utils.py:25: FutureWarning: jax.tree_map is deprecated, and will be removed in a future release. Use jax.tree_util.tree_map instead. + return jax.tree_map( + + Training...: 28% 1215/4393 [1:58:24<270:31:57, 306.46s/it] + Training...: 28% 1216/4393 [1:58:30<190:48:41, 216.22s/it] + Training...: 28% 1217/4393 [1:58:35<135:01:44, 153.06s/it] + Training...: 28% 1218/4393 [1:58:41<95:58:10, 108.82s/it]  + Training...: 28% 1219/4393 [1:58:46<68:37:29, 77.84s/it]  + Training...: 28% 1220/4393 [1:58:52<49:29:51, 56.16s/it] + Training...: 28% 1221/4393 [1:58:57<36:04:45, 40.95s/it] + Training...: 28% 1222/4393 [1:59:03<26:40:49, 30.29s/it] + Training...: 28% 1223/4393 [1:59:08<20:06:11, 22.83s/it] + Training...: 28% 1224/4393 [1:59:14<15:30:20, 17.61s/it] + Training...: 28% 1225/4393 [1:59:19<12:15:16, 13.93s/it] + Training...: 28% 1226/4393 [1:59:24<9:59:02, 11.35s/it]  + Training...: 28% 1227/4393 [1:59:30<8:22:44, 9.53s/it] + Training...: 28% 1228/4393 [1:59:35<7:13:30, 8.22s/it] + Training...: 28% 1229/4393 [1:59:40<6:25:14, 7.31s/it] + Training...: 28% 1230/4393 [1:59:45<5:50:36, 6.65s/it] + Training...: 28% 1231/4393 [1:59:50<5:25:52, 6.18s/it] + Training...: 28% 1232/4393 [1:59:55<5:06:41, 5.82s/it] + Training...: 28% 1233/4393 [2:00:00<4:52:53, 5.56s/it] + Training...: 28% 1234/4393 [2:00:05<4:41:40, 5.35s/it] + Training...: 28% 1235/4393 [2:00:09<4:29:13, 5.11s/it] + Training...: 28% 1236/4393 [2:00:14<4:20:18, 4.95s/it] + Training...: 28% 1237/4393 [2:00:18<4:11:00, 4.77s/it] + Training...: 28% 1238/4393 [2:00:23<4:01:27, 4.59s/it]/home/sanchitgandhi/hf/lib/python3.8/site-packages/flax/jax_utils.py:61: FutureWarning: jax.tree_map is deprecated, and will be removed in a future release. Use jax.tree_util.tree_map instead. + return jax.tree_map(lambda x: x[0], tree) +run_flax_speech_recognition_seq2seq.py:336: FutureWarning: jax.tree_map is deprecated, and will be removed in a future release. Use jax.tree_util.tree_map instead. + return jax.tree_map(lambda x: x.astype(jnp.float32) if x.dtype == jnp.bfloat16 else x, t) + +  + Training...: 28% 1238/4393 [2:00:27<4:01:27, 4.59s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:29:59<57:16:35, 20619.59s/it] + Training...: 28% 1239/4393 [2:00:27<4:01:17, 4.59s/it] + Training...: 28% 1240/4393 [2:00:31<3:50:21, 4.38s/it] + Training...: 28% 1241/4393 [2:00:35<3:36:33, 4.12s/it] + Training...: 28% 1242/4393 [2:00:38<3:22:52, 3.86s/it] + Training...: 28% 1243/4393 [2:00:41<3:09:04, 3.60s/it] + Training...: 28% 1244/4393 [2:00:44<2:54:42, 3.33s/it] + Training...: 28% 1245/4393 [2:00:46<2:39:31, 3.04s/it] + Training...: 28% 1246/4393 [2:00:48<2:25:09, 2.77s/it] + Training...: 28% 1247/4393 [2:00:50<2:10:41, 2.49s/it] + Training...: 28% 1248/4393 [2:00:51<1:56:21, 2.22s/it] + Training...: 28% 1249/4393 [2:00:53<1:41:57, 1.95s/it] + Training...: 28% 1250/4393 [2:00:54<1:27:21, 1.67s/it] + Training...: 28% 1251/4393 [2:01:00<2:38:17, 3.02s/it] + Training...: 28% 1252/4393 [2:01:06<3:26:59, 3.95s/it] + Training...: 29% 1253/4393 [2:01:12<4:02:11, 4.63s/it] + Training...: 29% 1254/4393 [2:01:18<4:23:19, 5.03s/it] + Training...: 29% 1255/4393 [2:01:24<4:36:31, 5.29s/it] + Training...: 29% 1256/4393 [2:01:30<4:44:42, 5.45s/it] + Training...: 29% 1257/4393 [2:01:36<4:50:18, 5.55s/it] + Training...: 29% 1258/4393 [2:01:42<4:53:00, 5.61s/it] + Training...: 29% 1259/4393 [2:01:47<4:56:04, 5.67s/it] + Training...: 29% 1260/4393 [2:01:53<4:55:48, 5.67s/it] + Training...: 29% 1261/4393 [2:01:59<4:57:15, 5.69s/it] + Training...: 29% 1262/4393 [2:02:04<4:55:48, 5.67s/it] + Training...: 29% 1263/4393 [2:02:10<4:55:55, 5.67s/it] +  + Training...: 29% 1263/4393 [2:02:17<4:55:55, 5.67s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:31:49<57:16:35, 20619.59s/it] + Training...: 29% 1264/4393 [2:02:17<5:08:28, 5.92s/it] + Training...: 29% 1265/4393 [2:02:22<5:03:48, 5.83s/it] + Training...: 29% 1266/4393 [2:02:28<4:58:54, 5.74s/it] + Training...: 29% 1267/4393 [2:02:33<4:56:36, 5.69s/it] + Training...: 29% 1268/4393 [2:02:39<4:53:18, 5.63s/it] + Training...: 29% 1269/4393 [2:02:44<4:49:54, 5.57s/it] + Training...: 29% 1270/4393 [2:02:50<4:46:56, 5.51s/it] + Training...: 29% 1271/4393 [2:02:55<4:44:37, 5.47s/it] + Training...: 29% 1272/4393 [2:03:00<4:42:00, 5.42s/it] + Training...: 29% 1273/4393 [2:03:06<4:40:01, 5.39s/it] + Training...: 29% 1274/4393 [2:03:11<4:38:42, 5.36s/it] + Training...: 29% 1275/4393 [2:03:16<4:35:48, 5.31s/it] + Training...: 29% 1276/4393 [2:03:21<4:32:56, 5.25s/it] + Training...: 29% 1277/4393 [2:03:26<4:30:12, 5.20s/it] + Training...: 29% 1278/4393 [2:03:31<4:28:49, 5.18s/it] + Training...: 29% 1279/4393 [2:03:36<4:26:59, 5.14s/it] + Training...: 29% 1280/4393 [2:03:41<4:23:28, 5.08s/it] + Training...: 29% 1281/4393 [2:03:46<4:20:41, 5.03s/it] + Training...: 29% 1282/4393 [2:03:51<4:16:54, 4.95s/it] + Training...: 29% 1283/4393 [2:03:56<4:14:50, 4.92s/it] + Training...: 29% 1284/4393 [2:04:00<4:09:52, 4.82s/it] + Training...: 29% 1285/4393 [2:04:05<4:05:06, 4.73s/it] + Training...: 29% 1286/4393 [2:04:09<3:58:52, 4.61s/it] + Training...: 29% 1287/4393 [2:04:13<3:52:17, 4.49s/it] + Training...: 29% 1288/4393 [2:04:18<3:44:54, 4.35s/it] +  + Training...: 29% 1288/4393 [2:04:22<3:44:54, 4.35s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:33:54<57:16:35, 20619.59s/it] + Training...: 29% 1289/4393 [2:04:22<3:45:14, 4.35s/it] + Training...: 29% 1290/4393 [2:04:25<3:31:37, 4.09s/it] + Training...: 29% 1291/4393 [2:04:29<3:18:01, 3.83s/it] + Training...: 29% 1292/4393 [2:04:32<3:05:51, 3.60s/it] + Training...: 29% 1293/4393 [2:04:34<2:53:47, 3.36s/it] + Training...: 29% 1294/4393 [2:04:37<2:41:44, 3.13s/it] + Training...: 29% 1295/4393 [2:04:39<2:29:22, 2.89s/it] + Training...: 30% 1296/4393 [2:04:41<2:15:45, 2.63s/it] + Training...: 30% 1297/4393 [2:04:43<2:01:27, 2.35s/it] + Training...: 30% 1298/4393 [2:04:45<1:47:49, 2.09s/it] + Training...: 30% 1299/4393 [2:04:46<1:34:22, 1.83s/it] + Training...: 30% 1300/4393 [2:04:47<1:20:39, 1.56s/it] + Training...: 30% 1301/4393 [2:04:53<2:32:15, 2.95s/it] + Training...: 30% 1302/4393 [2:04:59<3:20:36, 3.89s/it] + Training...: 30% 1303/4393 [2:05:05<3:54:46, 4.56s/it] + Training...: 30% 1304/4393 [2:05:11<4:17:46, 5.01s/it] + Training...: 30% 1305/4393 [2:05:17<4:32:20, 5.29s/it] + Training...: 30% 1306/4393 [2:05:23<4:42:10, 5.48s/it] + Training...: 30% 1307/4393 [2:05:29<4:48:04, 5.60s/it] + Training...: 30% 1308/4393 [2:05:35<4:51:33, 5.67s/it] + Training...: 30% 1309/4393 [2:05:41<4:54:20, 5.73s/it] + Training...: 30% 1310/4393 [2:05:46<4:54:46, 5.74s/it] + Training...: 30% 1311/4393 [2:05:52<4:55:11, 5.75s/it] + Training...: 30% 1312/4393 [2:05:58<4:54:01, 5.73s/it] + Training...: 30% 1313/4393 [2:06:04<4:53:54, 5.73s/it] +  + Training...: 30% 1313/4393 [2:06:10<4:53:54, 5.73s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:35:42<57:16:35, 20619.59s/it] + Training...: 30% 1314/4393 [2:06:10<5:04:08, 5.93s/it] + Training...: 30% 1315/4393 [2:06:16<4:59:13, 5.83s/it] + Training...: 30% 1316/4393 [2:06:21<4:53:10, 5.72s/it] + Training...: 30% 1317/4393 [2:06:27<4:50:14, 5.66s/it] + Training...: 30% 1318/4393 [2:06:32<4:47:14, 5.60s/it] + Training...: 30% 1319/4393 [2:06:38<4:44:56, 5.56s/it] + Training...: 30% 1320/4393 [2:06:43<4:42:10, 5.51s/it] + Training...: 30% 1321/4393 [2:06:48<4:40:21, 5.48s/it] + Training...: 30% 1322/4393 [2:06:54<4:38:45, 5.45s/it] + Training...: 30% 1323/4393 [2:06:59<4:34:53, 5.37s/it] + Training...: 30% 1324/4393 [2:07:04<4:32:06, 5.32s/it] + Training...: 30% 1325/4393 [2:07:09<4:29:55, 5.28s/it] + Training...: 30% 1326/4393 [2:07:14<4:27:33, 5.23s/it] + Training...: 30% 1327/4393 [2:07:19<4:25:35, 5.20s/it] + Training...: 30% 1328/4393 [2:07:25<4:23:45, 5.16s/it] + Training...: 30% 1329/4393 [2:07:30<4:21:03, 5.11s/it] + Training...: 30% 1330/4393 [2:07:35<4:18:11, 5.06s/it] + Training...: 30% 1331/4393 [2:07:39<4:14:19, 4.98s/it] + Training...: 30% 1332/4393 [2:07:44<4:10:19, 4.91s/it] + Training...: 30% 1333/4393 [2:07:49<4:10:23, 4.91s/it] + Training...: 30% 1334/4393 [2:07:54<4:08:55, 4.88s/it] + Training...: 30% 1335/4393 [2:07:59<4:07:24, 4.85s/it] + Training...: 30% 1336/4393 [2:08:03<4:02:47, 4.77s/it] + Training...: 30% 1337/4393 [2:08:08<3:57:41, 4.67s/it] + Training...: 30% 1338/4393 [2:08:12<3:51:24, 4.54s/it] +  + Training...: 30% 1338/4393 [2:08:16<3:51:24, 4.54s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:37:49<57:16:35, 20619.59s/it] + Training...: 30% 1339/4393 [2:08:16<3:52:04, 4.56s/it] + Training...: 31% 1340/4393 [2:08:20<3:39:24, 4.31s/it] + Training...: 31% 1341/4393 [2:08:24<3:27:16, 4.07s/it] + Training...: 31% 1342/4393 [2:08:27<3:15:32, 3.85s/it] + Training...: 31% 1343/4393 [2:08:30<3:01:08, 3.56s/it] + Training...: 31% 1344/4393 [2:08:33<2:47:14, 3.29s/it] + Training...: 31% 1345/4393 [2:08:35<2:32:53, 3.01s/it] + Training...: 31% 1346/4393 [2:08:37<2:19:03, 2.74s/it] + Training...: 31% 1347/4393 [2:08:39<2:04:32, 2.45s/it] + Training...: 31% 1348/4393 [2:08:40<1:51:11, 2.19s/it] + Training...: 31% 1349/4393 [2:08:42<1:37:37, 1.92s/it] + Training...: 31% 1350/4393 [2:08:43<1:23:56, 1.65s/it] + Training...: 31% 1351/4393 [2:08:49<2:36:51, 3.09s/it] + Training...: 31% 1352/4393 [2:08:55<3:25:43, 4.06s/it] + Training...: 31% 1353/4393 [2:09:02<4:00:41, 4.75s/it] + Training...: 31% 1354/4393 [2:09:08<4:22:50, 5.19s/it] + Training...: 31% 1355/4393 [2:09:14<4:38:20, 5.50s/it] + Training...: 31% 1356/4393 [2:09:20<4:46:54, 5.67s/it] + Training...: 31% 1357/4393 [2:09:27<4:54:41, 5.82s/it] + Training...: 31% 1358/4393 [2:09:33<4:59:10, 5.91s/it] + Training...: 31% 1359/4393 [2:09:39<5:02:50, 5.99s/it] + Training...: 31% 1360/4393 [2:09:45<5:04:02, 6.01s/it] + Training...: 31% 1361/4393 [2:09:51<5:04:17, 6.02s/it] + Training...: 31% 1362/4393 [2:09:57<5:02:53, 6.00s/it] + Training...: 31% 1363/4393 [2:10:03<5:02:33, 5.99s/it] +  + Training...: 31% 1363/4393 [2:10:09<5:02:33, 5.99s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:39:42<57:16:35, 20619.59s/it] + Training...: 31% 1364/4393 [2:10:09<5:12:48, 6.20s/it] + Training...: 31% 1365/4393 [2:10:15<5:08:59, 6.12s/it] + Training...: 31% 1366/4393 [2:10:21<5:05:12, 6.05s/it] + Training...: 31% 1367/4393 [2:10:27<5:00:48, 5.96s/it] + Training...: 31% 1368/4393 [2:10:33<4:57:59, 5.91s/it] + Training...: 31% 1369/4393 [2:10:39<4:55:50, 5.87s/it] + Training...: 31% 1370/4393 [2:10:44<4:52:56, 5.81s/it] + Training...: 31% 1371/4393 [2:10:50<4:50:20, 5.76s/it] + Training...: 31% 1372/4393 [2:10:56<4:47:21, 5.71s/it] + Training...: 31% 1373/4393 [2:11:01<4:44:50, 5.66s/it] + Training...: 31% 1374/4393 [2:11:07<4:41:38, 5.60s/it] + Training...: 31% 1375/4393 [2:11:12<4:39:33, 5.56s/it] + Training...: 31% 1376/4393 [2:11:17<4:37:09, 5.51s/it] + Training...: 31% 1377/4393 [2:11:23<4:34:45, 5.47s/it] + Training...: 31% 1378/4393 [2:11:28<4:30:53, 5.39s/it] + Training...: 31% 1379/4393 [2:11:33<4:28:29, 5.34s/it] + Training...: 31% 1380/4393 [2:11:38<4:25:35, 5.29s/it] + Training...: 31% 1381/4393 [2:11:44<4:24:05, 5.26s/it] + Training...: 31% 1382/4393 [2:11:49<4:21:27, 5.21s/it] + Training...: 31% 1383/4393 [2:11:54<4:17:09, 5.13s/it] + Training...: 32% 1384/4393 [2:11:59<4:14:10, 5.07s/it] + Training...: 32% 1385/4393 [2:12:03<4:09:34, 4.98s/it] + Training...: 32% 1386/4393 [2:12:08<4:03:50, 4.87s/it] + Training...: 32% 1387/4393 [2:12:12<3:58:18, 4.76s/it] + Training...: 32% 1388/4393 [2:12:17<3:52:47, 4.65s/it] +  + Training...: 32% 1388/4393 [2:12:21<3:52:47, 4.65s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:41:54<57:16:35, 20619.59s/it] + Training...: 32% 1389/4393 [2:12:21<3:53:01, 4.65s/it] + Training...: 32% 1390/4393 [2:12:25<3:40:54, 4.41s/it] + Training...: 32% 1391/4393 [2:12:29<3:27:46, 4.15s/it] + Training...: 32% 1392/4393 [2:12:32<3:14:21, 3.89s/it] + Training...: 32% 1393/4393 [2:12:35<3:00:06, 3.60s/it] + Training...: 32% 1394/4393 [2:12:38<2:46:56, 3.34s/it] + Training...: 32% 1395/4393 [2:12:40<2:33:11, 3.07s/it] + Training...: 32% 1396/4393 [2:12:42<2:20:10, 2.81s/it] + Training...: 32% 1397/4393 [2:12:44<2:06:09, 2.53s/it] + Training...: 32% 1398/4393 [2:12:46<1:52:07, 2.25s/it] + Training...: 32% 1399/4393 [2:12:47<1:37:39, 1.96s/it] + Training...: 32% 1400/4393 [2:12:48<1:23:12, 1.67s/it] + Training...: 32% 1401/4393 [2:12:54<2:31:35, 3.04s/it] + Training...: 32% 1402/4393 [2:13:01<3:17:46, 3.97s/it] + Training...: 32% 1403/4393 [2:13:07<3:47:58, 4.57s/it] + Training...: 32% 1404/4393 [2:13:13<4:11:27, 5.05s/it] + Training...: 32% 1405/4393 [2:13:19<4:24:52, 5.32s/it] + Training...: 32% 1406/4393 [2:13:25<4:33:22, 5.49s/it] + Training...: 32% 1407/4393 [2:13:30<4:37:59, 5.59s/it] + Training...: 32% 1408/4393 [2:13:36<4:41:36, 5.66s/it] + Training...: 32% 1409/4393 [2:13:42<4:43:02, 5.69s/it] + Training...: 32% 1410/4393 [2:13:48<4:43:11, 5.70s/it] + Training...: 32% 1411/4393 [2:13:53<4:43:54, 5.71s/it] + Training...: 32% 1412/4393 [2:13:59<4:42:17, 5.68s/it] + Training...: 32% 1413/4393 [2:14:05<4:42:00, 5.68s/it] +  + Training...: 32% 1413/4393 [2:14:11<4:42:00, 5.68s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:43:43<57:16:35, 20619.59s/it] + Training...: 32% 1414/4393 [2:14:11<4:52:35, 5.89s/it] + Training...: 32% 1415/4393 [2:14:17<4:48:18, 5.81s/it] + Training...: 32% 1416/4393 [2:14:22<4:45:15, 5.75s/it] + Training...: 32% 1417/4393 [2:14:28<4:42:38, 5.70s/it] + Training...: 32% 1418/4393 [2:14:33<4:39:18, 5.63s/it] + Training...: 32% 1419/4393 [2:14:39<4:37:13, 5.59s/it] + Training...: 32% 1420/4393 [2:14:44<4:34:55, 5.55s/it] + Training...: 32% 1421/4393 [2:14:50<4:32:10, 5.49s/it] + Training...: 32% 1422/4393 [2:14:55<4:30:07, 5.46s/it] + Training...: 32% 1423/4393 [2:15:00<4:28:00, 5.41s/it] + Training...: 32% 1424/4393 [2:15:06<4:25:24, 5.36s/it] + Training...: 32% 1425/4393 [2:15:11<4:24:03, 5.34s/it] + Training...: 32% 1426/4393 [2:15:16<4:21:36, 5.29s/it] + Training...: 32% 1427/4393 [2:15:21<4:19:33, 5.25s/it] + Training...: 33% 1428/4393 [2:15:26<4:17:24, 5.21s/it] + Training...: 33% 1429/4393 [2:15:31<4:14:25, 5.15s/it] + Training...: 33% 1430/4393 [2:15:36<4:10:39, 5.08s/it] + Training...: 33% 1431/4393 [2:15:41<4:07:43, 5.02s/it] + Training...: 33% 1432/4393 [2:15:46<4:03:47, 4.94s/it] + Training...: 33% 1433/4393 [2:15:51<4:01:27, 4.89s/it] + Training...: 33% 1434/4393 [2:15:55<3:57:16, 4.81s/it] + Training...: 33% 1435/4393 [2:16:00<3:53:54, 4.74s/it] + Training...: 33% 1436/4393 [2:16:04<3:48:25, 4.64s/it] + Training...: 33% 1437/4393 [2:16:08<3:42:14, 4.51s/it] + Training...: 33% 1438/4393 [2:16:13<3:35:27, 4.37s/it] +  + Training...: 33% 1438/4393 [2:16:17<3:35:27, 4.37s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:45:49<57:16:35, 20619.59s/it] + Training...: 33% 1439/4393 [2:16:17<3:35:31, 4.38s/it] + Training...: 33% 1440/4393 [2:16:20<3:23:14, 4.13s/it] + Training...: 33% 1441/4393 [2:16:24<3:11:42, 3.90s/it] + Training...: 33% 1442/4393 [2:16:27<3:00:18, 3.67s/it] + Training...: 33% 1443/4393 [2:16:30<2:47:45, 3.41s/it] + Training...: 33% 1444/4393 [2:16:32<2:35:09, 3.16s/it] + Training...: 33% 1445/4393 [2:16:35<2:22:21, 2.90s/it] + Training...: 33% 1446/4393 [2:16:37<2:09:45, 2.64s/it] + Training...: 33% 1447/4393 [2:16:38<1:57:26, 2.39s/it] + Training...: 33% 1448/4393 [2:16:40<1:44:56, 2.14s/it] + Training...: 33% 1449/4393 [2:16:41<1:31:48, 1.87s/it] + Training...: 33% 1450/4393 [2:16:42<1:18:51, 1.61s/it] + Training...: 33% 1451/4393 [2:16:49<2:28:37, 3.03s/it] + Training...: 33% 1452/4393 [2:16:55<3:15:58, 4.00s/it] + Training...: 33% 1453/4393 [2:17:01<3:46:26, 4.62s/it] + Training...: 33% 1454/4393 [2:17:07<4:06:40, 5.04s/it] + Training...: 33% 1455/4393 [2:17:13<4:20:59, 5.33s/it] + Training...: 33% 1456/4393 [2:17:19<4:29:57, 5.52s/it] + Training...: 33% 1457/4393 [2:17:25<4:35:59, 5.64s/it] + Training...: 33% 1458/4393 [2:17:31<4:39:30, 5.71s/it] + Training...: 33% 1459/4393 [2:17:37<4:42:01, 5.77s/it] + Training...: 33% 1460/4393 [2:17:42<4:42:13, 5.77s/it] + Training...: 33% 1461/4393 [2:17:48<4:41:10, 5.75s/it] + Training...: 33% 1462/4393 [2:17:54<4:40:25, 5.74s/it] + Training...: 33% 1463/4393 [2:18:00<4:39:20, 5.72s/it] +  + Training...: 33% 1463/4393 [2:18:06<4:39:20, 5.72s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:47:38<57:16:35, 20619.59s/it] + Training...: 33% 1464/4393 [2:18:06<4:49:58, 5.94s/it] + Training...: 33% 1465/4393 [2:18:12<4:44:35, 5.83s/it] + Training...: 33% 1466/4393 [2:18:17<4:40:04, 5.74s/it] + Training...: 33% 1467/4393 [2:18:23<4:38:00, 5.70s/it] + Training...: 33% 1468/4393 [2:18:28<4:34:27, 5.63s/it] + Training...: 33% 1469/4393 [2:18:34<4:31:24, 5.57s/it] + Training...: 33% 1470/4393 [2:18:39<4:29:04, 5.52s/it] + Training...: 33% 1471/4393 [2:18:44<4:26:23, 5.47s/it] + Training...: 34% 1472/4393 [2:18:50<4:24:46, 5.44s/it] + Training...: 34% 1473/4393 [2:18:55<4:22:47, 5.40s/it] + Training...: 34% 1474/4393 [2:19:00<4:20:56, 5.36s/it] + Training...: 34% 1475/4393 [2:19:06<4:19:01, 5.33s/it] + Training...: 34% 1476/4393 [2:19:11<4:18:18, 5.31s/it] + Training...: 34% 1477/4393 [2:19:16<4:18:30, 5.32s/it] + Training...: 34% 1478/4393 [2:19:21<4:15:55, 5.27s/it] + Training...: 34% 1479/4393 [2:19:26<4:12:15, 5.19s/it] + Training...: 34% 1480/4393 [2:19:31<4:07:54, 5.11s/it] + Training...: 34% 1481/4393 [2:19:36<4:05:13, 5.05s/it] + Training...: 34% 1482/4393 [2:19:41<4:03:04, 5.01s/it] + Training...: 34% 1483/4393 [2:19:46<3:58:53, 4.93s/it] + Training...: 34% 1484/4393 [2:19:50<3:54:44, 4.84s/it] + Training...: 34% 1485/4393 [2:19:55<3:50:11, 4.75s/it] + Training...: 34% 1486/4393 [2:19:59<3:45:16, 4.65s/it] + Training...: 34% 1487/4393 [2:20:04<3:39:32, 4.53s/it] + Training...: 34% 1488/4393 [2:20:08<3:32:19, 4.39s/it] +  + Training...: 34% 1488/4393 [2:20:12<3:32:19, 4.39s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:49:44<57:16:35, 20619.59s/it] + Training...: 34% 1489/4393 [2:20:12<3:33:17, 4.41s/it] + Training...: 34% 1490/4393 [2:20:16<3:20:20, 4.14s/it] + Training...: 34% 1491/4393 [2:20:19<3:07:36, 3.88s/it] + Training...: 34% 1492/4393 [2:20:22<2:54:50, 3.62s/it] + Training...: 34% 1493/4393 [2:20:25<2:42:03, 3.35s/it] + Training...: 34% 1494/4393 [2:20:27<2:29:11, 3.09s/it] + Training...: 34% 1495/4393 [2:20:29<2:16:23, 2.82s/it] + Training...: 34% 1496/4393 [2:20:31<2:04:13, 2.57s/it] + Training...: 34% 1497/4393 [2:20:33<1:51:04, 2.30s/it] + Training...: 34% 1498/4393 [2:20:34<1:38:02, 2.03s/it] + Training...: 34% 1499/4393 [2:20:36<1:25:58, 1.78s/it] + Training...: 34% 1500/4393 [2:20:37<1:13:51, 1.53s/it] + Training...: 34% 1501/4393 [2:20:43<2:21:43, 2.94s/it] + Training...: 34% 1502/4393 [2:20:49<3:09:13, 3.93s/it] + Training...: 34% 1503/4393 [2:20:55<3:39:00, 4.55s/it] + Training...: 34% 1504/4393 [2:21:01<3:57:27, 4.93s/it] + Training...: 34% 1505/4393 [2:21:07<4:11:39, 5.23s/it] + Training...: 34% 1506/4393 [2:21:13<4:22:20, 5.45s/it] + Training...: 34% 1507/4393 [2:21:19<4:32:02, 5.66s/it] + Training...: 34% 1508/4393 [2:21:25<4:34:32, 5.71s/it] + Training...: 34% 1509/4393 [2:21:30<4:35:17, 5.73s/it] + Training...: 34% 1510/4393 [2:21:36<4:34:24, 5.71s/it] + Training...: 34% 1511/4393 [2:21:42<4:35:17, 5.73s/it] + Training...: 34% 1512/4393 [2:21:48<4:35:07, 5.73s/it] + Training...: 34% 1513/4393 [2:21:53<4:34:00, 5.71s/it] +  + Training...: 34% 1513/4393 [2:22:00<4:34:00, 5.71s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:51:32<57:16:35, 20619.59s/it] + Training...: 34% 1514/4393 [2:22:00<4:44:40, 5.93s/it] + Training...: 34% 1515/4393 [2:22:05<4:40:17, 5.84s/it] + Training...: 35% 1516/4393 [2:22:11<4:35:06, 5.74s/it] + Training...: 35% 1517/4393 [2:22:16<4:32:06, 5.68s/it] + Training...: 35% 1518/4393 [2:22:22<4:28:36, 5.61s/it] + Training...: 35% 1519/4393 [2:22:27<4:27:42, 5.59s/it] + Training...: 35% 1520/4393 [2:22:33<4:27:44, 5.59s/it] + Training...: 35% 1521/4393 [2:22:38<4:24:03, 5.52s/it] + Training...: 35% 1522/4393 [2:22:44<4:21:31, 5.47s/it] + Training...: 35% 1523/4393 [2:22:49<4:20:12, 5.44s/it] + Training...: 35% 1524/4393 [2:22:54<4:18:08, 5.40s/it] + Training...: 35% 1525/4393 [2:23:00<4:16:01, 5.36s/it] + Training...: 35% 1526/4393 [2:23:05<4:12:48, 5.29s/it] + Training...: 35% 1527/4393 [2:23:10<4:09:56, 5.23s/it] + Training...: 35% 1528/4393 [2:23:15<4:06:52, 5.17s/it] + Training...: 35% 1529/4393 [2:23:20<4:04:31, 5.12s/it] + Training...: 35% 1530/4393 [2:23:25<4:01:22, 5.06s/it] + Training...: 35% 1531/4393 [2:23:30<3:58:36, 5.00s/it] + Training...: 35% 1532/4393 [2:23:34<3:54:59, 4.93s/it] + Training...: 35% 1533/4393 [2:23:39<3:51:49, 4.86s/it] + Training...: 35% 1534/4393 [2:23:44<3:47:15, 4.77s/it] + Training...: 35% 1535/4393 [2:23:48<3:43:27, 4.69s/it] + Training...: 35% 1536/4393 [2:23:53<3:38:02, 4.58s/it] + Training...: 35% 1537/4393 [2:23:57<3:32:38, 4.47s/it] + Training...: 35% 1538/4393 [2:24:01<3:27:39, 4.36s/it] +  + Training...: 35% 1538/4393 [2:24:05<3:27:39, 4.36s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:53:38<57:16:35, 20619.59s/it] + Training...: 35% 1539/4393 [2:24:05<3:30:19, 4.42s/it] + Training...: 35% 1540/4393 [2:24:09<3:17:59, 4.16s/it] + Training...: 35% 1541/4393 [2:24:12<3:05:27, 3.90s/it] + Training...: 35% 1542/4393 [2:24:15<2:53:05, 3.64s/it] + Training...: 35% 1543/4393 [2:24:18<2:40:14, 3.37s/it] + Training...: 35% 1544/4393 [2:24:21<2:27:24, 3.10s/it] + Training...: 35% 1545/4393 [2:24:23<2:14:35, 2.84s/it] + Training...: 35% 1546/4393 [2:24:25<2:01:47, 2.57s/it] + Training...: 35% 1547/4393 [2:24:26<1:48:53, 2.30s/it] + Training...: 35% 1548/4393 [2:24:28<1:37:49, 2.06s/it] + Training...: 35% 1549/4393 [2:24:29<1:25:38, 1.81s/it] + Training...: 35% 1550/4393 [2:24:30<1:13:15, 1.55s/it] + Training...: 35% 1551/4393 [2:24:36<2:19:00, 2.93s/it] + Training...: 35% 1552/4393 [2:24:42<3:06:10, 3.93s/it] + Training...: 35% 1553/4393 [2:24:49<3:36:45, 4.58s/it] + Training...: 35% 1554/4393 [2:24:55<3:56:42, 5.00s/it] + Training...: 35% 1555/4393 [2:25:01<4:11:35, 5.32s/it] + Training...: 35% 1556/4393 [2:25:06<4:20:13, 5.50s/it] + Training...: 35% 1557/4393 [2:25:12<4:26:36, 5.64s/it] + Training...: 35% 1558/4393 [2:25:19<4:33:50, 5.80s/it] + Training...: 35% 1559/4393 [2:25:25<4:36:11, 5.85s/it] + Training...: 36% 1560/4393 [2:25:30<4:35:00, 5.82s/it] + Training...: 36% 1561/4393 [2:25:36<4:33:17, 5.79s/it] + Training...: 36% 1562/4393 [2:25:42<4:31:43, 5.76s/it] + Training...: 36% 1563/4393 [2:25:47<4:30:40, 5.74s/it] +  + Training...: 36% 1563/4393 [2:25:54<4:30:40, 5.74s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:55:26<57:16:35, 20619.59s/it] + Training...: 36% 1564/4393 [2:25:54<4:40:42, 5.95s/it] + Training...: 36% 1565/4393 [2:26:00<4:36:32, 5.87s/it] + Training...: 36% 1566/4393 [2:26:05<4:32:54, 5.79s/it] + Training...: 36% 1567/4393 [2:26:11<4:30:33, 5.74s/it] + Training...: 36% 1568/4393 [2:26:16<4:27:43, 5.69s/it] + Training...: 36% 1569/4393 [2:26:22<4:27:51, 5.69s/it] + Training...: 36% 1570/4393 [2:26:28<4:25:56, 5.65s/it] + Training...: 36% 1571/4393 [2:26:33<4:23:49, 5.61s/it] + Training...: 36% 1572/4393 [2:26:39<4:20:24, 5.54s/it] + Training...: 36% 1573/4393 [2:26:44<4:18:51, 5.51s/it] + Training...: 36% 1574/4393 [2:26:49<4:16:44, 5.46s/it] + Training...: 36% 1575/4393 [2:26:55<4:15:10, 5.43s/it] + Training...: 36% 1576/4393 [2:27:00<4:15:05, 5.43s/it] + Training...: 36% 1577/4393 [2:27:05<4:12:49, 5.39s/it] + Training...: 36% 1578/4393 [2:27:11<4:10:36, 5.34s/it] + Training...: 36% 1579/4393 [2:27:16<4:08:15, 5.29s/it] + Training...: 36% 1580/4393 [2:27:21<4:05:25, 5.23s/it] + Training...: 36% 1581/4393 [2:27:26<4:03:00, 5.19s/it] + Training...: 36% 1582/4393 [2:27:31<3:59:54, 5.12s/it] + Training...: 36% 1583/4393 [2:27:36<3:58:29, 5.09s/it] + Training...: 36% 1584/4393 [2:27:41<3:57:02, 5.06s/it] + Training...: 36% 1585/4393 [2:27:46<3:51:46, 4.95s/it] + Training...: 36% 1586/4393 [2:27:50<3:47:00, 4.85s/it] + Training...: 36% 1587/4393 [2:27:55<3:42:06, 4.75s/it] + Training...: 36% 1588/4393 [2:27:59<3:34:32, 4.59s/it] +  + Training...: 36% 1588/4393 [2:28:04<3:34:32, 4.59s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:57:36<57:16:35, 20619.59s/it] + Training...: 36% 1589/4393 [2:28:04<3:34:39, 4.59s/it] + Training...: 36% 1590/4393 [2:28:07<3:21:08, 4.31s/it] + Training...: 36% 1591/4393 [2:28:11<3:08:00, 4.03s/it] + Training...: 36% 1592/4393 [2:28:14<2:55:10, 3.75s/it] + Training...: 36% 1593/4393 [2:28:16<2:41:26, 3.46s/it] + Training...: 36% 1594/4393 [2:28:19<2:28:23, 3.18s/it] + Training...: 36% 1595/4393 [2:28:21<2:15:37, 2.91s/it] + Training...: 36% 1596/4393 [2:28:23<2:03:20, 2.65s/it] + Training...: 36% 1597/4393 [2:28:25<1:51:36, 2.40s/it] + Training...: 36% 1598/4393 [2:28:27<1:39:31, 2.14s/it] + Training...: 36% 1599/4393 [2:28:28<1:27:16, 1.87s/it] + Training...: 36% 1600/4393 [2:28:29<1:14:36, 1.60s/it] + Training...: 36% 1601/4393 [2:28:35<2:18:42, 2.98s/it] + Training...: 36% 1602/4393 [2:28:41<3:03:54, 3.95s/it] + Training...: 36% 1603/4393 [2:28:47<3:33:20, 4.59s/it] + Training...: 37% 1604/4393 [2:28:53<3:53:46, 5.03s/it] + Training...: 37% 1605/4393 [2:29:00<4:08:44, 5.35s/it] + Training...: 37% 1606/4393 [2:29:05<4:15:52, 5.51s/it] + Training...: 37% 1607/4393 [2:29:11<4:20:22, 5.61s/it] + Training...: 37% 1608/4393 [2:29:17<4:23:01, 5.67s/it] + Training...: 37% 1609/4393 [2:29:23<4:24:48, 5.71s/it] + Training...: 37% 1610/4393 [2:29:29<4:25:45, 5.73s/it] + Training...: 37% 1611/4393 [2:29:34<4:25:39, 5.73s/it] + Training...: 37% 1612/4393 [2:29:40<4:24:34, 5.71s/it] + Training...: 37% 1613/4393 [2:29:46<4:23:27, 5.69s/it] +  + Training...: 37% 1613/4393 [2:29:52<4:23:27, 5.69s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [13:59:25<57:16:35, 20619.59s/it] + Training...: 37% 1614/4393 [2:29:52<4:35:00, 5.94s/it] + Training...: 37% 1615/4393 [2:29:58<4:30:59, 5.85s/it] + Training...: 37% 1616/4393 [2:30:03<4:26:07, 5.75s/it] + Training...: 37% 1617/4393 [2:30:09<4:22:50, 5.68s/it] + Training...: 37% 1618/4393 [2:30:14<4:19:16, 5.61s/it] + Training...: 37% 1619/4393 [2:30:20<4:16:19, 5.54s/it] + Training...: 37% 1620/4393 [2:30:25<4:16:49, 5.56s/it] + Training...: 37% 1621/4393 [2:30:31<4:15:10, 5.52s/it] + Training...: 37% 1622/4393 [2:30:36<4:12:13, 5.46s/it] + Training...: 37% 1623/4393 [2:30:41<4:09:04, 5.39s/it] + Training...: 37% 1624/4393 [2:30:47<4:06:17, 5.34s/it] + Training...: 37% 1625/4393 [2:30:52<4:05:16, 5.32s/it] + Training...: 37% 1626/4393 [2:30:57<4:02:28, 5.26s/it] + Training...: 37% 1627/4393 [2:31:02<3:59:51, 5.20s/it] + Training...: 37% 1628/4393 [2:31:07<3:56:27, 5.13s/it] + Training...: 37% 1629/4393 [2:31:12<3:54:21, 5.09s/it] + Training...: 37% 1630/4393 [2:31:17<3:50:57, 5.02s/it] + Training...: 37% 1631/4393 [2:31:22<3:48:14, 4.96s/it] + Training...: 37% 1632/4393 [2:31:26<3:45:26, 4.90s/it] + Training...: 37% 1633/4393 [2:31:31<3:42:27, 4.84s/it] + Training...: 37% 1634/4393 [2:31:36<3:39:19, 4.77s/it] + Training...: 37% 1635/4393 [2:31:40<3:34:59, 4.68s/it] + Training...: 37% 1636/4393 [2:31:44<3:29:28, 4.56s/it] + Training...: 37% 1637/4393 [2:31:49<3:23:50, 4.44s/it] + Training...: 37% 1638/4393 [2:31:53<3:17:55, 4.31s/it] +  + Training...: 37% 1638/4393 [2:31:57<3:17:55, 4.31s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:01:29<57:16:35, 20619.59s/it] + Training...: 37% 1639/4393 [2:31:57<3:19:39, 4.35s/it] + Training...: 37% 1640/4393 [2:32:01<3:09:25, 4.13s/it] + Training...: 37% 1641/4393 [2:32:04<2:59:38, 3.92s/it] + Training...: 37% 1642/4393 [2:32:07<2:49:38, 3.70s/it] + Training...: 37% 1643/4393 [2:32:10<2:38:05, 3.45s/it] + Training...: 37% 1644/4393 [2:32:13<2:25:25, 3.17s/it] + Training...: 37% 1645/4393 [2:32:15<2:12:49, 2.90s/it] + Training...: 37% 1646/4393 [2:32:17<2:00:52, 2.64s/it] + Training...: 37% 1647/4393 [2:32:19<1:49:04, 2.38s/it] + Training...: 38% 1648/4393 [2:32:20<1:37:04, 2.12s/it] + Training...: 38% 1649/4393 [2:32:21<1:25:03, 1.86s/it] + Training...: 38% 1650/4393 [2:32:22<1:12:50, 1.59s/it] + Training...: 38% 1651/4393 [2:32:29<2:16:29, 2.99s/it] + Training...: 38% 1652/4393 [2:32:35<3:00:56, 3.96s/it] + Training...: 38% 1653/4393 [2:32:41<3:30:13, 4.60s/it] + Training...: 38% 1654/4393 [2:32:47<3:47:15, 4.98s/it] + Training...: 38% 1655/4393 [2:32:53<3:59:42, 5.25s/it] + Training...: 38% 1656/4393 [2:32:59<4:07:39, 5.43s/it] + Training...: 38% 1657/4393 [2:33:04<4:13:32, 5.56s/it] + Training...: 38% 1658/4393 [2:33:10<4:16:59, 5.64s/it] + Training...: 38% 1659/4393 [2:33:16<4:19:11, 5.69s/it] + Training...: 38% 1660/4393 [2:33:22<4:20:22, 5.72s/it] + Training...: 38% 1661/4393 [2:33:28<4:21:35, 5.75s/it] + Training...: 38% 1662/4393 [2:33:33<4:20:50, 5.73s/it] + Training...: 38% 1663/4393 [2:33:39<4:19:16, 5.70s/it] +  + Training...: 38% 1663/4393 [2:33:45<4:19:16, 5.70s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:03:18<57:16:35, 20619.59s/it] + Training...: 38% 1664/4393 [2:33:45<4:28:04, 5.89s/it] + Training...: 38% 1665/4393 [2:33:51<4:24:16, 5.81s/it] + Training...: 38% 1666/4393 [2:33:56<4:20:01, 5.72s/it] + Training...: 38% 1667/4393 [2:34:02<4:16:32, 5.65s/it] + Training...: 38% 1668/4393 [2:34:07<4:12:51, 5.57s/it] + Training...: 38% 1669/4393 [2:34:13<4:10:31, 5.52s/it] + Training...: 38% 1670/4393 [2:34:18<4:08:08, 5.47s/it] + Training...: 38% 1671/4393 [2:34:23<4:05:32, 5.41s/it] + Training...: 38% 1672/4393 [2:34:29<4:03:35, 5.37s/it] + Training...: 38% 1673/4393 [2:34:34<4:03:00, 5.36s/it] + Training...: 38% 1674/4393 [2:34:39<4:02:09, 5.34s/it] + Training...: 38% 1675/4393 [2:34:45<4:00:37, 5.31s/it] + Training...: 38% 1676/4393 [2:34:50<3:58:43, 5.27s/it] + Training...: 38% 1677/4393 [2:34:55<3:55:49, 5.21s/it] + Training...: 38% 1678/4393 [2:35:00<3:52:54, 5.15s/it] + Training...: 38% 1679/4393 [2:35:05<3:49:31, 5.07s/it] + Training...: 38% 1680/4393 [2:35:10<3:46:25, 5.01s/it] + Training...: 38% 1681/4393 [2:35:14<3:42:55, 4.93s/it] + Training...: 38% 1682/4393 [2:35:19<3:39:18, 4.85s/it] + Training...: 38% 1683/4393 [2:35:24<3:35:36, 4.77s/it] + Training...: 38% 1684/4393 [2:35:28<3:31:57, 4.69s/it] + Training...: 38% 1685/4393 [2:35:33<3:29:01, 4.63s/it] + Training...: 38% 1686/4393 [2:35:37<3:24:56, 4.54s/it] + Training...: 38% 1687/4393 [2:35:41<3:18:30, 4.40s/it] + Training...: 38% 1688/4393 [2:35:45<3:11:29, 4.25s/it] +  + Training...: 38% 1688/4393 [2:35:49<3:11:29, 4.25s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:05:22<57:16:35, 20619.59s/it] + Training...: 38% 1689/4393 [2:35:49<3:12:49, 4.28s/it] + Training...: 38% 1690/4393 [2:35:53<3:02:17, 4.05s/it] + Training...: 38% 1691/4393 [2:35:56<2:51:43, 3.81s/it] + Training...: 39% 1692/4393 [2:35:59<2:40:51, 3.57s/it] + Training...: 39% 1693/4393 [2:36:02<2:30:10, 3.34s/it] + Training...: 39% 1694/4393 [2:36:04<2:18:42, 3.08s/it] + Training...: 39% 1695/4393 [2:36:06<2:07:00, 2.82s/it] + Training...: 39% 1696/4393 [2:36:08<1:55:34, 2.57s/it] + Training...: 39% 1697/4393 [2:36:10<1:44:37, 2.33s/it] + Training...: 39% 1698/4393 [2:36:12<1:33:32, 2.08s/it] + Training...: 39% 1699/4393 [2:36:13<1:22:02, 1.83s/it] + Training...: 39% 1700/4393 [2:36:14<1:10:32, 1.57s/it] + Training...: 39% 1701/4393 [2:36:20<2:14:02, 2.99s/it] + Training...: 39% 1702/4393 [2:36:26<2:57:57, 3.97s/it] + Training...: 39% 1703/4393 [2:36:33<3:26:31, 4.61s/it] + Training...: 39% 1704/4393 [2:36:38<3:43:55, 5.00s/it] + Training...: 39% 1705/4393 [2:36:44<3:56:59, 5.29s/it] + Training...: 39% 1706/4393 [2:36:50<4:04:36, 5.46s/it] + Training...: 39% 1707/4393 [2:36:56<4:11:15, 5.61s/it] + Training...: 39% 1708/4393 [2:37:02<4:14:16, 5.68s/it] + Training...: 39% 1709/4393 [2:37:08<4:15:48, 5.72s/it] + Training...: 39% 1710/4393 [2:37:14<4:15:36, 5.72s/it] + Training...: 39% 1711/4393 [2:37:19<4:16:31, 5.74s/it] + Training...: 39% 1712/4393 [2:37:25<4:16:26, 5.74s/it] + Training...: 39% 1713/4393 [2:37:31<4:18:30, 5.79s/it] +  + Training...: 39% 1713/4393 [2:37:38<4:18:30, 5.79s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:07:10<57:16:35, 20619.59s/it] + Training...: 39% 1714/4393 [2:37:38<4:28:59, 6.02s/it] + Training...: 39% 1715/4393 [2:37:43<4:23:42, 5.91s/it] + Training...: 39% 1716/4393 [2:37:49<4:18:50, 5.80s/it] + Training...: 39% 1717/4393 [2:37:54<4:15:04, 5.72s/it] + Training...: 39% 1718/4393 [2:38:00<4:11:09, 5.63s/it] + Training...: 39% 1719/4393 [2:38:05<4:09:29, 5.60s/it] + Training...: 39% 1720/4393 [2:38:11<4:07:52, 5.56s/it] + Training...: 39% 1721/4393 [2:38:16<4:06:03, 5.53s/it] + Training...: 39% 1722/4393 [2:38:22<4:03:46, 5.48s/it] + Training...: 39% 1723/4393 [2:38:27<4:01:53, 5.44s/it] + Training...: 39% 1724/4393 [2:38:32<4:01:58, 5.44s/it] + Training...: 39% 1725/4393 [2:38:38<4:01:43, 5.44s/it] + Training...: 39% 1726/4393 [2:38:43<4:00:28, 5.41s/it] + Training...: 39% 1727/4393 [2:38:48<3:59:14, 5.38s/it] + Training...: 39% 1728/4393 [2:38:54<3:57:20, 5.34s/it] + Training...: 39% 1729/4393 [2:38:59<3:54:55, 5.29s/it] + Training...: 39% 1730/4393 [2:39:04<3:52:38, 5.24s/it] + Training...: 39% 1731/4393 [2:39:09<3:49:36, 5.18s/it] + Training...: 39% 1732/4393 [2:39:14<3:47:39, 5.13s/it] + Training...: 39% 1733/4393 [2:39:19<3:44:16, 5.06s/it] + Training...: 39% 1734/4393 [2:39:24<3:40:42, 4.98s/it] + Training...: 39% 1735/4393 [2:39:29<3:37:18, 4.91s/it] + Training...: 40% 1736/4393 [2:39:33<3:33:44, 4.83s/it] + Training...: 40% 1737/4393 [2:39:38<3:29:41, 4.74s/it] + Training...: 40% 1738/4393 [2:39:42<3:24:01, 4.61s/it] +  + Training...: 40% 1738/4393 [2:39:47<3:24:01, 4.61s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:09:19<57:16:35, 20619.59s/it] + Training...: 40% 1739/4393 [2:39:47<3:26:52, 4.68s/it] + Training...: 40% 1740/4393 [2:39:51<3:15:53, 4.43s/it] + Training...: 40% 1741/4393 [2:39:54<3:05:34, 4.20s/it] + Training...: 40% 1742/4393 [2:39:58<2:54:30, 3.95s/it] + Training...: 40% 1743/4393 [2:40:01<2:42:48, 3.69s/it] + Training...: 40% 1744/4393 [2:40:04<2:30:19, 3.40s/it] + Training...: 40% 1745/4393 [2:40:06<2:17:30, 3.12s/it] + Training...: 40% 1746/4393 [2:40:08<2:05:05, 2.84s/it] + Training...: 40% 1747/4393 [2:40:10<1:52:34, 2.55s/it] + Training...: 40% 1748/4393 [2:40:12<1:40:27, 2.28s/it] + Training...: 40% 1749/4393 [2:40:13<1:28:05, 2.00s/it] + Training...: 40% 1750/4393 [2:40:14<1:15:19, 1.71s/it] + Training...: 40% 1751/4393 [2:40:20<2:14:03, 3.04s/it] + Training...: 40% 1752/4393 [2:40:26<2:55:05, 3.98s/it] + Training...: 40% 1753/4393 [2:40:32<3:21:48, 4.59s/it] + Training...: 40% 1754/4393 [2:40:38<3:40:14, 5.01s/it] + Training...: 40% 1755/4393 [2:40:44<3:51:55, 5.28s/it] + Training...: 40% 1756/4393 [2:40:50<4:00:15, 5.47s/it] + Training...: 40% 1757/4393 [2:40:56<4:05:41, 5.59s/it] + Training...: 40% 1758/4393 [2:41:02<4:08:48, 5.67s/it] + Training...: 40% 1759/4393 [2:41:08<4:09:55, 5.69s/it] + Training...: 40% 1760/4393 [2:41:13<4:09:56, 5.70s/it] + Training...: 40% 1761/4393 [2:41:19<4:09:58, 5.70s/it] + Training...: 40% 1762/4393 [2:41:25<4:10:10, 5.71s/it] + Training...: 40% 1763/4393 [2:41:31<4:10:49, 5.72s/it] +  + Training...: 40% 1763/4393 [2:41:37<4:10:49, 5.72s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:11:09<57:16:35, 20619.59s/it] + Training...: 40% 1764/4393 [2:41:37<4:20:51, 5.95s/it] + Training...: 40% 1765/4393 [2:41:43<4:15:16, 5.83s/it] + Training...: 40% 1766/4393 [2:41:48<4:11:17, 5.74s/it] + Training...: 40% 1767/4393 [2:41:54<4:08:29, 5.68s/it] + Training...: 40% 1768/4393 [2:41:59<4:05:40, 5.62s/it] + Training...: 40% 1769/4393 [2:42:05<4:04:27, 5.59s/it] + Training...: 40% 1770/4393 [2:42:10<4:02:53, 5.56s/it] + Training...: 40% 1771/4393 [2:42:15<4:00:21, 5.50s/it] + Training...: 40% 1772/4393 [2:42:21<3:57:33, 5.44s/it] + Training...: 40% 1773/4393 [2:42:26<3:57:23, 5.44s/it] + Training...: 40% 1774/4393 [2:42:32<3:56:02, 5.41s/it] + Training...: 40% 1775/4393 [2:42:37<3:53:53, 5.36s/it] + Training...: 40% 1776/4393 [2:42:42<3:50:49, 5.29s/it] + Training...: 40% 1777/4393 [2:42:47<3:48:13, 5.23s/it] + Training...: 40% 1778/4393 [2:42:52<3:45:22, 5.17s/it] + Training...: 40% 1779/4393 [2:42:57<3:43:26, 5.13s/it] + Training...: 41% 1780/4393 [2:43:02<3:40:05, 5.05s/it] + Training...: 41% 1781/4393 [2:43:07<3:36:45, 4.98s/it] + Training...: 41% 1782/4393 [2:43:11<3:33:04, 4.90s/it] + Training...: 41% 1783/4393 [2:43:16<3:29:34, 4.82s/it] + Training...: 41% 1784/4393 [2:43:21<3:25:31, 4.73s/it] + Training...: 41% 1785/4393 [2:43:25<3:21:27, 4.63s/it] + Training...: 41% 1786/4393 [2:43:29<3:16:33, 4.52s/it] + Training...: 41% 1787/4393 [2:43:33<3:10:37, 4.39s/it] + Training...: 41% 1788/4393 [2:43:37<3:05:11, 4.27s/it] +  + Training...: 41% 1788/4393 [2:43:42<3:05:11, 4.27s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:13:14<57:16:35, 20619.59s/it] + Training...: 41% 1789/4393 [2:43:42<3:06:24, 4.30s/it] + Training...: 41% 1790/4393 [2:43:45<2:56:24, 4.07s/it] + Training...: 41% 1791/4393 [2:43:49<2:45:33, 3.82s/it] + Training...: 41% 1792/4393 [2:43:51<2:34:32, 3.57s/it] + Training...: 41% 1793/4393 [2:43:54<2:23:41, 3.32s/it] + Training...: 41% 1794/4393 [2:43:57<2:12:47, 3.07s/it] + Training...: 41% 1795/4393 [2:43:59<2:02:10, 2.82s/it] + Training...: 41% 1796/4393 [2:44:01<1:51:35, 2.58s/it] + Training...: 41% 1797/4393 [2:44:03<1:41:01, 2.33s/it] + Training...: 41% 1798/4393 [2:44:04<1:30:43, 2.10s/it] + Training...: 41% 1799/4393 [2:44:06<1:20:34, 1.86s/it] + Training...: 41% 1800/4393 [2:44:07<1:09:31, 1.61s/it] + Training...: 41% 1801/4393 [2:44:13<2:09:12, 2.99s/it] + Training...: 41% 1802/4393 [2:44:19<2:52:25, 3.99s/it] + Training...: 41% 1803/4393 [2:44:25<3:19:43, 4.63s/it] + Training...: 41% 1804/4393 [2:44:31<3:37:29, 5.04s/it] + Training...: 41% 1805/4393 [2:44:37<3:50:11, 5.34s/it] + Training...: 41% 1806/4393 [2:44:43<3:57:25, 5.51s/it] + Training...: 41% 1807/4393 [2:44:49<4:01:55, 5.61s/it] + Training...: 41% 1808/4393 [2:44:55<4:03:50, 5.66s/it] + Training...: 41% 1809/4393 [2:45:01<4:06:38, 5.73s/it] + Training...: 41% 1810/4393 [2:45:06<4:06:44, 5.73s/it] + Training...: 41% 1811/4393 [2:45:12<4:07:02, 5.74s/it] + Training...: 41% 1812/4393 [2:45:18<4:06:06, 5.72s/it] + Training...: 41% 1813/4393 [2:45:24<4:05:40, 5.71s/it] +  + Training...: 41% 1813/4393 [2:45:30<4:05:40, 5.71s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:15:02<57:16:35, 20619.59s/it] + Training...: 41% 1814/4393 [2:45:30<4:14:53, 5.93s/it] + Training...: 41% 1815/4393 [2:45:36<4:11:59, 5.86s/it] + Training...: 41% 1816/4393 [2:45:41<4:07:52, 5.77s/it] + Training...: 41% 1817/4393 [2:45:47<4:05:12, 5.71s/it] + Training...: 41% 1818/4393 [2:45:52<4:01:46, 5.63s/it] + Training...: 41% 1819/4393 [2:45:58<3:59:53, 5.59s/it] + Training...: 41% 1820/4393 [2:46:03<3:57:35, 5.54s/it] + Training...: 41% 1821/4393 [2:46:09<3:56:12, 5.51s/it] + Training...: 41% 1822/4393 [2:46:14<3:53:53, 5.46s/it] + Training...: 41% 1823/4393 [2:46:19<3:52:14, 5.42s/it] + Training...: 42% 1824/4393 [2:46:25<3:49:40, 5.36s/it] + Training...: 42% 1825/4393 [2:46:30<3:48:21, 5.34s/it] + Training...: 42% 1826/4393 [2:46:35<3:48:06, 5.33s/it] + Training...: 42% 1827/4393 [2:46:40<3:46:53, 5.31s/it] + Training...: 42% 1828/4393 [2:46:45<3:43:37, 5.23s/it] + Training...: 42% 1829/4393 [2:46:51<3:41:25, 5.18s/it] + Training...: 42% 1830/4393 [2:46:55<3:38:35, 5.12s/it] + Training...: 42% 1831/4393 [2:47:00<3:35:35, 5.05s/it] + Training...: 42% 1832/4393 [2:47:05<3:32:28, 4.98s/it] + Training...: 42% 1833/4393 [2:47:10<3:29:34, 4.91s/it] + Training...: 42% 1834/4393 [2:47:15<3:25:45, 4.82s/it] + Training...: 42% 1835/4393 [2:47:19<3:22:27, 4.75s/it] + Training...: 42% 1836/4393 [2:47:24<3:17:59, 4.65s/it] + Training...: 42% 1837/4393 [2:47:28<3:13:52, 4.55s/it] + Training...: 42% 1838/4393 [2:47:32<3:08:20, 4.42s/it] +  + Training...: 42% 1838/4393 [2:47:37<3:08:20, 4.42s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:17:09<57:16:35, 20619.59s/it] + Training...: 42% 1839/4393 [2:47:37<3:10:47, 4.48s/it] + Training...: 42% 1840/4393 [2:47:40<3:00:50, 4.25s/it] + Training...: 42% 1841/4393 [2:47:44<2:50:50, 4.02s/it] + Training...: 42% 1842/4393 [2:47:47<2:40:16, 3.77s/it] + Training...: 42% 1843/4393 [2:47:50<2:30:07, 3.53s/it] + Training...: 42% 1844/4393 [2:47:53<2:18:29, 3.26s/it] + Training...: 42% 1845/4393 [2:47:55<2:06:00, 2.97s/it] + Training...: 42% 1846/4393 [2:47:57<1:54:11, 2.69s/it] + Training...: 42% 1847/4393 [2:47:59<1:42:50, 2.42s/it] + Training...: 42% 1848/4393 [2:48:00<1:31:49, 2.16s/it] + Training...: 42% 1849/4393 [2:48:02<1:20:37, 1.90s/it] + Training...: 42% 1850/4393 [2:48:03<1:09:27, 1.64s/it] + Training...: 42% 1851/4393 [2:48:09<2:07:26, 3.01s/it] + Training...: 42% 1852/4393 [2:48:15<2:48:25, 3.98s/it] + Training...: 42% 1853/4393 [2:48:21<3:15:18, 4.61s/it] + Training...: 42% 1854/4393 [2:48:27<3:31:58, 5.01s/it] + Training...: 42% 1855/4393 [2:48:33<3:43:52, 5.29s/it] + Training...: 42% 1856/4393 [2:48:39<3:50:52, 5.46s/it] + Training...: 42% 1857/4393 [2:48:45<3:56:22, 5.59s/it] + Training...: 42% 1858/4393 [2:48:51<3:58:54, 5.65s/it] + Training...: 42% 1859/4393 [2:48:56<4:00:23, 5.69s/it] + Training...: 42% 1860/4393 [2:49:02<4:02:12, 5.74s/it] + Training...: 42% 1861/4393 [2:49:08<4:03:38, 5.77s/it] + Training...: 42% 1862/4393 [2:49:14<4:02:40, 5.75s/it] + Training...: 42% 1863/4393 [2:49:19<4:01:36, 5.73s/it] +  + Training...: 42% 1863/4393 [2:49:26<4:01:36, 5.73s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:18:58<57:16:35, 20619.59s/it] + Training...: 42% 1864/4393 [2:49:26<4:11:17, 5.96s/it] + Training...: 42% 1865/4393 [2:49:32<4:07:04, 5.86s/it] + Training...: 42% 1866/4393 [2:49:37<4:03:33, 5.78s/it] + Training...: 42% 1867/4393 [2:49:43<3:59:54, 5.70s/it] + Training...: 43% 1868/4393 [2:49:48<3:56:19, 5.62s/it] + Training...: 43% 1869/4393 [2:49:54<3:54:11, 5.57s/it] + Training...: 43% 1870/4393 [2:49:59<3:51:26, 5.50s/it] + Training...: 43% 1871/4393 [2:50:04<3:49:26, 5.46s/it] + Training...: 43% 1872/4393 [2:50:10<3:48:01, 5.43s/it] + Training...: 43% 1873/4393 [2:50:15<3:46:32, 5.39s/it] + Training...: 43% 1874/4393 [2:50:20<3:44:38, 5.35s/it] + Training...: 43% 1875/4393 [2:50:25<3:42:53, 5.31s/it] + Training...: 43% 1876/4393 [2:50:31<3:40:19, 5.25s/it] + Training...: 43% 1877/4393 [2:50:36<3:37:51, 5.20s/it] + Training...: 43% 1878/4393 [2:50:41<3:35:44, 5.15s/it] + Training...: 43% 1879/4393 [2:50:46<3:33:16, 5.09s/it] + Training...: 43% 1880/4393 [2:50:51<3:31:23, 5.05s/it] + Training...: 43% 1881/4393 [2:50:55<3:29:37, 5.01s/it] + Training...: 43% 1882/4393 [2:51:00<3:26:52, 4.94s/it] + Training...: 43% 1883/4393 [2:51:05<3:23:33, 4.87s/it] + Training...: 43% 1884/4393 [2:51:10<3:20:45, 4.80s/it] + Training...: 43% 1885/4393 [2:51:14<3:16:40, 4.71s/it] + Training...: 43% 1886/4393 [2:51:18<3:12:06, 4.60s/it] + Training...: 43% 1887/4393 [2:51:23<3:06:53, 4.47s/it] + Training...: 43% 1888/4393 [2:51:27<3:00:54, 4.33s/it] +  + Training...: 43% 1888/4393 [2:51:31<3:00:54, 4.33s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:21:03<57:16:35, 20619.59s/it] + Training...: 43% 1889/4393 [2:51:31<3:01:12, 4.34s/it] + Training...: 43% 1890/4393 [2:51:35<2:51:34, 4.11s/it] + Training...: 43% 1891/4393 [2:51:38<2:41:07, 3.86s/it] + Training...: 43% 1892/4393 [2:51:41<2:30:46, 3.62s/it] + Training...: 43% 1893/4393 [2:51:44<2:20:10, 3.36s/it] + Training...: 43% 1894/4393 [2:51:46<2:08:52, 3.09s/it] + Training...: 43% 1895/4393 [2:51:48<1:57:38, 2.83s/it] + Training...: 43% 1896/4393 [2:51:50<1:46:11, 2.55s/it] + Training...: 43% 1897/4393 [2:51:52<1:34:50, 2.28s/it] + Training...: 43% 1898/4393 [2:51:53<1:23:53, 2.02s/it] + Training...: 43% 1899/4393 [2:51:54<1:13:29, 1.77s/it] + Training...: 43% 1900/4393 [2:51:55<1:03:36, 1.53s/it] + Training...: 43% 1901/4393 [2:52:02<2:01:14, 2.92s/it] + Training...: 43% 1902/4393 [2:52:08<2:41:41, 3.89s/it] + Training...: 43% 1903/4393 [2:52:14<3:08:25, 4.54s/it] + Training...: 43% 1904/4393 [2:52:20<3:25:40, 4.96s/it] + Training...: 43% 1905/4393 [2:52:26<3:38:03, 5.26s/it] + Training...: 43% 1906/4393 [2:52:32<3:46:02, 5.45s/it] + Training...: 43% 1907/4393 [2:52:37<3:51:10, 5.58s/it] + Training...: 43% 1908/4393 [2:52:43<3:53:30, 5.64s/it] + Training...: 43% 1909/4393 [2:52:49<3:56:11, 5.71s/it] + Training...: 43% 1910/4393 [2:52:55<3:56:52, 5.72s/it] + Training...: 44% 1911/4393 [2:53:01<3:59:27, 5.79s/it] + Training...: 44% 1912/4393 [2:53:07<3:59:08, 5.78s/it] + Training...: 44% 1913/4393 [2:53:12<3:57:00, 5.73s/it] +  + Training...: 44% 1913/4393 [2:53:19<3:57:00, 5.73s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:22:51<57:16:35, 20619.59s/it] + Training...: 44% 1914/4393 [2:53:19<4:06:14, 5.96s/it] + Training...: 44% 1915/4393 [2:53:24<4:01:44, 5.85s/it] + Training...: 44% 1916/4393 [2:53:30<3:58:22, 5.77s/it] + Training...: 44% 1917/4393 [2:53:35<3:55:47, 5.71s/it] + Training...: 44% 1918/4393 [2:53:41<3:53:23, 5.66s/it] + Training...: 44% 1919/4393 [2:53:47<3:52:09, 5.63s/it] + Training...: 44% 1920/4393 [2:53:52<3:49:25, 5.57s/it] + Training...: 44% 1921/4393 [2:53:57<3:47:19, 5.52s/it] + Training...: 44% 1922/4393 [2:54:03<3:46:16, 5.49s/it] + Training...: 44% 1923/4393 [2:54:08<3:45:18, 5.47s/it] + Training...: 44% 1924/4393 [2:54:13<3:42:16, 5.40s/it] + Training...: 44% 1925/4393 [2:54:19<3:40:02, 5.35s/it] + Training...: 44% 1926/4393 [2:54:24<3:36:59, 5.28s/it] + Training...: 44% 1927/4393 [2:54:29<3:34:35, 5.22s/it] + Training...: 44% 1928/4393 [2:54:34<3:31:51, 5.16s/it] + Training...: 44% 1929/4393 [2:54:39<3:29:19, 5.10s/it] + Training...: 44% 1930/4393 [2:54:44<3:26:55, 5.04s/it] + Training...: 44% 1931/4393 [2:54:49<3:24:48, 4.99s/it] + Training...: 44% 1932/4393 [2:54:53<3:22:38, 4.94s/it] + Training...: 44% 1933/4393 [2:54:58<3:19:42, 4.87s/it] + Training...: 44% 1934/4393 [2:55:03<3:15:09, 4.76s/it] + Training...: 44% 1935/4393 [2:55:07<3:11:21, 4.67s/it] + Training...: 44% 1936/4393 [2:55:11<3:06:30, 4.55s/it] + Training...: 44% 1937/4393 [2:55:16<3:01:34, 4.44s/it] + Training...: 44% 1938/4393 [2:55:20<2:57:24, 4.34s/it] +  + Training...: 44% 1938/4393 [2:55:24<2:57:24, 4.34s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:24:56<57:16:35, 20619.59s/it] + Training...: 44% 1939/4393 [2:55:24<2:57:39, 4.34s/it] + Training...: 44% 1940/4393 [2:55:28<2:48:35, 4.12s/it] + Training...: 44% 1941/4393 [2:55:31<2:39:08, 3.89s/it] + Training...: 44% 1942/4393 [2:55:34<2:29:24, 3.66s/it] + Training...: 44% 1943/4393 [2:55:37<2:19:31, 3.42s/it] + Training...: 44% 1944/4393 [2:55:40<2:08:47, 3.16s/it] + Training...: 44% 1945/4393 [2:55:42<1:58:32, 2.91s/it] + Training...: 44% 1946/4393 [2:55:44<1:47:49, 2.64s/it] + Training...: 44% 1947/4393 [2:55:46<1:36:50, 2.38s/it] + Training...: 44% 1948/4393 [2:55:47<1:26:14, 2.12s/it] + Training...: 44% 1949/4393 [2:55:48<1:16:04, 1.87s/it] + Training...: 44% 1950/4393 [2:55:49<1:05:35, 1.61s/it] + Training...: 44% 1951/4393 [2:55:56<2:01:45, 2.99s/it] + Training...: 44% 1952/4393 [2:56:02<2:39:55, 3.93s/it] + Training...: 44% 1953/4393 [2:56:08<3:06:22, 4.58s/it] + Training...: 44% 1954/4393 [2:56:14<3:25:35, 5.06s/it] + Training...: 45% 1955/4393 [2:56:20<3:38:57, 5.39s/it] + Training...: 45% 1956/4393 [2:56:26<3:45:01, 5.54s/it] + Training...: 45% 1957/4393 [2:56:32<3:48:59, 5.64s/it] + Training...: 45% 1958/4393 [2:56:38<3:51:17, 5.70s/it] + Training...: 45% 1959/4393 [2:56:44<3:51:48, 5.71s/it] + Training...: 45% 1960/4393 [2:56:49<3:52:51, 5.74s/it] + Training...: 45% 1961/4393 [2:56:55<3:52:09, 5.73s/it] + Training...: 45% 1962/4393 [2:57:01<3:51:28, 5.71s/it] + Training...: 45% 1963/4393 [2:57:06<3:51:36, 5.72s/it] +  + Training...: 45% 1963/4393 [2:57:13<3:51:36, 5.72s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:26:45<57:16:35, 20619.59s/it] + Training...: 45% 1964/4393 [2:57:13<3:59:49, 5.92s/it] + Training...: 45% 1965/4393 [2:57:19<3:56:36, 5.85s/it] + Training...: 45% 1966/4393 [2:57:24<3:52:54, 5.76s/it] + Training...: 45% 1967/4393 [2:57:30<3:49:32, 5.68s/it] + Training...: 45% 1968/4393 [2:57:35<3:47:23, 5.63s/it] + Training...: 45% 1969/4393 [2:57:41<3:48:09, 5.65s/it] + Training...: 45% 1970/4393 [2:57:46<3:45:29, 5.58s/it] + Training...: 45% 1971/4393 [2:57:52<3:43:49, 5.54s/it] + Training...: 45% 1972/4393 [2:57:57<3:41:31, 5.49s/it] + Training...: 45% 1973/4393 [2:58:02<3:38:52, 5.43s/it] + Training...: 45% 1974/4393 [2:58:08<3:36:32, 5.37s/it] + Training...: 45% 1975/4393 [2:58:13<3:34:05, 5.31s/it] + Training...: 45% 1976/4393 [2:58:18<3:31:48, 5.26s/it] + Training...: 45% 1977/4393 [2:58:23<3:29:45, 5.21s/it] + Training...: 45% 1978/4393 [2:58:28<3:27:08, 5.15s/it] + Training...: 45% 1979/4393 [2:58:33<3:24:53, 5.09s/it] + Training...: 45% 1980/4393 [2:58:38<3:22:20, 5.03s/it] + Training...: 45% 1981/4393 [2:58:43<3:20:25, 4.99s/it] + Training...: 45% 1982/4393 [2:58:47<3:17:37, 4.92s/it] + Training...: 45% 1983/4393 [2:58:52<3:14:22, 4.84s/it] + Training...: 45% 1984/4393 [2:58:57<3:11:29, 4.77s/it] + Training...: 45% 1985/4393 [2:59:01<3:07:35, 4.67s/it] + Training...: 45% 1986/4393 [2:59:05<3:02:39, 4.55s/it] + Training...: 45% 1987/4393 [2:59:10<2:57:19, 4.42s/it] + Training...: 45% 1988/4393 [2:59:13<2:50:57, 4.27s/it] +  + Training...: 45% 1988/4393 [2:59:18<2:50:57, 4.27s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:28:50<57:16:35, 20619.59s/it] + Training...: 45% 1989/4393 [2:59:18<2:51:28, 4.28s/it] + Training...: 45% 1990/4393 [2:59:21<2:41:45, 4.04s/it] + Training...: 45% 1991/4393 [2:59:24<2:32:19, 3.81s/it] + Training...: 45% 1992/4393 [2:59:27<2:22:24, 3.56s/it] + Training...: 45% 1993/4393 [2:59:30<2:12:50, 3.32s/it] + Training...: 45% 1994/4393 [2:59:33<2:02:08, 3.05s/it] + Training...: 45% 1995/4393 [2:59:35<1:51:47, 2.80s/it] + Training...: 45% 1996/4393 [2:59:37<1:41:54, 2.55s/it] + Training...: 45% 1997/4393 [2:59:39<1:32:15, 2.31s/it] + Training...: 45% 1998/4393 [2:59:40<1:22:24, 2.06s/it] + Training...: 46% 1999/4393 [2:59:41<1:12:38, 1.82s/it] + Training...: 46% 2000/4393 [2:59:42<1:02:17, 1.56s/it] + Training...: 46% 2001/4393 [2:59:49<1:57:46, 2.95s/it] + Training...: 46% 2002/4393 [2:59:55<2:35:54, 3.91s/it] + Training...: 46% 2003/4393 [3:00:01<3:03:35, 4.61s/it] + Training...: 46% 2004/4393 [3:00:07<3:20:20, 5.03s/it] + Training...: 46% 2005/4393 [3:00:13<3:30:36, 5.29s/it] + Training...: 46% 2006/4393 [3:00:19<3:37:58, 5.48s/it] + Training...: 46% 2007/4393 [3:00:25<3:43:06, 5.61s/it] + Training...: 46% 2008/4393 [3:00:30<3:44:45, 5.65s/it] + Training...: 46% 2009/4393 [3:00:36<3:46:34, 5.70s/it] + Training...: 46% 2010/4393 [3:00:42<3:46:49, 5.71s/it] + Training...: 46% 2011/4393 [3:00:48<3:47:12, 5.72s/it] + Training...: 46% 2012/4393 [3:00:53<3:46:26, 5.71s/it] + Training...: 46% 2013/4393 [3:00:59<3:45:16, 5.68s/it] +  + Training...: 46% 2013/4393 [3:01:05<3:45:16, 5.68s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:30:38<57:16:35, 20619.59s/it] + Training...: 46% 2014/4393 [3:01:05<3:53:58, 5.90s/it] + Training...: 46% 2015/4393 [3:01:11<3:49:44, 5.80s/it] + Training...: 46% 2016/4393 [3:01:17<3:46:57, 5.73s/it] + Training...: 46% 2017/4393 [3:01:22<3:45:18, 5.69s/it] + Training...: 46% 2018/4393 [3:01:28<3:42:58, 5.63s/it] + Training...: 46% 2019/4393 [3:01:33<3:40:19, 5.57s/it] + Training...: 46% 2020/4393 [3:01:38<3:38:08, 5.52s/it] + Training...: 46% 2021/4393 [3:01:44<3:36:18, 5.47s/it] + Training...: 46% 2022/4393 [3:01:49<3:33:54, 5.41s/it] + Training...: 46% 2023/4393 [3:01:54<3:32:39, 5.38s/it] + Training...: 46% 2024/4393 [3:02:00<3:30:57, 5.34s/it] + Training...: 46% 2025/4393 [3:02:05<3:30:24, 5.33s/it] + Training...: 46% 2026/4393 [3:02:10<3:29:30, 5.31s/it] + Training...: 46% 2027/4393 [3:02:15<3:28:00, 5.28s/it] + Training...: 46% 2028/4393 [3:02:20<3:25:33, 5.21s/it] + Training...: 46% 2029/4393 [3:02:25<3:23:04, 5.15s/it] + Training...: 46% 2030/4393 [3:02:30<3:20:27, 5.09s/it] + Training...: 46% 2031/4393 [3:02:35<3:17:44, 5.02s/it] + Training...: 46% 2032/4393 [3:02:40<3:14:44, 4.95s/it] + Training...: 46% 2033/4393 [3:02:45<3:11:57, 4.88s/it] + Training...: 46% 2034/4393 [3:02:49<3:09:46, 4.83s/it] + Training...: 46% 2035/4393 [3:02:54<3:06:31, 4.75s/it] + Training...: 46% 2036/4393 [3:02:58<3:02:43, 4.65s/it] + Training...: 46% 2037/4393 [3:03:03<2:58:27, 4.54s/it] + Training...: 46% 2038/4393 [3:03:07<2:53:04, 4.41s/it] +  + Training...: 46% 2038/4393 [3:03:11<2:53:04, 4.41s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:32:44<57:16:35, 20619.59s/it] + Training...: 46% 2039/4393 [3:03:11<2:54:40, 4.45s/it] + Training...: 46% 2040/4393 [3:03:15<2:45:36, 4.22s/it] + Training...: 46% 2041/4393 [3:03:19<2:36:42, 4.00s/it] + Training...: 46% 2042/4393 [3:03:22<2:26:47, 3.75s/it] + Training...: 47% 2043/4393 [3:03:25<2:17:01, 3.50s/it] + Training...: 47% 2044/4393 [3:03:27<2:07:24, 3.25s/it] + Training...: 47% 2045/4393 [3:03:30<1:56:44, 2.98s/it] + Training...: 47% 2046/4393 [3:03:32<1:46:17, 2.72s/it] + Training...: 47% 2047/4393 [3:03:34<1:35:44, 2.45s/it] + Training...: 47% 2048/4393 [3:03:35<1:25:35, 2.19s/it] + Training...: 47% 2049/4393 [3:03:36<1:14:53, 1.92s/it] + Training...: 47% 2050/4393 [3:03:37<1:03:59, 1.64s/it] + Training...: 47% 2051/4393 [3:03:44<1:57:57, 3.02s/it] + Training...: 47% 2052/4393 [3:03:50<2:34:02, 3.95s/it] + Training...: 47% 2053/4393 [3:03:56<2:57:45, 4.56s/it] + Training...: 47% 2054/4393 [3:04:02<3:13:44, 4.97s/it] + Training...: 47% 2055/4393 [3:04:08<3:25:22, 5.27s/it] + Training...: 47% 2056/4393 [3:04:14<3:32:15, 5.45s/it] + Training...: 47% 2057/4393 [3:04:20<3:37:52, 5.60s/it] + Training...: 47% 2058/4393 [3:04:25<3:40:02, 5.65s/it] + Training...: 47% 2059/4393 [3:04:31<3:42:04, 5.71s/it] + Training...: 47% 2060/4393 [3:04:37<3:43:16, 5.74s/it] + Training...: 47% 2061/4393 [3:04:43<3:44:44, 5.78s/it] + Training...: 47% 2062/4393 [3:04:49<3:43:19, 5.75s/it] + Training...: 47% 2063/4393 [3:04:54<3:42:12, 5.72s/it] +  + Training...: 47% 2063/4393 [3:05:01<3:42:12, 5.72s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:34:33<57:16:35, 20619.59s/it] + Training...: 47% 2064/4393 [3:05:01<3:49:57, 5.92s/it] + Training...: 47% 2065/4393 [3:05:06<3:46:03, 5.83s/it] + Training...: 47% 2066/4393 [3:05:12<3:42:15, 5.73s/it] + Training...: 47% 2067/4393 [3:05:17<3:39:19, 5.66s/it] + Training...: 47% 2068/4393 [3:05:23<3:35:54, 5.57s/it] + Training...: 47% 2069/4393 [3:05:28<3:34:42, 5.54s/it] + Training...: 47% 2070/4393 [3:05:33<3:31:54, 5.47s/it] + Training...: 47% 2071/4393 [3:05:39<3:31:16, 5.46s/it] + Training...: 47% 2072/4393 [3:05:44<3:29:51, 5.43s/it] + Training...: 47% 2073/4393 [3:05:49<3:28:46, 5.40s/it] + Training...: 47% 2074/4393 [3:05:55<3:26:51, 5.35s/it] + Training...: 47% 2075/4393 [3:06:00<3:25:28, 5.32s/it] + Training...: 47% 2076/4393 [3:06:05<3:23:21, 5.27s/it] + Training...: 47% 2077/4393 [3:06:10<3:21:54, 5.23s/it] + Training...: 47% 2078/4393 [3:06:15<3:18:44, 5.15s/it] + Training...: 47% 2079/4393 [3:06:20<3:17:57, 5.13s/it] + Training...: 47% 2080/4393 [3:06:25<3:16:27, 5.10s/it] + Training...: 47% 2081/4393 [3:06:30<3:13:35, 5.02s/it] + Training...: 47% 2082/4393 [3:06:35<3:09:39, 4.92s/it] + Training...: 47% 2083/4393 [3:06:39<3:06:41, 4.85s/it] + Training...: 47% 2084/4393 [3:06:44<3:03:45, 4.77s/it] + Training...: 47% 2085/4393 [3:06:49<3:00:21, 4.69s/it] + Training...: 47% 2086/4393 [3:06:53<2:56:36, 4.59s/it] + Training...: 48% 2087/4393 [3:06:57<2:53:15, 4.51s/it] + Training...: 48% 2088/4393 [3:07:01<2:48:40, 4.39s/it] +  + Training...: 48% 2088/4393 [3:07:06<2:48:40, 4.39s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:36:38<57:16:35, 20619.59s/it] + Training...: 48% 2089/4393 [3:07:06<2:49:02, 4.40s/it] + Training...: 48% 2090/4393 [3:07:09<2:39:14, 4.15s/it] + Training...: 48% 2091/4393 [3:07:13<2:29:48, 3.90s/it] + Training...: 48% 2092/4393 [3:07:16<2:19:05, 3.63s/it] + Training...: 48% 2093/4393 [3:07:18<2:08:02, 3.34s/it] + Training...: 48% 2094/4393 [3:07:21<1:56:57, 3.05s/it] + Training...: 48% 2095/4393 [3:07:23<1:46:45, 2.79s/it] + Training...: 48% 2096/4393 [3:07:25<1:37:05, 2.54s/it] + Training...: 48% 2097/4393 [3:07:27<1:27:29, 2.29s/it] + Training...: 48% 2098/4393 [3:07:28<1:18:31, 2.05s/it] + Training...: 48% 2099/4393 [3:07:29<1:09:21, 1.81s/it] + Training...: 48% 2100/4393 [3:07:30<59:45, 1.56s/it]  + Training...: 48% 2101/4393 [3:07:36<1:52:46, 2.95s/it] + Training...: 48% 2102/4393 [3:07:43<2:29:31, 3.92s/it] + Training...: 48% 2103/4393 [3:07:49<2:53:28, 4.55s/it] + Training...: 48% 2104/4393 [3:07:55<3:09:09, 4.96s/it] + Training...: 48% 2105/4393 [3:08:01<3:20:29, 5.26s/it] + Training...: 48% 2106/4393 [3:08:06<3:26:52, 5.43s/it] + Training...: 48% 2107/4393 [3:08:12<3:33:22, 5.60s/it] + Training...: 48% 2108/4393 [3:08:18<3:37:13, 5.70s/it] + Training...: 48% 2109/4393 [3:08:24<3:37:50, 5.72s/it] + Training...: 48% 2110/4393 [3:08:30<3:36:57, 5.70s/it] + Training...: 48% 2111/4393 [3:08:35<3:36:35, 5.69s/it] + Training...: 48% 2112/4393 [3:08:41<3:35:58, 5.68s/it] + Training...: 48% 2113/4393 [3:08:47<3:35:14, 5.66s/it] +  + Training...: 48% 2113/4393 [3:08:53<3:35:14, 5.66s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:38:25<57:16:35, 20619.59s/it] + Training...: 48% 2114/4393 [3:08:53<3:43:25, 5.88s/it] + Training...: 48% 2115/4393 [3:08:59<3:39:31, 5.78s/it] + Training...: 48% 2116/4393 [3:09:04<3:36:47, 5.71s/it] + Training...: 48% 2117/4393 [3:09:10<3:34:21, 5.65s/it] + Training...: 48% 2118/4393 [3:09:15<3:32:16, 5.60s/it] + Training...: 48% 2119/4393 [3:09:21<3:30:06, 5.54s/it] + Training...: 48% 2120/4393 [3:09:26<3:28:34, 5.51s/it] + Training...: 48% 2121/4393 [3:09:31<3:26:38, 5.46s/it] + Training...: 48% 2122/4393 [3:09:37<3:26:39, 5.46s/it] + Training...: 48% 2123/4393 [3:09:42<3:26:55, 5.47s/it] + Training...: 48% 2124/4393 [3:09:47<3:23:45, 5.39s/it] + Training...: 48% 2125/4393 [3:09:53<3:20:35, 5.31s/it] + Training...: 48% 2126/4393 [3:09:58<3:18:25, 5.25s/it] + Training...: 48% 2127/4393 [3:10:03<3:15:56, 5.19s/it] + Training...: 48% 2128/4393 [3:10:08<3:13:48, 5.13s/it] + Training...: 48% 2129/4393 [3:10:13<3:11:18, 5.07s/it] + Training...: 48% 2130/4393 [3:10:18<3:08:38, 5.00s/it] + Training...: 49% 2131/4393 [3:10:22<3:06:08, 4.94s/it] + Training...: 49% 2132/4393 [3:10:27<3:03:28, 4.87s/it] + Training...: 49% 2133/4393 [3:10:32<3:00:57, 4.80s/it] + Training...: 49% 2134/4393 [3:10:36<2:59:04, 4.76s/it] + Training...: 49% 2135/4393 [3:10:41<2:56:04, 4.68s/it] + Training...: 49% 2136/4393 [3:10:45<2:52:06, 4.58s/it] + Training...: 49% 2137/4393 [3:10:49<2:47:03, 4.44s/it] + Training...: 49% 2138/4393 [3:10:53<2:41:36, 4.30s/it] +  + Training...: 49% 2138/4393 [3:10:58<2:41:36, 4.30s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:40:30<57:16:35, 20619.59s/it] + Training...: 49% 2139/4393 [3:10:58<2:41:52, 4.31s/it] + Training...: 49% 2140/4393 [3:11:01<2:31:56, 4.05s/it] + Training...: 49% 2141/4393 [3:11:04<2:22:03, 3.78s/it] + Training...: 49% 2142/4393 [3:11:07<2:12:40, 3.54s/it] + Training...: 49% 2143/4393 [3:11:10<2:04:04, 3.31s/it] + Training...: 49% 2144/4393 [3:11:13<1:55:43, 3.09s/it] + Training...: 49% 2145/4393 [3:11:15<1:46:10, 2.83s/it] + Training...: 49% 2146/4393 [3:11:17<1:37:05, 2.59s/it] + Training...: 49% 2147/4393 [3:11:19<1:28:01, 2.35s/it] + Training...: 49% 2148/4393 [3:11:20<1:18:45, 2.10s/it] + Training...: 49% 2149/4393 [3:11:21<1:09:35, 1.86s/it] + Training...: 49% 2150/4393 [3:11:22<59:55, 1.60s/it]  + Training...: 49% 2151/4393 [3:11:29<1:51:22, 2.98s/it] + Training...: 49% 2152/4393 [3:11:35<2:27:49, 3.96s/it] + Training...: 49% 2153/4393 [3:11:41<2:51:05, 4.58s/it] + Training...: 49% 2154/4393 [3:11:47<3:05:44, 4.98s/it] + Training...: 49% 2155/4393 [3:11:53<3:17:10, 5.29s/it] + Training...: 49% 2156/4393 [3:11:59<3:23:22, 5.45s/it] + Training...: 49% 2157/4393 [3:12:05<3:28:29, 5.59s/it] + Training...: 49% 2158/4393 [3:12:10<3:30:39, 5.66s/it] + Training...: 49% 2159/4393 [3:12:16<3:32:18, 5.70s/it] + Training...: 49% 2160/4393 [3:12:22<3:33:03, 5.72s/it] + Training...: 49% 2161/4393 [3:12:28<3:32:57, 5.72s/it] + Training...: 49% 2162/4393 [3:12:33<3:32:22, 5.71s/it] + Training...: 49% 2163/4393 [3:12:39<3:31:53, 5.70s/it] +  + Training...: 49% 2163/4393 [3:12:45<3:31:53, 5.70s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:42:18<57:16:35, 20619.59s/it] + Training...: 49% 2164/4393 [3:12:45<3:39:27, 5.91s/it] + Training...: 49% 2165/4393 [3:12:51<3:36:47, 5.84s/it] + Training...: 49% 2166/4393 [3:12:57<3:35:01, 5.79s/it] + Training...: 49% 2167/4393 [3:13:02<3:31:59, 5.71s/it] + Training...: 49% 2168/4393 [3:13:08<3:28:28, 5.62s/it] + Training...: 49% 2169/4393 [3:13:13<3:25:50, 5.55s/it] + Training...: 49% 2170/4393 [3:13:18<3:22:53, 5.48s/it] + Training...: 49% 2171/4393 [3:13:24<3:20:50, 5.42s/it] + Training...: 49% 2172/4393 [3:13:29<3:19:01, 5.38s/it] + Training...: 49% 2173/4393 [3:13:34<3:17:31, 5.34s/it] + Training...: 49% 2174/4393 [3:13:39<3:15:53, 5.30s/it] + Training...: 50% 2175/4393 [3:13:45<3:15:02, 5.28s/it] + Training...: 50% 2176/4393 [3:13:50<3:12:59, 5.22s/it] + Training...: 50% 2177/4393 [3:13:55<3:10:59, 5.17s/it] + Training...: 50% 2178/4393 [3:14:00<3:09:04, 5.12s/it] + Training...: 50% 2179/4393 [3:14:05<3:07:34, 5.08s/it] + Training...: 50% 2180/4393 [3:14:10<3:05:49, 5.04s/it] + Training...: 50% 2181/4393 [3:14:15<3:03:56, 4.99s/it] + Training...: 50% 2182/4393 [3:14:19<3:01:14, 4.92s/it] + Training...: 50% 2183/4393 [3:14:24<2:59:15, 4.87s/it] + Training...: 50% 2184/4393 [3:14:29<2:56:14, 4.79s/it] + Training...: 50% 2185/4393 [3:14:33<2:52:33, 4.69s/it] + Training...: 50% 2186/4393 [3:14:37<2:48:13, 4.57s/it] + Training...: 50% 2187/4393 [3:14:42<2:42:35, 4.42s/it] + Training...: 50% 2188/4393 [3:14:45<2:37:33, 4.29s/it] +  + Training...: 50% 2188/4393 [3:14:50<2:37:33, 4.29s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:44:22<57:16:35, 20619.59s/it] + Training...: 50% 2189/4393 [3:14:50<2:39:28, 4.34s/it] + Training...: 50% 2190/4393 [3:14:54<2:31:12, 4.12s/it] + Training...: 50% 2191/4393 [3:14:57<2:21:59, 3.87s/it] + Training...: 50% 2192/4393 [3:15:00<2:12:26, 3.61s/it] + Training...: 50% 2193/4393 [3:15:03<2:03:18, 3.36s/it] + Training...: 50% 2194/4393 [3:15:05<1:53:28, 3.10s/it] + Training...: 50% 2195/4393 [3:15:07<1:44:04, 2.84s/it] + Training...: 50% 2196/4393 [3:15:09<1:34:45, 2.59s/it] + Training...: 50% 2197/4393 [3:15:11<1:25:48, 2.34s/it] + Training...: 50% 2198/4393 [3:15:13<1:16:52, 2.10s/it] + Training...: 50% 2199/4393 [3:15:14<1:07:45, 1.85s/it] + Training...: 50% 2200/4393 [3:15:15<58:13, 1.59s/it]  + Training...: 50% 2201/4393 [3:15:21<1:48:19, 2.97s/it] + Training...: 50% 2202/4393 [3:15:27<2:23:21, 3.93s/it] + Training...: 50% 2203/4393 [3:15:33<2:45:20, 4.53s/it] + Training...: 50% 2204/4393 [3:15:39<3:00:43, 4.95s/it] + Training...: 50% 2205/4393 [3:15:45<3:11:40, 5.26s/it] + Training...: 50% 2206/4393 [3:15:51<3:18:08, 5.44s/it] + Training...: 50% 2207/4393 [3:15:57<3:22:56, 5.57s/it] + Training...: 50% 2208/4393 [3:16:03<3:25:33, 5.64s/it] + Training...: 50% 2209/4393 [3:16:09<3:27:52, 5.71s/it] + Training...: 50% 2210/4393 [3:16:14<3:28:01, 5.72s/it] + Training...: 50% 2211/4393 [3:16:20<3:29:34, 5.76s/it] + Training...: 50% 2212/4393 [3:16:26<3:29:13, 5.76s/it] + Training...: 50% 2213/4393 [3:16:32<3:28:37, 5.74s/it] +  + Training...: 50% 2213/4393 [3:16:38<3:28:37, 5.74s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:46:10<57:16:35, 20619.59s/it] + Training...: 50% 2214/4393 [3:16:38<3:35:46, 5.94s/it] + Training...: 50% 2215/4393 [3:16:44<3:33:01, 5.87s/it] + Training...: 50% 2216/4393 [3:16:49<3:29:15, 5.77s/it] + Training...: 50% 2217/4393 [3:16:55<3:25:59, 5.68s/it] + Training...: 50% 2218/4393 [3:17:00<3:23:45, 5.62s/it] + Training...: 51% 2219/4393 [3:17:06<3:22:26, 5.59s/it] + Training...: 51% 2220/4393 [3:17:11<3:20:42, 5.54s/it] + Training...: 51% 2221/4393 [3:17:17<3:20:08, 5.53s/it] + Training...: 51% 2222/4393 [3:17:22<3:18:13, 5.48s/it] + Training...: 51% 2223/4393 [3:17:27<3:18:10, 5.48s/it] + Training...: 51% 2224/4393 [3:17:33<3:16:40, 5.44s/it] + Training...: 51% 2225/4393 [3:17:38<3:15:05, 5.40s/it] + Training...: 51% 2226/4393 [3:17:43<3:13:27, 5.36s/it] + Training...: 51% 2227/4393 [3:17:49<3:11:49, 5.31s/it] + Training...: 51% 2228/4393 [3:17:54<3:09:19, 5.25s/it] + Training...: 51% 2229/4393 [3:17:59<3:06:58, 5.18s/it] + Training...: 51% 2230/4393 [3:18:04<3:04:37, 5.12s/it] + Training...: 51% 2231/4393 [3:18:09<3:01:38, 5.04s/it] + Training...: 51% 2232/4393 [3:18:13<2:59:12, 4.98s/it] + Training...: 51% 2233/4393 [3:18:18<2:56:49, 4.91s/it] + Training...: 51% 2234/4393 [3:18:23<2:57:56, 4.95s/it] + Training...: 51% 2235/4393 [3:18:28<2:53:26, 4.82s/it] + Training...: 51% 2236/4393 [3:18:32<2:48:34, 4.69s/it] + Training...: 51% 2237/4393 [3:18:36<2:43:56, 4.56s/it] + Training...: 51% 2238/4393 [3:18:40<2:39:04, 4.43s/it] +  + Training...: 51% 2238/4393 [3:18:45<2:39:04, 4.43s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:48:17<57:16:35, 20619.59s/it] + Training...: 51% 2239/4393 [3:18:45<2:41:01, 4.49s/it] + Training...: 51% 2240/4393 [3:18:49<2:31:18, 4.22s/it] + Training...: 51% 2241/4393 [3:18:52<2:21:28, 3.94s/it] + Training...: 51% 2242/4393 [3:18:55<2:13:07, 3.71s/it] + Training...: 51% 2243/4393 [3:18:58<2:04:10, 3.47s/it] + Training...: 51% 2244/4393 [3:19:01<1:55:02, 3.21s/it] + Training...: 51% 2245/4393 [3:19:03<1:45:19, 2.94s/it] + Training...: 51% 2246/4393 [3:19:05<1:35:50, 2.68s/it] + Training...: 51% 2247/4393 [3:19:07<1:26:41, 2.42s/it] + Training...: 51% 2248/4393 [3:19:08<1:17:53, 2.18s/it] + Training...: 51% 2249/4393 [3:19:10<1:08:33, 1.92s/it] + Training...: 51% 2250/4393 [3:19:11<58:55, 1.65s/it]  + Training...: 51% 2251/4393 [3:19:17<1:46:53, 2.99s/it] + Training...: 51% 2252/4393 [3:19:23<2:20:30, 3.94s/it] + Training...: 51% 2253/4393 [3:19:29<2:42:56, 4.57s/it] + Training...: 51% 2254/4393 [3:19:35<2:57:31, 4.98s/it] + Training...: 51% 2255/4393 [3:19:41<3:09:15, 5.31s/it] + Training...: 51% 2256/4393 [3:19:47<3:16:57, 5.53s/it] + Training...: 51% 2257/4393 [3:19:53<3:21:41, 5.67s/it] + Training...: 51% 2258/4393 [3:19:59<3:23:14, 5.71s/it] + Training...: 51% 2259/4393 [3:20:05<3:24:16, 5.74s/it] + Training...: 51% 2260/4393 [3:20:11<3:24:19, 5.75s/it] + Training...: 51% 2261/4393 [3:20:16<3:25:26, 5.78s/it] + Training...: 51% 2262/4393 [3:20:22<3:24:18, 5.75s/it] + Training...: 52% 2263/4393 [3:20:28<3:23:40, 5.74s/it] +  + Training...: 52% 2263/4393 [3:20:34<3:23:40, 5.74s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:50:07<57:16:35, 20619.59s/it] + Training...: 52% 2264/4393 [3:20:34<3:30:57, 5.95s/it] + Training...: 52% 2265/4393 [3:20:40<3:27:29, 5.85s/it] + Training...: 52% 2266/4393 [3:20:46<3:25:37, 5.80s/it] + Training...: 52% 2267/4393 [3:20:51<3:23:12, 5.74s/it] + Training...: 52% 2268/4393 [3:20:57<3:20:21, 5.66s/it] + Training...: 52% 2269/4393 [3:21:02<3:18:38, 5.61s/it] + Training...: 52% 2270/4393 [3:21:07<3:16:05, 5.54s/it] + Training...: 52% 2271/4393 [3:21:13<3:14:40, 5.50s/it] + Training...: 52% 2272/4393 [3:21:18<3:12:50, 5.46s/it] + Training...: 52% 2273/4393 [3:21:24<3:13:59, 5.49s/it] + Training...: 52% 2274/4393 [3:21:29<3:11:49, 5.43s/it] + Training...: 52% 2275/4393 [3:21:34<3:09:39, 5.37s/it] + Training...: 52% 2276/4393 [3:21:40<3:07:30, 5.31s/it] + Training...: 52% 2277/4393 [3:21:45<3:06:49, 5.30s/it] + Training...: 52% 2278/4393 [3:21:50<3:04:42, 5.24s/it] + Training...: 52% 2279/4393 [3:21:55<3:02:39, 5.18s/it] + Training...: 52% 2280/4393 [3:22:00<3:00:23, 5.12s/it] + Training...: 52% 2281/4393 [3:22:05<2:58:19, 5.07s/it] + Training...: 52% 2282/4393 [3:22:10<2:56:07, 5.01s/it] + Training...: 52% 2283/4393 [3:22:15<2:53:51, 4.94s/it] + Training...: 52% 2284/4393 [3:22:19<2:50:51, 4.86s/it] + Training...: 52% 2285/4393 [3:22:24<2:48:27, 4.79s/it] + Training...: 52% 2286/4393 [3:22:28<2:44:48, 4.69s/it] + Training...: 52% 2287/4393 [3:22:33<2:41:33, 4.60s/it] + Training...: 52% 2288/4393 [3:22:37<2:36:48, 4.47s/it] +  + Training...: 52% 2288/4393 [3:22:42<2:36:48, 4.47s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:52:14<57:16:35, 20619.59s/it] + Training...: 52% 2289/4393 [3:22:42<2:40:07, 4.57s/it] + Training...: 52% 2290/4393 [3:22:45<2:32:12, 4.34s/it] + Training...: 52% 2291/4393 [3:22:49<2:23:20, 4.09s/it] + Training...: 52% 2292/4393 [3:22:52<2:14:07, 3.83s/it] + Training...: 52% 2293/4393 [3:22:55<2:04:36, 3.56s/it] + Training...: 52% 2294/4393 [3:22:58<1:54:36, 3.28s/it] + Training...: 52% 2295/4393 [3:23:00<1:44:54, 3.00s/it] + Training...: 52% 2296/4393 [3:23:02<1:35:23, 2.73s/it] + Training...: 52% 2297/4393 [3:23:04<1:25:18, 2.44s/it] + Training...: 52% 2298/4393 [3:23:05<1:15:03, 2.15s/it] + Training...: 52% 2299/4393 [3:23:07<1:04:52, 1.86s/it] + Training...: 52% 2300/4393 [3:23:08<55:22, 1.59s/it]  + Training...: 52% 2301/4393 [3:23:14<1:43:08, 2.96s/it] + Training...: 52% 2302/4393 [3:23:20<2:18:17, 3.97s/it] + Training...: 52% 2303/4393 [3:23:26<2:39:31, 4.58s/it] + Training...: 52% 2304/4393 [3:23:32<2:53:33, 4.98s/it] + Training...: 52% 2305/4393 [3:23:38<3:03:25, 5.27s/it] + Training...: 52% 2306/4393 [3:23:44<3:09:26, 5.45s/it] + Training...: 53% 2307/4393 [3:23:50<3:14:05, 5.58s/it] + Training...: 53% 2308/4393 [3:23:55<3:16:19, 5.65s/it] + Training...: 53% 2309/4393 [3:24:01<3:17:54, 5.70s/it] + Training...: 53% 2310/4393 [3:24:07<3:18:07, 5.71s/it] + Training...: 53% 2311/4393 [3:24:13<3:19:28, 5.75s/it] + Training...: 53% 2312/4393 [3:24:19<3:20:31, 5.78s/it] + Training...: 53% 2313/4393 [3:24:24<3:19:57, 5.77s/it] +  + Training...: 53% 2313/4393 [3:24:31<3:19:57, 5.77s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:54:03<57:16:35, 20619.59s/it] + Training...: 53% 2314/4393 [3:24:31<3:27:18, 5.98s/it] + Training...: 53% 2315/4393 [3:24:36<3:22:44, 5.85s/it] + Training...: 53% 2316/4393 [3:24:42<3:19:30, 5.76s/it] + Training...: 53% 2317/4393 [3:24:48<3:17:33, 5.71s/it] + Training...: 53% 2318/4393 [3:24:53<3:15:07, 5.64s/it] + Training...: 53% 2319/4393 [3:24:59<3:13:04, 5.59s/it] + Training...: 53% 2320/4393 [3:25:04<3:11:15, 5.54s/it] + Training...: 53% 2321/4393 [3:25:09<3:09:25, 5.49s/it] + Training...: 53% 2322/4393 [3:25:15<3:08:49, 5.47s/it] + Training...: 53% 2323/4393 [3:25:20<3:07:24, 5.43s/it] + Training...: 53% 2324/4393 [3:25:25<3:05:30, 5.38s/it] + Training...: 53% 2325/4393 [3:25:31<3:03:22, 5.32s/it] + Training...: 53% 2326/4393 [3:25:36<3:01:14, 5.26s/it] + Training...: 53% 2327/4393 [3:25:41<3:00:26, 5.24s/it] + Training...: 53% 2328/4393 [3:25:46<2:58:49, 5.20s/it] + Training...: 53% 2329/4393 [3:25:51<2:56:41, 5.14s/it] + Training...: 53% 2330/4393 [3:25:56<2:53:44, 5.05s/it] + Training...: 53% 2331/4393 [3:26:01<2:50:49, 4.97s/it] + Training...: 53% 2332/4393 [3:26:05<2:48:06, 4.89s/it] + Training...: 53% 2333/4393 [3:26:10<2:45:37, 4.82s/it] + Training...: 53% 2334/4393 [3:26:15<2:42:44, 4.74s/it] + Training...: 53% 2335/4393 [3:26:19<2:39:58, 4.66s/it] + Training...: 53% 2336/4393 [3:26:23<2:36:33, 4.57s/it] + Training...: 53% 2337/4393 [3:26:28<2:32:27, 4.45s/it] + Training...: 53% 2338/4393 [3:26:32<2:28:16, 4.33s/it] +  + Training...: 53% 2338/4393 [3:26:36<2:28:16, 4.33s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:56:08<57:16:35, 20619.59s/it] + Training...: 53% 2339/4393 [3:26:36<2:28:47, 4.35s/it] + Training...: 53% 2340/4393 [3:26:40<2:21:14, 4.13s/it] + Training...: 53% 2341/4393 [3:26:43<2:13:29, 3.90s/it] + Training...: 53% 2342/4393 [3:26:46<2:05:00, 3.66s/it] + Training...: 53% 2343/4393 [3:26:49<1:56:06, 3.40s/it] + Training...: 53% 2344/4393 [3:26:51<1:47:19, 3.14s/it] + Training...: 53% 2345/4393 [3:26:54<1:37:24, 2.85s/it] + Training...: 53% 2346/4393 [3:26:55<1:28:00, 2.58s/it] + Training...: 53% 2347/4393 [3:26:57<1:18:58, 2.32s/it] + Training...: 53% 2348/4393 [3:26:59<1:10:29, 2.07s/it] + Training...: 53% 2349/4393 [3:27:00<1:01:40, 1.81s/it] + Training...: 53% 2350/4393 [3:27:01<52:46, 1.55s/it]  + Training...: 54% 2351/4393 [3:27:07<1:42:05, 3.00s/it] + Training...: 54% 2352/4393 [3:27:13<2:15:25, 3.98s/it] + Training...: 54% 2353/4393 [3:27:20<2:39:02, 4.68s/it] + Training...: 54% 2354/4393 [3:27:26<2:53:41, 5.11s/it] + Training...: 54% 2355/4393 [3:27:32<3:04:14, 5.42s/it] + Training...: 54% 2356/4393 [3:27:38<3:08:24, 5.55s/it] + Training...: 54% 2357/4393 [3:27:44<3:11:16, 5.64s/it] + Training...: 54% 2358/4393 [3:27:50<3:12:30, 5.68s/it] + Training...: 54% 2359/4393 [3:27:55<3:13:21, 5.70s/it] + Training...: 54% 2360/4393 [3:28:01<3:13:13, 5.70s/it] + Training...: 54% 2361/4393 [3:28:07<3:13:09, 5.70s/it] + Training...: 54% 2362/4393 [3:28:12<3:12:43, 5.69s/it] + Training...: 54% 2363/4393 [3:28:18<3:12:26, 5.69s/it] +  + Training...: 54% 2363/4393 [3:28:25<3:12:26, 5.69s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [14:57:57<57:16:35, 20619.59s/it] + Training...: 54% 2364/4393 [3:28:25<3:20:21, 5.92s/it] + Training...: 54% 2365/4393 [3:28:30<3:19:09, 5.89s/it] + Training...: 54% 2366/4393 [3:28:36<3:17:00, 5.83s/it] + Training...: 54% 2367/4393 [3:28:42<3:14:20, 5.76s/it] + Training...: 54% 2368/4393 [3:28:47<3:11:33, 5.68s/it] + Training...: 54% 2369/4393 [3:28:53<3:09:15, 5.61s/it] + Training...: 54% 2370/4393 [3:28:58<3:06:41, 5.54s/it] + Training...: 54% 2371/4393 [3:29:03<3:05:00, 5.49s/it] + Training...: 54% 2372/4393 [3:29:09<3:02:43, 5.42s/it] + Training...: 54% 2373/4393 [3:29:14<3:01:30, 5.39s/it] + Training...: 54% 2374/4393 [3:29:19<3:00:21, 5.36s/it] + Training...: 54% 2375/4393 [3:29:24<2:58:28, 5.31s/it] + Training...: 54% 2376/4393 [3:29:30<2:57:01, 5.27s/it] + Training...: 54% 2377/4393 [3:29:35<2:55:42, 5.23s/it] + Training...: 54% 2378/4393 [3:29:40<2:54:12, 5.19s/it] + Training...: 54% 2379/4393 [3:29:45<2:52:40, 5.14s/it] + Training...: 54% 2380/4393 [3:29:50<2:50:15, 5.07s/it] + Training...: 54% 2381/4393 [3:29:55<2:48:13, 5.02s/it] + Training...: 54% 2382/4393 [3:29:59<2:45:41, 4.94s/it] + Training...: 54% 2383/4393 [3:30:04<2:43:30, 4.88s/it] + Training...: 54% 2384/4393 [3:30:09<2:41:09, 4.81s/it] + Training...: 54% 2385/4393 [3:30:13<2:38:50, 4.75s/it] + Training...: 54% 2386/4393 [3:30:18<2:35:50, 4.66s/it] + Training...: 54% 2387/4393 [3:30:22<2:33:08, 4.58s/it] + Training...: 54% 2388/4393 [3:30:26<2:28:37, 4.45s/it] +  + Training...: 54% 2388/4393 [3:30:31<2:28:37, 4.45s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:00:03<57:16:35, 20619.59s/it] + Training...: 54% 2389/4393 [3:30:31<2:29:45, 4.48s/it] + Training...: 54% 2390/4393 [3:30:35<2:22:03, 4.26s/it] + Training...: 54% 2391/4393 [3:30:38<2:14:47, 4.04s/it] + Training...: 54% 2392/4393 [3:30:41<2:06:42, 3.80s/it] + Training...: 54% 2393/4393 [3:30:44<1:57:58, 3.54s/it] + Training...: 54% 2394/4393 [3:30:47<1:49:13, 3.28s/it] + Training...: 55% 2395/4393 [3:30:49<1:40:45, 3.03s/it] + Training...: 55% 2396/4393 [3:30:52<1:32:05, 2.77s/it] + Training...: 55% 2397/4393 [3:30:53<1:23:12, 2.50s/it] + Training...: 55% 2398/4393 [3:30:55<1:14:18, 2.23s/it] + Training...: 55% 2399/4393 [3:30:56<1:05:09, 1.96s/it] + Training...: 55% 2400/4393 [3:30:57<55:43, 1.68s/it]  + Training...: 55% 2401/4393 [3:31:04<1:43:09, 3.11s/it] + Training...: 55% 2402/4393 [3:31:10<2:14:24, 4.05s/it] + Training...: 55% 2403/4393 [3:31:16<2:34:26, 4.66s/it] + Training...: 55% 2404/4393 [3:31:22<2:48:15, 5.08s/it] + Training...: 55% 2405/4393 [3:31:28<2:59:06, 5.41s/it] + Training...: 55% 2406/4393 [3:31:34<3:04:07, 5.56s/it] + Training...: 55% 2407/4393 [3:31:40<3:07:44, 5.67s/it] + Training...: 55% 2408/4393 [3:31:46<3:09:29, 5.73s/it] + Training...: 55% 2409/4393 [3:31:52<3:10:43, 5.77s/it] + Training...: 55% 2410/4393 [3:31:58<3:10:20, 5.76s/it] + Training...: 55% 2411/4393 [3:32:04<3:10:43, 5.77s/it] + Training...: 55% 2412/4393 [3:32:09<3:08:57, 5.72s/it] + Training...: 55% 2413/4393 [3:32:15<3:07:57, 5.70s/it] +  + Training...: 55% 2413/4393 [3:32:21<3:07:57, 5.70s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:01:54<57:16:35, 20619.59s/it] + Training...: 55% 2414/4393 [3:32:21<3:15:13, 5.92s/it] + Training...: 55% 2415/4393 [3:32:27<3:11:01, 5.79s/it] + Training...: 55% 2416/4393 [3:32:32<3:09:02, 5.74s/it] + Training...: 55% 2417/4393 [3:32:38<3:07:11, 5.68s/it] + Training...: 55% 2418/4393 [3:32:43<3:06:17, 5.66s/it] + Training...: 55% 2419/4393 [3:32:49<3:03:40, 5.58s/it] + Training...: 55% 2420/4393 [3:32:54<3:02:02, 5.54s/it] + Training...: 55% 2421/4393 [3:33:00<2:59:44, 5.47s/it] + Training...: 55% 2422/4393 [3:33:05<2:58:08, 5.42s/it] + Training...: 55% 2423/4393 [3:33:10<2:56:51, 5.39s/it] + Training...: 55% 2424/4393 [3:33:15<2:55:02, 5.33s/it] + Training...: 55% 2425/4393 [3:33:21<2:53:38, 5.29s/it] + Training...: 55% 2426/4393 [3:33:26<2:52:16, 5.26s/it] + Training...: 55% 2427/4393 [3:33:31<2:50:47, 5.21s/it] + Training...: 55% 2428/4393 [3:33:36<2:49:33, 5.18s/it] + Training...: 55% 2429/4393 [3:33:41<2:48:02, 5.13s/it] + Training...: 55% 2430/4393 [3:33:46<2:46:41, 5.09s/it] + Training...: 55% 2431/4393 [3:33:51<2:45:23, 5.06s/it] + Training...: 55% 2432/4393 [3:33:56<2:43:32, 5.00s/it] + Training...: 55% 2433/4393 [3:34:01<2:41:09, 4.93s/it] + Training...: 55% 2434/4393 [3:34:05<2:37:59, 4.84s/it] + Training...: 55% 2435/4393 [3:34:10<2:34:10, 4.72s/it] + Training...: 55% 2436/4393 [3:34:14<2:30:08, 4.60s/it] + Training...: 55% 2437/4393 [3:34:18<2:25:43, 4.47s/it] + Training...: 55% 2438/4393 [3:34:22<2:20:33, 4.31s/it] +  + Training...: 55% 2438/4393 [3:34:27<2:20:33, 4.31s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:03:59<57:16:35, 20619.59s/it] + Training...: 56% 2439/4393 [3:34:27<2:20:45, 4.32s/it] + Training...: 56% 2440/4393 [3:34:30<2:11:59, 4.06s/it] + Training...: 56% 2441/4393 [3:34:33<2:03:26, 3.79s/it] + Training...: 56% 2442/4393 [3:34:36<1:55:21, 3.55s/it] + Training...: 56% 2443/4393 [3:34:39<1:47:07, 3.30s/it] + Training...: 56% 2444/4393 [3:34:41<1:38:33, 3.03s/it] + Training...: 56% 2445/4393 [3:34:43<1:30:27, 2.79s/it] + Training...: 56% 2446/4393 [3:34:45<1:22:06, 2.53s/it] + Training...: 56% 2447/4393 [3:34:47<1:14:11, 2.29s/it] + Training...: 56% 2448/4393 [3:34:49<1:06:26, 2.05s/it] + Training...: 56% 2449/4393 [3:34:50<58:16, 1.80s/it]  + Training...: 56% 2450/4393 [3:34:51<50:37, 1.56s/it] + Training...: 56% 2451/4393 [3:34:57<1:37:13, 3.00s/it] + Training...: 56% 2452/4393 [3:35:03<2:07:48, 3.95s/it] + Training...: 56% 2453/4393 [3:35:09<2:28:26, 4.59s/it] + Training...: 56% 2454/4393 [3:35:15<2:42:00, 5.01s/it] + Training...: 56% 2455/4393 [3:35:21<2:51:37, 5.31s/it] + Training...: 56% 2456/4393 [3:35:27<2:57:00, 5.48s/it] + Training...: 56% 2457/4393 [3:35:33<3:00:51, 5.61s/it] + Training...: 56% 2458/4393 [3:35:39<3:03:05, 5.68s/it] + Training...: 56% 2459/4393 [3:35:45<3:04:19, 5.72s/it] + Training...: 56% 2460/4393 [3:35:51<3:05:28, 5.76s/it] + Training...: 56% 2461/4393 [3:35:56<3:05:19, 5.76s/it] + Training...: 56% 2462/4393 [3:36:02<3:05:56, 5.78s/it] + Training...: 56% 2463/4393 [3:36:08<3:06:38, 5.80s/it] +  + Training...: 56% 2463/4393 [3:36:15<3:06:38, 5.80s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:05:47<57:16:35, 20619.59s/it] + Training...: 56% 2464/4393 [3:36:15<3:14:10, 6.04s/it] + Training...: 56% 2465/4393 [3:36:20<3:10:30, 5.93s/it] + Training...: 56% 2466/4393 [3:36:26<3:06:58, 5.82s/it] + Training...: 56% 2467/4393 [3:36:32<3:04:20, 5.74s/it] + Training...: 56% 2468/4393 [3:36:37<3:02:22, 5.68s/it] + Training...: 56% 2469/4393 [3:36:43<3:01:07, 5.65s/it] + Training...: 56% 2470/4393 [3:36:48<2:58:43, 5.58s/it] + Training...: 56% 2471/4393 [3:36:54<2:57:08, 5.53s/it] + Training...: 56% 2472/4393 [3:36:59<2:55:36, 5.49s/it] + Training...: 56% 2473/4393 [3:37:04<2:53:53, 5.43s/it] + Training...: 56% 2474/4393 [3:37:10<2:53:05, 5.41s/it] + Training...: 56% 2475/4393 [3:37:15<2:51:43, 5.37s/it] + Training...: 56% 2476/4393 [3:37:20<2:50:13, 5.33s/it] + Training...: 56% 2477/4393 [3:37:25<2:48:33, 5.28s/it] + Training...: 56% 2478/4393 [3:37:30<2:47:03, 5.23s/it] + Training...: 56% 2479/4393 [3:37:35<2:45:20, 5.18s/it] + Training...: 56% 2480/4393 [3:37:40<2:43:40, 5.13s/it] + Training...: 56% 2481/4393 [3:37:45<2:41:23, 5.06s/it] + Training...: 56% 2482/4393 [3:37:50<2:40:01, 5.02s/it] + Training...: 57% 2483/4393 [3:37:55<2:38:34, 4.98s/it] + Training...: 57% 2484/4393 [3:38:00<2:35:15, 4.88s/it] + Training...: 57% 2485/4393 [3:38:04<2:31:37, 4.77s/it] + Training...: 57% 2486/4393 [3:38:09<2:27:43, 4.65s/it] + Training...: 57% 2487/4393 [3:38:13<2:23:20, 4.51s/it] + Training...: 57% 2488/4393 [3:38:17<2:18:19, 4.36s/it] +  + Training...: 57% 2488/4393 [3:38:21<2:18:19, 4.36s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:07:54<57:16:35, 20619.59s/it] + Training...: 57% 2489/4393 [3:38:21<2:18:50, 4.38s/it] + Training...: 57% 2490/4393 [3:38:25<2:10:51, 4.13s/it] + Training...: 57% 2491/4393 [3:38:28<2:03:30, 3.90s/it] + Training...: 57% 2492/4393 [3:38:31<1:56:33, 3.68s/it] + Training...: 57% 2493/4393 [3:38:34<1:48:57, 3.44s/it] + Training...: 57% 2494/4393 [3:38:37<1:41:22, 3.20s/it] + Training...: 57% 2495/4393 [3:38:39<1:32:54, 2.94s/it] + Training...: 57% 2496/4393 [3:38:41<1:24:31, 2.67s/it] + Training...: 57% 2497/4393 [3:38:43<1:16:00, 2.41s/it] + Training...: 57% 2498/4393 [3:38:45<1:07:40, 2.14s/it] + Training...: 57% 2499/4393 [3:38:46<59:12, 1.88s/it]  + Training...: 57% 2500/4393 [3:38:47<50:49, 1.61s/it] + Training...: 57% 2501/4393 [3:38:53<1:35:00, 3.01s/it] + Training...: 57% 2502/4393 [3:39:00<2:06:57, 4.03s/it] + Training...: 57% 2503/4393 [3:39:06<2:27:54, 4.70s/it] + Training...: 57% 2504/4393 [3:39:12<2:42:13, 5.15s/it] + Training...: 57% 2505/4393 [3:39:18<2:51:26, 5.45s/it] + Training...: 57% 2506/4393 [3:39:24<2:58:09, 5.66s/it] + Training...: 57% 2507/4393 [3:39:30<3:00:16, 5.74s/it] + Training...: 57% 2508/4393 [3:39:36<3:00:10, 5.74s/it] + Training...: 57% 2509/4393 [3:39:42<3:00:30, 5.75s/it] + Training...: 57% 2510/4393 [3:39:47<2:59:56, 5.73s/it] + Training...: 57% 2511/4393 [3:39:53<3:00:00, 5.74s/it] + Training...: 57% 2512/4393 [3:39:59<2:58:54, 5.71s/it] + Training...: 57% 2513/4393 [3:40:04<2:58:44, 5.70s/it] +  + Training...: 57% 2513/4393 [3:40:11<2:58:44, 5.70s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:09:43<57:16:35, 20619.59s/it] + Training...: 57% 2514/4393 [3:40:11<3:05:48, 5.93s/it] + Training...: 57% 2515/4393 [3:40:17<3:02:29, 5.83s/it] + Training...: 57% 2516/4393 [3:40:22<3:00:03, 5.76s/it] + Training...: 57% 2517/4393 [3:40:28<2:57:48, 5.69s/it] + Training...: 57% 2518/4393 [3:40:33<2:57:09, 5.67s/it] + Training...: 57% 2519/4393 [3:40:39<2:55:36, 5.62s/it] + Training...: 57% 2520/4393 [3:40:44<2:53:14, 5.55s/it] + Training...: 57% 2521/4393 [3:40:50<2:52:23, 5.53s/it] + Training...: 57% 2522/4393 [3:40:55<2:51:38, 5.50s/it] + Training...: 57% 2523/4393 [3:41:00<2:49:42, 5.44s/it] + Training...: 57% 2524/4393 [3:41:06<2:47:39, 5.38s/it] + Training...: 57% 2525/4393 [3:41:11<2:45:39, 5.32s/it] + Training...: 58% 2526/4393 [3:41:16<2:43:46, 5.26s/it] + Training...: 58% 2527/4393 [3:41:21<2:41:57, 5.21s/it] + Training...: 58% 2528/4393 [3:41:26<2:40:06, 5.15s/it] + Training...: 58% 2529/4393 [3:41:31<2:38:20, 5.10s/it] + Training...: 58% 2530/4393 [3:41:36<2:36:40, 5.05s/it] + Training...: 58% 2531/4393 [3:41:41<2:34:56, 4.99s/it] + Training...: 58% 2532/4393 [3:41:46<2:32:42, 4.92s/it] + Training...: 58% 2533/4393 [3:41:50<2:30:04, 4.84s/it] + Training...: 58% 2534/4393 [3:41:55<2:27:21, 4.76s/it] + Training...: 58% 2535/4393 [3:41:59<2:24:56, 4.68s/it] + Training...: 58% 2536/4393 [3:42:04<2:22:38, 4.61s/it] + Training...: 58% 2537/4393 [3:42:08<2:18:34, 4.48s/it] + Training...: 58% 2538/4393 [3:42:12<2:14:20, 4.35s/it] +  + Training...: 58% 2538/4393 [3:42:16<2:14:20, 4.35s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:11:49<57:16:35, 20619.59s/it] + Training...: 58% 2539/4393 [3:42:16<2:14:21, 4.35s/it] + Training...: 58% 2540/4393 [3:42:20<2:07:08, 4.12s/it] + Training...: 58% 2541/4393 [3:42:23<1:59:30, 3.87s/it] + Training...: 58% 2542/4393 [3:42:26<1:52:28, 3.65s/it] + Training...: 58% 2543/4393 [3:42:29<1:44:39, 3.39s/it] + Training...: 58% 2544/4393 [3:42:32<1:37:07, 3.15s/it] + Training...: 58% 2545/4393 [3:42:34<1:29:57, 2.92s/it] + Training...: 58% 2546/4393 [3:42:36<1:22:10, 2.67s/it] + Training...: 58% 2547/4393 [3:42:38<1:13:52, 2.40s/it] + Training...: 58% 2548/4393 [3:42:39<1:05:59, 2.15s/it] + Training...: 58% 2549/4393 [3:42:41<57:40, 1.88s/it]  + Training...: 58% 2550/4393 [3:42:42<49:14, 1.60s/it] + Training...: 58% 2551/4393 [3:42:48<1:32:56, 3.03s/it] + Training...: 58% 2552/4393 [3:42:54<2:03:07, 4.01s/it] + Training...: 58% 2553/4393 [3:43:00<2:21:58, 4.63s/it] + Training...: 58% 2554/4393 [3:43:06<2:33:50, 5.02s/it] + Training...: 58% 2555/4393 [3:43:12<2:42:31, 5.31s/it] + Training...: 58% 2556/4393 [3:43:18<2:47:54, 5.48s/it] + Training...: 58% 2557/4393 [3:43:24<2:51:27, 5.60s/it] + Training...: 58% 2558/4393 [3:43:30<2:53:20, 5.67s/it] + Training...: 58% 2559/4393 [3:43:36<2:54:56, 5.72s/it] + Training...: 58% 2560/4393 [3:43:42<2:54:55, 5.73s/it] + Training...: 58% 2561/4393 [3:43:47<2:54:49, 5.73s/it] + Training...: 58% 2562/4393 [3:43:53<2:54:15, 5.71s/it] + Training...: 58% 2563/4393 [3:43:59<2:54:40, 5.73s/it] +  + Training...: 58% 2563/4393 [3:44:05<2:54:40, 5.73s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:13:37<57:16:35, 20619.59s/it] + Training...: 58% 2564/4393 [3:44:05<3:01:18, 5.95s/it] + Training...: 58% 2565/4393 [3:44:11<2:58:32, 5.86s/it] + Training...: 58% 2566/4393 [3:44:16<2:55:58, 5.78s/it] + Training...: 58% 2567/4393 [3:44:22<2:54:29, 5.73s/it] + Training...: 58% 2568/4393 [3:44:28<2:52:28, 5.67s/it] + Training...: 58% 2569/4393 [3:44:33<2:50:31, 5.61s/it] + Training...: 59% 2570/4393 [3:44:38<2:48:38, 5.55s/it] + Training...: 59% 2571/4393 [3:44:44<2:46:46, 5.49s/it] + Training...: 59% 2572/4393 [3:44:49<2:45:09, 5.44s/it] + Training...: 59% 2573/4393 [3:44:54<2:44:19, 5.42s/it] + Training...: 59% 2574/4393 [3:45:00<2:42:52, 5.37s/it] + Training...: 59% 2575/4393 [3:45:05<2:41:53, 5.34s/it] + Training...: 59% 2576/4393 [3:45:10<2:40:13, 5.29s/it] + Training...: 59% 2577/4393 [3:45:15<2:38:30, 5.24s/it] + Training...: 59% 2578/4393 [3:45:20<2:36:36, 5.18s/it] + Training...: 59% 2579/4393 [3:45:25<2:36:33, 5.18s/it] + Training...: 59% 2580/4393 [3:45:31<2:35:18, 5.14s/it] + Training...: 59% 2581/4393 [3:45:36<2:33:42, 5.09s/it] + Training...: 59% 2582/4393 [3:45:40<2:31:31, 5.02s/it] + Training...: 59% 2583/4393 [3:45:45<2:28:33, 4.92s/it] + Training...: 59% 2584/4393 [3:45:50<2:25:55, 4.84s/it] + Training...: 59% 2585/4393 [3:45:54<2:23:42, 4.77s/it] + Training...: 59% 2586/4393 [3:45:59<2:20:35, 4.67s/it] + Training...: 59% 2587/4393 [3:46:03<2:16:51, 4.55s/it] + Training...: 59% 2588/4393 [3:46:07<2:12:16, 4.40s/it] +  + Training...: 59% 2588/4393 [3:46:12<2:12:16, 4.40s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:15:44<57:16:35, 20619.59s/it] + Training...: 59% 2589/4393 [3:46:12<2:12:52, 4.42s/it] + Training...: 59% 2590/4393 [3:46:15<2:05:36, 4.18s/it] + Training...: 59% 2591/4393 [3:46:19<1:58:35, 3.95s/it] + Training...: 59% 2592/4393 [3:46:22<1:51:25, 3.71s/it] + Training...: 59% 2593/4393 [3:46:25<1:44:02, 3.47s/it] + Training...: 59% 2594/4393 [3:46:27<1:36:57, 3.23s/it] + Training...: 59% 2595/4393 [3:46:30<1:29:33, 2.99s/it] + Training...: 59% 2596/4393 [3:46:32<1:21:33, 2.72s/it] + Training...: 59% 2597/4393 [3:46:34<1:13:05, 2.44s/it] + Training...: 59% 2598/4393 [3:46:35<1:05:00, 2.17s/it] + Training...: 59% 2599/4393 [3:46:36<56:46, 1.90s/it]  + Training...: 59% 2600/4393 [3:46:37<48:38, 1.63s/it] + Training...: 59% 2601/4393 [3:46:44<1:29:38, 3.00s/it] + Training...: 59% 2602/4393 [3:46:50<1:58:16, 3.96s/it] + Training...: 59% 2603/4393 [3:46:56<2:16:38, 4.58s/it] + Training...: 59% 2604/4393 [3:47:02<2:28:27, 4.98s/it] + Training...: 59% 2605/4393 [3:47:08<2:36:55, 5.27s/it] + Training...: 59% 2606/4393 [3:47:14<2:42:48, 5.47s/it] + Training...: 59% 2607/4393 [3:47:20<2:46:37, 5.60s/it] + Training...: 59% 2608/4393 [3:47:25<2:48:53, 5.68s/it] + Training...: 59% 2609/4393 [3:47:31<2:50:45, 5.74s/it] + Training...: 59% 2610/4393 [3:47:37<2:53:29, 5.84s/it] + Training...: 59% 2611/4393 [3:47:43<2:52:24, 5.80s/it] + Training...: 59% 2612/4393 [3:47:49<2:50:57, 5.76s/it] + Training...: 59% 2613/4393 [3:47:54<2:50:38, 5.75s/it] +  + Training...: 59% 2613/4393 [3:48:01<2:50:38, 5.75s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:17:33<57:16:35, 20619.59s/it] + Training...: 60% 2614/4393 [3:48:01<2:57:15, 5.98s/it] + Training...: 60% 2615/4393 [3:48:07<2:53:42, 5.86s/it] + Training...: 60% 2616/4393 [3:48:12<2:50:46, 5.77s/it] + Training...: 60% 2617/4393 [3:48:18<2:48:15, 5.68s/it] + Training...: 60% 2618/4393 [3:48:23<2:45:56, 5.61s/it] + Training...: 60% 2619/4393 [3:48:29<2:44:39, 5.57s/it] + Training...: 60% 2620/4393 [3:48:34<2:43:05, 5.52s/it] + Training...: 60% 2621/4393 [3:48:39<2:42:25, 5.50s/it] + Training...: 60% 2622/4393 [3:48:45<2:41:06, 5.46s/it] + Training...: 60% 2623/4393 [3:48:50<2:39:26, 5.41s/it] + Training...: 60% 2624/4393 [3:48:55<2:37:54, 5.36s/it] + Training...: 60% 2625/4393 [3:49:00<2:36:34, 5.31s/it] + Training...: 60% 2626/4393 [3:49:06<2:35:44, 5.29s/it] + Training...: 60% 2627/4393 [3:49:11<2:34:14, 5.24s/it] + Training...: 60% 2628/4393 [3:49:16<2:32:56, 5.20s/it] + Training...: 60% 2629/4393 [3:49:21<2:32:18, 5.18s/it] + Training...: 60% 2630/4393 [3:49:26<2:31:11, 5.15s/it] + Training...: 60% 2631/4393 [3:49:31<2:29:21, 5.09s/it] + Training...: 60% 2632/4393 [3:49:36<2:26:20, 4.99s/it] + Training...: 60% 2633/4393 [3:49:41<2:24:14, 4.92s/it] + Training...: 60% 2634/4393 [3:49:45<2:21:58, 4.84s/it] + Training...: 60% 2635/4393 [3:49:50<2:19:25, 4.76s/it] + Training...: 60% 2636/4393 [3:49:54<2:17:04, 4.68s/it] + Training...: 60% 2637/4393 [3:49:59<2:13:45, 4.57s/it] + Training...: 60% 2638/4393 [3:50:03<2:09:56, 4.44s/it] +  + Training...: 60% 2638/4393 [3:50:07<2:09:56, 4.44s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:19:40<57:16:35, 20619.59s/it] + Training...: 60% 2639/4393 [3:50:07<2:11:02, 4.48s/it] + Training...: 60% 2640/4393 [3:50:11<2:03:51, 4.24s/it] + Training...: 60% 2641/4393 [3:50:14<1:56:31, 3.99s/it] + Training...: 60% 2642/4393 [3:50:18<1:49:25, 3.75s/it] + Training...: 60% 2643/4393 [3:50:21<1:41:52, 3.49s/it] + Training...: 60% 2644/4393 [3:50:23<1:34:35, 3.24s/it] + Training...: 60% 2645/4393 [3:50:25<1:26:20, 2.96s/it] + Training...: 60% 2646/4393 [3:50:28<1:18:26, 2.69s/it] + Training...: 60% 2647/4393 [3:50:29<1:11:03, 2.44s/it] + Training...: 60% 2648/4393 [3:50:31<1:03:25, 2.18s/it] + Training...: 60% 2649/4393 [3:50:32<55:29, 1.91s/it]  + Training...: 60% 2650/4393 [3:50:33<47:47, 1.65s/it] + Training...: 60% 2651/4393 [3:50:40<1:27:50, 3.03s/it] + Training...: 60% 2652/4393 [3:50:46<1:55:26, 3.98s/it] + Training...: 60% 2653/4393 [3:50:52<2:13:01, 4.59s/it] + Training...: 60% 2654/4393 [3:50:58<2:26:08, 5.04s/it] + Training...: 60% 2655/4393 [3:51:04<2:34:56, 5.35s/it] + Training...: 60% 2656/4393 [3:51:10<2:38:48, 5.49s/it] + Training...: 60% 2657/4393 [3:51:16<2:41:52, 5.59s/it] + Training...: 61% 2658/4393 [3:51:21<2:43:21, 5.65s/it] + Training...: 61% 2659/4393 [3:51:27<2:44:33, 5.69s/it] + Training...: 61% 2660/4393 [3:51:33<2:44:07, 5.68s/it] + Training...: 61% 2661/4393 [3:51:39<2:44:19, 5.69s/it] + Training...: 61% 2662/4393 [3:51:44<2:44:21, 5.70s/it] + Training...: 61% 2663/4393 [3:51:50<2:43:51, 5.68s/it] +  + Training...: 61% 2663/4393 [3:51:56<2:43:51, 5.68s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:21:29<57:16:35, 20619.59s/it] + Training...: 61% 2664/4393 [3:51:56<2:51:00, 5.93s/it] + Training...: 61% 2665/4393 [3:52:02<2:49:10, 5.87s/it] + Training...: 61% 2666/4393 [3:52:08<2:45:43, 5.76s/it] + Training...: 61% 2667/4393 [3:52:13<2:45:07, 5.74s/it] + Training...: 61% 2668/4393 [3:52:19<2:44:31, 5.72s/it] + Training...: 61% 2669/4393 [3:52:24<2:42:37, 5.66s/it] + Training...: 61% 2670/4393 [3:52:30<2:42:12, 5.65s/it] + Training...: 61% 2671/4393 [3:52:35<2:39:32, 5.56s/it] + Training...: 61% 2672/4393 [3:52:41<2:37:14, 5.48s/it] + Training...: 61% 2673/4393 [3:52:46<2:35:08, 5.41s/it] + Training...: 61% 2674/4393 [3:52:51<2:32:40, 5.33s/it] + Training...: 61% 2675/4393 [3:52:56<2:31:05, 5.28s/it] + Training...: 61% 2676/4393 [3:53:01<2:30:10, 5.25s/it] + Training...: 61% 2677/4393 [3:53:07<2:29:00, 5.21s/it] + Training...: 61% 2678/4393 [3:53:12<2:27:39, 5.17s/it] + Training...: 61% 2679/4393 [3:53:17<2:26:24, 5.13s/it] + Training...: 61% 2680/4393 [3:53:22<2:24:22, 5.06s/it] + Training...: 61% 2681/4393 [3:53:27<2:22:56, 5.01s/it] + Training...: 61% 2682/4393 [3:53:31<2:21:17, 4.95s/it] + Training...: 61% 2683/4393 [3:53:36<2:19:21, 4.89s/it] + Training...: 61% 2684/4393 [3:53:41<2:16:28, 4.79s/it] + Training...: 61% 2685/4393 [3:53:45<2:13:55, 4.70s/it] + Training...: 61% 2686/4393 [3:53:49<2:10:35, 4.59s/it] + Training...: 61% 2687/4393 [3:53:54<2:07:10, 4.47s/it] + Training...: 61% 2688/4393 [3:53:58<2:02:21, 4.31s/it] +  + Training...: 61% 2688/4393 [3:54:02<2:02:21, 4.31s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:23:34<57:16:35, 20619.59s/it] + Training...: 61% 2689/4393 [3:54:02<2:02:51, 4.33s/it] + Training...: 61% 2690/4393 [3:54:05<1:55:16, 4.06s/it] + Training...: 61% 2691/4393 [3:54:09<1:48:26, 3.82s/it] + Training...: 61% 2692/4393 [3:54:12<1:41:21, 3.58s/it] + Training...: 61% 2693/4393 [3:54:14<1:34:57, 3.35s/it] + Training...: 61% 2694/4393 [3:54:17<1:27:15, 3.08s/it] + Training...: 61% 2695/4393 [3:54:19<1:19:21, 2.80s/it] + Training...: 61% 2696/4393 [3:54:21<1:11:53, 2.54s/it] + Training...: 61% 2697/4393 [3:54:23<1:04:39, 2.29s/it] + Training...: 61% 2698/4393 [3:54:24<57:50, 2.05s/it]  + Training...: 61% 2699/4393 [3:54:25<50:56, 1.80s/it] + Training...: 61% 2700/4393 [3:54:26<43:43, 1.55s/it] + Training...: 61% 2701/4393 [3:54:33<1:22:56, 2.94s/it] + Training...: 62% 2702/4393 [3:54:39<1:50:25, 3.92s/it] + Training...: 62% 2703/4393 [3:54:45<2:08:10, 4.55s/it] + Training...: 62% 2704/4393 [3:54:51<2:19:39, 4.96s/it] + Training...: 62% 2705/4393 [3:54:57<2:27:30, 5.24s/it] + Training...: 62% 2706/4393 [3:55:02<2:32:22, 5.42s/it] + Training...: 62% 2707/4393 [3:55:08<2:35:51, 5.55s/it] + Training...: 62% 2708/4393 [3:55:14<2:38:13, 5.63s/it] + Training...: 62% 2709/4393 [3:55:20<2:39:34, 5.69s/it] + Training...: 62% 2710/4393 [3:55:26<2:39:47, 5.70s/it] + Training...: 62% 2711/4393 [3:55:31<2:39:50, 5.70s/it] + Training...: 62% 2712/4393 [3:55:37<2:40:02, 5.71s/it] + Training...: 62% 2713/4393 [3:55:43<2:40:09, 5.72s/it] +  + Training...: 62% 2713/4393 [3:55:49<2:40:09, 5.72s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:25:22<57:16:35, 20619.59s/it] + Training...: 62% 2714/4393 [3:55:49<2:45:56, 5.93s/it] + Training...: 62% 2715/4393 [3:55:55<2:42:37, 5.81s/it] + Training...: 62% 2716/4393 [3:56:00<2:39:51, 5.72s/it] + Training...: 62% 2717/4393 [3:56:06<2:38:00, 5.66s/it] + Training...: 62% 2718/4393 [3:56:11<2:36:23, 5.60s/it] + Training...: 62% 2719/4393 [3:56:17<2:34:53, 5.55s/it] + Training...: 62% 2720/4393 [3:56:22<2:33:28, 5.50s/it] + Training...: 62% 2721/4393 [3:56:27<2:32:11, 5.46s/it] + Training...: 62% 2722/4393 [3:56:33<2:30:27, 5.40s/it] + Training...: 62% 2723/4393 [3:56:38<2:29:35, 5.37s/it] + Training...: 62% 2724/4393 [3:56:44<2:30:04, 5.39s/it] + Training...: 62% 2725/4393 [3:56:49<2:29:35, 5.38s/it] + Training...: 62% 2726/4393 [3:56:54<2:27:28, 5.31s/it] + Training...: 62% 2727/4393 [3:56:59<2:25:29, 5.24s/it] + Training...: 62% 2728/4393 [3:57:04<2:23:29, 5.17s/it] + Training...: 62% 2729/4393 [3:57:09<2:21:28, 5.10s/it] + Training...: 62% 2730/4393 [3:57:14<2:19:25, 5.03s/it] + Training...: 62% 2731/4393 [3:57:19<2:17:27, 4.96s/it] + Training...: 62% 2732/4393 [3:57:23<2:15:06, 4.88s/it] + Training...: 62% 2733/4393 [3:57:28<2:12:58, 4.81s/it] + Training...: 62% 2734/4393 [3:57:33<2:10:36, 4.72s/it] + Training...: 62% 2735/4393 [3:57:37<2:07:57, 4.63s/it] + Training...: 62% 2736/4393 [3:57:41<2:05:25, 4.54s/it] + Training...: 62% 2737/4393 [3:57:45<2:01:40, 4.41s/it] + Training...: 62% 2738/4393 [3:57:49<1:57:48, 4.27s/it] +  + Training...: 62% 2738/4393 [3:57:54<1:57:48, 4.27s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:27:26<57:16:35, 20619.59s/it] + Training...: 62% 2739/4393 [3:57:54<1:58:25, 4.30s/it] + Training...: 62% 2740/4393 [3:57:57<1:52:26, 4.08s/it] + Training...: 62% 2741/4393 [3:58:01<1:46:03, 3.85s/it] + Training...: 62% 2742/4393 [3:58:04<1:39:16, 3.61s/it] + Training...: 62% 2743/4393 [3:58:06<1:32:14, 3.35s/it] + Training...: 62% 2744/4393 [3:58:09<1:25:43, 3.12s/it] + Training...: 62% 2745/4393 [3:58:11<1:18:21, 2.85s/it] + Training...: 63% 2746/4393 [3:58:13<1:10:44, 2.58s/it] + Training...: 63% 2747/4393 [3:58:15<1:03:28, 2.31s/it] + Training...: 63% 2748/4393 [3:58:16<56:38, 2.07s/it]  + Training...: 63% 2749/4393 [3:58:18<50:01, 1.83s/it] + Training...: 63% 2750/4393 [3:58:19<42:54, 1.57s/it] + Training...: 63% 2751/4393 [3:58:25<1:20:56, 2.96s/it] + Training...: 63% 2752/4393 [3:58:31<1:46:52, 3.91s/it] + Training...: 63% 2753/4393 [3:58:37<2:04:29, 4.55s/it] + Training...: 63% 2754/4393 [3:58:43<2:16:29, 5.00s/it] + Training...: 63% 2755/4393 [3:58:49<2:25:39, 5.34s/it] + Training...: 63% 2756/4393 [3:58:55<2:29:43, 5.49s/it] + Training...: 63% 2757/4393 [3:59:01<2:32:33, 5.60s/it] + Training...: 63% 2758/4393 [3:59:07<2:33:55, 5.65s/it] + Training...: 63% 2759/4393 [3:59:12<2:35:16, 5.70s/it] + Training...: 63% 2760/4393 [3:59:18<2:35:32, 5.71s/it] + Training...: 63% 2761/4393 [3:59:24<2:35:46, 5.73s/it] + Training...: 63% 2762/4393 [3:59:30<2:35:14, 5.71s/it] + Training...: 63% 2763/4393 [3:59:35<2:35:05, 5.71s/it] +  + Training...: 63% 2763/4393 [3:59:42<2:35:05, 5.71s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:29:14<57:16:35, 20619.59s/it] + Training...: 63% 2764/4393 [3:59:42<2:41:00, 5.93s/it] + Training...: 63% 2765/4393 [3:59:47<2:38:20, 5.84s/it] + Training...: 63% 2766/4393 [3:59:53<2:35:33, 5.74s/it] + Training...: 63% 2767/4393 [3:59:58<2:33:53, 5.68s/it] + Training...: 63% 2768/4393 [4:00:04<2:31:52, 5.61s/it] + Training...: 63% 2769/4393 [4:00:09<2:30:33, 5.56s/it] + Training...: 63% 2770/4393 [4:00:15<2:28:59, 5.51s/it] + Training...: 63% 2771/4393 [4:00:20<2:29:40, 5.54s/it] + Training...: 63% 2772/4393 [4:00:26<2:28:42, 5.50s/it] + Training...: 63% 2773/4393 [4:00:31<2:26:45, 5.44s/it] + Training...: 63% 2774/4393 [4:00:36<2:24:58, 5.37s/it] + Training...: 63% 2775/4393 [4:00:41<2:23:38, 5.33s/it] + Training...: 63% 2776/4393 [4:00:47<2:22:29, 5.29s/it] + Training...: 63% 2777/4393 [4:00:52<2:20:44, 5.23s/it] + Training...: 63% 2778/4393 [4:00:57<2:19:15, 5.17s/it] + Training...: 63% 2779/4393 [4:01:02<2:17:53, 5.13s/it] + Training...: 63% 2780/4393 [4:01:07<2:15:47, 5.05s/it] + Training...: 63% 2781/4393 [4:01:11<2:13:48, 4.98s/it] + Training...: 63% 2782/4393 [4:01:16<2:12:20, 4.93s/it] + Training...: 63% 2783/4393 [4:01:21<2:10:35, 4.87s/it] + Training...: 63% 2784/4393 [4:01:26<2:08:45, 4.80s/it] + Training...: 63% 2785/4393 [4:01:30<2:06:18, 4.71s/it] + Training...: 63% 2786/4393 [4:01:35<2:03:32, 4.61s/it] + Training...: 63% 2787/4393 [4:01:39<2:00:28, 4.50s/it] + Training...: 63% 2788/4393 [4:01:43<1:57:08, 4.38s/it] +  + Training...: 63% 2788/4393 [4:01:47<1:57:08, 4.38s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:31:20<57:16:35, 20619.59s/it] + Training...: 63% 2789/4393 [4:01:47<1:58:18, 4.43s/it] + Training...: 64% 2790/4393 [4:01:51<1:52:04, 4.19s/it] + Training...: 64% 2791/4393 [4:01:54<1:45:23, 3.95s/it] + Training...: 64% 2792/4393 [4:01:58<1:39:07, 3.72s/it] + Training...: 64% 2793/4393 [4:02:01<1:33:16, 3.50s/it] + Training...: 64% 2794/4393 [4:02:03<1:27:27, 3.28s/it] + Training...: 64% 2795/4393 [4:02:06<1:20:11, 3.01s/it] + Training...: 64% 2796/4393 [4:02:08<1:12:21, 2.72s/it] + Training...: 64% 2797/4393 [4:02:10<1:04:44, 2.43s/it] + Training...: 64% 2798/4393 [4:02:11<57:26, 2.16s/it]  + Training...: 64% 2799/4393 [4:02:12<49:51, 1.88s/it] + Training...: 64% 2800/4393 [4:02:13<42:30, 1.60s/it] + Training...: 64% 2801/4393 [4:02:19<1:19:06, 2.98s/it] + Training...: 64% 2802/4393 [4:02:26<1:44:36, 3.95s/it] + Training...: 64% 2803/4393 [4:02:32<2:00:46, 4.56s/it] + Training...: 64% 2804/4393 [4:02:38<2:11:51, 4.98s/it] + Training...: 64% 2805/4393 [4:02:44<2:19:31, 5.27s/it] + Training...: 64% 2806/4393 [4:02:49<2:24:10, 5.45s/it] + Training...: 64% 2807/4393 [4:02:55<2:27:29, 5.58s/it] + Training...: 64% 2808/4393 [4:03:01<2:29:11, 5.65s/it] + Training...: 64% 2809/4393 [4:03:07<2:30:02, 5.68s/it] + Training...: 64% 2810/4393 [4:03:13<2:29:51, 5.68s/it] + Training...: 64% 2811/4393 [4:03:18<2:30:13, 5.70s/it] + Training...: 64% 2812/4393 [4:03:24<2:29:41, 5.68s/it] + Training...: 64% 2813/4393 [4:03:30<2:29:47, 5.69s/it] +  + Training...: 64% 2813/4393 [4:03:36<2:29:47, 5.69s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:33:08<57:16:35, 20619.59s/it] + Training...: 64% 2814/4393 [4:03:36<2:35:51, 5.92s/it] + Training...: 64% 2815/4393 [4:03:42<2:32:48, 5.81s/it] + Training...: 64% 2816/4393 [4:03:47<2:30:42, 5.73s/it] + Training...: 64% 2817/4393 [4:03:53<2:29:12, 5.68s/it] + Training...: 64% 2818/4393 [4:03:58<2:27:37, 5.62s/it] + Training...: 64% 2819/4393 [4:04:04<2:25:56, 5.56s/it] + Training...: 64% 2820/4393 [4:04:09<2:24:20, 5.51s/it] + Training...: 64% 2821/4393 [4:04:14<2:22:52, 5.45s/it] + Training...: 64% 2822/4393 [4:04:20<2:21:29, 5.40s/it] + Training...: 64% 2823/4393 [4:04:25<2:20:15, 5.36s/it] + Training...: 64% 2824/4393 [4:04:30<2:19:04, 5.32s/it] + Training...: 64% 2825/4393 [4:04:35<2:17:50, 5.27s/it] + Training...: 64% 2826/4393 [4:04:40<2:16:50, 5.24s/it] + Training...: 64% 2827/4393 [4:04:46<2:15:40, 5.20s/it] + Training...: 64% 2828/4393 [4:04:51<2:15:05, 5.18s/it] + Training...: 64% 2829/4393 [4:04:56<2:14:21, 5.15s/it] + Training...: 64% 2830/4393 [4:05:01<2:12:59, 5.10s/it] + Training...: 64% 2831/4393 [4:05:06<2:10:47, 5.02s/it] + Training...: 64% 2832/4393 [4:05:10<2:08:13, 4.93s/it] + Training...: 64% 2833/4393 [4:05:15<2:05:37, 4.83s/it] + Training...: 65% 2834/4393 [4:05:19<2:03:22, 4.75s/it] + Training...: 65% 2835/4393 [4:05:24<2:00:54, 4.66s/it] + Training...: 65% 2836/4393 [4:05:28<1:58:02, 4.55s/it] + Training...: 65% 2837/4393 [4:05:32<1:54:30, 4.42s/it] + Training...: 65% 2838/4393 [4:05:36<1:50:37, 4.27s/it] +  + Training...: 65% 2838/4393 [4:05:41<1:50:37, 4.27s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:35:13<57:16:35, 20619.59s/it] + Training...: 65% 2839/4393 [4:05:41<1:51:00, 4.29s/it] + Training...: 65% 2840/4393 [4:05:44<1:44:26, 4.04s/it] + Training...: 65% 2841/4393 [4:05:47<1:38:08, 3.79s/it] + Training...: 65% 2842/4393 [4:05:50<1:31:37, 3.54s/it] + Training...: 65% 2843/4393 [4:05:53<1:25:41, 3.32s/it] + Training...: 65% 2844/4393 [4:05:56<1:19:29, 3.08s/it] + Training...: 65% 2845/4393 [4:05:58<1:13:01, 2.83s/it] + Training...: 65% 2846/4393 [4:06:00<1:06:23, 2.57s/it] + Training...: 65% 2847/4393 [4:06:01<59:43, 2.32s/it]  + Training...: 65% 2848/4393 [4:06:03<53:35, 2.08s/it] + Training...: 65% 2849/4393 [4:06:04<47:02, 1.83s/it] + Training...: 65% 2850/4393 [4:06:05<40:25, 1.57s/it] + Training...: 65% 2851/4393 [4:06:11<1:15:45, 2.95s/it] + Training...: 65% 2852/4393 [4:06:18<1:40:24, 3.91s/it] + Training...: 65% 2853/4393 [4:06:24<1:57:07, 4.56s/it] + Training...: 65% 2854/4393 [4:06:30<2:07:22, 4.97s/it] + Training...: 65% 2855/4393 [4:06:35<2:14:54, 5.26s/it] + Training...: 65% 2856/4393 [4:06:41<2:19:40, 5.45s/it] + Training...: 65% 2857/4393 [4:06:47<2:22:26, 5.56s/it] + Training...: 65% 2858/4393 [4:06:53<2:24:31, 5.65s/it] + Training...: 65% 2859/4393 [4:06:59<2:25:57, 5.71s/it] + Training...: 65% 2860/4393 [4:07:05<2:26:52, 5.75s/it] + Training...: 65% 2861/4393 [4:07:11<2:28:01, 5.80s/it] + Training...: 65% 2862/4393 [4:07:16<2:26:58, 5.76s/it] + Training...: 65% 2863/4393 [4:07:22<2:26:40, 5.75s/it] +  + Training...: 65% 2863/4393 [4:07:29<2:26:40, 5.75s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:37:01<57:16:35, 20619.59s/it] + Training...: 65% 2864/4393 [4:07:29<2:32:34, 5.99s/it] + Training...: 65% 2865/4393 [4:07:34<2:29:19, 5.86s/it] + Training...: 65% 2866/4393 [4:07:40<2:27:01, 5.78s/it] + Training...: 65% 2867/4393 [4:07:45<2:24:42, 5.69s/it] + Training...: 65% 2868/4393 [4:07:51<2:22:48, 5.62s/it] + Training...: 65% 2869/4393 [4:07:56<2:21:22, 5.57s/it] + Training...: 65% 2870/4393 [4:08:01<2:19:45, 5.51s/it] + Training...: 65% 2871/4393 [4:08:07<2:18:44, 5.47s/it] + Training...: 65% 2872/4393 [4:08:12<2:18:15, 5.45s/it] + Training...: 65% 2873/4393 [4:08:18<2:17:51, 5.44s/it] + Training...: 65% 2874/4393 [4:08:23<2:17:17, 5.42s/it] + Training...: 65% 2875/4393 [4:08:28<2:16:52, 5.41s/it] + Training...: 65% 2876/4393 [4:08:34<2:15:48, 5.37s/it] + Training...: 65% 2877/4393 [4:08:39<2:15:07, 5.35s/it] + Training...: 66% 2878/4393 [4:08:44<2:14:07, 5.31s/it] + Training...: 66% 2879/4393 [4:08:49<2:13:29, 5.29s/it] + Training...: 66% 2880/4393 [4:08:55<2:11:54, 5.23s/it] + Training...: 66% 2881/4393 [4:09:00<2:09:27, 5.14s/it] + Training...: 66% 2882/4393 [4:09:04<2:07:52, 5.08s/it] + Training...: 66% 2883/4393 [4:09:09<2:05:56, 5.00s/it] + Training...: 66% 2884/4393 [4:09:14<2:03:45, 4.92s/it] + Training...: 66% 2885/4393 [4:09:19<2:01:57, 4.85s/it] + Training...: 66% 2886/4393 [4:09:23<1:59:03, 4.74s/it] + Training...: 66% 2887/4393 [4:09:27<1:55:53, 4.62s/it] + Training...: 66% 2888/4393 [4:09:32<1:52:25, 4.48s/it] +  + Training...: 66% 2888/4393 [4:09:36<1:52:25, 4.48s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:39:09<57:16:35, 20619.59s/it] + Training...: 66% 2889/4393 [4:09:36<1:53:34, 4.53s/it] + Training...: 66% 2890/4393 [4:09:40<1:47:13, 4.28s/it] + Training...: 66% 2891/4393 [4:09:43<1:40:25, 4.01s/it] + Training...: 66% 2892/4393 [4:09:47<1:33:47, 3.75s/it] + Training...: 66% 2893/4393 [4:09:49<1:27:16, 3.49s/it] + Training...: 66% 2894/4393 [4:09:52<1:20:32, 3.22s/it] + Training...: 66% 2895/4393 [4:09:54<1:13:47, 2.96s/it] + Training...: 66% 2896/4393 [4:09:56<1:06:54, 2.68s/it] + Training...: 66% 2897/4393 [4:09:58<59:54, 2.40s/it]  + Training...: 66% 2898/4393 [4:10:00<53:07, 2.13s/it] + Training...: 66% 2899/4393 [4:10:01<46:34, 1.87s/it] + Training...: 66% 2900/4393 [4:10:02<39:53, 1.60s/it] + Training...: 66% 2901/4393 [4:10:08<1:15:15, 3.03s/it] + Training...: 66% 2902/4393 [4:10:14<1:39:12, 3.99s/it] + Training...: 66% 2903/4393 [4:10:21<1:54:31, 4.61s/it] + Training...: 66% 2904/4393 [4:10:26<2:04:11, 5.00s/it] + Training...: 66% 2905/4393 [4:10:32<2:11:04, 5.29s/it] + Training...: 66% 2906/4393 [4:10:38<2:15:00, 5.45s/it] + Training...: 66% 2907/4393 [4:10:44<2:17:50, 5.57s/it] + Training...: 66% 2908/4393 [4:10:50<2:19:37, 5.64s/it] + Training...: 66% 2909/4393 [4:10:56<2:20:36, 5.68s/it] + Training...: 66% 2910/4393 [4:11:01<2:21:05, 5.71s/it] + Training...: 66% 2911/4393 [4:11:07<2:21:18, 5.72s/it] + Training...: 66% 2912/4393 [4:11:13<2:20:52, 5.71s/it] + Training...: 66% 2913/4393 [4:11:19<2:20:29, 5.70s/it] +  + Training...: 66% 2913/4393 [4:11:25<2:20:29, 5.70s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:40:57<57:16:35, 20619.59s/it] + Training...: 66% 2914/4393 [4:11:25<2:25:58, 5.92s/it] + Training...: 66% 2915/4393 [4:11:31<2:24:17, 5.86s/it] + Training...: 66% 2916/4393 [4:11:36<2:21:39, 5.75s/it] + Training...: 66% 2917/4393 [4:11:42<2:20:01, 5.69s/it] + Training...: 66% 2918/4393 [4:11:47<2:18:28, 5.63s/it] + Training...: 66% 2919/4393 [4:11:53<2:16:48, 5.57s/it] + Training...: 66% 2920/4393 [4:11:58<2:16:15, 5.55s/it] + Training...: 66% 2921/4393 [4:12:04<2:15:03, 5.51s/it] + Training...: 67% 2922/4393 [4:12:09<2:13:51, 5.46s/it] + Training...: 67% 2923/4393 [4:12:14<2:13:14, 5.44s/it] + Training...: 67% 2924/4393 [4:12:20<2:12:24, 5.41s/it] + Training...: 67% 2925/4393 [4:12:25<2:11:29, 5.37s/it] + Training...: 67% 2926/4393 [4:12:30<2:11:03, 5.36s/it] + Training...: 67% 2927/4393 [4:12:35<2:09:52, 5.32s/it] + Training...: 67% 2928/4393 [4:12:41<2:07:55, 5.24s/it] + Training...: 67% 2929/4393 [4:12:46<2:06:19, 5.18s/it] + Training...: 67% 2930/4393 [4:12:51<2:05:09, 5.13s/it] + Training...: 67% 2931/4393 [4:12:56<2:04:04, 5.09s/it] + Training...: 67% 2932/4393 [4:13:01<2:02:53, 5.05s/it] + Training...: 67% 2933/4393 [4:13:05<2:00:08, 4.94s/it] + Training...: 67% 2934/4393 [4:13:10<1:57:42, 4.84s/it] + Training...: 67% 2935/4393 [4:13:14<1:55:01, 4.73s/it] + Training...: 67% 2936/4393 [4:13:19<1:52:12, 4.62s/it] + Training...: 67% 2937/4393 [4:13:23<1:49:11, 4.50s/it] + Training...: 67% 2938/4393 [4:13:27<1:45:59, 4.37s/it] +  + Training...: 67% 2938/4393 [4:13:31<1:45:59, 4.37s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:43:04<57:16:35, 20619.59s/it] + Training...: 67% 2939/4393 [4:13:31<1:46:45, 4.41s/it] + Training...: 67% 2940/4393 [4:13:35<1:40:23, 4.15s/it] + Training...: 67% 2941/4393 [4:13:38<1:34:24, 3.90s/it] + Training...: 67% 2942/4393 [4:13:41<1:28:26, 3.66s/it] + Training...: 67% 2943/4393 [4:13:44<1:22:18, 3.41s/it] + Training...: 67% 2944/4393 [4:13:47<1:16:25, 3.16s/it] + Training...: 67% 2945/4393 [4:13:49<1:10:10, 2.91s/it] + Training...: 67% 2946/4393 [4:13:51<1:04:34, 2.68s/it] + Training...: 67% 2947/4393 [4:13:53<58:29, 2.43s/it]  + Training...: 67% 2948/4393 [4:13:55<52:14, 2.17s/it] + Training...: 67% 2949/4393 [4:13:56<45:46, 1.90s/it] + Training...: 67% 2950/4393 [4:13:57<39:25, 1.64s/it] + Training...: 67% 2951/4393 [4:14:03<1:12:30, 3.02s/it] + Training...: 67% 2952/4393 [4:14:09<1:35:11, 3.96s/it] + Training...: 67% 2953/4393 [4:14:15<1:50:13, 4.59s/it] + Training...: 67% 2954/4393 [4:14:22<2:01:53, 5.08s/it] + Training...: 67% 2955/4393 [4:14:28<2:08:15, 5.35s/it] + Training...: 67% 2956/4393 [4:14:34<2:12:03, 5.51s/it] + Training...: 67% 2957/4393 [4:14:39<2:14:19, 5.61s/it] + Training...: 67% 2958/4393 [4:14:45<2:15:32, 5.67s/it] + Training...: 67% 2959/4393 [4:14:51<2:16:22, 5.71s/it] + Training...: 67% 2960/4393 [4:14:57<2:16:29, 5.71s/it] + Training...: 67% 2961/4393 [4:15:02<2:16:36, 5.72s/it] + Training...: 67% 2962/4393 [4:15:08<2:16:26, 5.72s/it] + Training...: 67% 2963/4393 [4:15:14<2:15:39, 5.69s/it] +  + Training...: 67% 2963/4393 [4:15:20<2:15:39, 5.69s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:44:53<57:16:35, 20619.59s/it] + Training...: 67% 2964/4393 [4:15:20<2:21:10, 5.93s/it]Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |) +Step... (10025 | Loss: 0.08830500394105911, Learning Rate: 8.075959340203553e-05, Gradient Norm: 0.5834270119667053) +Step... (10050 | Loss: 0.09379326552152634, Learning Rate: 8.070908370427787e-05, Gradient Norm: 0.5947138667106628) +Step... (10075 | Loss: 0.09823241084814072, Learning Rate: 8.065858128247783e-05, Gradient Norm: 0.6362175345420837) +Step... (10100 | Loss: 0.06507033109664917, Learning Rate: 8.060807886067778e-05, Gradient Norm: 0.44929641485214233) +Step... (10125 | Loss: 0.08986574411392212, Learning Rate: 8.055756916292012e-05, Gradient Norm: 0.5437695384025574) +Step... (10150 | Loss: 0.07754779607057571, Learning Rate: 8.050706674112007e-05, Gradient Norm: 0.5932506918907166) +Step... (10175 | Loss: 0.06109367311000824, Learning Rate: 8.045656431932002e-05, Gradient Norm: 0.48796671628952026) +Step... (10200 | Loss: 0.049698445945978165, Learning Rate: 8.040605462156236e-05, Gradient Norm: 0.8630101680755615) +Step... (10225 | Loss: 0.10865950584411621, Learning Rate: 8.035555219976231e-05, Gradient Norm: 0.8655366897583008) +Step... (10250 | Loss: 0.09409575164318085, Learning Rate: 8.030504977796227e-05, Gradient Norm: 0.4608936905860901) +Step... (10275 | Loss: 0.05792190134525299, Learning Rate: 8.02545400802046e-05, Gradient Norm: 0.48416876792907715) +Step... (10300 | Loss: 0.07172185927629471, Learning Rate: 8.020403765840456e-05, Gradient Norm: 0.4808516502380371) +Step... (10325 | Loss: 0.11542685329914093, Learning Rate: 8.015353523660451e-05, Gradient Norm: 0.5415076017379761) +Step... (10350 | Loss: 0.09445095807313919, Learning Rate: 8.010302553884685e-05, Gradient Norm: 0.7509729266166687) +Step... (10375 | Loss: 0.0917729064822197, Learning Rate: 8.00525231170468e-05, Gradient Norm: 0.7841768264770508) +Step... (10400 | Loss: 0.09692507982254028, Learning Rate: 8.000202069524676e-05, Gradient Norm: 0.4939047396183014) +Step... (10425 | Loss: 0.10967368632555008, Learning Rate: 7.99515109974891e-05, Gradient Norm: 0.6144236922264099) +Step... (10450 | Loss: 0.08952171355485916, Learning Rate: 7.990100857568905e-05, Gradient Norm: 0.4610723853111267) +Step... (10475 | Loss: 0.09783611446619034, Learning Rate: 7.9850506153889e-05, Gradient Norm: 1.0754166841506958) +Step... (10500 | Loss: 0.09750466048717499, Learning Rate: 7.979999645613134e-05, Gradient Norm: 1.0963139533996582) +Step... (10525 | Loss: 0.12679395079612732, Learning Rate: 7.974949403433129e-05, Gradient Norm: 0.8020402789115906) +Step... (10550 | Loss: 0.09429805725812912, Learning Rate: 7.969899161253124e-05, Gradient Norm: 1.0350173711776733) +Step... (10575 | Loss: 0.10793081670999527, Learning Rate: 7.964848191477358e-05, Gradient Norm: 0.6100561618804932) +Step... (10600 | Loss: 0.08640165627002716, Learning Rate: 7.959797949297354e-05, Gradient Norm: 0.7656195163726807) +Step... (10625 | Loss: 0.08220628648996353, Learning Rate: 7.954747707117349e-05, Gradient Norm: 0.5055153965950012) +Step... (10650 | Loss: 0.05531294271349907, Learning Rate: 7.949696737341583e-05, Gradient Norm: 0.3716430068016052) +Step... (10675 | Loss: 0.10128755122423172, Learning Rate: 7.944646495161578e-05, Gradient Norm: 0.6751567125320435) +Step... (10700 | Loss: 0.09419593960046768, Learning Rate: 7.939596252981573e-05, Gradient Norm: 0.5146047472953796) +Step... (10725 | Loss: 0.09003331512212753, Learning Rate: 7.934545283205807e-05, Gradient Norm: 0.6622697710990906) +Step... (10750 | Loss: 0.08137448877096176, Learning Rate: 7.929495041025802e-05, Gradient Norm: 0.4655037224292755) +Step... (10775 | Loss: 0.11204398423433304, Learning Rate: 7.924444798845798e-05, Gradient Norm: 0.795628011226654) +Step... (10800 | Loss: 0.07448437809944153, Learning Rate: 7.919393829070032e-05, Gradient Norm: 0.8185940384864807) +Step... (10825 | Loss: 0.08662033081054688, Learning Rate: 7.914343586890027e-05, Gradient Norm: 1.1144791841506958) +Step... (10850 | Loss: 0.08633825182914734, Learning Rate: 7.909292617114261e-05, Gradient Norm: 0.5531063675880432) +Step... (10875 | Loss: 0.08835204690694809, Learning Rate: 7.904242374934256e-05, Gradient Norm: 0.5773115158081055) +Step... (10900 | Loss: 0.07958956062793732, Learning Rate: 7.899192132754251e-05, Gradient Norm: 0.456502765417099) +Step... (10925 | Loss: 0.08420322835445404, Learning Rate: 7.894141162978485e-05, Gradient Norm: 0.6213755011558533) +Step... (10950 | Loss: 0.09074170887470245, Learning Rate: 7.88909092079848e-05, Gradient Norm: 0.5302470326423645) +Step... (10975 | Loss: 0.12342730909585953, Learning Rate: 7.884039951022714e-05, Gradient Norm: 0.6465903520584106) +Step... (11000 | Loss: 0.08631566911935806, Learning Rate: 7.87898970884271e-05, Gradient Norm: 0.8472545146942139) +Step... (11025 | Loss: 0.09519127011299133, Learning Rate: 7.873938739066944e-05, Gradient Norm: 1.0324699878692627) +Step... (11050 | Loss: 0.07494750618934631, Learning Rate: 7.868888496886939e-05, Gradient Norm: 0.47489726543426514) +Step... (11075 | Loss: 0.08446040004491806, Learning Rate: 7.863838254706934e-05, Gradient Norm: 0.4666726291179657) +Step... (11100 | Loss: 0.11089751124382019, Learning Rate: 7.858787284931168e-05, Gradient Norm: 0.5858631134033203) +Step... (11125 | Loss: 0.08997974544763565, Learning Rate: 7.853737042751163e-05, Gradient Norm: 0.5574468970298767) +Step... (11150 | Loss: 0.0976918414235115, Learning Rate: 7.848686800571159e-05, Gradient Norm: 0.7729601263999939) +Step... (11175 | Loss: 0.11596120148897171, Learning Rate: 7.843635830795392e-05, Gradient Norm: 0.6145332455635071) +Step... (11200 | Loss: 0.14628462493419647, Learning Rate: 7.838585588615388e-05, Gradient Norm: 0.7486353516578674) +Step... (11225 | Loss: 0.08686906844377518, Learning Rate: 7.833535346435383e-05, Gradient Norm: 0.6828690767288208) +Step... (11250 | Loss: 0.08807458728551865, Learning Rate: 7.828484376659617e-05, Gradient Norm: 0.48905521631240845) +Step... (11275 | Loss: 0.0733414962887764, Learning Rate: 7.823434134479612e-05, Gradient Norm: 0.6179749965667725) +Step... (11300 | Loss: 0.08363909274339676, Learning Rate: 7.818383892299607e-05, Gradient Norm: 0.5266070365905762) +Step... (11325 | Loss: 0.09379070997238159, Learning Rate: 7.813332922523841e-05, Gradient Norm: 2.2078020572662354) +Step... (11350 | Loss: 0.12219930440187454, Learning Rate: 7.808282680343837e-05, Gradient Norm: 0.7328922748565674) +Step... (11375 | Loss: 0.12322746962308884, Learning Rate: 7.803232438163832e-05, Gradient Norm: 0.7159468531608582) +Step... (11400 | Loss: 0.07447610795497894, Learning Rate: 7.798181468388066e-05, Gradient Norm: 0.4881691336631775) +Step... (11425 | Loss: 0.11930616945028305, Learning Rate: 7.793131226208061e-05, Gradient Norm: 0.7435767650604248) +Step... (11450 | Loss: 0.12145199626684189, Learning Rate: 7.788080984028056e-05, Gradient Norm: 0.5640356540679932) +Step... (11475 | Loss: 0.062346503138542175, Learning Rate: 7.78303001425229e-05, Gradient Norm: 0.7859042882919312) +Step... (11500 | Loss: 0.07021258771419525, Learning Rate: 7.777979772072285e-05, Gradient Norm: 0.4334392547607422) +Step... (11525 | Loss: 0.09780385345220566, Learning Rate: 7.772929529892281e-05, Gradient Norm: 0.5772568583488464) +Step... (11550 | Loss: 0.07436276227235794, Learning Rate: 7.767878560116515e-05, Gradient Norm: 0.450169175863266) +Step... (11575 | Loss: 0.08741279691457748, Learning Rate: 7.76282831793651e-05, Gradient Norm: 0.5794440507888794) +Step... (11600 | Loss: 0.07495800405740738, Learning Rate: 7.757777348160744e-05, Gradient Norm: 0.43371060490608215) +Step... (11625 | Loss: 0.11196848750114441, Learning Rate: 7.752727105980739e-05, Gradient Norm: 0.6372781991958618) +Step... (11650 | Loss: 0.08077137917280197, Learning Rate: 7.747676863800734e-05, Gradient Norm: 13.57120132446289) +Step... (11675 | Loss: 0.12308752536773682, Learning Rate: 7.742625894024968e-05, Gradient Norm: 0.7368566393852234) +Step... (11700 | Loss: 0.057600561529397964, Learning Rate: 7.737575651844963e-05, Gradient Norm: 0.3913513720035553) +Step... (11725 | Loss: 0.10907137393951416, Learning Rate: 7.732525409664959e-05, Gradient Norm: 0.5861091613769531) + + Training...: 67% 2965/4393 [4:15:26<2:20:08, 5.89s/it] + Training...: 68% 2966/4393 [4:15:32<2:17:47, 5.79s/it] + Training...: 68% 2967/4393 [4:15:37<2:16:06, 5.73s/it] + Training...: 68% 2968/4393 [4:15:43<2:14:11, 5.65s/it] + Training...: 68% 2969/4393 [4:15:48<2:12:52, 5.60s/it] + Training...: 68% 2970/4393 [4:15:54<2:11:19, 5.54s/it] + Training...: 68% 2971/4393 [4:15:59<2:09:57, 5.48s/it] + Training...: 68% 2972/4393 [4:16:04<2:08:38, 5.43s/it] + Training...: 68% 2973/4393 [4:16:10<2:07:21, 5.38s/it] + Training...: 68% 2974/4393 [4:16:15<2:05:45, 5.32s/it] + Training...: 68% 2975/4393 [4:16:20<2:04:49, 5.28s/it] + Training...: 68% 2976/4393 [4:16:25<2:03:18, 5.22s/it] + Training...: 68% 2977/4393 [4:16:30<2:02:07, 5.17s/it] + Training...: 68% 2978/4393 [4:16:35<2:00:50, 5.12s/it] + Training...: 68% 2979/4393 [4:16:40<1:59:39, 5.08s/it] + Training...: 68% 2980/4393 [4:16:45<1:58:28, 5.03s/it] + Training...: 68% 2981/4393 [4:16:50<1:58:29, 5.04s/it] + Training...: 68% 2982/4393 [4:16:55<1:57:18, 4.99s/it] + Training...: 68% 2983/4393 [4:16:59<1:54:52, 4.89s/it] + Training...: 68% 2984/4393 [4:17:04<1:52:19, 4.78s/it] + Training...: 68% 2985/4393 [4:17:09<1:50:36, 4.71s/it] + Training...: 68% 2986/4393 [4:17:13<1:47:32, 4.59s/it] + Training...: 68% 2987/4393 [4:17:17<1:44:38, 4.47s/it] + Training...: 68% 2988/4393 [4:17:21<1:41:22, 4.33s/it] +  + Training...: 68% 2988/4393 [4:17:25<1:41:22, 4.33s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:46:58<57:16:35, 20619.59s/it] + Training...: 68% 2989/4393 [4:17:25<1:41:39, 4.34s/it] + Training...: 68% 2990/4393 [4:17:29<1:36:05, 4.11s/it] + Training...: 68% 2991/4393 [4:17:32<1:31:01, 3.90s/it] + Training...: 68% 2992/4393 [4:17:36<1:25:31, 3.66s/it] + Training...: 68% 2993/4393 [4:17:38<1:19:51, 3.42s/it] + Training...: 68% 2994/4393 [4:17:41<1:13:36, 3.16s/it] + Training...: 68% 2995/4393 [4:17:43<1:07:24, 2.89s/it] + Training...: 68% 2996/4393 [4:17:45<1:01:38, 2.65s/it] + Training...: 68% 2997/4393 [4:17:47<55:35, 2.39s/it]  + Training...: 68% 2998/4393 [4:17:49<49:49, 2.14s/it] + Training...: 68% 2999/4393 [4:17:50<43:38, 1.88s/it] + Training...: 68% 3000/4393 [4:17:51<37:32, 1.62s/it] + Training...: 68% 3001/4393 [4:17:57<1:09:50, 3.01s/it] + Training...: 68% 3002/4393 [4:18:04<1:33:03, 4.01s/it] + Training...: 68% 3003/4393 [4:18:10<1:48:31, 4.68s/it] + Training...: 68% 3004/4393 [4:18:16<1:57:34, 5.08s/it] + Training...: 68% 3005/4393 [4:18:22<2:03:53, 5.36s/it] + Training...: 68% 3006/4393 [4:18:28<2:08:00, 5.54s/it] + Training...: 68% 3007/4393 [4:18:34<2:10:24, 5.65s/it] + Training...: 68% 3008/4393 [4:18:40<2:11:57, 5.72s/it] + Training...: 68% 3009/4393 [4:18:45<2:13:13, 5.78s/it] + Training...: 69% 3010/4393 [4:18:51<2:13:13, 5.78s/it] + Training...: 69% 3011/4393 [4:18:57<2:13:03, 5.78s/it] + Training...: 69% 3012/4393 [4:19:03<2:13:54, 5.82s/it] + Training...: 69% 3013/4393 [4:19:09<2:14:10, 5.83s/it] +  + Training...: 69% 3013/4393 [4:19:15<2:14:10, 5.83s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:48:48<57:16:35, 20619.59s/it] + Training...: 69% 3014/4393 [4:19:15<2:18:54, 6.04s/it] + Training...: 69% 3015/4393 [4:19:21<2:15:40, 5.91s/it] + Training...: 69% 3016/4393 [4:19:26<2:13:13, 5.81s/it] + Training...: 69% 3017/4393 [4:19:32<2:12:10, 5.76s/it] + Training...: 69% 3018/4393 [4:19:38<2:09:54, 5.67s/it] + Training...: 69% 3019/4393 [4:19:43<2:08:20, 5.60s/it] + Training...: 69% 3020/4393 [4:19:48<2:06:49, 5.54s/it] + Training...: 69% 3021/4393 [4:19:54<2:05:50, 5.50s/it] + Training...: 69% 3022/4393 [4:19:59<2:04:32, 5.45s/it] + Training...: 69% 3023/4393 [4:20:04<2:03:27, 5.41s/it] + Training...: 69% 3024/4393 [4:20:10<2:02:30, 5.37s/it] + Training...: 69% 3025/4393 [4:20:15<2:01:38, 5.33s/it] + Training...: 69% 3026/4393 [4:20:20<2:00:23, 5.28s/it] + Training...: 69% 3027/4393 [4:20:25<1:59:20, 5.24s/it] + Training...: 69% 3028/4393 [4:20:30<1:58:09, 5.19s/it] + Training...: 69% 3029/4393 [4:20:35<1:57:04, 5.15s/it] + Training...: 69% 3030/4393 [4:20:40<1:55:40, 5.09s/it] + Training...: 69% 3031/4393 [4:20:45<1:54:23, 5.04s/it] + Training...: 69% 3032/4393 [4:20:50<1:53:13, 4.99s/it] + Training...: 69% 3033/4393 [4:20:55<1:51:40, 4.93s/it] + Training...: 69% 3034/4393 [4:21:00<1:50:35, 4.88s/it] + Training...: 69% 3035/4393 [4:21:05<1:49:41, 4.85s/it] + Training...: 69% 3036/4393 [4:21:09<1:47:01, 4.73s/it] + Training...: 69% 3037/4393 [4:21:13<1:43:59, 4.60s/it] + Training...: 69% 3038/4393 [4:21:17<1:40:31, 4.45s/it] +  + Training...: 69% 3038/4393 [4:21:22<1:40:31, 4.45s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:50:54<57:16:35, 20619.59s/it] + Training...: 69% 3039/4393 [4:21:22<1:41:07, 4.48s/it] + Training...: 69% 3040/4393 [4:21:26<1:35:34, 4.24s/it] + Training...: 69% 3041/4393 [4:21:29<1:29:59, 3.99s/it] + Training...: 69% 3042/4393 [4:21:32<1:24:36, 3.76s/it] + Training...: 69% 3043/4393 [4:21:35<1:18:49, 3.50s/it] + Training...: 69% 3044/4393 [4:21:38<1:12:53, 3.24s/it] + Training...: 69% 3045/4393 [4:21:40<1:06:33, 2.96s/it] + Training...: 69% 3046/4393 [4:21:42<1:00:41, 2.70s/it] + Training...: 69% 3047/4393 [4:21:44<54:36, 2.43s/it]  + Training...: 69% 3048/4393 [4:21:45<48:21, 2.16s/it] + Training...: 69% 3049/4393 [4:21:47<41:43, 1.86s/it] + Training...: 69% 3050/4393 [4:21:48<35:22, 1.58s/it] + Training...: 69% 3051/4393 [4:21:54<1:06:28, 2.97s/it] + Training...: 69% 3052/4393 [4:22:00<1:28:04, 3.94s/it] + Training...: 69% 3053/4393 [4:22:06<1:42:04, 4.57s/it] + Training...: 70% 3054/4393 [4:22:12<1:51:17, 4.99s/it] + Training...: 70% 3055/4393 [4:22:18<1:57:23, 5.26s/it] + Training...: 70% 3056/4393 [4:22:24<2:01:26, 5.45s/it] + Training...: 70% 3057/4393 [4:22:30<2:04:09, 5.58s/it] + Training...: 70% 3058/4393 [4:22:36<2:08:37, 5.78s/it] + Training...: 70% 3059/4393 [4:22:42<2:08:35, 5.78s/it] + Training...: 70% 3060/4393 [4:22:48<2:08:29, 5.78s/it] + Training...: 70% 3061/4393 [4:22:53<2:08:07, 5.77s/it] + Training...: 70% 3062/4393 [4:22:59<2:07:06, 5.73s/it] + Training...: 70% 3063/4393 [4:23:05<2:06:28, 5.71s/it] +  + Training...: 70% 3063/4393 [4:23:11<2:06:28, 5.71s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:52:43<57:16:35, 20619.59s/it] + Training...: 70% 3064/4393 [4:23:11<2:10:56, 5.91s/it] + Training...: 70% 3065/4393 [4:23:17<2:09:13, 5.84s/it] + Training...: 70% 3066/4393 [4:23:22<2:07:11, 5.75s/it] + Training...: 70% 3067/4393 [4:23:28<2:05:23, 5.67s/it] + Training...: 70% 3068/4393 [4:23:33<2:04:05, 5.62s/it] + Training...: 70% 3069/4393 [4:23:39<2:02:42, 5.56s/it] + Training...: 70% 3070/4393 [4:23:44<2:02:11, 5.54s/it] + Training...: 70% 3071/4393 [4:23:49<2:00:43, 5.48s/it] + Training...: 70% 3072/4393 [4:23:55<1:59:32, 5.43s/it] + Training...: 70% 3073/4393 [4:24:00<1:59:00, 5.41s/it] + Training...: 70% 3074/4393 [4:24:05<1:58:19, 5.38s/it] + Training...: 70% 3075/4393 [4:24:11<1:56:57, 5.32s/it] + Training...: 70% 3076/4393 [4:24:16<1:55:42, 5.27s/it] + Training...: 70% 3077/4393 [4:24:21<1:55:49, 5.28s/it] + Training...: 70% 3078/4393 [4:24:26<1:55:18, 5.26s/it] + Training...: 70% 3079/4393 [4:24:31<1:53:48, 5.20s/it] + Training...: 70% 3080/4393 [4:24:36<1:51:54, 5.11s/it] + Training...: 70% 3081/4393 [4:24:41<1:50:23, 5.05s/it] + Training...: 70% 3082/4393 [4:24:46<1:48:30, 4.97s/it] + Training...: 70% 3083/4393 [4:24:51<1:46:55, 4.90s/it] + Training...: 70% 3084/4393 [4:24:55<1:45:00, 4.81s/it] + Training...: 70% 3085/4393 [4:25:00<1:43:21, 4.74s/it] + Training...: 70% 3086/4393 [4:25:04<1:40:48, 4.63s/it] + Training...: 70% 3087/4393 [4:25:08<1:38:15, 4.51s/it] + Training...: 70% 3088/4393 [4:25:12<1:35:23, 4.39s/it] +  + Training...: 70% 3088/4393 [4:25:17<1:35:23, 4.39s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:54:49<57:16:35, 20619.59s/it] + Training...: 70% 3089/4393 [4:25:17<1:36:26, 4.44s/it] + Training...: 70% 3090/4393 [4:25:21<1:31:09, 4.20s/it] + Training...: 70% 3091/4393 [4:25:24<1:26:06, 3.97s/it] + Training...: 70% 3092/4393 [4:25:27<1:21:11, 3.74s/it] + Training...: 70% 3093/4393 [4:25:30<1:16:02, 3.51s/it] + Training...: 70% 3094/4393 [4:25:33<1:10:28, 3.25s/it] + Training...: 70% 3095/4393 [4:25:35<1:04:43, 2.99s/it] + Training...: 70% 3096/4393 [4:25:37<59:07, 2.73s/it]  + Training...: 70% 3097/4393 [4:25:39<53:25, 2.47s/it] + Training...: 71% 3098/4393 [4:25:41<47:37, 2.21s/it] + Training...: 71% 3099/4393 [4:25:42<41:38, 1.93s/it] + Training...: 71% 3100/4393 [4:25:43<35:29, 1.65s/it] + Training...: 71% 3101/4393 [4:25:49<1:04:54, 3.01s/it] + Training...: 71% 3102/4393 [4:25:56<1:25:15, 3.96s/it] + Training...: 71% 3103/4393 [4:26:02<1:38:50, 4.60s/it] + Training...: 71% 3104/4393 [4:26:08<1:48:16, 5.04s/it] + Training...: 71% 3105/4393 [4:26:14<1:55:11, 5.37s/it] + Training...: 71% 3106/4393 [4:26:20<1:58:30, 5.53s/it] + Training...: 71% 3107/4393 [4:26:26<2:00:50, 5.64s/it] + Training...: 71% 3108/4393 [4:26:32<2:02:22, 5.71s/it] + Training...: 71% 3109/4393 [4:26:38<2:04:31, 5.82s/it] + Training...: 71% 3110/4393 [4:26:43<2:04:49, 5.84s/it] + Training...: 71% 3111/4393 [4:26:49<2:04:09, 5.81s/it] + Training...: 71% 3112/4393 [4:26:55<2:03:27, 5.78s/it] + Training...: 71% 3113/4393 [4:27:01<2:02:42, 5.75s/it] +  + Training...: 71% 3113/4393 [4:27:07<2:02:42, 5.75s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:56:39<57:16:35, 20619.59s/it] + Training...: 71% 3114/4393 [4:27:07<2:07:17, 5.97s/it] + Training...: 71% 3115/4393 [4:27:13<2:04:54, 5.86s/it] + Training...: 71% 3116/4393 [4:27:18<2:02:36, 5.76s/it] + Training...: 71% 3117/4393 [4:27:24<2:01:09, 5.70s/it] + Training...: 71% 3118/4393 [4:27:29<1:59:20, 5.62s/it] + Training...: 71% 3119/4393 [4:27:35<1:58:43, 5.59s/it] + Training...: 71% 3120/4393 [4:27:40<1:57:45, 5.55s/it] + Training...: 71% 3121/4393 [4:27:46<1:56:33, 5.50s/it] + Training...: 71% 3122/4393 [4:27:51<1:55:05, 5.43s/it] + Training...: 71% 3123/4393 [4:27:56<1:54:08, 5.39s/it] + Training...: 71% 3124/4393 [4:28:02<1:53:58, 5.39s/it] + Training...: 71% 3125/4393 [4:28:07<1:53:37, 5.38s/it] + Training...: 71% 3126/4393 [4:28:12<1:51:50, 5.30s/it] + Training...: 71% 3127/4393 [4:28:17<1:52:03, 5.31s/it] + Training...: 71% 3128/4393 [4:28:22<1:50:47, 5.26s/it] + Training...: 71% 3129/4393 [4:28:27<1:49:12, 5.18s/it] + Training...: 71% 3130/4393 [4:28:32<1:47:51, 5.12s/it] + Training...: 71% 3131/4393 [4:28:37<1:45:51, 5.03s/it] + Training...: 71% 3132/4393 [4:28:42<1:43:58, 4.95s/it] + Training...: 71% 3133/4393 [4:28:47<1:42:12, 4.87s/it] + Training...: 71% 3134/4393 [4:28:51<1:40:46, 4.80s/it] + Training...: 71% 3135/4393 [4:28:56<1:39:16, 4.74s/it] + Training...: 71% 3136/4393 [4:29:00<1:37:02, 4.63s/it] + Training...: 71% 3137/4393 [4:29:05<1:34:25, 4.51s/it] + Training...: 71% 3138/4393 [4:29:09<1:31:11, 4.36s/it] +  + Training...: 71% 3138/4393 [4:29:13<1:31:11, 4.36s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [15:58:45<57:16:35, 20619.59s/it] + Training...: 71% 3139/4393 [4:29:13<1:31:43, 4.39s/it] + Training...: 71% 3140/4393 [4:29:17<1:27:28, 4.19s/it] + Training...: 72% 3141/4393 [4:29:20<1:21:59, 3.93s/it] + Training...: 72% 3142/4393 [4:29:23<1:16:54, 3.69s/it] + Training...: 72% 3143/4393 [4:29:26<1:11:20, 3.42s/it] + Training...: 72% 3144/4393 [4:29:29<1:05:30, 3.15s/it] + Training...: 72% 3145/4393 [4:29:31<59:41, 2.87s/it]  + Training...: 72% 3146/4393 [4:29:33<54:05, 2.60s/it] + Training...: 72% 3147/4393 [4:29:34<48:27, 2.33s/it] + Training...: 72% 3148/4393 [4:29:36<42:58, 2.07s/it] + Training...: 72% 3149/4393 [4:29:37<37:41, 1.82s/it] + Training...: 72% 3150/4393 [4:29:38<32:18, 1.56s/it] + Training...: 72% 3151/4393 [4:29:44<1:01:10, 2.96s/it] + Training...: 72% 3152/4393 [4:29:50<1:21:00, 3.92s/it] + Training...: 72% 3153/4393 [4:29:56<1:33:52, 4.54s/it] + Training...: 72% 3154/4393 [4:30:02<1:42:27, 4.96s/it] + Training...: 72% 3155/4393 [4:30:08<1:48:37, 5.26s/it] + Training...: 72% 3156/4393 [4:30:14<1:52:47, 5.47s/it] + Training...: 72% 3157/4393 [4:30:20<1:55:37, 5.61s/it] + Training...: 72% 3158/4393 [4:30:26<1:56:49, 5.68s/it] + Training...: 72% 3159/4393 [4:30:32<1:57:39, 5.72s/it] + Training...: 72% 3160/4393 [4:30:38<1:58:44, 5.78s/it] + Training...: 72% 3161/4393 [4:30:44<1:59:15, 5.81s/it] + Training...: 72% 3162/4393 [4:30:49<1:58:50, 5.79s/it] + Training...: 72% 3163/4393 [4:30:55<1:57:55, 5.75s/it] +  + Training...: 72% 3163/4393 [4:31:02<1:57:55, 5.75s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:00:34<57:16:35, 20619.59s/it] + Training...: 72% 3164/4393 [4:31:02<2:01:52, 5.95s/it] + Training...: 72% 3165/4393 [4:31:07<1:59:15, 5.83s/it] + Training...: 72% 3166/4393 [4:31:13<1:57:27, 5.74s/it] + Training...: 72% 3167/4393 [4:31:18<1:56:06, 5.68s/it] + Training...: 72% 3168/4393 [4:31:24<1:54:47, 5.62s/it] + Training...: 72% 3169/4393 [4:31:29<1:54:59, 5.64s/it] + Training...: 72% 3170/4393 [4:31:35<1:54:51, 5.63s/it] + Training...: 72% 3171/4393 [4:31:40<1:53:34, 5.58s/it] + Training...: 72% 3172/4393 [4:31:46<1:51:55, 5.50s/it] + Training...: 72% 3173/4393 [4:31:51<1:50:42, 5.44s/it] + Training...: 72% 3174/4393 [4:31:56<1:49:25, 5.39s/it] + Training...: 72% 3175/4393 [4:32:01<1:48:06, 5.33s/it] + Training...: 72% 3176/4393 [4:32:07<1:47:08, 5.28s/it] + Training...: 72% 3177/4393 [4:32:12<1:46:30, 5.26s/it] + Training...: 72% 3178/4393 [4:32:17<1:45:21, 5.20s/it] + Training...: 72% 3179/4393 [4:32:22<1:44:03, 5.14s/it] + Training...: 72% 3180/4393 [4:32:27<1:42:47, 5.08s/it] + Training...: 72% 3181/4393 [4:32:32<1:41:48, 5.04s/it] + Training...: 72% 3182/4393 [4:32:37<1:40:31, 4.98s/it] + Training...: 72% 3183/4393 [4:32:42<1:39:57, 4.96s/it] + Training...: 72% 3184/4393 [4:32:46<1:38:13, 4.87s/it] + Training...: 73% 3185/4393 [4:32:51<1:37:01, 4.82s/it] + Training...: 73% 3186/4393 [4:32:55<1:35:05, 4.73s/it] + Training...: 73% 3187/4393 [4:33:00<1:33:39, 4.66s/it] + Training...: 73% 3188/4393 [4:33:04<1:31:12, 4.54s/it] +  + Training...: 73% 3188/4393 [4:33:09<1:31:12, 4.54s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:02:41<57:16:35, 20619.59s/it] + Training...: 73% 3189/4393 [4:33:09<1:31:29, 4.56s/it] + Training...: 73% 3190/4393 [4:33:13<1:26:59, 4.34s/it] + Training...: 73% 3191/4393 [4:33:16<1:22:40, 4.13s/it] + Training...: 73% 3192/4393 [4:33:20<1:18:18, 3.91s/it] + Training...: 73% 3193/4393 [4:33:23<1:13:47, 3.69s/it] + Training...: 73% 3194/4393 [4:33:26<1:08:00, 3.40s/it] + Training...: 73% 3195/4393 [4:33:28<1:02:00, 3.11s/it] + Training...: 73% 3196/4393 [4:33:30<55:45, 2.80s/it]  + Training...: 73% 3197/4393 [4:33:32<49:55, 2.50s/it] + Training...: 73% 3198/4393 [4:33:33<44:07, 2.22s/it] + Training...: 73% 3199/4393 [4:33:35<38:13, 1.92s/it] + Training...: 73% 3200/4393 [4:33:36<32:43, 1.65s/it] + Training...: 73% 3201/4393 [4:33:42<1:00:11, 3.03s/it] + Training...: 73% 3202/4393 [4:33:48<1:19:04, 3.98s/it] + Training...: 73% 3203/4393 [4:33:54<1:30:55, 4.58s/it] + Training...: 73% 3204/4393 [4:34:00<1:38:42, 4.98s/it] + Training...: 73% 3205/4393 [4:34:06<1:44:17, 5.27s/it] + Training...: 73% 3206/4393 [4:34:12<1:47:25, 5.43s/it] + Training...: 73% 3207/4393 [4:34:18<1:49:37, 5.55s/it] + Training...: 73% 3208/4393 [4:34:23<1:51:09, 5.63s/it] + Training...: 73% 3209/4393 [4:34:29<1:51:53, 5.67s/it] + Training...: 73% 3210/4393 [4:34:35<1:52:17, 5.70s/it] + Training...: 73% 3211/4393 [4:34:41<1:52:28, 5.71s/it] + Training...: 73% 3212/4393 [4:34:46<1:51:47, 5.68s/it] + Training...: 73% 3213/4393 [4:34:52<1:51:34, 5.67s/it] +  + Training...: 73% 3213/4393 [4:34:58<1:51:34, 5.67s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:04:31<57:16:35, 20619.59s/it] + Training...: 73% 3214/4393 [4:34:58<1:55:27, 5.88s/it] + Training...: 73% 3215/4393 [4:35:04<1:53:42, 5.79s/it] + Training...: 73% 3216/4393 [4:35:09<1:51:50, 5.70s/it] + Training...: 73% 3217/4393 [4:35:15<1:50:33, 5.64s/it] + Training...: 73% 3218/4393 [4:35:20<1:49:34, 5.59s/it] + Training...: 73% 3219/4393 [4:35:26<1:48:35, 5.55s/it] + Training...: 73% 3220/4393 [4:35:31<1:47:18, 5.49s/it] + Training...: 73% 3221/4393 [4:35:37<1:46:55, 5.47s/it] + Training...: 73% 3222/4393 [4:35:42<1:45:55, 5.43s/it] + Training...: 73% 3223/4393 [4:35:47<1:44:59, 5.38s/it] + Training...: 73% 3224/4393 [4:35:52<1:43:50, 5.33s/it] + Training...: 73% 3225/4393 [4:35:58<1:43:48, 5.33s/it] + Training...: 73% 3226/4393 [4:36:03<1:43:19, 5.31s/it] + Training...: 73% 3227/4393 [4:36:08<1:42:12, 5.26s/it] + Training...: 73% 3228/4393 [4:36:13<1:40:56, 5.20s/it] + Training...: 74% 3229/4393 [4:36:18<1:39:45, 5.14s/it] + Training...: 74% 3230/4393 [4:36:23<1:38:24, 5.08s/it] + Training...: 74% 3231/4393 [4:36:28<1:37:19, 5.03s/it] + Training...: 74% 3232/4393 [4:36:33<1:35:46, 4.95s/it] + Training...: 74% 3233/4393 [4:36:38<1:34:45, 4.90s/it] + Training...: 74% 3234/4393 [4:36:42<1:33:13, 4.83s/it] + Training...: 74% 3235/4393 [4:36:47<1:32:05, 4.77s/it] + Training...: 74% 3236/4393 [4:36:51<1:30:18, 4.68s/it] + Training...: 74% 3237/4393 [4:36:56<1:28:16, 4.58s/it] + Training...: 74% 3238/4393 [4:37:00<1:25:30, 4.44s/it] +  + Training...: 74% 3238/4393 [4:37:04<1:25:30, 4.44s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:06:37<57:16:35, 20619.59s/it] + Training...: 74% 3239/4393 [4:37:04<1:26:07, 4.48s/it] + Training...: 74% 3240/4393 [4:37:08<1:21:14, 4.23s/it] + Training...: 74% 3241/4393 [4:37:11<1:16:29, 3.98s/it] + Training...: 74% 3242/4393 [4:37:15<1:11:35, 3.73s/it] + Training...: 74% 3243/4393 [4:37:17<1:06:06, 3.45s/it] + Training...: 74% 3244/4393 [4:37:20<1:00:50, 3.18s/it] + Training...: 74% 3245/4393 [4:37:22<55:54, 2.92s/it]  + Training...: 74% 3246/4393 [4:37:24<50:56, 2.66s/it] + Training...: 74% 3247/4393 [4:37:26<45:58, 2.41s/it] + Training...: 74% 3248/4393 [4:37:28<41:08, 2.16s/it] + Training...: 74% 3249/4393 [4:37:29<36:26, 1.91s/it] + Training...: 74% 3250/4393 [4:37:30<31:21, 1.65s/it] + Training...: 74% 3251/4393 [4:37:36<57:28, 3.02s/it] + Training...: 74% 3252/4393 [4:37:42<1:15:44, 3.98s/it] + Training...: 74% 3253/4393 [4:37:49<1:27:25, 4.60s/it] + Training...: 74% 3254/4393 [4:37:55<1:35:16, 5.02s/it] + Training...: 74% 3255/4393 [4:38:01<1:40:51, 5.32s/it] + Training...: 74% 3256/4393 [4:38:06<1:44:05, 5.49s/it] + Training...: 74% 3257/4393 [4:38:12<1:47:00, 5.65s/it] + Training...: 74% 3258/4393 [4:38:18<1:48:28, 5.73s/it] + Training...: 74% 3259/4393 [4:38:24<1:49:31, 5.79s/it] + Training...: 74% 3260/4393 [4:38:30<1:49:39, 5.81s/it] + Training...: 74% 3261/4393 [4:38:36<1:49:34, 5.81s/it] + Training...: 74% 3262/4393 [4:38:42<1:49:15, 5.80s/it] + Training...: 74% 3263/4393 [4:38:48<1:49:07, 5.79s/it] +  + Training...: 74% 3263/4393 [4:38:54<1:49:07, 5.79s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:08:27<57:16:35, 20619.59s/it] + Training...: 74% 3264/4393 [4:38:54<1:54:41, 6.09s/it] + Training...: 74% 3265/4393 [4:39:00<1:52:11, 5.97s/it] + Training...: 74% 3266/4393 [4:39:05<1:49:27, 5.83s/it] + Training...: 74% 3267/4393 [4:39:11<1:47:31, 5.73s/it] + Training...: 74% 3268/4393 [4:39:16<1:46:08, 5.66s/it] + Training...: 74% 3269/4393 [4:39:22<1:44:56, 5.60s/it] + Training...: 74% 3270/4393 [4:39:27<1:43:37, 5.54s/it] + Training...: 74% 3271/4393 [4:39:33<1:42:49, 5.50s/it] + Training...: 74% 3272/4393 [4:39:38<1:42:02, 5.46s/it] + Training...: 75% 3273/4393 [4:39:44<1:41:30, 5.44s/it] + Training...: 75% 3274/4393 [4:39:49<1:40:30, 5.39s/it] + Training...: 75% 3275/4393 [4:39:54<1:39:16, 5.33s/it] + Training...: 75% 3276/4393 [4:39:59<1:38:13, 5.28s/it] + Training...: 75% 3277/4393 [4:40:04<1:37:26, 5.24s/it] + Training...: 75% 3278/4393 [4:40:09<1:36:21, 5.19s/it] + Training...: 75% 3279/4393 [4:40:14<1:35:39, 5.15s/it] + Training...: 75% 3280/4393 [4:40:19<1:34:37, 5.10s/it] + Training...: 75% 3281/4393 [4:40:24<1:33:22, 5.04s/it] + Training...: 75% 3282/4393 [4:40:29<1:32:34, 5.00s/it] + Training...: 75% 3283/4393 [4:40:34<1:31:31, 4.95s/it] + Training...: 75% 3284/4393 [4:40:39<1:29:53, 4.86s/it] + Training...: 75% 3285/4393 [4:40:43<1:27:50, 4.76s/it] + Training...: 75% 3286/4393 [4:40:48<1:25:41, 4.64s/it] + Training...: 75% 3287/4393 [4:40:52<1:23:14, 4.52s/it] + Training...: 75% 3288/4393 [4:40:56<1:21:00, 4.40s/it] +  + Training...: 75% 3288/4393 [4:41:00<1:21:00, 4.40s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:10:33<57:16:35, 20619.59s/it] + Training...: 75% 3289/4393 [4:41:00<1:21:56, 4.45s/it] + Training...: 75% 3290/4393 [4:41:04<1:17:19, 4.21s/it] + Training...: 75% 3291/4393 [4:41:08<1:13:08, 3.98s/it] + Training...: 75% 3292/4393 [4:41:11<1:08:35, 3.74s/it] + Training...: 75% 3293/4393 [4:41:14<1:04:08, 3.50s/it] + Training...: 75% 3294/4393 [4:41:16<59:31, 3.25s/it]  + Training...: 75% 3295/4393 [4:41:19<54:24, 2.97s/it] + Training...: 75% 3296/4393 [4:41:21<49:29, 2.71s/it] + Training...: 75% 3297/4393 [4:41:23<44:39, 2.45s/it] + Training...: 75% 3298/4393 [4:41:24<39:49, 2.18s/it] + Training...: 75% 3299/4393 [4:41:25<34:47, 1.91s/it] + Training...: 75% 3300/4393 [4:41:26<29:53, 1.64s/it] + Training...: 75% 3301/4393 [4:41:33<54:24, 2.99s/it] + Training...: 75% 3302/4393 [4:41:39<1:11:32, 3.93s/it] + Training...: 75% 3303/4393 [4:41:45<1:22:56, 4.57s/it] + Training...: 75% 3304/4393 [4:41:51<1:30:21, 4.98s/it] + Training...: 75% 3305/4393 [4:41:57<1:35:44, 5.28s/it] + Training...: 75% 3306/4393 [4:42:03<1:38:46, 5.45s/it] + Training...: 75% 3307/4393 [4:42:08<1:40:32, 5.55s/it] + Training...: 75% 3308/4393 [4:42:14<1:42:53, 5.69s/it] + Training...: 75% 3309/4393 [4:42:20<1:44:10, 5.77s/it] + Training...: 75% 3310/4393 [4:42:26<1:44:10, 5.77s/it] + Training...: 75% 3311/4393 [4:42:32<1:43:49, 5.76s/it] + Training...: 75% 3312/4393 [4:42:37<1:43:17, 5.73s/it] + Training...: 75% 3313/4393 [4:42:43<1:42:52, 5.72s/it] +  + Training...: 75% 3313/4393 [4:42:50<1:42:52, 5.72s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:12:22<57:16:35, 20619.59s/it] + Training...: 75% 3314/4393 [4:42:50<1:46:40, 5.93s/it] + Training...: 75% 3315/4393 [4:42:55<1:44:40, 5.83s/it] + Training...: 75% 3316/4393 [4:43:01<1:42:45, 5.73s/it] + Training...: 76% 3317/4393 [4:43:06<1:41:38, 5.67s/it] + Training...: 76% 3318/4393 [4:43:12<1:40:37, 5.62s/it] + Training...: 76% 3319/4393 [4:43:17<1:40:33, 5.62s/it] + Training...: 76% 3320/4393 [4:43:23<1:39:54, 5.59s/it] + Training...: 76% 3321/4393 [4:43:28<1:38:49, 5.53s/it] + Training...: 76% 3322/4393 [4:43:34<1:37:58, 5.49s/it] + Training...: 76% 3323/4393 [4:43:39<1:37:25, 5.46s/it] + Training...: 76% 3324/4393 [4:43:44<1:36:46, 5.43s/it] + Training...: 76% 3325/4393 [4:43:50<1:35:23, 5.36s/it] + Training...: 76% 3326/4393 [4:43:55<1:34:06, 5.29s/it] + Training...: 76% 3327/4393 [4:44:00<1:33:20, 5.25s/it] + Training...: 76% 3328/4393 [4:44:05<1:32:34, 5.22s/it] + Training...: 76% 3329/4393 [4:44:10<1:32:00, 5.19s/it] + Training...: 76% 3330/4393 [4:44:15<1:30:38, 5.12s/it] + Training...: 76% 3331/4393 [4:44:20<1:29:41, 5.07s/it] + Training...: 76% 3332/4393 [4:44:25<1:27:49, 4.97s/it] + Training...: 76% 3333/4393 [4:44:30<1:26:46, 4.91s/it] + Training...: 76% 3334/4393 [4:44:34<1:25:22, 4.84s/it] + Training...: 76% 3335/4393 [4:44:39<1:24:13, 4.78s/it] + Training...: 76% 3336/4393 [4:44:43<1:23:07, 4.72s/it] + Training...: 76% 3337/4393 [4:44:48<1:20:56, 4.60s/it] + Training...: 76% 3338/4393 [4:44:52<1:18:25, 4.46s/it] +  + Training...: 76% 3338/4393 [4:44:57<1:18:25, 4.46s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:14:29<57:16:35, 20619.59s/it] + Training...: 76% 3339/4393 [4:44:57<1:19:29, 4.53s/it] + Training...: 76% 3340/4393 [4:45:00<1:15:13, 4.29s/it] + Training...: 76% 3341/4393 [4:45:04<1:10:41, 4.03s/it] + Training...: 76% 3342/4393 [4:45:07<1:06:03, 3.77s/it] + Training...: 76% 3343/4393 [4:45:10<1:01:19, 3.50s/it] + Training...: 76% 3344/4393 [4:45:12<56:28, 3.23s/it]  + Training...: 76% 3345/4393 [4:45:15<51:28, 2.95s/it] + Training...: 76% 3346/4393 [4:45:17<46:41, 2.68s/it] + Training...: 76% 3347/4393 [4:45:18<42:03, 2.41s/it] + Training...: 76% 3348/4393 [4:45:20<37:19, 2.14s/it] + Training...: 76% 3349/4393 [4:45:21<32:44, 1.88s/it] + Training...: 76% 3350/4393 [4:45:22<28:18, 1.63s/it] + Training...: 76% 3351/4393 [4:45:28<51:47, 2.98s/it] + Training...: 76% 3352/4393 [4:45:35<1:08:15, 3.93s/it] + Training...: 76% 3353/4393 [4:45:41<1:18:47, 4.55s/it] + Training...: 76% 3354/4393 [4:45:47<1:26:09, 4.98s/it] + Training...: 76% 3355/4393 [4:45:53<1:31:08, 5.27s/it] + Training...: 76% 3356/4393 [4:45:58<1:33:40, 5.42s/it] + Training...: 76% 3357/4393 [4:46:04<1:35:52, 5.55s/it] + Training...: 76% 3358/4393 [4:46:10<1:36:43, 5.61s/it] + Training...: 76% 3359/4393 [4:46:16<1:37:34, 5.66s/it] + Training...: 76% 3360/4393 [4:46:21<1:37:54, 5.69s/it] + Training...: 77% 3361/4393 [4:46:27<1:38:09, 5.71s/it] + Training...: 77% 3362/4393 [4:46:33<1:37:53, 5.70s/it] + Training...: 77% 3363/4393 [4:46:38<1:37:34, 5.68s/it] +  + Training...: 77% 3363/4393 [4:46:45<1:37:34, 5.68s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:16:17<57:16:35, 20619.59s/it] + Training...: 77% 3364/4393 [4:46:45<1:41:20, 5.91s/it] + Training...: 77% 3365/4393 [4:46:51<1:39:39, 5.82s/it] + Training...: 77% 3366/4393 [4:46:56<1:37:52, 5.72s/it] + Training...: 77% 3367/4393 [4:47:02<1:37:36, 5.71s/it] + Training...: 77% 3368/4393 [4:47:07<1:37:16, 5.69s/it] + Training...: 77% 3369/4393 [4:47:13<1:36:23, 5.65s/it] + Training...: 77% 3370/4393 [4:47:18<1:35:09, 5.58s/it] + Training...: 77% 3371/4393 [4:47:24<1:33:55, 5.51s/it] + Training...: 77% 3372/4393 [4:47:29<1:32:40, 5.45s/it] + Training...: 77% 3373/4393 [4:47:34<1:32:01, 5.41s/it] + Training...: 77% 3374/4393 [4:47:40<1:31:20, 5.38s/it] + Training...: 77% 3375/4393 [4:47:45<1:30:40, 5.34s/it] + Training...: 77% 3376/4393 [4:47:50<1:29:33, 5.28s/it] + Training...: 77% 3377/4393 [4:47:55<1:28:46, 5.24s/it] + Training...: 77% 3378/4393 [4:48:00<1:27:48, 5.19s/it] + Training...: 77% 3379/4393 [4:48:05<1:27:06, 5.15s/it] + Training...: 77% 3380/4393 [4:48:10<1:25:44, 5.08s/it] + Training...: 77% 3381/4393 [4:48:15<1:24:43, 5.02s/it] + Training...: 77% 3382/4393 [4:48:20<1:23:23, 4.95s/it] + Training...: 77% 3383/4393 [4:48:25<1:22:23, 4.89s/it] + Training...: 77% 3384/4393 [4:48:29<1:20:59, 4.82s/it] + Training...: 77% 3385/4393 [4:48:34<1:20:08, 4.77s/it] + Training...: 77% 3386/4393 [4:48:38<1:18:02, 4.65s/it] + Training...: 77% 3387/4393 [4:48:43<1:15:43, 4.52s/it] + Training...: 77% 3388/4393 [4:48:47<1:13:08, 4.37s/it] +  + Training...: 77% 3388/4393 [4:48:51<1:13:08, 4.37s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:18:23<57:16:35, 20619.59s/it] + Training...: 77% 3389/4393 [4:48:51<1:14:04, 4.43s/it] + Training...: 77% 3390/4393 [4:48:55<1:10:13, 4.20s/it] + Training...: 77% 3391/4393 [4:48:58<1:06:11, 3.96s/it] + Training...: 77% 3392/4393 [4:49:01<1:01:57, 3.71s/it] + Training...: 77% 3393/4393 [4:49:04<57:38, 3.46s/it]  + Training...: 77% 3394/4393 [4:49:07<53:13, 3.20s/it] + Training...: 77% 3395/4393 [4:49:09<48:26, 2.91s/it] + Training...: 77% 3396/4393 [4:49:11<43:57, 2.65s/it] + Training...: 77% 3397/4393 [4:49:13<39:39, 2.39s/it] + Training...: 77% 3398/4393 [4:49:14<35:28, 2.14s/it] + Training...: 77% 3399/4393 [4:49:16<31:14, 1.89s/it] + Training...: 77% 3400/4393 [4:49:17<26:47, 1.62s/it] + Training...: 77% 3401/4393 [4:49:23<49:19, 2.98s/it] + Training...: 77% 3402/4393 [4:49:29<1:05:15, 3.95s/it] + Training...: 77% 3403/4393 [4:49:35<1:15:43, 4.59s/it] + Training...: 77% 3404/4393 [4:49:41<1:22:30, 5.01s/it] + Training...: 78% 3405/4393 [4:49:47<1:27:09, 5.29s/it] + Training...: 78% 3406/4393 [4:49:53<1:29:53, 5.46s/it] + Training...: 78% 3407/4393 [4:49:59<1:31:39, 5.58s/it] + Training...: 78% 3408/4393 [4:50:05<1:32:46, 5.65s/it] + Training...: 78% 3409/4393 [4:50:10<1:33:28, 5.70s/it] + Training...: 78% 3410/4393 [4:50:16<1:33:32, 5.71s/it] + Training...: 78% 3411/4393 [4:50:22<1:34:51, 5.80s/it] + Training...: 78% 3412/4393 [4:50:28<1:34:28, 5.78s/it] + Training...: 78% 3413/4393 [4:50:34<1:33:37, 5.73s/it] +  + Training...: 78% 3413/4393 [4:50:40<1:33:37, 5.73s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:20:12<57:16:35, 20619.59s/it] + Training...: 78% 3414/4393 [4:50:40<1:36:52, 5.94s/it] + Training...: 78% 3415/4393 [4:50:46<1:35:02, 5.83s/it] + Training...: 78% 3416/4393 [4:50:51<1:33:18, 5.73s/it] + Training...: 78% 3417/4393 [4:50:57<1:32:16, 5.67s/it] + Training...: 78% 3418/4393 [4:51:02<1:31:15, 5.62s/it] + Training...: 78% 3419/4393 [4:51:07<1:30:23, 5.57s/it] + Training...: 78% 3420/4393 [4:51:13<1:29:24, 5.51s/it] + Training...: 78% 3421/4393 [4:51:18<1:28:28, 5.46s/it] + Training...: 78% 3422/4393 [4:51:23<1:27:28, 5.41s/it] + Training...: 78% 3423/4393 [4:51:29<1:26:36, 5.36s/it] + Training...: 78% 3424/4393 [4:51:34<1:25:24, 5.29s/it] + Training...: 78% 3425/4393 [4:51:39<1:25:42, 5.31s/it] + Training...: 78% 3426/4393 [4:51:44<1:25:22, 5.30s/it] + Training...: 78% 3427/4393 [4:51:50<1:24:09, 5.23s/it] + Training...: 78% 3428/4393 [4:51:54<1:22:36, 5.14s/it] + Training...: 78% 3429/4393 [4:51:59<1:21:28, 5.07s/it] + Training...: 78% 3430/4393 [4:52:04<1:20:13, 5.00s/it] + Training...: 78% 3431/4393 [4:52:09<1:19:17, 4.95s/it] + Training...: 78% 3432/4393 [4:52:14<1:18:02, 4.87s/it] + Training...: 78% 3433/4393 [4:52:18<1:17:10, 4.82s/it] + Training...: 78% 3434/4393 [4:52:23<1:15:32, 4.73s/it] + Training...: 78% 3435/4393 [4:52:27<1:13:51, 4.63s/it] + Training...: 78% 3436/4393 [4:52:32<1:11:58, 4.51s/it] + Training...: 78% 3437/4393 [4:52:36<1:09:53, 4.39s/it] + Training...: 78% 3438/4393 [4:52:40<1:07:26, 4.24s/it] +  + Training...: 78% 3438/4393 [4:52:44<1:07:26, 4.24s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:22:16<57:16:35, 20619.59s/it] + Training...: 78% 3439/4393 [4:52:44<1:07:21, 4.24s/it] + Training...: 78% 3440/4393 [4:52:47<1:03:17, 3.99s/it] + Training...: 78% 3441/4393 [4:52:50<59:34, 3.75s/it]  + Training...: 78% 3442/4393 [4:52:53<55:41, 3.51s/it] + Training...: 78% 3443/4393 [4:52:56<51:56, 3.28s/it] + Training...: 78% 3444/4393 [4:52:59<48:00, 3.04s/it] + Training...: 78% 3445/4393 [4:53:01<43:57, 2.78s/it] + Training...: 78% 3446/4393 [4:53:03<40:20, 2.56s/it] + Training...: 78% 3447/4393 [4:53:05<36:18, 2.30s/it] + Training...: 78% 3448/4393 [4:53:06<32:24, 2.06s/it] + Training...: 79% 3449/4393 [4:53:07<28:35, 1.82s/it] + Training...: 79% 3450/4393 [4:53:08<24:37, 1.57s/it] + Training...: 79% 3451/4393 [4:53:15<47:04, 3.00s/it] + Training...: 79% 3452/4393 [4:53:21<1:02:16, 3.97s/it] + Training...: 79% 3453/4393 [4:53:27<1:11:38, 4.57s/it] + Training...: 79% 3454/4393 [4:53:33<1:18:10, 5.00s/it] + Training...: 79% 3455/4393 [4:53:39<1:22:26, 5.27s/it] + Training...: 79% 3456/4393 [4:53:45<1:25:00, 5.44s/it] + Training...: 79% 3457/4393 [4:53:50<1:26:43, 5.56s/it] + Training...: 79% 3458/4393 [4:53:56<1:27:31, 5.62s/it] + Training...: 79% 3459/4393 [4:54:02<1:28:11, 5.66s/it] + Training...: 79% 3460/4393 [4:54:08<1:28:07, 5.67s/it] + Training...: 79% 3461/4393 [4:54:13<1:28:20, 5.69s/it] + Training...: 79% 3462/4393 [4:54:19<1:28:06, 5.68s/it] + Training...: 79% 3463/4393 [4:54:25<1:28:07, 5.69s/it] +  + Training...: 79% 3463/4393 [4:54:31<1:28:07, 5.69s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:24:03<57:16:35, 20619.59s/it] + Training...: 79% 3464/4393 [4:54:31<1:31:23, 5.90s/it] + Training...: 79% 3465/4393 [4:54:37<1:29:41, 5.80s/it] + Training...: 79% 3466/4393 [4:54:42<1:28:16, 5.71s/it] + Training...: 79% 3467/4393 [4:54:48<1:27:11, 5.65s/it] + Training...: 79% 3468/4393 [4:54:53<1:26:06, 5.59s/it] + Training...: 79% 3469/4393 [4:54:59<1:25:42, 5.57s/it] + Training...: 79% 3470/4393 [4:55:04<1:25:19, 5.55s/it] + Training...: 79% 3471/4393 [4:55:10<1:24:42, 5.51s/it] + Training...: 79% 3472/4393 [4:55:15<1:23:53, 5.47s/it] + Training...: 79% 3473/4393 [4:55:20<1:22:52, 5.40s/it] + Training...: 79% 3474/4393 [4:55:25<1:21:33, 5.32s/it] + Training...: 79% 3475/4393 [4:55:30<1:20:41, 5.27s/it] + Training...: 79% 3476/4393 [4:55:36<1:21:34, 5.34s/it] + Training...: 79% 3477/4393 [4:55:41<1:20:28, 5.27s/it] + Training...: 79% 3478/4393 [4:55:46<1:19:05, 5.19s/it] + Training...: 79% 3479/4393 [4:55:51<1:18:03, 5.12s/it] + Training...: 79% 3480/4393 [4:55:56<1:17:39, 5.10s/it] + Training...: 79% 3481/4393 [4:56:01<1:16:22, 5.02s/it] + Training...: 79% 3482/4393 [4:56:06<1:15:29, 4.97s/it] + Training...: 79% 3483/4393 [4:56:10<1:14:16, 4.90s/it] + Training...: 79% 3484/4393 [4:56:15<1:12:49, 4.81s/it] + Training...: 79% 3485/4393 [4:56:20<1:11:03, 4.70s/it] + Training...: 79% 3486/4393 [4:56:24<1:08:57, 4.56s/it] + Training...: 79% 3487/4393 [4:56:28<1:06:45, 4.42s/it] + Training...: 79% 3488/4393 [4:56:32<1:04:32, 4.28s/it] +  + Training...: 79% 3488/4393 [4:56:36<1:04:32, 4.28s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:26:08<57:16:35, 20619.59s/it] + Training...: 79% 3489/4393 [4:56:36<1:04:25, 4.28s/it] + Training...: 79% 3490/4393 [4:56:40<1:00:40, 4.03s/it] + Training...: 79% 3491/4393 [4:56:43<57:09, 3.80s/it]  + Training...: 79% 3492/4393 [4:56:46<53:31, 3.56s/it] + Training...: 80% 3493/4393 [4:56:49<50:02, 3.34s/it] + Training...: 80% 3494/4393 [4:56:51<46:08, 3.08s/it] + Training...: 80% 3495/4393 [4:56:53<42:09, 2.82s/it] + Training...: 80% 3496/4393 [4:56:55<38:07, 2.55s/it] + Training...: 80% 3497/4393 [4:56:57<34:19, 2.30s/it] + Training...: 80% 3498/4393 [4:56:58<30:46, 2.06s/it] + Training...: 80% 3499/4393 [4:57:00<26:59, 1.81s/it] + Training...: 80% 3500/4393 [4:57:01<23:06, 1.55s/it] + Training...: 80% 3501/4393 [4:57:07<43:50, 2.95s/it] + Training...: 80% 3502/4393 [4:57:13<57:51, 3.90s/it] + Training...: 80% 3503/4393 [4:57:19<1:07:17, 4.54s/it] + Training...: 80% 3504/4393 [4:57:25<1:13:26, 4.96s/it] + Training...: 80% 3505/4393 [4:57:31<1:17:37, 5.25s/it] + Training...: 80% 3506/4393 [4:57:37<1:20:40, 5.46s/it] + Training...: 80% 3507/4393 [4:57:43<1:22:14, 5.57s/it] + Training...: 80% 3508/4393 [4:57:48<1:22:55, 5.62s/it] + Training...: 80% 3509/4393 [4:57:54<1:23:47, 5.69s/it] + Training...: 80% 3510/4393 [4:58:00<1:24:28, 5.74s/it] + Training...: 80% 3511/4393 [4:58:06<1:24:47, 5.77s/it] + Training...: 80% 3512/4393 [4:58:12<1:24:48, 5.78s/it] + Training...: 80% 3513/4393 [4:58:17<1:24:21, 5.75s/it] +  + Training...: 80% 3513/4393 [4:58:24<1:24:21, 5.75s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:27:56<57:16:35, 20619.59s/it] + Training...: 80% 3514/4393 [4:58:24<1:27:17, 5.96s/it] + Training...: 80% 3515/4393 [4:58:29<1:26:00, 5.88s/it] + Training...: 80% 3516/4393 [4:58:35<1:24:37, 5.79s/it] + Training...: 80% 3517/4393 [4:58:41<1:23:26, 5.71s/it] + Training...: 80% 3518/4393 [4:58:46<1:22:44, 5.67s/it] + Training...: 80% 3519/4393 [4:58:52<1:21:43, 5.61s/it] + Training...: 80% 3520/4393 [4:58:57<1:20:54, 5.56s/it] + Training...: 80% 3521/4393 [4:59:03<1:20:15, 5.52s/it] + Training...: 80% 3522/4393 [4:59:08<1:19:35, 5.48s/it] + Training...: 80% 3523/4393 [4:59:13<1:18:48, 5.44s/it] + Training...: 80% 3524/4393 [4:59:18<1:17:46, 5.37s/it] + Training...: 80% 3525/4393 [4:59:24<1:16:56, 5.32s/it] + Training...: 80% 3526/4393 [4:59:29<1:16:12, 5.27s/it] + Training...: 80% 3527/4393 [4:59:34<1:15:21, 5.22s/it] + Training...: 80% 3528/4393 [4:59:39<1:14:52, 5.19s/it] + Training...: 80% 3529/4393 [4:59:44<1:14:53, 5.20s/it] + Training...: 80% 3530/4393 [4:59:49<1:13:52, 5.14s/it] + Training...: 80% 3531/4393 [4:59:54<1:12:23, 5.04s/it] + Training...: 80% 3532/4393 [4:59:59<1:10:53, 4.94s/it] + Training...: 80% 3533/4393 [5:00:03<1:09:30, 4.85s/it] + Training...: 80% 3534/4393 [5:00:08<1:08:25, 4.78s/it] + Training...: 80% 3535/4393 [5:00:12<1:06:47, 4.67s/it] + Training...: 80% 3536/4393 [5:00:17<1:05:23, 4.58s/it] + Training...: 81% 3537/4393 [5:00:21<1:03:41, 4.46s/it] + Training...: 81% 3538/4393 [5:00:25<1:01:32, 4.32s/it] +  + Training...: 81% 3538/4393 [5:00:29<1:01:32, 4.32s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:30:02<57:16:35, 20619.59s/it] + Training...: 81% 3539/4393 [5:00:29<1:01:28, 4.32s/it] + Training...: 81% 3540/4393 [5:00:33<58:13, 4.10s/it]  + Training...: 81% 3541/4393 [5:00:36<55:05, 3.88s/it] + Training...: 81% 3542/4393 [5:00:39<51:51, 3.66s/it] + Training...: 81% 3543/4393 [5:00:42<48:12, 3.40s/it] + Training...: 81% 3544/4393 [5:00:45<44:45, 3.16s/it] + Training...: 81% 3545/4393 [5:00:47<41:12, 2.92s/it] + Training...: 81% 3546/4393 [5:00:49<37:34, 2.66s/it] + Training...: 81% 3547/4393 [5:00:51<33:48, 2.40s/it] + Training...: 81% 3548/4393 [5:00:53<29:55, 2.13s/it] + Training...: 81% 3549/4393 [5:00:54<26:12, 1.86s/it] + Training...: 81% 3550/4393 [5:00:55<22:32, 1.60s/it] + Training...: 81% 3551/4393 [5:01:01<43:52, 3.13s/it] + Training...: 81% 3552/4393 [5:01:08<58:04, 4.14s/it] + Training...: 81% 3553/4393 [5:01:14<1:06:05, 4.72s/it] + Training...: 81% 3554/4393 [5:01:20<1:11:18, 5.10s/it] + Training...: 81% 3555/4393 [5:01:26<1:15:03, 5.37s/it] + Training...: 81% 3556/4393 [5:01:32<1:16:58, 5.52s/it] + Training...: 81% 3557/4393 [5:01:38<1:18:12, 5.61s/it] + Training...: 81% 3558/4393 [5:01:44<1:19:10, 5.69s/it] + Training...: 81% 3559/4393 [5:01:49<1:19:30, 5.72s/it] + Training...: 81% 3560/4393 [5:01:55<1:19:31, 5.73s/it] + Training...: 81% 3561/4393 [5:02:01<1:19:33, 5.74s/it] + Training...: 81% 3562/4393 [5:02:07<1:19:10, 5.72s/it] + Training...: 81% 3563/4393 [5:02:12<1:18:48, 5.70s/it] +  + Training...: 81% 3563/4393 [5:02:19<1:18:48, 5.70s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:31:51<57:16:35, 20619.59s/it] + Training...: 81% 3564/4393 [5:02:19<1:21:40, 5.91s/it] + Training...: 81% 3565/4393 [5:02:24<1:20:17, 5.82s/it] + Training...: 81% 3566/4393 [5:02:30<1:20:06, 5.81s/it] + Training...: 81% 3567/4393 [5:02:36<1:19:29, 5.77s/it] + Training...: 81% 3568/4393 [5:02:41<1:18:16, 5.69s/it] + Training...: 81% 3569/4393 [5:02:47<1:17:32, 5.65s/it] + Training...: 81% 3570/4393 [5:02:52<1:16:37, 5.59s/it] + Training...: 81% 3571/4393 [5:02:58<1:15:41, 5.52s/it] + Training...: 81% 3572/4393 [5:03:03<1:14:37, 5.45s/it] + Training...: 81% 3573/4393 [5:03:08<1:13:59, 5.41s/it] + Training...: 81% 3574/4393 [5:03:13<1:13:07, 5.36s/it] + Training...: 81% 3575/4393 [5:03:19<1:12:39, 5.33s/it] + Training...: 81% 3576/4393 [5:03:24<1:11:59, 5.29s/it] + Training...: 81% 3577/4393 [5:03:29<1:11:03, 5.22s/it] + Training...: 81% 3578/4393 [5:03:34<1:10:25, 5.18s/it] + Training...: 81% 3579/4393 [5:03:39<1:09:34, 5.13s/it] + Training...: 81% 3580/4393 [5:03:44<1:08:26, 5.05s/it] + Training...: 82% 3581/4393 [5:03:49<1:07:58, 5.02s/it] + Training...: 82% 3582/4393 [5:03:54<1:06:53, 4.95s/it] + Training...: 82% 3583/4393 [5:03:58<1:05:55, 4.88s/it] + Training...: 82% 3584/4393 [5:04:03<1:04:36, 4.79s/it] + Training...: 82% 3585/4393 [5:04:07<1:03:05, 4.68s/it] + Training...: 82% 3586/4393 [5:04:12<1:01:30, 4.57s/it] + Training...: 82% 3587/4393 [5:04:16<59:42, 4.44s/it]  + Training...: 82% 3588/4393 [5:04:20<57:57, 4.32s/it] +  + Training...: 82% 3588/4393 [5:04:24<57:57, 4.32s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:33:57<57:16:35, 20619.59s/it] + Training...: 82% 3589/4393 [5:04:24<58:14, 4.35s/it] + Training...: 82% 3590/4393 [5:04:28<55:02, 4.11s/it] + Training...: 82% 3591/4393 [5:04:31<51:51, 3.88s/it] + Training...: 82% 3592/4393 [5:04:34<48:43, 3.65s/it] + Training...: 82% 3593/4393 [5:04:37<45:30, 3.41s/it] + Training...: 82% 3594/4393 [5:04:40<42:19, 3.18s/it] + Training...: 82% 3595/4393 [5:04:42<39:15, 2.95s/it] + Training...: 82% 3596/4393 [5:04:44<35:45, 2.69s/it] + Training...: 82% 3597/4393 [5:04:46<32:04, 2.42s/it] + Training...: 82% 3598/4393 [5:04:48<28:26, 2.15s/it] + Training...: 82% 3599/4393 [5:04:49<24:45, 1.87s/it] + Training...: 82% 3600/4393 [5:04:50<21:10, 1.60s/it] + Training...: 82% 3601/4393 [5:04:56<39:24, 2.98s/it] + Training...: 82% 3602/4393 [5:05:02<51:51, 3.93s/it] + Training...: 82% 3603/4393 [5:05:08<59:55, 4.55s/it] + Training...: 82% 3604/4393 [5:05:14<1:05:32, 4.98s/it] + Training...: 82% 3605/4393 [5:05:20<1:09:13, 5.27s/it] + Training...: 82% 3606/4393 [5:05:26<1:11:19, 5.44s/it] + Training...: 82% 3607/4393 [5:05:32<1:12:37, 5.54s/it] + Training...: 82% 3608/4393 [5:05:37<1:13:30, 5.62s/it] + Training...: 82% 3609/4393 [5:05:43<1:14:07, 5.67s/it] + Training...: 82% 3610/4393 [5:05:49<1:14:27, 5.71s/it] + Training...: 82% 3611/4393 [5:05:55<1:14:22, 5.71s/it] + Training...: 82% 3612/4393 [5:06:00<1:14:10, 5.70s/it] + Training...: 82% 3613/4393 [5:06:06<1:13:51, 5.68s/it] +  + Training...: 82% 3613/4393 [5:06:13<1:13:51, 5.68s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:35:45<57:16:35, 20619.59s/it] + Training...: 82% 3614/4393 [5:06:13<1:16:43, 5.91s/it] + Training...: 82% 3615/4393 [5:06:18<1:15:21, 5.81s/it] + Training...: 82% 3616/4393 [5:06:24<1:14:27, 5.75s/it] + Training...: 82% 3617/4393 [5:06:29<1:13:35, 5.69s/it] + Training...: 82% 3618/4393 [5:06:35<1:13:23, 5.68s/it] + Training...: 82% 3619/4393 [5:06:40<1:12:56, 5.65s/it] + Training...: 82% 3620/4393 [5:06:46<1:11:48, 5.57s/it] + Training...: 82% 3621/4393 [5:06:51<1:11:04, 5.52s/it] + Training...: 82% 3622/4393 [5:06:57<1:10:07, 5.46s/it] + Training...: 82% 3623/4393 [5:07:02<1:09:23, 5.41s/it] + Training...: 82% 3624/4393 [5:07:07<1:08:35, 5.35s/it] + Training...: 83% 3625/4393 [5:07:12<1:08:09, 5.32s/it] + Training...: 83% 3626/4393 [5:07:17<1:07:11, 5.26s/it] + Training...: 83% 3627/4393 [5:07:23<1:06:39, 5.22s/it] + Training...: 83% 3628/4393 [5:07:28<1:05:49, 5.16s/it] + Training...: 83% 3629/4393 [5:07:33<1:04:56, 5.10s/it] + Training...: 83% 3630/4393 [5:07:38<1:04:12, 5.05s/it] + Training...: 83% 3631/4393 [5:07:42<1:03:29, 5.00s/it] + Training...: 83% 3632/4393 [5:07:47<1:02:40, 4.94s/it] + Training...: 83% 3633/4393 [5:07:52<1:01:31, 4.86s/it] + Training...: 83% 3634/4393 [5:07:56<1:00:25, 4.78s/it] + Training...: 83% 3635/4393 [5:08:01<59:15, 4.69s/it]  + Training...: 83% 3636/4393 [5:08:05<57:58, 4.59s/it] + Training...: 83% 3637/4393 [5:08:10<56:30, 4.48s/it] + Training...: 83% 3638/4393 [5:08:14<54:48, 4.36s/it] +  + Training...: 83% 3638/4393 [5:08:18<54:48, 4.36s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:37:50<57:16:35, 20619.59s/it] + Training...: 83% 3639/4393 [5:08:18<54:55, 4.37s/it] + Training...: 83% 3640/4393 [5:08:22<51:50, 4.13s/it] + Training...: 83% 3641/4393 [5:08:25<48:41, 3.88s/it] + Training...: 83% 3642/4393 [5:08:28<45:45, 3.66s/it] + Training...: 83% 3643/4393 [5:08:31<43:03, 3.44s/it] + Training...: 83% 3644/4393 [5:08:34<40:18, 3.23s/it] + Training...: 83% 3645/4393 [5:08:36<37:17, 2.99s/it] + Training...: 83% 3646/4393 [5:08:38<34:11, 2.75s/it] + Training...: 83% 3647/4393 [5:08:40<30:48, 2.48s/it] + Training...: 83% 3648/4393 [5:08:42<27:13, 2.19s/it] + Training...: 83% 3649/4393 [5:08:43<23:36, 1.90s/it] + Training...: 83% 3650/4393 [5:08:44<20:06, 1.62s/it] + Training...: 83% 3651/4393 [5:08:50<37:30, 3.03s/it] + Training...: 83% 3652/4393 [5:08:56<49:35, 4.01s/it] + Training...: 83% 3653/4393 [5:09:02<56:44, 4.60s/it] + Training...: 83% 3654/4393 [5:09:08<1:01:36, 5.00s/it] + Training...: 83% 3655/4393 [5:09:14<1:04:56, 5.28s/it] + Training...: 83% 3656/4393 [5:09:20<1:06:52, 5.44s/it] + Training...: 83% 3657/4393 [5:09:26<1:08:23, 5.57s/it] + Training...: 83% 3658/4393 [5:09:32<1:09:13, 5.65s/it] + Training...: 83% 3659/4393 [5:09:38<1:09:50, 5.71s/it] + Training...: 83% 3660/4393 [5:09:43<1:09:42, 5.71s/it] + Training...: 83% 3661/4393 [5:09:49<1:09:41, 5.71s/it] + Training...: 83% 3662/4393 [5:09:55<1:09:39, 5.72s/it] + Training...: 83% 3663/4393 [5:10:01<1:09:45, 5.73s/it] +  + Training...: 83% 3663/4393 [5:10:07<1:09:45, 5.73s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:39:39<57:16:35, 20619.59s/it] + Training...: 83% 3664/4393 [5:10:07<1:12:16, 5.95s/it] + Training...: 83% 3665/4393 [5:10:13<1:10:51, 5.84s/it] + Training...: 83% 3666/4393 [5:10:18<1:09:56, 5.77s/it] + Training...: 83% 3667/4393 [5:10:24<1:09:16, 5.73s/it] + Training...: 83% 3668/4393 [5:10:29<1:08:16, 5.65s/it] + Training...: 84% 3669/4393 [5:10:35<1:07:39, 5.61s/it] + Training...: 84% 3670/4393 [5:10:40<1:07:01, 5.56s/it] + Training...: 84% 3671/4393 [5:10:46<1:06:23, 5.52s/it] + Training...: 84% 3672/4393 [5:10:51<1:05:38, 5.46s/it] + Training...: 84% 3673/4393 [5:10:56<1:04:58, 5.42s/it] + Training...: 84% 3674/4393 [5:11:02<1:04:12, 5.36s/it] + Training...: 84% 3675/4393 [5:11:07<1:03:33, 5.31s/it] + Training...: 84% 3676/4393 [5:11:12<1:03:16, 5.29s/it] + Training...: 84% 3677/4393 [5:11:17<1:02:56, 5.27s/it] + Training...: 84% 3678/4393 [5:11:22<1:02:12, 5.22s/it] + Training...: 84% 3679/4393 [5:11:27<1:01:14, 5.15s/it] + Training...: 84% 3680/4393 [5:11:32<1:00:09, 5.06s/it] + Training...: 84% 3681/4393 [5:11:37<59:07, 4.98s/it]  + Training...: 84% 3682/4393 [5:11:42<58:07, 4.91s/it] + Training...: 84% 3683/4393 [5:11:46<57:10, 4.83s/it] + Training...: 84% 3684/4393 [5:11:51<56:18, 4.77s/it] + Training...: 84% 3685/4393 [5:11:56<55:18, 4.69s/it] + Training...: 84% 3686/4393 [5:12:00<53:50, 4.57s/it] + Training...: 84% 3687/4393 [5:12:04<52:21, 4.45s/it] + Training...: 84% 3688/4393 [5:12:08<50:35, 4.31s/it] +  + Training...: 84% 3688/4393 [5:12:12<50:35, 4.31s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:41:45<57:16:35, 20619.59s/it] + Training...: 84% 3689/4393 [5:12:12<50:39, 4.32s/it] + Training...: 84% 3690/4393 [5:12:16<47:39, 4.07s/it] + Training...: 84% 3691/4393 [5:12:19<44:54, 3.84s/it] + Training...: 84% 3692/4393 [5:12:22<41:58, 3.59s/it] + Training...: 84% 3693/4393 [5:12:25<39:07, 3.35s/it] + Training...: 84% 3694/4393 [5:12:28<36:22, 3.12s/it] + Training...: 84% 3695/4393 [5:12:30<33:16, 2.86s/it] + Training...: 84% 3696/4393 [5:12:32<30:19, 2.61s/it] + Training...: 84% 3697/4393 [5:12:34<27:18, 2.35s/it] + Training...: 84% 3698/4393 [5:12:35<24:24, 2.11s/it] + Training...: 84% 3699/4393 [5:12:36<21:27, 1.86s/it] + Training...: 84% 3700/4393 [5:12:37<18:26, 1.60s/it] + Training...: 84% 3701/4393 [5:12:43<34:00, 2.95s/it] + Training...: 84% 3702/4393 [5:12:50<45:14, 3.93s/it] + Training...: 84% 3703/4393 [5:12:56<52:25, 4.56s/it] + Training...: 84% 3704/4393 [5:13:02<56:57, 4.96s/it] + Training...: 84% 3705/4393 [5:13:08<1:00:26, 5.27s/it] + Training...: 84% 3706/4393 [5:13:13<1:02:29, 5.46s/it] + Training...: 84% 3707/4393 [5:13:19<1:03:51, 5.58s/it] + Training...: 84% 3708/4393 [5:13:25<1:04:27, 5.65s/it] + Training...: 84% 3709/4393 [5:13:31<1:04:48, 5.68s/it] + Training...: 84% 3710/4393 [5:13:37<1:04:41, 5.68s/it] + Training...: 84% 3711/4393 [5:13:42<1:04:51, 5.71s/it] + Training...: 84% 3712/4393 [5:13:48<1:04:43, 5.70s/it] + Training...: 85% 3713/4393 [5:13:54<1:04:34, 5.70s/it] +  + Training...: 85% 3713/4393 [5:14:00<1:04:34, 5.70s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:43:33<57:16:35, 20619.59s/it] + Training...: 85% 3714/4393 [5:14:00<1:07:10, 5.94s/it] + Training...: 85% 3715/4393 [5:14:06<1:06:12, 5.86s/it] + Training...: 85% 3716/4393 [5:14:12<1:05:37, 5.82s/it] + Training...: 85% 3717/4393 [5:14:17<1:05:09, 5.78s/it] + Training...: 85% 3718/4393 [5:14:23<1:04:12, 5.71s/it] + Training...: 85% 3719/4393 [5:14:28<1:03:18, 5.64s/it] + Training...: 85% 3720/4393 [5:14:34<1:03:01, 5.62s/it] + Training...: 85% 3721/4393 [5:14:39<1:02:05, 5.54s/it] + Training...: 85% 3722/4393 [5:14:45<1:01:08, 5.47s/it] + Training...: 85% 3723/4393 [5:14:50<1:00:16, 5.40s/it] + Training...: 85% 3724/4393 [5:14:55<59:43, 5.36s/it]  + Training...: 85% 3725/4393 [5:15:00<59:05, 5.31s/it] + Training...: 85% 3726/4393 [5:15:05<58:27, 5.26s/it] + Training...: 85% 3727/4393 [5:15:10<57:44, 5.20s/it] + Training...: 85% 3728/4393 [5:15:15<57:04, 5.15s/it] + Training...: 85% 3729/4393 [5:15:21<56:36, 5.12s/it] + Training...: 85% 3730/4393 [5:15:25<55:48, 5.05s/it] + Training...: 85% 3731/4393 [5:15:30<55:03, 4.99s/it] + Training...: 85% 3732/4393 [5:15:35<54:13, 4.92s/it] + Training...: 85% 3733/4393 [5:15:40<53:30, 4.86s/it] + Training...: 85% 3734/4393 [5:15:44<52:52, 4.81s/it] + Training...: 85% 3735/4393 [5:15:49<52:40, 4.80s/it] + Training...: 85% 3736/4393 [5:15:54<51:55, 4.74s/it] + Training...: 85% 3737/4393 [5:15:58<50:21, 4.61s/it] + Training...: 85% 3738/4393 [5:16:02<48:28, 4.44s/it] +  + Training...: 85% 3738/4393 [5:16:07<48:28, 4.44s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:45:39<57:16:35, 20619.59s/it] + Training...: 85% 3739/4393 [5:16:07<48:39, 4.46s/it] + Training...: 85% 3740/4393 [5:16:10<45:53, 4.22s/it] + Training...: 85% 3741/4393 [5:16:14<43:05, 3.97s/it] + Training...: 85% 3742/4393 [5:16:17<40:15, 3.71s/it] + Training...: 85% 3743/4393 [5:16:20<37:25, 3.45s/it] + Training...: 85% 3744/4393 [5:16:22<34:45, 3.21s/it] + Training...: 85% 3745/4393 [5:16:25<31:47, 2.94s/it] + Training...: 85% 3746/4393 [5:16:27<28:53, 2.68s/it] + Training...: 85% 3747/4393 [5:16:29<26:02, 2.42s/it] + Training...: 85% 3748/4393 [5:16:30<23:13, 2.16s/it] + Training...: 85% 3749/4393 [5:16:31<20:26, 1.91s/it] + Training...: 85% 3750/4393 [5:16:32<17:40, 1.65s/it] + Training...: 85% 3751/4393 [5:16:39<32:44, 3.06s/it] + Training...: 85% 3752/4393 [5:16:45<42:53, 4.01s/it] + Training...: 85% 3753/4393 [5:16:51<49:24, 4.63s/it] + Training...: 85% 3754/4393 [5:16:57<53:27, 5.02s/it] + Training...: 85% 3755/4393 [5:17:03<57:08, 5.37s/it] + Training...: 85% 3756/4393 [5:17:09<58:50, 5.54s/it] + Training...: 86% 3757/4393 [5:17:15<59:51, 5.65s/it] + Training...: 86% 3758/4393 [5:17:21<1:00:11, 5.69s/it] + Training...: 86% 3759/4393 [5:17:27<1:00:17, 5.71s/it] + Training...: 86% 3760/4393 [5:17:32<1:00:17, 5.72s/it] + Training...: 86% 3761/4393 [5:17:38<1:00:07, 5.71s/it] + Training...: 86% 3762/4393 [5:17:44<59:58, 5.70s/it]  + Training...: 86% 3763/4393 [5:17:49<59:57, 5.71s/it] +  + Training...: 86% 3763/4393 [5:17:56<59:57, 5.71s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:47:28<57:16:35, 20619.59s/it] + Training...: 86% 3764/4393 [5:17:56<1:02:11, 5.93s/it] + Training...: 86% 3765/4393 [5:18:01<1:01:00, 5.83s/it] + Training...: 86% 3766/4393 [5:18:07<59:49, 5.73s/it]  + Training...: 86% 3767/4393 [5:18:12<59:05, 5.66s/it] + Training...: 86% 3768/4393 [5:18:18<58:33, 5.62s/it] + Training...: 86% 3769/4393 [5:18:24<58:29, 5.62s/it] + Training...: 86% 3770/4393 [5:18:29<58:39, 5.65s/it] + Training...: 86% 3771/4393 [5:18:35<59:05, 5.70s/it] + Training...: 86% 3772/4393 [5:18:41<57:56, 5.60s/it] + Training...: 86% 3773/4393 [5:18:46<56:50, 5.50s/it] + Training...: 86% 3774/4393 [5:18:51<55:48, 5.41s/it] + Training...: 86% 3775/4393 [5:18:56<54:57, 5.34s/it] + Training...: 86% 3776/4393 [5:19:01<54:10, 5.27s/it] + Training...: 86% 3777/4393 [5:19:06<53:35, 5.22s/it] + Training...: 86% 3778/4393 [5:19:11<52:55, 5.16s/it] + Training...: 86% 3779/4393 [5:19:16<52:15, 5.11s/it] + Training...: 86% 3780/4393 [5:19:21<51:18, 5.02s/it] + Training...: 86% 3781/4393 [5:19:26<50:46, 4.98s/it] + Training...: 86% 3782/4393 [5:19:31<49:53, 4.90s/it] + Training...: 86% 3783/4393 [5:19:35<48:58, 4.82s/it] + Training...: 86% 3784/4393 [5:19:40<48:09, 4.75s/it] + Training...: 86% 3785/4393 [5:19:44<47:09, 4.65s/it] + Training...: 86% 3786/4393 [5:19:49<46:00, 4.55s/it] + Training...: 86% 3787/4393 [5:19:53<44:52, 4.44s/it] + Training...: 86% 3788/4393 [5:19:57<43:15, 4.29s/it] +  + Training...: 86% 3788/4393 [5:20:01<43:15, 4.29s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:49:33<57:16:35, 20619.59s/it] + Training...: 86% 3789/4393 [5:20:01<43:04, 4.28s/it] + Training...: 86% 3790/4393 [5:20:05<40:27, 4.03s/it] + Training...: 86% 3791/4393 [5:20:08<38:02, 3.79s/it] + Training...: 86% 3792/4393 [5:20:11<35:33, 3.55s/it] + Training...: 86% 3793/4393 [5:20:14<33:10, 3.32s/it] + Training...: 86% 3794/4393 [5:20:16<30:38, 3.07s/it] + Training...: 86% 3795/4393 [5:20:18<28:12, 2.83s/it] + Training...: 86% 3796/4393 [5:20:20<25:39, 2.58s/it] + Training...: 86% 3797/4393 [5:20:22<22:59, 2.31s/it] + Training...: 86% 3798/4393 [5:20:23<20:20, 2.05s/it] + Training...: 86% 3799/4393 [5:20:25<17:48, 1.80s/it] + Training...: 87% 3800/4393 [5:20:26<15:19, 1.55s/it] + Training...: 87% 3801/4393 [5:20:32<29:10, 2.96s/it] + Training...: 87% 3802/4393 [5:20:38<38:39, 3.92s/it] + Training...: 87% 3803/4393 [5:20:44<45:07, 4.59s/it] + Training...: 87% 3804/4393 [5:20:50<49:33, 5.05s/it] + Training...: 87% 3805/4393 [5:20:56<52:01, 5.31s/it] + Training...: 87% 3806/4393 [5:21:02<53:35, 5.48s/it] + Training...: 87% 3807/4393 [5:21:08<54:43, 5.60s/it] + Training...: 87% 3808/4393 [5:21:14<55:09, 5.66s/it] + Training...: 87% 3809/4393 [5:21:20<55:35, 5.71s/it] + Training...: 87% 3810/4393 [5:21:25<55:28, 5.71s/it] + Training...: 87% 3811/4393 [5:21:31<55:30, 5.72s/it] + Training...: 87% 3812/4393 [5:21:37<55:12, 5.70s/it] + Training...: 87% 3813/4393 [5:21:42<54:52, 5.68s/it] +  + Training...: 87% 3813/4393 [5:21:49<54:52, 5.68s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:51:21<57:16:35, 20619.59s/it] + Training...: 87% 3814/4393 [5:21:49<56:58, 5.90s/it] + Training...: 87% 3815/4393 [5:21:54<56:02, 5.82s/it] + Training...: 87% 3816/4393 [5:22:00<55:01, 5.72s/it] + Training...: 87% 3817/4393 [5:22:05<54:23, 5.67s/it] + Training...: 87% 3818/4393 [5:22:11<53:37, 5.60s/it] + Training...: 87% 3819/4393 [5:22:16<53:06, 5.55s/it] + Training...: 87% 3820/4393 [5:22:22<52:24, 5.49s/it] + Training...: 87% 3821/4393 [5:22:27<51:47, 5.43s/it] + Training...: 87% 3822/4393 [5:22:32<51:21, 5.40s/it] + Training...: 87% 3823/4393 [5:22:38<51:06, 5.38s/it] + Training...: 87% 3824/4393 [5:22:43<50:39, 5.34s/it] + Training...: 87% 3825/4393 [5:22:48<50:30, 5.34s/it] + Training...: 87% 3826/4393 [5:22:53<50:07, 5.30s/it] + Training...: 87% 3827/4393 [5:22:59<49:37, 5.26s/it] + Training...: 87% 3828/4393 [5:23:04<48:51, 5.19s/it] + Training...: 87% 3829/4393 [5:23:09<48:04, 5.12s/it] + Training...: 87% 3830/4393 [5:23:13<47:16, 5.04s/it] + Training...: 87% 3831/4393 [5:23:18<46:26, 4.96s/it] + Training...: 87% 3832/4393 [5:23:23<45:45, 4.89s/it] + Training...: 87% 3833/4393 [5:23:28<45:05, 4.83s/it] + Training...: 87% 3834/4393 [5:23:32<44:06, 4.73s/it] + Training...: 87% 3835/4393 [5:23:37<43:18, 4.66s/it] + Training...: 87% 3836/4393 [5:23:41<42:18, 4.56s/it] + Training...: 87% 3837/4393 [5:23:45<41:00, 4.42s/it] + Training...: 87% 3838/4393 [5:23:49<39:36, 4.28s/it] +  + Training...: 87% 3838/4393 [5:23:53<39:36, 4.28s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:53:26<57:16:35, 20619.59s/it] + Training...: 87% 3839/4393 [5:23:53<39:53, 4.32s/it] + Training...: 87% 3840/4393 [5:23:57<37:40, 4.09s/it] + Training...: 87% 3841/4393 [5:24:00<35:26, 3.85s/it] + Training...: 87% 3842/4393 [5:24:03<33:15, 3.62s/it] + Training...: 87% 3843/4393 [5:24:06<31:10, 3.40s/it] + Training...: 88% 3844/4393 [5:24:09<28:34, 3.12s/it] + Training...: 88% 3845/4393 [5:24:11<26:04, 2.86s/it] + Training...: 88% 3846/4393 [5:24:13<23:42, 2.60s/it] + Training...: 88% 3847/4393 [5:24:15<21:15, 2.34s/it] + Training...: 88% 3848/4393 [5:24:16<18:54, 2.08s/it] + Training...: 88% 3849/4393 [5:24:17<16:31, 1.82s/it] + Training...: 88% 3850/4393 [5:24:18<14:13, 1.57s/it] + Training...: 88% 3851/4393 [5:24:25<26:40, 2.95s/it] + Training...: 88% 3852/4393 [5:24:31<35:17, 3.91s/it] + Training...: 88% 3853/4393 [5:24:37<40:59, 4.55s/it] + Training...: 88% 3854/4393 [5:24:43<44:47, 4.99s/it] + Training...: 88% 3855/4393 [5:24:49<47:42, 5.32s/it] + Training...: 88% 3856/4393 [5:24:55<49:09, 5.49s/it] + Training...: 88% 3857/4393 [5:25:01<49:58, 5.59s/it] + Training...: 88% 3858/4393 [5:25:06<50:15, 5.64s/it] + Training...: 88% 3859/4393 [5:25:12<50:43, 5.70s/it] + Training...: 88% 3860/4393 [5:25:18<50:46, 5.72s/it] + Training...: 88% 3861/4393 [5:25:24<50:46, 5.73s/it] + Training...: 88% 3862/4393 [5:25:29<50:30, 5.71s/it] + Training...: 88% 3863/4393 [5:25:35<50:14, 5.69s/it] +  + Training...: 88% 3863/4393 [5:25:41<50:14, 5.69s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:55:14<57:16:35, 20619.59s/it] + Training...: 88% 3864/4393 [5:25:41<52:14, 5.93s/it] + Training...: 88% 3865/4393 [5:25:47<51:21, 5.84s/it] + Training...: 88% 3866/4393 [5:25:53<50:29, 5.75s/it] + Training...: 88% 3867/4393 [5:25:58<49:57, 5.70s/it] + Training...: 88% 3868/4393 [5:26:04<49:11, 5.62s/it] + Training...: 88% 3869/4393 [5:26:09<48:48, 5.59s/it] + Training...: 88% 3870/4393 [5:26:15<48:13, 5.53s/it] + Training...: 88% 3871/4393 [5:26:20<48:31, 5.58s/it] + Training...: 88% 3872/4393 [5:26:26<47:54, 5.52s/it] + Training...: 88% 3873/4393 [5:26:31<47:11, 5.45s/it] + Training...: 88% 3874/4393 [5:26:36<46:27, 5.37s/it] + Training...: 88% 3875/4393 [5:26:41<45:47, 5.30s/it] + Training...: 88% 3876/4393 [5:26:46<45:18, 5.26s/it] + Training...: 88% 3877/4393 [5:26:51<44:49, 5.21s/it] + Training...: 88% 3878/4393 [5:26:56<44:15, 5.16s/it] + Training...: 88% 3879/4393 [5:27:02<43:59, 5.14s/it] + Training...: 88% 3880/4393 [5:27:06<43:16, 5.06s/it] + Training...: 88% 3881/4393 [5:27:11<42:37, 5.00s/it] + Training...: 88% 3882/4393 [5:27:16<41:49, 4.91s/it] + Training...: 88% 3883/4393 [5:27:21<41:09, 4.84s/it] + Training...: 88% 3884/4393 [5:27:25<40:18, 4.75s/it] + Training...: 88% 3885/4393 [5:27:30<39:28, 4.66s/it] + Training...: 88% 3886/4393 [5:27:34<38:31, 4.56s/it] + Training...: 88% 3887/4393 [5:27:38<37:23, 4.43s/it] + Training...: 89% 3888/4393 [5:27:42<36:20, 4.32s/it] +  + Training...: 89% 3888/4393 [5:27:47<36:20, 4.32s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:57:19<57:16:35, 20619.59s/it] + Training...: 89% 3889/4393 [5:27:47<36:35, 4.36s/it] + Training...: 89% 3890/4393 [5:27:50<34:31, 4.12s/it] + Training...: 89% 3891/4393 [5:27:54<32:43, 3.91s/it] + Training...: 89% 3892/4393 [5:27:57<30:35, 3.66s/it] + Training...: 89% 3893/4393 [5:28:00<28:27, 3.41s/it] + Training...: 89% 3894/4393 [5:28:02<26:25, 3.18s/it] + Training...: 89% 3895/4393 [5:28:04<24:10, 2.91s/it] + Training...: 89% 3896/4393 [5:28:06<21:55, 2.65s/it] + Training...: 89% 3897/4393 [5:28:08<19:31, 2.36s/it] + Training...: 89% 3898/4393 [5:28:10<17:16, 2.09s/it] + Training...: 89% 3899/4393 [5:28:11<15:09, 1.84s/it] + Training...: 89% 3900/4393 [5:28:12<12:56, 1.58s/it] + Training...: 89% 3901/4393 [5:28:18<24:11, 2.95s/it] + Training...: 89% 3902/4393 [5:28:24<32:16, 3.94s/it] + Training...: 89% 3903/4393 [5:28:30<37:32, 4.60s/it] + Training...: 89% 3904/4393 [5:28:36<40:43, 5.00s/it] + Training...: 89% 3905/4393 [5:28:42<42:58, 5.28s/it] + Training...: 89% 3906/4393 [5:28:48<44:19, 5.46s/it] + Training...: 89% 3907/4393 [5:28:54<45:21, 5.60s/it] + Training...: 89% 3908/4393 [5:29:00<45:35, 5.64s/it] + Training...: 89% 3909/4393 [5:29:06<45:48, 5.68s/it] + Training...: 89% 3910/4393 [5:29:11<45:46, 5.69s/it] + Training...: 89% 3911/4393 [5:29:17<45:50, 5.71s/it] + Training...: 89% 3912/4393 [5:29:23<45:46, 5.71s/it] + Training...: 89% 3913/4393 [5:29:28<45:30, 5.69s/it] +  + Training...: 89% 3913/4393 [5:29:35<45:30, 5.69s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [16:59:07<57:16:35, 20619.59s/it] + Training...: 89% 3914/4393 [5:29:35<47:11, 5.91s/it] + Training...: 89% 3915/4393 [5:29:40<46:27, 5.83s/it] + Training...: 89% 3916/4393 [5:29:46<45:42, 5.75s/it] + Training...: 89% 3917/4393 [5:29:52<45:00, 5.67s/it] + Training...: 89% 3918/4393 [5:29:57<44:25, 5.61s/it] + Training...: 89% 3919/4393 [5:30:02<43:57, 5.56s/it] + Training...: 89% 3920/4393 [5:30:08<43:32, 5.52s/it] + Training...: 89% 3921/4393 [5:30:13<43:02, 5.47s/it] + Training...: 89% 3922/4393 [5:30:19<42:31, 5.42s/it] + Training...: 89% 3923/4393 [5:30:24<42:01, 5.36s/it] + Training...: 89% 3924/4393 [5:30:29<41:23, 5.29s/it] + Training...: 89% 3925/4393 [5:30:34<40:59, 5.26s/it] + Training...: 89% 3926/4393 [5:30:39<40:35, 5.22s/it] + Training...: 89% 3927/4393 [5:30:44<40:11, 5.17s/it] + Training...: 89% 3928/4393 [5:30:49<39:54, 5.15s/it] + Training...: 89% 3929/4393 [5:30:54<39:32, 5.11s/it] + Training...: 89% 3930/4393 [5:30:59<39:00, 5.06s/it] + Training...: 89% 3931/4393 [5:31:04<38:26, 4.99s/it] + Training...: 90% 3932/4393 [5:31:09<37:40, 4.90s/it] + Training...: 90% 3933/4393 [5:31:13<36:53, 4.81s/it] + Training...: 90% 3934/4393 [5:31:18<36:24, 4.76s/it] + Training...: 90% 3935/4393 [5:31:23<35:45, 4.69s/it] + Training...: 90% 3936/4393 [5:31:27<34:55, 4.58s/it] + Training...: 90% 3937/4393 [5:31:31<33:50, 4.45s/it] + Training...: 90% 3938/4393 [5:31:35<32:45, 4.32s/it] +  + Training...: 90% 3938/4393 [5:31:39<32:45, 4.32s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:01:12<57:16:35, 20619.59s/it] + Training...: 90% 3939/4393 [5:31:39<32:52, 4.34s/it] + Training...: 90% 3940/4393 [5:31:43<31:10, 4.13s/it] + Training...: 90% 3941/4393 [5:31:47<29:28, 3.91s/it] + Training...: 90% 3942/4393 [5:31:50<27:48, 3.70s/it] + Training...: 90% 3943/4393 [5:31:53<25:54, 3.45s/it] + Training...: 90% 3944/4393 [5:31:55<24:05, 3.22s/it] + Training...: 90% 3945/4393 [5:31:58<22:00, 2.95s/it] + Training...: 90% 3946/4393 [5:32:00<19:59, 2.68s/it] + Training...: 90% 3947/4393 [5:32:01<17:52, 2.40s/it] + Training...: 90% 3948/4393 [5:32:03<15:48, 2.13s/it] + Training...: 90% 3949/4393 [5:32:04<13:47, 1.86s/it] + Training...: 90% 3950/4393 [5:32:05<11:44, 1.59s/it] + Training...: 90% 3951/4393 [5:32:11<21:40, 2.94s/it] + Training...: 90% 3952/4393 [5:32:17<28:37, 3.89s/it] + Training...: 90% 3953/4393 [5:32:23<33:23, 4.55s/it] + Training...: 90% 3954/4393 [5:32:29<36:19, 4.96s/it] + Training...: 90% 3955/4393 [5:32:35<38:15, 5.24s/it] + Training...: 90% 3956/4393 [5:32:41<39:27, 5.42s/it] + Training...: 90% 3957/4393 [5:32:47<40:20, 5.55s/it] + Training...: 90% 3958/4393 [5:32:53<40:49, 5.63s/it] + Training...: 90% 3959/4393 [5:32:59<41:03, 5.68s/it] + Training...: 90% 3960/4393 [5:33:04<41:01, 5.69s/it] + Training...: 90% 3961/4393 [5:33:10<41:02, 5.70s/it] + Training...: 90% 3962/4393 [5:33:16<40:56, 5.70s/it] + Training...: 90% 3963/4393 [5:33:22<41:09, 5.74s/it] +  + Training...: 90% 3963/4393 [5:33:28<41:09, 5.74s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:03:00<57:16:35, 20619.59s/it] + Training...: 90% 3964/4393 [5:33:28<42:53, 6.00s/it] + Training...: 90% 3965/4393 [5:33:34<41:55, 5.88s/it] + Training...: 90% 3966/4393 [5:33:39<41:04, 5.77s/it] + Training...: 90% 3967/4393 [5:33:45<40:30, 5.71s/it] + Training...: 90% 3968/4393 [5:33:50<39:57, 5.64s/it] + Training...: 90% 3969/4393 [5:33:56<39:29, 5.59s/it] + Training...: 90% 3970/4393 [5:34:01<38:58, 5.53s/it] + Training...: 90% 3971/4393 [5:34:06<38:29, 5.47s/it] + Training...: 90% 3972/4393 [5:34:12<38:08, 5.44s/it] + Training...: 90% 3973/4393 [5:34:17<37:52, 5.41s/it] + Training...: 90% 3974/4393 [5:34:22<37:36, 5.38s/it] + Training...: 90% 3975/4393 [5:34:28<37:11, 5.34s/it] + Training...: 91% 3976/4393 [5:34:33<36:39, 5.28s/it] + Training...: 91% 3977/4393 [5:34:38<36:22, 5.25s/it] + Training...: 91% 3978/4393 [5:34:43<36:09, 5.23s/it] + Training...: 91% 3979/4393 [5:34:48<35:48, 5.19s/it] + Training...: 91% 3980/4393 [5:34:53<35:08, 5.11s/it] + Training...: 91% 3981/4393 [5:34:58<34:49, 5.07s/it] + Training...: 91% 3982/4393 [5:35:03<34:30, 5.04s/it] + Training...: 91% 3983/4393 [5:35:08<33:49, 4.95s/it] + Training...: 91% 3984/4393 [5:35:13<33:02, 4.85s/it] + Training...: 91% 3985/4393 [5:35:17<32:18, 4.75s/it] + Training...: 91% 3986/4393 [5:35:21<31:31, 4.65s/it] + Training...: 91% 3987/4393 [5:35:26<30:36, 4.52s/it] + Training...: 91% 3988/4393 [5:35:30<29:40, 4.40s/it] +  + Training...: 91% 3988/4393 [5:35:34<29:40, 4.40s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:05:07<57:16:35, 20619.59s/it] + Training...: 91% 3989/4393 [5:35:34<29:56, 4.45s/it] + Training...: 91% 3990/4393 [5:35:38<28:17, 4.21s/it] + Training...: 91% 3991/4393 [5:35:41<26:32, 3.96s/it] + Training...: 91% 3992/4393 [5:35:44<24:37, 3.68s/it] + Training...: 91% 3993/4393 [5:35:47<22:42, 3.41s/it] + Training...: 91% 3994/4393 [5:35:50<20:45, 3.12s/it] + Training...: 91% 3995/4393 [5:35:52<18:57, 2.86s/it] + Training...: 91% 3996/4393 [5:35:54<17:13, 2.60s/it] + Training...: 91% 3997/4393 [5:35:56<15:30, 2.35s/it] + Training...: 91% 3998/4393 [5:35:57<13:51, 2.11s/it] + Training...: 91% 3999/4393 [5:35:58<12:09, 1.85s/it] + Training...: 91% 4000/4393 [5:35:59<10:25, 1.59s/it] + Training...: 91% 4001/4393 [5:36:06<19:20, 2.96s/it] + Training...: 91% 4002/4393 [5:36:12<25:28, 3.91s/it] + Training...: 91% 4003/4393 [5:36:18<29:31, 4.54s/it] + Training...: 91% 4004/4393 [5:36:24<32:19, 4.98s/it] + Training...: 91% 4005/4393 [5:36:30<34:32, 5.34s/it] + Training...: 91% 4006/4393 [5:36:36<35:32, 5.51s/it] + Training...: 91% 4007/4393 [5:36:42<36:05, 5.61s/it] + Training...: 91% 4008/4393 [5:36:47<36:26, 5.68s/it] + Training...: 91% 4009/4393 [5:36:53<36:33, 5.71s/it] + Training...: 91% 4010/4393 [5:36:59<36:32, 5.72s/it] + Training...: 91% 4011/4393 [5:37:05<36:31, 5.74s/it] + Training...: 91% 4012/4393 [5:37:11<36:20, 5.72s/it] + Training...: 91% 4013/4393 [5:37:16<36:13, 5.72s/it] +  + Training...: 91% 4013/4393 [5:37:23<36:13, 5.72s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:06:55<57:16:35, 20619.59s/it] + Training...: 91% 4014/4393 [5:37:23<37:41, 5.97s/it] + Training...: 91% 4015/4393 [5:37:28<36:57, 5.87s/it] + Training...: 91% 4016/4393 [5:37:34<36:19, 5.78s/it] + Training...: 91% 4017/4393 [5:37:39<35:44, 5.70s/it] + Training...: 91% 4018/4393 [5:37:45<35:15, 5.64s/it] + Training...: 91% 4019/4393 [5:37:50<34:53, 5.60s/it] + Training...: 92% 4020/4393 [5:37:56<34:58, 5.63s/it] + Training...: 92% 4021/4393 [5:38:02<34:29, 5.56s/it] + Training...: 92% 4022/4393 [5:38:07<33:56, 5.49s/it] + Training...: 92% 4023/4393 [5:38:12<33:32, 5.44s/it] + Training...: 92% 4024/4393 [5:38:17<33:05, 5.38s/it] + Training...: 92% 4025/4393 [5:38:23<32:42, 5.33s/it] + Training...: 92% 4026/4393 [5:38:28<32:16, 5.28s/it] + Training...: 92% 4027/4393 [5:38:33<31:53, 5.23s/it] + Training...: 92% 4028/4393 [5:38:38<31:25, 5.17s/it] + Training...: 92% 4029/4393 [5:38:43<31:01, 5.11s/it] + Training...: 92% 4030/4393 [5:38:48<30:34, 5.05s/it] + Training...: 92% 4031/4393 [5:38:53<30:10, 5.00s/it] + Training...: 92% 4032/4393 [5:38:58<29:47, 4.95s/it] + Training...: 92% 4033/4393 [5:39:02<29:15, 4.88s/it] + Training...: 92% 4034/4393 [5:39:07<28:36, 4.78s/it] + Training...: 92% 4035/4393 [5:39:11<27:55, 4.68s/it] + Training...: 92% 4036/4393 [5:39:16<27:11, 4.57s/it] + Training...: 92% 4037/4393 [5:39:20<26:23, 4.45s/it] + Training...: 92% 4038/4393 [5:39:24<25:28, 4.31s/it] +  + Training...: 92% 4038/4393 [5:39:28<25:28, 4.31s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:09:00<57:16:35, 20619.59s/it] + Training...: 92% 4039/4393 [5:39:28<25:27, 4.31s/it] + Training...: 92% 4040/4393 [5:39:32<23:55, 4.07s/it] + Training...: 92% 4041/4393 [5:39:35<22:22, 3.81s/it] + Training...: 92% 4042/4393 [5:39:38<20:53, 3.57s/it] + Training...: 92% 4043/4393 [5:39:41<19:31, 3.35s/it] + Training...: 92% 4044/4393 [5:39:43<18:01, 3.10s/it] + Training...: 92% 4045/4393 [5:39:45<16:29, 2.84s/it] + Training...: 92% 4046/4393 [5:39:47<14:59, 2.59s/it] + Training...: 92% 4047/4393 [5:39:49<13:29, 2.34s/it] + Training...: 92% 4048/4393 [5:39:51<12:06, 2.11s/it] + Training...: 92% 4049/4393 [5:39:52<10:38, 1.86s/it] + Training...: 92% 4050/4393 [5:39:53<09:08, 1.60s/it] + Training...: 92% 4051/4393 [5:39:59<16:57, 2.97s/it] + Training...: 92% 4052/4393 [5:40:05<22:25, 3.95s/it] + Training...: 92% 4053/4393 [5:40:11<25:51, 4.56s/it] + Training...: 92% 4054/4393 [5:40:17<28:08, 4.98s/it] + Training...: 92% 4055/4393 [5:40:23<29:43, 5.28s/it] + Training...: 92% 4056/4393 [5:40:29<30:40, 5.46s/it] + Training...: 92% 4057/4393 [5:40:35<31:15, 5.58s/it] + Training...: 92% 4058/4393 [5:40:41<31:29, 5.64s/it] + Training...: 92% 4059/4393 [5:40:47<31:38, 5.69s/it] + Training...: 92% 4060/4393 [5:40:52<31:33, 5.69s/it] + Training...: 92% 4061/4393 [5:40:58<31:24, 5.68s/it] + Training...: 92% 4062/4393 [5:41:04<31:12, 5.66s/it] + Training...: 92% 4063/4393 [5:41:09<31:06, 5.65s/it] +  + Training...: 92% 4063/4393 [5:41:16<31:06, 5.65s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:10:48<57:16:35, 20619.59s/it] + Training...: 93% 4064/4393 [5:41:16<32:05, 5.85s/it] + Training...: 93% 4065/4393 [5:41:21<31:35, 5.78s/it] + Training...: 93% 4066/4393 [5:41:27<31:13, 5.73s/it] + Training...: 93% 4067/4393 [5:41:32<30:39, 5.64s/it] + Training...: 93% 4068/4393 [5:41:38<30:16, 5.59s/it] + Training...: 93% 4069/4393 [5:41:43<30:00, 5.56s/it] + Training...: 93% 4070/4393 [5:41:49<29:40, 5.51s/it] + Training...: 93% 4071/4393 [5:41:54<29:23, 5.48s/it] + Training...: 93% 4072/4393 [5:41:59<29:00, 5.42s/it] + Training...: 93% 4073/4393 [5:42:04<28:36, 5.36s/it] + Training...: 93% 4074/4393 [5:42:10<28:19, 5.33s/it] + Training...: 93% 4075/4393 [5:42:15<27:58, 5.28s/it] + Training...: 93% 4076/4393 [5:42:20<27:39, 5.23s/it] + Training...: 93% 4077/4393 [5:42:25<27:25, 5.21s/it] + Training...: 93% 4078/4393 [5:42:30<27:02, 5.15s/it] + Training...: 93% 4079/4393 [5:42:35<26:35, 5.08s/it] + Training...: 93% 4080/4393 [5:42:40<26:14, 5.03s/it] + Training...: 93% 4081/4393 [5:42:45<26:07, 5.03s/it] + Training...: 93% 4082/4393 [5:42:50<25:45, 4.97s/it] + Training...: 93% 4083/4393 [5:42:55<25:22, 4.91s/it] + Training...: 93% 4084/4393 [5:42:59<24:47, 4.81s/it] + Training...: 93% 4085/4393 [5:43:04<24:16, 4.73s/it] + Training...: 93% 4086/4393 [5:43:08<23:40, 4.63s/it] + Training...: 93% 4087/4393 [5:43:12<22:56, 4.50s/it] + Training...: 93% 4088/4393 [5:43:16<22:13, 4.37s/it] +  + Training...: 93% 4088/4393 [5:43:21<22:13, 4.37s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:12:53<57:16:35, 20619.59s/it] + Training...: 93% 4089/4393 [5:43:21<22:26, 4.43s/it] + Training...: 93% 4090/4393 [5:43:25<21:13, 4.20s/it] + Training...: 93% 4091/4393 [5:43:28<20:00, 3.98s/it] + Training...: 93% 4092/4393 [5:43:31<18:50, 3.76s/it] + Training...: 93% 4093/4393 [5:43:34<17:35, 3.52s/it] + Training...: 93% 4094/4393 [5:43:37<16:15, 3.26s/it] + Training...: 93% 4095/4393 [5:43:39<14:55, 3.00s/it] + Training...: 93% 4096/4393 [5:43:41<13:30, 2.73s/it] + Training...: 93% 4097/4393 [5:43:43<12:05, 2.45s/it] + Training...: 93% 4098/4393 [5:43:45<10:46, 2.19s/it] + Training...: 93% 4099/4393 [5:43:46<09:28, 1.93s/it] + Training...: 93% 4100/4393 [5:43:47<08:07, 1.66s/it] + Training...: 93% 4101/4393 [5:43:53<14:42, 3.02s/it] + Training...: 93% 4102/4393 [5:44:00<19:23, 4.00s/it] + Training...: 93% 4103/4393 [5:44:06<22:18, 4.61s/it] + Training...: 93% 4104/4393 [5:44:12<24:10, 5.02s/it] + Training...: 93% 4105/4393 [5:44:18<25:37, 5.34s/it] + Training...: 93% 4106/4393 [5:44:24<26:19, 5.50s/it] + Training...: 93% 4107/4393 [5:44:30<26:44, 5.61s/it] + Training...: 94% 4108/4393 [5:44:35<26:56, 5.67s/it] + Training...: 94% 4109/4393 [5:44:41<27:03, 5.72s/it] + Training...: 94% 4110/4393 [5:44:47<26:59, 5.72s/it] + Training...: 94% 4111/4393 [5:44:53<27:00, 5.75s/it] + Training...: 94% 4112/4393 [5:44:59<27:10, 5.80s/it] + Training...: 94% 4113/4393 [5:45:04<26:54, 5.77s/it] +  + Training...: 94% 4113/4393 [5:45:11<26:54, 5.77s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:14:43<57:16:35, 20619.59s/it] + Training...: 94% 4114/4393 [5:45:11<27:39, 5.95s/it] + Training...: 94% 4115/4393 [5:45:16<27:01, 5.83s/it] + Training...: 94% 4116/4393 [5:45:22<26:30, 5.74s/it] + Training...: 94% 4117/4393 [5:45:27<26:09, 5.69s/it] + Training...: 94% 4118/4393 [5:45:33<25:45, 5.62s/it] + Training...: 94% 4119/4393 [5:45:38<25:28, 5.58s/it] + Training...: 94% 4120/4393 [5:45:44<25:08, 5.53s/it] + Training...: 94% 4121/4393 [5:45:49<24:49, 5.48s/it] + Training...: 94% 4122/4393 [5:45:54<24:34, 5.44s/it] + Training...: 94% 4123/4393 [5:46:00<24:22, 5.42s/it] + Training...: 94% 4124/4393 [5:46:05<24:02, 5.36s/it] + Training...: 94% 4125/4393 [5:46:10<23:45, 5.32s/it] + Training...: 94% 4126/4393 [5:46:15<23:24, 5.26s/it] + Training...: 94% 4127/4393 [5:46:20<23:07, 5.22s/it] + Training...: 94% 4128/4393 [5:46:26<22:48, 5.17s/it] + Training...: 94% 4129/4393 [5:46:30<22:25, 5.10s/it] + Training...: 94% 4130/4393 [5:46:35<22:06, 5.04s/it] + Training...: 94% 4131/4393 [5:46:40<21:44, 4.98s/it] + Training...: 94% 4132/4393 [5:46:45<21:27, 4.93s/it] + Training...: 94% 4133/4393 [5:46:50<21:13, 4.90s/it] + Training...: 94% 4134/4393 [5:46:54<20:45, 4.81s/it] + Training...: 94% 4135/4393 [5:46:59<20:23, 4.74s/it] + Training...: 94% 4136/4393 [5:47:03<19:53, 4.64s/it] + Training...: 94% 4137/4393 [5:47:08<19:25, 4.55s/it] + Training...: 94% 4138/4393 [5:47:12<18:48, 4.42s/it] +  + Training...: 94% 4138/4393 [5:47:16<18:48, 4.42s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:16:49<57:16:35, 20619.59s/it] + Training...: 94% 4139/4393 [5:47:16<18:46, 4.43s/it] + Training...: 94% 4140/4393 [5:47:20<17:39, 4.19s/it] + Training...: 94% 4141/4393 [5:47:23<16:35, 3.95s/it] + Training...: 94% 4142/4393 [5:47:27<15:31, 3.71s/it] + Training...: 94% 4143/4393 [5:47:29<14:22, 3.45s/it] + Training...: 94% 4144/4393 [5:47:32<13:13, 3.18s/it] + Training...: 94% 4145/4393 [5:47:34<12:04, 2.92s/it] + Training...: 94% 4146/4393 [5:47:36<10:56, 2.66s/it] + Training...: 94% 4147/4393 [5:47:38<09:51, 2.40s/it] + Training...: 94% 4148/4393 [5:47:40<08:44, 2.14s/it] + Training...: 94% 4149/4393 [5:47:41<07:35, 1.87s/it] + Training...: 94% 4150/4393 [5:47:42<06:30, 1.61s/it] + Training...: 94% 4151/4393 [5:47:48<12:02, 2.98s/it] + Training...: 95% 4152/4393 [5:47:54<15:54, 3.96s/it] + Training...: 95% 4153/4393 [5:48:00<18:19, 4.58s/it] + Training...: 95% 4154/4393 [5:48:06<19:55, 5.00s/it] + Training...: 95% 4155/4393 [5:48:12<20:59, 5.29s/it] + Training...: 95% 4156/4393 [5:48:18<21:34, 5.46s/it] + Training...: 95% 4157/4393 [5:48:24<21:53, 5.56s/it] + Training...: 95% 4158/4393 [5:48:30<22:04, 5.64s/it] + Training...: 95% 4159/4393 [5:48:36<22:08, 5.68s/it] + Training...: 95% 4160/4393 [5:48:41<22:10, 5.71s/it] + Training...: 95% 4161/4393 [5:48:47<22:10, 5.74s/it] + Training...: 95% 4162/4393 [5:48:53<22:01, 5.72s/it] + Training...: 95% 4163/4393 [5:48:58<21:52, 5.71s/it] +  + Training...: 95% 4163/4393 [5:49:05<21:52, 5.71s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:18:37<57:16:35, 20619.59s/it] + Training...: 95% 4164/4393 [5:49:05<22:42, 5.95s/it] + Training...: 95% 4165/4393 [5:49:11<22:14, 5.85s/it] + Training...: 95% 4166/4393 [5:49:16<21:48, 5.76s/it] + Training...: 95% 4167/4393 [5:49:22<21:30, 5.71s/it] + Training...: 95% 4168/4393 [5:49:27<21:12, 5.65s/it] + Training...: 95% 4169/4393 [5:49:33<20:52, 5.59s/it] + Training...: 95% 4170/4393 [5:49:38<20:39, 5.56s/it] + Training...: 95% 4171/4393 [5:49:44<20:33, 5.56s/it] + Training...: 95% 4172/4393 [5:49:49<20:14, 5.50s/it] + Training...: 95% 4173/4393 [5:49:54<19:56, 5.44s/it] + Training...: 95% 4174/4393 [5:50:00<19:40, 5.39s/it] + Training...: 95% 4175/4393 [5:50:05<19:26, 5.35s/it] + Training...: 95% 4176/4393 [5:50:10<19:08, 5.29s/it] + Training...: 95% 4177/4393 [5:50:15<18:56, 5.26s/it] + Training...: 95% 4178/4393 [5:50:20<18:41, 5.22s/it] + Training...: 95% 4179/4393 [5:50:25<18:27, 5.17s/it] + Training...: 95% 4180/4393 [5:50:30<18:07, 5.10s/it] + Training...: 95% 4181/4393 [5:50:35<17:46, 5.03s/it] + Training...: 95% 4182/4393 [5:50:40<17:26, 4.96s/it] + Training...: 95% 4183/4393 [5:50:45<17:10, 4.91s/it] + Training...: 95% 4184/4393 [5:50:50<16:52, 4.84s/it] + Training...: 95% 4185/4393 [5:50:54<16:30, 4.76s/it] + Training...: 95% 4186/4393 [5:50:59<16:05, 4.66s/it] + Training...: 95% 4187/4393 [5:51:03<15:40, 4.56s/it] + Training...: 95% 4188/4393 [5:51:07<15:04, 4.41s/it] +  + Training...: 95% 4188/4393 [5:51:11<15:04, 4.41s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:20:44<57:16:35, 20619.59s/it] + Training...: 95% 4189/4393 [5:51:11<15:00, 4.41s/it] + Training...: 95% 4190/4393 [5:51:15<14:04, 4.16s/it] + Training...: 95% 4191/4393 [5:51:18<13:11, 3.92s/it] + Training...: 95% 4192/4393 [5:51:21<12:19, 3.68s/it] + Training...: 95% 4193/4393 [5:51:24<11:21, 3.41s/it] + Training...: 95% 4194/4393 [5:51:27<10:27, 3.15s/it] + Training...: 95% 4195/4393 [5:51:29<09:35, 2.90s/it] + Training...: 96% 4196/4393 [5:51:31<08:45, 2.67s/it] + Training...: 96% 4197/4393 [5:51:33<07:51, 2.40s/it] + Training...: 96% 4198/4393 [5:51:35<06:59, 2.15s/it] + Training...: 96% 4199/4393 [5:51:36<06:05, 1.88s/it] + Training...: 96% 4200/4393 [5:51:37<05:13, 1.62s/it] + Training...: 96% 4201/4393 [5:51:43<09:33, 2.99s/it] + Training...: 96% 4202/4393 [5:51:49<12:34, 3.95s/it] + Training...: 96% 4203/4393 [5:51:55<14:27, 4.57s/it] + Training...: 96% 4204/4393 [5:52:01<15:45, 5.00s/it] + Training...: 96% 4205/4393 [5:52:07<16:37, 5.30s/it] + Training...: 96% 4206/4393 [5:52:13<17:10, 5.51s/it] + Training...: 96% 4207/4393 [5:52:19<17:29, 5.64s/it] + Training...: 96% 4208/4393 [5:52:25<17:30, 5.68s/it] + Training...: 96% 4209/4393 [5:52:31<17:32, 5.72s/it] + Training...: 96% 4210/4393 [5:52:36<17:29, 5.74s/it] + Training...: 96% 4211/4393 [5:52:42<17:23, 5.73s/it] + Training...: 96% 4212/4393 [5:52:48<17:15, 5.72s/it] + Training...: 96% 4213/4393 [5:52:54<17:05, 5.70s/it] +  + Training...: 96% 4213/4393 [5:53:00<17:05, 5.70s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:22:32<57:16:35, 20619.59s/it] + Training...: 96% 4214/4393 [5:53:00<17:41, 5.93s/it] + Training...: 96% 4215/4393 [5:53:06<17:19, 5.84s/it] + Training...: 96% 4216/4393 [5:53:11<16:58, 5.76s/it] + Training...: 96% 4217/4393 [5:53:17<16:40, 5.69s/it] + Training...: 96% 4218/4393 [5:53:22<16:28, 5.65s/it] + Training...: 96% 4219/4393 [5:53:28<16:18, 5.62s/it] + Training...: 96% 4220/4393 [5:53:33<16:05, 5.58s/it] + Training...: 96% 4221/4393 [5:53:39<15:52, 5.54s/it] + Training...: 96% 4222/4393 [5:53:44<15:38, 5.49s/it] + Training...: 96% 4223/4393 [5:53:50<15:31, 5.48s/it] + Training...: 96% 4224/4393 [5:53:55<15:17, 5.43s/it] + Training...: 96% 4225/4393 [5:54:00<15:02, 5.37s/it] + Training...: 96% 4226/4393 [5:54:05<14:46, 5.31s/it] + Training...: 96% 4227/4393 [5:54:10<14:32, 5.26s/it] + Training...: 96% 4228/4393 [5:54:16<14:18, 5.20s/it] + Training...: 96% 4229/4393 [5:54:21<14:03, 5.14s/it] + Training...: 96% 4230/4393 [5:54:25<13:48, 5.08s/it] + Training...: 96% 4231/4393 [5:54:30<13:32, 5.02s/it] + Training...: 96% 4232/4393 [5:54:35<13:17, 4.95s/it] + Training...: 96% 4233/4393 [5:54:40<13:03, 4.90s/it] + Training...: 96% 4234/4393 [5:54:45<12:46, 4.82s/it] + Training...: 96% 4235/4393 [5:54:49<12:29, 4.75s/it] + Training...: 96% 4236/4393 [5:54:54<12:10, 4.65s/it] + Training...: 96% 4237/4393 [5:54:58<11:48, 4.54s/it] + Training...: 96% 4238/4393 [5:55:02<11:24, 4.41s/it] +  + Training...: 96% 4238/4393 [5:55:06<11:24, 4.41s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:24:39<57:16:35, 20619.59s/it] + Training...: 96% 4239/4393 [5:55:06<11:25, 4.45s/it] + Training...: 97% 4240/4393 [5:55:10<10:43, 4.21s/it] + Training...: 97% 4241/4393 [5:55:14<10:04, 3.97s/it] + Training...: 97% 4242/4393 [5:55:17<09:26, 3.75s/it] + Training...: 97% 4243/4393 [5:55:20<08:42, 3.48s/it] + Training...: 97% 4244/4393 [5:55:22<07:57, 3.20s/it] + Training...: 97% 4245/4393 [5:55:25<07:14, 2.94s/it] + Training...: 97% 4246/4393 [5:55:27<06:33, 2.68s/it] + Training...: 97% 4247/4393 [5:55:28<05:51, 2.41s/it] + Training...: 97% 4248/4393 [5:55:30<05:12, 2.15s/it] + Training...: 97% 4249/4393 [5:55:31<04:31, 1.89s/it] + Training...: 97% 4250/4393 [5:55:32<03:52, 1.63s/it] + Training...: 97% 4251/4393 [5:55:39<07:11, 3.04s/it] + Training...: 97% 4252/4393 [5:55:45<09:29, 4.04s/it] + Training...: 97% 4253/4393 [5:55:51<10:54, 4.67s/it] + Training...: 97% 4254/4393 [5:55:57<11:43, 5.06s/it] + Training...: 97% 4255/4393 [5:56:03<12:14, 5.32s/it] + Training...: 97% 4256/4393 [5:56:09<12:36, 5.52s/it] + Training...: 97% 4257/4393 [5:56:15<12:48, 5.65s/it] + Training...: 97% 4258/4393 [5:56:21<12:50, 5.71s/it] + Training...: 97% 4259/4393 [5:56:27<12:53, 5.77s/it] + Training...: 97% 4260/4393 [5:56:32<12:44, 5.75s/it] + Training...: 97% 4261/4393 [5:56:38<12:41, 5.77s/it] + Training...: 97% 4262/4393 [5:56:44<12:32, 5.74s/it] + Training...: 97% 4263/4393 [5:56:50<12:24, 5.73s/it] +  + Training...: 97% 4263/4393 [5:56:56<12:24, 5.73s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:26:28<57:16:35, 20619.59s/it] + Training...: 97% 4264/4393 [5:56:56<12:46, 5.94s/it] + Training...: 97% 4265/4393 [5:57:02<12:29, 5.85s/it] + Training...: 97% 4266/4393 [5:57:07<12:14, 5.79s/it] + Training...: 97% 4267/4393 [5:57:13<11:59, 5.71s/it] + Training...: 97% 4268/4393 [5:57:18<11:50, 5.69s/it] + Training...: 97% 4269/4393 [5:57:24<11:45, 5.69s/it] + Training...: 97% 4270/4393 [5:57:30<11:29, 5.60s/it] + Training...: 97% 4271/4393 [5:57:35<11:15, 5.53s/it] + Training...: 97% 4272/4393 [5:57:40<11:02, 5.48s/it] + Training...: 97% 4273/4393 [5:57:46<10:50, 5.42s/it] + Training...: 97% 4274/4393 [5:57:51<10:39, 5.37s/it] + Training...: 97% 4275/4393 [5:57:56<10:28, 5.32s/it] + Training...: 97% 4276/4393 [5:58:01<10:17, 5.28s/it] + Training...: 97% 4277/4393 [5:58:06<10:08, 5.24s/it] + Training...: 97% 4278/4393 [5:58:11<09:56, 5.19s/it] + Training...: 97% 4279/4393 [5:58:16<09:46, 5.15s/it] + Training...: 97% 4280/4393 [5:58:21<09:33, 5.07s/it] + Training...: 97% 4281/4393 [5:58:26<09:21, 5.01s/it] + Training...: 97% 4282/4393 [5:58:31<09:09, 4.95s/it] + Training...: 97% 4283/4393 [5:58:36<08:57, 4.88s/it] + Training...: 98% 4284/4393 [5:58:40<08:43, 4.81s/it] + Training...: 98% 4285/4393 [5:58:45<08:36, 4.79s/it] + Training...: 98% 4286/4393 [5:58:50<08:24, 4.71s/it] + Training...: 98% 4287/4393 [5:58:54<08:06, 4.59s/it] + Training...: 98% 4288/4393 [5:58:58<07:44, 4.42s/it] +  + Training...: 98% 4288/4393 [5:59:02<07:44, 4.42s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:28:35<57:16:35, 20619.59s/it] + Training...: 98% 4289/4393 [5:59:02<07:41, 4.44s/it] + Training...: 98% 4290/4393 [5:59:06<07:12, 4.20s/it] + Training...: 98% 4291/4393 [5:59:10<06:46, 3.98s/it] + Training...: 98% 4292/4393 [5:59:13<06:16, 3.73s/it] + Training...: 98% 4293/4393 [5:59:16<05:49, 3.49s/it] + Training...: 98% 4294/4393 [5:59:18<05:19, 3.23s/it] + Training...: 98% 4295/4393 [5:59:21<04:46, 2.93s/it] + Training...: 98% 4296/4393 [5:59:22<04:15, 2.64s/it] + Training...: 98% 4297/4393 [5:59:24<03:47, 2.37s/it] + Training...: 98% 4298/4393 [5:59:26<03:20, 2.11s/it] + Training...: 98% 4299/4393 [5:59:27<02:54, 1.86s/it] + Training...: 98% 4300/4393 [5:59:28<02:28, 1.60s/it] + Training...: 98% 4301/4393 [5:59:34<04:34, 2.98s/it] + Training...: 98% 4302/4393 [5:59:40<06:00, 3.96s/it] + Training...: 98% 4303/4393 [5:59:47<06:54, 4.60s/it] + Training...: 98% 4304/4393 [5:59:53<07:26, 5.02s/it] + Training...: 98% 4305/4393 [5:59:58<07:46, 5.30s/it] + Training...: 98% 4306/4393 [6:00:04<07:57, 5.48s/it] + Training...: 98% 4307/4393 [6:00:10<08:04, 5.63s/it] + Training...: 98% 4308/4393 [6:00:16<08:04, 5.70s/it] + Training...: 98% 4309/4393 [6:00:22<08:01, 5.73s/it] + Training...: 98% 4310/4393 [6:00:28<07:55, 5.73s/it] + Training...: 98% 4311/4393 [6:00:34<07:50, 5.74s/it] + Training...: 98% 4312/4393 [6:00:39<07:45, 5.75s/it] + Training...: 98% 4313/4393 [6:00:45<07:40, 5.75s/it] +  + Training...: 98% 4313/4393 [6:00:52<07:40, 5.75s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:30:24<57:16:35, 20619.59s/it] + Training...: 98% 4314/4393 [6:00:52<07:52, 5.98s/it] + Training...: 98% 4315/4393 [6:00:57<07:37, 5.86s/it] + Training...: 98% 4316/4393 [6:01:03<07:26, 5.80s/it] + Training...: 98% 4317/4393 [6:01:08<07:16, 5.74s/it] + Training...: 98% 4318/4393 [6:01:14<07:06, 5.69s/it] + Training...: 98% 4319/4393 [6:01:19<06:55, 5.61s/it] + Training...: 98% 4320/4393 [6:01:25<06:45, 5.55s/it] + Training...: 98% 4321/4393 [6:01:30<06:36, 5.51s/it] + Training...: 98% 4322/4393 [6:01:36<06:27, 5.46s/it] + Training...: 98% 4323/4393 [6:01:41<06:20, 5.43s/it] + Training...: 98% 4324/4393 [6:01:46<06:10, 5.37s/it] + Training...: 98% 4325/4393 [6:01:52<06:04, 5.36s/it] + Training...: 98% 4326/4393 [6:01:57<05:53, 5.28s/it] + Training...: 98% 4327/4393 [6:02:02<05:43, 5.21s/it] + Training...: 99% 4328/4393 [6:02:07<05:34, 5.15s/it] + Training...: 99% 4329/4393 [6:02:12<05:26, 5.11s/it] + Training...: 99% 4330/4393 [6:02:17<05:17, 5.03s/it] + Training...: 99% 4331/4393 [6:02:21<05:08, 4.97s/it] + Training...: 99% 4332/4393 [6:02:26<04:59, 4.91s/it] + Training...: 99% 4333/4393 [6:02:31<04:51, 4.86s/it] + Training...: 99% 4334/4393 [6:02:35<04:41, 4.77s/it] + Training...: 99% 4335/4393 [6:02:40<04:31, 4.68s/it] + Training...: 99% 4336/4393 [6:02:44<04:20, 4.57s/it] + Training...: 99% 4337/4393 [6:02:48<04:09, 4.45s/it] + Training...: 99% 4338/4393 [6:02:52<03:57, 4.32s/it] +  + Training...: 99% 4338/4393 [6:02:57<03:57, 4.32s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:32:29<57:16:35, 20619.59s/it] + Training...: 99% 4339/4393 [6:02:57<03:54, 4.34s/it] + Training...: 99% 4340/4393 [6:03:00<03:38, 4.13s/it] + Training...: 99% 4341/4393 [6:03:04<03:22, 3.90s/it] + Training...: 99% 4342/4393 [6:03:07<03:06, 3.65s/it] + Training...: 99% 4343/4393 [6:03:10<02:50, 3.42s/it] + Training...: 99% 4344/4393 [6:03:12<02:35, 3.18s/it] + Training...: 99% 4345/4393 [6:03:15<02:19, 2.91s/it] + Training...: 99% 4346/4393 [6:03:17<02:04, 2.65s/it] + Training...: 99% 4347/4393 [6:03:18<01:49, 2.38s/it] + Training...: 99% 4348/4393 [6:03:20<01:35, 2.12s/it] + Training...: 99% 4349/4393 [6:03:21<01:21, 1.85s/it] + Training...: 99% 4350/4393 [6:03:22<01:08, 1.59s/it] + Training...: 99% 4351/4393 [6:03:28<02:04, 2.97s/it] + Training...: 99% 4352/4393 [6:03:34<02:39, 3.90s/it] + Training...: 99% 4353/4393 [6:03:40<03:01, 4.54s/it] + Training...: 99% 4354/4393 [6:03:46<03:12, 4.94s/it] + Training...: 99% 4355/4393 [6:03:52<03:19, 5.24s/it] + Training...: 99% 4356/4393 [6:03:58<03:21, 5.44s/it] + Training...: 99% 4357/4393 [6:04:04<03:22, 5.61s/it] + Training...: 99% 4358/4393 [6:04:10<03:19, 5.69s/it] + Training...: 99% 4359/4393 [6:04:16<03:14, 5.72s/it] + Training...: 99% 4360/4393 [6:04:22<03:08, 5.72s/it] + Training...: 99% 4361/4393 [6:04:27<03:02, 5.70s/it] + Training...: 99% 4362/4393 [6:04:33<02:55, 5.66s/it] + Training...: 99% 4363/4393 [6:04:38<02:49, 5.64s/it] +  + Training...: 99% 4363/4393 [6:04:45<02:49, 5.64s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:34:17<57:16:35, 20619.59s/it] + Training...: 99% 4364/4393 [6:04:45<02:50, 5.88s/it] + Training...: 99% 4365/4393 [6:04:50<02:41, 5.76s/it] + Training...: 99% 4366/4393 [6:04:56<02:32, 5.64s/it] + Training...: 99% 4367/4393 [6:05:01<02:25, 5.58s/it] + Training...: 99% 4368/4393 [6:05:06<02:17, 5.51s/it] + Training...: 99% 4369/4393 [6:05:12<02:11, 5.47s/it] + Training...: 99% 4370/4393 [6:05:17<02:04, 5.43s/it] + Training...: 99% 4371/4393 [6:05:22<01:58, 5.37s/it] + Training...: 100% 4372/4393 [6:05:28<01:51, 5.33s/it] + Training...: 100% 4373/4393 [6:05:33<01:46, 5.34s/it] + Training...: 100% 4374/4393 [6:05:38<01:41, 5.32s/it] + Training...: 100% 4375/4393 [6:05:43<01:34, 5.24s/it] + Training...: 100% 4376/4393 [6:05:48<01:27, 5.14s/it] + Training...: 100% 4377/4393 [6:05:53<01:20, 5.04s/it] + Training...: 100% 4378/4393 [6:05:58<01:14, 4.96s/it] + Training...: 100% 4379/4393 [6:06:02<01:08, 4.87s/it] + Training...: 100% 4380/4393 [6:06:07<01:02, 4.78s/it] + Training...: 100% 4381/4393 [6:06:11<00:56, 4.68s/it] + Training...: 100% 4382/4393 [6:06:16<00:49, 4.54s/it] + Training...: 100% 4383/4393 [6:06:20<00:43, 4.38s/it] + Training...: 100% 4384/4393 [6:06:24<00:37, 4.22s/it] + Training...: 100% 4385/4393 [6:06:27<00:32, 4.00s/it] + Training...: 100% 4386/4393 [6:06:30<00:26, 3.76s/it] + Training...: 100% 4387/4393 [6:06:33<00:20, 3.50s/it] + Training...: 100% 4388/4393 [6:06:36<00:16, 3.22s/it] +  + Training...: 100% 4388/4393 [6:06:38<00:16, 3.22s/it] Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 17% 2/12 [17:36:11<57:16:35, 20619.59s/it] + Training...: 100% 4389/4393 [6:06:38<00:12, 3.08s/it] + Training...: 100% 4390/4393 [6:06:40<00:08, 2.77s/it] + Training...: 100% 4391/4393 [6:06:42<00:04, 2.47s/it] + Training...: 100% 4392/4393 [6:06:44<00:02, 2.21s/it] + Training...: 100% 4393/4393 [6:06:45<00:00, 1.91s/it] Training...: 100% 4393/4393 [6:06:45<00:00, 5.01s/it] + Step... (10000/50000 | Eval Loss: 0.6138997077941895 | Eval wer: 0.05543913826697548 | Eval cer: 0.039964500651745845 |): 25% 3/12 [17:36:17<53:07:59, 21253.24s/it] + Training...: 0% 0/4393 [00:00