diff --git "a/exp/log/log-train-2022-12-09-10-39-23-5" "b/exp/log/log-train-2022-12-09-10-39-23-5"
new file mode 100644--- /dev/null
+++ "b/exp/log/log-train-2022-12-09-10-39-23-5"
@@ -0,0 +1,6042 @@
+2022-12-09 10:39:23,970 INFO [train.py:493] (5/8) Training started
+2022-12-09 10:39:23,970 INFO [train.py:494] (5/8) {'max_sent_len': 200, 'sos_id': 1, 'eos_id': 1, 'blank_id': 0, 'lr': 0.001, 'weight_decay': 1e-06, 'best_train_loss': inf, 'best_valid_loss': inf, 'best_train_epoch': -1, 'best_valid_epoch': -1, 'batch_idx_train': 0, 'log_interval': 200, 'reset_interval': 2000, 'valid_interval': 1000, 'nhead': 8, 'embedding_dim': 768, 'encoder_dim': 768, 'dim_feedforward': 2048, 'dropout': 0.1, 'env_info': {'k2-version': '1.22', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': '96c9a2aece2a3a7633da07740e24fa3d96f5498c', 'k2-git-date': 'Thu Nov 10 08:14:02 2022', 'lhotse-version': '1.10.0', 'torch-version': '1.12.1', 'torch-cuda-available': True, 'torch-cuda-version': '11.6', 'python-version': '3.8', 'icefall-git-branch': 'transformer_lm', 'icefall-git-sha1': '5b028fe-dirty', 'icefall-git-date': 'Thu Dec 8 23:29:10 2022', 'icefall-path': '/ceph-data4/yangxiaoyu/softwares/icefall_development/icefall_zipformer_mvq', 'k2-path': '/ceph-data4/yangxiaoyu/softwares/anaconda3/envs/k2_latest/lib/python3.8/site-packages/k2/__init__.py', 'lhotse-path': '/ceph-data4/yangxiaoyu/softwares/anaconda3/envs/k2_latest/lib/python3.8/site-packages/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-1-0307195509-567fcb96d6-kdztg', 'IP address': '10.177.22.10'}, 'world_size': 8, 'master_port': 12354, 'tensorboard': True, 'num_epochs': 20, 'start_epoch': 0, 'exp_dir': PosixPath('transformer_lm/exp_full_libri_16layer_maxlen200_8gpu'), 'use_fp16': False, 'batch_size': 70, 'lm_data': './transformer_lm/libri_lm_training_bpe500/sorted-lm-data-libri-lm_maxlen200.pt', 'lm_data_valid': './transformer_lm/libri_lm_training_bpe500/sorted_lm_data-test.pt', 'vocab_size': 500, 'num_layers': 16, 'tie_weights': True, 'seed': 42}
+2022-12-09 10:39:23,971 INFO [train.py:505] (5/8) Device: cuda:5
+2022-12-09 10:39:23,971 INFO [train.py:507] (5/8) About to create model
+2022-12-09 10:39:24,451 INFO [model.py:64] (5/8) Tying weights
+2022-12-09 10:39:24,452 INFO [train.py:520] (5/8) Number of model parameters: 98611638
+2022-12-09 10:39:27,367 INFO [train.py:539] (5/8) Loading LM training data from ./transformer_lm/libri_lm_training_bpe500/sorted-lm-data-libri-lm_maxlen200.pt
+2022-12-09 10:39:41,333 INFO [train.py:546] (5/8) Loading LM validation data from ./transformer_lm/libri_lm_training_bpe500/sorted_lm_data-test.pt
+2022-12-09 10:39:44,739 INFO [train.py:421] (5/8) Epoch 0, batch 0, loss[loss=86.05, over 910.00 frames. , ppl: 2.3611747049389668e+37] tot_loss[loss=86.05, over 910.00 frames. , ppl: 2.3611747049389668e+37], batch size: 70
+2022-12-09 10:39:44,761 INFO [distributed.py:995] (5/8) Reducer buckets have been rebuilt in this iteration.
+2022-12-09 10:41:21,499 INFO [train.py:421] (5/8) Epoch 0, batch 200, loss[loss=8.012, over 2520.00 frames. , ppl: 3016.910179523348] tot_loss[loss=17, over 470178.78 frames. , ppl: 24265995.273026533], batch size: 70
+2022-12-09 10:43:01,687 INFO [train.py:421] (5/8) Epoch 0, batch 400, loss[loss=6.723, over 1610.00 frames. , ppl: 831.4001193896454] tot_loss[loss=11.48, over 996108.36 frames. , ppl: 96892.29325165658], batch size: 70
+2022-12-09 10:44:42,160 INFO [train.py:421] (5/8) Epoch 0, batch 600, loss[loss=6.411, over 4340.00 frames. , ppl: 608.3337950406494] tot_loss[loss=9.696, over 1405177.63 frames. , ppl: 16254.29321759802], batch size: 70
+2022-12-09 10:46:22,383 INFO [train.py:421] (5/8) Epoch 0, batch 800, loss[loss=5.452, over 2520.00 frames. , ppl: 233.2571427669306] tot_loss[loss=8.647, over 1794794.77 frames. , ppl: 5691.136485985011], batch size: 70
+2022-12-09 10:48:02,301 INFO [train.py:421] (5/8) Epoch 0, batch 1000, loss[loss=5.433, over 3290.00 frames. , ppl: 228.86653344407708] tot_loss[loss=7.878, over 2134034.34 frames. , ppl: 2638.1188259676046], batch size: 70
+2022-12-09 10:48:02,301 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 10:48:03,029 INFO [train.py:452] (5/8) Epoch 0, validation: loss=5.053, over 211138.00 frames. , ppl: 156.50471436114304
+2022-12-09 10:49:43,714 INFO [train.py:421] (5/8) Epoch 0, batch 1200, loss[loss=5.313, over 2870.00 frames. , ppl: 202.98274691532495] tot_loss[loss=7.392, over 2462155.69 frames. , ppl: 1623.0491063667414], batch size: 70
+2022-12-09 10:51:23,207 INFO [train.py:421] (5/8) Epoch 0, batch 1400, loss[loss=5.195, over 1470.00 frames. , ppl: 180.43770185234834] tot_loss[loss=7.014, over 2761700.74 frames. , ppl: 1112.5993993965676], batch size: 70
+2022-12-09 10:53:02,893 INFO [train.py:421] (5/8) Epoch 0, batch 1600, loss[loss=4.83, over 6650.00 frames. , ppl: 125.17080897330224] tot_loss[loss=6.672, over 3017884.64 frames. , ppl: 790.0472940369909], batch size: 70
+2022-12-09 10:54:41,870 INFO [train.py:421] (5/8) Epoch 0, batch 1800, loss[loss=4.444, over 2800.00 frames. , ppl: 85.07531305314956] tot_loss[loss=6.363, over 3231939.64 frames. , ppl: 580.0731109621893], batch size: 70
+2022-12-09 10:56:19,726 INFO [train.py:421] (5/8) Epoch 0, batch 2000, loss[loss=4.368, over 2940.00 frames. , ppl: 78.92139688041773] tot_loss[loss=6.068, over 3470364.22 frames. , ppl: 431.61805276945967], batch size: 70
+2022-12-09 10:56:19,726 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 10:56:20,476 INFO [train.py:452] (5/8) Epoch 0, validation: loss=4.224, over 211138.00 frames. , ppl: 68.31401714385908
+2022-12-09 10:58:02,726 INFO [train.py:421] (5/8) Epoch 0, batch 2200, loss[loss=4.571, over 1050.00 frames. , ppl: 96.61911502051767] tot_loss[loss=5.836, over 3684745.07 frames. , ppl: 342.23693772842483], batch size: 70
+2022-12-09 10:59:45,494 INFO [train.py:421] (5/8) Epoch 0, batch 2400, loss[loss=4.266, over 6300.00 frames. , ppl: 71.26707658535413] tot_loss[loss=5.638, over 3866341.21 frames. , ppl: 280.993555126692], batch size: 70
+2022-12-09 11:01:26,791 INFO [train.py:421] (5/8) Epoch 0, batch 2600, loss[loss=4.205, over 4480.00 frames. , ppl: 67.02662282518068] tot_loss[loss=5.447, over 4044413.17 frames. , ppl: 232.09903674205495], batch size: 70
+2022-12-09 11:03:09,418 INFO [train.py:421] (5/8) Epoch 0, batch 2800, loss[loss=3.95, over 6230.00 frames. , ppl: 51.95073922701872] tot_loss[loss=5.263, over 4236028.19 frames. , ppl: 192.99034101742603], batch size: 70
+2022-12-09 11:04:49,047 INFO [train.py:421] (5/8) Epoch 0, batch 3000, loss[loss=3.752, over 6860.00 frames. , ppl: 42.591671637253775] tot_loss[loss=5.098, over 4345883.82 frames. , ppl: 163.68601365978293], batch size: 70
+2022-12-09 11:04:49,048 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 11:04:49,779 INFO [train.py:452] (5/8) Epoch 0, validation: loss=3.608, over 211138.00 frames. , ppl: 36.89311779083857
+2022-12-09 11:06:26,626 INFO [train.py:421] (5/8) Epoch 0, batch 3200, loss[loss=3.545, over 1470.00 frames. , ppl: 34.64371039603855] tot_loss[loss=4.934, over 4431916.39 frames. , ppl: 138.9425585282743], batch size: 70
+2022-12-09 11:08:06,512 INFO [train.py:421] (5/8) Epoch 0, batch 3400, loss[loss=3.503, over 3920.00 frames. , ppl: 33.2306709772559] tot_loss[loss=4.767, over 4548949.34 frames. , ppl: 117.5730155799896], batch size: 70
+2022-12-09 11:09:49,649 INFO [train.py:421] (5/8) Epoch 0, batch 3600, loss[loss=3.302, over 5390.00 frames. , ppl: 27.166284892834696] tot_loss[loss=4.622, over 4598049.76 frames. , ppl: 101.72376209281818], batch size: 70
+2022-12-09 11:11:30,325 INFO [train.py:421] (5/8) Epoch 0, batch 3800, loss[loss=3.8, over 490.00 frames. , ppl: 44.68521818220669] tot_loss[loss=4.484, over 4641225.44 frames. , ppl: 88.59609480135197], batch size: 70
+2022-12-09 11:13:10,811 INFO [train.py:421] (5/8) Epoch 0, batch 4000, loss[loss=3.332, over 1260.00 frames. , ppl: 27.987924757119757] tot_loss[loss=4.345, over 4720648.34 frames. , ppl: 77.06657218689152], batch size: 70
+2022-12-09 11:13:10,811 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 11:13:11,561 INFO [train.py:452] (5/8) Epoch 0, validation: loss=3.125, over 211138.00 frames. , ppl: 22.762158928396982
+2022-12-09 11:14:48,443 INFO [train.py:421] (5/8) Epoch 0, batch 4200, loss[loss=3.166, over 2310.00 frames. , ppl: 23.714469267434257] tot_loss[loss=4.221, over 4771884.37 frames. , ppl: 68.09597270032843], batch size: 70
+2022-12-09 11:16:24,327 INFO [train.py:421] (5/8) Epoch 0, batch 4400, loss[loss=3.075, over 4620.00 frames. , ppl: 21.646001564822893] tot_loss[loss=4.099, over 4852101.47 frames. , ppl: 60.29831450485702], batch size: 70
+2022-12-09 11:18:03,018 INFO [train.py:421] (5/8) Epoch 0, batch 4600, loss[loss=3.041, over 2030.00 frames. , ppl: 20.921342300591952] tot_loss[loss=3.991, over 4914233.99 frames. , ppl: 54.10074417117683], batch size: 70
+2022-12-09 11:19:43,497 INFO [train.py:421] (5/8) Epoch 0, batch 4800, loss[loss=2.911, over 910.00 frames. , ppl: 18.376875466831816] tot_loss[loss=3.9, over 4922292.79 frames. , ppl: 49.3955900032872], batch size: 70
+2022-12-09 11:21:24,016 INFO [train.py:421] (5/8) Epoch 0, batch 5000, loss[loss=2.925, over 6440.00 frames. , ppl: 18.625677453326798] tot_loss[loss=3.802, over 5009966.68 frames. , ppl: 44.79152458608022], batch size: 70
+2022-12-09 11:21:24,016 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 11:21:24,746 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.954, over 211138.00 frames. , ppl: 19.189241084043
+2022-12-09 11:23:04,384 INFO [train.py:421] (5/8) Epoch 0, batch 5200, loss[loss=3.082, over 980.00 frames. , ppl: 21.805099264009986] tot_loss[loss=3.718, over 5058693.06 frames. , ppl: 41.182434260080306], batch size: 70
+2022-12-09 11:24:42,560 INFO [train.py:421] (5/8) Epoch 0, batch 5400, loss[loss=2.95, over 5110.00 frames. , ppl: 19.097032027571004] tot_loss[loss=3.645, over 5084925.03 frames. , ppl: 38.27362797455936], batch size: 70
+2022-12-09 11:26:20,378 INFO [train.py:421] (5/8) Epoch 0, batch 5600, loss[loss=2.892, over 2800.00 frames. , ppl: 18.020717034219395] tot_loss[loss=3.576, over 5114460.66 frames. , ppl: 35.72829957249627], batch size: 70
+2022-12-09 11:28:00,840 INFO [train.py:421] (5/8) Epoch 0, batch 5800, loss[loss=2.871, over 1470.00 frames. , ppl: 17.657762040507006] tot_loss[loss=3.508, over 5179212.15 frames. , ppl: 33.37480750822278], batch size: 70
+2022-12-09 11:29:42,214 INFO [train.py:421] (5/8) Epoch 0, batch 6000, loss[loss=3.568, over 560.00 frames. , ppl: 35.44178829035215] tot_loss[loss=3.445, over 5249963.32 frames. , ppl: 31.347961302090315], batch size: 70
+2022-12-09 11:29:42,215 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 11:29:42,974 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.875, over 211138.00 frames. , ppl: 17.725366404663895
+2022-12-09 11:31:26,526 INFO [train.py:421] (5/8) Epoch 0, batch 6200, loss[loss=2.859, over 7560.00 frames. , ppl: 17.44614066016889] tot_loss[loss=3.394, over 5262939.22 frames. , ppl: 29.771489976818195], batch size: 70
+2022-12-09 11:33:07,632 INFO [train.py:421] (5/8) Epoch 0, batch 6400, loss[loss=2.852, over 4480.00 frames. , ppl: 17.316158827613453] tot_loss[loss=3.35, over 5234545.63 frames. , ppl: 28.501184218877665], batch size: 70
+2022-12-09 11:34:47,513 INFO [train.py:421] (5/8) Epoch 0, batch 6600, loss[loss=2.856, over 3010.00 frames. , ppl: 17.38387489929852] tot_loss[loss=3.306, over 5239181.85 frames. , ppl: 27.285263295251298], batch size: 70
+2022-12-09 11:36:26,660 INFO [train.py:421] (5/8) Epoch 0, batch 6800, loss[loss=2.85, over 2590.00 frames. , ppl: 17.284008111550015] tot_loss[loss=3.266, over 5234335.31 frames. , ppl: 26.218082005404145], batch size: 70
+2022-12-09 11:38:05,181 INFO [train.py:421] (5/8) Epoch 0, batch 7000, loss[loss=2.791, over 3710.00 frames. , ppl: 16.29855890322734] tot_loss[loss=3.227, over 5248167.11 frames. , ppl: 25.193154660890997], batch size: 70
+2022-12-09 11:38:05,182 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 11:38:05,970 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.816, over 211138.00 frames. , ppl: 16.709705528719027
+2022-12-09 11:39:47,610 INFO [train.py:421] (5/8) Epoch 0, batch 7200, loss[loss=2.988, over 770.00 frames. , ppl: 19.855543166715687] tot_loss[loss=3.188, over 5297172.80 frames. , ppl: 24.233598664054853], batch size: 70
+2022-12-09 11:41:23,910 INFO [train.py:421] (5/8) Epoch 0, batch 7400, loss[loss=2.899, over 2730.00 frames. , ppl: 18.155399174333297] tot_loss[loss=3.157, over 5280806.96 frames. , ppl: 23.50000308077021], batch size: 70
+2022-12-09 11:43:02,498 INFO [train.py:421] (5/8) Epoch 0, batch 7600, loss[loss=2.832, over 1400.00 frames. , ppl: 16.981900301526565] tot_loss[loss=3.126, over 5294520.97 frames. , ppl: 22.779826276045323], batch size: 70
+2022-12-09 11:44:44,305 INFO [train.py:421] (5/8) Epoch 0, batch 7800, loss[loss=2.715, over 2870.00 frames. , ppl: 15.11206327381725] tot_loss[loss=3.097, over 5305521.12 frames. , ppl: 22.126745458521217], batch size: 70
+2022-12-09 11:46:23,635 INFO [train.py:421] (5/8) Epoch 0, batch 8000, loss[loss=2.791, over 2800.00 frames. , ppl: 16.29193160377458] tot_loss[loss=3.07, over 5324406.08 frames. , ppl: 21.53897437252353], batch size: 70
+2022-12-09 11:46:23,636 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 11:46:24,382 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.773, over 211138.00 frames. , ppl: 16.01179093613074
+2022-12-09 11:47:58,403 INFO [train.py:421] (5/8) Epoch 0, batch 8200, loss[loss=2.929, over 1680.00 frames. , ppl: 18.70165679643222] tot_loss[loss=3.046, over 5325838.13 frames. , ppl: 21.022086849100805], batch size: 70
+2022-12-09 11:49:42,155 INFO [train.py:421] (5/8) Epoch 0, batch 8400, loss[loss=2.834, over 2100.00 frames. , ppl: 17.00729854544945] tot_loss[loss=3.02, over 5368028.42 frames. , ppl: 20.498763810506023], batch size: 70
+2022-12-09 11:51:19,108 INFO [train.py:421] (5/8) Epoch 0, batch 8600, loss[loss=2.877, over 1750.00 frames. , ppl: 17.755583961051293] tot_loss[loss=2.999, over 5367170.16 frames. , ppl: 20.074453923026113], batch size: 70
+2022-12-09 11:53:00,437 INFO [train.py:421] (5/8) Epoch 0, batch 8800, loss[loss=2.766, over 5180.00 frames. , ppl: 15.902773648953824] tot_loss[loss=2.978, over 5388334.90 frames. , ppl: 19.646889986045426], batch size: 70
+2022-12-09 11:54:40,041 INFO [train.py:421] (5/8) Epoch 0, batch 9000, loss[loss=2.813, over 3080.00 frames. , ppl: 16.657472564633185] tot_loss[loss=2.959, over 5415947.50 frames. , ppl: 19.27764145314262], batch size: 70
+2022-12-09 11:54:40,042 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 11:54:40,785 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.745, over 211138.00 frames. , ppl: 15.565240729798518
+2022-12-09 11:56:21,412 INFO [train.py:421] (5/8) Epoch 0, batch 9200, loss[loss=2.644, over 1610.00 frames. , ppl: 14.072281948656013] tot_loss[loss=2.941, over 5426433.87 frames. , ppl: 18.932434122955698], batch size: 70
+2022-12-09 11:58:04,161 INFO [train.py:421] (5/8) Epoch 0, batch 9400, loss[loss=2.695, over 5320.00 frames. , ppl: 14.801519829184896] tot_loss[loss=2.926, over 5400073.01 frames. , ppl: 18.6495606193241], batch size: 70
+2022-12-09 11:59:43,941 INFO [train.py:421] (5/8) Epoch 0, batch 9600, loss[loss=2.783, over 1470.00 frames. , ppl: 16.173651865532147] tot_loss[loss=2.911, over 5387034.31 frames. , ppl: 18.38278430231368], batch size: 70
+2022-12-09 12:01:21,907 INFO [train.py:421] (5/8) Epoch 0, batch 9800, loss[loss=2.856, over 1260.00 frames. , ppl: 17.396077824307646] tot_loss[loss=2.897, over 5371837.71 frames. , ppl: 18.120941009259045], batch size: 70
+2022-12-09 12:03:00,591 INFO [train.py:421] (5/8) Epoch 0, batch 10000, loss[loss=2.702, over 2520.00 frames. , ppl: 14.914563189652945] tot_loss[loss=2.883, over 5373780.82 frames. , ppl: 17.87419865265306], batch size: 70
+2022-12-09 12:03:00,591 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 12:03:01,337 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.716, over 211138.00 frames. , ppl: 15.11237530515154
+2022-12-09 12:04:40,067 INFO [train.py:421] (5/8) Epoch 0, batch 10200, loss[loss=2.776, over 3640.00 frames. , ppl: 16.060229408390974] tot_loss[loss=2.87, over 5385970.68 frames. , ppl: 17.63075077067721], batch size: 70
+2022-12-09 12:06:22,433 INFO [train.py:421] (5/8) Epoch 0, batch 10400, loss[loss=2.708, over 910.00 frames. , ppl: 14.998519141906222] tot_loss[loss=2.857, over 5396375.21 frames. , ppl: 17.404632649215046], batch size: 70
+2022-12-09 12:08:01,855 INFO [train.py:421] (5/8) Epoch 0, batch 10600, loss[loss=2.587, over 2100.00 frames. , ppl: 13.295165642585573] tot_loss[loss=2.844, over 5428191.14 frames. , ppl: 17.19004826919663], batch size: 70
+2022-12-09 12:09:45,721 INFO [train.py:421] (5/8) Epoch 0, batch 10800, loss[loss=2.716, over 2800.00 frames. , ppl: 15.124214735626127] tot_loss[loss=2.832, over 5482874.32 frames. , ppl: 16.984198651785256], batch size: 70
+2022-12-09 12:11:29,859 INFO [train.py:421] (5/8) Epoch 0, batch 11000, loss[loss=2.655, over 8610.00 frames. , ppl: 14.217915853158154] tot_loss[loss=2.824, over 5436932.94 frames. , ppl: 16.8508331220314], batch size: 70
+2022-12-09 12:11:29,859 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 12:11:30,607 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.692, over 211138.00 frames. , ppl: 14.756337129748474
+2022-12-09 12:13:13,813 INFO [train.py:421] (5/8) Epoch 0, batch 11200, loss[loss=2.803, over 1050.00 frames. , ppl: 16.4890223517023] tot_loss[loss=2.816, over 5386939.72 frames. , ppl: 16.70986687652632], batch size: 70
+2022-12-09 12:14:55,302 INFO [train.py:421] (5/8) Epoch 0, batch 11400, loss[loss=2.617, over 4130.00 frames. , ppl: 13.6879381725829] tot_loss[loss=2.804, over 5449926.91 frames. , ppl: 16.518576204877444], batch size: 70
+2022-12-09 12:16:35,945 INFO [train.py:421] (5/8) Epoch 0, batch 11600, loss[loss=2.947, over 910.00 frames. , ppl: 19.05776992305224] tot_loss[loss=2.797, over 5437526.09 frames. , ppl: 16.40198289651038], batch size: 70
+2022-12-09 12:18:18,183 INFO [train.py:421] (5/8) Epoch 0, batch 11800, loss[loss=2.641, over 6580.00 frames. , ppl: 14.027243139702279] tot_loss[loss=2.79, over 5435699.47 frames. , ppl: 16.27541691105066], batch size: 70
+2022-12-09 12:19:58,700 INFO [train.py:421] (5/8) Epoch 0, batch 12000, loss[loss=3.015, over 700.00 frames. , ppl: 20.38479625797094] tot_loss[loss=2.78, over 5477289.27 frames. , ppl: 16.119188193255738], batch size: 70
+2022-12-09 12:19:58,701 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 12:19:59,461 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.67, over 211138.00 frames. , ppl: 14.443332034874045
+2022-12-09 12:21:38,235 INFO [train.py:421] (5/8) Epoch 0, batch 12200, loss[loss=2.796, over 1120.00 frames. , ppl: 16.38326339071302] tot_loss[loss=2.773, over 5472580.24 frames. , ppl: 16.00752573965361], batch size: 70
+2022-12-09 12:23:19,260 INFO [train.py:421] (5/8) Epoch 0, batch 12400, loss[loss=2.626, over 2730.00 frames. , ppl: 13.812484828595561] tot_loss[loss=2.765, over 5480535.26 frames. , ppl: 15.882555379558875], batch size: 70
+2022-12-09 12:25:01,000 INFO [train.py:421] (5/8) Epoch 0, batch 12600, loss[loss=2.734, over 2240.00 frames. , ppl: 15.394107571389181] tot_loss[loss=2.758, over 5504152.65 frames. , ppl: 15.773997493196322], batch size: 70
+2022-12-09 12:26:41,331 INFO [train.py:421] (5/8) Epoch 0, batch 12800, loss[loss=2.615, over 6720.00 frames. , ppl: 13.665005580671616] tot_loss[loss=2.751, over 5527658.58 frames. , ppl: 15.654802476937567], batch size: 70
+2022-12-09 12:28:17,849 INFO [train.py:421] (5/8) Epoch 0, batch 13000, loss[loss=2.601, over 7210.00 frames. , ppl: 13.481197584721563] tot_loss[loss=2.745, over 5510732.97 frames. , ppl: 15.570636049072476], batch size: 70
+2022-12-09 12:28:17,850 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 12:28:18,579 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.655, over 211138.00 frames. , ppl: 14.220370751795132
+2022-12-09 12:29:56,775 INFO [train.py:421] (5/8) Epoch 0, batch 13200, loss[loss=2.674, over 2800.00 frames. , ppl: 14.500078348175437] tot_loss[loss=2.739, over 5546864.66 frames. , ppl: 15.475821059627918], batch size: 70
+2022-12-09 12:31:35,007 INFO [train.py:421] (5/8) Epoch 0, batch 13400, loss[loss=2.69, over 1960.00 frames. , ppl: 14.732122933294844] tot_loss[loss=2.735, over 5516653.27 frames. , ppl: 15.408599495104243], batch size: 70
+2022-12-09 12:33:14,489 INFO [train.py:421] (5/8) Epoch 0, batch 13600, loss[loss=2.686, over 2660.00 frames. , ppl: 14.668303683853676] tot_loss[loss=2.729, over 5533386.88 frames. , ppl: 15.317606156490674], batch size: 70
+2022-12-09 12:34:55,745 INFO [train.py:421] (5/8) Epoch 0, batch 13800, loss[loss=2.515, over 3710.00 frames. , ppl: 12.364344674078556] tot_loss[loss=2.724, over 5511520.77 frames. , ppl: 15.238298554001538], batch size: 70
+2022-12-09 12:36:40,250 INFO [train.py:421] (5/8) Epoch 0, batch 14000, loss[loss=2.708, over 2240.00 frames. , ppl: 15.005973505654392] tot_loss[loss=2.718, over 5550788.09 frames. , ppl: 15.15520002351132], batch size: 70
+2022-12-09 12:36:40,250 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 12:36:41,012 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.637, over 211138.00 frames. , ppl: 13.971456542635757
+2022-12-09 12:38:24,303 INFO [train.py:421] (5/8) Epoch 0, batch 14200, loss[loss=2.756, over 1050.00 frames. , ppl: 15.73264734834605] tot_loss[loss=2.713, over 5547942.78 frames. , ppl: 15.081057260034175], batch size: 70
+2022-12-09 12:40:02,569 INFO [train.py:421] (5/8) Epoch 0, batch 14400, loss[loss=2.678, over 3150.00 frames. , ppl: 14.559378654992765] tot_loss[loss=2.709, over 5535726.20 frames. , ppl: 15.016601473606018], batch size: 70
+2022-12-09 12:41:41,073 INFO [train.py:421] (5/8) Epoch 0, batch 14600, loss[loss=2.819, over 980.00 frames. , ppl: 16.767785277281764] tot_loss[loss=2.706, over 5507976.07 frames. , ppl: 14.96308346715884], batch size: 70
+2022-12-09 12:43:22,501 INFO [train.py:421] (5/8) Epoch 0, batch 14800, loss[loss=2.634, over 2800.00 frames. , ppl: 13.932496393200385] tot_loss[loss=2.702, over 5486899.67 frames. , ppl: 14.905209249104836], batch size: 70
+2022-12-09 12:45:00,526 INFO [train.py:421] (5/8) Epoch 0, batch 15000, loss[loss=2.607, over 7210.00 frames. , ppl: 13.556630433054568] tot_loss[loss=2.699, over 5456931.34 frames. , ppl: 14.864110757952306], batch size: 70
+2022-12-09 12:45:00,526 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 12:45:01,254 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.624, over 211138.00 frames. , ppl: 13.790181361567887
+2022-12-09 12:46:44,176 INFO [train.py:421] (5/8) Epoch 0, batch 15200, loss[loss=2.842, over 1820.00 frames. , ppl: 17.155309067217395] tot_loss[loss=2.695, over 5449609.13 frames. , ppl: 14.804845578851703], batch size: 70
+2022-12-09 12:48:23,217 INFO [train.py:421] (5/8) Epoch 0, batch 15400, loss[loss=2.602, over 4550.00 frames. , ppl: 13.486447170062025] tot_loss[loss=2.692, over 5432719.17 frames. , ppl: 14.757023927318222], batch size: 70
+2022-12-09 12:50:02,334 INFO [train.py:421] (5/8) Epoch 0, batch 15600, loss[loss=2.995, over 840.00 frames. , ppl: 19.982305952412492] tot_loss[loss=2.688, over 5449566.22 frames. , ppl: 14.698949615863663], batch size: 70
+2022-12-09 12:51:41,369 INFO [train.py:421] (5/8) Epoch 0, batch 15800, loss[loss=2.858, over 1190.00 frames. , ppl: 17.430722485280334] tot_loss[loss=2.683, over 5456518.49 frames. , ppl: 14.631792855748314], batch size: 70
+2022-12-09 12:53:18,247 INFO [train.py:421] (5/8) Epoch 0, batch 16000, loss[loss=3.36, over 490.00 frames. , ppl: 28.78113922528711] tot_loss[loss=2.679, over 5481718.29 frames. , ppl: 14.57215230664834], batch size: 70
+2022-12-09 12:53:18,248 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 12:53:18,996 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.614, over 211138.00 frames. , ppl: 13.652930714419472
+2022-12-09 12:54:59,739 INFO [train.py:421] (5/8) Epoch 0, batch 16200, loss[loss=2.716, over 1400.00 frames. , ppl: 15.11949179356127] tot_loss[loss=2.674, over 5535278.03 frames. , ppl: 14.497222547510344], batch size: 70
+2022-12-09 12:56:38,652 INFO [train.py:421] (5/8) Epoch 0, batch 16400, loss[loss=2.726, over 1190.00 frames. , ppl: 15.264482092071445] tot_loss[loss=2.67, over 5556038.79 frames. , ppl: 14.443536770942867], batch size: 70
+2022-12-09 12:58:19,165 INFO [train.py:421] (5/8) Epoch 0, batch 16600, loss[loss=2.698, over 1120.00 frames. , ppl: 14.855295640031787] tot_loss[loss=2.666, over 5582944.71 frames. , ppl: 14.380154135128874], batch size: 70
+2022-12-09 13:00:02,388 INFO [train.py:421] (5/8) Epoch 0, batch 16800, loss[loss=2.602, over 2590.00 frames. , ppl: 13.492183852138393] tot_loss[loss=2.662, over 5632036.26 frames. , ppl: 14.322362257972676], batch size: 70
+2022-12-09 13:01:41,561 INFO [train.py:421] (5/8) Epoch 0, batch 17000, loss[loss=2.714, over 2450.00 frames. , ppl: 15.094769513884568] tot_loss[loss=2.659, over 5622222.58 frames. , ppl: 14.275465060231712], batch size: 70
+2022-12-09 13:01:41,561 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 13:01:42,309 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.601, over 211138.00 frames. , ppl: 13.482567378311968
+2022-12-09 13:03:23,534 INFO [train.py:421] (5/8) Epoch 0, batch 17200, loss[loss=2.671, over 1260.00 frames. , ppl: 14.460758162631993] tot_loss[loss=2.656, over 5579681.13 frames. , ppl: 14.237757865931718], batch size: 70
+2022-12-09 13:05:01,428 INFO [train.py:421] (5/8) Epoch 0, batch 17400, loss[loss=2.643, over 3430.00 frames. , ppl: 14.058977354175132] tot_loss[loss=2.653, over 5575427.97 frames. , ppl: 14.194504569237491], batch size: 70
+2022-12-09 13:06:45,195 INFO [train.py:421] (5/8) Epoch 0, batch 17600, loss[loss=2.545, over 4340.00 frames. , ppl: 12.748914854789499] tot_loss[loss=2.65, over 5583205.46 frames. , ppl: 14.158642087851213], batch size: 70
+2022-12-09 13:08:28,817 INFO [train.py:421] (5/8) Epoch 0, batch 17800, loss[loss=2.578, over 7000.00 frames. , ppl: 13.170123497011744] tot_loss[loss=2.647, over 5612311.25 frames. , ppl: 14.11084021956833], batch size: 70
+2022-12-09 13:10:04,651 INFO [train.py:421] (5/8) Epoch 0, batch 18000, loss[loss=2.672, over 1400.00 frames. , ppl: 14.474372523654065] tot_loss[loss=2.645, over 5555446.04 frames. , ppl: 14.08342186966637], batch size: 70
+2022-12-09 13:10:04,651 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 13:10:05,409 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.595, over 211138.00 frames. , ppl: 13.393325029986839
+2022-12-09 13:11:44,569 INFO [train.py:421] (5/8) Epoch 0, batch 18200, loss[loss=2.557, over 6510.00 frames. , ppl: 12.897293061015699] tot_loss[loss=2.643, over 5537240.70 frames. , ppl: 14.057974802302079], batch size: 70
+2022-12-09 13:13:26,207 INFO [train.py:421] (5/8) Epoch 0, batch 18400, loss[loss=2.675, over 1680.00 frames. , ppl: 14.511915400166927] tot_loss[loss=2.641, over 5516999.38 frames. , ppl: 14.022326365461376], batch size: 70
+2022-12-09 13:15:03,885 INFO [train.py:421] (5/8) Epoch 0, batch 18600, loss[loss=2.622, over 3290.00 frames. , ppl: 13.762760400297482] tot_loss[loss=2.639, over 5469280.37 frames. , ppl: 13.995687773248964], batch size: 70
+2022-12-09 13:16:42,944 INFO [train.py:421] (5/8) Epoch 0, batch 18800, loss[loss=2.594, over 2520.00 frames. , ppl: 13.38736211131872] tot_loss[loss=2.637, over 5449699.95 frames. , ppl: 13.976149936248621], batch size: 70
+2022-12-09 13:18:25,114 INFO [train.py:421] (5/8) Epoch 0, batch 19000, loss[loss=2.543, over 4970.00 frames. , ppl: 12.712737643078645] tot_loss[loss=2.634, over 5497178.83 frames. , ppl: 13.928455460682189], batch size: 70
+2022-12-09 13:18:25,115 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 13:18:25,840 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.582, over 211138.00 frames. , ppl: 13.223891354355008
+2022-12-09 13:20:07,289 INFO [train.py:421] (5/8) Epoch 0, batch 19200, loss[loss=2.557, over 7000.00 frames. , ppl: 12.901088181260809] tot_loss[loss=2.63, over 5512732.25 frames. , ppl: 13.880272613659463], batch size: 70
+2022-12-09 13:21:45,761 INFO [train.py:421] (5/8) Epoch 0, batch 19400, loss[loss=2.71, over 1750.00 frames. , ppl: 15.027426323260745] tot_loss[loss=2.629, over 5475256.42 frames. , ppl: 13.859603314243763], batch size: 70
+2022-12-09 13:23:23,704 INFO [train.py:421] (5/8) Epoch 0, batch 19600, loss[loss=2.629, over 1190.00 frames. , ppl: 13.856283877231794] tot_loss[loss=2.627, over 5466799.04 frames. , ppl: 13.833788398434326], batch size: 70
+2022-12-09 13:25:02,660 INFO [train.py:421] (5/8) Epoch 0, batch 19800, loss[loss=2.572, over 7280.00 frames. , ppl: 13.094336039990946] tot_loss[loss=2.625, over 5471616.00 frames. , ppl: 13.802865444771568], batch size: 70
+2022-12-09 13:26:43,266 INFO [train.py:421] (5/8) Epoch 0, batch 20000, loss[loss=2.6, over 5180.00 frames. , ppl: 13.462433433080639] tot_loss[loss=2.621, over 5525057.82 frames. , ppl: 13.755156293632567], batch size: 70
+2022-12-09 13:26:43,266 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 13:26:44,026 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.574, over 211138.00 frames. , ppl: 13.115298232215096
+2022-12-09 13:28:25,521 INFO [train.py:421] (5/8) Epoch 0, batch 20200, loss[loss=2.665, over 3430.00 frames. , ppl: 14.366661842191435] tot_loss[loss=2.619, over 5548312.41 frames. , ppl: 13.725750755233094], batch size: 70
+2022-12-09 13:30:02,452 INFO [train.py:421] (5/8) Epoch 0, batch 20400, loss[loss=2.56, over 3290.00 frames. , ppl: 12.93650233730424] tot_loss[loss=2.617, over 5536482.10 frames. , ppl: 13.689494462425023], batch size: 70
+2022-12-09 13:31:42,412 INFO [train.py:421] (5/8) Epoch 0, batch 20600, loss[loss=2.811, over 1120.00 frames. , ppl: 16.621830163519242] tot_loss[loss=2.616, over 5484346.30 frames. , ppl: 13.679823478336221], batch size: 70
+2022-12-09 13:33:22,028 INFO [train.py:421] (5/8) Epoch 0, batch 20800, loss[loss=2.579, over 1400.00 frames. , ppl: 13.180385416921704] tot_loss[loss=2.615, over 5492885.66 frames. , ppl: 13.661750141974139], batch size: 70
+2022-12-09 13:35:01,472 INFO [train.py:421] (5/8) Epoch 0, batch 21000, loss[loss=2.718, over 980.00 frames. , ppl: 15.152024717263657] tot_loss[loss=2.612, over 5513226.86 frames. , ppl: 13.629050292343903], batch size: 70
+2022-12-09 13:35:01,472 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 13:35:02,241 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.567, over 211138.00 frames. , ppl: 13.024711781587353
+2022-12-09 13:36:41,742 INFO [train.py:421] (5/8) Epoch 0, batch 21200, loss[loss=2.575, over 3850.00 frames. , ppl: 13.13397540009287] tot_loss[loss=2.61, over 5512495.60 frames. , ppl: 13.605545556399193], batch size: 70
+2022-12-09 13:38:26,895 INFO [train.py:421] (5/8) Epoch 0, batch 21400, loss[loss=2.712, over 1890.00 frames. , ppl: 15.06505076281476] tot_loss[loss=2.608, over 5532126.55 frames. , ppl: 13.577247880886585], batch size: 70
+2022-12-09 13:40:04,883 INFO [train.py:421] (5/8) Epoch 0, batch 21600, loss[loss=2.534, over 8050.00 frames. , ppl: 12.601764697050246] tot_loss[loss=2.607, over 5516957.34 frames. , ppl: 13.560547451084945], batch size: 70
+2022-12-09 13:41:41,329 INFO [train.py:421] (5/8) Epoch 0, batch 21800, loss[loss=2.581, over 3990.00 frames. , ppl: 13.208429367221802] tot_loss[loss=2.606, over 5473720.37 frames. , ppl: 13.550588673510552], batch size: 70
+2022-12-09 13:43:22,212 INFO [train.py:421] (5/8) Epoch 0, batch 22000, loss[loss=2.629, over 1190.00 frames. , ppl: 13.866606931593585] tot_loss[loss=2.604, over 5481643.77 frames. , ppl: 13.516479589723886], batch size: 70
+2022-12-09 13:43:22,213 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 13:43:22,962 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.558, over 211138.00 frames. , ppl: 12.907422572347382
+2022-12-09 13:45:08,010 INFO [train.py:421] (5/8) Epoch 0, batch 22200, loss[loss=2.7, over 1050.00 frames. , ppl: 14.886389790446724] tot_loss[loss=2.602, over 5499297.43 frames. , ppl: 13.496125144893904], batch size: 70
+2022-12-09 13:46:46,076 INFO [train.py:421] (5/8) Epoch 0, batch 22400, loss[loss=2.593, over 3430.00 frames. , ppl: 13.368729440347215] tot_loss[loss=2.601, over 5513003.13 frames. , ppl: 13.471736184654793], batch size: 70
+2022-12-09 13:48:21,209 INFO [train.py:421] (5/8) Epoch 0, batch 22600, loss[loss=2.688, over 980.00 frames. , ppl: 14.706135797334596] tot_loss[loss=2.6, over 5439760.80 frames. , ppl: 13.46413384433882], batch size: 70
+2022-12-09 13:50:00,289 INFO [train.py:421] (5/8) Epoch 0, batch 22800, loss[loss=2.502, over 5950.00 frames. , ppl: 12.201692181021334] tot_loss[loss=2.599, over 5419640.75 frames. , ppl: 13.447421201452595], batch size: 70
+2022-12-09 13:51:39,299 INFO [train.py:421] (5/8) Epoch 0, batch 23000, loss[loss=2.533, over 4200.00 frames. , ppl: 12.584987688230651] tot_loss[loss=2.597, over 5411929.33 frames. , ppl: 13.429221013871055], batch size: 70
+2022-12-09 13:51:39,300 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 13:51:40,059 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.55, over 211138.00 frames. , ppl: 12.811174547900729
+2022-12-09 13:53:19,944 INFO [train.py:421] (5/8) Epoch 0, batch 23200, loss[loss=2.564, over 2240.00 frames. , ppl: 12.992395023071403] tot_loss[loss=2.595, over 5437735.80 frames. , ppl: 13.392921970078248], batch size: 70
+2022-12-09 13:55:01,941 INFO [train.py:421] (5/8) Epoch 0, batch 23400, loss[loss=2.522, over 5460.00 frames. , ppl: 12.448942611867754] tot_loss[loss=2.592, over 5449390.25 frames. , ppl: 13.356277249691695], batch size: 70
+2022-12-09 13:56:43,808 INFO [train.py:421] (5/8) Epoch 0, batch 23600, loss[loss=2.976, over 630.00 frames. , ppl: 19.606681102014154] tot_loss[loss=2.591, over 5438781.58 frames. , ppl: 13.3393504867914], batch size: 70
+2022-12-09 13:58:26,280 INFO [train.py:421] (5/8) Epoch 0, batch 23800, loss[loss=2.715, over 1120.00 frames. , ppl: 15.106839947069314] tot_loss[loss=2.589, over 5436657.57 frames. , ppl: 13.317494437035144], batch size: 70
+2022-12-09 14:00:07,995 INFO [train.py:421] (5/8) Epoch 0, batch 24000, loss[loss=2.471, over 3990.00 frames. , ppl: 11.83474746134186] tot_loss[loss=2.587, over 5496546.31 frames. , ppl: 13.284380007663763], batch size: 70
+2022-12-09 14:00:07,996 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 14:00:08,746 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.543, over 211138.00 frames. , ppl: 12.713002157513388
+2022-12-09 14:01:49,803 INFO [train.py:421] (5/8) Epoch 0, batch 24200, loss[loss=2.68, over 1750.00 frames. , ppl: 14.583160410767732] tot_loss[loss=2.585, over 5480984.52 frames. , ppl: 13.267326567578804], batch size: 70
+2022-12-09 14:03:30,660 INFO [train.py:421] (5/8) Epoch 0, batch 24400, loss[loss=2.613, over 1260.00 frames. , ppl: 13.64218107728201] tot_loss[loss=2.584, over 5494031.09 frames. , ppl: 13.247080495299759], batch size: 70
+2022-12-09 14:05:11,531 INFO [train.py:421] (5/8) Epoch 0, batch 24600, loss[loss=2.932, over 910.00 frames. , ppl: 18.764692698619545] tot_loss[loss=2.583, over 5483821.07 frames. , ppl: 13.239480395931148], batch size: 70
+2022-12-09 14:06:52,052 INFO [train.py:421] (5/8) Epoch 0, batch 24800, loss[loss=2.574, over 2590.00 frames. , ppl: 13.124006303988056] tot_loss[loss=2.582, over 5469057.42 frames. , ppl: 13.223234387582782], batch size: 70
+2022-12-09 14:08:35,041 INFO [train.py:421] (5/8) Epoch 0, batch 25000, loss[loss=2.464, over 3850.00 frames. , ppl: 11.752023831307998] tot_loss[loss=2.58, over 5430717.79 frames. , ppl: 13.20184202021965], batch size: 70
+2022-12-09 14:08:35,041 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 14:08:35,788 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.538, over 211138.00 frames. , ppl: 12.65462202047239
+2022-12-09 14:10:12,802 INFO [train.py:421] (5/8) Epoch 0, batch 25200, loss[loss=3.004, over 700.00 frames. , ppl: 20.15621816978675] tot_loss[loss=2.578, over 5433835.87 frames. , ppl: 13.173211451798881], batch size: 70
+2022-12-09 14:11:51,083 INFO [train.py:421] (5/8) Epoch 0, batch 25400, loss[loss=2.509, over 2170.00 frames. , ppl: 12.295875032079145] tot_loss[loss=2.577, over 5409444.42 frames. , ppl: 13.154573736326608], batch size: 70
+2022-12-09 14:13:29,055 INFO [train.py:421] (5/8) Epoch 0, batch 25600, loss[loss=2.665, over 1260.00 frames. , ppl: 14.365721430471398] tot_loss[loss=2.575, over 5407897.30 frames. , ppl: 13.136962742299175], batch size: 70
+2022-12-09 14:15:06,161 INFO [train.py:421] (5/8) Epoch 0, batch 25800, loss[loss=2.538, over 2730.00 frames. , ppl: 12.650183276085809] tot_loss[loss=2.574, over 5426647.87 frames. , ppl: 13.12259394378442], batch size: 70
+2022-12-09 14:16:45,358 INFO [train.py:421] (5/8) Epoch 0, batch 26000, loss[loss=2.619, over 910.00 frames. , ppl: 13.716297653819913] tot_loss[loss=2.574, over 5377628.06 frames. , ppl: 13.12047008065686], batch size: 70
+2022-12-09 14:16:45,359 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 14:16:46,124 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.534, over 211138.00 frames. , ppl: 12.606376393324691
+2022-12-09 14:18:23,650 INFO [train.py:421] (5/8) Epoch 0, batch 26200, loss[loss=2.317, over 3360.00 frames. , ppl: 10.140957077217797] tot_loss[loss=2.573, over 5390384.03 frames. , ppl: 13.102762476821665], batch size: 70
+2022-12-09 14:20:04,885 INFO [train.py:421] (5/8) Epoch 0, batch 26400, loss[loss=2.653, over 2030.00 frames. , ppl: 14.193726062796642] tot_loss[loss=2.571, over 5414723.76 frames. , ppl: 13.08117364692725], batch size: 70
+2022-12-09 14:21:42,869 INFO [train.py:421] (5/8) Epoch 0, batch 26600, loss[loss=2.539, over 3710.00 frames. , ppl: 12.667269182856398] tot_loss[loss=2.57, over 5430547.50 frames. , ppl: 13.061409328663624], batch size: 70
+2022-12-09 14:23:20,634 INFO [train.py:421] (5/8) Epoch 0, batch 26800, loss[loss=2.604, over 2660.00 frames. , ppl: 13.51679834762773] tot_loss[loss=2.568, over 5473925.10 frames. , ppl: 13.04213091788419], batch size: 70
+2022-12-09 14:25:00,562 INFO [train.py:421] (5/8) Epoch 0, batch 27000, loss[loss=2.634, over 1680.00 frames. , ppl: 13.928559321842007] tot_loss[loss=2.566, over 5469647.39 frames. , ppl: 13.017552071038146], batch size: 70
+2022-12-09 14:25:00,562 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 14:25:01,326 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.527, over 211138.00 frames. , ppl: 12.51153246345996
+2022-12-09 14:26:38,297 INFO [train.py:421] (5/8) Epoch 0, batch 27200, loss[loss=2.626, over 1890.00 frames. , ppl: 13.821425050016767] tot_loss[loss=2.565, over 5479858.48 frames. , ppl: 12.996335226243362], batch size: 70
+2022-12-09 14:28:19,675 INFO [train.py:421] (5/8) Epoch 0, batch 27400, loss[loss=2.577, over 1750.00 frames. , ppl: 13.15994851154395] tot_loss[loss=2.561, over 5530928.86 frames. , ppl: 12.955025227691246], batch size: 70
+2022-12-09 14:30:01,585 INFO [train.py:421] (5/8) Epoch 0, batch 27600, loss[loss=2.548, over 2730.00 frames. , ppl: 12.777892445870945] tot_loss[loss=2.56, over 5492182.85 frames. , ppl: 12.941648443911083], batch size: 70
+2022-12-09 14:31:41,951 INFO [train.py:421] (5/8) Epoch 0, batch 27800, loss[loss=2.56, over 1750.00 frames. , ppl: 12.934875316272164] tot_loss[loss=2.559, over 5480864.05 frames. , ppl: 12.923994458148268], batch size: 70
+2022-12-09 14:33:19,871 INFO [train.py:421] (5/8) Epoch 0, batch 28000, loss[loss=3.162, over 490.00 frames. , ppl: 23.612566232195817] tot_loss[loss=2.559, over 5455367.10 frames. , ppl: 12.924210202849123], batch size: 70
+2022-12-09 14:33:19,871 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 14:33:20,620 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.521, over 211138.00 frames. , ppl: 12.4409233226362
+2022-12-09 14:35:00,358 INFO [train.py:421] (5/8) Epoch 0, batch 28200, loss[loss=2.487, over 2800.00 frames. , ppl: 12.026742870504604] tot_loss[loss=2.559, over 5427099.58 frames. , ppl: 12.921575146221883], batch size: 70
+2022-12-09 14:36:41,343 INFO [train.py:421] (5/8) Epoch 0, batch 28400, loss[loss=2.569, over 1820.00 frames. , ppl: 13.05104989933963] tot_loss[loss=2.558, over 5437359.57 frames. , ppl: 12.908117362356236], batch size: 70
+2022-12-09 14:38:25,477 INFO [train.py:421] (5/8) Epoch 0, batch 28600, loss[loss=2.548, over 2100.00 frames. , ppl: 12.781273262479187] tot_loss[loss=2.557, over 5418317.58 frames. , ppl: 12.895468067925425], batch size: 70
+2022-12-09 14:40:02,956 INFO [train.py:421] (5/8) Epoch 0, batch 28800, loss[loss=2.479, over 5040.00 frames. , ppl: 11.925862812431818] tot_loss[loss=2.555, over 5426834.08 frames. , ppl: 12.872652148855336], batch size: 70
+2022-12-09 14:41:43,067 INFO [train.py:421] (5/8) Epoch 0, batch 29000, loss[loss=2.486, over 3220.00 frames. , ppl: 12.014737552405125] tot_loss[loss=2.555, over 5396119.24 frames. , ppl: 12.872623329199463], batch size: 70
+2022-12-09 14:41:43,068 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 14:41:43,812 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.513, over 211138.00 frames. , ppl: 12.34438956706025
+2022-12-09 14:43:21,607 INFO [train.py:421] (5/8) Epoch 0, batch 29200, loss[loss=2.534, over 3500.00 frames. , ppl: 12.609392393074149] tot_loss[loss=2.553, over 5435815.42 frames. , ppl: 12.846980700547393], batch size: 70
+2022-12-09 14:45:00,642 INFO [train.py:421] (5/8) Epoch 0, batch 29400, loss[loss=2.557, over 2870.00 frames. , ppl: 12.893673311001235] tot_loss[loss=2.552, over 5444263.80 frames. , ppl: 12.834686932340938], batch size: 70
+2022-12-09 14:46:41,805 INFO [train.py:421] (5/8) Epoch 0, batch 29600, loss[loss=2.54, over 4200.00 frames. , ppl: 12.68525316491565] tot_loss[loss=2.549, over 5490780.21 frames. , ppl: 12.797184206099582], batch size: 70
+2022-12-09 14:48:22,706 INFO [train.py:421] (5/8) Epoch 0, batch 29800, loss[loss=2.744, over 1540.00 frames. , ppl: 15.541991542143716] tot_loss[loss=2.55, over 5441383.47 frames. , ppl: 12.806321138978012], batch size: 70
+2022-12-09 14:50:00,643 INFO [train.py:421] (5/8) Epoch 0, batch 30000, loss[loss=2.492, over 910.00 frames. , ppl: 12.08505604174166] tot_loss[loss=2.548, over 5456814.81 frames. , ppl: 12.786853119466734], batch size: 70
+2022-12-09 14:50:00,643 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 14:50:01,403 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.512, over 211138.00 frames. , ppl: 12.331413290958714
+2022-12-09 14:51:44,211 INFO [train.py:421] (5/8) Epoch 0, batch 30200, loss[loss=2.726, over 980.00 frames. , ppl: 15.264945137990702] tot_loss[loss=2.547, over 5465320.33 frames. , ppl: 12.771984022702972], batch size: 70
+2022-12-09 14:53:26,920 INFO [train.py:421] (5/8) Epoch 0, batch 30400, loss[loss=2.949, over 700.00 frames. , ppl: 19.081607030544443] tot_loss[loss=2.545, over 5503173.45 frames. , ppl: 12.740673926654251], batch size: 70
+2022-12-09 14:55:12,237 INFO [train.py:421] (5/8) Epoch 0, batch 30600, loss[loss=2.775, over 630.00 frames. , ppl: 16.041268743636355] tot_loss[loss=2.543, over 5521651.85 frames. , ppl: 12.721107890675668], batch size: 70
+2022-12-09 14:56:52,333 INFO [train.py:421] (5/8) Epoch 0, batch 30800, loss[loss=2.55, over 2310.00 frames. , ppl: 12.80434822153702] tot_loss[loss=2.542, over 5530001.28 frames. , ppl: 12.705502110251878], batch size: 70
+2022-12-09 14:58:33,196 INFO [train.py:421] (5/8) Epoch 0, batch 31000, loss[loss=2.601, over 2590.00 frames. , ppl: 13.47367136677685] tot_loss[loss=2.54, over 5510879.60 frames. , ppl: 12.685738959280025], batch size: 70
+2022-12-09 14:58:33,196 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 14:58:33,957 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.506, over 211138.00 frames. , ppl: 12.259548686709403
+2022-12-09 15:00:11,919 INFO [train.py:421] (5/8) Epoch 0, batch 31200, loss[loss=2.505, over 8050.00 frames. , ppl: 12.24714201427386] tot_loss[loss=2.54, over 5493453.31 frames. , ppl: 12.677971419699153], batch size: 70
+2022-12-09 15:01:49,406 INFO [train.py:421] (5/8) Epoch 0, batch 31400, loss[loss=2.469, over 3780.00 frames. , ppl: 11.810280399040115] tot_loss[loss=2.538, over 5514617.88 frames. , ppl: 12.64905326700141], batch size: 70
+2022-12-09 15:03:31,993 INFO [train.py:421] (5/8) Epoch 0, batch 31600, loss[loss=2.648, over 1330.00 frames. , ppl: 14.130606913664938] tot_loss[loss=2.537, over 5495979.94 frames. , ppl: 12.64663357702659], batch size: 70
+2022-12-09 15:05:12,046 INFO [train.py:421] (5/8) Epoch 0, batch 31800, loss[loss=2.511, over 2450.00 frames. , ppl: 12.322174822720422] tot_loss[loss=2.536, over 5503596.10 frames. , ppl: 12.6335591058983], batch size: 70
+2022-12-09 15:06:52,420 INFO [train.py:421] (5/8) Epoch 0, batch 32000, loss[loss=2.778, over 770.00 frames. , ppl: 16.086408305730902] tot_loss[loss=2.537, over 5479511.79 frames. , ppl: 12.6374019757405], batch size: 70
+2022-12-09 15:06:52,421 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 15:06:53,178 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.502, over 211138.00 frames. , ppl: 12.211804295938897
+2022-12-09 15:08:31,772 INFO [train.py:421] (5/8) Epoch 0, batch 32200, loss[loss=2.549, over 1330.00 frames. , ppl: 12.788108279372084] tot_loss[loss=2.537, over 5493736.43 frames. , ppl: 12.635720815722543], batch size: 70
+2022-12-09 15:10:13,702 INFO [train.py:421] (5/8) Epoch 0, batch 32400, loss[loss=2.619, over 910.00 frames. , ppl: 13.720125279762376] tot_loss[loss=2.535, over 5494318.57 frames. , ppl: 12.614846453739375], batch size: 70
+2022-12-09 15:11:52,318 INFO [train.py:421] (5/8) Epoch 0, batch 32600, loss[loss=2.6, over 1960.00 frames. , ppl: 13.465213932200264] tot_loss[loss=2.534, over 5492350.13 frames. , ppl: 12.604501993298268], batch size: 70
+2022-12-09 15:13:32,200 INFO [train.py:421] (5/8) Epoch 0, batch 32800, loss[loss=2.454, over 2730.00 frames. , ppl: 11.635591723992006] tot_loss[loss=2.533, over 5499426.12 frames. , ppl: 12.592262579877067], batch size: 70
+2022-12-09 15:15:11,964 INFO [train.py:421] (5/8) Epoch 0, batch 33000, loss[loss=2.627, over 1190.00 frames. , ppl: 13.832748795918825] tot_loss[loss=2.533, over 5492110.38 frames. , ppl: 12.585418959341325], batch size: 70
+2022-12-09 15:15:11,964 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 15:15:12,723 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.498, over 211138.00 frames. , ppl: 12.152444554503326
+2022-12-09 15:16:47,661 INFO [train.py:421] (5/8) Epoch 0, batch 33200, loss[loss=2.514, over 3500.00 frames. , ppl: 12.351518584928712] tot_loss[loss=2.532, over 5481720.92 frames. , ppl: 12.575463690820852], batch size: 70
+2022-12-09 15:18:24,732 INFO [train.py:421] (5/8) Epoch 0, batch 33400, loss[loss=2.449, over 4340.00 frames. , ppl: 11.57829899213689] tot_loss[loss=2.529, over 5521942.80 frames. , ppl: 12.536823197832613], batch size: 70
+2022-12-09 15:20:01,826 INFO [train.py:421] (5/8) Epoch 0, batch 33600, loss[loss=2.594, over 1820.00 frames. , ppl: 13.378215316694645] tot_loss[loss=2.528, over 5487704.91 frames. , ppl: 12.534466528391366], batch size: 70
+2022-12-09 15:21:40,604 INFO [train.py:421] (5/8) Epoch 0, batch 33800, loss[loss=2.623, over 910.00 frames. , ppl: 13.770875774311026] tot_loss[loss=2.528, over 5427985.20 frames. , ppl: 12.534371314189388], batch size: 70
+2022-12-09 15:23:22,961 INFO [train.py:421] (5/8) Epoch 0, batch 34000, loss[loss=3.162, over 560.00 frames. , ppl: 23.620583881906313] tot_loss[loss=2.528, over 5410454.90 frames. , ppl: 12.530037505835997], batch size: 70
+2022-12-09 15:23:22,962 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 15:23:23,718 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.494, over 211138.00 frames. , ppl: 12.106278320487071
+2022-12-09 15:25:05,072 INFO [train.py:421] (5/8) Epoch 0, batch 34200, loss[loss=2.659, over 770.00 frames. , ppl: 14.281885479542384] tot_loss[loss=2.527, over 5405028.90 frames. , ppl: 12.51991055923411], batch size: 70
+2022-12-09 15:26:48,264 INFO [train.py:421] (5/8) Epoch 0, batch 34400, loss[loss=2.448, over 5390.00 frames. , ppl: 11.5699018924011] tot_loss[loss=2.527, over 5355342.99 frames. , ppl: 12.513051419307695], batch size: 70
+2022-12-09 15:28:27,299 INFO [train.py:421] (5/8) Epoch 0, batch 34600, loss[loss=2.52, over 2100.00 frames. , ppl: 12.423875572252108] tot_loss[loss=2.526, over 5312614.11 frames. , ppl: 12.50768464745317], batch size: 70
+2022-12-09 15:30:05,467 INFO [train.py:421] (5/8) Epoch 0, batch 34800, loss[loss=2.749, over 1610.00 frames. , ppl: 15.632610808400269] tot_loss[loss=2.527, over 5277007.52 frames. , ppl: 12.510356343986512], batch size: 70
+2022-12-09 15:31:45,504 INFO [train.py:421] (5/8) Epoch 0, batch 35000, loss[loss=2.527, over 2590.00 frames. , ppl: 12.517240848506846] tot_loss[loss=2.526, over 5305260.62 frames. , ppl: 12.500297937362095], batch size: 70
+2022-12-09 15:31:45,504 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 15:31:46,264 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.491, over 211138.00 frames. , ppl: 12.072202954749265
+2022-12-09 15:33:26,023 INFO [train.py:421] (5/8) Epoch 0, batch 35200, loss[loss=2.502, over 980.00 frames. , ppl: 12.202866321663308] tot_loss[loss=2.525, over 5297811.76 frames. , ppl: 12.49264705918772], batch size: 70
+2022-12-09 15:35:05,233 INFO [train.py:421] (5/8) Epoch 0, batch 35400, loss[loss=2.471, over 5530.00 frames. , ppl: 11.838240083732067] tot_loss[loss=2.524, over 5295939.09 frames. , ppl: 12.481296419052846], batch size: 70
+2022-12-09 15:36:46,624 INFO [train.py:421] (5/8) Epoch 0, batch 35600, loss[loss=2.533, over 2030.00 frames. , ppl: 12.588045027183615] tot_loss[loss=2.523, over 5358527.39 frames. , ppl: 12.465361891948108], batch size: 70
+2022-12-09 15:38:27,733 INFO [train.py:421] (5/8) Epoch 0, batch 35800, loss[loss=2.605, over 1120.00 frames. , ppl: 13.532745598091063] tot_loss[loss=2.522, over 5353606.74 frames. , ppl: 12.459072753730872], batch size: 70
+2022-12-09 15:40:08,921 INFO [train.py:421] (5/8) Epoch 0, batch 36000, loss[loss=2.544, over 2660.00 frames. , ppl: 12.732370392337966] tot_loss[loss=2.521, over 5378290.03 frames. , ppl: 12.437898951822962], batch size: 70
+2022-12-09 15:40:08,922 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 15:40:09,682 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.487, over 211138.00 frames. , ppl: 12.026263233378076
+2022-12-09 15:41:50,312 INFO [train.py:421] (5/8) Epoch 0, batch 36200, loss[loss=2.669, over 1190.00 frames. , ppl: 14.419211621454604] tot_loss[loss=2.52, over 5363794.84 frames. , ppl: 12.427305742145514], batch size: 70
+2022-12-09 15:43:29,553 INFO [train.py:421] (5/8) Epoch 0, batch 36400, loss[loss=2.435, over 4200.00 frames. , ppl: 11.415723157248621] tot_loss[loss=2.519, over 5392259.19 frames. , ppl: 12.411819353842404], batch size: 70
+2022-12-09 15:45:09,050 INFO [train.py:421] (5/8) Epoch 0, batch 36600, loss[loss=2.505, over 2870.00 frames. , ppl: 12.240826750037817] tot_loss[loss=2.516, over 5436001.07 frames. , ppl: 12.375828325378547], batch size: 70
+2022-12-09 15:46:47,616 INFO [train.py:421] (5/8) Epoch 0, batch 36800, loss[loss=2.481, over 2730.00 frames. , ppl: 11.951265790479866] tot_loss[loss=2.515, over 5463282.03 frames. , ppl: 12.364412400390307], batch size: 70
+2022-12-09 15:48:30,558 INFO [train.py:421] (5/8) Epoch 0, batch 37000, loss[loss=2.56, over 1260.00 frames. , ppl: 12.934728047363503] tot_loss[loss=2.515, over 5435536.68 frames. , ppl: 12.36486505252047], batch size: 70
+2022-12-09 15:48:30,559 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 15:48:31,316 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.483, over 211138.00 frames. , ppl: 11.975344738756288
+2022-12-09 15:50:06,979 INFO [train.py:421] (5/8) Epoch 0, batch 37200, loss[loss=2.54, over 2380.00 frames. , ppl: 12.674064711012338] tot_loss[loss=2.514, over 5411556.25 frames. , ppl: 12.359012917134038], batch size: 70
+2022-12-09 15:51:48,848 INFO [train.py:421] (5/8) Epoch 0, batch 37400, loss[loss=2.982, over 630.00 frames. , ppl: 19.733208059901802] tot_loss[loss=2.515, over 5392584.31 frames. , ppl: 12.363595782066213], batch size: 70
+2022-12-09 15:53:29,771 INFO [train.py:421] (5/8) Epoch 0, batch 37600, loss[loss=2.668, over 910.00 frames. , ppl: 14.416619063798734] tot_loss[loss=2.513, over 5429399.15 frames. , ppl: 12.345539036434278], batch size: 70
+2022-12-09 15:55:11,151 INFO [train.py:421] (5/8) Epoch 0, batch 37800, loss[loss=2.447, over 1470.00 frames. , ppl: 11.557969963831892] tot_loss[loss=2.512, over 5493718.86 frames. , ppl: 12.32394620813499], batch size: 70
+2022-12-09 15:56:53,633 INFO [train.py:421] (5/8) Epoch 0, batch 38000, loss[loss=2.568, over 1330.00 frames. , ppl: 13.036042237777584] tot_loss[loss=2.51, over 5488974.29 frames. , ppl: 12.303362718014352], batch size: 70
+2022-12-09 15:56:53,633 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 15:56:54,399 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.477, over 211138.00 frames. , ppl: 11.90581072793103
+2022-12-09 15:58:34,659 INFO [train.py:421] (5/8) Epoch 0, batch 38200, loss[loss=2.445, over 1190.00 frames. , ppl: 11.529294001558968] tot_loss[loss=2.509, over 5494860.58 frames. , ppl: 12.29824992959388], batch size: 70
+2022-12-09 16:00:16,309 INFO [train.py:421] (5/8) Epoch 0, batch 38400, loss[loss=2.54, over 2590.00 frames. , ppl: 12.685481554715443] tot_loss[loss=2.507, over 5542482.84 frames. , ppl: 12.269421485856833], batch size: 70
+2022-12-09 16:01:53,959 INFO [train.py:421] (5/8) Epoch 0, batch 38600, loss[loss=2.766, over 700.00 frames. , ppl: 15.889394204028878] tot_loss[loss=2.506, over 5524522.41 frames. , ppl: 12.253918670702332], batch size: 70
+2022-12-09 16:03:33,397 INFO [train.py:421] (5/8) Epoch 0, batch 38800, loss[loss=2.792, over 980.00 frames. , ppl: 16.316571893450558] tot_loss[loss=2.506, over 5524355.63 frames. , ppl: 12.255067659053955], batch size: 70
+2022-12-09 16:05:12,806 INFO [train.py:421] (5/8) Epoch 0, batch 39000, loss[loss=2.618, over 2030.00 frames. , ppl: 13.707162119540959] tot_loss[loss=2.504, over 5569295.90 frames. , ppl: 12.234574275288752], batch size: 70
+2022-12-09 16:05:12,806 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 16:05:13,565 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.476, over 211138.00 frames. , ppl: 11.897084242143393
+2022-12-09 16:06:53,444 INFO [train.py:421] (5/8) Epoch 0, batch 39200, loss[loss=2.424, over 4060.00 frames. , ppl: 11.289502878080274] tot_loss[loss=2.504, over 5589206.38 frames. , ppl: 12.227993750262586], batch size: 70
+2022-12-09 16:08:35,763 INFO [train.py:421] (5/8) Epoch 0, batch 39400, loss[loss=2.507, over 1890.00 frames. , ppl: 12.269867672667687] tot_loss[loss=2.503, over 5572047.54 frames. , ppl: 12.218299821456892], batch size: 70
+2022-12-09 16:10:13,107 INFO [train.py:421] (5/8) Epoch 0, batch 39600, loss[loss=2.63, over 1960.00 frames. , ppl: 13.87586318100866] tot_loss[loss=2.503, over 5553895.09 frames. , ppl: 12.214622588402463], batch size: 70
+2022-12-09 16:11:57,772 INFO [train.py:421] (5/8) Epoch 0, batch 39800, loss[loss=2.412, over 6020.00 frames. , ppl: 11.158358548251838] tot_loss[loss=2.501, over 5599089.61 frames. , ppl: 12.198315754873422], batch size: 70
+2022-12-09 16:13:37,671 INFO [train.py:421] (5/8) Epoch 0, batch 40000, loss[loss=2.487, over 3710.00 frames. , ppl: 12.030742567370423] tot_loss[loss=2.501, over 5576572.94 frames. , ppl: 12.195031169387821], batch size: 70
+2022-12-09 16:13:37,672 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 16:13:38,430 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.473, over 211138.00 frames. , ppl: 11.854175758684555
+2022-12-09 16:15:17,654 INFO [train.py:421] (5/8) Epoch 0, batch 40200, loss[loss=2.66, over 1470.00 frames. , ppl: 14.301263767272422] tot_loss[loss=2.5, over 5581847.76 frames. , ppl: 12.181937870526388], batch size: 70
+2022-12-09 16:16:59,417 INFO [train.py:421] (5/8) Epoch 0, batch 40400, loss[loss=2.743, over 840.00 frames. , ppl: 15.533201952852387] tot_loss[loss=2.498, over 5625482.64 frames. , ppl: 12.156592530717818], batch size: 70
+2022-12-09 16:18:39,717 INFO [train.py:421] (5/8) Epoch 0, batch 40600, loss[loss=2.431, over 3570.00 frames. , ppl: 11.370128203963791] tot_loss[loss=2.497, over 5616659.48 frames. , ppl: 12.150111491256029], batch size: 70
+2022-12-09 16:20:18,282 INFO [train.py:421] (5/8) Epoch 0, batch 40800, loss[loss=2.574, over 2730.00 frames. , ppl: 13.11651735567393] tot_loss[loss=2.496, over 5629613.14 frames. , ppl: 12.128056801362904], batch size: 70
+2022-12-09 16:21:58,938 INFO [train.py:421] (5/8) Epoch 0, batch 41000, loss[loss=2.475, over 11830.00 frames. , ppl: 11.884936447209208] tot_loss[loss=2.496, over 5567604.78 frames. , ppl: 12.139182773949955], batch size: 70
+2022-12-09 16:21:58,938 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 16:21:59,668 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.47, over 211138.00 frames. , ppl: 11.826051100421541
+2022-12-09 16:23:39,477 INFO [train.py:421] (5/8) Epoch 0, batch 41200, loss[loss=2.478, over 3290.00 frames. , ppl: 11.916942787897076] tot_loss[loss=2.496, over 5541571.72 frames. , ppl: 12.136765588309807], batch size: 70
+2022-12-09 16:25:18,888 INFO [train.py:421] (5/8) Epoch 0, batch 41400, loss[loss=2.934, over 840.00 frames. , ppl: 18.80914748493415] tot_loss[loss=2.496, over 5534482.85 frames. , ppl: 12.128198787502876], batch size: 70
+2022-12-09 16:26:57,209 INFO [train.py:421] (5/8) Epoch 0, batch 41600, loss[loss=3.147, over 560.00 frames. , ppl: 23.267868261826873] tot_loss[loss=2.495, over 5534276.40 frames. , ppl: 12.117275980640342], batch size: 70
+2022-12-09 16:28:34,642 INFO [train.py:421] (5/8) Epoch 0, batch 41800, loss[loss=2.436, over 1820.00 frames. , ppl: 11.42709933982708] tot_loss[loss=2.494, over 5560647.71 frames. , ppl: 12.107140515188261], batch size: 70
+2022-12-09 16:30:16,901 INFO [train.py:421] (5/8) Epoch 0, batch 42000, loss[loss=2.843, over 700.00 frames. , ppl: 17.160517627378844] tot_loss[loss=2.493, over 5588491.93 frames. , ppl: 12.10105402048639], batch size: 70
+2022-12-09 16:30:16,901 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 16:30:17,662 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.468, over 211138.00 frames. , ppl: 11.797221378196006
+2022-12-09 16:31:58,193 INFO [train.py:421] (5/8) Epoch 0, batch 42200, loss[loss=3.048, over 560.00 frames. , ppl: 21.081752467565117] tot_loss[loss=2.493, over 5566216.51 frames. , ppl: 12.10037918357813], batch size: 70
+2022-12-09 16:33:38,750 INFO [train.py:421] (5/8) Epoch 0, batch 42400, loss[loss=2.417, over 4620.00 frames. , ppl: 11.214198918267092] tot_loss[loss=2.493, over 5519792.10 frames. , ppl: 12.101732604987758], batch size: 70
+2022-12-09 16:35:21,172 INFO [train.py:421] (5/8) Epoch 0, batch 42600, loss[loss=2.59, over 1960.00 frames. , ppl: 13.332842990147062] tot_loss[loss=2.492, over 5539271.50 frames. , ppl: 12.084975633937415], batch size: 70
+2022-12-09 16:37:02,087 INFO [train.py:421] (5/8) Epoch 0, batch 42800, loss[loss=2.425, over 3500.00 frames. , ppl: 11.306765096410917] tot_loss[loss=2.491, over 5546536.60 frames. , ppl: 12.069309283040363], batch size: 70
+2022-12-09 16:38:43,436 INFO [train.py:421] (5/8) Epoch 0, batch 43000, loss[loss=2.45, over 5110.00 frames. , ppl: 11.585576551786499] tot_loss[loss=2.492, over 5513204.56 frames. , ppl: 12.079479402685159], batch size: 70
+2022-12-09 16:38:43,437 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 16:38:44,197 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.462, over 211138.00 frames. , ppl: 11.728416107172176
+2022-12-09 16:40:25,946 INFO [train.py:421] (5/8) Epoch 0, batch 43200, loss[loss=2.584, over 1680.00 frames. , ppl: 13.244042495111081] tot_loss[loss=2.489, over 5549905.53 frames. , ppl: 12.054980873815827], batch size: 70
+2022-12-09 16:42:04,147 INFO [train.py:421] (5/8) Epoch 0, batch 43400, loss[loss=2.557, over 2450.00 frames. , ppl: 12.89614477421698] tot_loss[loss=2.49, over 5485773.50 frames. , ppl: 12.062884983270699], batch size: 70
+2022-12-09 16:43:44,124 INFO [train.py:421] (5/8) Epoch 0, batch 43600, loss[loss=2.382, over 3220.00 frames. , ppl: 10.825480481242609] tot_loss[loss=2.49, over 5452397.27 frames. , ppl: 12.061802286181614], batch size: 70
+2022-12-09 16:45:29,464 INFO [train.py:421] (5/8) Epoch 0, batch 43800, loss[loss=2.539, over 2030.00 frames. , ppl: 12.665333692218052] tot_loss[loss=2.489, over 5493947.29 frames. , ppl: 12.04565576419923], batch size: 70
+2022-12-09 16:47:12,055 INFO [train.py:421] (5/8) Epoch 0, batch 44000, loss[loss=2.48, over 2170.00 frames. , ppl: 11.936244089233515] tot_loss[loss=2.489, over 5460029.57 frames. , ppl: 12.044824032071185], batch size: 70
+2022-12-09 16:47:12,055 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 16:47:12,815 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.46, over 211138.00 frames. , ppl: 11.703982858703393
+2022-12-09 16:48:52,628 INFO [train.py:421] (5/8) Epoch 0, batch 44200, loss[loss=2.512, over 2730.00 frames. , ppl: 12.33465643022332] tot_loss[loss=2.489, over 5417703.00 frames. , ppl: 12.051224635683466], batch size: 70
+2022-12-09 16:50:32,147 INFO [train.py:421] (5/8) Epoch 0, batch 44400, loss[loss=2.472, over 2800.00 frames. , ppl: 11.851833263356626] tot_loss[loss=2.489, over 5424029.53 frames. , ppl: 12.054380541432357], batch size: 70
+2022-12-09 16:52:13,144 INFO [train.py:421] (5/8) Epoch 0, batch 44600, loss[loss=2.628, over 1820.00 frames. , ppl: 13.848568594532429] tot_loss[loss=2.489, over 5394013.37 frames. , ppl: 12.053669405414375], batch size: 70
+2022-12-09 16:53:54,414 INFO [train.py:421] (5/8) Epoch 0, batch 44800, loss[loss=2.821, over 700.00 frames. , ppl: 16.79084111206714] tot_loss[loss=2.488, over 5392455.10 frames. , ppl: 12.038543351392597], batch size: 70
+2022-12-09 16:55:31,012 INFO [train.py:421] (5/8) Epoch 0, batch 45000, loss[loss=2.539, over 2380.00 frames. , ppl: 12.661626196808825] tot_loss[loss=2.488, over 5398566.70 frames. , ppl: 12.035199874087084], batch size: 70
+2022-12-09 16:55:31,012 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 16:55:31,771 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.459, over 211138.00 frames. , ppl: 11.693083254495477
+2022-12-09 16:57:12,294 INFO [train.py:421] (5/8) Epoch 0, batch 45200, loss[loss=2.615, over 1820.00 frames. , ppl: 13.668490992081159] tot_loss[loss=2.488, over 5398970.66 frames. , ppl: 12.033854306360627], batch size: 70
+2022-12-09 16:58:52,879 INFO [train.py:421] (5/8) Epoch 0, batch 45400, loss[loss=2.554, over 1120.00 frames. , ppl: 12.853317460444558] tot_loss[loss=2.487, over 5413020.49 frames. , ppl: 12.026268280320481], batch size: 70
+2022-12-09 17:00:33,559 INFO [train.py:421] (5/8) Epoch 0, batch 45600, loss[loss=2.456, over 4550.00 frames. , ppl: 11.656786756051812] tot_loss[loss=2.487, over 5400334.51 frames. , ppl: 12.020685992207566], batch size: 70
+2022-12-09 17:02:15,306 INFO [train.py:421] (5/8) Epoch 0, batch 45800, loss[loss=2.745, over 840.00 frames. , ppl: 15.56948315927846] tot_loss[loss=2.486, over 5401394.66 frames. , ppl: 12.0131316821058], batch size: 70
+2022-12-09 17:03:53,494 INFO [train.py:421] (5/8) Epoch 0, batch 46000, loss[loss=3.68, over 420.00 frames. , ppl: 39.639766613797406] tot_loss[loss=2.484, over 5441334.42 frames. , ppl: 11.994160140081046], batch size: 70
+2022-12-09 17:03:53,494 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 17:03:54,254 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.455, over 211138.00 frames. , ppl: 11.65209699143015
+2022-12-09 17:05:29,359 INFO [train.py:421] (5/8) Epoch 0, batch 46200, loss[loss=2.502, over 2240.00 frames. , ppl: 12.200895030707148] tot_loss[loss=2.482, over 5487571.96 frames. , ppl: 11.970286539642775], batch size: 70
+2022-12-09 17:07:09,593 INFO [train.py:421] (5/8) Epoch 0, batch 46400, loss[loss=2.382, over 6230.00 frames. , ppl: 10.824736152939813] tot_loss[loss=2.481, over 5497005.84 frames. , ppl: 11.956594984009941], batch size: 70
+2022-12-09 17:08:51,898 INFO [train.py:421] (5/8) Epoch 0, batch 46600, loss[loss=2.503, over 980.00 frames. , ppl: 12.213075923033767] tot_loss[loss=2.482, over 5478181.94 frames. , ppl: 11.961527228125647], batch size: 70
+2022-12-09 17:10:31,043 INFO [train.py:421] (5/8) Epoch 0, batch 46800, loss[loss=2.427, over 7560.00 frames. , ppl: 11.32107866481751] tot_loss[loss=2.48, over 5492829.46 frames. , ppl: 11.94320380091805], batch size: 70
+2022-12-09 17:12:09,903 INFO [train.py:421] (5/8) Epoch 0, batch 47000, loss[loss=2.614, over 1120.00 frames. , ppl: 13.652962900818379] tot_loss[loss=2.479, over 5497601.72 frames. , ppl: 11.931067654068492], batch size: 70
+2022-12-09 17:12:09,904 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 17:12:10,666 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.453, over 211138.00 frames. , ppl: 11.621817980016578
+2022-12-09 17:13:49,903 INFO [train.py:421] (5/8) Epoch 0, batch 47200, loss[loss=2.442, over 4550.00 frames. , ppl: 11.491928479362642] tot_loss[loss=2.478, over 5534757.76 frames. , ppl: 11.914249718975636], batch size: 70
+2022-12-09 17:15:28,651 INFO [train.py:421] (5/8) Epoch 0, batch 47400, loss[loss=2.719, over 980.00 frames. , ppl: 15.168755953903187] tot_loss[loss=2.477, over 5578270.77 frames. , ppl: 11.90490042647277], batch size: 70
+2022-12-09 17:17:08,453 INFO [train.py:421] (5/8) Epoch 0, batch 47600, loss[loss=2.427, over 4410.00 frames. , ppl: 11.321582577147096] tot_loss[loss=2.476, over 5607156.21 frames. , ppl: 11.894351233612229], batch size: 70
+2022-12-09 17:18:46,707 INFO [train.py:421] (5/8) Epoch 0, batch 47800, loss[loss=2.464, over 2660.00 frames. , ppl: 11.753438525276687] tot_loss[loss=2.477, over 5531964.43 frames. , ppl: 11.904214884081965], batch size: 70
+2022-12-09 17:20:33,545 INFO [train.py:421] (5/8) Epoch 0, batch 48000, loss[loss=3.143, over 560.00 frames. , ppl: 23.161983542528034] tot_loss[loss=2.476, over 5515647.55 frames. , ppl: 11.896551126950351], batch size: 70
+2022-12-09 17:20:33,546 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 17:20:34,309 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.453, over 211138.00 frames. , ppl: 11.621976231748523
+2022-12-09 17:22:13,220 INFO [train.py:421] (5/8) Epoch 0, batch 48200, loss[loss=2.656, over 1820.00 frames. , ppl: 14.24244215312542] tot_loss[loss=2.475, over 5536640.11 frames. , ppl: 11.883757069997925], batch size: 70
+2022-12-09 17:23:50,812 INFO [train.py:421] (5/8) Epoch 0, batch 48400, loss[loss=3.094, over 630.00 frames. , ppl: 22.063117766942742] tot_loss[loss=2.474, over 5552631.86 frames. , ppl: 11.872265043708543], batch size: 70
+2022-12-09 17:25:30,262 INFO [train.py:421] (5/8) Epoch 0, batch 48600, loss[loss=2.451, over 1470.00 frames. , ppl: 11.598143306422315] tot_loss[loss=2.474, over 5549721.99 frames. , ppl: 11.869065004653086], batch size: 70
+2022-12-09 17:27:09,468 INFO [train.py:421] (5/8) Epoch 0, batch 48800, loss[loss=2.452, over 1820.00 frames. , ppl: 11.607190101682558] tot_loss[loss=2.474, over 5500817.72 frames. , ppl: 11.868558917146451], batch size: 70
+2022-12-09 17:28:45,143 INFO [train.py:421] (5/8) Epoch 0, batch 49000, loss[loss=2.675, over 1540.00 frames. , ppl: 14.516565305754643] tot_loss[loss=2.473, over 5505677.23 frames. , ppl: 11.860879080302748], batch size: 70
+2022-12-09 17:28:45,144 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 17:28:45,901 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.449, over 211138.00 frames. , ppl: 11.574441333093757
+2022-12-09 17:30:31,206 INFO [train.py:421] (5/8) Epoch 0, batch 49200, loss[loss=2.48, over 2100.00 frames. , ppl: 11.944241220061627] tot_loss[loss=2.472, over 5496242.23 frames. , ppl: 11.851822112347566], batch size: 70
+2022-12-09 17:32:12,942 INFO [train.py:421] (5/8) Epoch 0, batch 49400, loss[loss=2.391, over 3430.00 frames. , ppl: 10.925037712999233] tot_loss[loss=2.472, over 5550904.56 frames. , ppl: 11.851647939795427], batch size: 70
+2022-12-09 17:33:55,713 INFO [train.py:421] (5/8) Epoch 0, batch 49600, loss[loss=2.623, over 1540.00 frames. , ppl: 13.77876116908032] tot_loss[loss=2.471, over 5575463.90 frames. , ppl: 11.834761575295074], batch size: 70
+2022-12-09 17:35:36,197 INFO [train.py:421] (5/8) Epoch 0, batch 49800, loss[loss=2.953, over 630.00 frames. , ppl: 19.17242772704594] tot_loss[loss=2.471, over 5565003.30 frames. , ppl: 11.82933527306023], batch size: 70
+2022-12-09 17:37:18,309 INFO [train.py:421] (5/8) Epoch 0, batch 50000, loss[loss=2.423, over 3850.00 frames. , ppl: 11.28014768328516] tot_loss[loss=2.47, over 5564029.39 frames. , ppl: 11.820387614972983], batch size: 70
+2022-12-09 17:37:18,309 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 17:37:19,072 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.445, over 211138.00 frames. , ppl: 11.528131384960666
+2022-12-09 17:38:55,888 INFO [train.py:421] (5/8) Epoch 0, batch 50200, loss[loss=2.492, over 2100.00 frames. , ppl: 12.087233737184498] tot_loss[loss=2.47, over 5522175.07 frames. , ppl: 11.828303310853483], batch size: 70
+2022-12-09 17:40:41,944 INFO [train.py:421] (5/8) Epoch 0, batch 50400, loss[loss=2.786, over 910.00 frames. , ppl: 16.22239205946298] tot_loss[loss=2.471, over 5511548.81 frames. , ppl: 11.829136285145562], batch size: 70
+2022-12-09 17:42:23,309 INFO [train.py:421] (5/8) Epoch 0, batch 50600, loss[loss=2.475, over 8890.00 frames. , ppl: 11.88593149166485] tot_loss[loss=2.471, over 5503420.50 frames. , ppl: 11.839221439004133], batch size: 70
+2022-12-09 17:44:03,399 INFO [train.py:421] (5/8) Epoch 0, batch 50800, loss[loss=2.432, over 1260.00 frames. , ppl: 11.384611809891902] tot_loss[loss=2.471, over 5482895.43 frames. , ppl: 11.838776335895798], batch size: 70
+2022-12-09 17:45:43,892 INFO [train.py:421] (5/8) Epoch 0, batch 51000, loss[loss=2.891, over 630.00 frames. , ppl: 18.01562819643931] tot_loss[loss=2.47, over 5496492.48 frames. , ppl: 11.821594361563744], batch size: 70
+2022-12-09 17:45:43,893 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 17:45:44,651 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.442, over 211138.00 frames. , ppl: 11.496775701837747
+2022-12-09 17:47:23,507 INFO [train.py:421] (5/8) Epoch 0, batch 51200, loss[loss=2.57, over 1330.00 frames. , ppl: 13.059552639454044] tot_loss[loss=2.471, over 5482045.54 frames. , ppl: 11.830101479175537], batch size: 70
+2022-12-09 17:49:00,219 INFO [train.py:421] (5/8) Epoch 0, batch 51400, loss[loss=2.384, over 4830.00 frames. , ppl: 10.846771856807056] tot_loss[loss=2.471, over 5461630.27 frames. , ppl: 11.834500012625545], batch size: 70
+2022-12-09 17:50:40,892 INFO [train.py:421] (5/8) Epoch 0, batch 51600, loss[loss=2.361, over 4130.00 frames. , ppl: 10.604718471938478] tot_loss[loss=2.47, over 5462870.28 frames. , ppl: 11.828354170437782], batch size: 70
+2022-12-09 17:52:21,992 INFO [train.py:421] (5/8) Epoch 0, batch 51800, loss[loss=2.416, over 4760.00 frames. , ppl: 11.199929745748593] tot_loss[loss=2.47, over 5447001.69 frames. , ppl: 11.827861740983701], batch size: 70
+2022-12-09 17:54:03,060 INFO [train.py:421] (5/8) Epoch 0, batch 52000, loss[loss=2.75, over 840.00 frames. , ppl: 15.63589552454243] tot_loss[loss=2.47, over 5442913.46 frames. , ppl: 11.825465968564275], batch size: 70
+2022-12-09 17:54:03,061 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 17:54:03,822 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.441, over 211138.00 frames. , ppl: 11.488961168872661
+2022-12-09 17:55:46,383 INFO [train.py:421] (5/8) Epoch 0, batch 52200, loss[loss=2.709, over 1190.00 frames. , ppl: 15.011431944183974] tot_loss[loss=2.469, over 5463578.38 frames. , ppl: 11.813258994688015], batch size: 70
+2022-12-09 17:57:27,048 INFO [train.py:421] (5/8) Epoch 0, batch 52400, loss[loss=2.702, over 1260.00 frames. , ppl: 14.906323524286055] tot_loss[loss=2.468, over 5468195.37 frames. , ppl: 11.803866785018457], batch size: 70
+2022-12-09 17:59:04,826 INFO [train.py:421] (5/8) Epoch 0, batch 52600, loss[loss=2.909, over 700.00 frames. , ppl: 18.335462362658298] tot_loss[loss=2.469, over 5412761.35 frames. , ppl: 11.810040247328752], batch size: 70
+2022-12-09 18:00:44,978 INFO [train.py:421] (5/8) Epoch 0, batch 52800, loss[loss=2.584, over 980.00 frames. , ppl: 13.245709895219813] tot_loss[loss=2.468, over 5435178.77 frames. , ppl: 11.79461430397101], batch size: 70
+2022-12-09 18:02:25,843 INFO [train.py:421] (5/8) Epoch 0, batch 53000, loss[loss=2.42, over 2450.00 frames. , ppl: 11.249697045282335] tot_loss[loss=2.467, over 5429381.88 frames. , ppl: 11.791198298045726], batch size: 70
+2022-12-09 18:02:25,843 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 18:02:26,592 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.439, over 211138.00 frames. , ppl: 11.45966239282214
+2022-12-09 18:04:04,874 INFO [train.py:421] (5/8) Epoch 0, batch 53200, loss[loss=2.399, over 6510.00 frames. , ppl: 11.007957403386401] tot_loss[loss=2.466, over 5432052.13 frames. , ppl: 11.776281427367179], batch size: 70
+2022-12-09 18:05:43,609 INFO [train.py:421] (5/8) Epoch 0, batch 53400, loss[loss=2.519, over 1680.00 frames. , ppl: 12.411310151975739] tot_loss[loss=2.465, over 5437733.40 frames. , ppl: 11.768168049523963], batch size: 70
+2022-12-09 18:07:22,503 INFO [train.py:421] (5/8) Epoch 0, batch 53600, loss[loss=2.373, over 4480.00 frames. , ppl: 10.73436823841262] tot_loss[loss=2.464, over 5451937.69 frames. , ppl: 11.75739705259131], batch size: 70
+2022-12-09 18:09:01,078 INFO [train.py:421] (5/8) Epoch 0, batch 53800, loss[loss=2.426, over 2310.00 frames. , ppl: 11.311715470712288] tot_loss[loss=2.464, over 5444723.42 frames. , ppl: 11.753822529299423], batch size: 70
+2022-12-09 18:10:37,935 INFO [train.py:421] (5/8) Epoch 0, batch 54000, loss[loss=2.385, over 3220.00 frames. , ppl: 10.862976516136616] tot_loss[loss=2.463, over 5463915.61 frames. , ppl: 11.744222623975794], batch size: 70
+2022-12-09 18:10:37,936 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 18:10:38,697 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.437, over 211138.00 frames. , ppl: 11.437813533531314
+2022-12-09 18:12:17,929 INFO [train.py:421] (5/8) Epoch 0, batch 54200, loss[loss=2.573, over 770.00 frames. , ppl: 13.106148027196333] tot_loss[loss=2.463, over 5447904.57 frames. , ppl: 11.741710701784712], batch size: 70
+2022-12-09 18:13:57,231 INFO [train.py:421] (5/8) Epoch 0, batch 54400, loss[loss=2.54, over 1260.00 frames. , ppl: 12.68544488452284] tot_loss[loss=2.464, over 5416471.22 frames. , ppl: 11.751502288459095], batch size: 70
+2022-12-09 18:15:38,992 INFO [train.py:421] (5/8) Epoch 0, batch 54600, loss[loss=2.465, over 1960.00 frames. , ppl: 11.760571884826975] tot_loss[loss=2.464, over 5428259.61 frames. , ppl: 11.74779705904401], batch size: 70
+2022-12-09 18:17:17,688 INFO [train.py:421] (5/8) Epoch 0, batch 54800, loss[loss=2.494, over 2450.00 frames. , ppl: 12.106730762510836] tot_loss[loss=2.462, over 5446440.17 frames. , ppl: 11.732361777663616], batch size: 70
+2022-12-09 18:18:54,917 INFO [train.py:421] (5/8) Epoch 0, batch 55000, loss[loss=2.536, over 2380.00 frames. , ppl: 12.628557063744793] tot_loss[loss=2.462, over 5453588.96 frames. , ppl: 11.727841986266478], batch size: 70
+2022-12-09 18:18:54,918 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 18:18:55,677 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.435, over 211138.00 frames. , ppl: 11.413029548583495
+2022-12-09 18:20:34,862 INFO [train.py:421] (5/8) Epoch 0, batch 55200, loss[loss=2.52, over 2800.00 frames. , ppl: 12.428713702392393] tot_loss[loss=2.461, over 5453807.11 frames. , ppl: 11.715417526426746], batch size: 70
+2022-12-09 18:22:16,325 INFO [train.py:421] (5/8) Epoch 0, batch 55400, loss[loss=2.384, over 4690.00 frames. , ppl: 10.846783705963219] tot_loss[loss=2.459, over 5462134.08 frames. , ppl: 11.697103303018846], batch size: 70
+2022-12-09 18:23:55,809 INFO [train.py:421] (5/8) Epoch 0, batch 55600, loss[loss=2.435, over 5390.00 frames. , ppl: 11.410283982412796] tot_loss[loss=2.459, over 5451851.80 frames. , ppl: 11.695115050116964], batch size: 70
+2022-12-09 18:25:34,331 INFO [train.py:421] (5/8) Epoch 0, batch 55800, loss[loss=2.522, over 2380.00 frames. , ppl: 12.448933574543242] tot_loss[loss=2.459, over 5448470.11 frames. , ppl: 11.688102484458197], batch size: 70
+2022-12-09 18:27:11,392 INFO [train.py:421] (5/8) Epoch 0, batch 56000, loss[loss=2.486, over 2660.00 frames. , ppl: 12.010409842014592] tot_loss[loss=2.459, over 5426134.93 frames. , ppl: 11.690333947325882], batch size: 70
+2022-12-09 18:27:11,393 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 18:27:12,151 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.435, over 211138.00 frames. , ppl: 11.411955259320733
+2022-12-09 18:28:51,286 INFO [train.py:421] (5/8) Epoch 0, batch 56200, loss[loss=2.541, over 1120.00 frames. , ppl: 12.696224554071875] tot_loss[loss=2.458, over 5439760.58 frames. , ppl: 11.676609217698738], batch size: 70
+2022-12-09 18:30:31,855 INFO [train.py:421] (5/8) Epoch 0, batch 56400, loss[loss=2.453, over 2310.00 frames. , ppl: 11.62727100121859] tot_loss[loss=2.457, over 5458686.86 frames. , ppl: 11.66871938843204], batch size: 70
+2022-12-09 18:32:12,558 INFO [train.py:421] (5/8) Epoch 0, batch 56600, loss[loss=2.384, over 4270.00 frames. , ppl: 10.845638204306452] tot_loss[loss=2.455, over 5512955.22 frames. , ppl: 11.651927074185418], batch size: 70
+2022-12-09 18:33:53,495 INFO [train.py:421] (5/8) Epoch 0, batch 56800, loss[loss=2.714, over 910.00 frames. , ppl: 15.09543669130064] tot_loss[loss=2.456, over 5474709.73 frames. , ppl: 11.656193214875406], batch size: 70
+2022-12-09 18:35:37,629 INFO [train.py:421] (5/8) Epoch 0, batch 57000, loss[loss=2.445, over 7630.00 frames. , ppl: 11.53230533612291] tot_loss[loss=2.456, over 5463080.55 frames. , ppl: 11.66161572568387], batch size: 70
+2022-12-09 18:35:37,629 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 18:35:38,358 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.428, over 211138.00 frames. , ppl: 11.336884940977864
+2022-12-09 18:37:20,806 INFO [train.py:421] (5/8) Epoch 0, batch 57200, loss[loss=2.383, over 4270.00 frames. , ppl: 10.838085440251472] tot_loss[loss=2.455, over 5500666.11 frames. , ppl: 11.650711720774604], batch size: 70
+2022-12-09 18:39:00,405 INFO [train.py:421] (5/8) Epoch 0, batch 57400, loss[loss=2.611, over 1400.00 frames. , ppl: 13.614949095666885] tot_loss[loss=2.453, over 5577382.28 frames. , ppl: 11.626073099789586], batch size: 70
+2022-12-09 18:40:42,505 INFO [train.py:421] (5/8) Epoch 0, batch 57600, loss[loss=2.356, over 6440.00 frames. , ppl: 10.553111515284217] tot_loss[loss=2.453, over 5587171.89 frames. , ppl: 11.61932072825182], batch size: 70
+2022-12-09 18:42:20,984 INFO [train.py:421] (5/8) Epoch 0, batch 57800, loss[loss=2.439, over 2870.00 frames. , ppl: 11.459747143509684] tot_loss[loss=2.453, over 5588174.16 frames. , ppl: 11.62010871824522], batch size: 70
+2022-12-09 18:44:01,540 INFO [train.py:421] (5/8) Epoch 0, batch 58000, loss[loss=2.333, over 3990.00 frames. , ppl: 10.312733015052023] tot_loss[loss=2.453, over 5564422.41 frames. , ppl: 11.620014629074467], batch size: 70
+2022-12-09 18:44:01,540 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 18:44:02,301 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.427, over 211138.00 frames. , ppl: 11.325093455985686
+2022-12-09 18:45:45,034 INFO [train.py:421] (5/8) Epoch 0, batch 58200, loss[loss=2.415, over 7210.00 frames. , ppl: 11.189616373171486] tot_loss[loss=2.453, over 5552427.70 frames. , ppl: 11.620689619214088], batch size: 70
+2022-12-09 18:47:28,520 INFO [train.py:421] (5/8) Epoch 0, batch 58400, loss[loss=2.479, over 1750.00 frames. , ppl: 11.924976243854955] tot_loss[loss=2.453, over 5566912.30 frames. , ppl: 11.622478835072108], batch size: 70
+2022-12-09 18:49:11,522 INFO [train.py:421] (5/8) Epoch 0, batch 58600, loss[loss=2.559, over 2170.00 frames. , ppl: 12.926609250166097] tot_loss[loss=2.453, over 5547491.33 frames. , ppl: 11.622050069814653], batch size: 70
+2022-12-09 18:50:48,039 INFO [train.py:421] (5/8) Epoch 0, batch 58800, loss[loss=2.438, over 1960.00 frames. , ppl: 11.451495312893206] tot_loss[loss=2.452, over 5530134.85 frames. , ppl: 11.615542195314834], batch size: 70
+2022-12-09 18:52:28,262 INFO [train.py:421] (5/8) Epoch 0, batch 59000, loss[loss=2.43, over 3920.00 frames. , ppl: 11.36064119487204] tot_loss[loss=2.452, over 5507025.78 frames. , ppl: 11.615357044469054], batch size: 70
+2022-12-09 18:52:28,262 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 18:52:29,026 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.429, over 211138.00 frames. , ppl: 11.349824230730555
+2022-12-09 18:54:09,183 INFO [train.py:421] (5/8) Epoch 0, batch 59200, loss[loss=2.416, over 3430.00 frames. , ppl: 11.20381990572302] tot_loss[loss=2.453, over 5489024.93 frames. , ppl: 11.618956483255522], batch size: 70
+2022-12-09 18:55:49,440 INFO [train.py:421] (5/8) Epoch 0, batch 59400, loss[loss=2.482, over 840.00 frames. , ppl: 11.960847300000625] tot_loss[loss=2.453, over 5478006.96 frames. , ppl: 11.620962983230106], batch size: 70
+2022-12-09 18:57:31,373 INFO [train.py:421] (5/8) Epoch 0, batch 59600, loss[loss=2.388, over 8260.00 frames. , ppl: 10.89668475286379] tot_loss[loss=2.452, over 5488967.18 frames. , ppl: 11.613109876200568], batch size: 70
+2022-12-09 18:59:09,088 INFO [train.py:421] (5/8) Epoch 0, batch 59800, loss[loss=2.451, over 1470.00 frames. , ppl: 11.602050386198494] tot_loss[loss=2.453, over 5425647.54 frames. , ppl: 11.625264420097725], batch size: 70
+2022-12-09 19:00:50,598 INFO [train.py:421] (5/8) Epoch 0, batch 60000, loss[loss=2.53, over 2100.00 frames. , ppl: 12.556953803730552] tot_loss[loss=2.454, over 5382779.11 frames. , ppl: 11.631341506348402], batch size: 70
+2022-12-09 19:00:50,598 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 19:00:51,357 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.426, over 211138.00 frames. , ppl: 11.316449249268606
+2022-12-09 19:02:33,547 INFO [train.py:421] (5/8) Epoch 0, batch 60200, loss[loss=2.471, over 4200.00 frames. , ppl: 11.83868196608631] tot_loss[loss=2.452, over 5417164.22 frames. , ppl: 11.612869245967488], batch size: 70
+2022-12-09 19:04:11,496 INFO [train.py:421] (5/8) Epoch 0, batch 60400, loss[loss=2.356, over 4200.00 frames. , ppl: 10.551903483261952] tot_loss[loss=2.451, over 5439840.20 frames. , ppl: 11.596891675138838], batch size: 70
+2022-12-09 19:05:47,668 INFO [train.py:421] (5/8) Epoch 0, batch 60600, loss[loss=2.645, over 1190.00 frames. , ppl: 14.079499344994137] tot_loss[loss=2.45, over 5443340.09 frames. , ppl: 11.586220120733424], batch size: 70
+2022-12-09 19:07:33,494 INFO [train.py:421] (5/8) Epoch 0, batch 60800, loss[loss=2.561, over 2170.00 frames. , ppl: 12.951388688028032] tot_loss[loss=2.448, over 5506697.34 frames. , ppl: 11.565251947472047], batch size: 70
+2022-12-09 19:09:14,067 INFO [train.py:421] (5/8) Epoch 0, batch 61000, loss[loss=2.699, over 980.00 frames. , ppl: 14.86460791269953] tot_loss[loss=2.446, over 5544165.71 frames. , ppl: 11.546459291260195], batch size: 70
+2022-12-09 19:09:14,068 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 19:09:14,814 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.424, over 211138.00 frames. , ppl: 11.292349414932128
+2022-12-09 19:10:56,125 INFO [train.py:421] (5/8) Epoch 0, batch 61200, loss[loss=2.697, over 980.00 frames. , ppl: 14.8364058390258] tot_loss[loss=2.445, over 5544666.41 frames. , ppl: 11.534928702902137], batch size: 70
+2022-12-09 19:12:36,985 INFO [train.py:421] (5/8) Epoch 0, batch 61400, loss[loss=2.543, over 1400.00 frames. , ppl: 12.714679849396017] tot_loss[loss=2.446, over 5520067.69 frames. , ppl: 11.537604966827006], batch size: 70
+2022-12-09 19:14:17,498 INFO [train.py:421] (5/8) Epoch 0, batch 61600, loss[loss=3.723, over 420.00 frames. , ppl: 41.40723068894303] tot_loss[loss=2.445, over 5524620.40 frames. , ppl: 11.526315922766594], batch size: 70
+2022-12-09 19:16:00,806 INFO [train.py:421] (5/8) Epoch 0, batch 61800, loss[loss=3.294, over 490.00 frames. , ppl: 26.94776790450401] tot_loss[loss=2.443, over 5538443.27 frames. , ppl: 11.511086979981975], batch size: 70
+2022-12-09 19:17:37,942 INFO [train.py:421] (5/8) Epoch 0, batch 62000, loss[loss=2.402, over 7630.00 frames. , ppl: 11.044862768415454] tot_loss[loss=2.443, over 5521028.16 frames. , ppl: 11.512563187460755], batch size: 70
+2022-12-09 19:17:37,942 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 19:17:38,663 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.422, over 211138.00 frames. , ppl: 11.26730360842701
+2022-12-09 19:19:16,403 INFO [train.py:421] (5/8) Epoch 0, batch 62200, loss[loss=2.396, over 3710.00 frames. , ppl: 10.979244455351113] tot_loss[loss=2.444, over 5504766.83 frames. , ppl: 11.516987774004162], batch size: 70
+2022-12-09 19:20:55,497 INFO [train.py:421] (5/8) Epoch 0, batch 62400, loss[loss=2.364, over 3360.00 frames. , ppl: 10.632102279293212] tot_loss[loss=2.444, over 5499196.08 frames. , ppl: 11.520636905030846], batch size: 70
+2022-12-09 19:22:38,366 INFO [train.py:421] (5/8) Epoch 0, batch 62600, loss[loss=2.541, over 1050.00 frames. , ppl: 12.692602522796843] tot_loss[loss=2.443, over 5533650.85 frames. , ppl: 11.510338597149506], batch size: 70
+2022-12-09 19:24:18,394 INFO [train.py:421] (5/8) Epoch 0, batch 62800, loss[loss=2.327, over 6580.00 frames. , ppl: 10.246773958695961] tot_loss[loss=2.443, over 5508927.34 frames. , ppl: 11.512299359364231], batch size: 70
+2022-12-09 19:25:57,774 INFO [train.py:421] (5/8) Epoch 0, batch 63000, loss[loss=2.545, over 2520.00 frames. , ppl: 12.74201529314149] tot_loss[loss=2.443, over 5505871.46 frames. , ppl: 11.513250938061594], batch size: 70
+2022-12-09 19:25:57,775 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 19:25:58,533 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.419, over 211138.00 frames. , ppl: 11.238974307884806
+2022-12-09 19:27:42,798 INFO [train.py:421] (5/8) Epoch 0, batch 63200, loss[loss=2.384, over 3430.00 frames. , ppl: 10.8453404553419] tot_loss[loss=2.442, over 5556030.38 frames. , ppl: 11.501196520475492], batch size: 70
+2022-12-09 19:29:19,322 INFO [train.py:421] (5/8) Epoch 0, batch 63400, loss[loss=2.41, over 3290.00 frames. , ppl: 11.138953933696142] tot_loss[loss=2.441, over 5594368.99 frames. , ppl: 11.479382557621296], batch size: 70
+2022-12-09 19:30:58,088 INFO [train.py:421] (5/8) Epoch 0, batch 63600, loss[loss=2.446, over 2590.00 frames. , ppl: 11.5450048810232] tot_loss[loss=2.44, over 5582818.96 frames. , ppl: 11.477255847322434], batch size: 70
+2022-12-09 19:32:37,332 INFO [train.py:421] (5/8) Epoch 0, batch 63800, loss[loss=2.451, over 3150.00 frames. , ppl: 11.599102615360062] tot_loss[loss=2.439, over 5600328.92 frames. , ppl: 11.464087897525385], batch size: 70
+2022-12-09 19:34:18,414 INFO [train.py:421] (5/8) Epoch 0, batch 64000, loss[loss=2.358, over 3360.00 frames. , ppl: 10.567624780985412] tot_loss[loss=2.44, over 5544564.99 frames. , ppl: 11.476204613730703], batch size: 70
+2022-12-09 19:34:18,415 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 19:34:19,175 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.418, over 211138.00 frames. , ppl: 11.224827241902219
+2022-12-09 19:35:57,968 INFO [train.py:421] (5/8) Epoch 0, batch 64200, loss[loss=2.556, over 1400.00 frames. , ppl: 12.878356295237815] tot_loss[loss=2.441, over 5529355.86 frames. , ppl: 11.482493863574332], batch size: 70
+2022-12-09 19:37:37,733 INFO [train.py:421] (5/8) Epoch 0, batch 64400, loss[loss=3.693, over 420.00 frames. , ppl: 40.1768779606827] tot_loss[loss=2.441, over 5499664.91 frames. , ppl: 11.482480806947704], batch size: 70
+2022-12-09 19:39:14,505 INFO [train.py:421] (5/8) Epoch 0, batch 64600, loss[loss=2.379, over 7700.00 frames. , ppl: 10.797642471405998] tot_loss[loss=2.441, over 5478981.01 frames. , ppl: 11.482484862216532], batch size: 70
+2022-12-09 19:40:56,610 INFO [train.py:421] (5/8) Epoch 0, batch 64800, loss[loss=2.431, over 3710.00 frames. , ppl: 11.365203962876896] tot_loss[loss=2.44, over 5478637.56 frames. , ppl: 11.47452585298387], batch size: 70
+2022-12-09 19:42:37,261 INFO [train.py:421] (5/8) Epoch 0, batch 65000, loss[loss=2.407, over 3220.00 frames. , ppl: 11.096884935620613] tot_loss[loss=2.439, over 5494475.64 frames. , ppl: 11.459177155901909], batch size: 70
+2022-12-09 19:42:37,262 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 19:42:38,022 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.417, over 211138.00 frames. , ppl: 11.211114467296362
+2022-12-09 19:44:16,899 INFO [train.py:421] (5/8) Epoch 0, batch 65200, loss[loss=2.382, over 4620.00 frames. , ppl: 10.827806398509932] tot_loss[loss=2.438, over 5504371.22 frames. , ppl: 11.451367628031983], batch size: 70
+2022-12-09 19:45:56,839 INFO [train.py:421] (5/8) Epoch 0, batch 65400, loss[loss=2.511, over 2240.00 frames. , ppl: 12.316004199113708] tot_loss[loss=2.437, over 5519290.06 frames. , ppl: 11.442896169596098], batch size: 70
+2022-12-09 19:47:35,891 INFO [train.py:421] (5/8) Epoch 0, batch 65600, loss[loss=2.458, over 1750.00 frames. , ppl: 11.679805538085882] tot_loss[loss=2.436, over 5543816.26 frames. , ppl: 11.429737935422132], batch size: 70
+2022-12-09 19:49:13,457 INFO [train.py:421] (5/8) Epoch 0, batch 65800, loss[loss=2.482, over 4060.00 frames. , ppl: 11.962338984218905] tot_loss[loss=2.436, over 5533773.88 frames. , ppl: 11.429001890590435], batch size: 70
+2022-12-09 19:50:53,677 INFO [train.py:421] (5/8) Epoch 0, batch 66000, loss[loss=2.676, over 1120.00 frames. , ppl: 14.521320967141042] tot_loss[loss=2.435, over 5566256.22 frames. , ppl: 11.417200784015389], batch size: 70
+2022-12-09 19:50:53,678 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 19:50:54,438 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.415, over 211138.00 frames. , ppl: 11.192550351531178
+2022-12-09 19:52:35,267 INFO [train.py:421] (5/8) Epoch 0, batch 66200, loss[loss=3.021, over 560.00 frames. , ppl: 20.51309027630907] tot_loss[loss=2.434, over 5579924.89 frames. , ppl: 11.407898639098338], batch size: 70
+2022-12-09 19:54:17,400 INFO [train.py:421] (5/8) Epoch 0, batch 66400, loss[loss=2.473, over 2100.00 frames. , ppl: 11.856097140767774] tot_loss[loss=2.433, over 5626043.33 frames. , ppl: 11.396552389260794], batch size: 70
+2022-12-09 19:56:01,926 INFO [train.py:421] (5/8) Epoch 0, batch 66600, loss[loss=3.056, over 560.00 frames. , ppl: 21.236921102906674] tot_loss[loss=2.434, over 5606273.49 frames. , ppl: 11.399583235451805], batch size: 70
+2022-12-09 19:57:44,179 INFO [train.py:421] (5/8) Epoch 0, batch 66800, loss[loss=3.472, over 490.00 frames. , ppl: 32.216295538466724] tot_loss[loss=2.431, over 5656945.75 frames. , ppl: 11.374152105049484], batch size: 70
+2022-12-09 19:59:25,912 INFO [train.py:421] (5/8) Epoch 0, batch 67000, loss[loss=2.376, over 3360.00 frames. , ppl: 10.757808459150695] tot_loss[loss=2.432, over 5609115.11 frames. , ppl: 11.385081892403331], batch size: 70
+2022-12-09 19:59:25,913 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 19:59:26,673 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.412, over 211138.00 frames. , ppl: 11.1568600153867
+2022-12-09 20:01:08,429 INFO [train.py:421] (5/8) Epoch 0, batch 67200, loss[loss=2.517, over 1330.00 frames. , ppl: 12.39734459846825] tot_loss[loss=2.432, over 5613430.26 frames. , ppl: 11.382931239698738], batch size: 70
+2022-12-09 20:02:47,732 INFO [train.py:421] (5/8) Epoch 0, batch 67400, loss[loss=2.593, over 1680.00 frames. , ppl: 13.366287168159353] tot_loss[loss=2.432, over 5596426.41 frames. , ppl: 11.38469242761162], batch size: 70
+2022-12-09 20:04:25,980 INFO [train.py:421] (5/8) Epoch 0, batch 67600, loss[loss=2.348, over 3920.00 frames. , ppl: 10.46731853011907] tot_loss[loss=2.432, over 5594524.17 frames. , ppl: 11.382459382891321], batch size: 70
+2022-12-09 20:06:07,197 INFO [train.py:421] (5/8) Epoch 0, batch 67800, loss[loss=3.065, over 560.00 frames. , ppl: 21.439046979381946] tot_loss[loss=2.433, over 5577890.18 frames. , ppl: 11.394163242833665], batch size: 70
+2022-12-09 20:07:47,435 INFO [train.py:421] (5/8) Epoch 0, batch 68000, loss[loss=3.069, over 560.00 frames. , ppl: 21.52995339743697] tot_loss[loss=2.433, over 5556042.58 frames. , ppl: 11.39731003164616], batch size: 70
+2022-12-09 20:07:47,436 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 20:07:48,196 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.413, over 211138.00 frames. , ppl: 11.167173836606553
+2022-12-09 20:09:29,900 INFO [train.py:421] (5/8) Epoch 0, batch 68200, loss[loss=2.494, over 1680.00 frames. , ppl: 12.11516541381493] tot_loss[loss=2.433, over 5545472.84 frames. , ppl: 11.398535974086661], batch size: 70
+2022-12-09 20:11:13,456 INFO [train.py:421] (5/8) Epoch 0, batch 68400, loss[loss=2.508, over 1890.00 frames. , ppl: 12.284607130792605] tot_loss[loss=2.434, over 5518228.94 frames. , ppl: 11.402170009654633], batch size: 70
+2022-12-09 20:12:55,461 INFO [train.py:421] (5/8) Epoch 0, batch 68600, loss[loss=2.513, over 1610.00 frames. , ppl: 12.338357461328767] tot_loss[loss=2.433, over 5518882.63 frames. , ppl: 11.396005201884883], batch size: 70
+2022-12-09 20:14:35,033 INFO [train.py:421] (5/8) Epoch 0, batch 68800, loss[loss=2.329, over 10080.00 frames. , ppl: 10.269145353457908] tot_loss[loss=2.435, over 5458983.41 frames. , ppl: 11.413298954815312], batch size: 70
+2022-12-09 20:16:15,817 INFO [train.py:421] (5/8) Epoch 0, batch 69000, loss[loss=2.562, over 2730.00 frames. , ppl: 12.960804984138719] tot_loss[loss=2.434, over 5439379.46 frames. , ppl: 11.409130081047483], batch size: 70
+2022-12-09 20:16:15,817 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 20:16:16,582 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.409, over 211138.00 frames. , ppl: 11.12763404873676
+2022-12-09 20:17:58,588 INFO [train.py:421] (5/8) Epoch 0, batch 69200, loss[loss=2.381, over 1680.00 frames. , ppl: 10.820845112242756] tot_loss[loss=2.433, over 5461663.63 frames. , ppl: 11.3976637047968], batch size: 70
+2022-12-09 20:19:40,454 INFO [train.py:421] (5/8) Epoch 0, batch 69400, loss[loss=2.39, over 2520.00 frames. , ppl: 10.908414950666561] tot_loss[loss=2.435, over 5385251.21 frames. , ppl: 11.4182520488562], batch size: 70
+2022-12-09 20:21:21,257 INFO [train.py:421] (5/8) Epoch 0, batch 69600, loss[loss=2.674, over 1190.00 frames. , ppl: 14.500022867537673] tot_loss[loss=2.436, over 5353239.98 frames. , ppl: 11.422705630761342], batch size: 70
+2022-12-09 20:23:02,999 INFO [train.py:421] (5/8) Epoch 0, batch 69800, loss[loss=2.531, over 1680.00 frames. , ppl: 12.561472819353211] tot_loss[loss=2.434, over 5398786.32 frames. , ppl: 11.406987331198394], batch size: 70
+2022-12-09 20:24:45,759 INFO [train.py:421] (5/8) Epoch 0, batch 70000, loss[loss=2.385, over 2450.00 frames. , ppl: 10.862383046922114] tot_loss[loss=2.433, over 5420235.40 frames. , ppl: 11.39123371776338], batch size: 70
+2022-12-09 20:24:45,759 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 20:24:46,509 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.41, over 211138.00 frames. , ppl: 11.135786255375706
+2022-12-09 20:26:29,616 INFO [train.py:421] (5/8) Epoch 0, batch 70200, loss[loss=2.457, over 2310.00 frames. , ppl: 11.670689187308321] tot_loss[loss=2.433, over 5415519.40 frames. , ppl: 11.39162703456868], batch size: 70
+2022-12-09 20:28:07,706 INFO [train.py:421] (5/8) Epoch 0, batch 70400, loss[loss=2.525, over 1610.00 frames. , ppl: 12.492283731539818] tot_loss[loss=2.433, over 5432558.21 frames. , ppl: 11.391965398374232], batch size: 70
+2022-12-09 20:29:47,709 INFO [train.py:421] (5/8) Epoch 0, batch 70600, loss[loss=2.486, over 1890.00 frames. , ppl: 12.008777180775146] tot_loss[loss=2.433, over 5411908.08 frames. , ppl: 11.39186899545469], batch size: 70
+2022-12-09 20:31:33,136 INFO [train.py:421] (5/8) Epoch 0, batch 70800, loss[loss=2.433, over 4830.00 frames. , ppl: 11.398029531716181] tot_loss[loss=2.432, over 5464252.32 frames. , ppl: 11.377556355603158], batch size: 70
+2022-12-09 20:33:13,806 INFO [train.py:421] (5/8) Epoch 0, batch 71000, loss[loss=2.356, over 4760.00 frames. , ppl: 10.553875250774745] tot_loss[loss=2.431, over 5436099.78 frames. , ppl: 11.374723711390889], batch size: 70
+2022-12-09 20:33:13,807 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 20:33:14,538 INFO [train.py:452] (5/8) Epoch 0, validation: loss=2.406, over 211138.00 frames. , ppl: 11.087587669142305
+2022-12-09 20:34:53,367 INFO [train.py:421] (5/8) Epoch 0, batch 71200, loss[loss=2.46, over 1890.00 frames. , ppl: 11.701832736836089] tot_loss[loss=2.432, over 5427913.11 frames. , ppl: 11.377827643133724], batch size: 70
+2022-12-09 20:36:32,752 INFO [train.py:421] (5/8) Epoch 0, batch 71400, loss[loss=2.539, over 1190.00 frames. , ppl: 12.665546675032955] tot_loss[loss=2.432, over 5380689.91 frames. , ppl: 11.385024456087365], batch size: 70
+2022-12-09 20:38:12,564 INFO [train.py:421] (5/8) Epoch 0, batch 71600, loss[loss=2.42, over 4760.00 frames. , ppl: 11.24507304771547] tot_loss[loss=2.432, over 5382643.88 frames. , ppl: 11.384478036525088], batch size: 70
+2022-12-09 20:39:51,525 INFO [train.py:421] (5/8) Epoch 0, batch 71800, loss[loss=2.407, over 2520.00 frames. , ppl: 11.100689499512505] tot_loss[loss=2.431, over 5389001.01 frames. , ppl: 11.374948141396482], batch size: 70
+2022-12-09 20:41:07,511 INFO [train.py:421] (5/8) Epoch 1, batch 0, loss[loss=2.335, over 4200.00 frames. , ppl: 10.32530573199784] tot_loss[loss=2.335, over 4200.00 frames. , ppl: 10.32530573199784], batch size: 70
+2022-12-09 20:42:47,122 INFO [train.py:421] (5/8) Epoch 1, batch 200, loss[loss=2.479, over 2380.00 frames. , ppl: 11.925526530164513] tot_loss[loss=2.426, over 518480.41 frames. , ppl: 11.311686756140572], batch size: 70
+2022-12-09 20:44:26,254 INFO [train.py:421] (5/8) Epoch 1, batch 400, loss[loss=2.367, over 3080.00 frames. , ppl: 10.660058497041637] tot_loss[loss=2.42, over 999319.15 frames. , ppl: 11.250673298563768], batch size: 70
+2022-12-09 20:46:09,602 INFO [train.py:421] (5/8) Epoch 1, batch 600, loss[loss=2.474, over 1400.00 frames. , ppl: 11.86902906543314] tot_loss[loss=2.422, over 1384807.96 frames. , ppl: 11.26632476273606], batch size: 70
+2022-12-09 20:47:54,142 INFO [train.py:421] (5/8) Epoch 1, batch 800, loss[loss=2.375, over 1890.00 frames. , ppl: 10.750927083213279] tot_loss[loss=2.42, over 1785310.43 frames. , ppl: 11.242443326947747], batch size: 70
+2022-12-09 20:49:35,988 INFO [train.py:421] (5/8) Epoch 1, batch 1000, loss[loss=2.424, over 2870.00 frames. , ppl: 11.295126609044484] tot_loss[loss=2.42, over 2146134.35 frames. , ppl: 11.242053612547322], batch size: 70
+2022-12-09 20:49:35,988 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 20:49:36,749 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.403, over 211138.00 frames. , ppl: 11.059317373352997
+2022-12-09 20:51:14,565 INFO [train.py:421] (5/8) Epoch 1, batch 1200, loss[loss=2.65, over 980.00 frames. , ppl: 14.160358823277233] tot_loss[loss=2.419, over 2466296.51 frames. , ppl: 11.23398008175757], batch size: 70
+2022-12-09 20:52:52,434 INFO [train.py:421] (5/8) Epoch 1, batch 1400, loss[loss=2.336, over 3850.00 frames. , ppl: 10.343287553908773] tot_loss[loss=2.419, over 2742680.11 frames. , ppl: 11.238248992533235], batch size: 70
+2022-12-09 20:54:34,022 INFO [train.py:421] (5/8) Epoch 1, batch 1600, loss[loss=2.389, over 2450.00 frames. , ppl: 10.898224959836552] tot_loss[loss=2.42, over 3005783.43 frames. , ppl: 11.245956092910468], batch size: 70
+2022-12-09 20:56:14,666 INFO [train.py:421] (5/8) Epoch 1, batch 1800, loss[loss=2.572, over 2310.00 frames. , ppl: 13.096354379350103] tot_loss[loss=2.419, over 3248616.26 frames. , ppl: 11.238469536405152], batch size: 70
+2022-12-09 20:57:57,543 INFO [train.py:421] (5/8) Epoch 1, batch 2000, loss[loss=2.399, over 5810.00 frames. , ppl: 11.011931863036436] tot_loss[loss=2.42, over 3434859.38 frames. , ppl: 11.248298877714914], batch size: 70
+2022-12-09 20:57:57,543 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 20:57:58,304 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.404, over 211138.00 frames. , ppl: 11.068564594231407
+2022-12-09 20:59:42,326 INFO [train.py:421] (5/8) Epoch 1, batch 2200, loss[loss=2.54, over 3150.00 frames. , ppl: 12.679690625597019] tot_loss[loss=2.42, over 3640520.59 frames. , ppl: 11.244839267978445], batch size: 70
+2022-12-09 21:01:20,653 INFO [train.py:421] (5/8) Epoch 1, batch 2400, loss[loss=2.535, over 1680.00 frames. , ppl: 12.610651694707185] tot_loss[loss=2.42, over 3793269.31 frames. , ppl: 11.246491401837456], batch size: 70
+2022-12-09 21:03:02,871 INFO [train.py:421] (5/8) Epoch 1, batch 2600, loss[loss=4.174, over 350.00 frames. , ppl: 64.99367123588767] tot_loss[loss=2.421, over 3910206.99 frames. , ppl: 11.255231757737558], batch size: 70
+2022-12-09 21:04:41,925 INFO [train.py:421] (5/8) Epoch 1, batch 2800, loss[loss=2.426, over 1330.00 frames. , ppl: 11.316744942128233] tot_loss[loss=2.42, over 4058146.92 frames. , ppl: 11.250554199309738], batch size: 70
+2022-12-09 21:06:24,186 INFO [train.py:421] (5/8) Epoch 1, batch 3000, loss[loss=2.637, over 2170.00 frames. , ppl: 13.971305838216956] tot_loss[loss=2.42, over 4180453.51 frames. , ppl: 11.25122651440874], batch size: 70
+2022-12-09 21:06:24,187 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 21:06:24,931 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.404, over 211138.00 frames. , ppl: 11.068386028582267
+2022-12-09 21:08:06,436 INFO [train.py:421] (5/8) Epoch 1, batch 3200, loss[loss=2.299, over 3500.00 frames. , ppl: 9.966309744157495] tot_loss[loss=2.421, over 4283188.72 frames. , ppl: 11.257255523560781], batch size: 70
+2022-12-09 21:09:45,260 INFO [train.py:421] (5/8) Epoch 1, batch 3400, loss[loss=3.217, over 490.00 frames. , ppl: 24.941890727019178] tot_loss[loss=2.421, over 4394975.78 frames. , ppl: 11.25290225583801], batch size: 70
+2022-12-09 21:11:24,245 INFO [train.py:421] (5/8) Epoch 1, batch 3600, loss[loss=2.358, over 2030.00 frames. , ppl: 10.568401347331443] tot_loss[loss=2.42, over 4471622.44 frames. , ppl: 11.25130838883594], batch size: 70
+2022-12-09 21:13:06,632 INFO [train.py:421] (5/8) Epoch 1, batch 3800, loss[loss=2.399, over 6440.00 frames. , ppl: 11.008888829438307] tot_loss[loss=2.421, over 4537609.62 frames. , ppl: 11.259577918225899], batch size: 70
+2022-12-09 21:14:40,525 INFO [train.py:421] (5/8) Epoch 1, batch 4000, loss[loss=2.311, over 5810.00 frames. , ppl: 10.082606130121658] tot_loss[loss=2.422, over 4578740.06 frames. , ppl: 11.271139327735263], batch size: 70
+2022-12-09 21:14:40,525 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 21:14:41,285 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.405, over 211138.00 frames. , ppl: 11.079054234738054
+2022-12-09 21:16:21,688 INFO [train.py:421] (5/8) Epoch 1, batch 4200, loss[loss=2.384, over 2590.00 frames. , ppl: 10.848886246113205] tot_loss[loss=2.423, over 4612307.83 frames. , ppl: 11.274612332350818], batch size: 70
+2022-12-09 21:18:00,256 INFO [train.py:421] (5/8) Epoch 1, batch 4400, loss[loss=2.361, over 4690.00 frames. , ppl: 10.606179649300746] tot_loss[loss=2.424, over 4647980.66 frames. , ppl: 11.287198692804921], batch size: 70
+2022-12-09 21:19:45,683 INFO [train.py:421] (5/8) Epoch 1, batch 4600, loss[loss=2.271, over 8400.00 frames. , ppl: 9.685294690940712] tot_loss[loss=2.423, over 4736248.79 frames. , ppl: 11.27862514144206], batch size: 70
+2022-12-09 21:21:25,276 INFO [train.py:421] (5/8) Epoch 1, batch 4800, loss[loss=2.47, over 3080.00 frames. , ppl: 11.826466294044925] tot_loss[loss=2.421, over 4822336.60 frames. , ppl: 11.255650010114982], batch size: 70
+2022-12-09 21:23:05,012 INFO [train.py:421] (5/8) Epoch 1, batch 5000, loss[loss=2.315, over 5180.00 frames. , ppl: 10.125051191383358] tot_loss[loss=2.419, over 4940068.78 frames. , ppl: 11.234546134904399], batch size: 70
+2022-12-09 21:23:05,013 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 21:23:05,774 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.399, over 211138.00 frames. , ppl: 11.009013517670198
+2022-12-09 21:24:42,080 INFO [train.py:421] (5/8) Epoch 1, batch 5200, loss[loss=2.492, over 2170.00 frames. , ppl: 12.083433140024491] tot_loss[loss=2.419, over 5000084.54 frames. , ppl: 11.236249816629579], batch size: 70
+2022-12-09 21:26:20,268 INFO [train.py:421] (5/8) Epoch 1, batch 5400, loss[loss=2.419, over 2240.00 frames. , ppl: 11.232043367657113] tot_loss[loss=2.419, over 5029590.41 frames. , ppl: 11.234025732255555], batch size: 70
+2022-12-09 21:28:00,127 INFO [train.py:421] (5/8) Epoch 1, batch 5600, loss[loss=2.375, over 5460.00 frames. , ppl: 10.753370920487932] tot_loss[loss=2.417, over 5097283.77 frames. , ppl: 11.217286032008573], batch size: 70
+2022-12-09 21:29:38,417 INFO [train.py:421] (5/8) Epoch 1, batch 5800, loss[loss=3.226, over 560.00 frames. , ppl: 25.181394045187442] tot_loss[loss=2.419, over 5093060.54 frames. , ppl: 11.230211931166375], batch size: 70
+2022-12-09 21:31:19,588 INFO [train.py:421] (5/8) Epoch 1, batch 6000, loss[loss=2.382, over 11620.00 frames. , ppl: 10.82149277620896] tot_loss[loss=2.418, over 5131449.26 frames. , ppl: 11.225462721699543], batch size: 70
+2022-12-09 21:31:19,588 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 21:31:20,318 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.4, over 211138.00 frames. , ppl: 11.020588453227733
+2022-12-09 21:33:01,336 INFO [train.py:421] (5/8) Epoch 1, batch 6200, loss[loss=2.739, over 770.00 frames. , ppl: 15.46652817489963] tot_loss[loss=2.418, over 5171305.75 frames. , ppl: 11.227363506747224], batch size: 70
+2022-12-09 21:34:44,217 INFO [train.py:421] (5/8) Epoch 1, batch 6400, loss[loss=2.513, over 1540.00 frames. , ppl: 12.343274042302813] tot_loss[loss=2.419, over 5167591.16 frames. , ppl: 11.236102358844523], batch size: 70
+2022-12-09 21:36:22,197 INFO [train.py:421] (5/8) Epoch 1, batch 6600, loss[loss=2.48, over 2030.00 frames. , ppl: 11.942671335533841] tot_loss[loss=2.419, over 5188513.35 frames. , ppl: 11.235109464416617], batch size: 70
+2022-12-09 21:38:05,940 INFO [train.py:421] (5/8) Epoch 1, batch 6800, loss[loss=2.419, over 2240.00 frames. , ppl: 11.230190088468389] tot_loss[loss=2.418, over 5232282.13 frames. , ppl: 11.223760481324533], batch size: 70
+2022-12-09 21:39:45,054 INFO [train.py:421] (5/8) Epoch 1, batch 7000, loss[loss=2.675, over 840.00 frames. , ppl: 14.508356017471606] tot_loss[loss=2.418, over 5272458.18 frames. , ppl: 11.22232696298809], batch size: 70
+2022-12-09 21:39:45,054 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 21:39:45,806 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.398, over 211138.00 frames. , ppl: 11.000016301250483
+2022-12-09 21:41:28,146 INFO [train.py:421] (5/8) Epoch 1, batch 7200, loss[loss=2.407, over 2170.00 frames. , ppl: 11.104423805839117] tot_loss[loss=2.417, over 5289375.09 frames. , ppl: 11.216652426634038], batch size: 70
+2022-12-09 21:43:10,158 INFO [train.py:421] (5/8) Epoch 1, batch 7400, loss[loss=2.401, over 2660.00 frames. , ppl: 11.029143162503487] tot_loss[loss=2.418, over 5310133.12 frames. , ppl: 11.221139146741859], batch size: 70
+2022-12-09 21:44:48,881 INFO [train.py:421] (5/8) Epoch 1, batch 7600, loss[loss=2.364, over 4270.00 frames. , ppl: 10.637496875357318] tot_loss[loss=2.417, over 5324466.71 frames. , ppl: 11.209526973333244], batch size: 70
+2022-12-09 21:46:31,987 INFO [train.py:421] (5/8) Epoch 1, batch 7800, loss[loss=2.297, over 6510.00 frames. , ppl: 9.940043337499219] tot_loss[loss=2.417, over 5333787.66 frames. , ppl: 11.21119169613692], batch size: 70
+2022-12-09 21:48:11,093 INFO [train.py:421] (5/8) Epoch 1, batch 8000, loss[loss=2.479, over 2380.00 frames. , ppl: 11.933169845453877] tot_loss[loss=2.417, over 5308127.58 frames. , ppl: 11.216114172573203], batch size: 70
+2022-12-09 21:48:11,093 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 21:48:11,857 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.397, over 211138.00 frames. , ppl: 10.99293966703717
+2022-12-09 21:49:52,602 INFO [train.py:421] (5/8) Epoch 1, batch 8200, loss[loss=2.683, over 840.00 frames. , ppl: 14.632651573950303] tot_loss[loss=2.416, over 5329581.25 frames. , ppl: 11.205921239455295], batch size: 70
+2022-12-09 21:51:31,078 INFO [train.py:421] (5/8) Epoch 1, batch 8400, loss[loss=3.733, over 420.00 frames. , ppl: 41.82454237671949] tot_loss[loss=2.416, over 5339071.11 frames. , ppl: 11.201191004149965], batch size: 70
+2022-12-09 21:53:08,554 INFO [train.py:421] (5/8) Epoch 1, batch 8600, loss[loss=2.672, over 770.00 frames. , ppl: 14.469482503297064] tot_loss[loss=2.416, over 5317808.65 frames. , ppl: 11.201941312269543], batch size: 70
+2022-12-09 21:54:48,827 INFO [train.py:421] (5/8) Epoch 1, batch 8800, loss[loss=2.518, over 700.00 frames. , ppl: 12.402686728753357] tot_loss[loss=2.416, over 5330396.28 frames. , ppl: 11.203106844913878], batch size: 70
+2022-12-09 21:56:32,516 INFO [train.py:421] (5/8) Epoch 1, batch 9000, loss[loss=2.582, over 1750.00 frames. , ppl: 13.224831824250558] tot_loss[loss=2.414, over 5389356.21 frames. , ppl: 11.18071736906216], batch size: 70
+2022-12-09 21:56:32,516 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 21:56:33,273 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.396, over 211138.00 frames. , ppl: 10.981821240677286
+2022-12-09 21:58:14,745 INFO [train.py:421] (5/8) Epoch 1, batch 9200, loss[loss=2.533, over 1680.00 frames. , ppl: 12.595959122317035] tot_loss[loss=2.414, over 5375300.96 frames. , ppl: 11.17881772119249], batch size: 70
+2022-12-09 21:59:56,481 INFO [train.py:421] (5/8) Epoch 1, batch 9400, loss[loss=2.344, over 3850.00 frames. , ppl: 10.426600573699526] tot_loss[loss=2.414, over 5379419.22 frames. , ppl: 11.174694289693134], batch size: 70
+2022-12-09 22:01:34,866 INFO [train.py:421] (5/8) Epoch 1, batch 9600, loss[loss=3.044, over 630.00 frames. , ppl: 20.994053901080864] tot_loss[loss=2.413, over 5397766.40 frames. , ppl: 11.167076359813775], batch size: 70
+2022-12-09 22:03:13,418 INFO [train.py:421] (5/8) Epoch 1, batch 9800, loss[loss=2.326, over 2310.00 frames. , ppl: 10.232407997185476] tot_loss[loss=2.412, over 5463170.99 frames. , ppl: 11.154347881249581], batch size: 70
+2022-12-09 22:04:51,393 INFO [train.py:421] (5/8) Epoch 1, batch 10000, loss[loss=2.539, over 1470.00 frames. , ppl: 12.66869826385467] tot_loss[loss=2.412, over 5425909.61 frames. , ppl: 11.159347444060357], batch size: 70
+2022-12-09 22:04:51,394 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 22:04:52,141 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.395, over 211138.00 frames. , ppl: 10.972836503018678
+2022-12-09 22:06:33,532 INFO [train.py:421] (5/8) Epoch 1, batch 10200, loss[loss=2.457, over 1960.00 frames. , ppl: 11.671527440629747] tot_loss[loss=2.411, over 5449063.04 frames. , ppl: 11.145948990123568], batch size: 70
+2022-12-09 22:08:13,474 INFO [train.py:421] (5/8) Epoch 1, batch 10400, loss[loss=2.306, over 5530.00 frames. , ppl: 10.033138432451945] tot_loss[loss=2.409, over 5510785.20 frames. , ppl: 11.125946486563015], batch size: 70
+2022-12-09 22:09:54,345 INFO [train.py:421] (5/8) Epoch 1, batch 10600, loss[loss=2.346, over 3500.00 frames. , ppl: 10.441607535603929] tot_loss[loss=2.409, over 5510733.91 frames. , ppl: 11.127323304821202], batch size: 70
+2022-12-09 22:11:32,870 INFO [train.py:421] (5/8) Epoch 1, batch 10800, loss[loss=2.424, over 4340.00 frames. , ppl: 11.287915389346496] tot_loss[loss=2.41, over 5499353.68 frames. , ppl: 11.131103397061723], batch size: 70
+2022-12-09 22:13:12,635 INFO [train.py:421] (5/8) Epoch 1, batch 11000, loss[loss=2.34, over 1540.00 frames. , ppl: 10.378007082335294] tot_loss[loss=2.41, over 5508963.19 frames. , ppl: 11.134325697230347], batch size: 70
+2022-12-09 22:13:12,636 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 22:13:13,393 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.394, over 211138.00 frames. , ppl: 10.959451037049323
+2022-12-09 22:14:51,704 INFO [train.py:421] (5/8) Epoch 1, batch 11200, loss[loss=2.512, over 1470.00 frames. , ppl: 12.329485832821863] tot_loss[loss=2.41, over 5516059.98 frames. , ppl: 11.133022979271031], batch size: 70
+2022-12-09 22:16:32,959 INFO [train.py:421] (5/8) Epoch 1, batch 11400, loss[loss=2.329, over 4410.00 frames. , ppl: 10.270253433244742] tot_loss[loss=2.409, over 5530497.99 frames. , ppl: 11.126026955167223], batch size: 70
+2022-12-09 22:18:09,022 INFO [train.py:421] (5/8) Epoch 1, batch 11600, loss[loss=2.38, over 2310.00 frames. , ppl: 10.800561110531461] tot_loss[loss=2.409, over 5532167.44 frames. , ppl: 11.127992460906624], batch size: 70
+2022-12-09 22:19:53,238 INFO [train.py:421] (5/8) Epoch 1, batch 11800, loss[loss=2.504, over 1190.00 frames. , ppl: 12.232031802179053] tot_loss[loss=2.409, over 5538787.28 frames. , ppl: 11.12798564897679], batch size: 70
+2022-12-09 22:21:30,851 INFO [train.py:421] (5/8) Epoch 1, batch 12000, loss[loss=2.554, over 2450.00 frames. , ppl: 12.861040261938344] tot_loss[loss=2.409, over 5547264.78 frames. , ppl: 11.12386613188088], batch size: 70
+2022-12-09 22:21:30,852 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 22:21:31,611 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.393, over 211138.00 frames. , ppl: 10.943533776107909
+2022-12-09 22:23:19,195 INFO [train.py:421] (5/8) Epoch 1, batch 12200, loss[loss=2.314, over 3920.00 frames. , ppl: 10.11299367744459] tot_loss[loss=2.409, over 5567362.00 frames. , ppl: 11.11804015641683], batch size: 70
+2022-12-09 22:24:55,362 INFO [train.py:421] (5/8) Epoch 1, batch 12400, loss[loss=2.513, over 1400.00 frames. , ppl: 12.34338714671072] tot_loss[loss=2.408, over 5565354.84 frames. , ppl: 11.10958157373759], batch size: 70
+2022-12-09 22:26:36,489 INFO [train.py:421] (5/8) Epoch 1, batch 12600, loss[loss=2.339, over 3500.00 frames. , ppl: 10.37553192644639] tot_loss[loss=2.408, over 5542469.53 frames. , ppl: 11.112956365020802], batch size: 70
+2022-12-09 22:28:14,893 INFO [train.py:421] (5/8) Epoch 1, batch 12800, loss[loss=2.728, over 770.00 frames. , ppl: 15.295379570473479] tot_loss[loss=2.408, over 5540391.50 frames. , ppl: 11.115447093553403], batch size: 70
+2022-12-09 22:29:53,722 INFO [train.py:421] (5/8) Epoch 1, batch 13000, loss[loss=2.904, over 630.00 frames. , ppl: 18.248734764050592] tot_loss[loss=2.409, over 5505799.64 frames. , ppl: 11.125593182335392], batch size: 70
+2022-12-09 22:29:53,722 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 22:29:54,481 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.391, over 211138.00 frames. , ppl: 10.927798136667734
+2022-12-09 22:31:36,554 INFO [train.py:421] (5/8) Epoch 1, batch 13200, loss[loss=2.43, over 2170.00 frames. , ppl: 11.355974335513364] tot_loss[loss=2.409, over 5486459.24 frames. , ppl: 11.124601662126192], batch size: 70
+2022-12-09 22:33:15,727 INFO [train.py:421] (5/8) Epoch 1, batch 13400, loss[loss=2.664, over 1050.00 frames. , ppl: 14.352256540605525] tot_loss[loss=2.408, over 5514244.44 frames. , ppl: 11.116229765194772], batch size: 70
+2022-12-09 22:34:54,052 INFO [train.py:421] (5/8) Epoch 1, batch 13600, loss[loss=2.41, over 1610.00 frames. , ppl: 11.13473849103746] tot_loss[loss=2.408, over 5493359.47 frames. , ppl: 11.11443899220638], batch size: 70
+2022-12-09 22:36:30,646 INFO [train.py:421] (5/8) Epoch 1, batch 13800, loss[loss=2.374, over 2590.00 frames. , ppl: 10.734968536096307] tot_loss[loss=2.409, over 5480792.00 frames. , ppl: 11.11982069334975], batch size: 70
+2022-12-09 22:38:12,925 INFO [train.py:421] (5/8) Epoch 1, batch 14000, loss[loss=2.386, over 4200.00 frames. , ppl: 10.874637267721486] tot_loss[loss=2.407, over 5538449.83 frames. , ppl: 11.102117489488952], batch size: 70
+2022-12-09 22:38:12,925 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 22:38:13,682 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.39, over 211138.00 frames. , ppl: 10.914509612851669
+2022-12-09 22:39:52,639 INFO [train.py:421] (5/8) Epoch 1, batch 14200, loss[loss=2.483, over 1610.00 frames. , ppl: 11.974055776364606] tot_loss[loss=2.407, over 5530898.30 frames. , ppl: 11.103198095663144], batch size: 70
+2022-12-09 22:41:35,427 INFO [train.py:421] (5/8) Epoch 1, batch 14400, loss[loss=2.354, over 1260.00 frames. , ppl: 10.531627846474008] tot_loss[loss=2.407, over 5567216.38 frames. , ppl: 11.100383375553916], batch size: 70
+2022-12-09 22:43:15,960 INFO [train.py:421] (5/8) Epoch 1, batch 14600, loss[loss=4.103, over 350.00 frames. , ppl: 60.544242102645114] tot_loss[loss=2.408, over 5524229.69 frames. , ppl: 11.111645863908976], batch size: 70
+2022-12-09 22:44:53,639 INFO [train.py:421] (5/8) Epoch 1, batch 14800, loss[loss=2.485, over 2170.00 frames. , ppl: 11.995581341727123] tot_loss[loss=2.409, over 5474170.28 frames. , ppl: 11.12725120451949], batch size: 70
+2022-12-09 22:46:32,361 INFO [train.py:421] (5/8) Epoch 1, batch 15000, loss[loss=2.381, over 5320.00 frames. , ppl: 10.815709041458947] tot_loss[loss=2.41, over 5433919.88 frames. , ppl: 11.135248243577442], batch size: 70
+2022-12-09 22:46:32,361 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 22:46:33,092 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.389, over 211138.00 frames. , ppl: 10.905875331584172
+2022-12-09 22:48:15,334 INFO [train.py:421] (5/8) Epoch 1, batch 15200, loss[loss=2.345, over 7210.00 frames. , ppl: 10.435657250625177] tot_loss[loss=2.409, over 5510353.33 frames. , ppl: 11.120440180657019], batch size: 70
+2022-12-09 22:49:57,178 INFO [train.py:421] (5/8) Epoch 1, batch 15400, loss[loss=2.303, over 3570.00 frames. , ppl: 10.008110471765685] tot_loss[loss=2.407, over 5557252.39 frames. , ppl: 11.09775652243574], batch size: 70
+2022-12-09 22:51:38,862 INFO [train.py:421] (5/8) Epoch 1, batch 15600, loss[loss=2.394, over 4620.00 frames. , ppl: 10.962270053732656] tot_loss[loss=2.408, over 5488699.98 frames. , ppl: 11.107540193067623], batch size: 70
+2022-12-09 22:53:18,733 INFO [train.py:421] (5/8) Epoch 1, batch 15800, loss[loss=2.54, over 2030.00 frames. , ppl: 12.68157850592673] tot_loss[loss=2.407, over 5494090.97 frames. , ppl: 11.105979699937262], batch size: 70
+2022-12-09 22:55:01,063 INFO [train.py:421] (5/8) Epoch 1, batch 16000, loss[loss=2.445, over 2730.00 frames. , ppl: 11.536068054651123] tot_loss[loss=2.406, over 5538451.27 frames. , ppl: 11.093753636509998], batch size: 70
+2022-12-09 22:55:01,064 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 22:55:01,823 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.387, over 211138.00 frames. , ppl: 10.886231072245774
+2022-12-09 22:56:40,411 INFO [train.py:421] (5/8) Epoch 1, batch 16200, loss[loss=2.359, over 5250.00 frames. , ppl: 10.58038941160665] tot_loss[loss=2.406, over 5530045.04 frames. , ppl: 11.084143794693084], batch size: 70
+2022-12-09 22:58:20,171 INFO [train.py:421] (5/8) Epoch 1, batch 16400, loss[loss=2.611, over 1820.00 frames. , ppl: 13.614943251339028] tot_loss[loss=2.405, over 5525668.88 frames. , ppl: 11.077139798467861], batch size: 70
+2022-12-09 23:00:00,604 INFO [train.py:421] (5/8) Epoch 1, batch 16600, loss[loss=2.406, over 3150.00 frames. , ppl: 11.086335962519744] tot_loss[loss=2.405, over 5527455.14 frames. , ppl: 11.073752885635333], batch size: 70
+2022-12-09 23:01:39,180 INFO [train.py:421] (5/8) Epoch 1, batch 16800, loss[loss=2.399, over 1750.00 frames. , ppl: 11.017051362152008] tot_loss[loss=2.403, over 5547521.08 frames. , ppl: 11.060568899442872], batch size: 70
+2022-12-09 23:03:20,369 INFO [train.py:421] (5/8) Epoch 1, batch 17000, loss[loss=2.347, over 7070.00 frames. , ppl: 10.456197905933225] tot_loss[loss=2.403, over 5562218.54 frames. , ppl: 11.058889612252727], batch size: 70
+2022-12-09 23:03:20,369 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 23:03:21,099 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.388, over 211138.00 frames. , ppl: 10.88762972140702
+2022-12-09 23:04:58,074 INFO [train.py:421] (5/8) Epoch 1, batch 17200, loss[loss=2.503, over 1400.00 frames. , ppl: 12.223407344750019] tot_loss[loss=2.404, over 5533091.51 frames. , ppl: 11.066659471813997], batch size: 70
+2022-12-09 23:06:36,552 INFO [train.py:421] (5/8) Epoch 1, batch 17400, loss[loss=2.562, over 1120.00 frames. , ppl: 12.961372288505924] tot_loss[loss=2.403, over 5580832.40 frames. , ppl: 11.054690039525427], batch size: 70
+2022-12-09 23:08:16,407 INFO [train.py:421] (5/8) Epoch 1, batch 17600, loss[loss=2.373, over 1680.00 frames. , ppl: 10.724646431763654] tot_loss[loss=2.404, over 5550095.65 frames. , ppl: 11.062473288727702], batch size: 70
+2022-12-09 23:09:57,151 INFO [train.py:421] (5/8) Epoch 1, batch 17800, loss[loss=3.623, over 420.00 frames. , ppl: 37.45307168940434] tot_loss[loss=2.405, over 5503266.09 frames. , ppl: 11.075015432428257], batch size: 70
+2022-12-09 23:11:36,545 INFO [train.py:421] (5/8) Epoch 1, batch 18000, loss[loss=2.942, over 630.00 frames. , ppl: 18.957391232912137] tot_loss[loss=2.403, over 5525440.16 frames. , ppl: 11.060279022220485], batch size: 70
+2022-12-09 23:11:36,545 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 23:11:37,293 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.386, over 211138.00 frames. , ppl: 10.869570686577362
+2022-12-09 23:13:24,105 INFO [train.py:421] (5/8) Epoch 1, batch 18200, loss[loss=2.327, over 7980.00 frames. , ppl: 10.243278456176503] tot_loss[loss=2.403, over 5573089.21 frames. , ppl: 11.051885244130075], batch size: 70
+2022-12-09 23:15:06,589 INFO [train.py:421] (5/8) Epoch 1, batch 18400, loss[loss=2.447, over 4900.00 frames. , ppl: 11.556120412798252] tot_loss[loss=2.403, over 5578813.59 frames. , ppl: 11.051972670395061], batch size: 70
+2022-12-09 23:16:40,257 INFO [train.py:421] (5/8) Epoch 1, batch 18600, loss[loss=2.377, over 2940.00 frames. , ppl: 10.773695714812215] tot_loss[loss=2.404, over 5550182.32 frames. , ppl: 11.066274450907239], batch size: 70
+2022-12-09 23:18:19,348 INFO [train.py:421] (5/8) Epoch 1, batch 18800, loss[loss=2.404, over 1820.00 frames. , ppl: 11.071831603746315] tot_loss[loss=2.404, over 5509722.36 frames. , ppl: 11.071258590286394], batch size: 70
+2022-12-09 23:19:58,914 INFO [train.py:421] (5/8) Epoch 1, batch 19000, loss[loss=2.494, over 3430.00 frames. , ppl: 12.109410704586114] tot_loss[loss=2.404, over 5511498.06 frames. , ppl: 11.063498994564924], batch size: 70
+2022-12-09 23:19:58,914 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 23:19:59,659 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.385, over 211138.00 frames. , ppl: 10.86154424255148
+2022-12-09 23:21:39,313 INFO [train.py:421] (5/8) Epoch 1, batch 19200, loss[loss=4.266, over 350.00 frames. , ppl: 71.2490113369857] tot_loss[loss=2.404, over 5482704.41 frames. , ppl: 11.063388900670674], batch size: 70
+2022-12-09 23:23:22,550 INFO [train.py:421] (5/8) Epoch 1, batch 19400, loss[loss=2.411, over 1610.00 frames. , ppl: 11.144079664457694] tot_loss[loss=2.404, over 5485715.55 frames. , ppl: 11.062082477752245], batch size: 70
+2022-12-09 23:25:00,345 INFO [train.py:421] (5/8) Epoch 1, batch 19600, loss[loss=2.309, over 7000.00 frames. , ppl: 10.066075395461175] tot_loss[loss=2.403, over 5505934.65 frames. , ppl: 11.051095639290104], batch size: 70
+2022-12-09 23:26:38,753 INFO [train.py:421] (5/8) Epoch 1, batch 19800, loss[loss=2.292, over 4130.00 frames. , ppl: 9.89350237393249] tot_loss[loss=2.402, over 5519372.05 frames. , ppl: 11.050733041495244], batch size: 70
+2022-12-09 23:28:20,837 INFO [train.py:421] (5/8) Epoch 1, batch 20000, loss[loss=2.339, over 3430.00 frames. , ppl: 10.374810575083638] tot_loss[loss=2.404, over 5478164.65 frames. , ppl: 11.064155858747057], batch size: 70
+2022-12-09 23:28:20,838 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 23:28:21,595 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.383, over 211138.00 frames. , ppl: 10.840206792981473
+2022-12-09 23:30:07,793 INFO [train.py:421] (5/8) Epoch 1, batch 20200, loss[loss=2.636, over 980.00 frames. , ppl: 13.962719766209199] tot_loss[loss=2.403, over 5513676.78 frames. , ppl: 11.054591822696791], batch size: 70
+2022-12-09 23:31:46,079 INFO [train.py:421] (5/8) Epoch 1, batch 20400, loss[loss=2.529, over 1470.00 frames. , ppl: 12.54223682709384] tot_loss[loss=2.404, over 5467205.82 frames. , ppl: 11.068907533279205], batch size: 70
+2022-12-09 23:33:26,950 INFO [train.py:421] (5/8) Epoch 1, batch 20600, loss[loss=2.389, over 5530.00 frames. , ppl: 10.897636858711948] tot_loss[loss=2.403, over 5481781.13 frames. , ppl: 11.059674769521289], batch size: 70
+2022-12-09 23:35:04,516 INFO [train.py:421] (5/8) Epoch 1, batch 20800, loss[loss=2.321, over 3150.00 frames. , ppl: 10.180899765820767] tot_loss[loss=2.403, over 5487433.65 frames. , ppl: 11.057833154814237], batch size: 70
+2022-12-09 23:36:44,985 INFO [train.py:421] (5/8) Epoch 1, batch 21000, loss[loss=2.456, over 2450.00 frames. , ppl: 11.660954213084961] tot_loss[loss=2.403, over 5489894.76 frames. , ppl: 11.061675700321928], batch size: 70
+2022-12-09 23:36:44,985 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 23:36:45,739 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.381, over 211138.00 frames. , ppl: 10.817335716309938
+2022-12-09 23:38:28,716 INFO [train.py:421] (5/8) Epoch 1, batch 21200, loss[loss=2.78, over 770.00 frames. , ppl: 16.12167059901114] tot_loss[loss=2.403, over 5495911.04 frames. , ppl: 11.052759325377894], batch size: 70
+2022-12-09 23:40:05,767 INFO [train.py:421] (5/8) Epoch 1, batch 21400, loss[loss=2.595, over 1750.00 frames. , ppl: 13.40203083065324] tot_loss[loss=2.403, over 5484628.85 frames. , ppl: 11.058580219877962], batch size: 70
+2022-12-09 23:41:45,963 INFO [train.py:421] (5/8) Epoch 1, batch 21600, loss[loss=2.404, over 2730.00 frames. , ppl: 11.064524797811211] tot_loss[loss=2.404, over 5480662.40 frames. , ppl: 11.066412026544159], batch size: 70
+2022-12-09 23:43:25,698 INFO [train.py:421] (5/8) Epoch 1, batch 21800, loss[loss=2.533, over 1260.00 frames. , ppl: 12.58630649673554] tot_loss[loss=2.404, over 5459684.86 frames. , ppl: 11.068111411010465], batch size: 70
+2022-12-09 23:45:06,450 INFO [train.py:421] (5/8) Epoch 1, batch 22000, loss[loss=2.338, over 3080.00 frames. , ppl: 10.364973335115229] tot_loss[loss=2.402, over 5485848.79 frames. , ppl: 11.05056206570755], batch size: 70
+2022-12-09 23:45:06,450 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 23:45:07,208 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.382, over 211138.00 frames. , ppl: 10.821137268370757
+2022-12-09 23:46:46,113 INFO [train.py:421] (5/8) Epoch 1, batch 22200, loss[loss=3.577, over 420.00 frames. , ppl: 35.768251242495204] tot_loss[loss=2.401, over 5499931.09 frames. , ppl: 11.033085200275716], batch size: 70
+2022-12-09 23:48:27,624 INFO [train.py:421] (5/8) Epoch 1, batch 22400, loss[loss=2.903, over 560.00 frames. , ppl: 18.23662870194056] tot_loss[loss=2.401, over 5514164.93 frames. , ppl: 11.029368699864532], batch size: 70
+2022-12-09 23:50:05,039 INFO [train.py:421] (5/8) Epoch 1, batch 22600, loss[loss=4.281, over 350.00 frames. , ppl: 72.34204912689945] tot_loss[loss=2.4, over 5510524.98 frames. , ppl: 11.020528340431028], batch size: 70
+2022-12-09 23:51:43,332 INFO [train.py:421] (5/8) Epoch 1, batch 22800, loss[loss=2.359, over 3500.00 frames. , ppl: 10.58366086354835] tot_loss[loss=2.4, over 5503091.67 frames. , ppl: 11.02041304378347], batch size: 70
+2022-12-09 23:53:27,617 INFO [train.py:421] (5/8) Epoch 1, batch 23000, loss[loss=2.461, over 2170.00 frames. , ppl: 11.722171339334842] tot_loss[loss=2.4, over 5492341.38 frames. , ppl: 11.020473157205478], batch size: 70
+2022-12-09 23:53:27,618 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-09 23:53:28,375 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.38, over 211138.00 frames. , ppl: 10.800644888892075
+2022-12-09 23:55:09,816 INFO [train.py:421] (5/8) Epoch 1, batch 23200, loss[loss=2.526, over 2100.00 frames. , ppl: 12.499692638169767] tot_loss[loss=2.4, over 5518829.03 frames. , ppl: 11.018121927588409], batch size: 70
+2022-12-09 23:56:52,353 INFO [train.py:421] (5/8) Epoch 1, batch 23400, loss[loss=2.479, over 1610.00 frames. , ppl: 11.93362032485075] tot_loss[loss=2.4, over 5488395.82 frames. , ppl: 11.021502553347466], batch size: 70
+2022-12-09 23:58:28,672 INFO [train.py:421] (5/8) Epoch 1, batch 23600, loss[loss=2.317, over 3010.00 frames. , ppl: 10.145429755376718] tot_loss[loss=2.401, over 5455475.53 frames. , ppl: 11.029582083980234], batch size: 70
+2022-12-10 00:00:06,981 INFO [train.py:421] (5/8) Epoch 1, batch 23800, loss[loss=2.548, over 1680.00 frames. , ppl: 12.776257044567773] tot_loss[loss=2.4, over 5489069.53 frames. , ppl: 11.019609500170146], batch size: 70
+2022-12-10 00:01:47,276 INFO [train.py:421] (5/8) Epoch 1, batch 24000, loss[loss=2.411, over 1750.00 frames. , ppl: 11.142137541970103] tot_loss[loss=2.399, over 5498386.87 frames. , ppl: 11.008948446044343], batch size: 70
+2022-12-10 00:01:47,277 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 00:01:48,035 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.379, over 211138.00 frames. , ppl: 10.78979920368168
+2022-12-10 00:03:26,227 INFO [train.py:421] (5/8) Epoch 1, batch 24200, loss[loss=2.425, over 1960.00 frames. , ppl: 11.298110879353757] tot_loss[loss=2.398, over 5491447.13 frames. , ppl: 11.004605370656757], batch size: 70
+2022-12-10 00:05:05,819 INFO [train.py:421] (5/8) Epoch 1, batch 24400, loss[loss=2.351, over 4270.00 frames. , ppl: 10.492653655504522] tot_loss[loss=2.397, over 5497771.94 frames. , ppl: 10.988970244202923], batch size: 70
+2022-12-10 00:06:43,341 INFO [train.py:421] (5/8) Epoch 1, batch 24600, loss[loss=2.292, over 4830.00 frames. , ppl: 9.890581704131286] tot_loss[loss=2.398, over 5436453.03 frames. , ppl: 11.002394355673422], batch size: 70
+2022-12-10 00:08:25,288 INFO [train.py:421] (5/8) Epoch 1, batch 24800, loss[loss=2.405, over 2170.00 frames. , ppl: 11.074325889180145] tot_loss[loss=2.397, over 5455436.41 frames. , ppl: 10.98921866172817], batch size: 70
+2022-12-10 00:10:06,626 INFO [train.py:421] (5/8) Epoch 1, batch 25000, loss[loss=2.797, over 700.00 frames. , ppl: 16.399550749425448] tot_loss[loss=2.398, over 5409942.71 frames. , ppl: 10.995907928640802], batch size: 70
+2022-12-10 00:10:06,627 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 00:10:07,386 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.38, over 211138.00 frames. , ppl: 10.800197296869092
+2022-12-10 00:11:43,292 INFO [train.py:421] (5/8) Epoch 1, batch 25200, loss[loss=2.319, over 5530.00 frames. , ppl: 10.167021318110875] tot_loss[loss=2.398, over 5404010.10 frames. , ppl: 11.003217093134083], batch size: 70
+2022-12-10 00:13:24,087 INFO [train.py:421] (5/8) Epoch 1, batch 25400, loss[loss=2.47, over 2030.00 frames. , ppl: 11.820829474660353] tot_loss[loss=2.399, over 5379783.39 frames. , ppl: 11.008177160960354], batch size: 70
+2022-12-10 00:15:07,555 INFO [train.py:421] (5/8) Epoch 1, batch 25600, loss[loss=2.435, over 1400.00 frames. , ppl: 11.413370340178098] tot_loss[loss=2.398, over 5426070.71 frames. , ppl: 11.004445900856927], batch size: 70
+2022-12-10 00:16:48,181 INFO [train.py:421] (5/8) Epoch 1, batch 25800, loss[loss=3.441, over 490.00 frames. , ppl: 31.205227802520564] tot_loss[loss=2.398, over 5432248.39 frames. , ppl: 11.004540539338382], batch size: 70
+2022-12-10 00:18:27,901 INFO [train.py:421] (5/8) Epoch 1, batch 26000, loss[loss=2.547, over 1610.00 frames. , ppl: 12.763582288684747] tot_loss[loss=2.397, over 5462163.20 frames. , ppl: 10.993940868494262], batch size: 70
+2022-12-10 00:18:27,901 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 00:18:28,648 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.379, over 211138.00 frames. , ppl: 10.799350119895008
+2022-12-10 00:20:08,713 INFO [train.py:421] (5/8) Epoch 1, batch 26200, loss[loss=2.718, over 1050.00 frames. , ppl: 15.151198831218684] tot_loss[loss=2.397, over 5513674.18 frames. , ppl: 10.985260056390638], batch size: 70
+2022-12-10 00:21:44,686 INFO [train.py:421] (5/8) Epoch 1, batch 26400, loss[loss=2.412, over 2100.00 frames. , ppl: 11.159142472100338] tot_loss[loss=2.397, over 5519700.27 frames. , ppl: 10.993736759630973], batch size: 70
+2022-12-10 00:23:22,416 INFO [train.py:421] (5/8) Epoch 1, batch 26600, loss[loss=2.639, over 1610.00 frames. , ppl: 14.003622446343913] tot_loss[loss=2.398, over 5511193.20 frames. , ppl: 10.998147511768465], batch size: 70
+2022-12-10 00:25:04,487 INFO [train.py:421] (5/8) Epoch 1, batch 26800, loss[loss=2.559, over 1190.00 frames. , ppl: 12.92726286267601] tot_loss[loss=2.397, over 5530618.94 frames. , ppl: 10.991446247842088], batch size: 70
+2022-12-10 00:26:48,139 INFO [train.py:421] (5/8) Epoch 1, batch 27000, loss[loss=2.431, over 1330.00 frames. , ppl: 11.375089747269687] tot_loss[loss=2.397, over 5528608.23 frames. , ppl: 10.986908293648874], batch size: 70
+2022-12-10 00:26:48,139 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 00:26:48,899 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.379, over 211138.00 frames. , ppl: 10.791004984020308
+2022-12-10 00:28:29,579 INFO [train.py:421] (5/8) Epoch 1, batch 27200, loss[loss=2.473, over 1820.00 frames. , ppl: 11.856652738694754] tot_loss[loss=2.397, over 5481303.20 frames. , ppl: 10.992494494676851], batch size: 70
+2022-12-10 00:30:09,129 INFO [train.py:421] (5/8) Epoch 1, batch 27400, loss[loss=2.673, over 770.00 frames. , ppl: 14.478918048495855] tot_loss[loss=2.396, over 5485032.29 frames. , ppl: 10.981258619031047], batch size: 70
+2022-12-10 00:31:48,114 INFO [train.py:421] (5/8) Epoch 1, batch 27600, loss[loss=2.325, over 1610.00 frames. , ppl: 10.231375359831011] tot_loss[loss=2.396, over 5485917.59 frames. , ppl: 10.98328381442358], batch size: 70
+2022-12-10 00:33:29,296 INFO [train.py:421] (5/8) Epoch 1, batch 27800, loss[loss=4.176, over 350.00 frames. , ppl: 65.13522811262156] tot_loss[loss=2.396, over 5525484.97 frames. , ppl: 10.975004002192186], batch size: 70
+2022-12-10 00:35:07,734 INFO [train.py:421] (5/8) Epoch 1, batch 28000, loss[loss=2.334, over 5670.00 frames. , ppl: 10.323350964905492] tot_loss[loss=2.396, over 5525761.07 frames. , ppl: 10.974643959692324], batch size: 70
+2022-12-10 00:35:07,735 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 00:35:08,463 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.377, over 211138.00 frames. , ppl: 10.77500384045564
+2022-12-10 00:36:52,791 INFO [train.py:421] (5/8) Epoch 1, batch 28200, loss[loss=2.4, over 3080.00 frames. , ppl: 11.024448660139168] tot_loss[loss=2.396, over 5510413.87 frames. , ppl: 10.979095095345322], batch size: 70
+2022-12-10 00:38:30,675 INFO [train.py:421] (5/8) Epoch 1, batch 28400, loss[loss=2.476, over 1680.00 frames. , ppl: 11.888546698357928] tot_loss[loss=2.396, over 5508077.38 frames. , ppl: 10.983190337393111], batch size: 70
+2022-12-10 00:40:09,191 INFO [train.py:421] (5/8) Epoch 1, batch 28600, loss[loss=2.414, over 1680.00 frames. , ppl: 11.177288416187016] tot_loss[loss=2.397, over 5506653.00 frames. , ppl: 10.98655835316104], batch size: 70
+2022-12-10 00:41:48,510 INFO [train.py:421] (5/8) Epoch 1, batch 28800, loss[loss=2.395, over 2170.00 frames. , ppl: 10.96896020473552] tot_loss[loss=2.396, over 5519311.57 frames. , ppl: 10.974577068711618], batch size: 70
+2022-12-10 00:43:30,618 INFO [train.py:421] (5/8) Epoch 1, batch 29000, loss[loss=2.551, over 1610.00 frames. , ppl: 12.821798147044868] tot_loss[loss=2.396, over 5502471.96 frames. , ppl: 10.977991494185225], batch size: 70
+2022-12-10 00:43:30,619 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 00:43:31,364 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.375, over 211138.00 frames. , ppl: 10.749719593753966
+2022-12-10 00:45:08,480 INFO [train.py:421] (5/8) Epoch 1, batch 29200, loss[loss=2.424, over 4270.00 frames. , ppl: 11.2963529858561] tot_loss[loss=2.395, over 5529978.81 frames. , ppl: 10.970543653098137], batch size: 70
+2022-12-10 00:46:48,118 INFO [train.py:421] (5/8) Epoch 1, batch 29400, loss[loss=2.914, over 630.00 frames. , ppl: 18.4371397420342] tot_loss[loss=2.396, over 5485743.40 frames. , ppl: 10.977444276369813], batch size: 70
+2022-12-10 00:48:31,855 INFO [train.py:421] (5/8) Epoch 1, batch 29600, loss[loss=2.539, over 1330.00 frames. , ppl: 12.665719952868752] tot_loss[loss=2.396, over 5474056.36 frames. , ppl: 10.978358010279791], batch size: 70
+2022-12-10 00:50:13,023 INFO [train.py:421] (5/8) Epoch 1, batch 29800, loss[loss=2.345, over 3290.00 frames. , ppl: 10.432130969508536] tot_loss[loss=2.394, over 5515376.79 frames. , ppl: 10.959204431335111], batch size: 70
+2022-12-10 00:51:49,940 INFO [train.py:421] (5/8) Epoch 1, batch 30000, loss[loss=2.463, over 1890.00 frames. , ppl: 11.743296201727158] tot_loss[loss=2.394, over 5488011.82 frames. , ppl: 10.960383927379215], batch size: 70
+2022-12-10 00:51:49,940 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 00:51:50,686 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.373, over 211138.00 frames. , ppl: 10.73370182407569
+2022-12-10 00:53:27,749 INFO [train.py:421] (5/8) Epoch 1, batch 30200, loss[loss=2.499, over 1540.00 frames. , ppl: 12.174747945329536] tot_loss[loss=2.395, over 5487251.68 frames. , ppl: 10.972160452967836], batch size: 70
+2022-12-10 00:55:09,215 INFO [train.py:421] (5/8) Epoch 1, batch 30400, loss[loss=2.49, over 1610.00 frames. , ppl: 12.062749631575231] tot_loss[loss=2.396, over 5478748.59 frames. , ppl: 10.976038999038947], batch size: 70
+2022-12-10 00:56:50,132 INFO [train.py:421] (5/8) Epoch 1, batch 30600, loss[loss=2.452, over 1540.00 frames. , ppl: 11.610203478911908] tot_loss[loss=2.395, over 5495993.33 frames. , ppl: 10.964452323927095], batch size: 70
+2022-12-10 00:58:29,442 INFO [train.py:421] (5/8) Epoch 1, batch 30800, loss[loss=2.302, over 4970.00 frames. , ppl: 9.998612351709012] tot_loss[loss=2.393, over 5520819.59 frames. , ppl: 10.947792585388235], batch size: 70
+2022-12-10 01:00:09,321 INFO [train.py:421] (5/8) Epoch 1, batch 31000, loss[loss=2.325, over 5180.00 frames. , ppl: 10.226036157142005] tot_loss[loss=2.392, over 5518780.49 frames. , ppl: 10.937543933516015], batch size: 70
+2022-12-10 01:00:09,322 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 01:00:10,085 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.373, over 211138.00 frames. , ppl: 10.729127435277624
+2022-12-10 01:01:48,656 INFO [train.py:421] (5/8) Epoch 1, batch 31200, loss[loss=2.374, over 9310.00 frames. , ppl: 10.73820095054758] tot_loss[loss=2.393, over 5500958.61 frames. , ppl: 10.943732197216422], batch size: 70
+2022-12-10 01:03:31,225 INFO [train.py:421] (5/8) Epoch 1, batch 31400, loss[loss=2.315, over 6930.00 frames. , ppl: 10.127085014188788] tot_loss[loss=2.392, over 5526501.15 frames. , ppl: 10.933177365293217], batch size: 70
+2022-12-10 01:05:10,005 INFO [train.py:421] (5/8) Epoch 1, batch 31600, loss[loss=2.236, over 4620.00 frames. , ppl: 9.356887766972552] tot_loss[loss=2.391, over 5540318.24 frames. , ppl: 10.923246619725534], batch size: 70
+2022-12-10 01:06:49,277 INFO [train.py:421] (5/8) Epoch 1, batch 31800, loss[loss=2.403, over 2310.00 frames. , ppl: 11.05468629583193] tot_loss[loss=2.39, over 5552840.00 frames. , ppl: 10.917384238895327], batch size: 70
+2022-12-10 01:08:31,499 INFO [train.py:421] (5/8) Epoch 1, batch 32000, loss[loss=3.235, over 490.00 frames. , ppl: 25.396653078097053] tot_loss[loss=2.391, over 5532068.13 frames. , ppl: 10.921889018760316], batch size: 70
+2022-12-10 01:08:31,500 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 01:08:32,284 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.372, over 211138.00 frames. , ppl: 10.718152095318677
+2022-12-10 01:10:13,339 INFO [train.py:421] (5/8) Epoch 1, batch 32200, loss[loss=2.374, over 4270.00 frames. , ppl: 10.739438517040226] tot_loss[loss=2.392, over 5526371.29 frames. , ppl: 10.933279577227578], batch size: 70
+2022-12-10 01:11:53,782 INFO [train.py:421] (5/8) Epoch 1, batch 32400, loss[loss=2.297, over 8540.00 frames. , ppl: 9.941968047525084] tot_loss[loss=2.392, over 5501385.54 frames. , ppl: 10.933448625494016], batch size: 70
+2022-12-10 01:13:33,493 INFO [train.py:421] (5/8) Epoch 1, batch 32600, loss[loss=2.426, over 2310.00 frames. , ppl: 11.315223667860627] tot_loss[loss=2.392, over 5504829.59 frames. , ppl: 10.930450064434886], batch size: 70
+2022-12-10 01:15:16,739 INFO [train.py:421] (5/8) Epoch 1, batch 32800, loss[loss=2.441, over 2660.00 frames. , ppl: 11.489447575486947] tot_loss[loss=2.393, over 5474875.01 frames. , ppl: 10.943976909386924], batch size: 70
+2022-12-10 01:16:56,205 INFO [train.py:421] (5/8) Epoch 1, batch 33000, loss[loss=2.35, over 5530.00 frames. , ppl: 10.489018127812093] tot_loss[loss=2.392, over 5537439.32 frames. , ppl: 10.937798903705419], batch size: 70
+2022-12-10 01:16:56,205 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 01:16:56,953 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.371, over 211138.00 frames. , ppl: 10.707864689524108
+2022-12-10 01:18:40,611 INFO [train.py:421] (5/8) Epoch 1, batch 33200, loss[loss=2.348, over 6720.00 frames. , ppl: 10.461022947676224] tot_loss[loss=2.392, over 5553682.67 frames. , ppl: 10.933294457802807], batch size: 70
+2022-12-10 01:20:20,452 INFO [train.py:421] (5/8) Epoch 1, batch 33400, loss[loss=3.285, over 490.00 frames. , ppl: 26.71251906040151] tot_loss[loss=2.391, over 5560955.45 frames. , ppl: 10.927786722536084], batch size: 70
+2022-12-10 01:21:57,275 INFO [train.py:421] (5/8) Epoch 1, batch 33600, loss[loss=2.347, over 4340.00 frames. , ppl: 10.458763466145498] tot_loss[loss=2.39, over 5607147.11 frames. , ppl: 10.910259195411694], batch size: 70
+2022-12-10 01:23:38,645 INFO [train.py:421] (5/8) Epoch 1, batch 33800, loss[loss=2.358, over 5110.00 frames. , ppl: 10.574354578598367] tot_loss[loss=2.391, over 5585885.17 frames. , ppl: 10.921712709953798], batch size: 70
+2022-12-10 01:25:17,021 INFO [train.py:421] (5/8) Epoch 1, batch 34000, loss[loss=2.429, over 2940.00 frames. , ppl: 11.347780512848695] tot_loss[loss=2.391, over 5563142.94 frames. , ppl: 10.922031874724748], batch size: 70
+2022-12-10 01:25:17,021 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 01:25:17,782 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.373, over 211138.00 frames. , ppl: 10.73305843282348
+2022-12-10 01:27:03,120 INFO [train.py:421] (5/8) Epoch 1, batch 34200, loss[loss=2.687, over 1050.00 frames. , ppl: 14.693621741138273] tot_loss[loss=2.389, over 5600650.00 frames. , ppl: 10.904214267542528], batch size: 70
+2022-12-10 01:28:41,223 INFO [train.py:421] (5/8) Epoch 1, batch 34400, loss[loss=2.366, over 1890.00 frames. , ppl: 10.655275829333357] tot_loss[loss=2.389, over 5574910.27 frames. , ppl: 10.905800534738141], batch size: 70
+2022-12-10 01:30:22,448 INFO [train.py:421] (5/8) Epoch 1, batch 34600, loss[loss=2.555, over 1750.00 frames. , ppl: 12.865946109407599] tot_loss[loss=2.388, over 5605284.17 frames. , ppl: 10.892300144025757], batch size: 70
+2022-12-10 01:32:03,740 INFO [train.py:421] (5/8) Epoch 1, batch 34800, loss[loss=2.46, over 1890.00 frames. , ppl: 11.700061295104765] tot_loss[loss=2.388, over 5560608.78 frames. , ppl: 10.896097802899073], batch size: 70
+2022-12-10 01:33:43,070 INFO [train.py:421] (5/8) Epoch 1, batch 35000, loss[loss=2.79, over 700.00 frames. , ppl: 16.28538705323956] tot_loss[loss=2.388, over 5568667.07 frames. , ppl: 10.894012519653106], batch size: 70
+2022-12-10 01:33:43,070 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 01:33:43,828 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.372, over 211138.00 frames. , ppl: 10.721890226540799
+2022-12-10 01:35:22,932 INFO [train.py:421] (5/8) Epoch 1, batch 35200, loss[loss=2.795, over 630.00 frames. , ppl: 16.355664096197778] tot_loss[loss=2.389, over 5524132.46 frames. , ppl: 10.899716800937627], batch size: 70
+2022-12-10 01:37:03,047 INFO [train.py:421] (5/8) Epoch 1, batch 35400, loss[loss=2.405, over 1400.00 frames. , ppl: 11.078011062410058] tot_loss[loss=2.389, over 5560321.96 frames. , ppl: 10.900033361550017], batch size: 70
+2022-12-10 01:38:46,644 INFO [train.py:421] (5/8) Epoch 1, batch 35600, loss[loss=2.381, over 3990.00 frames. , ppl: 10.816246430703533] tot_loss[loss=2.388, over 5559290.05 frames. , ppl: 10.890060122322947], batch size: 70
+2022-12-10 01:40:24,015 INFO [train.py:421] (5/8) Epoch 1, batch 35800, loss[loss=2.625, over 1470.00 frames. , ppl: 13.80411565526487] tot_loss[loss=2.387, over 5550930.15 frames. , ppl: 10.886153753137993], batch size: 70
+2022-12-10 01:42:02,433 INFO [train.py:421] (5/8) Epoch 1, batch 36000, loss[loss=2.284, over 5110.00 frames. , ppl: 9.819428864021276] tot_loss[loss=2.388, over 5569404.48 frames. , ppl: 10.887177258299555], batch size: 70
+2022-12-10 01:42:02,433 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 01:42:03,181 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.371, over 211138.00 frames. , ppl: 10.709730213282327
+2022-12-10 01:43:40,127 INFO [train.py:421] (5/8) Epoch 1, batch 36200, loss[loss=2.388, over 3080.00 frames. , ppl: 10.889064652008857] tot_loss[loss=2.387, over 5565881.87 frames. , ppl: 10.885564568576086], batch size: 70
+2022-12-10 01:45:21,067 INFO [train.py:421] (5/8) Epoch 1, batch 36400, loss[loss=2.601, over 1610.00 frames. , ppl: 13.473615042235192] tot_loss[loss=2.387, over 5579642.06 frames. , ppl: 10.882809245995302], batch size: 70
+2022-12-10 01:47:02,497 INFO [train.py:421] (5/8) Epoch 1, batch 36600, loss[loss=2.315, over 11480.00 frames. , ppl: 10.125921376433308] tot_loss[loss=2.386, over 5602007.68 frames. , ppl: 10.871287218001333], batch size: 70
+2022-12-10 01:48:44,159 INFO [train.py:421] (5/8) Epoch 1, batch 36800, loss[loss=2.358, over 2940.00 frames. , ppl: 10.566527904871268] tot_loss[loss=2.386, over 5605298.56 frames. , ppl: 10.869452456357738], batch size: 70
+2022-12-10 01:50:20,592 INFO [train.py:421] (5/8) Epoch 1, batch 37000, loss[loss=2.392, over 1470.00 frames. , ppl: 10.932117667829662] tot_loss[loss=2.386, over 5618056.22 frames. , ppl: 10.867062850282416], batch size: 70
+2022-12-10 01:50:20,593 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 01:50:21,353 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.371, over 211138.00 frames. , ppl: 10.706359193662431
+2022-12-10 01:52:04,192 INFO [train.py:421] (5/8) Epoch 1, batch 37200, loss[loss=2.294, over 4270.00 frames. , ppl: 9.913756687630583] tot_loss[loss=2.385, over 5619883.54 frames. , ppl: 10.863557566571933], batch size: 70
+2022-12-10 01:53:44,587 INFO [train.py:421] (5/8) Epoch 1, batch 37400, loss[loss=2.349, over 4620.00 frames. , ppl: 10.478335642663797] tot_loss[loss=2.385, over 5636350.45 frames. , ppl: 10.853971449774983], batch size: 70
+2022-12-10 01:55:24,444 INFO [train.py:421] (5/8) Epoch 1, batch 37600, loss[loss=3.219, over 490.00 frames. , ppl: 25.004216321015274] tot_loss[loss=2.384, over 5635745.62 frames. , ppl: 10.850488016634795], batch size: 70
+2022-12-10 01:57:05,223 INFO [train.py:421] (5/8) Epoch 1, batch 37800, loss[loss=2.266, over 5390.00 frames. , ppl: 9.640977557455383] tot_loss[loss=2.385, over 5577836.25 frames. , ppl: 10.86421533280801], batch size: 70
+2022-12-10 01:58:47,483 INFO [train.py:421] (5/8) Epoch 1, batch 38000, loss[loss=2.32, over 5600.00 frames. , ppl: 10.172569410510793] tot_loss[loss=2.385, over 5602330.92 frames. , ppl: 10.856244600630708], batch size: 70
+2022-12-10 01:58:47,484 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 01:58:48,230 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.367, over 211138.00 frames. , ppl: 10.660825546799813
+2022-12-10 02:00:27,335 INFO [train.py:421] (5/8) Epoch 1, batch 38200, loss[loss=2.233, over 4830.00 frames. , ppl: 9.332287120469108] tot_loss[loss=2.383, over 5639888.24 frames. , ppl: 10.83465426561422], batch size: 70
+2022-12-10 02:02:09,572 INFO [train.py:421] (5/8) Epoch 1, batch 38400, loss[loss=2.311, over 4130.00 frames. , ppl: 10.084248213017409] tot_loss[loss=2.383, over 5618127.19 frames. , ppl: 10.841230168503307], batch size: 70
+2022-12-10 02:03:49,951 INFO [train.py:421] (5/8) Epoch 1, batch 38600, loss[loss=2.486, over 980.00 frames. , ppl: 12.012508254024693] tot_loss[loss=2.383, over 5617607.63 frames. , ppl: 10.842044891274574], batch size: 70
+2022-12-10 02:05:31,055 INFO [train.py:421] (5/8) Epoch 1, batch 38800, loss[loss=2.499, over 980.00 frames. , ppl: 12.172779956131263] tot_loss[loss=2.383, over 5611024.18 frames. , ppl: 10.83951506589614], batch size: 70
+2022-12-10 02:07:09,882 INFO [train.py:421] (5/8) Epoch 1, batch 39000, loss[loss=2.674, over 1260.00 frames. , ppl: 14.493454614520282] tot_loss[loss=2.384, over 5561597.07 frames. , ppl: 10.84810810117517], batch size: 70
+2022-12-10 02:07:09,882 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 02:07:10,643 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.368, over 211138.00 frames. , ppl: 10.674933246579991
+2022-12-10 02:08:52,861 INFO [train.py:421] (5/8) Epoch 1, batch 39200, loss[loss=2.549, over 1890.00 frames. , ppl: 12.794087305180943] tot_loss[loss=2.384, over 5573510.68 frames. , ppl: 10.844230883969773], batch size: 70
+2022-12-10 02:10:35,768 INFO [train.py:421] (5/8) Epoch 1, batch 39400, loss[loss=2.441, over 3780.00 frames. , ppl: 11.482909964856626] tot_loss[loss=2.383, over 5567364.11 frames. , ppl: 10.838865827697541], batch size: 70
+2022-12-10 02:12:14,142 INFO [train.py:421] (5/8) Epoch 1, batch 39600, loss[loss=2.367, over 1750.00 frames. , ppl: 10.660231010544903] tot_loss[loss=2.384, over 5519906.51 frames. , ppl: 10.848777066864967], batch size: 70
+2022-12-10 02:13:56,725 INFO [train.py:421] (5/8) Epoch 1, batch 39800, loss[loss=2.436, over 1190.00 frames. , ppl: 11.4248570646055] tot_loss[loss=2.384, over 5492191.44 frames. , ppl: 10.850036136243572], batch size: 70
+2022-12-10 02:15:38,369 INFO [train.py:421] (5/8) Epoch 1, batch 40000, loss[loss=2.383, over 1610.00 frames. , ppl: 10.839755191818826] tot_loss[loss=2.385, over 5457697.67 frames. , ppl: 10.861244844309542], batch size: 70
+2022-12-10 02:15:38,369 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 02:15:39,131 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.368, over 211138.00 frames. , ppl: 10.68034759574071
+2022-12-10 02:17:21,026 INFO [train.py:421] (5/8) Epoch 1, batch 40200, loss[loss=2.56, over 840.00 frames. , ppl: 12.932323510620495] tot_loss[loss=2.386, over 5439315.40 frames. , ppl: 10.867940664985875], batch size: 70
+2022-12-10 02:19:03,094 INFO [train.py:421] (5/8) Epoch 1, batch 40400, loss[loss=2.318, over 2240.00 frames. , ppl: 10.159866072106286] tot_loss[loss=2.387, over 5421919.16 frames. , ppl: 10.875394459883532], batch size: 70
+2022-12-10 02:20:42,136 INFO [train.py:421] (5/8) Epoch 1, batch 40600, loss[loss=2.348, over 4410.00 frames. , ppl: 10.461122558408306] tot_loss[loss=2.386, over 5402750.31 frames. , ppl: 10.874677080514246], batch size: 70
+2022-12-10 02:22:24,716 INFO [train.py:421] (5/8) Epoch 1, batch 40800, loss[loss=2.385, over 2240.00 frames. , ppl: 10.864406022894537] tot_loss[loss=2.385, over 5419747.82 frames. , ppl: 10.859732314673119], batch size: 70
+2022-12-10 02:24:04,605 INFO [train.py:421] (5/8) Epoch 1, batch 41000, loss[loss=2.418, over 3220.00 frames. , ppl: 11.218438452610574] tot_loss[loss=2.384, over 5428824.71 frames. , ppl: 10.853012011135265], batch size: 70
+2022-12-10 02:24:04,606 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 02:24:05,368 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.367, over 211138.00 frames. , ppl: 10.668684715535875
+2022-12-10 02:25:40,536 INFO [train.py:421] (5/8) Epoch 1, batch 41200, loss[loss=2.4, over 3570.00 frames. , ppl: 11.025456225457178] tot_loss[loss=2.385, over 5397479.94 frames. , ppl: 10.861132643406325], batch size: 70
+2022-12-10 02:27:22,733 INFO [train.py:421] (5/8) Epoch 1, batch 41400, loss[loss=2.388, over 3360.00 frames. , ppl: 10.889948396662895] tot_loss[loss=2.385, over 5396644.13 frames. , ppl: 10.86073649541046], batch size: 70
+2022-12-10 02:29:04,790 INFO [train.py:421] (5/8) Epoch 1, batch 41600, loss[loss=2.384, over 3430.00 frames. , ppl: 10.844556182590038] tot_loss[loss=2.384, over 5447269.75 frames. , ppl: 10.848356766092781], batch size: 70
+2022-12-10 02:30:42,785 INFO [train.py:421] (5/8) Epoch 1, batch 41800, loss[loss=2.339, over 5810.00 frames. , ppl: 10.37327131465643] tot_loss[loss=2.384, over 5454908.70 frames. , ppl: 10.85058527472237], batch size: 70
+2022-12-10 02:32:23,880 INFO [train.py:421] (5/8) Epoch 1, batch 42000, loss[loss=2.339, over 3990.00 frames. , ppl: 10.366684237855113] tot_loss[loss=2.385, over 5393904.81 frames. , ppl: 10.859022739193994], batch size: 70
+2022-12-10 02:32:23,880 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 02:32:24,639 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.364, over 211138.00 frames. , ppl: 10.63275418399596
+2022-12-10 02:34:04,507 INFO [train.py:421] (5/8) Epoch 1, batch 42200, loss[loss=2.248, over 2730.00 frames. , ppl: 9.465899150205873] tot_loss[loss=2.385, over 5376251.60 frames. , ppl: 10.85534354317583], batch size: 70
+2022-12-10 02:35:44,538 INFO [train.py:421] (5/8) Epoch 1, batch 42400, loss[loss=2.264, over 6370.00 frames. , ppl: 9.61776227179434] tot_loss[loss=2.384, over 5363781.70 frames. , ppl: 10.852653599842352], batch size: 70
+2022-12-10 02:37:26,135 INFO [train.py:421] (5/8) Epoch 1, batch 42600, loss[loss=2.623, over 980.00 frames. , ppl: 13.776080251402169] tot_loss[loss=2.385, over 5352778.00 frames. , ppl: 10.854621375681559], batch size: 70
+2022-12-10 02:39:06,181 INFO [train.py:421] (5/8) Epoch 1, batch 42800, loss[loss=2.445, over 980.00 frames. , ppl: 11.524899001471873] tot_loss[loss=2.386, over 5339416.68 frames. , ppl: 10.866073661675776], batch size: 70
+2022-12-10 02:40:45,345 INFO [train.py:421] (5/8) Epoch 1, batch 43000, loss[loss=3.298, over 490.00 frames. , ppl: 27.06689633422421] tot_loss[loss=2.384, over 5375214.31 frames. , ppl: 10.846880354342638], batch size: 70
+2022-12-10 02:40:45,345 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 02:40:46,105 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.364, over 211138.00 frames. , ppl: 10.630343507571958
+2022-12-10 02:42:23,863 INFO [train.py:421] (5/8) Epoch 1, batch 43200, loss[loss=2.524, over 700.00 frames. , ppl: 12.475798305006487] tot_loss[loss=2.385, over 5345543.01 frames. , ppl: 10.854287219498604], batch size: 70
+2022-12-10 02:44:06,635 INFO [train.py:421] (5/8) Epoch 1, batch 43400, loss[loss=2.271, over 7350.00 frames. , ppl: 9.68998315077283] tot_loss[loss=2.382, over 5411504.68 frames. , ppl: 10.82976178743632], batch size: 70
+2022-12-10 02:45:46,645 INFO [train.py:421] (5/8) Epoch 1, batch 43600, loss[loss=2.316, over 4060.00 frames. , ppl: 10.134791965882645] tot_loss[loss=2.382, over 5403919.99 frames. , ppl: 10.828201967542777], batch size: 70
+2022-12-10 02:47:26,672 INFO [train.py:421] (5/8) Epoch 1, batch 43800, loss[loss=2.336, over 1610.00 frames. , ppl: 10.335875228157542] tot_loss[loss=2.383, over 5381179.44 frames. , ppl: 10.832924334124705], batch size: 70
+2022-12-10 02:49:05,961 INFO [train.py:421] (5/8) Epoch 1, batch 44000, loss[loss=2.831, over 770.00 frames. , ppl: 16.967419807377407] tot_loss[loss=2.382, over 5400521.23 frames. , ppl: 10.831586435335948], batch size: 70
+2022-12-10 02:49:05,961 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 02:49:06,722 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.365, over 211138.00 frames. , ppl: 10.648347077171886
+2022-12-10 02:50:46,016 INFO [train.py:421] (5/8) Epoch 1, batch 44200, loss[loss=2.46, over 1610.00 frames. , ppl: 11.700544347162635] tot_loss[loss=2.383, over 5371560.11 frames. , ppl: 10.841782531427917], batch size: 70
+2022-12-10 02:52:25,664 INFO [train.py:421] (5/8) Epoch 1, batch 44400, loss[loss=2.612, over 1330.00 frames. , ppl: 13.627436421668214] tot_loss[loss=2.384, over 5320473.39 frames. , ppl: 10.852883238206532], batch size: 70
+2022-12-10 02:54:07,176 INFO [train.py:421] (5/8) Epoch 1, batch 44600, loss[loss=2.897, over 700.00 frames. , ppl: 18.113699433865868] tot_loss[loss=2.385, over 5322188.47 frames. , ppl: 10.859426543818067], batch size: 70
+2022-12-10 02:55:46,318 INFO [train.py:421] (5/8) Epoch 1, batch 44800, loss[loss=2.531, over 1120.00 frames. , ppl: 12.563159653806268] tot_loss[loss=2.385, over 5327429.39 frames. , ppl: 10.857338412664848], batch size: 70
+2022-12-10 02:57:25,620 INFO [train.py:421] (5/8) Epoch 1, batch 45000, loss[loss=2.558, over 1470.00 frames. , ppl: 12.913040213772579] tot_loss[loss=2.384, over 5370388.13 frames. , ppl: 10.852473951665207], batch size: 70
+2022-12-10 02:57:25,620 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 02:57:26,378 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.364, over 211138.00 frames. , ppl: 10.628203939461411
+2022-12-10 02:59:10,686 INFO [train.py:421] (5/8) Epoch 1, batch 45200, loss[loss=2.533, over 910.00 frames. , ppl: 12.59552460499795] tot_loss[loss=2.385, over 5396833.70 frames. , ppl: 10.855636447634314], batch size: 70
+2022-12-10 03:00:53,880 INFO [train.py:421] (5/8) Epoch 1, batch 45400, loss[loss=2.973, over 560.00 frames. , ppl: 19.54539981880279] tot_loss[loss=2.384, over 5414653.30 frames. , ppl: 10.852145471533472], batch size: 70
+2022-12-10 03:02:33,640 INFO [train.py:421] (5/8) Epoch 1, batch 45600, loss[loss=2.399, over 2170.00 frames. , ppl: 11.017378792705209] tot_loss[loss=2.383, over 5429018.62 frames. , ppl: 10.840811348655224], batch size: 70
+2022-12-10 03:04:17,137 INFO [train.py:421] (5/8) Epoch 1, batch 45800, loss[loss=2.475, over 1540.00 frames. , ppl: 11.877445197681235] tot_loss[loss=2.384, over 5406818.11 frames. , ppl: 10.850587449232094], batch size: 70
+2022-12-10 03:05:58,453 INFO [train.py:421] (5/8) Epoch 1, batch 46000, loss[loss=2.781, over 700.00 frames. , ppl: 16.140111117685002] tot_loss[loss=2.384, over 5392828.42 frames. , ppl: 10.853340559472569], batch size: 70
+2022-12-10 03:05:58,453 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 03:05:59,200 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.363, over 211138.00 frames. , ppl: 10.624262598332244
+2022-12-10 03:07:41,815 INFO [train.py:421] (5/8) Epoch 1, batch 46200, loss[loss=4.168, over 350.00 frames. , ppl: 64.59255722064044] tot_loss[loss=2.383, over 5429591.65 frames. , ppl: 10.835794224173227], batch size: 70
+2022-12-10 03:09:19,254 INFO [train.py:421] (5/8) Epoch 1, batch 46400, loss[loss=3.359, over 490.00 frames. , ppl: 28.7567857174746] tot_loss[loss=2.383, over 5421873.50 frames. , ppl: 10.83462285092571], batch size: 70
+2022-12-10 03:11:00,974 INFO [train.py:421] (5/8) Epoch 1, batch 46600, loss[loss=2.311, over 2380.00 frames. , ppl: 10.079579922780367] tot_loss[loss=2.382, over 5425908.01 frames. , ppl: 10.82981660777812], batch size: 70
+2022-12-10 03:12:37,299 INFO [train.py:421] (5/8) Epoch 1, batch 46800, loss[loss=2.454, over 1260.00 frames. , ppl: 11.632660844086942] tot_loss[loss=2.381, over 5463807.71 frames. , ppl: 10.818402152673956], batch size: 70
+2022-12-10 03:14:18,443 INFO [train.py:421] (5/8) Epoch 1, batch 47000, loss[loss=2.556, over 910.00 frames. , ppl: 12.884978809538442] tot_loss[loss=2.38, over 5493828.18 frames. , ppl: 10.803784783984554], batch size: 70
+2022-12-10 03:14:18,443 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 03:14:19,231 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.363, over 211138.00 frames. , ppl: 10.618849962774785
+2022-12-10 03:15:58,272 INFO [train.py:421] (5/8) Epoch 1, batch 47200, loss[loss=2.334, over 2100.00 frames. , ppl: 10.317686045437581] tot_loss[loss=2.379, over 5472640.70 frames. , ppl: 10.797308962664099], batch size: 70
+2022-12-10 03:17:35,094 INFO [train.py:421] (5/8) Epoch 1, batch 47400, loss[loss=3.78, over 420.00 frames. , ppl: 43.80964675953984] tot_loss[loss=2.379, over 5462807.11 frames. , ppl: 10.79892736312393], batch size: 70
+2022-12-10 03:19:18,529 INFO [train.py:421] (5/8) Epoch 1, batch 47600, loss[loss=2.916, over 630.00 frames. , ppl: 18.470128407444076] tot_loss[loss=2.38, over 5432510.94 frames. , ppl: 10.808091027360962], batch size: 70
+2022-12-10 03:20:57,292 INFO [train.py:421] (5/8) Epoch 1, batch 47800, loss[loss=2.516, over 1470.00 frames. , ppl: 12.375425495353067] tot_loss[loss=2.381, over 5391375.82 frames. , ppl: 10.81893359044281], batch size: 70
+2022-12-10 03:22:37,621 INFO [train.py:421] (5/8) Epoch 1, batch 48000, loss[loss=2.955, over 630.00 frames. , ppl: 19.198178086142523] tot_loss[loss=2.382, over 5347810.19 frames. , ppl: 10.827346593199595], batch size: 70
+2022-12-10 03:22:37,621 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 03:22:38,366 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.361, over 211138.00 frames. , ppl: 10.600952749654093
+2022-12-10 03:24:15,525 INFO [train.py:421] (5/8) Epoch 1, batch 48200, loss[loss=2.305, over 4130.00 frames. , ppl: 10.022837714415397] tot_loss[loss=2.382, over 5329901.12 frames. , ppl: 10.82654655775883], batch size: 70
+2022-12-10 03:25:55,553 INFO [train.py:421] (5/8) Epoch 1, batch 48400, loss[loss=2.321, over 5950.00 frames. , ppl: 10.188232960855176] tot_loss[loss=2.383, over 5304616.68 frames. , ppl: 10.834741115376477], batch size: 70
+2022-12-10 03:27:34,538 INFO [train.py:421] (5/8) Epoch 1, batch 48600, loss[loss=2.276, over 2940.00 frames. , ppl: 9.734157354612895] tot_loss[loss=2.383, over 5308728.13 frames. , ppl: 10.833489442093006], batch size: 70
+2022-12-10 03:29:11,885 INFO [train.py:421] (5/8) Epoch 1, batch 48800, loss[loss=2.449, over 1540.00 frames. , ppl: 11.574792929227966] tot_loss[loss=2.382, over 5308377.31 frames. , ppl: 10.82744886079146], batch size: 70
+2022-12-10 03:30:50,523 INFO [train.py:421] (5/8) Epoch 1, batch 49000, loss[loss=2.36, over 2520.00 frames. , ppl: 10.592659781693557] tot_loss[loss=2.382, over 5333649.78 frames. , ppl: 10.823816192173108], batch size: 70
+2022-12-10 03:30:50,524 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 03:30:51,283 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.36, over 211138.00 frames. , ppl: 10.585974491786418
+2022-12-10 03:32:29,120 INFO [train.py:421] (5/8) Epoch 1, batch 49200, loss[loss=2.357, over 3640.00 frames. , ppl: 10.559598685574855] tot_loss[loss=2.382, over 5321345.04 frames. , ppl: 10.826251740410646], batch size: 70
+2022-12-10 03:34:05,518 INFO [train.py:421] (5/8) Epoch 1, batch 49400, loss[loss=3.093, over 560.00 frames. , ppl: 22.045023760422062] tot_loss[loss=2.383, over 5291354.28 frames. , ppl: 10.832821719595684], batch size: 70
+2022-12-10 03:35:47,724 INFO [train.py:421] (5/8) Epoch 1, batch 49600, loss[loss=2.376, over 2240.00 frames. , ppl: 10.759251522529407] tot_loss[loss=2.382, over 5290758.17 frames. , ppl: 10.830831430161144], batch size: 70
+2022-12-10 03:37:28,142 INFO [train.py:421] (5/8) Epoch 1, batch 49800, loss[loss=2.421, over 2380.00 frames. , ppl: 11.26216274039444] tot_loss[loss=2.382, over 5297038.40 frames. , ppl: 10.825001434486861], batch size: 70
+2022-12-10 03:39:06,578 INFO [train.py:421] (5/8) Epoch 1, batch 50000, loss[loss=2.371, over 4760.00 frames. , ppl: 10.705628132799793] tot_loss[loss=2.381, over 5324970.68 frames. , ppl: 10.81927357654073], batch size: 70
+2022-12-10 03:39:06,578 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 03:39:07,339 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.359, over 211138.00 frames. , ppl: 10.58257820728752
+2022-12-10 03:40:43,902 INFO [train.py:421] (5/8) Epoch 1, batch 50200, loss[loss=2.452, over 2730.00 frames. , ppl: 11.613475958557798] tot_loss[loss=2.381, over 5342365.65 frames. , ppl: 10.81582772570373], batch size: 70
+2022-12-10 03:42:22,999 INFO [train.py:421] (5/8) Epoch 1, batch 50400, loss[loss=2.554, over 980.00 frames. , ppl: 12.862342426159723] tot_loss[loss=2.382, over 5351873.48 frames. , ppl: 10.82165000138107], batch size: 70
+2022-12-10 03:44:03,135 INFO [train.py:421] (5/8) Epoch 1, batch 50600, loss[loss=3.505, over 490.00 frames. , ppl: 33.2834056024656] tot_loss[loss=2.382, over 5330749.24 frames. , ppl: 10.824237219107117], batch size: 70
+2022-12-10 03:45:44,370 INFO [train.py:421] (5/8) Epoch 1, batch 50800, loss[loss=2.368, over 2310.00 frames. , ppl: 10.674893223852617] tot_loss[loss=2.381, over 5329367.32 frames. , ppl: 10.821009503320614], batch size: 70
+2022-12-10 03:47:26,021 INFO [train.py:421] (5/8) Epoch 1, batch 51000, loss[loss=2.332, over 4060.00 frames. , ppl: 10.296140017630721] tot_loss[loss=2.381, over 5376825.33 frames. , ppl: 10.810791772345771], batch size: 70
+2022-12-10 03:47:26,022 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 03:47:26,769 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.361, over 211138.00 frames. , ppl: 10.598020656585552
+2022-12-10 03:49:07,095 INFO [train.py:421] (5/8) Epoch 1, batch 51200, loss[loss=2.687, over 840.00 frames. , ppl: 14.682571858568853] tot_loss[loss=2.379, over 5443805.16 frames. , ppl: 10.79801426655784], batch size: 70
+2022-12-10 03:50:49,016 INFO [train.py:421] (5/8) Epoch 1, batch 51400, loss[loss=2.409, over 2310.00 frames. , ppl: 11.119189309080959] tot_loss[loss=2.379, over 5462509.45 frames. , ppl: 10.794318521150107], batch size: 70
+2022-12-10 03:52:29,258 INFO [train.py:421] (5/8) Epoch 1, batch 51600, loss[loss=2.36, over 6020.00 frames. , ppl: 10.58931117501247] tot_loss[loss=2.378, over 5472846.04 frames. , ppl: 10.787753112293382], batch size: 70
+2022-12-10 03:54:10,138 INFO [train.py:421] (5/8) Epoch 1, batch 51800, loss[loss=2.489, over 1610.00 frames. , ppl: 12.048855741720946] tot_loss[loss=2.378, over 5466930.85 frames. , ppl: 10.778842489077997], batch size: 70
+2022-12-10 03:55:47,567 INFO [train.py:421] (5/8) Epoch 1, batch 52000, loss[loss=2.425, over 2520.00 frames. , ppl: 11.300495114247626] tot_loss[loss=2.376, over 5487925.82 frames. , ppl: 10.762068902274047], batch size: 70
+2022-12-10 03:55:47,568 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 03:55:48,316 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.359, over 211138.00 frames. , ppl: 10.580972870980991
+2022-12-10 03:57:25,692 INFO [train.py:421] (5/8) Epoch 1, batch 52200, loss[loss=2.527, over 1610.00 frames. , ppl: 12.510374424864978] tot_loss[loss=2.376, over 5453980.10 frames. , ppl: 10.766602627167508], batch size: 70
+2022-12-10 03:59:05,804 INFO [train.py:421] (5/8) Epoch 1, batch 52400, loss[loss=2.333, over 7070.00 frames. , ppl: 10.304007833827646] tot_loss[loss=2.375, over 5514736.81 frames. , ppl: 10.750950625648892], batch size: 70
+2022-12-10 04:00:46,728 INFO [train.py:421] (5/8) Epoch 1, batch 52600, loss[loss=2.41, over 2170.00 frames. , ppl: 11.12857454688648] tot_loss[loss=2.375, over 5525141.44 frames. , ppl: 10.751381659158346], batch size: 70
+2022-12-10 04:02:26,552 INFO [train.py:421] (5/8) Epoch 1, batch 52800, loss[loss=2.376, over 4900.00 frames. , ppl: 10.760659483162927] tot_loss[loss=2.373, over 5612635.17 frames. , ppl: 10.729773292968542], batch size: 70
+2022-12-10 04:04:06,247 INFO [train.py:421] (5/8) Epoch 1, batch 53000, loss[loss=2.303, over 4970.00 frames. , ppl: 10.008320427629553] tot_loss[loss=2.372, over 5630460.35 frames. , ppl: 10.722985250504752], batch size: 70
+2022-12-10 04:04:06,247 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 04:04:07,006 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.358, over 211138.00 frames. , ppl: 10.567887116951528
+2022-12-10 04:05:52,844 INFO [train.py:421] (5/8) Epoch 1, batch 53200, loss[loss=2.281, over 7280.00 frames. , ppl: 9.787220380279006] tot_loss[loss=2.371, over 5666914.57 frames. , ppl: 10.710390463616669], batch size: 70
+2022-12-10 04:07:33,548 INFO [train.py:421] (5/8) Epoch 1, batch 53400, loss[loss=2.276, over 6090.00 frames. , ppl: 9.73720341229657] tot_loss[loss=2.372, over 5655679.69 frames. , ppl: 10.714127311991964], batch size: 70
+2022-12-10 04:09:19,575 INFO [train.py:421] (5/8) Epoch 1, batch 53600, loss[loss=2.438, over 2310.00 frames. , ppl: 11.451842326244378] tot_loss[loss=2.37, over 5683973.32 frames. , ppl: 10.699945202119368], batch size: 70
+2022-12-10 04:10:59,120 INFO [train.py:421] (5/8) Epoch 1, batch 53800, loss[loss=2.376, over 2590.00 frames. , ppl: 10.763538246275878] tot_loss[loss=2.37, over 5659554.06 frames. , ppl: 10.702543651090775], batch size: 70
+2022-12-10 04:12:36,568 INFO [train.py:421] (5/8) Epoch 1, batch 54000, loss[loss=2.42, over 2310.00 frames. , ppl: 11.246407016172267] tot_loss[loss=2.371, over 5641041.78 frames. , ppl: 10.70333314823929], batch size: 70
+2022-12-10 04:12:36,568 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 04:12:37,329 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.356, over 211138.00 frames. , ppl: 10.55320704955568
+2022-12-10 04:14:15,872 INFO [train.py:421] (5/8) Epoch 1, batch 54200, loss[loss=2.804, over 770.00 frames. , ppl: 16.507156410512593] tot_loss[loss=2.372, over 5607969.86 frames. , ppl: 10.72181637317945], batch size: 70
+2022-12-10 04:15:51,797 INFO [train.py:421] (5/8) Epoch 1, batch 54400, loss[loss=2.438, over 1960.00 frames. , ppl: 11.446332851223929] tot_loss[loss=2.373, over 5584108.37 frames. , ppl: 10.729112806787242], batch size: 70
+2022-12-10 04:17:31,784 INFO [train.py:421] (5/8) Epoch 1, batch 54600, loss[loss=2.447, over 2450.00 frames. , ppl: 11.556491220602044] tot_loss[loss=2.374, over 5557189.63 frames. , ppl: 10.735268229819852], batch size: 70
+2022-12-10 04:19:11,953 INFO [train.py:421] (5/8) Epoch 1, batch 54800, loss[loss=2.448, over 1540.00 frames. , ppl: 11.564928584637485] tot_loss[loss=2.375, over 5517663.98 frames. , ppl: 10.748284510280326], batch size: 70
+2022-12-10 04:20:51,927 INFO [train.py:421] (5/8) Epoch 1, batch 55000, loss[loss=2.504, over 1330.00 frames. , ppl: 12.225577947961986] tot_loss[loss=2.373, over 5556956.34 frames. , ppl: 10.734368775202256], batch size: 70
+2022-12-10 04:20:51,927 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 04:20:52,675 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.355, over 211138.00 frames. , ppl: 10.542236908630983
+2022-12-10 04:22:35,384 INFO [train.py:421] (5/8) Epoch 1, batch 55200, loss[loss=2.435, over 3150.00 frames. , ppl: 11.411777741938424] tot_loss[loss=2.373, over 5606209.27 frames. , ppl: 10.724698852420694], batch size: 70
+2022-12-10 04:24:13,649 INFO [train.py:421] (5/8) Epoch 1, batch 55400, loss[loss=2.331, over 5670.00 frames. , ppl: 10.28728466126819] tot_loss[loss=2.373, over 5593367.29 frames. , ppl: 10.73062923651013], batch size: 70
+2022-12-10 04:25:54,098 INFO [train.py:421] (5/8) Epoch 1, batch 55600, loss[loss=2.353, over 5530.00 frames. , ppl: 10.518752454916578] tot_loss[loss=2.373, over 5573698.95 frames. , ppl: 10.731538081239638], batch size: 70
+2022-12-10 04:27:33,018 INFO [train.py:421] (5/8) Epoch 1, batch 55800, loss[loss=2.31, over 5600.00 frames. , ppl: 10.076209762829135] tot_loss[loss=2.373, over 5589154.76 frames. , ppl: 10.725005910817853], batch size: 70
+2022-12-10 04:29:13,169 INFO [train.py:421] (5/8) Epoch 1, batch 56000, loss[loss=2.403, over 3080.00 frames. , ppl: 11.056194743910197] tot_loss[loss=2.373, over 5595424.94 frames. , ppl: 10.730778347601072], batch size: 70
+2022-12-10 04:29:13,169 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 04:29:13,929 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.355, over 211138.00 frames. , ppl: 10.539062115723747
+2022-12-10 04:30:54,329 INFO [train.py:421] (5/8) Epoch 1, batch 56200, loss[loss=2.268, over 4690.00 frames. , ppl: 9.660755169151331] tot_loss[loss=2.372, over 5607058.37 frames. , ppl: 10.723927108451258], batch size: 70
+2022-12-10 04:32:32,226 INFO [train.py:421] (5/8) Epoch 1, batch 56400, loss[loss=2.348, over 2100.00 frames. , ppl: 10.460429472696081] tot_loss[loss=2.373, over 5553934.27 frames. , ppl: 10.732432555775985], batch size: 70
+2022-12-10 04:34:13,605 INFO [train.py:421] (5/8) Epoch 1, batch 56600, loss[loss=2.905, over 910.00 frames. , ppl: 18.266695183625185] tot_loss[loss=2.373, over 5547410.42 frames. , ppl: 10.734754376887938], batch size: 70
+2022-12-10 04:35:52,343 INFO [train.py:421] (5/8) Epoch 1, batch 56800, loss[loss=2.374, over 3500.00 frames. , ppl: 10.7423744519175] tot_loss[loss=2.373, over 5547546.17 frames. , ppl: 10.73218442966159], batch size: 70
+2022-12-10 04:37:33,496 INFO [train.py:421] (5/8) Epoch 1, batch 57000, loss[loss=2.329, over 1960.00 frames. , ppl: 10.267575825776865] tot_loss[loss=2.372, over 5563699.14 frames. , ppl: 10.719013762635424], batch size: 70
+2022-12-10 04:37:33,496 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 04:37:34,257 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.354, over 211138.00 frames. , ppl: 10.531887643629647
+2022-12-10 04:39:15,041 INFO [train.py:421] (5/8) Epoch 1, batch 57200, loss[loss=2.342, over 3010.00 frames. , ppl: 10.402708029206213] tot_loss[loss=2.373, over 5557779.97 frames. , ppl: 10.727617352273016], batch size: 70
+2022-12-10 04:40:55,985 INFO [train.py:421] (5/8) Epoch 1, batch 57400, loss[loss=2.471, over 2100.00 frames. , ppl: 11.835929616351867] tot_loss[loss=2.373, over 5549119.68 frames. , ppl: 10.725853800700563], batch size: 70
+2022-12-10 04:42:35,563 INFO [train.py:421] (5/8) Epoch 1, batch 57600, loss[loss=2.399, over 1330.00 frames. , ppl: 11.00725514558029] tot_loss[loss=2.373, over 5534921.56 frames. , ppl: 10.730352950255561], batch size: 70
+2022-12-10 04:44:14,430 INFO [train.py:421] (5/8) Epoch 1, batch 57800, loss[loss=2.407, over 2030.00 frames. , ppl: 11.103605311003454] tot_loss[loss=2.372, over 5565970.98 frames. , ppl: 10.718656229983313], batch size: 70
+2022-12-10 04:45:53,663 INFO [train.py:421] (5/8) Epoch 1, batch 58000, loss[loss=2.524, over 1330.00 frames. , ppl: 12.476591281548009] tot_loss[loss=2.372, over 5547349.41 frames. , ppl: 10.722475401504251], batch size: 70
+2022-12-10 04:45:53,664 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 04:45:54,423 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.353, over 211138.00 frames. , ppl: 10.515346212674748
+2022-12-10 04:47:35,224 INFO [train.py:421] (5/8) Epoch 1, batch 58200, loss[loss=2.34, over 3850.00 frames. , ppl: 10.37700582827308] tot_loss[loss=2.371, over 5579270.54 frames. , ppl: 10.703419276172218], batch size: 70
+2022-12-10 04:49:15,831 INFO [train.py:421] (5/8) Epoch 1, batch 58400, loss[loss=2.298, over 4690.00 frames. , ppl: 9.958317589965167] tot_loss[loss=2.372, over 5514776.42 frames. , ppl: 10.722914002368059], batch size: 70
+2022-12-10 04:50:58,466 INFO [train.py:421] (5/8) Epoch 1, batch 58600, loss[loss=2.746, over 910.00 frames. , ppl: 15.58244132756225] tot_loss[loss=2.372, over 5536252.42 frames. , ppl: 10.723645569130529], batch size: 70
+2022-12-10 04:52:37,253 INFO [train.py:421] (5/8) Epoch 1, batch 58800, loss[loss=2.254, over 7280.00 frames. , ppl: 9.524333167072763] tot_loss[loss=2.373, over 5531572.43 frames. , ppl: 10.724286146072705], batch size: 70
+2022-12-10 04:54:20,777 INFO [train.py:421] (5/8) Epoch 1, batch 59000, loss[loss=2.921, over 630.00 frames. , ppl: 18.55446550232416] tot_loss[loss=2.371, over 5546903.07 frames. , ppl: 10.712909293528654], batch size: 70
+2022-12-10 04:54:20,778 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 04:54:21,536 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.352, over 211138.00 frames. , ppl: 10.503966120477648
+2022-12-10 04:55:59,455 INFO [train.py:421] (5/8) Epoch 1, batch 59200, loss[loss=2.401, over 2800.00 frames. , ppl: 11.034550664453537] tot_loss[loss=2.372, over 5510486.02 frames. , ppl: 10.719732069624715], batch size: 70
+2022-12-10 04:57:42,858 INFO [train.py:421] (5/8) Epoch 1, batch 59400, loss[loss=2.427, over 1820.00 frames. , ppl: 11.320662226071228] tot_loss[loss=2.373, over 5497303.30 frames. , ppl: 10.725531390819372], batch size: 70
+2022-12-10 04:59:24,822 INFO [train.py:421] (5/8) Epoch 1, batch 59600, loss[loss=2.287, over 3990.00 frames. , ppl: 9.8493723102438] tot_loss[loss=2.373, over 5487049.36 frames. , ppl: 10.727082151366893], batch size: 70
+2022-12-10 05:01:09,076 INFO [train.py:421] (5/8) Epoch 1, batch 59800, loss[loss=2.288, over 2800.00 frames. , ppl: 9.853523103201079] tot_loss[loss=2.372, over 5488062.08 frames. , ppl: 10.715559417397738], batch size: 70
+2022-12-10 05:02:50,323 INFO [train.py:421] (5/8) Epoch 1, batch 60000, loss[loss=2.749, over 630.00 frames. , ppl: 15.634105104702739] tot_loss[loss=2.372, over 5485583.98 frames. , ppl: 10.721069732755243], batch size: 70
+2022-12-10 05:02:50,324 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 05:02:51,100 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.353, over 211138.00 frames. , ppl: 10.521841308858221
+2022-12-10 05:04:30,567 INFO [train.py:421] (5/8) Epoch 1, batch 60200, loss[loss=2.504, over 910.00 frames. , ppl: 12.227478053453678] tot_loss[loss=2.372, over 5496372.87 frames. , ppl: 10.713842051685228], batch size: 70
+2022-12-10 05:06:11,857 INFO [train.py:421] (5/8) Epoch 1, batch 60400, loss[loss=2.283, over 7000.00 frames. , ppl: 9.806095530897958] tot_loss[loss=2.371, over 5507679.23 frames. , ppl: 10.71291010781861], batch size: 70
+2022-12-10 05:07:53,808 INFO [train.py:421] (5/8) Epoch 1, batch 60600, loss[loss=2.385, over 2240.00 frames. , ppl: 10.862284275801796] tot_loss[loss=2.371, over 5491826.05 frames. , ppl: 10.711815972459098], batch size: 70
+2022-12-10 05:09:33,763 INFO [train.py:421] (5/8) Epoch 1, batch 60800, loss[loss=2.307, over 3080.00 frames. , ppl: 10.048083549367542] tot_loss[loss=2.371, over 5496765.57 frames. , ppl: 10.708424636728951], batch size: 70
+2022-12-10 05:11:18,590 INFO [train.py:421] (5/8) Epoch 1, batch 61000, loss[loss=2.357, over 2170.00 frames. , ppl: 10.556638801676902] tot_loss[loss=2.37, over 5511085.73 frames. , ppl: 10.70089453225178], batch size: 70
+2022-12-10 05:11:18,590 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 05:11:19,337 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.352, over 211138.00 frames. , ppl: 10.508142781873898
+2022-12-10 05:12:58,120 INFO [train.py:421] (5/8) Epoch 1, batch 61200, loss[loss=5.033, over 280.00 frames. , ppl: 153.3315853167261] tot_loss[loss=2.371, over 5505618.64 frames. , ppl: 10.708095295884867], batch size: 70
+2022-12-10 05:14:39,407 INFO [train.py:421] (5/8) Epoch 1, batch 61400, loss[loss=2.526, over 1120.00 frames. , ppl: 12.508426043518904] tot_loss[loss=2.37, over 5529515.13 frames. , ppl: 10.700061704942838], batch size: 70
+2022-12-10 05:16:19,246 INFO [train.py:421] (5/8) Epoch 1, batch 61600, loss[loss=2.654, over 770.00 frames. , ppl: 14.215776393366358] tot_loss[loss=2.371, over 5510505.20 frames. , ppl: 10.710712533765262], batch size: 70
+2022-12-10 05:17:59,811 INFO [train.py:421] (5/8) Epoch 1, batch 61800, loss[loss=2.316, over 5810.00 frames. , ppl: 10.137371607970442] tot_loss[loss=2.372, over 5483207.80 frames. , ppl: 10.72104394032309], batch size: 70
+2022-12-10 05:19:41,042 INFO [train.py:421] (5/8) Epoch 1, batch 62000, loss[loss=3.206, over 490.00 frames. , ppl: 24.689401520174837] tot_loss[loss=2.372, over 5501440.83 frames. , ppl: 10.714552635900697], batch size: 70
+2022-12-10 05:19:41,042 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 05:19:41,789 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.351, over 211138.00 frames. , ppl: 10.500484247218756
+2022-12-10 05:21:23,028 INFO [train.py:421] (5/8) Epoch 1, batch 62200, loss[loss=2.875, over 700.00 frames. , ppl: 17.716777421026684] tot_loss[loss=2.372, over 5504606.05 frames. , ppl: 10.719389860509315], batch size: 70
+2022-12-10 05:23:03,310 INFO [train.py:421] (5/8) Epoch 1, batch 62400, loss[loss=2.442, over 1960.00 frames. , ppl: 11.491848230221988] tot_loss[loss=2.371, over 5500469.73 frames. , ppl: 10.712538189083894], batch size: 70
+2022-12-10 05:24:44,528 INFO [train.py:421] (5/8) Epoch 1, batch 62600, loss[loss=2.26, over 6020.00 frames. , ppl: 9.582341760723212] tot_loss[loss=2.37, over 5544462.10 frames. , ppl: 10.696297409825641], batch size: 70
+2022-12-10 05:26:22,733 INFO [train.py:421] (5/8) Epoch 1, batch 62800, loss[loss=2.477, over 1820.00 frames. , ppl: 11.907913366903617] tot_loss[loss=2.371, over 5504706.81 frames. , ppl: 10.707444109760038], batch size: 70
+2022-12-10 05:28:05,201 INFO [train.py:421] (5/8) Epoch 1, batch 63000, loss[loss=2.487, over 1050.00 frames. , ppl: 12.027183311047429] tot_loss[loss=2.37, over 5521630.21 frames. , ppl: 10.699920720738561], batch size: 70
+2022-12-10 05:28:05,201 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 05:28:05,959 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.352, over 211138.00 frames. , ppl: 10.50832786224836
+2022-12-10 05:29:47,018 INFO [train.py:421] (5/8) Epoch 1, batch 63200, loss[loss=2.435, over 1960.00 frames. , ppl: 11.417187301247733] tot_loss[loss=2.37, over 5520138.62 frames. , ppl: 10.698992884149067], batch size: 70
+2022-12-10 05:31:24,706 INFO [train.py:421] (5/8) Epoch 1, batch 63400, loss[loss=3.205, over 560.00 frames. , ppl: 24.645668842167286] tot_loss[loss=2.37, over 5531512.20 frames. , ppl: 10.699343901247788], batch size: 70
+2022-12-10 05:33:03,873 INFO [train.py:421] (5/8) Epoch 1, batch 63600, loss[loss=2.333, over 7350.00 frames. , ppl: 10.31208860475894] tot_loss[loss=2.37, over 5535239.66 frames. , ppl: 10.696370685112312], batch size: 70
+2022-12-10 05:34:41,284 INFO [train.py:421] (5/8) Epoch 1, batch 63800, loss[loss=2.486, over 910.00 frames. , ppl: 12.009157477532446] tot_loss[loss=2.369, over 5533387.70 frames. , ppl: 10.691618623820865], batch size: 70
+2022-12-10 05:36:22,762 INFO [train.py:421] (5/8) Epoch 1, batch 64000, loss[loss=2.692, over 770.00 frames. , ppl: 14.767405865153204] tot_loss[loss=2.369, over 5573545.88 frames. , ppl: 10.688554213649798], batch size: 70
+2022-12-10 05:36:22,762 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 05:36:23,521 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.351, over 211138.00 frames. , ppl: 10.500185854649434
+2022-12-10 05:38:06,567 INFO [train.py:421] (5/8) Epoch 1, batch 64200, loss[loss=2.421, over 2520.00 frames. , ppl: 11.254802971911595] tot_loss[loss=2.37, over 5572067.39 frames. , ppl: 10.692796624624021], batch size: 70
+2022-12-10 05:39:47,998 INFO [train.py:421] (5/8) Epoch 1, batch 64400, loss[loss=2.347, over 3640.00 frames. , ppl: 10.453358261825333] tot_loss[loss=2.37, over 5550320.50 frames. , ppl: 10.699045961615596], batch size: 70
+2022-12-10 05:41:29,595 INFO [train.py:421] (5/8) Epoch 1, batch 64600, loss[loss=2.293, over 6020.00 frames. , ppl: 9.90018684136552] tot_loss[loss=2.369, over 5602887.16 frames. , ppl: 10.688145424162746], batch size: 70
+2022-12-10 05:43:06,386 INFO [train.py:421] (5/8) Epoch 1, batch 64800, loss[loss=2.252, over 7910.00 frames. , ppl: 9.50748421291946] tot_loss[loss=2.369, over 5599057.85 frames. , ppl: 10.681748096703771], batch size: 70
+2022-12-10 05:44:45,913 INFO [train.py:421] (5/8) Epoch 1, batch 65000, loss[loss=2.269, over 2660.00 frames. , ppl: 9.67111100598519] tot_loss[loss=2.367, over 5651144.59 frames. , ppl: 10.664160952780364], batch size: 70
+2022-12-10 05:44:45,914 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 05:44:46,659 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.351, over 211138.00 frames. , ppl: 10.499536258484806
+2022-12-10 05:46:27,214 INFO [train.py:421] (5/8) Epoch 1, batch 65200, loss[loss=2.28, over 9520.00 frames. , ppl: 9.776641898521207] tot_loss[loss=2.367, over 5649828.44 frames. , ppl: 10.66190967747806], batch size: 70
+2022-12-10 05:48:06,794 INFO [train.py:421] (5/8) Epoch 1, batch 65400, loss[loss=2.363, over 1960.00 frames. , ppl: 10.624886574171217] tot_loss[loss=2.365, over 5720205.54 frames. , ppl: 10.64312282613914], batch size: 70
+2022-12-10 05:49:47,266 INFO [train.py:421] (5/8) Epoch 1, batch 65600, loss[loss=2.478, over 1890.00 frames. , ppl: 11.912067715286778] tot_loss[loss=2.367, over 5647114.31 frames. , ppl: 10.663100199796158], batch size: 70
+2022-12-10 05:51:25,721 INFO [train.py:421] (5/8) Epoch 1, batch 65800, loss[loss=2.334, over 3080.00 frames. , ppl: 10.320598970468595] tot_loss[loss=2.367, over 5622763.87 frames. , ppl: 10.66462332747785], batch size: 70
+2022-12-10 05:53:05,416 INFO [train.py:421] (5/8) Epoch 1, batch 66000, loss[loss=2.296, over 4270.00 frames. , ppl: 9.933967634736108] tot_loss[loss=2.368, over 5618540.11 frames. , ppl: 10.67329258817825], batch size: 70
+2022-12-10 05:53:05,417 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 05:53:06,133 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.349, over 211138.00 frames. , ppl: 10.479744419827652
+2022-12-10 05:54:45,806 INFO [train.py:421] (5/8) Epoch 1, batch 66200, loss[loss=2.208, over 4060.00 frames. , ppl: 9.093001932700043] tot_loss[loss=2.368, over 5590044.01 frames. , ppl: 10.678462664878039], batch size: 70
+2022-12-10 05:56:26,059 INFO [train.py:421] (5/8) Epoch 1, batch 66400, loss[loss=2.371, over 2310.00 frames. , ppl: 10.710828532949263] tot_loss[loss=2.369, over 5545873.62 frames. , ppl: 10.688089964108581], batch size: 70
+2022-12-10 05:58:06,911 INFO [train.py:421] (5/8) Epoch 1, batch 66600, loss[loss=2.565, over 1190.00 frames. , ppl: 13.006475271715477] tot_loss[loss=2.367, over 5612447.75 frames. , ppl: 10.663836694393883], batch size: 70
+2022-12-10 05:59:44,036 INFO [train.py:421] (5/8) Epoch 1, batch 66800, loss[loss=2.271, over 5250.00 frames. , ppl: 9.685326218644013] tot_loss[loss=2.367, over 5611382.53 frames. , ppl: 10.66687273345966], batch size: 70
+2022-12-10 06:01:21,599 INFO [train.py:421] (5/8) Epoch 1, batch 67000, loss[loss=2.283, over 5110.00 frames. , ppl: 9.804770647609175] tot_loss[loss=2.366, over 5643026.60 frames. , ppl: 10.650772138874016], batch size: 70
+2022-12-10 06:01:21,600 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 06:01:22,329 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.348, over 211138.00 frames. , ppl: 10.464386029268178
+2022-12-10 06:02:59,568 INFO [train.py:421] (5/8) Epoch 1, batch 67200, loss[loss=2.414, over 980.00 frames. , ppl: 11.173788589806295] tot_loss[loss=2.366, over 5632508.91 frames. , ppl: 10.653868342274334], batch size: 70
+2022-12-10 06:04:38,155 INFO [train.py:421] (5/8) Epoch 1, batch 67400, loss[loss=2.428, over 1680.00 frames. , ppl: 11.338790465229597] tot_loss[loss=2.365, over 5669148.09 frames. , ppl: 10.644964954434236], batch size: 70
+2022-12-10 06:06:18,488 INFO [train.py:421] (5/8) Epoch 1, batch 67600, loss[loss=2.484, over 1260.00 frames. , ppl: 11.985843960629254] tot_loss[loss=2.365, over 5655738.02 frames. , ppl: 10.642620490327028], batch size: 70
+2022-12-10 06:08:00,614 INFO [train.py:421] (5/8) Epoch 1, batch 67800, loss[loss=2.277, over 4130.00 frames. , ppl: 9.749431447062175] tot_loss[loss=2.365, over 5649572.55 frames. , ppl: 10.642665002176452], batch size: 70
+2022-12-10 06:09:41,544 INFO [train.py:421] (5/8) Epoch 1, batch 68000, loss[loss=2.595, over 1120.00 frames. , ppl: 13.397173171209435] tot_loss[loss=2.366, over 5623725.11 frames. , ppl: 10.654936724862473], batch size: 70
+2022-12-10 06:09:41,545 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 06:09:42,303 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.35, over 211138.00 frames. , ppl: 10.481405758605055
+2022-12-10 06:11:26,715 INFO [train.py:421] (5/8) Epoch 1, batch 68200, loss[loss=2.752, over 700.00 frames. , ppl: 15.666114531558996] tot_loss[loss=2.367, over 5577962.33 frames. , ppl: 10.663676461401943], batch size: 70
+2022-12-10 06:13:04,743 INFO [train.py:421] (5/8) Epoch 1, batch 68400, loss[loss=2.266, over 7000.00 frames. , ppl: 9.643187178028901] tot_loss[loss=2.367, over 5527761.18 frames. , ppl: 10.667357374058284], batch size: 70
+2022-12-10 06:14:45,555 INFO [train.py:421] (5/8) Epoch 1, batch 68600, loss[loss=2.366, over 3360.00 frames. , ppl: 10.65107572913184] tot_loss[loss=2.368, over 5510258.14 frames. , ppl: 10.672984997856295], batch size: 70
+2022-12-10 06:16:28,484 INFO [train.py:421] (5/8) Epoch 1, batch 68800, loss[loss=2.642, over 1050.00 frames. , ppl: 14.042111505083636] tot_loss[loss=2.367, over 5528658.55 frames. , ppl: 10.667640923483107], batch size: 70
+2022-12-10 06:18:08,237 INFO [train.py:421] (5/8) Epoch 1, batch 69000, loss[loss=2.544, over 910.00 frames. , ppl: 12.72917007585484] tot_loss[loss=2.368, over 5491825.96 frames. , ppl: 10.672371976400935], batch size: 70
+2022-12-10 06:18:08,238 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 06:18:08,996 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.348, over 211138.00 frames. , ppl: 10.467406639264686
+2022-12-10 06:19:44,539 INFO [train.py:421] (5/8) Epoch 1, batch 69200, loss[loss=2.439, over 1680.00 frames. , ppl: 11.457670029690918] tot_loss[loss=2.368, over 5473050.57 frames. , ppl: 10.67198325453401], batch size: 70
+2022-12-10 06:21:26,218 INFO [train.py:421] (5/8) Epoch 1, batch 69400, loss[loss=2.342, over 3990.00 frames. , ppl: 10.399639852566974] tot_loss[loss=2.367, over 5505214.12 frames. , ppl: 10.660720797849564], batch size: 70
+2022-12-10 06:23:06,573 INFO [train.py:421] (5/8) Epoch 1, batch 69600, loss[loss=2.542, over 1190.00 frames. , ppl: 12.709153671859475] tot_loss[loss=2.367, over 5479378.50 frames. , ppl: 10.66254054890111], batch size: 70
+2022-12-10 06:24:43,354 INFO [train.py:421] (5/8) Epoch 1, batch 69800, loss[loss=2.485, over 1470.00 frames. , ppl: 11.997602425509413] tot_loss[loss=2.369, over 5414896.79 frames. , ppl: 10.681778263570932], batch size: 70
+2022-12-10 06:26:25,254 INFO [train.py:421] (5/8) Epoch 1, batch 70000, loss[loss=2.426, over 1680.00 frames. , ppl: 11.310952605766396] tot_loss[loss=2.368, over 5445783.60 frames. , ppl: 10.675027758090435], batch size: 70
+2022-12-10 06:26:25,254 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 06:26:26,013 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.348, over 211138.00 frames. , ppl: 10.463410326196138
+2022-12-10 06:28:04,613 INFO [train.py:421] (5/8) Epoch 1, batch 70200, loss[loss=2.872, over 840.00 frames. , ppl: 17.672029217174035] tot_loss[loss=2.368, over 5453765.03 frames. , ppl: 10.671221476215782], batch size: 70
+2022-12-10 06:29:39,976 INFO [train.py:421] (5/8) Epoch 1, batch 70400, loss[loss=2.379, over 3430.00 frames. , ppl: 10.79024016543502] tot_loss[loss=2.367, over 5466273.80 frames. , ppl: 10.66111519617484], batch size: 70
+2022-12-10 06:31:16,084 INFO [train.py:421] (5/8) Epoch 1, batch 70600, loss[loss=2.363, over 2870.00 frames. , ppl: 10.628034532711021] tot_loss[loss=2.366, over 5503165.50 frames. , ppl: 10.652416688153048], batch size: 70
+2022-12-10 06:32:52,996 INFO [train.py:421] (5/8) Epoch 1, batch 70800, loss[loss=2.586, over 840.00 frames. , ppl: 13.272322630474305] tot_loss[loss=2.367, over 5445615.33 frames. , ppl: 10.664306467061916], batch size: 70
+2022-12-10 06:34:35,544 INFO [train.py:421] (5/8) Epoch 1, batch 71000, loss[loss=2.443, over 3360.00 frames. , ppl: 11.511079030824535] tot_loss[loss=2.368, over 5414354.64 frames. , ppl: 10.675194084548743], batch size: 70
+2022-12-10 06:34:35,544 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 06:34:36,305 INFO [train.py:452] (5/8) Epoch 1, validation: loss=2.347, over 211138.00 frames. , ppl: 10.45096802929164
+2022-12-10 06:36:12,870 INFO [train.py:421] (5/8) Epoch 1, batch 71200, loss[loss=2.477, over 1120.00 frames. , ppl: 11.900930135497852] tot_loss[loss=2.369, over 5394162.45 frames. , ppl: 10.68245823340505], batch size: 70
+2022-12-10 06:37:52,675 INFO [train.py:421] (5/8) Epoch 1, batch 71400, loss[loss=2.393, over 1120.00 frames. , ppl: 10.951020159647413] tot_loss[loss=2.369, over 5391010.65 frames. , ppl: 10.687726744142854], batch size: 70
+2022-12-10 06:39:34,127 INFO [train.py:421] (5/8) Epoch 1, batch 71600, loss[loss=3.597, over 420.00 frames. , ppl: 36.5040053891713] tot_loss[loss=2.368, over 5398706.41 frames. , ppl: 10.680122007688302], batch size: 70
+2022-12-10 06:41:16,765 INFO [train.py:421] (5/8) Epoch 1, batch 71800, loss[loss=2.654, over 840.00 frames. , ppl: 14.212141316303367] tot_loss[loss=2.367, over 5446236.23 frames. , ppl: 10.670376546069555], batch size: 70
+2022-12-10 06:42:32,558 INFO [train.py:421] (5/8) Epoch 2, batch 0, loss[loss=2.775, over 700.00 frames. , ppl: 16.037983718201776] tot_loss[loss=2.775, over 700.00 frames. , ppl: 16.037983718201776], batch size: 70
+2022-12-10 06:44:11,579 INFO [train.py:421] (5/8) Epoch 2, batch 200, loss[loss=2.447, over 2590.00 frames. , ppl: 11.558611844838818] tot_loss[loss=2.364, over 515272.89 frames. , ppl: 10.63104343633669], batch size: 70
+2022-12-10 06:45:50,262 INFO [train.py:421] (5/8) Epoch 2, batch 400, loss[loss=2.272, over 3010.00 frames. , ppl: 9.696138413323448] tot_loss[loss=2.356, over 981372.98 frames. , ppl: 10.553751532339025], batch size: 70
+2022-12-10 06:47:30,606 INFO [train.py:421] (5/8) Epoch 2, batch 600, loss[loss=2.422, over 1400.00 frames. , ppl: 11.262890817425278] tot_loss[loss=2.357, over 1408695.12 frames. , ppl: 10.556152935329441], batch size: 70
+2022-12-10 06:49:11,810 INFO [train.py:421] (5/8) Epoch 2, batch 800, loss[loss=2.364, over 2940.00 frames. , ppl: 10.629468471296942] tot_loss[loss=2.357, over 1807052.21 frames. , ppl: 10.56054050764657], batch size: 70
+2022-12-10 06:50:52,992 INFO [train.py:421] (5/8) Epoch 2, batch 1000, loss[loss=2.398, over 2240.00 frames. , ppl: 11.000133026808003] tot_loss[loss=2.356, over 2181702.32 frames. , ppl: 10.5479674515083], batch size: 70
+2022-12-10 06:50:52,992 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 06:50:53,753 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.346, over 211138.00 frames. , ppl: 10.441904495381296
+2022-12-10 06:52:33,130 INFO [train.py:421] (5/8) Epoch 2, batch 1200, loss[loss=2.91, over 630.00 frames. , ppl: 18.3579275482603] tot_loss[loss=2.357, over 2493732.13 frames. , ppl: 10.55475209964109], batch size: 70
+2022-12-10 06:54:14,266 INFO [train.py:421] (5/8) Epoch 2, batch 1400, loss[loss=2.338, over 3640.00 frames. , ppl: 10.360832062363102] tot_loss[loss=2.359, over 2772888.51 frames. , ppl: 10.580456247103356], batch size: 70
+2022-12-10 06:55:55,748 INFO [train.py:421] (5/8) Epoch 2, batch 1600, loss[loss=2.472, over 1470.00 frames. , ppl: 11.846818506671903] tot_loss[loss=2.36, over 3024209.26 frames. , ppl: 10.587150465405402], batch size: 70
+2022-12-10 06:57:35,263 INFO [train.py:421] (5/8) Epoch 2, batch 1800, loss[loss=2.324, over 3780.00 frames. , ppl: 10.219762867282407] tot_loss[loss=2.361, over 3218584.03 frames. , ppl: 10.599569681767337], batch size: 70
+2022-12-10 06:59:14,992 INFO [train.py:421] (5/8) Epoch 2, batch 2000, loss[loss=2.574, over 1610.00 frames. , ppl: 13.11994019189783] tot_loss[loss=2.36, over 3430418.97 frames. , ppl: 10.595634405464612], batch size: 70
+2022-12-10 06:59:14,993 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 06:59:15,755 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.345, over 211138.00 frames. , ppl: 10.430619322580018
+2022-12-10 07:00:58,121 INFO [train.py:421] (5/8) Epoch 2, batch 2200, loss[loss=2.355, over 5250.00 frames. , ppl: 10.535696512628009] tot_loss[loss=2.36, over 3646473.84 frames. , ppl: 10.587416558304183], batch size: 70
+2022-12-10 07:02:37,017 INFO [train.py:421] (5/8) Epoch 2, batch 2400, loss[loss=2.287, over 4830.00 frames. , ppl: 9.84364342235659] tot_loss[loss=2.361, over 3779557.51 frames. , ppl: 10.600486437753359], batch size: 70
+2022-12-10 07:04:16,524 INFO [train.py:421] (5/8) Epoch 2, batch 2600, loss[loss=2.7, over 840.00 frames. , ppl: 14.885644762726738] tot_loss[loss=2.361, over 3942414.94 frames. , ppl: 10.604431929327815], batch size: 70
+2022-12-10 07:05:57,378 INFO [train.py:421] (5/8) Epoch 2, batch 2800, loss[loss=3.551, over 420.00 frames. , ppl: 34.85752188618613] tot_loss[loss=2.361, over 4090518.08 frames. , ppl: 10.599605155985765], batch size: 70
+2022-12-10 07:07:34,871 INFO [train.py:421] (5/8) Epoch 2, batch 3000, loss[loss=2.429, over 1610.00 frames. , ppl: 11.352027370544313] tot_loss[loss=2.362, over 4192274.60 frames. , ppl: 10.607922020098469], batch size: 70
+2022-12-10 07:07:34,871 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 07:07:35,617 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.346, over 211138.00 frames. , ppl: 10.439906381190209
+2022-12-10 07:09:13,521 INFO [train.py:421] (5/8) Epoch 2, batch 3200, loss[loss=2.275, over 6160.00 frames. , ppl: 9.724135202219024] tot_loss[loss=2.362, over 4308028.14 frames. , ppl: 10.608475659576696], batch size: 70
+2022-12-10 07:10:54,074 INFO [train.py:421] (5/8) Epoch 2, batch 3400, loss[loss=2.514, over 1330.00 frames. , ppl: 12.359402956611541] tot_loss[loss=2.361, over 4400499.37 frames. , ppl: 10.606526435915175], batch size: 70
+2022-12-10 07:12:38,050 INFO [train.py:421] (5/8) Epoch 2, batch 3600, loss[loss=2.309, over 2450.00 frames. , ppl: 10.06667144128821] tot_loss[loss=2.359, over 4505478.87 frames. , ppl: 10.585489280326192], batch size: 70
+2022-12-10 07:14:17,898 INFO [train.py:421] (5/8) Epoch 2, batch 3800, loss[loss=2.317, over 6300.00 frames. , ppl: 10.148986232463791] tot_loss[loss=2.36, over 4586627.62 frames. , ppl: 10.589512112404156], batch size: 70
+2022-12-10 07:15:59,649 INFO [train.py:421] (5/8) Epoch 2, batch 4000, loss[loss=3.328, over 490.00 frames. , ppl: 27.87026188080791] tot_loss[loss=2.359, over 4690353.38 frames. , ppl: 10.581154082377022], batch size: 70
+2022-12-10 07:15:59,649 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 07:16:00,396 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.345, over 211138.00 frames. , ppl: 10.432709850132296
+2022-12-10 07:17:44,062 INFO [train.py:421] (5/8) Epoch 2, batch 4200, loss[loss=2.394, over 1610.00 frames. , ppl: 10.955214013415608] tot_loss[loss=2.358, over 4794575.83 frames. , ppl: 10.574561573607724], batch size: 70
+2022-12-10 07:19:26,559 INFO [train.py:421] (5/8) Epoch 2, batch 4400, loss[loss=2.369, over 2310.00 frames. , ppl: 10.688284494657752] tot_loss[loss=2.358, over 4865055.78 frames. , ppl: 10.567859023225859], batch size: 70
+2022-12-10 07:21:09,714 INFO [train.py:421] (5/8) Epoch 2, batch 4600, loss[loss=2.305, over 2940.00 frames. , ppl: 10.027532459423087] tot_loss[loss=2.358, over 4920548.09 frames. , ppl: 10.570534875279034], batch size: 70
+2022-12-10 07:22:49,446 INFO [train.py:421] (5/8) Epoch 2, batch 4800, loss[loss=2.364, over 2240.00 frames. , ppl: 10.630407465745375] tot_loss[loss=2.357, over 4990933.71 frames. , ppl: 10.561013814417093], batch size: 70
+2022-12-10 07:24:27,438 INFO [train.py:421] (5/8) Epoch 2, batch 5000, loss[loss=2.278, over 7350.00 frames. , ppl: 9.75792030502244] tot_loss[loss=2.357, over 5021248.33 frames. , ppl: 10.561022265408273], batch size: 70
+2022-12-10 07:24:27,439 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 07:24:28,168 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.343, over 211138.00 frames. , ppl: 10.417051893133054
+2022-12-10 07:26:10,651 INFO [train.py:421] (5/8) Epoch 2, batch 5200, loss[loss=2.626, over 910.00 frames. , ppl: 13.814184615089065] tot_loss[loss=2.357, over 5110525.07 frames. , ppl: 10.55644075707955], batch size: 70
+2022-12-10 07:27:50,416 INFO [train.py:421] (5/8) Epoch 2, batch 5400, loss[loss=2.54, over 1190.00 frames. , ppl: 12.684886216462917] tot_loss[loss=2.357, over 5138729.29 frames. , ppl: 10.559958476952467], batch size: 70
+2022-12-10 07:29:28,146 INFO [train.py:421] (5/8) Epoch 2, batch 5600, loss[loss=2.574, over 910.00 frames. , ppl: 13.123843578049948] tot_loss[loss=2.358, over 5154764.12 frames. , ppl: 10.567924717864745], batch size: 70
+2022-12-10 07:31:06,307 INFO [train.py:421] (5/8) Epoch 2, batch 5800, loss[loss=2.348, over 2940.00 frames. , ppl: 10.461245335731945] tot_loss[loss=2.357, over 5216502.91 frames. , ppl: 10.555296952920704], batch size: 70
+2022-12-10 07:32:43,819 INFO [train.py:421] (5/8) Epoch 2, batch 6000, loss[loss=2.333, over 3850.00 frames. , ppl: 10.312532284820865] tot_loss[loss=2.356, over 5288660.17 frames. , ppl: 10.54594291702513], batch size: 70
+2022-12-10 07:32:43,820 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 07:32:44,568 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.343, over 211138.00 frames. , ppl: 10.413985701421621
+2022-12-10 07:34:23,914 INFO [train.py:421] (5/8) Epoch 2, batch 6200, loss[loss=2.378, over 2730.00 frames. , ppl: 10.78049892028499] tot_loss[loss=2.355, over 5288011.87 frames. , ppl: 10.541138656579598], batch size: 70
+2022-12-10 07:36:04,956 INFO [train.py:421] (5/8) Epoch 2, batch 6400, loss[loss=2.512, over 910.00 frames. , ppl: 12.328247032700698] tot_loss[loss=2.355, over 5321489.66 frames. , ppl: 10.543129166758838], batch size: 70
+2022-12-10 07:37:46,153 INFO [train.py:421] (5/8) Epoch 2, batch 6600, loss[loss=2.285, over 4760.00 frames. , ppl: 9.82563502182383] tot_loss[loss=2.354, over 5396018.56 frames. , ppl: 10.531989680492506], batch size: 70
+2022-12-10 07:39:25,805 INFO [train.py:421] (5/8) Epoch 2, batch 6800, loss[loss=2.288, over 4830.00 frames. , ppl: 9.860052807467431] tot_loss[loss=2.354, over 5421241.29 frames. , ppl: 10.531766486936798], batch size: 70
+2022-12-10 07:41:05,167 INFO [train.py:421] (5/8) Epoch 2, batch 7000, loss[loss=2.368, over 2730.00 frames. , ppl: 10.680887807331457] tot_loss[loss=2.354, over 5500298.27 frames. , ppl: 10.522518563297979], batch size: 70
+2022-12-10 07:41:05,168 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 07:41:05,947 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.342, over 211138.00 frames. , ppl: 10.402274966255137
+2022-12-10 07:42:45,100 INFO [train.py:421] (5/8) Epoch 2, batch 7200, loss[loss=2.449, over 1820.00 frames. , ppl: 11.577178373432734] tot_loss[loss=2.354, over 5496162.05 frames. , ppl: 10.526229365564829], batch size: 70
+2022-12-10 07:44:25,815 INFO [train.py:421] (5/8) Epoch 2, batch 7400, loss[loss=2.302, over 2940.00 frames. , ppl: 9.999008281653026] tot_loss[loss=2.353, over 5522492.99 frames. , ppl: 10.521880391814195], batch size: 70
+2022-12-10 07:46:04,861 INFO [train.py:421] (5/8) Epoch 2, batch 7600, loss[loss=2.712, over 980.00 frames. , ppl: 15.06611499913337] tot_loss[loss=2.354, over 5505698.91 frames. , ppl: 10.530044256092497], batch size: 70
+2022-12-10 07:47:41,332 INFO [train.py:421] (5/8) Epoch 2, batch 7800, loss[loss=2.363, over 3290.00 frames. , ppl: 10.627864026769233] tot_loss[loss=2.355, over 5503145.02 frames. , ppl: 10.53655508535715], batch size: 70
+2022-12-10 07:49:26,401 INFO [train.py:421] (5/8) Epoch 2, batch 8000, loss[loss=2.527, over 980.00 frames. , ppl: 12.514411238429608] tot_loss[loss=2.355, over 5530926.59 frames. , ppl: 10.534924188143846], batch size: 70
+2022-12-10 07:49:26,401 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 07:49:27,161 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.343, over 211138.00 frames. , ppl: 10.408854278509716
+2022-12-10 07:51:05,854 INFO [train.py:421] (5/8) Epoch 2, batch 8200, loss[loss=2.318, over 2800.00 frames. , ppl: 10.152752369832587] tot_loss[loss=2.355, over 5526178.87 frames. , ppl: 10.534016727208146], batch size: 70
+2022-12-10 07:52:44,237 INFO [train.py:421] (5/8) Epoch 2, batch 8400, loss[loss=2.639, over 980.00 frames. , ppl: 13.99637586794353] tot_loss[loss=2.354, over 5515651.41 frames. , ppl: 10.532218730333737], batch size: 70
+2022-12-10 07:54:24,068 INFO [train.py:421] (5/8) Epoch 2, batch 8600, loss[loss=2.461, over 1190.00 frames. , ppl: 11.713920083912592] tot_loss[loss=2.354, over 5517483.34 frames. , ppl: 10.52636472161019], batch size: 70
+2022-12-10 07:56:01,956 INFO [train.py:421] (5/8) Epoch 2, batch 8800, loss[loss=2.447, over 1470.00 frames. , ppl: 11.5571061889641] tot_loss[loss=2.354, over 5526625.79 frames. , ppl: 10.524565682169301], batch size: 70
+2022-12-10 07:57:42,022 INFO [train.py:421] (5/8) Epoch 2, batch 9000, loss[loss=2.33, over 2450.00 frames. , ppl: 10.282473544105917] tot_loss[loss=2.353, over 5554035.53 frames. , ppl: 10.52037533257404], batch size: 70
+2022-12-10 07:57:42,022 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 07:57:42,768 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.342, over 211138.00 frames. , ppl: 10.399154636076556
+2022-12-10 07:59:22,462 INFO [train.py:421] (5/8) Epoch 2, batch 9200, loss[loss=2.441, over 1960.00 frames. , ppl: 11.485251175271124] tot_loss[loss=2.354, over 5525718.63 frames. , ppl: 10.532419951929047], batch size: 70
+2022-12-10 08:00:56,951 INFO [train.py:421] (5/8) Epoch 2, batch 9400, loss[loss=2.353, over 2940.00 frames. , ppl: 10.516797062849031] tot_loss[loss=2.356, over 5466353.46 frames. , ppl: 10.545977283150595], batch size: 70
+2022-12-10 08:02:37,352 INFO [train.py:421] (5/8) Epoch 2, batch 9600, loss[loss=2.489, over 1260.00 frames. , ppl: 12.046096441032498] tot_loss[loss=2.356, over 5449748.79 frames. , ppl: 10.551942216797345], batch size: 70
+2022-12-10 08:04:17,627 INFO [train.py:421] (5/8) Epoch 2, batch 9800, loss[loss=2.304, over 4340.00 frames. , ppl: 10.01354024915979] tot_loss[loss=2.356, over 5436348.67 frames. , ppl: 10.551365153614231], batch size: 70
+2022-12-10 08:06:01,213 INFO [train.py:421] (5/8) Epoch 2, batch 10000, loss[loss=2.455, over 2660.00 frames. , ppl: 11.650317851757226] tot_loss[loss=2.355, over 5492872.39 frames. , ppl: 10.534173552103272], batch size: 70
+2022-12-10 08:06:01,214 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 08:06:01,975 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.343, over 211138.00 frames. , ppl: 10.415013831030345
+2022-12-10 08:07:39,459 INFO [train.py:421] (5/8) Epoch 2, batch 10200, loss[loss=2.546, over 1190.00 frames. , ppl: 12.76027806981242] tot_loss[loss=2.355, over 5496054.13 frames. , ppl: 10.533342115551484], batch size: 70
+2022-12-10 08:09:21,336 INFO [train.py:421] (5/8) Epoch 2, batch 10400, loss[loss=2.353, over 3080.00 frames. , ppl: 10.517583005906] tot_loss[loss=2.354, over 5524438.18 frames. , ppl: 10.525582019794436], batch size: 70
+2022-12-10 08:11:01,483 INFO [train.py:421] (5/8) Epoch 2, batch 10600, loss[loss=2.283, over 5110.00 frames. , ppl: 9.807495489637297] tot_loss[loss=2.354, over 5502600.50 frames. , ppl: 10.524072704862549], batch size: 70
+2022-12-10 08:12:42,387 INFO [train.py:421] (5/8) Epoch 2, batch 10800, loss[loss=2.365, over 1680.00 frames. , ppl: 10.647415731877738] tot_loss[loss=2.355, over 5430624.43 frames. , ppl: 10.539957825741379], batch size: 70
+2022-12-10 08:14:21,837 INFO [train.py:421] (5/8) Epoch 2, batch 11000, loss[loss=2.659, over 770.00 frames. , ppl: 14.283062885210668] tot_loss[loss=2.355, over 5439853.80 frames. , ppl: 10.53779073610685], batch size: 70
+2022-12-10 08:14:21,837 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 08:14:22,596 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.342, over 211138.00 frames. , ppl: 10.400310603854379
+2022-12-10 08:16:01,434 INFO [train.py:421] (5/8) Epoch 2, batch 11200, loss[loss=2.169, over 1890.00 frames. , ppl: 8.74937714914424] tot_loss[loss=2.354, over 5453981.68 frames. , ppl: 10.530163997382706], batch size: 70
+2022-12-10 08:17:42,445 INFO [train.py:421] (5/8) Epoch 2, batch 11400, loss[loss=2.282, over 5180.00 frames. , ppl: 9.798768618346916] tot_loss[loss=2.354, over 5501716.04 frames. , ppl: 10.528204398099822], batch size: 70
+2022-12-10 08:19:25,307 INFO [train.py:421] (5/8) Epoch 2, batch 11600, loss[loss=2.364, over 1400.00 frames. , ppl: 10.634867707329386] tot_loss[loss=2.353, over 5521397.35 frames. , ppl: 10.520373080181741], batch size: 70
+2022-12-10 08:21:08,341 INFO [train.py:421] (5/8) Epoch 2, batch 11800, loss[loss=2.395, over 3500.00 frames. , ppl: 10.971773098991948] tot_loss[loss=2.355, over 5492230.46 frames. , ppl: 10.537849528104786], batch size: 70
+2022-12-10 08:22:51,113 INFO [train.py:421] (5/8) Epoch 2, batch 12000, loss[loss=2.308, over 5110.00 frames. , ppl: 10.050929880406494] tot_loss[loss=2.356, over 5475593.07 frames. , ppl: 10.543944975308847], batch size: 70
+2022-12-10 08:22:51,113 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 08:22:51,875 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.342, over 211138.00 frames. , ppl: 10.405858265858988
+2022-12-10 08:24:33,719 INFO [train.py:421] (5/8) Epoch 2, batch 12200, loss[loss=2.27, over 3430.00 frames. , ppl: 9.679887508098012] tot_loss[loss=2.356, over 5477305.29 frames. , ppl: 10.545841793029853], batch size: 70
+2022-12-10 08:26:16,337 INFO [train.py:421] (5/8) Epoch 2, batch 12400, loss[loss=2.272, over 12180.00 frames. , ppl: 9.698935260051266] tot_loss[loss=2.355, over 5486714.57 frames. , ppl: 10.541553195075306], batch size: 70
+2022-12-10 08:27:55,050 INFO [train.py:421] (5/8) Epoch 2, batch 12600, loss[loss=2.764, over 910.00 frames. , ppl: 15.86148773671756] tot_loss[loss=2.356, over 5481825.58 frames. , ppl: 10.54765533049407], batch size: 70
+2022-12-10 08:29:33,545 INFO [train.py:421] (5/8) Epoch 2, batch 12800, loss[loss=2.367, over 3290.00 frames. , ppl: 10.664953053489882] tot_loss[loss=2.355, over 5494714.17 frames. , ppl: 10.541478100942577], batch size: 70
+2022-12-10 08:31:14,850 INFO [train.py:421] (5/8) Epoch 2, batch 13000, loss[loss=2.238, over 3640.00 frames. , ppl: 9.375198022418674] tot_loss[loss=2.356, over 5497190.09 frames. , ppl: 10.546593897979559], batch size: 70
+2022-12-10 08:31:14,851 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 08:31:15,609 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.345, over 211138.00 frames. , ppl: 10.435761475361208
+2022-12-10 08:32:58,546 INFO [train.py:421] (5/8) Epoch 2, batch 13200, loss[loss=2.28, over 5810.00 frames. , ppl: 9.77313285242001] tot_loss[loss=2.356, over 5474150.04 frames. , ppl: 10.547963453198388], batch size: 70
+2022-12-10 08:34:37,207 INFO [train.py:421] (5/8) Epoch 2, batch 13400, loss[loss=2.366, over 2310.00 frames. , ppl: 10.652929209934731] tot_loss[loss=2.355, over 5493471.89 frames. , ppl: 10.533940468103927], batch size: 70
+2022-12-10 08:36:17,992 INFO [train.py:421] (5/8) Epoch 2, batch 13600, loss[loss=2.489, over 980.00 frames. , ppl: 12.044088632211558] tot_loss[loss=2.355, over 5471677.10 frames. , ppl: 10.53513606547333], batch size: 70
+2022-12-10 08:38:00,730 INFO [train.py:421] (5/8) Epoch 2, batch 13800, loss[loss=2.246, over 6370.00 frames. , ppl: 9.45090024028489] tot_loss[loss=2.354, over 5499523.55 frames. , ppl: 10.530861841275765], batch size: 70
+2022-12-10 08:39:39,916 INFO [train.py:421] (5/8) Epoch 2, batch 14000, loss[loss=2.311, over 4480.00 frames. , ppl: 10.081051878373339] tot_loss[loss=2.355, over 5473924.97 frames. , ppl: 10.535334717945346], batch size: 70
+2022-12-10 08:39:39,917 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 08:39:40,646 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.34, over 211138.00 frames. , ppl: 10.38519312924439
+2022-12-10 08:41:22,077 INFO [train.py:421] (5/8) Epoch 2, batch 14200, loss[loss=2.511, over 1050.00 frames. , ppl: 12.317499475351877] tot_loss[loss=2.354, over 5491132.00 frames. , ppl: 10.52972159324353], batch size: 70
+2022-12-10 08:43:05,914 INFO [train.py:421] (5/8) Epoch 2, batch 14400, loss[loss=2.32, over 6790.00 frames. , ppl: 10.17632821999816] tot_loss[loss=2.353, over 5519874.77 frames. , ppl: 10.518289807665674], batch size: 70
+2022-12-10 08:44:46,235 INFO [train.py:421] (5/8) Epoch 2, batch 14600, loss[loss=2.404, over 1610.00 frames. , ppl: 11.07264475430052] tot_loss[loss=2.354, over 5497800.21 frames. , ppl: 10.522425294421486], batch size: 70
+2022-12-10 08:46:23,519 INFO [train.py:421] (5/8) Epoch 2, batch 14800, loss[loss=2.484, over 2940.00 frames. , ppl: 11.992134022171804] tot_loss[loss=2.354, over 5498385.55 frames. , ppl: 10.526698662144518], batch size: 70
+2022-12-10 08:48:03,535 INFO [train.py:421] (5/8) Epoch 2, batch 15000, loss[loss=2.262, over 3640.00 frames. , ppl: 9.60218095839657] tot_loss[loss=2.355, over 5465991.62 frames. , ppl: 10.537248859216607], batch size: 70
+2022-12-10 08:48:03,536 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 08:48:04,267 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.339, over 211138.00 frames. , ppl: 10.36868412657422
+2022-12-10 08:49:43,366 INFO [train.py:421] (5/8) Epoch 2, batch 15200, loss[loss=2.517, over 980.00 frames. , ppl: 12.387276067855717] tot_loss[loss=2.355, over 5436274.57 frames. , ppl: 10.54083493785647], batch size: 70
+2022-12-10 08:51:24,495 INFO [train.py:421] (5/8) Epoch 2, batch 15400, loss[loss=2.531, over 1190.00 frames. , ppl: 12.56222664785693] tot_loss[loss=2.354, over 5461467.71 frames. , ppl: 10.527525960585413], batch size: 70
+2022-12-10 08:53:07,425 INFO [train.py:421] (5/8) Epoch 2, batch 15600, loss[loss=2.432, over 1120.00 frames. , ppl: 11.378857494704624] tot_loss[loss=2.353, over 5486622.10 frames. , ppl: 10.51769898190507], batch size: 70
+2022-12-10 08:54:45,435 INFO [train.py:421] (5/8) Epoch 2, batch 15800, loss[loss=2.42, over 3010.00 frames. , ppl: 11.245852747404669] tot_loss[loss=2.353, over 5481085.90 frames. , ppl: 10.515905413010467], batch size: 70
+2022-12-10 08:56:24,653 INFO [train.py:421] (5/8) Epoch 2, batch 16000, loss[loss=2.577, over 980.00 frames. , ppl: 13.153856207176222] tot_loss[loss=2.353, over 5467135.50 frames. , ppl: 10.520201755386882], batch size: 70
+2022-12-10 08:56:24,653 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 08:56:25,397 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.34, over 211138.00 frames. , ppl: 10.380303272822726
+2022-12-10 08:58:07,170 INFO [train.py:421] (5/8) Epoch 2, batch 16200, loss[loss=2.502, over 1680.00 frames. , ppl: 12.20957900199028] tot_loss[loss=2.354, over 5465228.87 frames. , ppl: 10.525554295756713], batch size: 70
+2022-12-10 08:59:47,484 INFO [train.py:421] (5/8) Epoch 2, batch 16400, loss[loss=2.315, over 3010.00 frames. , ppl: 10.124122742974935] tot_loss[loss=2.353, over 5496338.12 frames. , ppl: 10.512902952719111], batch size: 70
+2022-12-10 09:01:27,431 INFO [train.py:421] (5/8) Epoch 2, batch 16600, loss[loss=2.283, over 4970.00 frames. , ppl: 9.809199078658475] tot_loss[loss=2.35, over 5580240.96 frames. , ppl: 10.489855395565648], batch size: 70
+2022-12-10 09:03:04,464 INFO [train.py:421] (5/8) Epoch 2, batch 16800, loss[loss=2.406, over 2380.00 frames. , ppl: 11.088664662709235] tot_loss[loss=2.351, over 5564995.55 frames. , ppl: 10.497231128815319], batch size: 70
+2022-12-10 09:04:43,292 INFO [train.py:421] (5/8) Epoch 2, batch 17000, loss[loss=2.801, over 630.00 frames. , ppl: 16.46342830100631] tot_loss[loss=2.351, over 5584204.27 frames. , ppl: 10.495956827167886], batch size: 70
+2022-12-10 09:04:43,292 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 09:04:44,079 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.34, over 211138.00 frames. , ppl: 10.380625913933697
+2022-12-10 09:06:22,856 INFO [train.py:421] (5/8) Epoch 2, batch 17200, loss[loss=2.372, over 2170.00 frames. , ppl: 10.716096606499029] tot_loss[loss=2.35, over 5598953.50 frames. , ppl: 10.4850937119669], batch size: 70
+2022-12-10 09:08:07,700 INFO [train.py:421] (5/8) Epoch 2, batch 17400, loss[loss=2.411, over 2520.00 frames. , ppl: 11.145285184379418] tot_loss[loss=2.348, over 5659154.37 frames. , ppl: 10.469787239449627], batch size: 70
+2022-12-10 09:09:46,666 INFO [train.py:421] (5/8) Epoch 2, batch 17600, loss[loss=2.272, over 5880.00 frames. , ppl: 9.698913264063307] tot_loss[loss=2.349, over 5651763.59 frames. , ppl: 10.470402677836441], batch size: 70
+2022-12-10 09:11:23,680 INFO [train.py:421] (5/8) Epoch 2, batch 17800, loss[loss=2.325, over 2450.00 frames. , ppl: 10.231560609078466] tot_loss[loss=2.348, over 5653727.32 frames. , ppl: 10.46952484881885], batch size: 70
+2022-12-10 09:13:01,348 INFO [train.py:421] (5/8) Epoch 2, batch 18000, loss[loss=2.634, over 770.00 frames. , ppl: 13.93337020553732] tot_loss[loss=2.349, over 5635733.48 frames. , ppl: 10.471585068054347], batch size: 70
+2022-12-10 09:13:01,349 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 09:13:02,108 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.339, over 211138.00 frames. , ppl: 10.371116820501767
+2022-12-10 09:14:39,902 INFO [train.py:421] (5/8) Epoch 2, batch 18200, loss[loss=2.381, over 1750.00 frames. , ppl: 10.81478071811714] tot_loss[loss=2.35, over 5613220.73 frames. , ppl: 10.481990466973203], batch size: 70
+2022-12-10 09:16:18,845 INFO [train.py:421] (5/8) Epoch 2, batch 18400, loss[loss=3.027, over 560.00 frames. , ppl: 20.6272691160561] tot_loss[loss=2.351, over 5571036.59 frames. , ppl: 10.499716807476236], batch size: 70
+2022-12-10 09:17:58,062 INFO [train.py:421] (5/8) Epoch 2, batch 18600, loss[loss=2.248, over 4270.00 frames. , ppl: 9.470241093055893] tot_loss[loss=2.352, over 5533138.02 frames. , ppl: 10.505282014901265], batch size: 70
+2022-12-10 09:19:35,458 INFO [train.py:421] (5/8) Epoch 2, batch 18800, loss[loss=2.301, over 3640.00 frames. , ppl: 9.985987643970622] tot_loss[loss=2.351, over 5527898.21 frames. , ppl: 10.500291941025573], batch size: 70
+2022-12-10 09:21:19,016 INFO [train.py:421] (5/8) Epoch 2, batch 19000, loss[loss=2.522, over 1680.00 frames. , ppl: 12.450369658243172] tot_loss[loss=2.352, over 5504974.52 frames. , ppl: 10.507895691652822], batch size: 70
+2022-12-10 09:21:19,017 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 09:21:19,781 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.337, over 211138.00 frames. , ppl: 10.346198405029323
+2022-12-10 09:23:00,682 INFO [train.py:421] (5/8) Epoch 2, batch 19200, loss[loss=2.671, over 770.00 frames. , ppl: 14.451848665740542] tot_loss[loss=2.352, over 5482076.22 frames. , ppl: 10.510660878401422], batch size: 70
+2022-12-10 09:24:40,865 INFO [train.py:421] (5/8) Epoch 2, batch 19400, loss[loss=3.327, over 490.00 frames. , ppl: 27.856261114191902] tot_loss[loss=2.353, over 5464787.03 frames. , ppl: 10.513082920087806], batch size: 70
+2022-12-10 09:26:20,689 INFO [train.py:421] (5/8) Epoch 2, batch 19600, loss[loss=2.304, over 2800.00 frames. , ppl: 10.010573461830683] tot_loss[loss=2.353, over 5456523.51 frames. , ppl: 10.515352762145191], batch size: 70
+2022-12-10 09:28:00,785 INFO [train.py:421] (5/8) Epoch 2, batch 19800, loss[loss=2.285, over 7420.00 frames. , ppl: 9.824792935779136] tot_loss[loss=2.353, over 5432013.41 frames. , ppl: 10.52222586560829], batch size: 70
+2022-12-10 09:29:37,936 INFO [train.py:421] (5/8) Epoch 2, batch 20000, loss[loss=2.665, over 840.00 frames. , ppl: 14.368790607465732] tot_loss[loss=2.353, over 5452555.76 frames. , ppl: 10.517255075139662], batch size: 70
+2022-12-10 09:29:37,936 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 09:29:38,694 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.339, over 211138.00 frames. , ppl: 10.367246266120773
+2022-12-10 09:31:19,579 INFO [train.py:421] (5/8) Epoch 2, batch 20200, loss[loss=2.301, over 8750.00 frames. , ppl: 9.988555177829952] tot_loss[loss=2.352, over 5461159.75 frames. , ppl: 10.509408320719418], batch size: 70
+2022-12-10 09:32:58,804 INFO [train.py:421] (5/8) Epoch 2, batch 20400, loss[loss=2.396, over 1680.00 frames. , ppl: 10.975848983840354] tot_loss[loss=2.352, over 5476690.75 frames. , ppl: 10.511500301501092], batch size: 70
+2022-12-10 09:34:39,600 INFO [train.py:421] (5/8) Epoch 2, batch 20600, loss[loss=2.503, over 1750.00 frames. , ppl: 12.21664865337623] tot_loss[loss=2.352, over 5485089.65 frames. , ppl: 10.510170689017368], batch size: 70
+2022-12-10 09:36:18,799 INFO [train.py:421] (5/8) Epoch 2, batch 20800, loss[loss=2.259, over 4900.00 frames. , ppl: 9.568874821102085] tot_loss[loss=2.352, over 5485476.21 frames. , ppl: 10.507611528156712], batch size: 70
+2022-12-10 09:37:55,988 INFO [train.py:421] (5/8) Epoch 2, batch 21000, loss[loss=2.166, over 8890.00 frames. , ppl: 8.720197762336142] tot_loss[loss=2.351, over 5491912.42 frames. , ppl: 10.499867752235055], batch size: 70
+2022-12-10 09:37:55,989 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 09:37:56,718 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.338, over 211138.00 frames. , ppl: 10.362088796314158
+2022-12-10 09:39:34,400 INFO [train.py:421] (5/8) Epoch 2, batch 21200, loss[loss=2.453, over 1610.00 frames. , ppl: 11.62421328073618] tot_loss[loss=2.351, over 5482040.50 frames. , ppl: 10.499125818107345], batch size: 70
+2022-12-10 09:41:12,856 INFO [train.py:421] (5/8) Epoch 2, batch 21400, loss[loss=2.392, over 2030.00 frames. , ppl: 10.940429706121266] tot_loss[loss=2.352, over 5464240.39 frames. , ppl: 10.505924944051005], batch size: 70
+2022-12-10 09:42:51,082 INFO [train.py:421] (5/8) Epoch 2, batch 21600, loss[loss=2.673, over 1050.00 frames. , ppl: 14.487347316708666] tot_loss[loss=2.352, over 5470449.03 frames. , ppl: 10.504195965942394], batch size: 70
+2022-12-10 09:44:31,993 INFO [train.py:421] (5/8) Epoch 2, batch 21800, loss[loss=2.251, over 4270.00 frames. , ppl: 9.493646246668193] tot_loss[loss=2.352, over 5468640.09 frames. , ppl: 10.50612464101121], batch size: 70
+2022-12-10 09:46:13,455 INFO [train.py:421] (5/8) Epoch 2, batch 22000, loss[loss=2.444, over 2380.00 frames. , ppl: 11.520833877124536] tot_loss[loss=2.351, over 5500103.56 frames. , ppl: 10.499569591012532], batch size: 70
+2022-12-10 09:46:13,456 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 09:46:14,216 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.337, over 211138.00 frames. , ppl: 10.353811838414043
+2022-12-10 09:47:52,620 INFO [train.py:421] (5/8) Epoch 2, batch 22200, loss[loss=2.36, over 2240.00 frames. , ppl: 10.585900881820551] tot_loss[loss=2.351, over 5501189.02 frames. , ppl: 10.5006680226959], batch size: 70
+2022-12-10 09:49:28,392 INFO [train.py:421] (5/8) Epoch 2, batch 22400, loss[loss=2.44, over 2100.00 frames. , ppl: 11.473416887991734] tot_loss[loss=2.35, over 5525861.13 frames. , ppl: 10.490441115867393], batch size: 70
+2022-12-10 09:51:04,810 INFO [train.py:421] (5/8) Epoch 2, batch 22600, loss[loss=2.283, over 4200.00 frames. , ppl: 9.80171559604018] tot_loss[loss=2.351, over 5489991.97 frames. , ppl: 10.499307183582259], batch size: 70
+2022-12-10 09:52:46,947 INFO [train.py:421] (5/8) Epoch 2, batch 22800, loss[loss=2.256, over 2800.00 frames. , ppl: 9.547339753107263] tot_loss[loss=2.352, over 5459172.10 frames. , ppl: 10.50558348506302], batch size: 70
+2022-12-10 09:54:25,175 INFO [train.py:421] (5/8) Epoch 2, batch 23000, loss[loss=2.272, over 5460.00 frames. , ppl: 9.69543353573863] tot_loss[loss=2.352, over 5463366.39 frames. , ppl: 10.505064280697136], batch size: 70
+2022-12-10 09:54:25,176 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 09:54:25,907 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.337, over 211138.00 frames. , ppl: 10.35141844156331
+2022-12-10 09:56:05,775 INFO [train.py:421] (5/8) Epoch 2, batch 23200, loss[loss=2.272, over 5600.00 frames. , ppl: 9.694918781231067] tot_loss[loss=2.351, over 5494973.19 frames. , ppl: 10.497174028952355], batch size: 70
+2022-12-10 09:57:48,629 INFO [train.py:421] (5/8) Epoch 2, batch 23400, loss[loss=2.547, over 910.00 frames. , ppl: 12.771522274421251] tot_loss[loss=2.35, over 5524700.80 frames. , ppl: 10.488755983143012], batch size: 70
+2022-12-10 09:59:32,019 INFO [train.py:421] (5/8) Epoch 2, batch 23600, loss[loss=2.36, over 1820.00 frames. , ppl: 10.588418355814326] tot_loss[loss=2.35, over 5530513.20 frames. , ppl: 10.486327440006507], batch size: 70
+2022-12-10 10:01:14,603 INFO [train.py:421] (5/8) Epoch 2, batch 23800, loss[loss=2.536, over 1890.00 frames. , ppl: 12.62324429254174] tot_loss[loss=2.349, over 5598188.01 frames. , ppl: 10.480198754277303], batch size: 70
+2022-12-10 10:02:54,344 INFO [train.py:421] (5/8) Epoch 2, batch 24000, loss[loss=2.414, over 1610.00 frames. , ppl: 11.177913508094145] tot_loss[loss=2.35, over 5570335.48 frames. , ppl: 10.480812101145702], batch size: 70
+2022-12-10 10:02:54,345 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 10:02:55,105 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.335, over 211138.00 frames. , ppl: 10.32757755196688
+2022-12-10 10:04:39,373 INFO [train.py:421] (5/8) Epoch 2, batch 24200, loss[loss=2.453, over 2520.00 frames. , ppl: 11.625180062493303] tot_loss[loss=2.351, over 5516322.50 frames. , ppl: 10.494491101844938], batch size: 70
+2022-12-10 10:06:16,094 INFO [train.py:421] (5/8) Epoch 2, batch 24400, loss[loss=2.331, over 3010.00 frames. , ppl: 10.290503188050598] tot_loss[loss=2.351, over 5478345.25 frames. , ppl: 10.497790820954243], batch size: 70
+2022-12-10 10:07:52,407 INFO [train.py:421] (5/8) Epoch 2, batch 24600, loss[loss=2.526, over 1470.00 frames. , ppl: 12.501797438487634] tot_loss[loss=2.352, over 5456915.04 frames. , ppl: 10.50822921359035], batch size: 70
+2022-12-10 10:09:36,242 INFO [train.py:421] (5/8) Epoch 2, batch 24800, loss[loss=2.421, over 2380.00 frames. , ppl: 11.252898946722917] tot_loss[loss=2.351, over 5465397.29 frames. , ppl: 10.497678638203762], batch size: 70
+2022-12-10 10:11:14,032 INFO [train.py:421] (5/8) Epoch 2, batch 25000, loss[loss=2.605, over 1120.00 frames. , ppl: 13.537557747606677] tot_loss[loss=2.351, over 5477245.54 frames. , ppl: 10.492190047580731], batch size: 70
+2022-12-10 10:11:14,033 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 10:11:14,793 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.336, over 211138.00 frames. , ppl: 10.337237971792588
+2022-12-10 10:12:54,450 INFO [train.py:421] (5/8) Epoch 2, batch 25200, loss[loss=2.318, over 4620.00 frames. , ppl: 10.150278907808824] tot_loss[loss=2.351, over 5470393.06 frames. , ppl: 10.493832649267706], batch size: 70
+2022-12-10 10:14:32,510 INFO [train.py:421] (5/8) Epoch 2, batch 25400, loss[loss=2.564, over 1610.00 frames. , ppl: 12.985285645339479] tot_loss[loss=2.35, over 5492499.97 frames. , ppl: 10.4869932809433], batch size: 70
+2022-12-10 10:16:11,534 INFO [train.py:421] (5/8) Epoch 2, batch 25600, loss[loss=2.268, over 2170.00 frames. , ppl: 9.660568093605855] tot_loss[loss=2.35, over 5514327.56 frames. , ppl: 10.486217891666056], batch size: 70
+2022-12-10 10:17:49,159 INFO [train.py:421] (5/8) Epoch 2, batch 25800, loss[loss=2.78, over 630.00 frames. , ppl: 16.11317901173813] tot_loss[loss=2.35, over 5535432.22 frames. , ppl: 10.487608755855602], batch size: 70
+2022-12-10 10:19:31,238 INFO [train.py:421] (5/8) Epoch 2, batch 26000, loss[loss=2.318, over 3220.00 frames. , ppl: 10.159024337878273] tot_loss[loss=2.35, over 5539516.70 frames. , ppl: 10.486626143728026], batch size: 70
+2022-12-10 10:19:31,238 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 10:19:31,994 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.335, over 211138.00 frames. , ppl: 10.33015502328756
+2022-12-10 10:21:14,006 INFO [train.py:421] (5/8) Epoch 2, batch 26200, loss[loss=2.289, over 5320.00 frames. , ppl: 9.866361598155198] tot_loss[loss=2.351, over 5526728.61 frames. , ppl: 10.492670510236152], batch size: 70
+2022-12-10 10:22:55,797 INFO [train.py:421] (5/8) Epoch 2, batch 26400, loss[loss=2.417, over 2520.00 frames. , ppl: 11.216269676166421] tot_loss[loss=2.35, over 5540968.23 frames. , ppl: 10.486494769006924], batch size: 70
+2022-12-10 10:24:38,325 INFO [train.py:421] (5/8) Epoch 2, batch 26600, loss[loss=2.286, over 4760.00 frames. , ppl: 9.834246383618954] tot_loss[loss=2.35, over 5522571.88 frames. , ppl: 10.490311176955801], batch size: 70
+2022-12-10 10:26:19,546 INFO [train.py:421] (5/8) Epoch 2, batch 26800, loss[loss=2.407, over 2450.00 frames. , ppl: 11.103306035122504] tot_loss[loss=2.35, over 5536041.90 frames. , ppl: 10.482587145320098], batch size: 70
+2022-12-10 10:27:59,904 INFO [train.py:421] (5/8) Epoch 2, batch 27000, loss[loss=2.213, over 3290.00 frames. , ppl: 9.14142353632687] tot_loss[loss=2.35, over 5507559.97 frames. , ppl: 10.485702266331279], batch size: 70
+2022-12-10 10:27:59,905 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 10:28:00,666 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.335, over 211138.00 frames. , ppl: 10.332974773515254
+2022-12-10 10:29:43,667 INFO [train.py:421] (5/8) Epoch 2, batch 27200, loss[loss=2.37, over 2310.00 frames. , ppl: 10.699177412044856] tot_loss[loss=2.35, over 5529786.01 frames. , ppl: 10.48277910030614], batch size: 70
+2022-12-10 10:31:24,575 INFO [train.py:421] (5/8) Epoch 2, batch 27400, loss[loss=2.289, over 3080.00 frames. , ppl: 9.865571655416437] tot_loss[loss=2.349, over 5540198.58 frames. , ppl: 10.47546030278772], batch size: 70
+2022-12-10 10:33:03,927 INFO [train.py:421] (5/8) Epoch 2, batch 27600, loss[loss=2.492, over 1120.00 frames. , ppl: 12.09078808851196] tot_loss[loss=2.349, over 5557858.14 frames. , ppl: 10.470720772997451], batch size: 70
+2022-12-10 10:34:44,448 INFO [train.py:421] (5/8) Epoch 2, batch 27800, loss[loss=2.217, over 12180.00 frames. , ppl: 9.182355049344098] tot_loss[loss=2.349, over 5533596.64 frames. , ppl: 10.471138281483595], batch size: 70
+2022-12-10 10:36:25,384 INFO [train.py:421] (5/8) Epoch 2, batch 28000, loss[loss=2.609, over 1470.00 frames. , ppl: 13.58972800979527] tot_loss[loss=2.35, over 5497885.31 frames. , ppl: 10.482464592674782], batch size: 70
+2022-12-10 10:36:25,385 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 10:36:26,134 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.333, over 211138.00 frames. , ppl: 10.313298466854016
+2022-12-10 10:38:05,396 INFO [train.py:421] (5/8) Epoch 2, batch 28200, loss[loss=2.389, over 2100.00 frames. , ppl: 10.899308117437027] tot_loss[loss=2.349, over 5527172.94 frames. , ppl: 10.475079719154033], batch size: 70
+2022-12-10 10:39:46,483 INFO [train.py:421] (5/8) Epoch 2, batch 28400, loss[loss=2.281, over 3710.00 frames. , ppl: 9.786363472806] tot_loss[loss=2.349, over 5534985.80 frames. , ppl: 10.471649804423171], batch size: 70
+2022-12-10 10:41:25,661 INFO [train.py:421] (5/8) Epoch 2, batch 28600, loss[loss=2.388, over 1750.00 frames. , ppl: 10.890047844364759] tot_loss[loss=2.35, over 5477663.71 frames. , ppl: 10.481896253549122], batch size: 70
+2022-12-10 10:43:03,861 INFO [train.py:421] (5/8) Epoch 2, batch 28800, loss[loss=2.542, over 1540.00 frames. , ppl: 12.699654272350037] tot_loss[loss=2.35, over 5456790.57 frames. , ppl: 10.487461454485992], batch size: 70
+2022-12-10 10:44:44,078 INFO [train.py:421] (5/8) Epoch 2, batch 29000, loss[loss=2.251, over 5180.00 frames. , ppl: 9.50130731212147] tot_loss[loss=2.35, over 5453881.05 frames. , ppl: 10.484654099168548], batch size: 70
+2022-12-10 10:44:44,079 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 10:44:44,826 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.335, over 211138.00 frames. , ppl: 10.328367846981726
+2022-12-10 10:46:24,236 INFO [train.py:421] (5/8) Epoch 2, batch 29200, loss[loss=2.207, over 10360.00 frames. , ppl: 9.088173366569443] tot_loss[loss=2.351, over 5446130.66 frames. , ppl: 10.49380421815072], batch size: 70
+2022-12-10 10:48:02,851 INFO [train.py:421] (5/8) Epoch 2, batch 29400, loss[loss=2.34, over 2240.00 frames. , ppl: 10.383099573842616] tot_loss[loss=2.349, over 5488904.27 frames. , ppl: 10.478762507573814], batch size: 70
+2022-12-10 10:49:40,055 INFO [train.py:421] (5/8) Epoch 2, batch 29600, loss[loss=2.36, over 1400.00 frames. , ppl: 10.589817507110055] tot_loss[loss=2.35, over 5460949.96 frames. , ppl: 10.486824477945298], batch size: 70
+2022-12-10 10:51:22,076 INFO [train.py:421] (5/8) Epoch 2, batch 29800, loss[loss=3.419, over 490.00 frames. , ppl: 30.549996976691812] tot_loss[loss=2.351, over 5431977.53 frames. , ppl: 10.49196877759986], batch size: 70
+2022-12-10 10:53:03,452 INFO [train.py:421] (5/8) Epoch 2, batch 30000, loss[loss=2.399, over 1820.00 frames. , ppl: 11.016672090040768] tot_loss[loss=2.351, over 5434414.42 frames. , ppl: 10.491751214905317], batch size: 70
+2022-12-10 10:53:03,453 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 10:53:04,216 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.333, over 211138.00 frames. , ppl: 10.311935441408778
+2022-12-10 10:54:45,306 INFO [train.py:421] (5/8) Epoch 2, batch 30200, loss[loss=2.461, over 3360.00 frames. , ppl: 11.71163643690884] tot_loss[loss=2.35, over 5458685.88 frames. , ppl: 10.487245692218705], batch size: 70
+2022-12-10 10:56:29,740 INFO [train.py:421] (5/8) Epoch 2, batch 30400, loss[loss=2.389, over 2800.00 frames. , ppl: 10.906391750099734] tot_loss[loss=2.35, over 5475539.41 frames. , ppl: 10.490248125423367], batch size: 70
+2022-12-10 10:58:09,969 INFO [train.py:421] (5/8) Epoch 2, batch 30600, loss[loss=2.419, over 1400.00 frames. , ppl: 11.2391229172955] tot_loss[loss=2.35, over 5506226.05 frames. , ppl: 10.480918073600513], batch size: 70
+2022-12-10 10:59:47,280 INFO [train.py:421] (5/8) Epoch 2, batch 30800, loss[loss=2.551, over 1680.00 frames. , ppl: 12.816122373853679] tot_loss[loss=2.349, over 5507859.70 frames. , ppl: 10.477909465656005], batch size: 70
+2022-12-10 11:01:28,588 INFO [train.py:421] (5/8) Epoch 2, batch 31000, loss[loss=2.526, over 1470.00 frames. , ppl: 12.507336221835383] tot_loss[loss=2.349, over 5536468.90 frames. , ppl: 10.470399612294429], batch size: 70
+2022-12-10 11:01:28,588 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 11:01:29,350 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.334, over 211138.00 frames. , ppl: 10.318448451765484
+2022-12-10 11:03:12,432 INFO [train.py:421] (5/8) Epoch 2, batch 31200, loss[loss=2.284, over 4550.00 frames. , ppl: 9.815615167846145] tot_loss[loss=2.348, over 5560547.91 frames. , ppl: 10.463524563146022], batch size: 70
+2022-12-10 11:04:53,783 INFO [train.py:421] (5/8) Epoch 2, batch 31400, loss[loss=2.428, over 1960.00 frames. , ppl: 11.340315935254832] tot_loss[loss=2.349, over 5557538.46 frames. , ppl: 10.47067095035779], batch size: 70
+2022-12-10 11:06:34,444 INFO [train.py:421] (5/8) Epoch 2, batch 31600, loss[loss=2.673, over 840.00 frames. , ppl: 14.485704412369275] tot_loss[loss=2.349, over 5517897.06 frames. , ppl: 10.47716765766866], batch size: 70
+2022-12-10 11:08:17,173 INFO [train.py:421] (5/8) Epoch 2, batch 31800, loss[loss=2.362, over 1540.00 frames. , ppl: 10.616473953124585] tot_loss[loss=2.349, over 5505842.31 frames. , ppl: 10.476860960027155], batch size: 70
+2022-12-10 11:09:56,529 INFO [train.py:421] (5/8) Epoch 2, batch 32000, loss[loss=2.382, over 4270.00 frames. , ppl: 10.826719108770133] tot_loss[loss=2.349, over 5495659.89 frames. , ppl: 10.479283275214158], batch size: 70
+2022-12-10 11:09:56,530 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 11:09:57,292 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.334, over 211138.00 frames. , ppl: 10.31484792506888
+2022-12-10 11:11:36,906 INFO [train.py:421] (5/8) Epoch 2, batch 32200, loss[loss=2.431, over 1960.00 frames. , ppl: 11.36500619854034] tot_loss[loss=2.35, over 5458014.90 frames. , ppl: 10.484114959996692], batch size: 70
+2022-12-10 11:13:14,855 INFO [train.py:421] (5/8) Epoch 2, batch 32400, loss[loss=2.559, over 1050.00 frames. , ppl: 12.917314728386039] tot_loss[loss=2.35, over 5464724.45 frames. , ppl: 10.488044083361755], batch size: 70
+2022-12-10 11:14:55,253 INFO [train.py:421] (5/8) Epoch 2, batch 32600, loss[loss=2.504, over 1400.00 frames. , ppl: 12.22849438408136] tot_loss[loss=2.35, over 5462567.36 frames. , ppl: 10.482146528925604], batch size: 70
+2022-12-10 11:16:38,284 INFO [train.py:421] (5/8) Epoch 2, batch 32800, loss[loss=2.544, over 1120.00 frames. , ppl: 12.724812347496108] tot_loss[loss=2.349, over 5495301.96 frames. , ppl: 10.47685446870881], batch size: 70
+2022-12-10 11:18:17,211 INFO [train.py:421] (5/8) Epoch 2, batch 33000, loss[loss=2.413, over 2940.00 frames. , ppl: 11.169842073051463] tot_loss[loss=2.349, over 5500674.42 frames. , ppl: 10.477066898168037], batch size: 70
+2022-12-10 11:18:17,211 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 11:18:17,973 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.332, over 211138.00 frames. , ppl: 10.303007731813349
+2022-12-10 11:20:00,440 INFO [train.py:421] (5/8) Epoch 2, batch 33200, loss[loss=2.371, over 2870.00 frames. , ppl: 10.704508579010021] tot_loss[loss=2.348, over 5529100.02 frames. , ppl: 10.460049243674566], batch size: 70
+2022-12-10 11:21:39,768 INFO [train.py:421] (5/8) Epoch 2, batch 33400, loss[loss=2.281, over 1960.00 frames. , ppl: 9.786355974780959] tot_loss[loss=2.346, over 5561641.51 frames. , ppl: 10.44875568099338], batch size: 70
+2022-12-10 11:23:22,970 INFO [train.py:421] (5/8) Epoch 2, batch 33600, loss[loss=2.473, over 1890.00 frames. , ppl: 11.854363597556953] tot_loss[loss=2.347, over 5548050.60 frames. , ppl: 10.451027957256128], batch size: 70
+2022-12-10 11:25:04,055 INFO [train.py:421] (5/8) Epoch 2, batch 33800, loss[loss=2.282, over 2030.00 frames. , ppl: 9.791927902442131] tot_loss[loss=2.345, over 5595664.78 frames. , ppl: 10.429805487082218], batch size: 70
+2022-12-10 11:26:42,453 INFO [train.py:421] (5/8) Epoch 2, batch 34000, loss[loss=2.448, over 3290.00 frames. , ppl: 11.562139779938832] tot_loss[loss=2.344, over 5604057.49 frames. , ppl: 10.425927628976783], batch size: 70
+2022-12-10 11:26:42,453 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 11:26:43,186 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.331, over 211138.00 frames. , ppl: 10.291255765342557
+2022-12-10 11:28:20,403 INFO [train.py:421] (5/8) Epoch 2, batch 34200, loss[loss=2.506, over 1400.00 frames. , ppl: 12.254371650364924] tot_loss[loss=2.345, over 5583727.08 frames. , ppl: 10.434325114820668], batch size: 70
+2022-12-10 11:30:00,226 INFO [train.py:421] (5/8) Epoch 2, batch 34400, loss[loss=2.409, over 1400.00 frames. , ppl: 11.12796319021985] tot_loss[loss=2.345, over 5588283.66 frames. , ppl: 10.433631517209786], batch size: 70
+2022-12-10 11:31:39,796 INFO [train.py:421] (5/8) Epoch 2, batch 34600, loss[loss=2.412, over 2520.00 frames. , ppl: 11.15558013102493] tot_loss[loss=2.345, over 5565873.60 frames. , ppl: 10.429886798011529], batch size: 70
+2022-12-10 11:33:17,810 INFO [train.py:421] (5/8) Epoch 2, batch 34800, loss[loss=2.536, over 1260.00 frames. , ppl: 12.635021948889523] tot_loss[loss=2.345, over 5530168.07 frames. , ppl: 10.436227390538997], batch size: 70
+2022-12-10 11:35:00,057 INFO [train.py:421] (5/8) Epoch 2, batch 35000, loss[loss=2.496, over 980.00 frames. , ppl: 12.13544029966972] tot_loss[loss=2.345, over 5542986.77 frames. , ppl: 10.435858533924572], batch size: 70
+2022-12-10 11:35:00,058 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 11:35:00,819 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.332, over 211138.00 frames. , ppl: 10.297171983552662
+2022-12-10 11:36:40,680 INFO [train.py:421] (5/8) Epoch 2, batch 35200, loss[loss=2.514, over 1050.00 frames. , ppl: 12.3565948704435] tot_loss[loss=2.347, over 5485669.27 frames. , ppl: 10.457428263447065], batch size: 70
+2022-12-10 11:38:22,509 INFO [train.py:421] (5/8) Epoch 2, batch 35400, loss[loss=2.264, over 2800.00 frames. , ppl: 9.620733889572382] tot_loss[loss=2.347, over 5472735.21 frames. , ppl: 10.456431200111101], batch size: 70
+2022-12-10 11:40:03,278 INFO [train.py:421] (5/8) Epoch 2, batch 35600, loss[loss=2.362, over 2310.00 frames. , ppl: 10.609405771510337] tot_loss[loss=2.348, over 5471359.38 frames. , ppl: 10.461252129167246], batch size: 70
+2022-12-10 11:41:42,466 INFO [train.py:421] (5/8) Epoch 2, batch 35800, loss[loss=2.418, over 2380.00 frames. , ppl: 11.220479236352164] tot_loss[loss=2.348, over 5477487.16 frames. , ppl: 10.461128761190954], batch size: 70
+2022-12-10 11:43:21,035 INFO [train.py:421] (5/8) Epoch 2, batch 36000, loss[loss=2.474, over 2520.00 frames. , ppl: 11.87292275727392] tot_loss[loss=2.348, over 5479698.71 frames. , ppl: 10.465229906938376], batch size: 70
+2022-12-10 11:43:21,035 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 11:43:21,794 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.331, over 211138.00 frames. , ppl: 10.288832667020564
+2022-12-10 11:45:02,829 INFO [train.py:421] (5/8) Epoch 2, batch 36200, loss[loss=3.226, over 490.00 frames. , ppl: 25.16663659065276] tot_loss[loss=2.348, over 5473621.99 frames. , ppl: 10.466816769103948], batch size: 70
+2022-12-10 11:46:46,962 INFO [train.py:421] (5/8) Epoch 2, batch 36400, loss[loss=2.433, over 1540.00 frames. , ppl: 11.394587290715743] tot_loss[loss=2.348, over 5472583.86 frames. , ppl: 10.461774022941409], batch size: 70
+2022-12-10 11:48:25,554 INFO [train.py:421] (5/8) Epoch 2, batch 36600, loss[loss=2.26, over 2730.00 frames. , ppl: 9.586033599732556] tot_loss[loss=2.348, over 5463625.36 frames. , ppl: 10.467032917143932], batch size: 70
+2022-12-10 11:50:04,140 INFO [train.py:421] (5/8) Epoch 2, batch 36800, loss[loss=2.346, over 2100.00 frames. , ppl: 10.440482540192408] tot_loss[loss=2.349, over 5448211.98 frames. , ppl: 10.478974241294377], batch size: 70
+2022-12-10 11:51:43,180 INFO [train.py:421] (5/8) Epoch 2, batch 37000, loss[loss=2.424, over 1120.00 frames. , ppl: 11.296357507002378] tot_loss[loss=2.349, over 5470356.39 frames. , ppl: 10.471458563111211], batch size: 70
+2022-12-10 11:51:43,181 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 11:51:43,927 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.333, over 211138.00 frames. , ppl: 10.313814418221853
+2022-12-10 11:53:27,767 INFO [train.py:421] (5/8) Epoch 2, batch 37200, loss[loss=2.244, over 6090.00 frames. , ppl: 9.434632724378268] tot_loss[loss=2.347, over 5519222.24 frames. , ppl: 10.456893850396112], batch size: 70
+2022-12-10 11:55:05,711 INFO [train.py:421] (5/8) Epoch 2, batch 37400, loss[loss=2.54, over 1820.00 frames. , ppl: 12.676683196457713] tot_loss[loss=2.347, over 5496891.83 frames. , ppl: 10.453661251388507], batch size: 70
+2022-12-10 11:56:46,031 INFO [train.py:421] (5/8) Epoch 2, batch 37600, loss[loss=2.339, over 2310.00 frames. , ppl: 10.368155037267934] tot_loss[loss=2.348, over 5467335.47 frames. , ppl: 10.465066522831346], batch size: 70
+2022-12-10 11:58:28,997 INFO [train.py:421] (5/8) Epoch 2, batch 37800, loss[loss=2.303, over 4200.00 frames. , ppl: 10.002748778136532] tot_loss[loss=2.348, over 5457598.30 frames. , ppl: 10.461669491315805], batch size: 70
+2022-12-10 12:00:10,593 INFO [train.py:421] (5/8) Epoch 2, batch 38000, loss[loss=2.27, over 6930.00 frames. , ppl: 9.679386628453436] tot_loss[loss=2.348, over 5462083.96 frames. , ppl: 10.460008863240017], batch size: 70
+2022-12-10 12:00:10,594 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 12:00:11,327 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.329, over 211138.00 frames. , ppl: 10.269646244641864
+2022-12-10 12:01:54,474 INFO [train.py:421] (5/8) Epoch 2, batch 38200, loss[loss=2.823, over 700.00 frames. , ppl: 16.83373670106517] tot_loss[loss=2.348, over 5438549.58 frames. , ppl: 10.460668689168438], batch size: 70
+2022-12-10 12:03:35,678 INFO [train.py:421] (5/8) Epoch 2, batch 38400, loss[loss=2.62, over 1120.00 frames. , ppl: 13.730429750396333] tot_loss[loss=2.348, over 5413401.47 frames. , ppl: 10.462620962508014], batch size: 70
+2022-12-10 12:05:13,549 INFO [train.py:421] (5/8) Epoch 2, batch 38600, loss[loss=2.312, over 2940.00 frames. , ppl: 10.093920662487621] tot_loss[loss=2.347, over 5402024.00 frames. , ppl: 10.458396951486154], batch size: 70
+2022-12-10 12:06:50,147 INFO [train.py:421] (5/8) Epoch 2, batch 38800, loss[loss=2.32, over 3290.00 frames. , ppl: 10.173154983595968] tot_loss[loss=2.347, over 5376322.71 frames. , ppl: 10.458014926829135], batch size: 70
+2022-12-10 12:08:30,334 INFO [train.py:421] (5/8) Epoch 2, batch 39000, loss[loss=2.319, over 2520.00 frames. , ppl: 10.167193381675322] tot_loss[loss=2.347, over 5419239.45 frames. , ppl: 10.451954989566499], batch size: 70
+2022-12-10 12:08:30,334 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 12:08:31,096 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.328, over 211138.00 frames. , ppl: 10.258752219498415
+2022-12-10 12:10:14,114 INFO [train.py:421] (5/8) Epoch 2, batch 39200, loss[loss=2.285, over 9100.00 frames. , ppl: 9.824281811383301] tot_loss[loss=2.346, over 5465937.13 frames. , ppl: 10.444432573687632], batch size: 70
+2022-12-10 12:11:55,247 INFO [train.py:421] (5/8) Epoch 2, batch 39400, loss[loss=2.614, over 770.00 frames. , ppl: 13.656307916744574] tot_loss[loss=2.346, over 5467517.54 frames. , ppl: 10.441967267926413], batch size: 70
+2022-12-10 12:13:32,407 INFO [train.py:421] (5/8) Epoch 2, batch 39600, loss[loss=2.415, over 2310.00 frames. , ppl: 11.189384353463774] tot_loss[loss=2.347, over 5444051.22 frames. , ppl: 10.450659128299373], batch size: 70
+2022-12-10 12:15:17,688 INFO [train.py:421] (5/8) Epoch 2, batch 39800, loss[loss=2.442, over 2100.00 frames. , ppl: 11.498470272014176] tot_loss[loss=2.347, over 5430436.64 frames. , ppl: 10.459073931293153], batch size: 70
+2022-12-10 12:16:56,217 INFO [train.py:421] (5/8) Epoch 2, batch 40000, loss[loss=2.385, over 1820.00 frames. , ppl: 10.854167611292212] tot_loss[loss=2.348, over 5418765.73 frames. , ppl: 10.463978501162199], batch size: 70
+2022-12-10 12:16:56,218 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 12:16:56,977 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.331, over 211138.00 frames. , ppl: 10.292756209643162
+2022-12-10 12:18:37,689 INFO [train.py:421] (5/8) Epoch 2, batch 40200, loss[loss=2.328, over 1960.00 frames. , ppl: 10.254656383727244] tot_loss[loss=2.347, over 5448647.36 frames. , ppl: 10.449416310346887], batch size: 70
+2022-12-10 12:20:17,609 INFO [train.py:421] (5/8) Epoch 2, batch 40400, loss[loss=2.279, over 1260.00 frames. , ppl: 9.762424818752729] tot_loss[loss=2.347, over 5416180.66 frames. , ppl: 10.455380120959287], batch size: 70
+2022-12-10 12:22:00,252 INFO [train.py:421] (5/8) Epoch 2, batch 40600, loss[loss=2.312, over 3570.00 frames. , ppl: 10.09063147252812] tot_loss[loss=2.345, over 5443593.98 frames. , ppl: 10.435775973627582], batch size: 70
+2022-12-10 12:23:41,677 INFO [train.py:421] (5/8) Epoch 2, batch 40800, loss[loss=2.373, over 2030.00 frames. , ppl: 10.724410667152776] tot_loss[loss=2.345, over 5468444.69 frames. , ppl: 10.435554186520624], batch size: 70
+2022-12-10 12:25:27,354 INFO [train.py:421] (5/8) Epoch 2, batch 41000, loss[loss=2.291, over 5740.00 frames. , ppl: 9.88211007110435] tot_loss[loss=2.345, over 5459804.42 frames. , ppl: 10.434374058742334], batch size: 70
+2022-12-10 12:25:27,354 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 12:25:28,086 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.329, over 211138.00 frames. , ppl: 10.27130012072034
+2022-12-10 12:27:08,688 INFO [train.py:421] (5/8) Epoch 2, batch 41200, loss[loss=2.326, over 1890.00 frames. , ppl: 10.239163512839147] tot_loss[loss=2.346, over 5416979.75 frames. , ppl: 10.439948625939635], batch size: 70
+2022-12-10 12:28:47,118 INFO [train.py:421] (5/8) Epoch 2, batch 41400, loss[loss=2.392, over 1960.00 frames. , ppl: 10.931509449169155] tot_loss[loss=2.345, over 5432466.38 frames. , ppl: 10.436269148712631], batch size: 70
+2022-12-10 12:30:25,486 INFO [train.py:421] (5/8) Epoch 2, batch 41600, loss[loss=2.374, over 910.00 frames. , ppl: 10.743626018821939] tot_loss[loss=2.346, over 5429547.38 frames. , ppl: 10.440438416607023], batch size: 70
+2022-12-10 12:32:02,259 INFO [train.py:421] (5/8) Epoch 2, batch 41800, loss[loss=2.707, over 770.00 frames. , ppl: 14.986362751617612] tot_loss[loss=2.347, over 5403142.21 frames. , ppl: 10.45267042575336], batch size: 70
+2022-12-10 12:33:43,237 INFO [train.py:421] (5/8) Epoch 2, batch 42000, loss[loss=2.318, over 1540.00 frames. , ppl: 10.1522975226029] tot_loss[loss=2.346, over 5414469.81 frames. , ppl: 10.445670652822068], batch size: 70
+2022-12-10 12:33:43,237 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 12:33:43,997 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.331, over 211138.00 frames. , ppl: 10.283807052673856
+2022-12-10 12:35:25,663 INFO [train.py:421] (5/8) Epoch 2, batch 42200, loss[loss=2.659, over 910.00 frames. , ppl: 14.279278811881875] tot_loss[loss=2.346, over 5419595.24 frames. , ppl: 10.440550196670028], batch size: 70
+2022-12-10 12:37:03,123 INFO [train.py:421] (5/8) Epoch 2, batch 42400, loss[loss=2.348, over 2240.00 frames. , ppl: 10.468039570371442] tot_loss[loss=2.345, over 5444622.88 frames. , ppl: 10.436177306004733], batch size: 70
+2022-12-10 12:38:43,240 INFO [train.py:421] (5/8) Epoch 2, batch 42600, loss[loss=2.294, over 3710.00 frames. , ppl: 9.915461935069098] tot_loss[loss=2.345, over 5448267.04 frames. , ppl: 10.434423875336748], batch size: 70
+2022-12-10 12:40:22,491 INFO [train.py:421] (5/8) Epoch 2, batch 42800, loss[loss=2.253, over 7280.00 frames. , ppl: 9.518087631736275] tot_loss[loss=2.345, over 5479831.22 frames. , ppl: 10.435602386823579], batch size: 70
+2022-12-10 12:42:04,985 INFO [train.py:421] (5/8) Epoch 2, batch 43000, loss[loss=2.507, over 1680.00 frames. , ppl: 12.273225868516425] tot_loss[loss=2.346, over 5470050.77 frames. , ppl: 10.44016646461131], batch size: 70
+2022-12-10 12:42:04,986 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 12:42:05,730 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.328, over 211138.00 frames. , ppl: 10.261654238592559
+2022-12-10 12:43:44,242 INFO [train.py:421] (5/8) Epoch 2, batch 43200, loss[loss=2.324, over 5530.00 frames. , ppl: 10.221545207895495] tot_loss[loss=2.347, over 5448326.33 frames. , ppl: 10.45240821221933], batch size: 70
+2022-12-10 12:45:21,838 INFO [train.py:421] (5/8) Epoch 2, batch 43400, loss[loss=2.271, over 4480.00 frames. , ppl: 9.687813852699446] tot_loss[loss=2.347, over 5423939.33 frames. , ppl: 10.455830753306937], batch size: 70
+2022-12-10 12:47:02,706 INFO [train.py:421] (5/8) Epoch 2, batch 43600, loss[loss=2.428, over 1820.00 frames. , ppl: 11.333893833142431] tot_loss[loss=2.346, over 5474208.64 frames. , ppl: 10.441726522064455], batch size: 70
+2022-12-10 12:48:41,303 INFO [train.py:421] (5/8) Epoch 2, batch 43800, loss[loss=2.312, over 6650.00 frames. , ppl: 10.0911621830087] tot_loss[loss=2.347, over 5439125.63 frames. , ppl: 10.452420214318852], batch size: 70
+2022-12-10 12:50:21,826 INFO [train.py:421] (5/8) Epoch 2, batch 44000, loss[loss=2.484, over 1680.00 frames. , ppl: 11.98993441524528] tot_loss[loss=2.346, over 5458781.22 frames. , ppl: 10.444239554055914], batch size: 70
+2022-12-10 12:50:21,827 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 12:50:22,572 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.329, over 211138.00 frames. , ppl: 10.270837981475065
+2022-12-10 12:52:01,323 INFO [train.py:421] (5/8) Epoch 2, batch 44200, loss[loss=2.565, over 1470.00 frames. , ppl: 12.998499804629338] tot_loss[loss=2.346, over 5451054.04 frames. , ppl: 10.443429993410382], batch size: 70
+2022-12-10 12:53:43,812 INFO [train.py:421] (5/8) Epoch 2, batch 44400, loss[loss=2.263, over 6090.00 frames. , ppl: 9.610028996977489] tot_loss[loss=2.345, over 5496141.12 frames. , ppl: 10.435335020372705], batch size: 70
+2022-12-10 12:55:21,483 INFO [train.py:421] (5/8) Epoch 2, batch 44600, loss[loss=2.792, over 700.00 frames. , ppl: 16.30559728019168] tot_loss[loss=2.346, over 5482261.59 frames. , ppl: 10.440746204667967], batch size: 70
+2022-12-10 12:57:01,754 INFO [train.py:421] (5/8) Epoch 2, batch 44800, loss[loss=2.436, over 3220.00 frames. , ppl: 11.425143434785664] tot_loss[loss=2.345, over 5485547.99 frames. , ppl: 10.435919934348757], batch size: 70
+2022-12-10 12:58:44,440 INFO [train.py:421] (5/8) Epoch 2, batch 45000, loss[loss=2.392, over 2240.00 frames. , ppl: 10.93736711572271] tot_loss[loss=2.345, over 5455232.38 frames. , ppl: 10.435860632458809], batch size: 70
+2022-12-10 12:58:44,441 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 12:58:45,186 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.329, over 211138.00 frames. , ppl: 10.267509369063996
+2022-12-10 13:00:27,490 INFO [train.py:421] (5/8) Epoch 2, batch 45200, loss[loss=2.378, over 4620.00 frames. , ppl: 10.780056501035325] tot_loss[loss=2.345, over 5461930.82 frames. , ppl: 10.436905618508383], batch size: 70
+2022-12-10 13:02:10,351 INFO [train.py:421] (5/8) Epoch 2, batch 45400, loss[loss=2.284, over 3640.00 frames. , ppl: 9.813026354284716] tot_loss[loss=2.346, over 5473924.60 frames. , ppl: 10.443795155945562], batch size: 70
+2022-12-10 13:03:49,609 INFO [train.py:421] (5/8) Epoch 2, batch 45600, loss[loss=2.322, over 3780.00 frames. , ppl: 10.192727955685163] tot_loss[loss=2.345, over 5489534.26 frames. , ppl: 10.43662796096621], batch size: 70
+2022-12-10 13:05:28,544 INFO [train.py:421] (5/8) Epoch 2, batch 45800, loss[loss=2.511, over 840.00 frames. , ppl: 12.321405161060392] tot_loss[loss=2.345, over 5472462.90 frames. , ppl: 10.438008921341712], batch size: 70
+2022-12-10 13:07:05,405 INFO [train.py:421] (5/8) Epoch 2, batch 46000, loss[loss=2.207, over 4130.00 frames. , ppl: 9.09249029906656] tot_loss[loss=2.345, over 5478697.88 frames. , ppl: 10.43615223307493], batch size: 70
+2022-12-10 13:07:05,405 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 13:07:06,169 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.329, over 211138.00 frames. , ppl: 10.269302733935248
+2022-12-10 13:08:48,316 INFO [train.py:421] (5/8) Epoch 2, batch 46200, loss[loss=2.255, over 7210.00 frames. , ppl: 9.53336501770096] tot_loss[loss=2.345, over 5482909.29 frames. , ppl: 10.430468099352293], batch size: 70
+2022-12-10 13:10:28,306 INFO [train.py:421] (5/8) Epoch 2, batch 46400, loss[loss=2.486, over 1820.00 frames. , ppl: 12.01239591720764] tot_loss[loss=2.344, over 5502533.51 frames. , ppl: 10.418797267272726], batch size: 70
+2022-12-10 13:12:09,489 INFO [train.py:421] (5/8) Epoch 2, batch 46600, loss[loss=2.265, over 5040.00 frames. , ppl: 9.628742941999699] tot_loss[loss=2.343, over 5506260.37 frames. , ppl: 10.415156946499142], batch size: 70
+2022-12-10 13:13:46,920 INFO [train.py:421] (5/8) Epoch 2, batch 46800, loss[loss=2.561, over 1260.00 frames. , ppl: 12.944478581161091] tot_loss[loss=2.344, over 5502673.06 frames. , ppl: 10.417865100919359], batch size: 70
+2022-12-10 13:15:22,164 INFO [train.py:421] (5/8) Epoch 2, batch 47000, loss[loss=2.326, over 5040.00 frames. , ppl: 10.233383870922584] tot_loss[loss=2.344, over 5477150.75 frames. , ppl: 10.421136542283909], batch size: 70
+2022-12-10 13:15:22,164 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 13:15:22,923 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.329, over 211138.00 frames. , ppl: 10.265565897614955
+2022-12-10 13:17:01,370 INFO [train.py:421] (5/8) Epoch 2, batch 47200, loss[loss=2.282, over 4690.00 frames. , ppl: 9.800350271219136] tot_loss[loss=2.343, over 5487291.20 frames. , ppl: 10.413400556516859], batch size: 70
+2022-12-10 13:18:45,163 INFO [train.py:421] (5/8) Epoch 2, batch 47400, loss[loss=2.324, over 2100.00 frames. , ppl: 10.219568016890339] tot_loss[loss=2.344, over 5436147.35 frames. , ppl: 10.420748069339625], batch size: 70
+2022-12-10 13:20:26,207 INFO [train.py:421] (5/8) Epoch 2, batch 47600, loss[loss=3.53, over 490.00 frames. , ppl: 34.13921003272027] tot_loss[loss=2.344, over 5432149.00 frames. , ppl: 10.425617878055434], batch size: 70
+2022-12-10 13:22:08,948 INFO [train.py:421] (5/8) Epoch 2, batch 47800, loss[loss=2.212, over 4970.00 frames. , ppl: 9.137629234714225] tot_loss[loss=2.344, over 5459689.17 frames. , ppl: 10.420864497243462], batch size: 70
+2022-12-10 13:23:44,809 INFO [train.py:421] (5/8) Epoch 2, batch 48000, loss[loss=2.394, over 1610.00 frames. , ppl: 10.962401277064247] tot_loss[loss=2.345, over 5429710.07 frames. , ppl: 10.43242481832961], batch size: 70
+2022-12-10 13:23:44,809 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 13:23:45,556 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.328, over 211138.00 frames. , ppl: 10.26050912621634
+2022-12-10 13:25:25,335 INFO [train.py:421] (5/8) Epoch 2, batch 48200, loss[loss=2.849, over 630.00 frames. , ppl: 17.269110530385152] tot_loss[loss=2.344, over 5458071.02 frames. , ppl: 10.424125991055224], batch size: 70
+2022-12-10 13:27:08,712 INFO [train.py:421] (5/8) Epoch 2, batch 48400, loss[loss=2.358, over 3570.00 frames. , ppl: 10.572475835123925] tot_loss[loss=2.343, over 5517462.02 frames. , ppl: 10.410414062605222], batch size: 70
+2022-12-10 13:28:49,067 INFO [train.py:421] (5/8) Epoch 2, batch 48600, loss[loss=2.305, over 1960.00 frames. , ppl: 10.029015601763376] tot_loss[loss=2.342, over 5542978.52 frames. , ppl: 10.399094653657736], batch size: 70
+2022-12-10 13:30:27,731 INFO [train.py:421] (5/8) Epoch 2, batch 48800, loss[loss=2.489, over 1680.00 frames. , ppl: 12.047410600069838] tot_loss[loss=2.341, over 5551401.17 frames. , ppl: 10.392362296509834], batch size: 70
+2022-12-10 13:32:08,080 INFO [train.py:421] (5/8) Epoch 2, batch 49000, loss[loss=2.26, over 6230.00 frames. , ppl: 9.580960535525163] tot_loss[loss=2.341, over 5568077.26 frames. , ppl: 10.389242332640737], batch size: 70
+2022-12-10 13:32:08,081 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 13:32:08,810 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.328, over 211138.00 frames. , ppl: 10.253641134938565
+2022-12-10 13:33:44,603 INFO [train.py:421] (5/8) Epoch 2, batch 49200, loss[loss=2.683, over 980.00 frames. , ppl: 14.631784010814151] tot_loss[loss=2.341, over 5554095.03 frames. , ppl: 10.390663531239944], batch size: 70
+2022-12-10 13:35:23,414 INFO [train.py:421] (5/8) Epoch 2, batch 49400, loss[loss=2.384, over 1330.00 frames. , ppl: 10.851014958604274] tot_loss[loss=2.342, over 5532248.28 frames. , ppl: 10.40459315223442], batch size: 70
+2022-12-10 13:37:03,014 INFO [train.py:421] (5/8) Epoch 2, batch 49600, loss[loss=2.417, over 1190.00 frames. , ppl: 11.213046253499389] tot_loss[loss=2.343, over 5489676.91 frames. , ppl: 10.408701971911706], batch size: 70
+2022-12-10 13:38:43,472 INFO [train.py:421] (5/8) Epoch 2, batch 49800, loss[loss=2.322, over 2030.00 frames. , ppl: 10.200008332245691] tot_loss[loss=2.343, over 5460390.61 frames. , ppl: 10.410104260522255], batch size: 70
+2022-12-10 13:40:22,316 INFO [train.py:421] (5/8) Epoch 2, batch 50000, loss[loss=4.106, over 350.00 frames. , ppl: 60.719654232686956] tot_loss[loss=2.345, over 5399179.22 frames. , ppl: 10.432785381960189], batch size: 70
+2022-12-10 13:40:22,316 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 13:40:23,076 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.327, over 211138.00 frames. , ppl: 10.24865242939489
+2022-12-10 13:42:06,023 INFO [train.py:421] (5/8) Epoch 2, batch 50200, loss[loss=2.363, over 3500.00 frames. , ppl: 10.622054763756756] tot_loss[loss=2.345, over 5438502.90 frames. , ppl: 10.42892269797538], batch size: 70
+2022-12-10 13:43:50,046 INFO [train.py:421] (5/8) Epoch 2, batch 50400, loss[loss=2.469, over 1540.00 frames. , ppl: 11.807268085415053] tot_loss[loss=2.342, over 5506301.70 frames. , ppl: 10.402472079346158], batch size: 70
+2022-12-10 13:45:26,757 INFO [train.py:421] (5/8) Epoch 2, batch 50600, loss[loss=2.23, over 9870.00 frames. , ppl: 9.301321505798374] tot_loss[loss=2.343, over 5506099.30 frames. , ppl: 10.409714025817696], batch size: 70
+2022-12-10 13:47:06,669 INFO [train.py:421] (5/8) Epoch 2, batch 50800, loss[loss=2.383, over 2240.00 frames. , ppl: 10.838315574906943] tot_loss[loss=2.342, over 5489438.52 frames. , ppl: 10.406150229359188], batch size: 70
+2022-12-10 13:48:47,679 INFO [train.py:421] (5/8) Epoch 2, batch 51000, loss[loss=2.597, over 840.00 frames. , ppl: 13.426934388767876] tot_loss[loss=2.343, over 5470531.67 frames. , ppl: 10.412296125322023], batch size: 70
+2022-12-10 13:48:47,680 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 13:48:48,425 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.329, over 211138.00 frames. , ppl: 10.265193656171405
+2022-12-10 13:50:27,915 INFO [train.py:421] (5/8) Epoch 2, batch 51200, loss[loss=2.386, over 3360.00 frames. , ppl: 10.871162379170489] tot_loss[loss=2.343, over 5466710.56 frames. , ppl: 10.412294596400027], batch size: 70
+2022-12-10 13:52:05,836 INFO [train.py:421] (5/8) Epoch 2, batch 51400, loss[loss=2.259, over 5600.00 frames. , ppl: 9.575695474607128] tot_loss[loss=2.342, over 5500863.26 frames. , ppl: 10.406419615337164], batch size: 70
+2022-12-10 13:53:44,960 INFO [train.py:421] (5/8) Epoch 2, batch 51600, loss[loss=2.509, over 1120.00 frames. , ppl: 12.292235566716702] tot_loss[loss=2.342, over 5515471.58 frames. , ppl: 10.402455944836609], batch size: 70
+2022-12-10 13:55:22,668 INFO [train.py:421] (5/8) Epoch 2, batch 51800, loss[loss=2.233, over 4270.00 frames. , ppl: 9.32960005376172] tot_loss[loss=2.342, over 5518189.52 frames. , ppl: 10.401372879507317], batch size: 70
+2022-12-10 13:57:00,774 INFO [train.py:421] (5/8) Epoch 2, batch 52000, loss[loss=2.764, over 700.00 frames. , ppl: 15.862235006328314] tot_loss[loss=2.344, over 5435327.46 frames. , ppl: 10.425014464817913], batch size: 70
+2022-12-10 13:57:00,774 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 13:57:01,533 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.329, over 211138.00 frames. , ppl: 10.267428826974491
+2022-12-10 13:58:39,520 INFO [train.py:421] (5/8) Epoch 2, batch 52200, loss[loss=2.31, over 4200.00 frames. , ppl: 10.07506416515659] tot_loss[loss=2.344, over 5444063.21 frames. , ppl: 10.42302246689372], batch size: 70
+2022-12-10 14:00:16,937 INFO [train.py:421] (5/8) Epoch 2, batch 52400, loss[loss=2.268, over 3150.00 frames. , ppl: 9.655477989814042] tot_loss[loss=2.344, over 5413483.81 frames. , ppl: 10.419361243688542], batch size: 70
+2022-12-10 14:01:58,866 INFO [train.py:421] (5/8) Epoch 2, batch 52600, loss[loss=2.692, over 700.00 frames. , ppl: 14.759385496362215] tot_loss[loss=2.343, over 5411277.46 frames. , ppl: 10.416569678757849], batch size: 70
+2022-12-10 14:03:38,044 INFO [train.py:421] (5/8) Epoch 2, batch 52800, loss[loss=2.277, over 5040.00 frames. , ppl: 9.751767742111005] tot_loss[loss=2.343, over 5450915.07 frames. , ppl: 10.412357487555472], batch size: 70
+2022-12-10 14:05:17,825 INFO [train.py:421] (5/8) Epoch 2, batch 53000, loss[loss=2.259, over 7980.00 frames. , ppl: 9.576431241193553] tot_loss[loss=2.342, over 5485163.28 frames. , ppl: 10.402737579212857], batch size: 70
+2022-12-10 14:05:17,825 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 14:05:18,555 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.326, over 211138.00 frames. , ppl: 10.233852338507358
+2022-12-10 14:06:58,238 INFO [train.py:421] (5/8) Epoch 2, batch 53200, loss[loss=2.48, over 2660.00 frames. , ppl: 11.93836388955572] tot_loss[loss=2.343, over 5456002.10 frames. , ppl: 10.41481751513989], batch size: 70
+2022-12-10 14:08:37,562 INFO [train.py:421] (5/8) Epoch 2, batch 53400, loss[loss=2.962, over 630.00 frames. , ppl: 19.334060525781076] tot_loss[loss=2.342, over 5474339.38 frames. , ppl: 10.403233505905625], batch size: 70
+2022-12-10 14:10:19,462 INFO [train.py:421] (5/8) Epoch 2, batch 53600, loss[loss=2.833, over 630.00 frames. , ppl: 16.994010196723373] tot_loss[loss=2.342, over 5488570.56 frames. , ppl: 10.397540247913138], batch size: 70
+2022-12-10 14:11:57,899 INFO [train.py:421] (5/8) Epoch 2, batch 53800, loss[loss=2.361, over 1610.00 frames. , ppl: 10.600438244044236] tot_loss[loss=2.342, over 5478224.23 frames. , ppl: 10.400884894133725], batch size: 70
+2022-12-10 14:13:35,070 INFO [train.py:421] (5/8) Epoch 2, batch 54000, loss[loss=2.315, over 7070.00 frames. , ppl: 10.122035082721474] tot_loss[loss=2.343, over 5422961.46 frames. , ppl: 10.410545239317294], batch size: 70
+2022-12-10 14:13:35,071 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 14:13:35,833 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.325, over 211138.00 frames. , ppl: 10.222168662360373
+2022-12-10 14:15:13,891 INFO [train.py:421] (5/8) Epoch 2, batch 54200, loss[loss=3.155, over 490.00 frames. , ppl: 23.44538791688027] tot_loss[loss=2.342, over 5438524.82 frames. , ppl: 10.404549732437081], batch size: 70
+2022-12-10 14:16:54,955 INFO [train.py:421] (5/8) Epoch 2, batch 54400, loss[loss=2.648, over 910.00 frames. , ppl: 14.12454891805841] tot_loss[loss=2.342, over 5432151.87 frames. , ppl: 10.40323937178488], batch size: 70
+2022-12-10 14:18:34,363 INFO [train.py:421] (5/8) Epoch 2, batch 54600, loss[loss=2.506, over 1540.00 frames. , ppl: 12.26132584329766] tot_loss[loss=2.342, over 5445069.79 frames. , ppl: 10.398753515567543], batch size: 70
+2022-12-10 14:20:12,846 INFO [train.py:421] (5/8) Epoch 2, batch 54800, loss[loss=2.252, over 4690.00 frames. , ppl: 9.503217072234358] tot_loss[loss=2.341, over 5496513.89 frames. , ppl: 10.392116561267015], batch size: 70
+2022-12-10 14:21:50,926 INFO [train.py:421] (5/8) Epoch 2, batch 55000, loss[loss=2.29, over 3780.00 frames. , ppl: 9.875762259832607] tot_loss[loss=2.343, over 5464888.47 frames. , ppl: 10.408932174349697], batch size: 70
+2022-12-10 14:21:50,926 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 14:21:51,690 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.326, over 211138.00 frames. , ppl: 10.232852694220295
+2022-12-10 14:23:32,583 INFO [train.py:421] (5/8) Epoch 2, batch 55200, loss[loss=2.281, over 3780.00 frames. , ppl: 9.790249343583618] tot_loss[loss=2.344, over 5429882.88 frames. , ppl: 10.417790389039935], batch size: 70
+2022-12-10 14:25:08,970 INFO [train.py:421] (5/8) Epoch 2, batch 55400, loss[loss=2.45, over 1540.00 frames. , ppl: 11.585644610332604] tot_loss[loss=2.344, over 5422676.71 frames. , ppl: 10.423091844263812], batch size: 70
+2022-12-10 14:26:48,398 INFO [train.py:421] (5/8) Epoch 2, batch 55600, loss[loss=2.271, over 3920.00 frames. , ppl: 9.687784891245645] tot_loss[loss=2.343, over 5442217.75 frames. , ppl: 10.41485518625667], batch size: 70
+2022-12-10 14:28:34,501 INFO [train.py:421] (5/8) Epoch 2, batch 55800, loss[loss=2.511, over 910.00 frames. , ppl: 12.31407923045909] tot_loss[loss=2.343, over 5445574.36 frames. , ppl: 10.412423310269391], batch size: 70
+2022-12-10 14:30:15,194 INFO [train.py:421] (5/8) Epoch 2, batch 56000, loss[loss=2.505, over 1470.00 frames. , ppl: 12.247509348012326] tot_loss[loss=2.343, over 5475505.95 frames. , ppl: 10.407262229764395], batch size: 70
+2022-12-10 14:30:15,194 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 14:30:15,958 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.327, over 211138.00 frames. , ppl: 10.2458754038067
+2022-12-10 14:31:57,753 INFO [train.py:421] (5/8) Epoch 2, batch 56200, loss[loss=3.162, over 560.00 frames. , ppl: 23.618709764422544] tot_loss[loss=2.342, over 5519101.71 frames. , ppl: 10.397126088249763], batch size: 70
+2022-12-10 14:33:41,513 INFO [train.py:421] (5/8) Epoch 2, batch 56400, loss[loss=2.277, over 2170.00 frames. , ppl: 9.747198415447134] tot_loss[loss=2.343, over 5489513.83 frames. , ppl: 10.408421518688531], batch size: 70
+2022-12-10 14:35:23,939 INFO [train.py:421] (5/8) Epoch 2, batch 56600, loss[loss=2.653, over 1050.00 frames. , ppl: 14.19386660872935] tot_loss[loss=2.342, over 5498080.99 frames. , ppl: 10.404723324633743], batch size: 70
+2022-12-10 14:37:04,930 INFO [train.py:421] (5/8) Epoch 2, batch 56800, loss[loss=2.439, over 1470.00 frames. , ppl: 11.466684344590467] tot_loss[loss=2.341, over 5541380.15 frames. , ppl: 10.389317849520985], batch size: 70
+2022-12-10 14:38:48,269 INFO [train.py:421] (5/8) Epoch 2, batch 57000, loss[loss=2.41, over 2030.00 frames. , ppl: 11.136335784889722] tot_loss[loss=2.34, over 5586842.27 frames. , ppl: 10.376557482196997], batch size: 70
+2022-12-10 14:38:48,270 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 14:38:49,030 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.324, over 211138.00 frames. , ppl: 10.216746149309023
+2022-12-10 14:40:28,410 INFO [train.py:421] (5/8) Epoch 2, batch 57200, loss[loss=2.316, over 2100.00 frames. , ppl: 10.136490975014278] tot_loss[loss=2.34, over 5555056.29 frames. , ppl: 10.377504496752076], batch size: 70
+2022-12-10 14:42:06,873 INFO [train.py:421] (5/8) Epoch 2, batch 57400, loss[loss=2.394, over 2380.00 frames. , ppl: 10.958027652995131] tot_loss[loss=2.338, over 5590467.05 frames. , ppl: 10.364574497219213], batch size: 70
+2022-12-10 14:43:46,025 INFO [train.py:421] (5/8) Epoch 2, batch 57600, loss[loss=2.282, over 4480.00 frames. , ppl: 9.795910309890187] tot_loss[loss=2.337, over 5616533.79 frames. , ppl: 10.353712938245788], batch size: 70
+2022-12-10 14:45:24,393 INFO [train.py:421] (5/8) Epoch 2, batch 57800, loss[loss=2.606, over 1050.00 frames. , ppl: 13.551385496771903] tot_loss[loss=2.338, over 5591363.01 frames. , ppl: 10.365543310249642], batch size: 70
+2022-12-10 14:47:02,165 INFO [train.py:421] (5/8) Epoch 2, batch 58000, loss[loss=2.276, over 2310.00 frames. , ppl: 9.741042682879783] tot_loss[loss=2.34, over 5555024.47 frames. , ppl: 10.380463565459872], batch size: 70
+2022-12-10 14:47:02,165 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 14:47:02,926 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.325, over 211138.00 frames. , ppl: 10.22471528659933
+2022-12-10 14:48:40,205 INFO [train.py:421] (5/8) Epoch 2, batch 58200, loss[loss=2.314, over 1610.00 frames. , ppl: 10.116953932493102] tot_loss[loss=2.34, over 5522959.20 frames. , ppl: 10.38492292152147], batch size: 70
+2022-12-10 14:50:20,541 INFO [train.py:421] (5/8) Epoch 2, batch 58400, loss[loss=2.259, over 6300.00 frames. , ppl: 9.568968267601543] tot_loss[loss=2.342, over 5484926.66 frames. , ppl: 10.397491386953705], batch size: 70
+2022-12-10 14:51:59,928 INFO [train.py:421] (5/8) Epoch 2, batch 58600, loss[loss=2.255, over 6650.00 frames. , ppl: 9.530967450717412] tot_loss[loss=2.341, over 5511250.20 frames. , ppl: 10.388499852753293], batch size: 70
+2022-12-10 14:53:41,810 INFO [train.py:421] (5/8) Epoch 2, batch 58800, loss[loss=2.761, over 700.00 frames. , ppl: 15.822063311384877] tot_loss[loss=2.34, over 5552830.33 frames. , ppl: 10.3767896282472], batch size: 70
+2022-12-10 14:55:21,057 INFO [train.py:421] (5/8) Epoch 2, batch 59000, loss[loss=2.289, over 3220.00 frames. , ppl: 9.865032016164543] tot_loss[loss=2.341, over 5519744.17 frames. , ppl: 10.387187202590852], batch size: 70
+2022-12-10 14:55:21,058 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 14:55:21,801 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.325, over 211138.00 frames. , ppl: 10.225733811230693
+2022-12-10 14:57:02,593 INFO [train.py:421] (5/8) Epoch 2, batch 59200, loss[loss=2.326, over 4830.00 frames. , ppl: 10.240298748730067] tot_loss[loss=2.341, over 5514768.80 frames. , ppl: 10.388531631084803], batch size: 70
+2022-12-10 14:58:41,089 INFO [train.py:421] (5/8) Epoch 2, batch 59400, loss[loss=2.471, over 1400.00 frames. , ppl: 11.834389955848943] tot_loss[loss=2.342, over 5490716.77 frames. , ppl: 10.397288502714117], batch size: 70
+2022-12-10 15:00:19,437 INFO [train.py:421] (5/8) Epoch 2, batch 59600, loss[loss=2.257, over 3570.00 frames. , ppl: 9.551753409011534] tot_loss[loss=2.342, over 5493869.00 frames. , ppl: 10.397702033272534], batch size: 70
+2022-12-10 15:02:02,528 INFO [train.py:421] (5/8) Epoch 2, batch 59800, loss[loss=2.349, over 7560.00 frames. , ppl: 10.47374643036108] tot_loss[loss=2.34, over 5511226.50 frames. , ppl: 10.382526545269318], batch size: 70
+2022-12-10 15:03:44,826 INFO [train.py:421] (5/8) Epoch 2, batch 60000, loss[loss=3.613, over 420.00 frames. , ppl: 37.071575472163815] tot_loss[loss=2.338, over 5551864.12 frames. , ppl: 10.365442380333079], batch size: 70
+2022-12-10 15:03:44,827 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 15:03:45,573 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.324, over 211138.00 frames. , ppl: 10.21343657885031
+2022-12-10 15:05:24,002 INFO [train.py:421] (5/8) Epoch 2, batch 60200, loss[loss=2.325, over 3780.00 frames. , ppl: 10.224410806819042] tot_loss[loss=2.34, over 5484016.53 frames. , ppl: 10.385820272148699], batch size: 70
+2022-12-10 15:07:02,397 INFO [train.py:421] (5/8) Epoch 2, batch 60400, loss[loss=2.398, over 3360.00 frames. , ppl: 11.003727174793061] tot_loss[loss=2.34, over 5501157.67 frames. , ppl: 10.377539767688614], batch size: 70
+2022-12-10 15:08:40,165 INFO [train.py:421] (5/8) Epoch 2, batch 60600, loss[loss=2.456, over 2450.00 frames. , ppl: 11.663148285220391] tot_loss[loss=2.34, over 5510741.35 frames. , ppl: 10.38068156952301], batch size: 70
+2022-12-10 15:10:20,034 INFO [train.py:421] (5/8) Epoch 2, batch 60800, loss[loss=2.248, over 5390.00 frames. , ppl: 9.471158624263325] tot_loss[loss=2.339, over 5541544.17 frames. , ppl: 10.372061053605448], batch size: 70
+2022-12-10 15:11:55,175 INFO [train.py:421] (5/8) Epoch 2, batch 61000, loss[loss=2.437, over 1680.00 frames. , ppl: 11.442965433469334] tot_loss[loss=2.34, over 5554567.47 frames. , ppl: 10.384774646930786], batch size: 70
+2022-12-10 15:11:55,176 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 15:11:55,927 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.324, over 211138.00 frames. , ppl: 10.211778417126597
+2022-12-10 15:13:35,413 INFO [train.py:421] (5/8) Epoch 2, batch 61200, loss[loss=2.315, over 3150.00 frames. , ppl: 10.124648266356372] tot_loss[loss=2.342, over 5494328.88 frames. , ppl: 10.398525448583452], batch size: 70
+2022-12-10 15:15:14,572 INFO [train.py:421] (5/8) Epoch 2, batch 61400, loss[loss=2.46, over 1470.00 frames. , ppl: 11.708557759918513] tot_loss[loss=2.342, over 5475706.44 frames. , ppl: 10.399283574847864], batch size: 70
+2022-12-10 15:16:52,145 INFO [train.py:421] (5/8) Epoch 2, batch 61600, loss[loss=2.273, over 3010.00 frames. , ppl: 9.711713226032607] tot_loss[loss=2.342, over 5469339.37 frames. , ppl: 10.402000300753867], batch size: 70
+2022-12-10 15:18:32,314 INFO [train.py:421] (5/8) Epoch 2, batch 61800, loss[loss=2.402, over 1540.00 frames. , ppl: 11.044727060745304] tot_loss[loss=2.342, over 5447006.41 frames. , ppl: 10.406923108959958], batch size: 70
+2022-12-10 15:20:14,993 INFO [train.py:421] (5/8) Epoch 2, batch 62000, loss[loss=2.572, over 840.00 frames. , ppl: 13.09240050226337] tot_loss[loss=2.342, over 5462768.62 frames. , ppl: 10.39829092334631], batch size: 70
+2022-12-10 15:20:14,993 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 15:20:15,753 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.324, over 211138.00 frames. , ppl: 10.219063032959546
+2022-12-10 15:21:54,084 INFO [train.py:421] (5/8) Epoch 2, batch 62200, loss[loss=2.303, over 3080.00 frames. , ppl: 10.00436004463346] tot_loss[loss=2.343, over 5435925.51 frames. , ppl: 10.409068152077964], batch size: 70
+2022-12-10 15:23:35,865 INFO [train.py:421] (5/8) Epoch 2, batch 62400, loss[loss=2.27, over 3430.00 frames. , ppl: 9.680681263194787] tot_loss[loss=2.341, over 5468531.49 frames. , ppl: 10.393573562817673], batch size: 70
+2022-12-10 15:25:13,327 INFO [train.py:421] (5/8) Epoch 2, batch 62600, loss[loss=2.197, over 4200.00 frames. , ppl: 9.00045733121818] tot_loss[loss=2.341, over 5483165.28 frames. , ppl: 10.393027232362108], batch size: 70
+2022-12-10 15:26:54,709 INFO [train.py:421] (5/8) Epoch 2, batch 62800, loss[loss=2.539, over 1120.00 frames. , ppl: 12.667551875036226] tot_loss[loss=2.341, over 5482359.55 frames. , ppl: 10.388504805829175], batch size: 70
+2022-12-10 15:28:32,241 INFO [train.py:421] (5/8) Epoch 2, batch 63000, loss[loss=2.857, over 630.00 frames. , ppl: 17.416418433519745] tot_loss[loss=2.341, over 5471796.71 frames. , ppl: 10.389090769647918], batch size: 70
+2022-12-10 15:28:32,241 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 15:28:33,000 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.323, over 211138.00 frames. , ppl: 10.20283624301792
+2022-12-10 15:30:14,790 INFO [train.py:421] (5/8) Epoch 2, batch 63200, loss[loss=2.26, over 5670.00 frames. , ppl: 9.583186878434029] tot_loss[loss=2.342, over 5436689.48 frames. , ppl: 10.397343305691042], batch size: 70
+2022-12-10 15:31:59,064 INFO [train.py:421] (5/8) Epoch 2, batch 63400, loss[loss=2.392, over 1540.00 frames. , ppl: 10.939838465065014] tot_loss[loss=2.341, over 5472995.78 frames. , ppl: 10.389017562199918], batch size: 70
+2022-12-10 15:33:37,719 INFO [train.py:421] (5/8) Epoch 2, batch 63600, loss[loss=3.087, over 630.00 frames. , ppl: 21.911684876425642] tot_loss[loss=2.341, over 5473857.79 frames. , ppl: 10.386605545751761], batch size: 70
+2022-12-10 15:35:17,819 INFO [train.py:421] (5/8) Epoch 2, batch 63800, loss[loss=2.374, over 2520.00 frames. , ppl: 10.740731549771349] tot_loss[loss=2.341, over 5440754.50 frames. , ppl: 10.393813011414077], batch size: 70
+2022-12-10 15:36:57,203 INFO [train.py:421] (5/8) Epoch 2, batch 64000, loss[loss=2.52, over 980.00 frames. , ppl: 12.42395439372591] tot_loss[loss=2.341, over 5487013.61 frames. , ppl: 10.39095245276705], batch size: 70
+2022-12-10 15:36:57,204 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 15:36:57,966 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.324, over 211138.00 frames. , ppl: 10.220371427142487
+2022-12-10 15:38:37,736 INFO [train.py:421] (5/8) Epoch 2, batch 64200, loss[loss=2.251, over 7910.00 frames. , ppl: 9.498504293241952] tot_loss[loss=2.339, over 5546244.12 frames. , ppl: 10.367891480451787], batch size: 70
+2022-12-10 15:40:17,493 INFO [train.py:421] (5/8) Epoch 2, batch 64400, loss[loss=2.459, over 1050.00 frames. , ppl: 11.693337699581202] tot_loss[loss=2.339, over 5520948.40 frames. , ppl: 10.372862415282789], batch size: 70
+2022-12-10 15:41:55,772 INFO [train.py:421] (5/8) Epoch 2, batch 64600, loss[loss=2.538, over 1190.00 frames. , ppl: 12.654469064960315] tot_loss[loss=2.34, over 5492270.95 frames. , ppl: 10.381246746955432], batch size: 70
+2022-12-10 15:43:33,789 INFO [train.py:421] (5/8) Epoch 2, batch 64800, loss[loss=2.394, over 2170.00 frames. , ppl: 10.960226347685932] tot_loss[loss=2.338, over 5522992.61 frames. , ppl: 10.365646286995217], batch size: 70
+2022-12-10 15:45:14,653 INFO [train.py:421] (5/8) Epoch 2, batch 65000, loss[loss=2.416, over 2450.00 frames. , ppl: 11.20253829204402] tot_loss[loss=2.34, over 5489077.64 frames. , ppl: 10.382974381743919], batch size: 70
+2022-12-10 15:45:14,654 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 15:45:15,399 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.323, over 211138.00 frames. , ppl: 10.211031803223284
+2022-12-10 15:46:58,768 INFO [train.py:421] (5/8) Epoch 2, batch 65200, loss[loss=3.417, over 490.00 frames. , ppl: 30.472132266384786] tot_loss[loss=2.339, over 5543313.54 frames. , ppl: 10.370266194362559], batch size: 70
+2022-12-10 15:48:38,204 INFO [train.py:421] (5/8) Epoch 2, batch 65400, loss[loss=2.286, over 5600.00 frames. , ppl: 9.837228035156283] tot_loss[loss=2.339, over 5520615.12 frames. , ppl: 10.372564599681805], batch size: 70
+2022-12-10 15:50:19,932 INFO [train.py:421] (5/8) Epoch 2, batch 65600, loss[loss=2.82, over 630.00 frames. , ppl: 16.771090730077855] tot_loss[loss=2.338, over 5532472.44 frames. , ppl: 10.36442049728577], batch size: 70
+2022-12-10 15:51:59,510 INFO [train.py:421] (5/8) Epoch 2, batch 65800, loss[loss=2.304, over 2660.00 frames. , ppl: 10.011033236082742] tot_loss[loss=2.338, over 5518877.29 frames. , ppl: 10.36557631052715], batch size: 70
+2022-12-10 15:53:41,830 INFO [train.py:421] (5/8) Epoch 2, batch 66000, loss[loss=3.637, over 420.00 frames. , ppl: 37.9702088289762] tot_loss[loss=2.338, over 5555772.33 frames. , ppl: 10.357103574828042], batch size: 70
+2022-12-10 15:53:41,831 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 15:53:42,595 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.321, over 211138.00 frames. , ppl: 10.187981682837393
+2022-12-10 15:55:24,764 INFO [train.py:421] (5/8) Epoch 2, batch 66200, loss[loss=2.797, over 700.00 frames. , ppl: 16.39421226980979] tot_loss[loss=2.336, over 5604229.86 frames. , ppl: 10.340827554840448], batch size: 70
+2022-12-10 15:57:04,875 INFO [train.py:421] (5/8) Epoch 2, batch 66400, loss[loss=2.31, over 1610.00 frames. , ppl: 10.072243964868267] tot_loss[loss=2.337, over 5573603.64 frames. , ppl: 10.34902399700311], batch size: 70
+2022-12-10 15:58:43,336 INFO [train.py:421] (5/8) Epoch 2, batch 66600, loss[loss=2.485, over 1470.00 frames. , ppl: 12.004415019630146] tot_loss[loss=2.338, over 5540077.96 frames. , ppl: 10.357545449025961], batch size: 70
+2022-12-10 16:00:20,766 INFO [train.py:421] (5/8) Epoch 2, batch 66800, loss[loss=2.98, over 560.00 frames. , ppl: 19.68271630566817] tot_loss[loss=2.337, over 5518769.02 frames. , ppl: 10.354667022669362], batch size: 70
+2022-12-10 16:02:02,722 INFO [train.py:421] (5/8) Epoch 2, batch 67000, loss[loss=2.494, over 1190.00 frames. , ppl: 12.110182911649703] tot_loss[loss=2.336, over 5549320.64 frames. , ppl: 10.339528330317858], batch size: 70
+2022-12-10 16:02:02,722 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 16:02:03,483 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.323, over 211138.00 frames. , ppl: 10.203262098954827
+2022-12-10 16:03:41,290 INFO [train.py:421] (5/8) Epoch 2, batch 67200, loss[loss=2.585, over 910.00 frames. , ppl: 13.267874473563259] tot_loss[loss=2.335, over 5579965.69 frames. , ppl: 10.326036357660287], batch size: 70
+2022-12-10 16:05:20,727 INFO [train.py:421] (5/8) Epoch 2, batch 67400, loss[loss=2.29, over 3010.00 frames. , ppl: 9.874682660509784] tot_loss[loss=2.335, over 5567516.92 frames. , ppl: 10.32523107893898], batch size: 70
+2022-12-10 16:07:01,242 INFO [train.py:421] (5/8) Epoch 2, batch 67600, loss[loss=2.294, over 6020.00 frames. , ppl: 9.909816310480378] tot_loss[loss=2.335, over 5548994.92 frames. , ppl: 10.333748997929126], batch size: 70
+2022-12-10 16:08:44,331 INFO [train.py:421] (5/8) Epoch 2, batch 67800, loss[loss=2.446, over 1680.00 frames. , ppl: 11.536939579590571] tot_loss[loss=2.335, over 5541150.22 frames. , ppl: 10.334496455571061], batch size: 70
+2022-12-10 16:10:23,874 INFO [train.py:421] (5/8) Epoch 2, batch 68000, loss[loss=2.344, over 2520.00 frames. , ppl: 10.419956620991215] tot_loss[loss=2.335, over 5545481.89 frames. , ppl: 10.333277911196669], batch size: 70
+2022-12-10 16:10:23,875 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 16:10:24,633 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.323, over 211138.00 frames. , ppl: 10.201776210751579
+2022-12-10 16:12:06,495 INFO [train.py:421] (5/8) Epoch 2, batch 68200, loss[loss=3.623, over 420.00 frames. , ppl: 37.443058372224094] tot_loss[loss=2.336, over 5548400.76 frames. , ppl: 10.335647646902055], batch size: 70
+2022-12-10 16:13:47,743 INFO [train.py:421] (5/8) Epoch 2, batch 68400, loss[loss=2.429, over 1610.00 frames. , ppl: 11.351538496934445] tot_loss[loss=2.336, over 5533511.84 frames. , ppl: 10.336493325103243], batch size: 70
+2022-12-10 16:15:28,277 INFO [train.py:421] (5/8) Epoch 2, batch 68600, loss[loss=2.571, over 910.00 frames. , ppl: 13.07685282115231] tot_loss[loss=2.336, over 5540790.23 frames. , ppl: 10.337953812091325], batch size: 70
+2022-12-10 16:17:07,753 INFO [train.py:421] (5/8) Epoch 2, batch 68800, loss[loss=2.404, over 1190.00 frames. , ppl: 11.066172300198453] tot_loss[loss=2.336, over 5540713.97 frames. , ppl: 10.341223271936286], batch size: 70
+2022-12-10 16:18:50,975 INFO [train.py:421] (5/8) Epoch 2, batch 69000, loss[loss=2.234, over 6090.00 frames. , ppl: 9.340881780339739] tot_loss[loss=2.335, over 5579250.09 frames. , ppl: 10.324844580379759], batch size: 70
+2022-12-10 16:18:50,976 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 16:18:51,729 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.32, over 211138.00 frames. , ppl: 10.176567267443412
+2022-12-10 16:20:33,667 INFO [train.py:421] (5/8) Epoch 2, batch 69200, loss[loss=2.371, over 1890.00 frames. , ppl: 10.702762336645687] tot_loss[loss=2.334, over 5573765.98 frames. , ppl: 10.323736341637197], batch size: 70
+2022-12-10 16:22:12,410 INFO [train.py:421] (5/8) Epoch 2, batch 69400, loss[loss=2.409, over 2240.00 frames. , ppl: 11.119226327040224] tot_loss[loss=2.335, over 5560099.22 frames. , ppl: 10.325421645842368], batch size: 70
+2022-12-10 16:23:54,609 INFO [train.py:421] (5/8) Epoch 2, batch 69600, loss[loss=2.775, over 770.00 frames. , ppl: 16.031255006498675] tot_loss[loss=2.335, over 5561116.23 frames. , ppl: 10.324781706850553], batch size: 70
+2022-12-10 16:25:37,453 INFO [train.py:421] (5/8) Epoch 2, batch 69800, loss[loss=2.377, over 2310.00 frames. , ppl: 10.770611047694521] tot_loss[loss=2.336, over 5532622.26 frames. , ppl: 10.336649297920426], batch size: 70
+2022-12-10 16:27:18,878 INFO [train.py:421] (5/8) Epoch 2, batch 70000, loss[loss=2.422, over 1050.00 frames. , ppl: 11.272098315437297] tot_loss[loss=2.334, over 5581407.79 frames. , ppl: 10.317078311824737], batch size: 70
+2022-12-10 16:27:18,879 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 16:27:19,639 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.322, over 211138.00 frames. , ppl: 10.191266412638488
+2022-12-10 16:29:02,809 INFO [train.py:421] (5/8) Epoch 2, batch 70200, loss[loss=2.582, over 980.00 frames. , ppl: 13.228571868186776] tot_loss[loss=2.335, over 5548267.52 frames. , ppl: 10.327317980518782], batch size: 70
+2022-12-10 16:30:41,311 INFO [train.py:421] (5/8) Epoch 2, batch 70400, loss[loss=2.314, over 3710.00 frames. , ppl: 10.112810239708361] tot_loss[loss=2.334, over 5546113.27 frames. , ppl: 10.32166554130779], batch size: 70
+2022-12-10 16:32:24,246 INFO [train.py:421] (5/8) Epoch 2, batch 70600, loss[loss=2.27, over 3920.00 frames. , ppl: 9.683434416438896] tot_loss[loss=2.335, over 5523084.61 frames. , ppl: 10.33377349165209], batch size: 70
+2022-12-10 16:34:01,869 INFO [train.py:421] (5/8) Epoch 2, batch 70800, loss[loss=2.234, over 3080.00 frames. , ppl: 9.340847959449494] tot_loss[loss=2.336, over 5498234.03 frames. , ppl: 10.339089850633643], batch size: 70
+2022-12-10 16:35:42,854 INFO [train.py:421] (5/8) Epoch 2, batch 71000, loss[loss=2.585, over 1050.00 frames. , ppl: 13.259015476476257] tot_loss[loss=2.335, over 5552436.14 frames. , ppl: 10.32831155533357], batch size: 70
+2022-12-10 16:35:42,855 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 16:35:43,613 INFO [train.py:452] (5/8) Epoch 2, validation: loss=2.319, over 211138.00 frames. , ppl: 10.169496572570102
+2022-12-10 16:37:26,617 INFO [train.py:421] (5/8) Epoch 2, batch 71200, loss[loss=2.47, over 980.00 frames. , ppl: 11.821649895360324] tot_loss[loss=2.335, over 5549441.95 frames. , ppl: 10.329622444003244], batch size: 70
+2022-12-10 16:39:05,517 INFO [train.py:421] (5/8) Epoch 2, batch 71400, loss[loss=2.524, over 1260.00 frames. , ppl: 12.478098616659137] tot_loss[loss=2.336, over 5530752.82 frames. , ppl: 10.343695896047619], batch size: 70
+2022-12-10 16:40:47,647 INFO [train.py:421] (5/8) Epoch 2, batch 71600, loss[loss=2.261, over 4340.00 frames. , ppl: 9.59177320218421] tot_loss[loss=2.336, over 5536834.31 frames. , ppl: 10.34070509395776], batch size: 70
+2022-12-10 16:42:29,323 INFO [train.py:421] (5/8) Epoch 2, batch 71800, loss[loss=2.534, over 1190.00 frames. , ppl: 12.605839982860868] tot_loss[loss=2.337, over 5523190.41 frames. , ppl: 10.348994025401751], batch size: 70
+2022-12-10 16:43:45,587 INFO [train.py:421] (5/8) Epoch 3, batch 0, loss[loss=2.272, over 2240.00 frames. , ppl: 9.702926937690318] tot_loss[loss=2.272, over 2240.00 frames. , ppl: 9.702926937690318], batch size: 70
+2022-12-10 16:45:28,193 INFO [train.py:421] (5/8) Epoch 3, batch 200, loss[loss=2.843, over 700.00 frames. , ppl: 17.173866627420526] tot_loss[loss=2.332, over 522764.31 frames. , ppl: 10.294298183412604], batch size: 70
+2022-12-10 16:47:09,548 INFO [train.py:421] (5/8) Epoch 3, batch 400, loss[loss=2.181, over 11900.00 frames. , ppl: 8.851122162016132] tot_loss[loss=2.321, over 1037254.87 frames. , ppl: 10.181943951643508], batch size: 70
+2022-12-10 16:48:48,730 INFO [train.py:421] (5/8) Epoch 3, batch 600, loss[loss=2.195, over 5320.00 frames. , ppl: 8.97721963377368] tot_loss[loss=2.322, over 1457510.15 frames. , ppl: 10.19140755690154], batch size: 70
+2022-12-10 16:50:28,185 INFO [train.py:421] (5/8) Epoch 3, batch 800, loss[loss=2.25, over 6930.00 frames. , ppl: 9.487103459367104] tot_loss[loss=2.326, over 1810240.19 frames. , ppl: 10.236260879319746], batch size: 70
+2022-12-10 16:52:09,311 INFO [train.py:421] (5/8) Epoch 3, batch 1000, loss[loss=2.405, over 1190.00 frames. , ppl: 11.08107801754052] tot_loss[loss=2.324, over 2177961.66 frames. , ppl: 10.21926619406141], batch size: 70
+2022-12-10 16:52:09,311 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 16:52:10,090 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.319, over 211138.00 frames. , ppl: 10.164404381725806
+2022-12-10 16:53:53,057 INFO [train.py:421] (5/8) Epoch 3, batch 1200, loss[loss=2.593, over 770.00 frames. , ppl: 13.36908948080462] tot_loss[loss=2.325, over 2511785.55 frames. , ppl: 10.22238533758206], batch size: 70
+2022-12-10 16:55:35,512 INFO [train.py:421] (5/8) Epoch 3, batch 1400, loss[loss=2.183, over 3710.00 frames. , ppl: 8.87154143319919] tot_loss[loss=2.322, over 2857268.48 frames. , ppl: 10.198493853648666], batch size: 70
+2022-12-10 16:57:16,468 INFO [train.py:421] (5/8) Epoch 3, batch 1600, loss[loss=2.266, over 2030.00 frames. , ppl: 9.639582511990875] tot_loss[loss=2.324, over 3080977.38 frames. , ppl: 10.213965725215544], batch size: 70
+2022-12-10 16:58:58,235 INFO [train.py:421] (5/8) Epoch 3, batch 1800, loss[loss=2.293, over 2310.00 frames. , ppl: 9.907873877608875] tot_loss[loss=2.324, over 3325877.12 frames. , ppl: 10.212852317446778], batch size: 70
+2022-12-10 17:00:33,686 INFO [train.py:421] (5/8) Epoch 3, batch 2000, loss[loss=2.345, over 3500.00 frames. , ppl: 10.430559963603802] tot_loss[loss=2.327, over 3482130.59 frames. , ppl: 10.243421218468432], batch size: 70
+2022-12-10 17:00:33,686 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 17:00:34,445 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.321, over 211138.00 frames. , ppl: 10.184881917574328
+2022-12-10 17:02:13,490 INFO [train.py:421] (5/8) Epoch 3, batch 2200, loss[loss=2.261, over 5250.00 frames. , ppl: 9.595207594112955] tot_loss[loss=2.326, over 3674538.95 frames. , ppl: 10.232511025493258], batch size: 70
+2022-12-10 17:03:54,052 INFO [train.py:421] (5/8) Epoch 3, batch 2400, loss[loss=2.321, over 2310.00 frames. , ppl: 10.190445388333394] tot_loss[loss=2.325, over 3869811.77 frames. , ppl: 10.226476887767266], batch size: 70
+2022-12-10 17:05:32,333 INFO [train.py:421] (5/8) Epoch 3, batch 2600, loss[loss=2.257, over 4270.00 frames. , ppl: 9.557693219892583] tot_loss[loss=2.325, over 4011348.10 frames. , ppl: 10.230679486224847], batch size: 70
+2022-12-10 17:07:12,598 INFO [train.py:421] (5/8) Epoch 3, batch 2800, loss[loss=2.446, over 1470.00 frames. , ppl: 11.537329987224348] tot_loss[loss=2.325, over 4159107.63 frames. , ppl: 10.228045245952885], batch size: 70
+2022-12-10 17:08:55,682 INFO [train.py:421] (5/8) Epoch 3, batch 3000, loss[loss=2.29, over 4060.00 frames. , ppl: 9.878789654153616] tot_loss[loss=2.326, over 4270300.92 frames. , ppl: 10.236078746570414], batch size: 70
+2022-12-10 17:08:55,682 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 17:08:56,443 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.32, over 211138.00 frames. , ppl: 10.171098190408838
+2022-12-10 17:10:38,823 INFO [train.py:421] (5/8) Epoch 3, batch 3200, loss[loss=2.423, over 3360.00 frames. , ppl: 11.27867811565375] tot_loss[loss=2.327, over 4371449.42 frames. , ppl: 10.247111158370911], batch size: 70
+2022-12-10 17:12:18,589 INFO [train.py:421] (5/8) Epoch 3, batch 3400, loss[loss=2.397, over 1540.00 frames. , ppl: 10.984971705274155] tot_loss[loss=2.325, over 4509675.66 frames. , ppl: 10.2270189436226], batch size: 70
+2022-12-10 17:13:58,759 INFO [train.py:421] (5/8) Epoch 3, batch 3600, loss[loss=2.411, over 1680.00 frames. , ppl: 11.14968016904945] tot_loss[loss=2.324, over 4638068.16 frames. , ppl: 10.2181158068382], batch size: 70
+2022-12-10 17:15:38,952 INFO [train.py:421] (5/8) Epoch 3, batch 3800, loss[loss=2.543, over 1540.00 frames. , ppl: 12.713991105161616] tot_loss[loss=2.324, over 4768570.95 frames. , ppl: 10.21442026639246], batch size: 70
+2022-12-10 17:17:17,264 INFO [train.py:421] (5/8) Epoch 3, batch 4000, loss[loss=2.482, over 1050.00 frames. , ppl: 11.96612068504797] tot_loss[loss=2.325, over 4829118.11 frames. , ppl: 10.226834171651756], batch size: 70
+2022-12-10 17:17:17,265 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 17:17:18,025 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.319, over 211138.00 frames. , ppl: 10.163163321129922
+2022-12-10 17:18:57,037 INFO [train.py:421] (5/8) Epoch 3, batch 4200, loss[loss=2.206, over 7560.00 frames. , ppl: 9.0808973220016] tot_loss[loss=2.325, over 4922175.23 frames. , ppl: 10.224983831258152], batch size: 70
+2022-12-10 17:20:36,279 INFO [train.py:421] (5/8) Epoch 3, batch 4400, loss[loss=2.401, over 1890.00 frames. , ppl: 11.037196643107464] tot_loss[loss=2.325, over 4987256.99 frames. , ppl: 10.222708169581262], batch size: 70
+2022-12-10 17:22:15,217 INFO [train.py:421] (5/8) Epoch 3, batch 4600, loss[loss=2.351, over 5320.00 frames. , ppl: 10.49888608214702] tot_loss[loss=2.326, over 5037594.35 frames. , ppl: 10.23976296161952], batch size: 70
+2022-12-10 17:23:55,352 INFO [train.py:421] (5/8) Epoch 3, batch 4800, loss[loss=2.801, over 840.00 frames. , ppl: 16.46912182848187] tot_loss[loss=2.326, over 5105455.23 frames. , ppl: 10.233677217241423], batch size: 70
+2022-12-10 17:25:31,817 INFO [train.py:421] (5/8) Epoch 3, batch 5000, loss[loss=2.486, over 1750.00 frames. , ppl: 12.011682805496106] tot_loss[loss=2.326, over 5147783.35 frames. , ppl: 10.236460638943264], batch size: 70
+2022-12-10 17:25:31,818 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 17:25:32,565 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.319, over 211138.00 frames. , ppl: 10.167579177724292
+2022-12-10 17:27:12,749 INFO [train.py:421] (5/8) Epoch 3, batch 5200, loss[loss=2.423, over 3010.00 frames. , ppl: 11.274111918532984] tot_loss[loss=2.328, over 5168786.13 frames. , ppl: 10.253815181311007], batch size: 70
+2022-12-10 17:28:53,461 INFO [train.py:421] (5/8) Epoch 3, batch 5400, loss[loss=2.709, over 700.00 frames. , ppl: 15.007527331347232] tot_loss[loss=2.328, over 5212415.64 frames. , ppl: 10.252331476241388], batch size: 70
+2022-12-10 17:30:38,110 INFO [train.py:421] (5/8) Epoch 3, batch 5600, loss[loss=2.226, over 2730.00 frames. , ppl: 9.265469854672295] tot_loss[loss=2.327, over 5274233.25 frames. , ppl: 10.24234229359102], batch size: 70
+2022-12-10 17:32:21,557 INFO [train.py:421] (5/8) Epoch 3, batch 5800, loss[loss=2.189, over 8470.00 frames. , ppl: 8.923053497690212] tot_loss[loss=2.326, over 5327186.30 frames. , ppl: 10.234940383887265], batch size: 70
+2022-12-10 17:34:02,534 INFO [train.py:421] (5/8) Epoch 3, batch 6000, loss[loss=2.267, over 2310.00 frames. , ppl: 9.6481585259982] tot_loss[loss=2.325, over 5377702.06 frames. , ppl: 10.221966466618907], batch size: 70
+2022-12-10 17:34:02,535 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 17:34:03,279 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.317, over 211138.00 frames. , ppl: 10.146697899157441
+2022-12-10 17:35:42,402 INFO [train.py:421] (5/8) Epoch 3, batch 6200, loss[loss=2.257, over 4830.00 frames. , ppl: 9.553572752558182] tot_loss[loss=2.326, over 5341248.49 frames. , ppl: 10.237746996068347], batch size: 70
+2022-12-10 17:37:19,187 INFO [train.py:421] (5/8) Epoch 3, batch 6400, loss[loss=2.372, over 2310.00 frames. , ppl: 10.714526323226416] tot_loss[loss=2.327, over 5309523.19 frames. , ppl: 10.251714356943783], batch size: 70
+2022-12-10 17:38:59,188 INFO [train.py:421] (5/8) Epoch 3, batch 6600, loss[loss=2.854, over 630.00 frames. , ppl: 17.34969481447992] tot_loss[loss=2.327, over 5349949.59 frames. , ppl: 10.243437329389812], batch size: 70
+2022-12-10 17:40:37,939 INFO [train.py:421] (5/8) Epoch 3, batch 6800, loss[loss=2.373, over 2170.00 frames. , ppl: 10.731981799013122] tot_loss[loss=2.327, over 5363643.81 frames. , ppl: 10.25004141354205], batch size: 70
+2022-12-10 17:42:18,035 INFO [train.py:421] (5/8) Epoch 3, batch 7000, loss[loss=2.493, over 1120.00 frames. , ppl: 12.103113079386835] tot_loss[loss=2.328, over 5370730.49 frames. , ppl: 10.254357550167635], batch size: 70
+2022-12-10 17:42:18,036 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 17:42:18,803 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.318, over 211138.00 frames. , ppl: 10.159209485439936
+2022-12-10 17:44:00,387 INFO [train.py:421] (5/8) Epoch 3, batch 7200, loss[loss=2.329, over 2940.00 frames. , ppl: 10.26862290887869] tot_loss[loss=2.327, over 5371553.83 frames. , ppl: 10.25027292539188], batch size: 70
+2022-12-10 17:45:38,843 INFO [train.py:421] (5/8) Epoch 3, batch 7400, loss[loss=2.341, over 4690.00 frames. , ppl: 10.388354139824095] tot_loss[loss=2.327, over 5387927.75 frames. , ppl: 10.249303869313794], batch size: 70
+2022-12-10 17:47:16,693 INFO [train.py:421] (5/8) Epoch 3, batch 7600, loss[loss=2.288, over 2240.00 frames. , ppl: 9.853491743946321] tot_loss[loss=2.326, over 5405518.66 frames. , ppl: 10.238687233058846], batch size: 70
+2022-12-10 17:48:56,788 INFO [train.py:421] (5/8) Epoch 3, batch 7800, loss[loss=2.361, over 2870.00 frames. , ppl: 10.60102219644854] tot_loss[loss=2.327, over 5408302.72 frames. , ppl: 10.243507275658375], batch size: 70
+2022-12-10 17:50:40,362 INFO [train.py:421] (5/8) Epoch 3, batch 8000, loss[loss=2.421, over 2030.00 frames. , ppl: 11.262104259793901] tot_loss[loss=2.327, over 5425185.81 frames. , ppl: 10.244251915698472], batch size: 70
+2022-12-10 17:50:40,363 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 17:50:41,126 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.318, over 211138.00 frames. , ppl: 10.153032911836572
+2022-12-10 17:52:20,457 INFO [train.py:421] (5/8) Epoch 3, batch 8200, loss[loss=2.349, over 2170.00 frames. , ppl: 10.475617764789218] tot_loss[loss=2.327, over 5441416.42 frames. , ppl: 10.245396425771188], batch size: 70
+2022-12-10 17:53:58,391 INFO [train.py:421] (5/8) Epoch 3, batch 8400, loss[loss=2.426, over 1750.00 frames. , ppl: 11.312022197269279] tot_loss[loss=2.327, over 5448445.08 frames. , ppl: 10.246268740254527], batch size: 70
+2022-12-10 17:55:40,058 INFO [train.py:421] (5/8) Epoch 3, batch 8600, loss[loss=2.173, over 9590.00 frames. , ppl: 8.781278698379436] tot_loss[loss=2.327, over 5442187.75 frames. , ppl: 10.246124263892217], batch size: 70
+2022-12-10 17:57:23,082 INFO [train.py:421] (5/8) Epoch 3, batch 8800, loss[loss=2.176, over 8260.00 frames. , ppl: 8.809837326242485] tot_loss[loss=2.325, over 5505708.75 frames. , ppl: 10.231428817798212], batch size: 70
+2022-12-10 17:59:05,018 INFO [train.py:421] (5/8) Epoch 3, batch 9000, loss[loss=2.294, over 3290.00 frames. , ppl: 9.919269755061961] tot_loss[loss=2.325, over 5516021.76 frames. , ppl: 10.228361862570729], batch size: 70
+2022-12-10 17:59:05,018 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 17:59:05,778 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.318, over 211138.00 frames. , ppl: 10.158504303301722
+2022-12-10 18:00:47,962 INFO [train.py:421] (5/8) Epoch 3, batch 9200, loss[loss=2.201, over 7560.00 frames. , ppl: 9.032211720188405] tot_loss[loss=2.325, over 5518768.16 frames. , ppl: 10.227624843048998], batch size: 70
+2022-12-10 18:02:28,019 INFO [train.py:421] (5/8) Epoch 3, batch 9400, loss[loss=2.289, over 3710.00 frames. , ppl: 9.865370046777828] tot_loss[loss=2.325, over 5553824.39 frames. , ppl: 10.226832346860684], batch size: 70
+2022-12-10 18:04:09,142 INFO [train.py:421] (5/8) Epoch 3, batch 9600, loss[loss=2.216, over 6930.00 frames. , ppl: 9.172447681891772] tot_loss[loss=2.324, over 5581814.14 frames. , ppl: 10.218907877593903], batch size: 70
+2022-12-10 18:05:47,177 INFO [train.py:421] (5/8) Epoch 3, batch 9800, loss[loss=2.438, over 2170.00 frames. , ppl: 11.45376105361245] tot_loss[loss=2.324, over 5574667.44 frames. , ppl: 10.218732345252247], batch size: 70
+2022-12-10 18:07:29,482 INFO [train.py:421] (5/8) Epoch 3, batch 10000, loss[loss=2.398, over 1330.00 frames. , ppl: 11.000179469255743] tot_loss[loss=2.324, over 5590829.13 frames. , ppl: 10.220971795891948], batch size: 70
+2022-12-10 18:07:29,482 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 18:07:30,247 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.318, over 211138.00 frames. , ppl: 10.15942601173964
+2022-12-10 18:09:07,694 INFO [train.py:421] (5/8) Epoch 3, batch 10200, loss[loss=2.213, over 4270.00 frames. , ppl: 9.142589716634633] tot_loss[loss=2.325, over 5541020.70 frames. , ppl: 10.230388193210944], batch size: 70
+2022-12-10 18:10:48,659 INFO [train.py:421] (5/8) Epoch 3, batch 10400, loss[loss=2.245, over 5040.00 frames. , ppl: 9.441302393100747] tot_loss[loss=2.325, over 5552216.66 frames. , ppl: 10.226420536219882], batch size: 70
+2022-12-10 18:12:27,603 INFO [train.py:421] (5/8) Epoch 3, batch 10600, loss[loss=2.75, over 700.00 frames. , ppl: 15.643908574070837] tot_loss[loss=2.326, over 5520996.24 frames. , ppl: 10.23658366848508], batch size: 70
+2022-12-10 18:14:05,280 INFO [train.py:421] (5/8) Epoch 3, batch 10800, loss[loss=2.277, over 8050.00 frames. , ppl: 9.749418457151702] tot_loss[loss=2.328, over 5487552.96 frames. , ppl: 10.252705582390407], batch size: 70
+2022-12-10 18:15:47,986 INFO [train.py:421] (5/8) Epoch 3, batch 11000, loss[loss=2.323, over 7000.00 frames. , ppl: 10.205988025646272] tot_loss[loss=2.328, over 5483341.32 frames. , ppl: 10.256169524150106], batch size: 70
+2022-12-10 18:15:47,986 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 18:15:48,745 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.319, over 211138.00 frames. , ppl: 10.163211456412954
+2022-12-10 18:17:26,886 INFO [train.py:421] (5/8) Epoch 3, batch 11200, loss[loss=2.466, over 1330.00 frames. , ppl: 11.773615622688977] tot_loss[loss=2.329, over 5434457.33 frames. , ppl: 10.269984477906773], batch size: 70
+2022-12-10 18:19:04,367 INFO [train.py:421] (5/8) Epoch 3, batch 11400, loss[loss=3.425, over 420.00 frames. , ppl: 30.728788851782948] tot_loss[loss=2.328, over 5461752.47 frames. , ppl: 10.260637567177914], batch size: 70
+2022-12-10 18:20:41,828 INFO [train.py:421] (5/8) Epoch 3, batch 11600, loss[loss=2.362, over 2590.00 frames. , ppl: 10.60968234053175] tot_loss[loss=2.329, over 5449657.99 frames. , ppl: 10.265465014546068], batch size: 70
+2022-12-10 18:22:26,235 INFO [train.py:421] (5/8) Epoch 3, batch 11800, loss[loss=2.287, over 3010.00 frames. , ppl: 9.848792031367555] tot_loss[loss=2.329, over 5446118.74 frames. , ppl: 10.264513290368534], batch size: 70
+2022-12-10 18:24:10,841 INFO [train.py:421] (5/8) Epoch 3, batch 12000, loss[loss=2.315, over 910.00 frames. , ppl: 10.129296146778865] tot_loss[loss=2.328, over 5439540.42 frames. , ppl: 10.259633827258396], batch size: 70
+2022-12-10 18:24:10,841 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 18:24:11,627 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.317, over 211138.00 frames. , ppl: 10.149768016214665
+2022-12-10 18:25:51,545 INFO [train.py:421] (5/8) Epoch 3, batch 12200, loss[loss=2.777, over 630.00 frames. , ppl: 16.06277320335159] tot_loss[loss=2.328, over 5437773.91 frames. , ppl: 10.261696284410709], batch size: 70
+2022-12-10 18:27:29,144 INFO [train.py:421] (5/8) Epoch 3, batch 12400, loss[loss=2.339, over 1540.00 frames. , ppl: 10.369022943711945] tot_loss[loss=2.328, over 5469088.74 frames. , ppl: 10.261001293319392], batch size: 70
+2022-12-10 18:29:06,471 INFO [train.py:421] (5/8) Epoch 3, batch 12600, loss[loss=2.632, over 1120.00 frames. , ppl: 13.894935763924158] tot_loss[loss=2.329, over 5431758.10 frames. , ppl: 10.271596904776647], batch size: 70
+2022-12-10 18:30:46,103 INFO [train.py:421] (5/8) Epoch 3, batch 12800, loss[loss=2.315, over 2590.00 frames. , ppl: 10.12292688520306] tot_loss[loss=2.329, over 5452312.15 frames. , ppl: 10.266135032433333], batch size: 70
+2022-12-10 18:32:27,344 INFO [train.py:421] (5/8) Epoch 3, batch 13000, loss[loss=2.352, over 1540.00 frames. , ppl: 10.501946254995747] tot_loss[loss=2.329, over 5458431.28 frames. , ppl: 10.265032716245353], batch size: 70
+2022-12-10 18:32:27,345 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 18:32:28,107 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.318, over 211138.00 frames. , ppl: 10.154893455412658
+2022-12-10 18:34:13,397 INFO [train.py:421] (5/8) Epoch 3, batch 13200, loss[loss=2.289, over 5110.00 frames. , ppl: 9.865576797392865] tot_loss[loss=2.329, over 5466300.29 frames. , ppl: 10.265664202756971], batch size: 70
+2022-12-10 18:35:54,371 INFO [train.py:421] (5/8) Epoch 3, batch 13400, loss[loss=2.237, over 10220.00 frames. , ppl: 9.364968586426889] tot_loss[loss=2.329, over 5443186.21 frames. , ppl: 10.269200123386705], batch size: 70
+2022-12-10 18:37:33,219 INFO [train.py:421] (5/8) Epoch 3, batch 13600, loss[loss=2.408, over 1820.00 frames. , ppl: 11.111903645883283] tot_loss[loss=2.329, over 5451943.21 frames. , ppl: 10.26375266587264], batch size: 70
+2022-12-10 18:39:15,506 INFO [train.py:421] (5/8) Epoch 3, batch 13800, loss[loss=2.23, over 8260.00 frames. , ppl: 9.298226644104735] tot_loss[loss=2.328, over 5492822.10 frames. , ppl: 10.258096368528523], batch size: 70
+2022-12-10 18:40:59,171 INFO [train.py:421] (5/8) Epoch 3, batch 14000, loss[loss=3.119, over 490.00 frames. , ppl: 22.627106651927644] tot_loss[loss=2.329, over 5467866.28 frames. , ppl: 10.264604807821911], batch size: 70
+2022-12-10 18:40:59,171 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 18:40:59,950 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.317, over 211138.00 frames. , ppl: 10.146131741243801
+2022-12-10 18:42:39,841 INFO [train.py:421] (5/8) Epoch 3, batch 14200, loss[loss=2.277, over 5950.00 frames. , ppl: 9.750711954408445] tot_loss[loss=2.327, over 5468484.15 frames. , ppl: 10.250365406312797], batch size: 70
+2022-12-10 18:44:22,956 INFO [train.py:421] (5/8) Epoch 3, batch 14400, loss[loss=2.219, over 11760.00 frames. , ppl: 9.198859479979811] tot_loss[loss=2.328, over 5470300.90 frames. , ppl: 10.253145315116383], batch size: 70
+2022-12-10 18:46:05,764 INFO [train.py:421] (5/8) Epoch 3, batch 14600, loss[loss=2.728, over 700.00 frames. , ppl: 15.296549833600256] tot_loss[loss=2.329, over 5432954.29 frames. , ppl: 10.265872598111264], batch size: 70
+2022-12-10 18:47:46,146 INFO [train.py:421] (5/8) Epoch 3, batch 14800, loss[loss=2.275, over 3150.00 frames. , ppl: 9.728078853812193] tot_loss[loss=2.33, over 5440168.85 frames. , ppl: 10.274526701449474], batch size: 70
+2022-12-10 18:49:30,929 INFO [train.py:421] (5/8) Epoch 3, batch 15000, loss[loss=2.266, over 4900.00 frames. , ppl: 9.637508948627858] tot_loss[loss=2.331, over 5406035.14 frames. , ppl: 10.284122332994624], batch size: 70
+2022-12-10 18:49:30,929 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 18:49:31,678 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.315, over 211138.00 frames. , ppl: 10.126931133917406
+2022-12-10 18:51:11,705 INFO [train.py:421] (5/8) Epoch 3, batch 15200, loss[loss=2.301, over 3710.00 frames. , ppl: 9.98836995545512] tot_loss[loss=2.331, over 5402992.38 frames. , ppl: 10.283548527107154], batch size: 70
+2022-12-10 18:52:50,273 INFO [train.py:421] (5/8) Epoch 3, batch 15400, loss[loss=2.286, over 3780.00 frames. , ppl: 9.839532634664609] tot_loss[loss=2.331, over 5382342.41 frames. , ppl: 10.290740605081611], batch size: 70
+2022-12-10 18:54:28,574 INFO [train.py:421] (5/8) Epoch 3, batch 15600, loss[loss=2.509, over 1330.00 frames. , ppl: 12.295422604139725] tot_loss[loss=2.331, over 5379955.12 frames. , ppl: 10.292704022264111], batch size: 70
+2022-12-10 18:56:05,413 INFO [train.py:421] (5/8) Epoch 3, batch 15800, loss[loss=2.241, over 3920.00 frames. , ppl: 9.405054988540433] tot_loss[loss=2.331, over 5384887.76 frames. , ppl: 10.289223667758446], batch size: 70
+2022-12-10 18:57:45,639 INFO [train.py:421] (5/8) Epoch 3, batch 16000, loss[loss=2.286, over 4760.00 frames. , ppl: 9.832933016430387] tot_loss[loss=2.331, over 5389073.17 frames. , ppl: 10.284723470502875], batch size: 70
+2022-12-10 18:57:45,639 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 18:57:46,369 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.316, over 211138.00 frames. , ppl: 10.134522713241985
+2022-12-10 18:59:25,143 INFO [train.py:421] (5/8) Epoch 3, batch 16200, loss[loss=2.483, over 980.00 frames. , ppl: 11.976161450399923] tot_loss[loss=2.329, over 5435747.59 frames. , ppl: 10.267562471741455], batch size: 70
+2022-12-10 19:01:05,960 INFO [train.py:421] (5/8) Epoch 3, batch 16400, loss[loss=2.226, over 2730.00 frames. , ppl: 9.25950419986886] tot_loss[loss=2.33, over 5399699.83 frames. , ppl: 10.278419668308112], batch size: 70
+2022-12-10 19:02:45,402 INFO [train.py:421] (5/8) Epoch 3, batch 16600, loss[loss=2.601, over 770.00 frames. , ppl: 13.472985490946014] tot_loss[loss=2.331, over 5393949.45 frames. , ppl: 10.284673603054989], batch size: 70
+2022-12-10 19:04:28,700 INFO [train.py:421] (5/8) Epoch 3, batch 16800, loss[loss=2.275, over 2940.00 frames. , ppl: 9.724334583895285] tot_loss[loss=2.33, over 5407541.01 frames. , ppl: 10.27381307588695], batch size: 70
+2022-12-10 19:06:10,450 INFO [train.py:421] (5/8) Epoch 3, batch 17000, loss[loss=2.257, over 3570.00 frames. , ppl: 9.55022762490296] tot_loss[loss=2.327, over 5485391.58 frames. , ppl: 10.25093756783369], batch size: 70
+2022-12-10 19:06:10,451 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 19:06:11,210 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.317, over 211138.00 frames. , ppl: 10.149083017239782
+2022-12-10 19:07:50,801 INFO [train.py:421] (5/8) Epoch 3, batch 17200, loss[loss=2.227, over 4550.00 frames. , ppl: 9.270757830139337] tot_loss[loss=2.327, over 5470354.78 frames. , ppl: 10.251559300308797], batch size: 70
+2022-12-10 19:09:33,685 INFO [train.py:421] (5/8) Epoch 3, batch 17400, loss[loss=2.807, over 700.00 frames. , ppl: 16.558852673575178] tot_loss[loss=2.327, over 5491333.16 frames. , ppl: 10.24515377176923], batch size: 70
+2022-12-10 19:11:12,084 INFO [train.py:421] (5/8) Epoch 3, batch 17600, loss[loss=2.391, over 2170.00 frames. , ppl: 10.922361614310205] tot_loss[loss=2.327, over 5507439.58 frames. , ppl: 10.242435659112635], batch size: 70
+2022-12-10 19:12:53,522 INFO [train.py:421] (5/8) Epoch 3, batch 17800, loss[loss=5.057, over 280.00 frames. , ppl: 157.05056424919292] tot_loss[loss=2.326, over 5517885.76 frames. , ppl: 10.2418003585202], batch size: 70
+2022-12-10 19:14:33,707 INFO [train.py:421] (5/8) Epoch 3, batch 18000, loss[loss=2.382, over 1540.00 frames. , ppl: 10.82356388548986] tot_loss[loss=2.326, over 5528997.66 frames. , ppl: 10.236666378438485], batch size: 70
+2022-12-10 19:14:33,707 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 19:14:34,466 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.315, over 211138.00 frames. , ppl: 10.129355082846555
+2022-12-10 19:16:15,851 INFO [train.py:421] (5/8) Epoch 3, batch 18200, loss[loss=2.22, over 3500.00 frames. , ppl: 9.209242407520952] tot_loss[loss=2.326, over 5533556.57 frames. , ppl: 10.236009243290379], batch size: 70
+2022-12-10 19:17:56,507 INFO [train.py:421] (5/8) Epoch 3, batch 18400, loss[loss=2.361, over 2170.00 frames. , ppl: 10.604735782117892] tot_loss[loss=2.327, over 5528448.64 frames. , ppl: 10.242175136455767], batch size: 70
+2022-12-10 19:19:35,455 INFO [train.py:421] (5/8) Epoch 3, batch 18600, loss[loss=2.441, over 1120.00 frames. , ppl: 11.483647561526372] tot_loss[loss=2.327, over 5511313.35 frames. , ppl: 10.251967702836092], batch size: 70
+2022-12-10 19:21:19,541 INFO [train.py:421] (5/8) Epoch 3, batch 18800, loss[loss=3.541, over 420.00 frames. , ppl: 34.497670872830284] tot_loss[loss=2.326, over 5567492.28 frames. , ppl: 10.240859668653744], batch size: 70
+2022-12-10 19:22:57,474 INFO [train.py:421] (5/8) Epoch 3, batch 19000, loss[loss=2.333, over 2310.00 frames. , ppl: 10.308479462412409] tot_loss[loss=2.327, over 5549677.25 frames. , ppl: 10.245265931771447], batch size: 70
+2022-12-10 19:22:57,475 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 19:22:58,219 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.317, over 211138.00 frames. , ppl: 10.14855577978015
+2022-12-10 19:24:36,580 INFO [train.py:421] (5/8) Epoch 3, batch 19200, loss[loss=2.254, over 5950.00 frames. , ppl: 9.52879447401331] tot_loss[loss=2.326, over 5566196.21 frames. , ppl: 10.237194098534697], batch size: 70
+2022-12-10 19:26:16,122 INFO [train.py:421] (5/8) Epoch 3, batch 19400, loss[loss=2.495, over 1050.00 frames. , ppl: 12.120656905734162] tot_loss[loss=2.326, over 5560824.88 frames. , ppl: 10.232299561377076], batch size: 70
+2022-12-10 19:27:55,666 INFO [train.py:421] (5/8) Epoch 3, batch 19600, loss[loss=2.249, over 4060.00 frames. , ppl: 9.476965829859225] tot_loss[loss=2.324, over 5584419.42 frames. , ppl: 10.217280261556368], batch size: 70
+2022-12-10 19:29:36,902 INFO [train.py:421] (5/8) Epoch 3, batch 19800, loss[loss=2.352, over 2030.00 frames. , ppl: 10.508321210877902] tot_loss[loss=2.324, over 5580830.43 frames. , ppl: 10.215573754563867], batch size: 70
+2022-12-10 19:31:19,082 INFO [train.py:421] (5/8) Epoch 3, batch 20000, loss[loss=2.261, over 5180.00 frames. , ppl: 9.591708628784797] tot_loss[loss=2.324, over 5587406.73 frames. , ppl: 10.212788402586618], batch size: 70
+2022-12-10 19:31:19,082 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 19:31:19,811 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.315, over 211138.00 frames. , ppl: 10.12899827473271
+2022-12-10 19:33:00,567 INFO [train.py:421] (5/8) Epoch 3, batch 20200, loss[loss=2.638, over 1120.00 frames. , ppl: 13.983437902548399] tot_loss[loss=2.324, over 5594544.15 frames. , ppl: 10.213007577198468], batch size: 70
+2022-12-10 19:34:42,566 INFO [train.py:421] (5/8) Epoch 3, batch 20400, loss[loss=2.77, over 840.00 frames. , ppl: 15.951363779013176] tot_loss[loss=2.324, over 5564565.26 frames. , ppl: 10.217609778850969], batch size: 70
+2022-12-10 19:36:23,595 INFO [train.py:421] (5/8) Epoch 3, batch 20600, loss[loss=2.35, over 4550.00 frames. , ppl: 10.484206005196922] tot_loss[loss=2.324, over 5590030.08 frames. , ppl: 10.213686154186426], batch size: 70
+2022-12-10 19:38:05,419 INFO [train.py:421] (5/8) Epoch 3, batch 20800, loss[loss=2.541, over 1190.00 frames. , ppl: 12.69314903087385] tot_loss[loss=2.324, over 5588236.44 frames. , ppl: 10.212512439162037], batch size: 70
+2022-12-10 19:39:45,172 INFO [train.py:421] (5/8) Epoch 3, batch 21000, loss[loss=2.457, over 840.00 frames. , ppl: 11.672875697462736] tot_loss[loss=2.324, over 5595349.97 frames. , ppl: 10.213993812334088], batch size: 70
+2022-12-10 19:39:45,172 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 19:39:45,934 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.314, over 211138.00 frames. , ppl: 10.115766279020953
+2022-12-10 19:41:26,043 INFO [train.py:421] (5/8) Epoch 3, batch 21200, loss[loss=2.651, over 770.00 frames. , ppl: 14.166560546057775] tot_loss[loss=2.324, over 5583729.30 frames. , ppl: 10.219719535394267], batch size: 70
+2022-12-10 19:43:03,101 INFO [train.py:421] (5/8) Epoch 3, batch 21400, loss[loss=2.411, over 1890.00 frames. , ppl: 11.148977164691898] tot_loss[loss=2.326, over 5541368.64 frames. , ppl: 10.237565509447572], batch size: 70
+2022-12-10 19:44:44,725 INFO [train.py:421] (5/8) Epoch 3, batch 21600, loss[loss=2.162, over 6160.00 frames. , ppl: 8.687245199748777] tot_loss[loss=2.326, over 5530717.63 frames. , ppl: 10.238998500837774], batch size: 70
+2022-12-10 19:46:24,994 INFO [train.py:421] (5/8) Epoch 3, batch 21800, loss[loss=2.358, over 4970.00 frames. , ppl: 10.571937924125372] tot_loss[loss=2.325, over 5590243.76 frames. , ppl: 10.223854926024185], batch size: 70
+2022-12-10 19:48:03,110 INFO [train.py:421] (5/8) Epoch 3, batch 22000, loss[loss=2.197, over 5320.00 frames. , ppl: 8.995757823336488] tot_loss[loss=2.326, over 5542446.23 frames. , ppl: 10.234561687173699], batch size: 70
+2022-12-10 19:48:03,110 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 19:48:03,838 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.314, over 211138.00 frames. , ppl: 10.119331761662655
+2022-12-10 19:49:42,910 INFO [train.py:421] (5/8) Epoch 3, batch 22200, loss[loss=2.566, over 910.00 frames. , ppl: 13.009145804126579] tot_loss[loss=2.324, over 5581428.51 frames. , ppl: 10.22123394820928], batch size: 70
+2022-12-10 19:51:27,707 INFO [train.py:421] (5/8) Epoch 3, batch 22400, loss[loss=2.458, over 1330.00 frames. , ppl: 11.68238719485867] tot_loss[loss=2.324, over 5620171.99 frames. , ppl: 10.220038649800987], batch size: 70
+2022-12-10 19:53:09,801 INFO [train.py:421] (5/8) Epoch 3, batch 22600, loss[loss=2.414, over 2520.00 frames. , ppl: 11.174933965300513] tot_loss[loss=2.324, over 5628655.94 frames. , ppl: 10.220489542757699], batch size: 70
+2022-12-10 19:54:48,476 INFO [train.py:421] (5/8) Epoch 3, batch 22800, loss[loss=2.254, over 7700.00 frames. , ppl: 9.524990585657584] tot_loss[loss=2.324, over 5627825.71 frames. , ppl: 10.212122248314916], batch size: 70
+2022-12-10 19:56:25,441 INFO [train.py:421] (5/8) Epoch 3, batch 23000, loss[loss=2.256, over 3500.00 frames. , ppl: 9.549413807597851] tot_loss[loss=2.325, over 5570268.95 frames. , ppl: 10.222698658812838], batch size: 70
+2022-12-10 19:56:25,442 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 19:56:26,223 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.314, over 211138.00 frames. , ppl: 10.111284628584807
+2022-12-10 19:58:04,891 INFO [train.py:421] (5/8) Epoch 3, batch 23200, loss[loss=2.551, over 980.00 frames. , ppl: 12.822191179276778] tot_loss[loss=2.325, over 5532763.06 frames. , ppl: 10.228140931841203], batch size: 70
+2022-12-10 19:59:44,528 INFO [train.py:421] (5/8) Epoch 3, batch 23400, loss[loss=2.389, over 1120.00 frames. , ppl: 10.899603044925673] tot_loss[loss=2.323, over 5593956.56 frames. , ppl: 10.210873043643222], batch size: 70
+2022-12-10 20:01:22,837 INFO [train.py:421] (5/8) Epoch 3, batch 23600, loss[loss=2.312, over 2450.00 frames. , ppl: 10.092201473465266] tot_loss[loss=2.325, over 5547685.95 frames. , ppl: 10.224675982162681], batch size: 70
+2022-12-10 20:03:03,118 INFO [train.py:421] (5/8) Epoch 3, batch 23800, loss[loss=2.371, over 1120.00 frames. , ppl: 10.705672609358196] tot_loss[loss=2.325, over 5522321.36 frames. , ppl: 10.231094528030317], batch size: 70
+2022-12-10 20:04:45,087 INFO [train.py:421] (5/8) Epoch 3, batch 24000, loss[loss=2.298, over 2450.00 frames. , ppl: 9.958629811282837] tot_loss[loss=2.324, over 5549584.41 frames. , ppl: 10.216745874666456], batch size: 70
+2022-12-10 20:04:45,088 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 20:04:45,875 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.314, over 211138.00 frames. , ppl: 10.115662972133224
+2022-12-10 20:06:27,305 INFO [train.py:421] (5/8) Epoch 3, batch 24200, loss[loss=2.344, over 2940.00 frames. , ppl: 10.421239627596808] tot_loss[loss=2.323, over 5588319.50 frames. , ppl: 10.20997327106623], batch size: 70
+2022-12-10 20:08:06,570 INFO [train.py:421] (5/8) Epoch 3, batch 24400, loss[loss=2.425, over 1610.00 frames. , ppl: 11.302699029867261] tot_loss[loss=2.324, over 5570766.47 frames. , ppl: 10.218121107227395], batch size: 70
+2022-12-10 20:09:46,779 INFO [train.py:421] (5/8) Epoch 3, batch 24600, loss[loss=2.515, over 1260.00 frames. , ppl: 12.365424162579895] tot_loss[loss=2.324, over 5575483.49 frames. , ppl: 10.216906277646608], batch size: 70
+2022-12-10 20:11:29,174 INFO [train.py:421] (5/8) Epoch 3, batch 24800, loss[loss=2.703, over 910.00 frames. , ppl: 14.918312726483869] tot_loss[loss=2.323, over 5618714.11 frames. , ppl: 10.201294251382938], batch size: 70
+2022-12-10 20:13:09,481 INFO [train.py:421] (5/8) Epoch 3, batch 25000, loss[loss=2.242, over 3640.00 frames. , ppl: 9.40941977885253] tot_loss[loss=2.324, over 5588474.51 frames. , ppl: 10.213288815071218], batch size: 70
+2022-12-10 20:13:09,481 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 20:13:10,215 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.315, over 211138.00 frames. , ppl: 10.122748675677348
+2022-12-10 20:14:53,086 INFO [train.py:421] (5/8) Epoch 3, batch 25200, loss[loss=2.281, over 4410.00 frames. , ppl: 9.787049206049073] tot_loss[loss=2.322, over 5621749.88 frames. , ppl: 10.201120770400308], batch size: 70
+2022-12-10 20:16:34,327 INFO [train.py:421] (5/8) Epoch 3, batch 25400, loss[loss=2.484, over 1120.00 frames. , ppl: 11.986392641581661] tot_loss[loss=2.323, over 5598813.22 frames. , ppl: 10.208952823920107], batch size: 70
+2022-12-10 20:18:12,372 INFO [train.py:421] (5/8) Epoch 3, batch 25600, loss[loss=2.61, over 1260.00 frames. , ppl: 13.6057095069259] tot_loss[loss=2.324, over 5562992.81 frames. , ppl: 10.214881217415051], batch size: 70
+2022-12-10 20:19:49,967 INFO [train.py:421] (5/8) Epoch 3, batch 25800, loss[loss=2.144, over 3850.00 frames. , ppl: 8.531478557336099] tot_loss[loss=2.324, over 5537736.70 frames. , ppl: 10.211795591553555], batch size: 70
+2022-12-10 20:21:34,321 INFO [train.py:421] (5/8) Epoch 3, batch 26000, loss[loss=2.283, over 3850.00 frames. , ppl: 9.80282099506322] tot_loss[loss=2.323, over 5553861.18 frames. , ppl: 10.207725597755326], batch size: 70
+2022-12-10 20:21:34,321 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 20:21:35,067 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.316, over 211138.00 frames. , ppl: 10.132153014018082
+2022-12-10 20:23:08,526 INFO [train.py:421] (5/8) Epoch 3, batch 26200, loss[loss=2.547, over 1120.00 frames. , ppl: 12.766985421981735] tot_loss[loss=2.324, over 5529339.06 frames. , ppl: 10.213124544667526], batch size: 70
+2022-12-10 20:24:45,539 INFO [train.py:421] (5/8) Epoch 3, batch 26400, loss[loss=2.418, over 1260.00 frames. , ppl: 11.220830168286632] tot_loss[loss=2.325, over 5508860.11 frames. , ppl: 10.224525066249683], batch size: 70
+2022-12-10 20:26:27,607 INFO [train.py:421] (5/8) Epoch 3, batch 26600, loss[loss=2.345, over 2520.00 frames. , ppl: 10.435999785558034] tot_loss[loss=2.325, over 5509359.65 frames. , ppl: 10.222046810809323], batch size: 70
+2022-12-10 20:28:07,421 INFO [train.py:421] (5/8) Epoch 3, batch 26800, loss[loss=2.586, over 1050.00 frames. , ppl: 13.270803627743838] tot_loss[loss=2.325, over 5487546.97 frames. , ppl: 10.225243539731016], batch size: 70
+2022-12-10 20:29:49,943 INFO [train.py:421] (5/8) Epoch 3, batch 27000, loss[loss=2.448, over 1750.00 frames. , ppl: 11.56657438074413] tot_loss[loss=2.324, over 5538650.18 frames. , ppl: 10.213458027385405], batch size: 70
+2022-12-10 20:29:49,944 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 20:29:50,704 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.313, over 211138.00 frames. , ppl: 10.101093802370023
+2022-12-10 20:31:29,521 INFO [train.py:421] (5/8) Epoch 3, batch 27200, loss[loss=2.347, over 4620.00 frames. , ppl: 10.451565409176153] tot_loss[loss=2.325, over 5485937.60 frames. , ppl: 10.231338864365831], batch size: 70
+2022-12-10 20:33:11,514 INFO [train.py:421] (5/8) Epoch 3, batch 27400, loss[loss=2.41, over 1330.00 frames. , ppl: 11.138377034425996] tot_loss[loss=2.324, over 5559901.75 frames. , ppl: 10.214221562039812], batch size: 70
+2022-12-10 20:34:50,153 INFO [train.py:421] (5/8) Epoch 3, batch 27600, loss[loss=2.212, over 6230.00 frames. , ppl: 9.13743698065769] tot_loss[loss=2.323, over 5555251.07 frames. , ppl: 10.210417926094857], batch size: 70
+2022-12-10 20:36:32,924 INFO [train.py:421] (5/8) Epoch 3, batch 27800, loss[loss=2.765, over 840.00 frames. , ppl: 15.87882031140437] tot_loss[loss=2.323, over 5560363.29 frames. , ppl: 10.207841099078523], batch size: 70
+2022-12-10 20:38:09,408 INFO [train.py:421] (5/8) Epoch 3, batch 28000, loss[loss=2.46, over 1400.00 frames. , ppl: 11.707042737372708] tot_loss[loss=2.323, over 5566845.85 frames. , ppl: 10.204779413189998], batch size: 70
+2022-12-10 20:38:09,408 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 20:38:10,154 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.313, over 211138.00 frames. , ppl: 10.106116881121231
+2022-12-10 20:39:49,677 INFO [train.py:421] (5/8) Epoch 3, batch 28200, loss[loss=2.333, over 2870.00 frames. , ppl: 10.31192282129125] tot_loss[loss=2.324, over 5544225.63 frames. , ppl: 10.21406072596949], batch size: 70
+2022-12-10 20:41:31,344 INFO [train.py:421] (5/8) Epoch 3, batch 28400, loss[loss=2.25, over 11340.00 frames. , ppl: 9.488624830326495] tot_loss[loss=2.323, over 5582543.85 frames. , ppl: 10.20182012448681], batch size: 70
+2022-12-10 20:43:09,535 INFO [train.py:421] (5/8) Epoch 3, batch 28600, loss[loss=2.438, over 1120.00 frames. , ppl: 11.453722877994815] tot_loss[loss=2.324, over 5532007.11 frames. , ppl: 10.215257541571951], batch size: 70
+2022-12-10 20:44:46,042 INFO [train.py:421] (5/8) Epoch 3, batch 28800, loss[loss=2.481, over 1610.00 frames. , ppl: 11.954936709349141] tot_loss[loss=2.324, over 5528879.92 frames. , ppl: 10.21375486558223], batch size: 70
+2022-12-10 20:46:28,959 INFO [train.py:421] (5/8) Epoch 3, batch 29000, loss[loss=2.645, over 910.00 frames. , ppl: 14.086433356297157] tot_loss[loss=2.324, over 5517232.26 frames. , ppl: 10.218520049302915], batch size: 70
+2022-12-10 20:46:28,959 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 20:46:29,722 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.313, over 211138.00 frames. , ppl: 10.10355792431578
+2022-12-10 20:48:07,969 INFO [train.py:421] (5/8) Epoch 3, batch 29200, loss[loss=2.843, over 630.00 frames. , ppl: 17.174059299414807] tot_loss[loss=2.324, over 5541886.54 frames. , ppl: 10.212753659903226], batch size: 70
+2022-12-10 20:49:48,633 INFO [train.py:421] (5/8) Epoch 3, batch 29400, loss[loss=2.236, over 10570.00 frames. , ppl: 9.360304264298417] tot_loss[loss=2.325, over 5501909.82 frames. , ppl: 10.223196673352977], batch size: 70
+2022-12-10 20:51:26,256 INFO [train.py:421] (5/8) Epoch 3, batch 29600, loss[loss=2.333, over 2800.00 frames. , ppl: 10.30925994710125] tot_loss[loss=2.326, over 5463544.15 frames. , ppl: 10.238945333041137], batch size: 70
+2022-12-10 20:53:01,786 INFO [train.py:421] (5/8) Epoch 3, batch 29800, loss[loss=2.297, over 3290.00 frames. , ppl: 9.944381137828385] tot_loss[loss=2.325, over 5488032.73 frames. , ppl: 10.226018662988816], batch size: 70
+2022-12-10 20:54:41,555 INFO [train.py:421] (5/8) Epoch 3, batch 30000, loss[loss=2.35, over 3430.00 frames. , ppl: 10.487735833329335] tot_loss[loss=2.325, over 5482362.50 frames. , ppl: 10.2302730004547], batch size: 70
+2022-12-10 20:54:41,555 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 20:54:42,324 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.313, over 211138.00 frames. , ppl: 10.100455441623417
+2022-12-10 20:56:19,500 INFO [train.py:421] (5/8) Epoch 3, batch 30200, loss[loss=2.202, over 4970.00 frames. , ppl: 9.039808157978015] tot_loss[loss=2.325, over 5461693.49 frames. , ppl: 10.227516324744938], batch size: 70
+2022-12-10 20:58:04,785 INFO [train.py:421] (5/8) Epoch 3, batch 30400, loss[loss=2.204, over 5740.00 frames. , ppl: 9.063817690589431] tot_loss[loss=2.324, over 5492751.31 frames. , ppl: 10.217930010906974], batch size: 70
+2022-12-10 20:59:43,385 INFO [train.py:421] (5/8) Epoch 3, batch 30600, loss[loss=2.22, over 5740.00 frames. , ppl: 9.206923279579998] tot_loss[loss=2.324, over 5495216.21 frames. , ppl: 10.212803081855297], batch size: 70
+2022-12-10 21:01:20,859 INFO [train.py:421] (5/8) Epoch 3, batch 30800, loss[loss=2.499, over 910.00 frames. , ppl: 12.174733989245935] tot_loss[loss=2.324, over 5478180.22 frames. , ppl: 10.219942895308334], batch size: 70
+2022-12-10 21:02:59,411 INFO [train.py:421] (5/8) Epoch 3, batch 31000, loss[loss=2.477, over 1260.00 frames. , ppl: 11.901432843824278] tot_loss[loss=2.325, over 5462570.36 frames. , ppl: 10.224513001761657], batch size: 70
+2022-12-10 21:02:59,412 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 21:03:00,157 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.312, over 211138.00 frames. , ppl: 10.091295598195819
+2022-12-10 21:04:37,247 INFO [train.py:421] (5/8) Epoch 3, batch 31200, loss[loss=2.438, over 1890.00 frames. , ppl: 11.451678529560546] tot_loss[loss=2.325, over 5449020.25 frames. , ppl: 10.22690962823716], batch size: 70
+2022-12-10 21:06:17,344 INFO [train.py:421] (5/8) Epoch 3, batch 31400, loss[loss=2.375, over 1400.00 frames. , ppl: 10.755181736466463] tot_loss[loss=2.326, over 5412235.57 frames. , ppl: 10.23884254562602], batch size: 70
+2022-12-10 21:07:59,119 INFO [train.py:421] (5/8) Epoch 3, batch 31600, loss[loss=2.325, over 3360.00 frames. , ppl: 10.222943077437682] tot_loss[loss=2.326, over 5416782.41 frames. , ppl: 10.236706607383955], batch size: 70
+2022-12-10 21:09:38,631 INFO [train.py:421] (5/8) Epoch 3, batch 31800, loss[loss=2.322, over 3010.00 frames. , ppl: 10.193830628930248] tot_loss[loss=2.327, over 5375914.97 frames. , ppl: 10.244705989455097], batch size: 70
+2022-12-10 21:11:16,581 INFO [train.py:421] (5/8) Epoch 3, batch 32000, loss[loss=2.562, over 1120.00 frames. , ppl: 12.967340811923902] tot_loss[loss=2.328, over 5348572.57 frames. , ppl: 10.254108577328191], batch size: 70
+2022-12-10 21:11:16,581 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 21:11:17,340 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.313, over 211138.00 frames. , ppl: 10.101464578420366
+2022-12-10 21:12:58,748 INFO [train.py:421] (5/8) Epoch 3, batch 32200, loss[loss=2.225, over 3850.00 frames. , ppl: 9.253154205948658] tot_loss[loss=2.327, over 5363153.08 frames. , ppl: 10.248789380109066], batch size: 70
+2022-12-10 21:14:34,252 INFO [train.py:421] (5/8) Epoch 3, batch 32400, loss[loss=2.292, over 3780.00 frames. , ppl: 9.890438612259784] tot_loss[loss=2.328, over 5347644.56 frames. , ppl: 10.258135198313765], batch size: 70
+2022-12-10 21:16:14,641 INFO [train.py:421] (5/8) Epoch 3, batch 32600, loss[loss=2.278, over 4480.00 frames. , ppl: 9.752543895772442] tot_loss[loss=2.328, over 5377062.65 frames. , ppl: 10.26025218823], batch size: 70
+2022-12-10 21:17:51,484 INFO [train.py:421] (5/8) Epoch 3, batch 32800, loss[loss=2.22, over 6580.00 frames. , ppl: 9.20571526450314] tot_loss[loss=2.328, over 5375438.83 frames. , ppl: 10.257510860285866], batch size: 70
+2022-12-10 21:19:29,969 INFO [train.py:421] (5/8) Epoch 3, batch 33000, loss[loss=2.354, over 2310.00 frames. , ppl: 10.532827807325386] tot_loss[loss=2.329, over 5358604.74 frames. , ppl: 10.263370941137493], batch size: 70
+2022-12-10 21:19:29,970 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 21:19:30,729 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.314, over 211138.00 frames. , ppl: 10.111637819514526
+2022-12-10 21:21:05,638 INFO [train.py:421] (5/8) Epoch 3, batch 33200, loss[loss=2.338, over 3920.00 frames. , ppl: 10.357204433586988] tot_loss[loss=2.329, over 5359175.05 frames. , ppl: 10.264935381343397], batch size: 70
+2022-12-10 21:22:50,813 INFO [train.py:421] (5/8) Epoch 3, batch 33400, loss[loss=2.306, over 4200.00 frames. , ppl: 10.034032929884512] tot_loss[loss=2.327, over 5400268.07 frames. , ppl: 10.249115238379497], batch size: 70
+2022-12-10 21:24:31,309 INFO [train.py:421] (5/8) Epoch 3, batch 33600, loss[loss=2.217, over 3150.00 frames. , ppl: 9.180058481526927] tot_loss[loss=2.324, over 5481198.62 frames. , ppl: 10.216602245953265], batch size: 70
+2022-12-10 21:26:13,403 INFO [train.py:421] (5/8) Epoch 3, batch 33800, loss[loss=2.59, over 840.00 frames. , ppl: 13.332582253455024] tot_loss[loss=2.324, over 5454454.88 frames. , ppl: 10.220334814469819], batch size: 70
+2022-12-10 21:27:54,537 INFO [train.py:421] (5/8) Epoch 3, batch 34000, loss[loss=2.408, over 3150.00 frames. , ppl: 11.11243099471196] tot_loss[loss=2.324, over 5471139.80 frames. , ppl: 10.216764404369131], batch size: 70
+2022-12-10 21:27:54,537 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 21:27:55,295 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.31, over 211138.00 frames. , ppl: 10.071958753555176
+2022-12-10 21:29:32,476 INFO [train.py:421] (5/8) Epoch 3, batch 34200, loss[loss=2.428, over 2030.00 frames. , ppl: 11.339649612760429] tot_loss[loss=2.325, over 5446124.79 frames. , ppl: 10.224898489460891], batch size: 70
+2022-12-10 21:31:11,682 INFO [train.py:421] (5/8) Epoch 3, batch 34400, loss[loss=2.255, over 11597.00 frames. , ppl: 9.5368458277211] tot_loss[loss=2.324, over 5469548.84 frames. , ppl: 10.220492899257193], batch size: 70
+2022-12-10 21:32:50,594 INFO [train.py:421] (5/8) Epoch 3, batch 34600, loss[loss=2.575, over 980.00 frames. , ppl: 13.125083513577518] tot_loss[loss=2.324, over 5477634.20 frames. , ppl: 10.216949167717324], batch size: 70
+2022-12-10 21:34:29,955 INFO [train.py:421] (5/8) Epoch 3, batch 34800, loss[loss=2.262, over 8750.00 frames. , ppl: 9.597918051868454] tot_loss[loss=2.323, over 5520655.77 frames. , ppl: 10.203973477407585], batch size: 70
+2022-12-10 21:36:14,332 INFO [train.py:421] (5/8) Epoch 3, batch 35000, loss[loss=2.22, over 5390.00 frames. , ppl: 9.206271293750799] tot_loss[loss=2.321, over 5556245.59 frames. , ppl: 10.188876975455788], batch size: 70
+2022-12-10 21:36:14,332 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 21:36:15,077 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.311, over 211138.00 frames. , ppl: 10.083123013689438
+2022-12-10 21:37:54,319 INFO [train.py:421] (5/8) Epoch 3, batch 35200, loss[loss=2.398, over 2590.00 frames. , ppl: 10.999595924050519] tot_loss[loss=2.322, over 5533217.57 frames. , ppl: 10.191655003611988], batch size: 70
+2022-12-10 21:39:36,771 INFO [train.py:421] (5/8) Epoch 3, batch 35400, loss[loss=2.471, over 2100.00 frames. , ppl: 11.829802432561024] tot_loss[loss=2.321, over 5562119.72 frames. , ppl: 10.186881787919429], batch size: 70
+2022-12-10 21:41:19,923 INFO [train.py:421] (5/8) Epoch 3, batch 35600, loss[loss=2.344, over 3080.00 frames. , ppl: 10.422796817381762] tot_loss[loss=2.323, over 5540196.03 frames. , ppl: 10.201586734508862], batch size: 70
+2022-12-10 21:42:54,211 INFO [train.py:421] (5/8) Epoch 3, batch 35800, loss[loss=2.463, over 2450.00 frames. , ppl: 11.736051367280675] tot_loss[loss=2.323, over 5554871.93 frames. , ppl: 10.204208072125546], batch size: 70
+2022-12-10 21:44:36,435 INFO [train.py:421] (5/8) Epoch 3, batch 36000, loss[loss=2.484, over 1470.00 frames. , ppl: 11.984800913994542] tot_loss[loss=2.323, over 5545938.61 frames. , ppl: 10.210729238844516], batch size: 70
+2022-12-10 21:44:36,435 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 21:44:37,181 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.311, over 211138.00 frames. , ppl: 10.086282871808761
+2022-12-10 21:46:17,616 INFO [train.py:421] (5/8) Epoch 3, batch 36200, loss[loss=2.466, over 1050.00 frames. , ppl: 11.778795468655257] tot_loss[loss=2.324, over 5539110.80 frames. , ppl: 10.212501842205512], batch size: 70
+2022-12-10 21:47:56,635 INFO [train.py:421] (5/8) Epoch 3, batch 36400, loss[loss=2.235, over 4130.00 frames. , ppl: 9.349273094721267] tot_loss[loss=2.323, over 5550913.12 frames. , ppl: 10.211331973540675], batch size: 70
+2022-12-10 21:49:39,009 INFO [train.py:421] (5/8) Epoch 3, batch 36600, loss[loss=2.339, over 2170.00 frames. , ppl: 10.367514122693] tot_loss[loss=2.324, over 5528291.62 frames. , ppl: 10.218161121085965], batch size: 70
+2022-12-10 21:51:17,824 INFO [train.py:421] (5/8) Epoch 3, batch 36800, loss[loss=2.502, over 1470.00 frames. , ppl: 12.212250403212622] tot_loss[loss=2.324, over 5524287.96 frames. , ppl: 10.216604357715928], batch size: 70
+2022-12-10 21:53:00,121 INFO [train.py:421] (5/8) Epoch 3, batch 37000, loss[loss=2.612, over 700.00 frames. , ppl: 13.631338963713121] tot_loss[loss=2.325, over 5479938.40 frames. , ppl: 10.226813241947669], batch size: 70
+2022-12-10 21:53:00,122 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 21:53:00,870 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.312, over 211138.00 frames. , ppl: 10.091421060289873
+2022-12-10 21:54:40,044 INFO [train.py:421] (5/8) Epoch 3, batch 37200, loss[loss=3.01, over 560.00 frames. , ppl: 20.28106729254642] tot_loss[loss=2.326, over 5456214.44 frames. , ppl: 10.23349685385196], batch size: 70
+2022-12-10 21:56:18,567 INFO [train.py:421] (5/8) Epoch 3, batch 37400, loss[loss=2.321, over 5880.00 frames. , ppl: 10.182689534406997] tot_loss[loss=2.326, over 5462032.66 frames. , ppl: 10.235735769977012], batch size: 70
+2022-12-10 21:57:58,919 INFO [train.py:421] (5/8) Epoch 3, batch 37600, loss[loss=2.384, over 2310.00 frames. , ppl: 10.845143820428811] tot_loss[loss=2.326, over 5443949.24 frames. , ppl: 10.239877448263128], batch size: 70
+2022-12-10 21:59:41,112 INFO [train.py:421] (5/8) Epoch 3, batch 37800, loss[loss=3.231, over 490.00 frames. , ppl: 25.30453648634118] tot_loss[loss=2.326, over 5474475.86 frames. , ppl: 10.235786615789726], batch size: 70
+2022-12-10 22:01:19,423 INFO [train.py:421] (5/8) Epoch 3, batch 38000, loss[loss=2.42, over 1820.00 frames. , ppl: 11.245487612934452] tot_loss[loss=2.326, over 5428747.91 frames. , ppl: 10.238311573292796], batch size: 70
+2022-12-10 22:01:19,424 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 22:01:20,188 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.311, over 211138.00 frames. , ppl: 10.088428318438387
+2022-12-10 22:02:59,176 INFO [train.py:421] (5/8) Epoch 3, batch 38200, loss[loss=2.357, over 2100.00 frames. , ppl: 10.559708417551617] tot_loss[loss=2.325, over 5436815.72 frames. , ppl: 10.231615405540914], batch size: 70
+2022-12-10 22:04:43,418 INFO [train.py:421] (5/8) Epoch 3, batch 38400, loss[loss=2.86, over 700.00 frames. , ppl: 17.456679891996945] tot_loss[loss=2.323, over 5496175.57 frames. , ppl: 10.207737343272315], batch size: 70
+2022-12-10 22:06:21,541 INFO [train.py:421] (5/8) Epoch 3, batch 38600, loss[loss=2.265, over 3990.00 frames. , ppl: 9.633797141243216] tot_loss[loss=2.324, over 5474486.17 frames. , ppl: 10.214151017296642], batch size: 70
+2022-12-10 22:08:03,080 INFO [train.py:421] (5/8) Epoch 3, batch 38800, loss[loss=2.315, over 4130.00 frames. , ppl: 10.127796721561523] tot_loss[loss=2.322, over 5529329.62 frames. , ppl: 10.198563610063843], batch size: 70
+2022-12-10 22:09:44,350 INFO [train.py:421] (5/8) Epoch 3, batch 39000, loss[loss=2.271, over 4620.00 frames. , ppl: 9.692910588029653] tot_loss[loss=2.321, over 5576532.24 frames. , ppl: 10.185386442913943], batch size: 70
+2022-12-10 22:09:44,351 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 22:09:45,098 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.311, over 211138.00 frames. , ppl: 10.087220422135697
+2022-12-10 22:11:24,505 INFO [train.py:421] (5/8) Epoch 3, batch 39200, loss[loss=2.286, over 3570.00 frames. , ppl: 9.837686986261996] tot_loss[loss=2.319, over 5634372.97 frames. , ppl: 10.16733553352927], batch size: 70
+2022-12-10 22:13:02,104 INFO [train.py:421] (5/8) Epoch 3, batch 39400, loss[loss=2.52, over 1330.00 frames. , ppl: 12.424300071649961] tot_loss[loss=2.321, over 5558760.87 frames. , ppl: 10.184785523460526], batch size: 70
+2022-12-10 22:14:41,112 INFO [train.py:421] (5/8) Epoch 3, batch 39600, loss[loss=2.247, over 2380.00 frames. , ppl: 9.455752329651512] tot_loss[loss=2.321, over 5546970.80 frames. , ppl: 10.182649896489036], batch size: 70
+2022-12-10 22:16:21,149 INFO [train.py:421] (5/8) Epoch 3, batch 39800, loss[loss=2.286, over 2450.00 frames. , ppl: 9.832499358455973] tot_loss[loss=2.322, over 5510207.29 frames. , ppl: 10.194156164146632], batch size: 70
+2022-12-10 22:17:59,552 INFO [train.py:421] (5/8) Epoch 3, batch 40000, loss[loss=2.411, over 1540.00 frames. , ppl: 11.141847927596771] tot_loss[loss=2.321, over 5556482.10 frames. , ppl: 10.18099706556122], batch size: 70
+2022-12-10 22:17:59,552 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 22:18:00,311 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.311, over 211138.00 frames. , ppl: 10.0863605000554
+2022-12-10 22:19:42,201 INFO [train.py:421] (5/8) Epoch 3, batch 40200, loss[loss=2.227, over 6090.00 frames. , ppl: 9.274455431315825] tot_loss[loss=2.32, over 5574665.37 frames. , ppl: 10.171623487739962], batch size: 70
+2022-12-10 22:21:21,734 INFO [train.py:421] (5/8) Epoch 3, batch 40400, loss[loss=2.263, over 4760.00 frames. , ppl: 9.607281846922707] tot_loss[loss=2.32, over 5559186.56 frames. , ppl: 10.179238064794749], batch size: 70
+2022-12-10 22:23:02,835 INFO [train.py:421] (5/8) Epoch 3, batch 40600, loss[loss=2.238, over 5390.00 frames. , ppl: 9.371007984306308] tot_loss[loss=2.32, over 5567607.70 frames. , ppl: 10.172340356135466], batch size: 70
+2022-12-10 22:24:42,985 INFO [train.py:421] (5/8) Epoch 3, batch 40800, loss[loss=2.188, over 3360.00 frames. , ppl: 8.919216916733351] tot_loss[loss=2.321, over 5526731.72 frames. , ppl: 10.182991150208293], batch size: 70
+2022-12-10 22:26:20,554 INFO [train.py:421] (5/8) Epoch 3, batch 41000, loss[loss=2.674, over 840.00 frames. , ppl: 14.499811160530873] tot_loss[loss=2.322, over 5478863.80 frames. , ppl: 10.198197881523898], batch size: 70
+2022-12-10 22:26:20,554 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 22:26:21,301 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.312, over 211138.00 frames. , ppl: 10.09838814732227
+2022-12-10 22:28:02,987 INFO [train.py:421] (5/8) Epoch 3, batch 41200, loss[loss=2.483, over 1680.00 frames. , ppl: 11.973386086892914] tot_loss[loss=2.324, over 5438397.66 frames. , ppl: 10.215362482169962], batch size: 70
+2022-12-10 22:29:45,179 INFO [train.py:421] (5/8) Epoch 3, batch 41400, loss[loss=2.372, over 2380.00 frames. , ppl: 10.7224435455298] tot_loss[loss=2.323, over 5464019.17 frames. , ppl: 10.203260487737136], batch size: 70
+2022-12-10 22:31:24,818 INFO [train.py:421] (5/8) Epoch 3, batch 41600, loss[loss=2.463, over 1540.00 frames. , ppl: 11.74030269635327] tot_loss[loss=2.323, over 5462075.16 frames. , ppl: 10.208359209337383], batch size: 70
+2022-12-10 22:32:59,977 INFO [train.py:421] (5/8) Epoch 3, batch 41800, loss[loss=2.353, over 2660.00 frames. , ppl: 10.521941657250377] tot_loss[loss=2.323, over 5459331.70 frames. , ppl: 10.206369793369536], batch size: 70
+2022-12-10 22:34:40,203 INFO [train.py:421] (5/8) Epoch 3, batch 42000, loss[loss=2.506, over 1190.00 frames. , ppl: 12.255955892971313] tot_loss[loss=2.324, over 5445242.68 frames. , ppl: 10.21639262672938], batch size: 70
+2022-12-10 22:34:40,204 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 22:34:40,954 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.31, over 211138.00 frames. , ppl: 10.07229715383208
+2022-12-10 22:36:24,017 INFO [train.py:421] (5/8) Epoch 3, batch 42200, loss[loss=2.311, over 3570.00 frames. , ppl: 10.082158270051945] tot_loss[loss=2.323, over 5465489.46 frames. , ppl: 10.204038732977802], batch size: 70
+2022-12-10 22:38:04,419 INFO [train.py:421] (5/8) Epoch 3, batch 42400, loss[loss=2.387, over 3500.00 frames. , ppl: 10.88418508368736] tot_loss[loss=2.324, over 5427446.65 frames. , ppl: 10.219410632674984], batch size: 70
+2022-12-10 22:39:45,497 INFO [train.py:421] (5/8) Epoch 3, batch 42600, loss[loss=2.394, over 1400.00 frames. , ppl: 10.954427605805543] tot_loss[loss=2.324, over 5419656.99 frames. , ppl: 10.218657341906823], batch size: 70
+2022-12-10 22:41:24,895 INFO [train.py:421] (5/8) Epoch 3, batch 42800, loss[loss=2.192, over 2450.00 frames. , ppl: 8.955956464585167] tot_loss[loss=2.325, over 5382560.22 frames. , ppl: 10.229912803999433], batch size: 70
+2022-12-10 22:43:08,206 INFO [train.py:421] (5/8) Epoch 3, batch 43000, loss[loss=3.549, over 420.00 frames. , ppl: 34.76221730135812] tot_loss[loss=2.325, over 5388334.86 frames. , ppl: 10.229903678302094], batch size: 70
+2022-12-10 22:43:08,206 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 22:43:08,978 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.31, over 211138.00 frames. , ppl: 10.073643413765264
+2022-12-10 22:44:46,749 INFO [train.py:421] (5/8) Epoch 3, batch 43200, loss[loss=2.363, over 2030.00 frames. , ppl: 10.626603549935343] tot_loss[loss=2.325, over 5390379.22 frames. , ppl: 10.224855807766273], batch size: 70
+2022-12-10 22:46:25,132 INFO [train.py:421] (5/8) Epoch 3, batch 43400, loss[loss=2.399, over 2520.00 frames. , ppl: 11.006772247122043] tot_loss[loss=2.325, over 5398014.40 frames. , ppl: 10.226158649844173], batch size: 70
+2022-12-10 22:48:04,841 INFO [train.py:421] (5/8) Epoch 3, batch 43600, loss[loss=2.831, over 700.00 frames. , ppl: 16.95412134469962] tot_loss[loss=2.326, over 5401639.03 frames. , ppl: 10.23195205165169], batch size: 70
+2022-12-10 22:49:41,095 INFO [train.py:421] (5/8) Epoch 3, batch 43800, loss[loss=2.347, over 2590.00 frames. , ppl: 10.45275887529382] tot_loss[loss=2.326, over 5367992.40 frames. , ppl: 10.236242378984281], batch size: 70
+2022-12-10 22:51:19,560 INFO [train.py:421] (5/8) Epoch 3, batch 44000, loss[loss=2.719, over 700.00 frames. , ppl: 15.165261641464463] tot_loss[loss=2.326, over 5375894.47 frames. , ppl: 10.237799492183925], batch size: 70
+2022-12-10 22:51:19,561 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 22:51:20,323 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.312, over 211138.00 frames. , ppl: 10.093368910010755
+2022-12-10 22:53:02,987 INFO [train.py:421] (5/8) Epoch 3, batch 44200, loss[loss=2.264, over 2310.00 frames. , ppl: 9.617128071427006] tot_loss[loss=2.326, over 5360016.90 frames. , ppl: 10.241703932594046], batch size: 70
+2022-12-10 22:54:43,985 INFO [train.py:421] (5/8) Epoch 3, batch 44400, loss[loss=2.516, over 1120.00 frames. , ppl: 12.373075313213297] tot_loss[loss=2.326, over 5353686.24 frames. , ppl: 10.235979462545256], batch size: 70
+2022-12-10 22:56:24,870 INFO [train.py:421] (5/8) Epoch 3, batch 44600, loss[loss=2.348, over 1330.00 frames. , ppl: 10.462872332286015] tot_loss[loss=2.325, over 5375727.52 frames. , ppl: 10.223521150824281], batch size: 70
+2022-12-10 22:58:04,158 INFO [train.py:421] (5/8) Epoch 3, batch 44800, loss[loss=2.401, over 1470.00 frames. , ppl: 11.03226797984609] tot_loss[loss=2.325, over 5391836.75 frames. , ppl: 10.22166753881197], batch size: 70
+2022-12-10 22:59:45,431 INFO [train.py:421] (5/8) Epoch 3, batch 45000, loss[loss=2.207, over 5810.00 frames. , ppl: 9.088006821912415] tot_loss[loss=2.324, over 5386927.01 frames. , ppl: 10.21797976821188], batch size: 70
+2022-12-10 22:59:45,431 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 22:59:46,181 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.313, over 211138.00 frames. , ppl: 10.10122686156436
+2022-12-10 23:01:30,013 INFO [train.py:421] (5/8) Epoch 3, batch 45200, loss[loss=2.382, over 1750.00 frames. , ppl: 10.829235222707757] tot_loss[loss=2.324, over 5382287.17 frames. , ppl: 10.214364138837578], batch size: 70
+2022-12-10 23:03:11,270 INFO [train.py:421] (5/8) Epoch 3, batch 45400, loss[loss=2.388, over 1470.00 frames. , ppl: 10.887204380016524] tot_loss[loss=2.323, over 5391720.09 frames. , ppl: 10.211202985942968], batch size: 70
+2022-12-10 23:04:51,283 INFO [train.py:421] (5/8) Epoch 3, batch 45600, loss[loss=2.507, over 1120.00 frames. , ppl: 12.265042024416742] tot_loss[loss=2.324, over 5402399.34 frames. , ppl: 10.212690945048307], batch size: 70
+2022-12-10 23:06:30,767 INFO [train.py:421] (5/8) Epoch 3, batch 45800, loss[loss=2.6, over 1050.00 frames. , ppl: 13.46957458796199] tot_loss[loss=2.324, over 5423362.16 frames. , ppl: 10.213252818437763], batch size: 70
+2022-12-10 23:08:09,340 INFO [train.py:421] (5/8) Epoch 3, batch 46000, loss[loss=2.221, over 9380.00 frames. , ppl: 9.21424831706184] tot_loss[loss=2.324, over 5446642.12 frames. , ppl: 10.215196541863063], batch size: 70
+2022-12-10 23:08:09,340 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 23:08:10,128 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.31, over 211138.00 frames. , ppl: 10.07448734063421
+2022-12-10 23:09:50,389 INFO [train.py:421] (5/8) Epoch 3, batch 46200, loss[loss=3.59, over 420.00 frames. , ppl: 36.247141184790486] tot_loss[loss=2.323, over 5481186.47 frames. , ppl: 10.205035471633614], batch size: 70
+2022-12-10 23:11:28,888 INFO [train.py:421] (5/8) Epoch 3, batch 46400, loss[loss=2.469, over 1050.00 frames. , ppl: 11.806744605728273] tot_loss[loss=2.323, over 5452450.12 frames. , ppl: 10.205675259382017], batch size: 70
+2022-12-10 23:13:11,853 INFO [train.py:421] (5/8) Epoch 3, batch 46600, loss[loss=2.468, over 1890.00 frames. , ppl: 11.79970058203691] tot_loss[loss=2.322, over 5489421.49 frames. , ppl: 10.198833588878891], batch size: 70
+2022-12-10 23:14:52,146 INFO [train.py:421] (5/8) Epoch 3, batch 46800, loss[loss=2.381, over 1750.00 frames. , ppl: 10.811151254400807] tot_loss[loss=2.321, over 5486504.01 frames. , ppl: 10.190817314255474], batch size: 70
+2022-12-10 23:16:28,906 INFO [train.py:421] (5/8) Epoch 3, batch 47000, loss[loss=2.304, over 4060.00 frames. , ppl: 10.010219752032967] tot_loss[loss=2.323, over 5422271.18 frames. , ppl: 10.205947206578413], batch size: 70
+2022-12-10 23:16:28,907 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 23:16:29,653 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.307, over 211138.00 frames. , ppl: 10.048629133994561
+2022-12-10 23:18:11,342 INFO [train.py:421] (5/8) Epoch 3, batch 47200, loss[loss=2.589, over 910.00 frames. , ppl: 13.312229753124955] tot_loss[loss=2.322, over 5445258.10 frames. , ppl: 10.196437306951267], batch size: 70
+2022-12-10 23:19:51,597 INFO [train.py:421] (5/8) Epoch 3, batch 47400, loss[loss=2.335, over 2870.00 frames. , ppl: 10.327272925339518] tot_loss[loss=2.322, over 5449458.26 frames. , ppl: 10.196453132383898], batch size: 70
+2022-12-10 23:21:30,718 INFO [train.py:421] (5/8) Epoch 3, batch 47600, loss[loss=2.694, over 770.00 frames. , ppl: 14.784116858190176] tot_loss[loss=2.322, over 5429245.37 frames. , ppl: 10.198099331346404], batch size: 70
+2022-12-10 23:23:11,827 INFO [train.py:421] (5/8) Epoch 3, batch 47800, loss[loss=2.228, over 5740.00 frames. , ppl: 9.281289229683853] tot_loss[loss=2.322, over 5440865.96 frames. , ppl: 10.194565384958748], batch size: 70
+2022-12-10 23:24:53,396 INFO [train.py:421] (5/8) Epoch 3, batch 48000, loss[loss=2.488, over 1750.00 frames. , ppl: 12.035414545372564] tot_loss[loss=2.322, over 5418277.81 frames. , ppl: 10.196966510499294], batch size: 70
+2022-12-10 23:24:53,396 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 23:24:54,156 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.309, over 211138.00 frames. , ppl: 10.0626251790151
+2022-12-10 23:26:31,206 INFO [train.py:421] (5/8) Epoch 3, batch 48200, loss[loss=3.238, over 490.00 frames. , ppl: 25.48954315221427] tot_loss[loss=2.322, over 5404525.44 frames. , ppl: 10.195024964953669], batch size: 70
+2022-12-10 23:28:10,222 INFO [train.py:421] (5/8) Epoch 3, batch 48400, loss[loss=2.558, over 770.00 frames. , ppl: 12.904478009161068] tot_loss[loss=2.322, over 5401756.11 frames. , ppl: 10.195244251343302], batch size: 70
+2022-12-10 23:29:50,973 INFO [train.py:421] (5/8) Epoch 3, batch 48600, loss[loss=2.391, over 2660.00 frames. , ppl: 10.929790757440765] tot_loss[loss=2.323, over 5382843.52 frames. , ppl: 10.203363815135482], batch size: 70
+2022-12-10 23:31:26,390 INFO [train.py:421] (5/8) Epoch 3, batch 48800, loss[loss=2.374, over 4130.00 frames. , ppl: 10.744758800605336] tot_loss[loss=2.323, over 5398995.03 frames. , ppl: 10.202891599182362], batch size: 70
+2022-12-10 23:33:09,328 INFO [train.py:421] (5/8) Epoch 3, batch 49000, loss[loss=2.402, over 1960.00 frames. , ppl: 11.04823302361989] tot_loss[loss=2.322, over 5411514.88 frames. , ppl: 10.198211121884947], batch size: 70
+2022-12-10 23:33:09,329 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 23:33:10,087 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.309, over 211138.00 frames. , ppl: 10.064253163288134
+2022-12-10 23:34:51,170 INFO [train.py:421] (5/8) Epoch 3, batch 49200, loss[loss=2.351, over 2800.00 frames. , ppl: 10.497125134825861] tot_loss[loss=2.321, over 5447279.08 frames. , ppl: 10.190680307637976], batch size: 70
+2022-12-10 23:36:30,854 INFO [train.py:421] (5/8) Epoch 3, batch 49400, loss[loss=2.42, over 1540.00 frames. , ppl: 11.241733157713035] tot_loss[loss=2.322, over 5429251.73 frames. , ppl: 10.198427735093656], batch size: 70
+2022-12-10 23:38:10,928 INFO [train.py:421] (5/8) Epoch 3, batch 49600, loss[loss=2.62, over 1050.00 frames. , ppl: 13.740831356986204] tot_loss[loss=2.322, over 5450637.28 frames. , ppl: 10.195496448225535], batch size: 70
+2022-12-10 23:39:52,726 INFO [train.py:421] (5/8) Epoch 3, batch 49800, loss[loss=2.239, over 1540.00 frames. , ppl: 9.379287776777415] tot_loss[loss=2.321, over 5475158.82 frames. , ppl: 10.185221306796567], batch size: 70
+2022-12-10 23:41:29,040 INFO [train.py:421] (5/8) Epoch 3, batch 50000, loss[loss=2.281, over 5740.00 frames. , ppl: 9.789994811977662] tot_loss[loss=2.322, over 5471055.01 frames. , ppl: 10.191927510490551], batch size: 70
+2022-12-10 23:41:29,040 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 23:41:29,779 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.31, over 211138.00 frames. , ppl: 10.074415768120824
+2022-12-10 23:43:13,634 INFO [train.py:421] (5/8) Epoch 3, batch 50200, loss[loss=2.329, over 2870.00 frames. , ppl: 10.266066906230046] tot_loss[loss=2.321, over 5481471.82 frames. , ppl: 10.187095690414273], batch size: 70
+2022-12-10 23:44:54,564 INFO [train.py:421] (5/8) Epoch 3, batch 50400, loss[loss=2.247, over 2590.00 frames. , ppl: 9.46306271316177] tot_loss[loss=2.32, over 5540280.71 frames. , ppl: 10.173472559371183], batch size: 70
+2022-12-10 23:46:36,439 INFO [train.py:421] (5/8) Epoch 3, batch 50600, loss[loss=2.466, over 1400.00 frames. , ppl: 11.772322853989039] tot_loss[loss=2.321, over 5511983.59 frames. , ppl: 10.183764158775093], batch size: 70
+2022-12-10 23:48:17,653 INFO [train.py:421] (5/8) Epoch 3, batch 50800, loss[loss=2.413, over 2380.00 frames. , ppl: 11.16521594957115] tot_loss[loss=2.32, over 5522462.28 frames. , ppl: 10.174687296086912], batch size: 70
+2022-12-10 23:49:58,957 INFO [train.py:421] (5/8) Epoch 3, batch 51000, loss[loss=2.516, over 770.00 frames. , ppl: 12.37527414879967] tot_loss[loss=2.32, over 5534690.75 frames. , ppl: 10.176687714535428], batch size: 70
+2022-12-10 23:49:58,958 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 23:49:59,716 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.309, over 211138.00 frames. , ppl: 10.069029904707316
+2022-12-10 23:51:40,231 INFO [train.py:421] (5/8) Epoch 3, batch 51200, loss[loss=2.49, over 1120.00 frames. , ppl: 12.057988028467946] tot_loss[loss=2.32, over 5518258.56 frames. , ppl: 10.179047130834194], batch size: 70
+2022-12-10 23:53:19,771 INFO [train.py:421] (5/8) Epoch 3, batch 51400, loss[loss=2.558, over 980.00 frames. , ppl: 12.90540010616846] tot_loss[loss=2.321, over 5505399.53 frames. , ppl: 10.183408588212123], batch size: 70
+2022-12-10 23:55:01,899 INFO [train.py:421] (5/8) Epoch 3, batch 51600, loss[loss=2.301, over 1890.00 frames. , ppl: 9.986032593494858] tot_loss[loss=2.321, over 5499309.60 frames. , ppl: 10.184452926954469], batch size: 70
+2022-12-10 23:56:41,059 INFO [train.py:421] (5/8) Epoch 3, batch 51800, loss[loss=2.386, over 1540.00 frames. , ppl: 10.871670804858848] tot_loss[loss=2.322, over 5468430.62 frames. , ppl: 10.195233451651802], batch size: 70
+2022-12-10 23:58:20,736 INFO [train.py:421] (5/8) Epoch 3, batch 52000, loss[loss=2.509, over 980.00 frames. , ppl: 12.29685282027265] tot_loss[loss=2.322, over 5455310.86 frames. , ppl: 10.195162603074857], batch size: 70
+2022-12-10 23:58:20,736 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-10 23:58:21,496 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.309, over 211138.00 frames. , ppl: 10.064753676155952
+2022-12-11 00:00:01,560 INFO [train.py:421] (5/8) Epoch 3, batch 52200, loss[loss=2.519, over 770.00 frames. , ppl: 12.420302719720564] tot_loss[loss=2.322, over 5477904.47 frames. , ppl: 10.196654873259558], batch size: 70
+2022-12-11 00:01:45,812 INFO [train.py:421] (5/8) Epoch 3, batch 52400, loss[loss=2.544, over 1190.00 frames. , ppl: 12.731048700130643] tot_loss[loss=2.322, over 5486130.03 frames. , ppl: 10.193094818958551], batch size: 70
+2022-12-11 00:03:24,687 INFO [train.py:421] (5/8) Epoch 3, batch 52600, loss[loss=2.315, over 3500.00 frames. , ppl: 10.12632988357022] tot_loss[loss=2.32, over 5523326.35 frames. , ppl: 10.17474748946763], batch size: 70
+2022-12-11 00:05:03,993 INFO [train.py:421] (5/8) Epoch 3, batch 52800, loss[loss=2.502, over 1540.00 frames. , ppl: 12.202580839889643] tot_loss[loss=2.321, over 5482548.42 frames. , ppl: 10.18174996774877], batch size: 70
+2022-12-11 00:06:42,494 INFO [train.py:421] (5/8) Epoch 3, batch 53000, loss[loss=2.331, over 1890.00 frames. , ppl: 10.284035670453845] tot_loss[loss=2.321, over 5468060.62 frames. , ppl: 10.18445003727258], batch size: 70
+2022-12-11 00:06:42,494 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 00:06:43,239 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.311, over 211138.00 frames. , ppl: 10.088516415422001
+2022-12-11 00:08:25,740 INFO [train.py:421] (5/8) Epoch 3, batch 53200, loss[loss=2.198, over 6090.00 frames. , ppl: 9.00681443358676] tot_loss[loss=2.321, over 5475755.47 frames. , ppl: 10.187475084184824], batch size: 70
+2022-12-11 00:10:06,550 INFO [train.py:421] (5/8) Epoch 3, batch 53400, loss[loss=2.473, over 1260.00 frames. , ppl: 11.854169891380437] tot_loss[loss=2.32, over 5518780.80 frames. , ppl: 10.176612568811706], batch size: 70
+2022-12-11 00:11:48,855 INFO [train.py:421] (5/8) Epoch 3, batch 53600, loss[loss=2.418, over 1610.00 frames. , ppl: 11.226819822396159] tot_loss[loss=2.321, over 5508498.71 frames. , ppl: 10.185932275060743], batch size: 70
+2022-12-11 00:13:30,371 INFO [train.py:421] (5/8) Epoch 3, batch 53800, loss[loss=2.487, over 1260.00 frames. , ppl: 12.028459048372966] tot_loss[loss=2.32, over 5535479.22 frames. , ppl: 10.171334966857856], batch size: 70
+2022-12-11 00:15:11,539 INFO [train.py:421] (5/8) Epoch 3, batch 54000, loss[loss=2.387, over 1680.00 frames. , ppl: 10.877134146814457] tot_loss[loss=2.319, over 5554549.08 frames. , ppl: 10.163115905600822], batch size: 70
+2022-12-11 00:15:11,540 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 00:15:12,298 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.306, over 211138.00 frames. , ppl: 10.032199299773053
+2022-12-11 00:16:54,777 INFO [train.py:421] (5/8) Epoch 3, batch 54200, loss[loss=2.405, over 2450.00 frames. , ppl: 11.081444495045597] tot_loss[loss=2.319, over 5526512.35 frames. , ppl: 10.169729087583892], batch size: 70
+2022-12-11 00:18:34,608 INFO [train.py:421] (5/8) Epoch 3, batch 54400, loss[loss=2.407, over 3080.00 frames. , ppl: 11.103033050554206] tot_loss[loss=2.32, over 5519968.14 frames. , ppl: 10.174145092323934], batch size: 70
+2022-12-11 00:20:14,979 INFO [train.py:421] (5/8) Epoch 3, batch 54600, loss[loss=2.346, over 2380.00 frames. , ppl: 10.442504899809588] tot_loss[loss=2.321, over 5480096.80 frames. , ppl: 10.185629533907317], batch size: 70
+2022-12-11 00:21:53,921 INFO [train.py:421] (5/8) Epoch 3, batch 54800, loss[loss=2.768, over 840.00 frames. , ppl: 15.93041745620325] tot_loss[loss=2.322, over 5458163.93 frames. , ppl: 10.196908518191522], batch size: 70
+2022-12-11 00:23:33,753 INFO [train.py:421] (5/8) Epoch 3, batch 55000, loss[loss=2.854, over 630.00 frames. , ppl: 17.350125119786625] tot_loss[loss=2.321, over 5485001.43 frames. , ppl: 10.18714388463349], batch size: 70
+2022-12-11 00:23:33,754 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 00:23:34,513 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.307, over 211138.00 frames. , ppl: 10.042009979542168
+2022-12-11 00:25:14,234 INFO [train.py:421] (5/8) Epoch 3, batch 55200, loss[loss=2.251, over 5040.00 frames. , ppl: 9.500202629405944] tot_loss[loss=2.32, over 5497133.59 frames. , ppl: 10.18037125739553], batch size: 70
+2022-12-11 00:26:51,653 INFO [train.py:421] (5/8) Epoch 3, batch 55400, loss[loss=2.318, over 2800.00 frames. , ppl: 10.153694319263293] tot_loss[loss=2.319, over 5505337.57 frames. , ppl: 10.168076932399764], batch size: 70
+2022-12-11 00:28:28,804 INFO [train.py:421] (5/8) Epoch 3, batch 55600, loss[loss=2.908, over 560.00 frames. , ppl: 18.322732935263446] tot_loss[loss=2.32, over 5512966.42 frames. , ppl: 10.172220116553675], batch size: 70
+2022-12-11 00:30:05,107 INFO [train.py:421] (5/8) Epoch 3, batch 55800, loss[loss=2.281, over 2170.00 frames. , ppl: 9.789411117306086] tot_loss[loss=2.32, over 5522348.59 frames. , ppl: 10.173825856801148], batch size: 70
+2022-12-11 00:31:49,155 INFO [train.py:421] (5/8) Epoch 3, batch 56000, loss[loss=2.304, over 4200.00 frames. , ppl: 10.01089700447328] tot_loss[loss=2.319, over 5551400.52 frames. , ppl: 10.162736741620298], batch size: 70
+2022-12-11 00:31:49,155 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 00:31:49,915 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.306, over 211138.00 frames. , ppl: 10.029381468540643
+2022-12-11 00:33:27,283 INFO [train.py:421] (5/8) Epoch 3, batch 56200, loss[loss=2.399, over 2450.00 frames. , ppl: 11.015639190342378] tot_loss[loss=2.32, over 5534458.19 frames. , ppl: 10.174770499553823], batch size: 70
+2022-12-11 00:35:04,950 INFO [train.py:421] (5/8) Epoch 3, batch 56400, loss[loss=2.459, over 840.00 frames. , ppl: 11.691883191912765] tot_loss[loss=2.32, over 5529702.08 frames. , ppl: 10.180277574489955], batch size: 70
+2022-12-11 00:36:45,127 INFO [train.py:421] (5/8) Epoch 3, batch 56600, loss[loss=2.364, over 1890.00 frames. , ppl: 10.634410873808035] tot_loss[loss=2.321, over 5522487.86 frames. , ppl: 10.183406833273033], batch size: 70
+2022-12-11 00:38:27,987 INFO [train.py:421] (5/8) Epoch 3, batch 56800, loss[loss=3.049, over 560.00 frames. , ppl: 21.08863758672972] tot_loss[loss=2.321, over 5502496.58 frames. , ppl: 10.181434765781937], batch size: 70
+2022-12-11 00:40:12,326 INFO [train.py:421] (5/8) Epoch 3, batch 57000, loss[loss=2.245, over 4060.00 frames. , ppl: 9.441442435522415] tot_loss[loss=2.32, over 5514986.96 frames. , ppl: 10.17976939995911], batch size: 70
+2022-12-11 00:40:12,327 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 00:40:13,085 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.307, over 211138.00 frames. , ppl: 10.04296571105421
+2022-12-11 00:41:53,208 INFO [train.py:421] (5/8) Epoch 3, batch 57200, loss[loss=2.257, over 5110.00 frames. , ppl: 9.55265128824963] tot_loss[loss=2.321, over 5483479.58 frames. , ppl: 10.186236063786442], batch size: 70
+2022-12-11 00:43:33,444 INFO [train.py:421] (5/8) Epoch 3, batch 57400, loss[loss=2.374, over 3850.00 frames. , ppl: 10.741532784539515] tot_loss[loss=2.321, over 5483488.28 frames. , ppl: 10.186269890392712], batch size: 70
+2022-12-11 00:45:12,037 INFO [train.py:421] (5/8) Epoch 3, batch 57600, loss[loss=2.404, over 1960.00 frames. , ppl: 11.063345603966798] tot_loss[loss=2.321, over 5510716.10 frames. , ppl: 10.185410753252071], batch size: 70
+2022-12-11 00:46:53,806 INFO [train.py:421] (5/8) Epoch 3, batch 57800, loss[loss=2.37, over 1750.00 frames. , ppl: 10.702046542565952] tot_loss[loss=2.321, over 5505840.66 frames. , ppl: 10.185426657602871], batch size: 70
+2022-12-11 00:48:33,991 INFO [train.py:421] (5/8) Epoch 3, batch 58000, loss[loss=2.308, over 3710.00 frames. , ppl: 10.054379783735536] tot_loss[loss=2.321, over 5496635.80 frames. , ppl: 10.185439711948932], batch size: 70
+2022-12-11 00:48:33,992 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 00:48:34,750 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.306, over 211138.00 frames. , ppl: 10.037739283472558
+2022-12-11 00:50:13,898 INFO [train.py:421] (5/8) Epoch 3, batch 58200, loss[loss=2.459, over 1470.00 frames. , ppl: 11.69178415991914] tot_loss[loss=2.321, over 5498869.04 frames. , ppl: 10.19064637247861], batch size: 70
+2022-12-11 00:51:51,052 INFO [train.py:421] (5/8) Epoch 3, batch 58400, loss[loss=2.39, over 1120.00 frames. , ppl: 10.915233572027269] tot_loss[loss=2.322, over 5490619.44 frames. , ppl: 10.19494096522232], batch size: 70
+2022-12-11 00:53:28,837 INFO [train.py:421] (5/8) Epoch 3, batch 58600, loss[loss=2.316, over 4550.00 frames. , ppl: 10.137769756428375] tot_loss[loss=2.322, over 5473813.43 frames. , ppl: 10.193019293765085], batch size: 70
+2022-12-11 00:55:08,240 INFO [train.py:421] (5/8) Epoch 3, batch 58800, loss[loss=2.358, over 1610.00 frames. , ppl: 10.567738557400897] tot_loss[loss=2.321, over 5504441.37 frames. , ppl: 10.182832003851948], batch size: 70
+2022-12-11 00:56:50,214 INFO [train.py:421] (5/8) Epoch 3, batch 59000, loss[loss=2.394, over 3500.00 frames. , ppl: 10.952717638208009] tot_loss[loss=2.32, over 5541235.71 frames. , ppl: 10.178705555898661], batch size: 70
+2022-12-11 00:56:50,215 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 00:56:50,962 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.309, over 211138.00 frames. , ppl: 10.063399667509817
+2022-12-11 00:58:29,069 INFO [train.py:421] (5/8) Epoch 3, batch 59200, loss[loss=2.392, over 1470.00 frames. , ppl: 10.930449232476908] tot_loss[loss=2.322, over 5464912.84 frames. , ppl: 10.200688520449503], batch size: 70
+2022-12-11 01:00:05,133 INFO [train.py:421] (5/8) Epoch 3, batch 59400, loss[loss=2.326, over 3290.00 frames. , ppl: 10.236562569765947] tot_loss[loss=2.322, over 5464514.61 frames. , ppl: 10.198253222721384], batch size: 70
+2022-12-11 01:01:44,065 INFO [train.py:421] (5/8) Epoch 3, batch 59600, loss[loss=2.39, over 2170.00 frames. , ppl: 10.91236585584634] tot_loss[loss=2.323, over 5472124.06 frames. , ppl: 10.203312476326468], batch size: 70
+2022-12-11 01:03:24,733 INFO [train.py:421] (5/8) Epoch 3, batch 59800, loss[loss=2.491, over 770.00 frames. , ppl: 12.070014562093156] tot_loss[loss=2.322, over 5478901.25 frames. , ppl: 10.198271151351012], batch size: 70
+2022-12-11 01:05:04,163 INFO [train.py:421] (5/8) Epoch 3, batch 60000, loss[loss=2.317, over 3640.00 frames. , ppl: 10.141157049233946] tot_loss[loss=2.323, over 5450370.23 frames. , ppl: 10.208999154570547], batch size: 70
+2022-12-11 01:05:04,164 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 01:05:04,908 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.307, over 211138.00 frames. , ppl: 10.045027606490608
+2022-12-11 01:06:48,968 INFO [train.py:421] (5/8) Epoch 3, batch 60200, loss[loss=2.463, over 1050.00 frames. , ppl: 11.738361724349327] tot_loss[loss=2.322, over 5476296.47 frames. , ppl: 10.199297995266223], batch size: 70
+2022-12-11 01:08:28,495 INFO [train.py:421] (5/8) Epoch 3, batch 60400, loss[loss=2.518, over 1050.00 frames. , ppl: 12.401900194727004] tot_loss[loss=2.32, over 5539222.03 frames. , ppl: 10.18033418081315], batch size: 70
+2022-12-11 01:10:08,957 INFO [train.py:421] (5/8) Epoch 3, batch 60600, loss[loss=2.572, over 910.00 frames. , ppl: 13.093851250509159] tot_loss[loss=2.32, over 5529100.37 frames. , ppl: 10.176833686877337], batch size: 70
+2022-12-11 01:11:50,384 INFO [train.py:421] (5/8) Epoch 3, batch 60800, loss[loss=2.315, over 3430.00 frames. , ppl: 10.129927636905943] tot_loss[loss=2.32, over 5560580.78 frames. , ppl: 10.174681570395496], batch size: 70
+2022-12-11 01:13:29,060 INFO [train.py:421] (5/8) Epoch 3, batch 61000, loss[loss=2.501, over 1470.00 frames. , ppl: 12.189176708641837] tot_loss[loss=2.32, over 5541935.26 frames. , ppl: 10.177645670090483], batch size: 70
+2022-12-11 01:13:29,060 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 01:13:29,818 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.306, over 211138.00 frames. , ppl: 10.0369311167839
+2022-12-11 01:15:11,767 INFO [train.py:421] (5/8) Epoch 3, batch 61200, loss[loss=2.98, over 630.00 frames. , ppl: 19.680768995510228] tot_loss[loss=2.32, over 5555572.48 frames. , ppl: 10.17136769195396], batch size: 70
+2022-12-11 01:16:52,650 INFO [train.py:421] (5/8) Epoch 3, batch 61400, loss[loss=2.322, over 2310.00 frames. , ppl: 10.1953922356711] tot_loss[loss=2.319, over 5533922.82 frames. , ppl: 10.170304107034386], batch size: 70
+2022-12-11 01:18:30,898 INFO [train.py:421] (5/8) Epoch 3, batch 61600, loss[loss=2.43, over 1540.00 frames. , ppl: 11.362438770092538] tot_loss[loss=2.318, over 5574309.00 frames. , ppl: 10.156340345966546], batch size: 70
+2022-12-11 01:20:08,760 INFO [train.py:421] (5/8) Epoch 3, batch 61800, loss[loss=2.38, over 1820.00 frames. , ppl: 10.800958885305397] tot_loss[loss=2.318, over 5565351.19 frames. , ppl: 10.158629477335868], batch size: 70
+2022-12-11 01:21:46,207 INFO [train.py:421] (5/8) Epoch 3, batch 62000, loss[loss=3.252, over 490.00 frames. , ppl: 25.838083024288657] tot_loss[loss=2.319, over 5529164.53 frames. , ppl: 10.16859713911725], batch size: 70
+2022-12-11 01:21:46,207 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 01:21:46,967 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.307, over 211138.00 frames. , ppl: 10.04664679581944
+2022-12-11 01:23:28,132 INFO [train.py:421] (5/8) Epoch 3, batch 62200, loss[loss=2.326, over 2590.00 frames. , ppl: 10.239252523543367] tot_loss[loss=2.318, over 5546478.28 frames. , ppl: 10.15883549671112], batch size: 70
+2022-12-11 01:25:03,203 INFO [train.py:421] (5/8) Epoch 3, batch 62400, loss[loss=2.545, over 840.00 frames. , ppl: 12.748785644890814] tot_loss[loss=2.319, over 5514427.45 frames. , ppl: 10.16247837082096], batch size: 70
+2022-12-11 01:26:46,722 INFO [train.py:421] (5/8) Epoch 3, batch 62600, loss[loss=2.481, over 980.00 frames. , ppl: 11.953960659889603] tot_loss[loss=2.317, over 5575697.78 frames. , ppl: 10.142825983571724], batch size: 70
+2022-12-11 01:28:27,835 INFO [train.py:421] (5/8) Epoch 3, batch 62800, loss[loss=2.458, over 1820.00 frames. , ppl: 11.683190244053135] tot_loss[loss=2.316, over 5569723.82 frames. , ppl: 10.139545161286996], batch size: 70
+2022-12-11 01:30:02,615 INFO [train.py:421] (5/8) Epoch 3, batch 63000, loss[loss=2.243, over 13580.00 frames. , ppl: 9.42345181938151] tot_loss[loss=2.317, over 5565354.12 frames. , ppl: 10.144018972799415], batch size: 70
+2022-12-11 01:30:02,615 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 01:30:03,344 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.306, over 211138.00 frames. , ppl: 10.036726114257576
+2022-12-11 01:31:44,152 INFO [train.py:421] (5/8) Epoch 3, batch 63200, loss[loss=2.285, over 2240.00 frames. , ppl: 9.822034654840895] tot_loss[loss=2.317, over 5573151.13 frames. , ppl: 10.147060355759864], batch size: 70
+2022-12-11 01:33:23,261 INFO [train.py:421] (5/8) Epoch 3, batch 63400, loss[loss=2.337, over 840.00 frames. , ppl: 10.348828991533596] tot_loss[loss=2.317, over 5582413.17 frames. , ppl: 10.147372380947075], batch size: 70
+2022-12-11 01:35:03,944 INFO [train.py:421] (5/8) Epoch 3, batch 63600, loss[loss=2.546, over 1610.00 frames. , ppl: 12.755742167475518] tot_loss[loss=2.318, over 5565985.11 frames. , ppl: 10.151154981772898], batch size: 70
+2022-12-11 01:36:42,934 INFO [train.py:421] (5/8) Epoch 3, batch 63800, loss[loss=2.556, over 910.00 frames. , ppl: 12.890580147328835] tot_loss[loss=2.317, over 5577675.72 frames. , ppl: 10.144962007993442], batch size: 70
+2022-12-11 01:38:28,651 INFO [train.py:421] (5/8) Epoch 3, batch 64000, loss[loss=2.324, over 2450.00 frames. , ppl: 10.218685465819737] tot_loss[loss=2.317, over 5575398.42 frames. , ppl: 10.147107504855995], batch size: 70
+2022-12-11 01:38:28,652 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 01:38:29,412 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.306, over 211138.00 frames. , ppl: 10.036485464460732
+2022-12-11 01:40:09,937 INFO [train.py:421] (5/8) Epoch 3, batch 64200, loss[loss=2.871, over 700.00 frames. , ppl: 17.654653571810726] tot_loss[loss=2.319, over 5527681.22 frames. , ppl: 10.162504434115826], batch size: 70
+2022-12-11 01:41:46,095 INFO [train.py:421] (5/8) Epoch 3, batch 64400, loss[loss=2.439, over 1400.00 frames. , ppl: 11.463677103443699] tot_loss[loss=2.319, over 5515073.89 frames. , ppl: 10.166604612210318], batch size: 70
+2022-12-11 01:43:25,726 INFO [train.py:421] (5/8) Epoch 3, batch 64600, loss[loss=2.344, over 3360.00 frames. , ppl: 10.423530351528573] tot_loss[loss=2.319, over 5533939.94 frames. , ppl: 10.166260666395795], batch size: 70
+2022-12-11 01:45:05,816 INFO [train.py:421] (5/8) Epoch 3, batch 64800, loss[loss=2.322, over 1960.00 frames. , ppl: 10.192508924744937] tot_loss[loss=2.319, over 5541335.00 frames. , ppl: 10.16045855853479], batch size: 70
+2022-12-11 01:46:48,565 INFO [train.py:421] (5/8) Epoch 3, batch 65000, loss[loss=4.202, over 350.00 frames. , ppl: 66.78860656426725] tot_loss[loss=2.318, over 5544352.98 frames. , ppl: 10.1551499487192], batch size: 70
+2022-12-11 01:46:48,565 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 01:46:49,313 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.305, over 211138.00 frames. , ppl: 10.022263217124767
+2022-12-11 01:48:29,575 INFO [train.py:421] (5/8) Epoch 3, batch 65200, loss[loss=2.52, over 1330.00 frames. , ppl: 12.432547389799687] tot_loss[loss=2.318, over 5544181.16 frames. , ppl: 10.154252601253821], batch size: 70
+2022-12-11 01:50:10,551 INFO [train.py:421] (5/8) Epoch 3, batch 65400, loss[loss=2.228, over 4480.00 frames. , ppl: 9.285280401570827] tot_loss[loss=2.318, over 5529600.43 frames. , ppl: 10.155845780929283], batch size: 70
+2022-12-11 01:51:49,276 INFO [train.py:421] (5/8) Epoch 3, batch 65600, loss[loss=2.386, over 1960.00 frames. , ppl: 10.873857429418464] tot_loss[loss=2.319, over 5481015.62 frames. , ppl: 10.16590954454762], batch size: 70
+2022-12-11 01:53:27,733 INFO [train.py:421] (5/8) Epoch 3, batch 65800, loss[loss=2.659, over 700.00 frames. , ppl: 14.275093744269478] tot_loss[loss=2.319, over 5463103.74 frames. , ppl: 10.16785839460849], batch size: 70
+2022-12-11 01:55:11,465 INFO [train.py:421] (5/8) Epoch 3, batch 66000, loss[loss=3.585, over 420.00 frames. , ppl: 36.03604690896378] tot_loss[loss=2.32, over 5445200.70 frames. , ppl: 10.173864435066466], batch size: 70
+2022-12-11 01:55:11,466 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 01:55:12,223 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.305, over 211138.00 frames. , ppl: 10.026417513222102
+2022-12-11 01:56:53,388 INFO [train.py:421] (5/8) Epoch 3, batch 66200, loss[loss=2.898, over 560.00 frames. , ppl: 18.14157022435627] tot_loss[loss=2.319, over 5438353.62 frames. , ppl: 10.16475097105889], batch size: 70
+2022-12-11 01:58:33,167 INFO [train.py:421] (5/8) Epoch 3, batch 66400, loss[loss=2.309, over 3010.00 frames. , ppl: 10.062527989065734] tot_loss[loss=2.319, over 5417590.15 frames. , ppl: 10.167268594219141], batch size: 70
+2022-12-11 02:00:07,972 INFO [train.py:421] (5/8) Epoch 3, batch 66600, loss[loss=2.58, over 1050.00 frames. , ppl: 13.191008658589414] tot_loss[loss=2.319, over 5437581.34 frames. , ppl: 10.16186801366215], batch size: 70
+2022-12-11 02:01:49,359 INFO [train.py:421] (5/8) Epoch 3, batch 66800, loss[loss=2.316, over 2100.00 frames. , ppl: 10.132004459171133] tot_loss[loss=2.319, over 5427951.23 frames. , ppl: 10.168017865687844], batch size: 70
+2022-12-11 02:03:29,963 INFO [train.py:421] (5/8) Epoch 3, batch 67000, loss[loss=2.582, over 840.00 frames. , ppl: 13.222742239603935] tot_loss[loss=2.32, over 5422018.20 frames. , ppl: 10.17403572410247], batch size: 70
+2022-12-11 02:03:29,963 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 02:03:30,724 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.305, over 211138.00 frames. , ppl: 10.023177014556284
+2022-12-11 02:05:08,154 INFO [train.py:421] (5/8) Epoch 3, batch 67200, loss[loss=2.351, over 1890.00 frames. , ppl: 10.494846546638826] tot_loss[loss=2.321, over 5387165.33 frames. , ppl: 10.190230141384777], batch size: 70
+2022-12-11 02:06:47,972 INFO [train.py:421] (5/8) Epoch 3, batch 67400, loss[loss=2.287, over 3010.00 frames. , ppl: 9.846427766647656] tot_loss[loss=2.321, over 5401912.21 frames. , ppl: 10.182331559285242], batch size: 70
+2022-12-11 02:08:29,679 INFO [train.py:421] (5/8) Epoch 3, batch 67600, loss[loss=2.297, over 2940.00 frames. , ppl: 9.943477938491917] tot_loss[loss=2.321, over 5407598.29 frames. , ppl: 10.185412976077686], batch size: 70
+2022-12-11 02:10:11,841 INFO [train.py:421] (5/8) Epoch 3, batch 67800, loss[loss=2.228, over 3360.00 frames. , ppl: 9.284524119641565] tot_loss[loss=2.322, over 5368358.11 frames. , ppl: 10.196605060194543], batch size: 70
+2022-12-11 02:11:53,390 INFO [train.py:421] (5/8) Epoch 3, batch 68000, loss[loss=2.086, over 6790.00 frames. , ppl: 8.05165507250413] tot_loss[loss=2.321, over 5403873.13 frames. , ppl: 10.1844994533235], batch size: 70
+2022-12-11 02:11:53,390 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 02:11:54,123 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.305, over 211138.00 frames. , ppl: 10.025138400067002
+2022-12-11 02:13:31,897 INFO [train.py:421] (5/8) Epoch 3, batch 68200, loss[loss=2.228, over 11760.00 frames. , ppl: 9.282676417812786] tot_loss[loss=2.32, over 5438244.03 frames. , ppl: 10.172963797909055], batch size: 70
+2022-12-11 02:15:11,916 INFO [train.py:421] (5/8) Epoch 3, batch 68400, loss[loss=2.357, over 3220.00 frames. , ppl: 10.554487841788287] tot_loss[loss=2.319, over 5452628.16 frames. , ppl: 10.168540942501242], batch size: 70
+2022-12-11 02:16:58,787 INFO [train.py:421] (5/8) Epoch 3, batch 68600, loss[loss=2.278, over 910.00 frames. , ppl: 9.75834798235449] tot_loss[loss=2.319, over 5454569.69 frames. , ppl: 10.16145618936938], batch size: 70
+2022-12-11 02:18:39,312 INFO [train.py:421] (5/8) Epoch 3, batch 68800, loss[loss=2.36, over 1890.00 frames. , ppl: 10.590368116861713] tot_loss[loss=2.319, over 5435429.39 frames. , ppl: 10.168893222983915], batch size: 70
+2022-12-11 02:20:20,726 INFO [train.py:421] (5/8) Epoch 3, batch 69000, loss[loss=2.369, over 1680.00 frames. , ppl: 10.69025051342397] tot_loss[loss=2.32, over 5392535.80 frames. , ppl: 10.178835474669933], batch size: 70
+2022-12-11 02:20:20,726 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 02:20:21,480 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.305, over 211138.00 frames. , ppl: 10.019322126504301
+2022-12-11 02:22:00,180 INFO [train.py:421] (5/8) Epoch 3, batch 69200, loss[loss=2.582, over 1120.00 frames. , ppl: 13.223680951422331] tot_loss[loss=2.321, over 5393234.85 frames. , ppl: 10.189928230158193], batch size: 70
+2022-12-11 02:23:41,999 INFO [train.py:421] (5/8) Epoch 3, batch 69400, loss[loss=2.266, over 3500.00 frames. , ppl: 9.636680760789293] tot_loss[loss=2.322, over 5375079.72 frames. , ppl: 10.193921916439608], batch size: 70
+2022-12-11 02:25:20,934 INFO [train.py:421] (5/8) Epoch 3, batch 69600, loss[loss=2.437, over 1680.00 frames. , ppl: 11.436616533171147] tot_loss[loss=2.322, over 5370058.76 frames. , ppl: 10.194830546193028], batch size: 70
+2022-12-11 02:27:00,942 INFO [train.py:421] (5/8) Epoch 3, batch 69800, loss[loss=2.428, over 3010.00 frames. , ppl: 11.340357721042247] tot_loss[loss=2.32, over 5453865.94 frames. , ppl: 10.176974749155539], batch size: 70
+2022-12-11 02:28:40,242 INFO [train.py:421] (5/8) Epoch 3, batch 70000, loss[loss=2.359, over 1890.00 frames. , ppl: 10.583076398415766] tot_loss[loss=2.32, over 5459777.05 frames. , ppl: 10.17112822664578], batch size: 70
+2022-12-11 02:28:40,242 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 02:28:40,991 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.305, over 211138.00 frames. , ppl: 10.023684386112267
+2022-12-11 02:30:18,976 INFO [train.py:421] (5/8) Epoch 3, batch 70200, loss[loss=2.213, over 4200.00 frames. , ppl: 9.138642562137553] tot_loss[loss=2.319, over 5455504.98 frames. , ppl: 10.170092246754313], batch size: 70
+2022-12-11 02:31:59,181 INFO [train.py:421] (5/8) Epoch 3, batch 70400, loss[loss=2.268, over 2450.00 frames. , ppl: 9.661913997520061] tot_loss[loss=2.319, over 5472376.36 frames. , ppl: 10.164522110444084], batch size: 70
+2022-12-11 02:33:37,898 INFO [train.py:421] (5/8) Epoch 3, batch 70600, loss[loss=2.253, over 3570.00 frames. , ppl: 9.516983600698177] tot_loss[loss=2.317, over 5500357.48 frames. , ppl: 10.148651639181463], batch size: 70
+2022-12-11 02:35:16,197 INFO [train.py:421] (5/8) Epoch 3, batch 70800, loss[loss=3.232, over 490.00 frames. , ppl: 25.341832863278107] tot_loss[loss=2.318, over 5471786.63 frames. , ppl: 10.154113637629088], batch size: 70
+2022-12-11 02:36:54,122 INFO [train.py:421] (5/8) Epoch 3, batch 71000, loss[loss=2.223, over 5670.00 frames. , ppl: 9.234086414849973] tot_loss[loss=2.319, over 5441526.07 frames. , ppl: 10.16858175470497], batch size: 70
+2022-12-11 02:36:54,122 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 02:36:54,882 INFO [train.py:452] (5/8) Epoch 3, validation: loss=2.303, over 211138.00 frames. , ppl: 10.006615510581163
+2022-12-11 02:38:34,987 INFO [train.py:421] (5/8) Epoch 3, batch 71200, loss[loss=2.427, over 2100.00 frames. , ppl: 11.327793947217563] tot_loss[loss=2.319, over 5445311.57 frames. , ppl: 10.163292439378045], batch size: 70
+2022-12-11 02:40:17,301 INFO [train.py:421] (5/8) Epoch 3, batch 71400, loss[loss=2.336, over 2030.00 frames. , ppl: 10.334956271083568] tot_loss[loss=2.318, over 5475335.39 frames. , ppl: 10.159936435559443], batch size: 70
+2022-12-11 02:41:59,128 INFO [train.py:421] (5/8) Epoch 3, batch 71600, loss[loss=2.301, over 4690.00 frames. , ppl: 9.988801700117639] tot_loss[loss=2.317, over 5513406.56 frames. , ppl: 10.149640552741493], batch size: 70
+2022-12-11 02:43:42,979 INFO [train.py:421] (5/8) Epoch 3, batch 71800, loss[loss=2.583, over 1050.00 frames. , ppl: 13.23990899615875] tot_loss[loss=2.316, over 5559772.48 frames. , ppl: 10.13221937974544], batch size: 70
+2022-12-11 02:44:57,812 INFO [train.py:421] (5/8) Epoch 4, batch 0, loss[loss=2.334, over 2170.00 frames. , ppl: 10.321238515462724] tot_loss[loss=2.334, over 2170.00 frames. , ppl: 10.321238515462724], batch size: 70
+2022-12-11 02:46:37,238 INFO [train.py:421] (5/8) Epoch 4, batch 200, loss[loss=2.523, over 1260.00 frames. , ppl: 12.469689972413228] tot_loss[loss=2.32, over 497484.08 frames. , ppl: 10.180377602524743], batch size: 70
+2022-12-11 02:48:15,754 INFO [train.py:421] (5/8) Epoch 4, batch 400, loss[loss=2.259, over 7490.00 frames. , ppl: 9.569956810289042] tot_loss[loss=2.321, over 915839.97 frames. , ppl: 10.184267457844896], batch size: 70
+2022-12-11 02:49:54,209 INFO [train.py:421] (5/8) Epoch 4, batch 600, loss[loss=2.433, over 1190.00 frames. , ppl: 11.392313557801208] tot_loss[loss=2.318, over 1343269.56 frames. , ppl: 10.158011938901552], batch size: 70
+2022-12-11 02:51:35,083 INFO [train.py:421] (5/8) Epoch 4, batch 800, loss[loss=2.424, over 1330.00 frames. , ppl: 11.290186059842746] tot_loss[loss=2.313, over 1783593.11 frames. , ppl: 10.100743464361853], batch size: 70
+2022-12-11 02:53:15,879 INFO [train.py:421] (5/8) Epoch 4, batch 1000, loss[loss=2.414, over 1400.00 frames. , ppl: 11.178826737619747] tot_loss[loss=2.308, over 2171793.94 frames. , ppl: 10.055623695454749], batch size: 70
+2022-12-11 02:53:15,879 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 02:53:16,642 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.303, over 211138.00 frames. , ppl: 10.006100117136109
+2022-12-11 02:54:58,858 INFO [train.py:421] (5/8) Epoch 4, batch 1200, loss[loss=2.65, over 770.00 frames. , ppl: 14.15089756138138] tot_loss[loss=2.308, over 2483959.72 frames. , ppl: 10.05166647929557], batch size: 70
+2022-12-11 02:56:36,318 INFO [train.py:421] (5/8) Epoch 4, batch 1400, loss[loss=2.304, over 3500.00 frames. , ppl: 10.017922086402125] tot_loss[loss=2.308, over 2766149.01 frames. , ppl: 10.057005602136964], batch size: 70
+2022-12-11 02:58:17,502 INFO [train.py:421] (5/8) Epoch 4, batch 1600, loss[loss=2.277, over 3080.00 frames. , ppl: 9.742953138975246] tot_loss[loss=2.31, over 3002529.27 frames. , ppl: 10.073750480569615], batch size: 70
+2022-12-11 02:59:58,143 INFO [train.py:421] (5/8) Epoch 4, batch 1800, loss[loss=2.354, over 1890.00 frames. , ppl: 10.528959036069073] tot_loss[loss=2.31, over 3237394.09 frames. , ppl: 10.073191278103431], batch size: 70
+2022-12-11 03:01:36,803 INFO [train.py:421] (5/8) Epoch 4, batch 2000, loss[loss=2.248, over 5670.00 frames. , ppl: 9.464810959121353] tot_loss[loss=2.308, over 3483621.55 frames. , ppl: 10.053736269746043], batch size: 70
+2022-12-11 03:01:36,804 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 03:01:37,550 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.305, over 211138.00 frames. , ppl: 10.020806653868545
+2022-12-11 03:03:19,536 INFO [train.py:421] (5/8) Epoch 4, batch 2200, loss[loss=4.813, over 280.00 frames. , ppl: 123.04027872387532] tot_loss[loss=2.309, over 3658464.82 frames. , ppl: 10.063146999826182], batch size: 70
+2022-12-11 03:04:59,303 INFO [train.py:421] (5/8) Epoch 4, batch 2400, loss[loss=2.681, over 840.00 frames. , ppl: 14.601392271518941] tot_loss[loss=2.309, over 3832195.76 frames. , ppl: 10.060134918673734], batch size: 70
+2022-12-11 03:06:39,922 INFO [train.py:421] (5/8) Epoch 4, batch 2600, loss[loss=2.384, over 2520.00 frames. , ppl: 10.844555067631969] tot_loss[loss=2.308, over 3974455.69 frames. , ppl: 10.056764962905042], batch size: 70
+2022-12-11 03:08:21,018 INFO [train.py:421] (5/8) Epoch 4, batch 2800, loss[loss=2.327, over 1960.00 frames. , ppl: 10.243937478126366] tot_loss[loss=2.309, over 4110285.98 frames. , ppl: 10.061927120977016], batch size: 70
+2022-12-11 03:09:57,240 INFO [train.py:421] (5/8) Epoch 4, batch 3000, loss[loss=2.384, over 2100.00 frames. , ppl: 10.848851748900161] tot_loss[loss=2.309, over 4230146.55 frames. , ppl: 10.06513079813792], batch size: 70
+2022-12-11 03:09:57,240 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 03:09:58,000 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.303, over 211138.00 frames. , ppl: 10.00204158109846
+2022-12-11 03:11:38,666 INFO [train.py:421] (5/8) Epoch 4, batch 3200, loss[loss=2.944, over 630.00 frames. , ppl: 18.993728542455624] tot_loss[loss=2.309, over 4334094.16 frames. , ppl: 10.060816549182588], batch size: 70
+2022-12-11 03:13:21,675 INFO [train.py:421] (5/8) Epoch 4, batch 3400, loss[loss=2.325, over 1400.00 frames. , ppl: 10.22357923821393] tot_loss[loss=2.309, over 4430033.87 frames. , ppl: 10.067433703927797], batch size: 70
+2022-12-11 03:15:01,796 INFO [train.py:421] (5/8) Epoch 4, batch 3600, loss[loss=2.309, over 2030.00 frames. , ppl: 10.066638441623533] tot_loss[loss=2.31, over 4506346.87 frames. , ppl: 10.070478553777631], batch size: 70
+2022-12-11 03:16:43,478 INFO [train.py:421] (5/8) Epoch 4, batch 3800, loss[loss=2.36, over 1400.00 frames. , ppl: 10.591823232362419] tot_loss[loss=2.31, over 4624129.75 frames. , ppl: 10.073408018442567], batch size: 70
+2022-12-11 03:18:23,716 INFO [train.py:421] (5/8) Epoch 4, batch 4000, loss[loss=5.061, over 280.00 frames. , ppl: 157.82499876147784] tot_loss[loss=2.31, over 4671439.64 frames. , ppl: 10.07759067435927], batch size: 70
+2022-12-11 03:18:23,717 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 03:18:24,476 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.305, over 211138.00 frames. , ppl: 10.024831259142056
+2022-12-11 03:20:05,986 INFO [train.py:421] (5/8) Epoch 4, batch 4200, loss[loss=2.259, over 2520.00 frames. , ppl: 9.572213132790727] tot_loss[loss=2.308, over 4802650.40 frames. , ppl: 10.05597679956476], batch size: 70
+2022-12-11 03:21:49,026 INFO [train.py:421] (5/8) Epoch 4, batch 4400, loss[loss=2.428, over 2170.00 frames. , ppl: 11.340135128087558] tot_loss[loss=2.309, over 4861734.99 frames. , ppl: 10.061979741970168], batch size: 70
+2022-12-11 03:23:27,758 INFO [train.py:421] (5/8) Epoch 4, batch 4600, loss[loss=2.325, over 2730.00 frames. , ppl: 10.22489136340426] tot_loss[loss=2.308, over 4932453.29 frames. , ppl: 10.055079000466312], batch size: 70
+2022-12-11 03:25:07,823 INFO [train.py:421] (5/8) Epoch 4, batch 4800, loss[loss=2.318, over 2520.00 frames. , ppl: 10.151299877258175] tot_loss[loss=2.31, over 4970632.01 frames. , ppl: 10.070085021355], batch size: 70
+2022-12-11 03:26:49,307 INFO [train.py:421] (5/8) Epoch 4, batch 5000, loss[loss=2.228, over 9380.00 frames. , ppl: 9.279565724745169] tot_loss[loss=2.31, over 5040803.13 frames. , ppl: 10.069467465317276], batch size: 70
+2022-12-11 03:26:49,307 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 03:26:50,036 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.303, over 211138.00 frames. , ppl: 10.007444935075092
+2022-12-11 03:28:31,282 INFO [train.py:421] (5/8) Epoch 4, batch 5200, loss[loss=2.312, over 3920.00 frames. , ppl: 10.099156638722931] tot_loss[loss=2.309, over 5104252.04 frames. , ppl: 10.06929759205022], batch size: 70
+2022-12-11 03:30:12,449 INFO [train.py:421] (5/8) Epoch 4, batch 5400, loss[loss=2.376, over 1120.00 frames. , ppl: 10.766644198174337] tot_loss[loss=2.31, over 5113979.89 frames. , ppl: 10.076390727919046], batch size: 70
+2022-12-11 03:31:52,290 INFO [train.py:421] (5/8) Epoch 4, batch 5600, loss[loss=2.426, over 2380.00 frames. , ppl: 11.317613308271625] tot_loss[loss=2.309, over 5191937.55 frames. , ppl: 10.064464048655756], batch size: 70
+2022-12-11 03:33:32,829 INFO [train.py:421] (5/8) Epoch 4, batch 5800, loss[loss=2.6, over 1330.00 frames. , ppl: 13.468538471189307] tot_loss[loss=2.309, over 5211197.41 frames. , ppl: 10.067748407130104], batch size: 70
+2022-12-11 03:35:14,355 INFO [train.py:421] (5/8) Epoch 4, batch 6000, loss[loss=4.24, over 350.00 frames. , ppl: 69.37396955881374] tot_loss[loss=2.309, over 5277302.18 frames. , ppl: 10.06234752449739], batch size: 70
+2022-12-11 03:35:14,356 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 03:35:15,103 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.303, over 211138.00 frames. , ppl: 10.002629307959912
+2022-12-11 03:36:56,273 INFO [train.py:421] (5/8) Epoch 4, batch 6200, loss[loss=2.348, over 2590.00 frames. , ppl: 10.460076348710153] tot_loss[loss=2.308, over 5316379.25 frames. , ppl: 10.057245216351994], batch size: 70
+2022-12-11 03:38:34,613 INFO [train.py:421] (5/8) Epoch 4, batch 6400, loss[loss=2.579, over 1050.00 frames. , ppl: 13.177581573484987] tot_loss[loss=2.308, over 5351358.06 frames. , ppl: 10.059277395652394], batch size: 70
+2022-12-11 03:40:16,710 INFO [train.py:421] (5/8) Epoch 4, batch 6600, loss[loss=2.178, over 6090.00 frames. , ppl: 8.83191288682924] tot_loss[loss=2.308, over 5394197.05 frames. , ppl: 10.052165323756915], batch size: 70
+2022-12-11 03:41:57,721 INFO [train.py:421] (5/8) Epoch 4, batch 6800, loss[loss=2.987, over 560.00 frames. , ppl: 19.826061942092565] tot_loss[loss=2.307, over 5427098.97 frames. , ppl: 10.048608523391525], batch size: 70
+2022-12-11 03:43:38,826 INFO [train.py:421] (5/8) Epoch 4, batch 7000, loss[loss=2.283, over 2170.00 frames. , ppl: 9.807600752861479] tot_loss[loss=2.307, over 5448509.72 frames. , ppl: 10.041406908739008], batch size: 70
+2022-12-11 03:43:38,827 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 03:43:39,587 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.304, over 211138.00 frames. , ppl: 10.018172918207874
+2022-12-11 03:45:21,098 INFO [train.py:421] (5/8) Epoch 4, batch 7200, loss[loss=2.257, over 9660.00 frames. , ppl: 9.553765915191896] tot_loss[loss=2.307, over 5455417.69 frames. , ppl: 10.044723321752187], batch size: 70
+2022-12-11 03:47:01,090 INFO [train.py:421] (5/8) Epoch 4, batch 7400, loss[loss=2.207, over 4620.00 frames. , ppl: 9.091964613492856] tot_loss[loss=2.307, over 5462992.25 frames. , ppl: 10.048141049181192], batch size: 70
+2022-12-11 03:48:40,871 INFO [train.py:421] (5/8) Epoch 4, batch 7600, loss[loss=2.326, over 1750.00 frames. , ppl: 10.235705173235411] tot_loss[loss=2.307, over 5472293.03 frames. , ppl: 10.046439638875615], batch size: 70
+2022-12-11 03:50:23,892 INFO [train.py:421] (5/8) Epoch 4, batch 7800, loss[loss=2.328, over 1540.00 frames. , ppl: 10.25942357980425] tot_loss[loss=2.308, over 5459370.16 frames. , ppl: 10.0534434180817], batch size: 70
+2022-12-11 03:52:01,577 INFO [train.py:421] (5/8) Epoch 4, batch 8000, loss[loss=2.454, over 1260.00 frames. , ppl: 11.63941796080731] tot_loss[loss=2.308, over 5421275.10 frames. , ppl: 10.057517649134791], batch size: 70
+2022-12-11 03:52:01,577 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 03:52:02,323 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.304, over 211138.00 frames. , ppl: 10.015572487058913
+2022-12-11 03:53:42,801 INFO [train.py:421] (5/8) Epoch 4, batch 8200, loss[loss=2.374, over 2380.00 frames. , ppl: 10.738930649132191] tot_loss[loss=2.307, over 5469578.69 frames. , ppl: 10.046820064959057], batch size: 70
+2022-12-11 03:55:20,981 INFO [train.py:421] (5/8) Epoch 4, batch 8400, loss[loss=2.224, over 2940.00 frames. , ppl: 9.246210274666273] tot_loss[loss=2.308, over 5433085.37 frames. , ppl: 10.058788743359345], batch size: 70
+2022-12-11 03:56:59,259 INFO [train.py:421] (5/8) Epoch 4, batch 8600, loss[loss=2.308, over 2450.00 frames. , ppl: 10.052232633663921] tot_loss[loss=2.31, over 5431222.24 frames. , ppl: 10.07065935266816], batch size: 70
+2022-12-11 03:58:39,473 INFO [train.py:421] (5/8) Epoch 4, batch 8800, loss[loss=2.25, over 4270.00 frames. , ppl: 9.492159072728327] tot_loss[loss=2.309, over 5487473.20 frames. , ppl: 10.064140732916842], batch size: 70
+2022-12-11 04:00:15,936 INFO [train.py:421] (5/8) Epoch 4, batch 9000, loss[loss=2.347, over 2380.00 frames. , ppl: 10.455701761587733] tot_loss[loss=2.309, over 5474572.41 frames. , ppl: 10.06182065674169], batch size: 70
+2022-12-11 04:00:15,936 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 04:00:16,698 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.305, over 211138.00 frames. , ppl: 10.02282839694073
+2022-12-11 04:01:59,796 INFO [train.py:421] (5/8) Epoch 4, batch 9200, loss[loss=2.49, over 840.00 frames. , ppl: 12.062097845306285] tot_loss[loss=2.307, over 5532828.17 frames. , ppl: 10.048839246391667], batch size: 70
+2022-12-11 04:03:46,666 INFO [train.py:421] (5/8) Epoch 4, batch 9400, loss[loss=2.23, over 4410.00 frames. , ppl: 9.296155391167614] tot_loss[loss=2.306, over 5572457.71 frames. , ppl: 10.037498353432134], batch size: 70
+2022-12-11 04:05:23,921 INFO [train.py:421] (5/8) Epoch 4, batch 9600, loss[loss=2.24, over 2940.00 frames. , ppl: 9.390475556839043] tot_loss[loss=2.307, over 5539671.21 frames. , ppl: 10.042010006731125], batch size: 70
+2022-12-11 04:07:04,976 INFO [train.py:421] (5/8) Epoch 4, batch 9800, loss[loss=2.353, over 1610.00 frames. , ppl: 10.516072683156755] tot_loss[loss=2.307, over 5542606.23 frames. , ppl: 10.040876196316145], batch size: 70
+2022-12-11 04:08:44,055 INFO [train.py:421] (5/8) Epoch 4, batch 10000, loss[loss=2.266, over 3010.00 frames. , ppl: 9.642107673539545] tot_loss[loss=2.308, over 5478841.87 frames. , ppl: 10.05781361769033], batch size: 70
+2022-12-11 04:08:44,055 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 04:08:44,815 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.305, over 211138.00 frames. , ppl: 10.02091344153873
+2022-12-11 04:10:25,961 INFO [train.py:421] (5/8) Epoch 4, batch 10200, loss[loss=2.44, over 1260.00 frames. , ppl: 11.474689470618436] tot_loss[loss=2.307, over 5534455.38 frames. , ppl: 10.042699859593522], batch size: 70
+2022-12-11 04:12:05,325 INFO [train.py:421] (5/8) Epoch 4, batch 10400, loss[loss=2.269, over 2520.00 frames. , ppl: 9.667736212084789] tot_loss[loss=2.307, over 5540164.84 frames. , ppl: 10.04449840952975], batch size: 70
+2022-12-11 04:13:45,272 INFO [train.py:421] (5/8) Epoch 4, batch 10600, loss[loss=2.579, over 1890.00 frames. , ppl: 13.179068029907137] tot_loss[loss=2.307, over 5524274.40 frames. , ppl: 10.046179915043645], batch size: 70
+2022-12-11 04:15:25,275 INFO [train.py:421] (5/8) Epoch 4, batch 10800, loss[loss=2.477, over 1750.00 frames. , ppl: 11.90452436721378] tot_loss[loss=2.307, over 5530759.30 frames. , ppl: 10.048886725526916], batch size: 70
+2022-12-11 04:17:08,163 INFO [train.py:421] (5/8) Epoch 4, batch 11000, loss[loss=2.276, over 2870.00 frames. , ppl: 9.735540138455928] tot_loss[loss=2.306, over 5588585.07 frames. , ppl: 10.033524990603068], batch size: 70
+2022-12-11 04:17:08,163 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 04:17:08,910 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.303, over 211138.00 frames. , ppl: 10.005364098386623
+2022-12-11 04:18:47,977 INFO [train.py:421] (5/8) Epoch 4, batch 11200, loss[loss=2.322, over 2380.00 frames. , ppl: 10.192830809410355] tot_loss[loss=2.307, over 5566467.60 frames. , ppl: 10.043358437387054], batch size: 70
+2022-12-11 04:20:23,515 INFO [train.py:421] (5/8) Epoch 4, batch 11400, loss[loss=2.425, over 1050.00 frames. , ppl: 11.299654334878628] tot_loss[loss=2.309, over 5498507.55 frames. , ppl: 10.064348268098088], batch size: 70
+2022-12-11 04:22:05,511 INFO [train.py:421] (5/8) Epoch 4, batch 11600, loss[loss=2.23, over 4550.00 frames. , ppl: 9.297834350971945] tot_loss[loss=2.31, over 5468071.41 frames. , ppl: 10.075881328576456], batch size: 70
+2022-12-11 04:23:44,973 INFO [train.py:421] (5/8) Epoch 4, batch 11800, loss[loss=2.266, over 2590.00 frames. , ppl: 9.640580390992397] tot_loss[loss=2.312, over 5432497.60 frames. , ppl: 10.090861915834369], batch size: 70
+2022-12-11 04:25:25,266 INFO [train.py:421] (5/8) Epoch 4, batch 12000, loss[loss=2.153, over 8050.00 frames. , ppl: 8.609617992198531] tot_loss[loss=2.312, over 5423519.10 frames. , ppl: 10.093391868975965], batch size: 70
+2022-12-11 04:25:25,267 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 04:25:25,998 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.301, over 211138.00 frames. , ppl: 9.98219907371343
+2022-12-11 04:27:07,827 INFO [train.py:421] (5/8) Epoch 4, batch 12200, loss[loss=2.342, over 2030.00 frames. , ppl: 10.40601010662215] tot_loss[loss=2.311, over 5448862.82 frames. , ppl: 10.087057832207595], batch size: 70
+2022-12-11 04:28:46,443 INFO [train.py:421] (5/8) Epoch 4, batch 12400, loss[loss=2.463, over 1190.00 frames. , ppl: 11.744312238223522] tot_loss[loss=2.312, over 5446569.46 frames. , ppl: 10.090106152346351], batch size: 70
+2022-12-11 04:30:27,175 INFO [train.py:421] (5/8) Epoch 4, batch 12600, loss[loss=2.183, over 7630.00 frames. , ppl: 8.870667445025282] tot_loss[loss=2.311, over 5493552.29 frames. , ppl: 10.08669046634558], batch size: 70
+2022-12-11 04:32:07,001 INFO [train.py:421] (5/8) Epoch 4, batch 12800, loss[loss=2.54, over 1050.00 frames. , ppl: 12.675915504765506] tot_loss[loss=2.311, over 5512663.47 frames. , ppl: 10.082624201784643], batch size: 70
+2022-12-11 04:33:50,382 INFO [train.py:421] (5/8) Epoch 4, batch 13000, loss[loss=2.566, over 840.00 frames. , ppl: 13.01063532238932] tot_loss[loss=2.311, over 5512316.37 frames. , ppl: 10.080363103455479], batch size: 70
+2022-12-11 04:33:50,382 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 04:33:51,146 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.3, over 211138.00 frames. , ppl: 9.97099750372906
+2022-12-11 04:35:32,578 INFO [train.py:421] (5/8) Epoch 4, batch 13200, loss[loss=2.245, over 2940.00 frames. , ppl: 9.43930493302324] tot_loss[loss=2.311, over 5494420.63 frames. , ppl: 10.083268611944202], batch size: 70
+2022-12-11 04:37:11,847 INFO [train.py:421] (5/8) Epoch 4, batch 13400, loss[loss=2.41, over 1400.00 frames. , ppl: 11.135743684562895] tot_loss[loss=2.312, over 5479257.12 frames. , ppl: 10.08983158192938], batch size: 70
+2022-12-11 04:38:49,025 INFO [train.py:421] (5/8) Epoch 4, batch 13600, loss[loss=2.327, over 1750.00 frames. , ppl: 10.248163238711335] tot_loss[loss=2.312, over 5476783.47 frames. , ppl: 10.092493816471109], batch size: 70
+2022-12-11 04:40:27,700 INFO [train.py:421] (5/8) Epoch 4, batch 13800, loss[loss=2.41, over 1400.00 frames. , ppl: 11.13456888518613] tot_loss[loss=2.311, over 5486271.24 frames. , ppl: 10.083042927004138], batch size: 70
+2022-12-11 04:42:07,519 INFO [train.py:421] (5/8) Epoch 4, batch 14000, loss[loss=3.077, over 560.00 frames. , ppl: 21.699671055569624] tot_loss[loss=2.311, over 5499943.63 frames. , ppl: 10.079991393926608], batch size: 70
+2022-12-11 04:42:07,520 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 04:42:08,266 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.301, over 211138.00 frames. , ppl: 9.986692979928648
+2022-12-11 04:43:46,528 INFO [train.py:421] (5/8) Epoch 4, batch 14200, loss[loss=2.43, over 1400.00 frames. , ppl: 11.355166647259384] tot_loss[loss=2.311, over 5493054.63 frames. , ppl: 10.082989464648026], batch size: 70
+2022-12-11 04:45:27,608 INFO [train.py:421] (5/8) Epoch 4, batch 14400, loss[loss=2.453, over 2030.00 frames. , ppl: 11.62330285283075] tot_loss[loss=2.31, over 5561379.19 frames. , ppl: 10.077107686055397], batch size: 70
+2022-12-11 04:47:07,724 INFO [train.py:421] (5/8) Epoch 4, batch 14600, loss[loss=2.205, over 2730.00 frames. , ppl: 9.067787329671425] tot_loss[loss=2.311, over 5533596.70 frames. , ppl: 10.084701365992991], batch size: 70
+2022-12-11 04:48:49,492 INFO [train.py:421] (5/8) Epoch 4, batch 14800, loss[loss=2.439, over 1890.00 frames. , ppl: 11.462171356756672] tot_loss[loss=2.312, over 5537634.09 frames. , ppl: 10.092841860274556], batch size: 70
+2022-12-11 04:50:23,961 INFO [train.py:421] (5/8) Epoch 4, batch 15000, loss[loss=2.292, over 4410.00 frames. , ppl: 9.895109878710954] tot_loss[loss=2.313, over 5495167.86 frames. , ppl: 10.104107606312285], batch size: 70
+2022-12-11 04:50:23,962 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 04:50:24,691 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.3, over 211138.00 frames. , ppl: 9.975186652961758
+2022-12-11 04:52:04,475 INFO [train.py:421] (5/8) Epoch 4, batch 15200, loss[loss=2.206, over 5670.00 frames. , ppl: 9.081483853098165] tot_loss[loss=2.312, over 5510773.16 frames. , ppl: 10.094467487628206], batch size: 70
+2022-12-11 04:53:41,644 INFO [train.py:421] (5/8) Epoch 4, batch 15400, loss[loss=2.478, over 1260.00 frames. , ppl: 11.92115898448616] tot_loss[loss=2.313, over 5479126.67 frames. , ppl: 10.101906467742168], batch size: 70
+2022-12-11 04:55:25,232 INFO [train.py:421] (5/8) Epoch 4, batch 15600, loss[loss=2.385, over 1960.00 frames. , ppl: 10.855122381663996] tot_loss[loss=2.312, over 5480862.41 frames. , ppl: 10.09815661090064], batch size: 70
+2022-12-11 04:57:01,735 INFO [train.py:421] (5/8) Epoch 4, batch 15800, loss[loss=2.397, over 2380.00 frames. , ppl: 10.98917049662174] tot_loss[loss=2.313, over 5470173.51 frames. , ppl: 10.101783374012813], batch size: 70
+2022-12-11 04:58:46,138 INFO [train.py:421] (5/8) Epoch 4, batch 16000, loss[loss=2.354, over 1610.00 frames. , ppl: 10.531271185267656] tot_loss[loss=2.312, over 5482363.38 frames. , ppl: 10.098432509959588], batch size: 70
+2022-12-11 04:58:46,139 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 04:58:46,898 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.301, over 211138.00 frames. , ppl: 9.980866511809946
+2022-12-11 05:00:26,707 INFO [train.py:421] (5/8) Epoch 4, batch 16200, loss[loss=2.612, over 980.00 frames. , ppl: 13.629521948774123] tot_loss[loss=2.314, over 5449511.50 frames. , ppl: 10.110582964336334], batch size: 70
+2022-12-11 05:02:10,985 INFO [train.py:421] (5/8) Epoch 4, batch 16400, loss[loss=2.583, over 770.00 frames. , ppl: 13.233866767464635] tot_loss[loss=2.313, over 5452766.66 frames. , ppl: 10.102167898354333], batch size: 70
+2022-12-11 05:03:48,549 INFO [train.py:421] (5/8) Epoch 4, batch 16600, loss[loss=2.21, over 5040.00 frames. , ppl: 9.1173704905833] tot_loss[loss=2.313, over 5434777.95 frames. , ppl: 10.1095213616807], batch size: 70
+2022-12-11 05:05:30,768 INFO [train.py:421] (5/8) Epoch 4, batch 16800, loss[loss=2.321, over 2730.00 frames. , ppl: 10.184073779564857] tot_loss[loss=2.314, over 5412331.03 frames. , ppl: 10.11883019360633], batch size: 70
+2022-12-11 05:07:12,257 INFO [train.py:421] (5/8) Epoch 4, batch 17000, loss[loss=2.294, over 2730.00 frames. , ppl: 9.918317844236245] tot_loss[loss=2.313, over 5433801.95 frames. , ppl: 10.107512707017891], batch size: 70
+2022-12-11 05:07:12,257 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 05:07:13,004 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.302, over 211138.00 frames. , ppl: 9.993913175932885
+2022-12-11 05:08:54,242 INFO [train.py:421] (5/8) Epoch 4, batch 17200, loss[loss=2.299, over 2590.00 frames. , ppl: 9.966148911245748] tot_loss[loss=2.313, over 5430977.27 frames. , ppl: 10.10623023644333], batch size: 70
+2022-12-11 05:10:34,554 INFO [train.py:421] (5/8) Epoch 4, batch 17400, loss[loss=2.57, over 1050.00 frames. , ppl: 13.067884366359436] tot_loss[loss=2.313, over 5436092.68 frames. , ppl: 10.10744151760126], batch size: 70
+2022-12-11 05:12:12,660 INFO [train.py:421] (5/8) Epoch 4, batch 17600, loss[loss=2.349, over 2450.00 frames. , ppl: 10.472846644287152] tot_loss[loss=2.313, over 5444242.94 frames. , ppl: 10.107864705598379], batch size: 70
+2022-12-11 05:13:51,762 INFO [train.py:421] (5/8) Epoch 4, batch 17800, loss[loss=2.423, over 840.00 frames. , ppl: 11.275620085738828] tot_loss[loss=2.313, over 5446859.08 frames. , ppl: 10.106540239334365], batch size: 70
+2022-12-11 05:15:30,815 INFO [train.py:421] (5/8) Epoch 4, batch 18000, loss[loss=2.311, over 1890.00 frames. , ppl: 10.086367310253475] tot_loss[loss=2.313, over 5446244.72 frames. , ppl: 10.107081099966768], batch size: 70
+2022-12-11 05:15:30,815 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 05:15:31,578 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.302, over 211138.00 frames. , ppl: 9.995456073238609
+2022-12-11 05:17:10,754 INFO [train.py:421] (5/8) Epoch 4, batch 18200, loss[loss=2.274, over 2310.00 frames. , ppl: 9.716032872592814] tot_loss[loss=2.313, over 5458047.29 frames. , ppl: 10.102791690849093], batch size: 70
+2022-12-11 05:18:48,983 INFO [train.py:421] (5/8) Epoch 4, batch 18400, loss[loss=2.337, over 1680.00 frames. , ppl: 10.351865832248505] tot_loss[loss=2.312, over 5474373.04 frames. , ppl: 10.095215350877295], batch size: 70
+2022-12-11 05:20:26,265 INFO [train.py:421] (5/8) Epoch 4, batch 18600, loss[loss=2.303, over 2240.00 frames. , ppl: 10.002443523127075] tot_loss[loss=2.311, over 5482417.51 frames. , ppl: 10.088718134376002], batch size: 70
+2022-12-11 05:22:06,444 INFO [train.py:421] (5/8) Epoch 4, batch 18800, loss[loss=2.327, over 1960.00 frames. , ppl: 10.250886381537388] tot_loss[loss=2.311, over 5516434.80 frames. , ppl: 10.080538933808985], batch size: 70
+2022-12-11 05:23:48,712 INFO [train.py:421] (5/8) Epoch 4, batch 19000, loss[loss=2.435, over 1330.00 frames. , ppl: 11.417839409617887] tot_loss[loss=2.311, over 5518382.25 frames. , ppl: 10.081596784224688], batch size: 70
+2022-12-11 05:23:48,712 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 05:23:49,478 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.3, over 211138.00 frames. , ppl: 9.977298131379836
+2022-12-11 05:25:29,342 INFO [train.py:421] (5/8) Epoch 4, batch 19200, loss[loss=2.862, over 630.00 frames. , ppl: 17.49142239620794] tot_loss[loss=2.31, over 5538543.51 frames. , ppl: 10.070271356355107], batch size: 70
+2022-12-11 05:27:11,514 INFO [train.py:421] (5/8) Epoch 4, batch 19400, loss[loss=2.396, over 1610.00 frames. , ppl: 10.976054259203678] tot_loss[loss=2.31, over 5534693.00 frames. , ppl: 10.075436169563556], batch size: 70
+2022-12-11 05:28:50,710 INFO [train.py:421] (5/8) Epoch 4, batch 19600, loss[loss=2.67, over 770.00 frames. , ppl: 14.433604761147134] tot_loss[loss=2.311, over 5535750.56 frames. , ppl: 10.086142714773793], batch size: 70
+2022-12-11 05:30:32,726 INFO [train.py:421] (5/8) Epoch 4, batch 19800, loss[loss=2.402, over 1050.00 frames. , ppl: 11.046476050385284] tot_loss[loss=2.311, over 5522910.81 frames. , ppl: 10.088138455778948], batch size: 70
+2022-12-11 05:32:10,663 INFO [train.py:421] (5/8) Epoch 4, batch 20000, loss[loss=2.582, over 700.00 frames. , ppl: 13.230056923058733] tot_loss[loss=2.311, over 5518306.01 frames. , ppl: 10.083237073781739], batch size: 70
+2022-12-11 05:32:10,664 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 05:32:11,424 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.302, over 211138.00 frames. , ppl: 9.990693533416067
+2022-12-11 05:33:50,978 INFO [train.py:421] (5/8) Epoch 4, batch 20200, loss[loss=2.311, over 3010.00 frames. , ppl: 10.086286954613989] tot_loss[loss=2.31, over 5532000.22 frames. , ppl: 10.073644549852308], batch size: 70
+2022-12-11 05:35:32,013 INFO [train.py:421] (5/8) Epoch 4, batch 20400, loss[loss=2.291, over 6300.00 frames. , ppl: 9.883798361285262] tot_loss[loss=2.312, over 5454790.16 frames. , ppl: 10.093404759017709], batch size: 70
+2022-12-11 05:37:16,459 INFO [train.py:421] (5/8) Epoch 4, batch 20600, loss[loss=2.325, over 3640.00 frames. , ppl: 10.223802371275111] tot_loss[loss=2.311, over 5479565.77 frames. , ppl: 10.084299815921797], batch size: 70
+2022-12-11 05:38:53,403 INFO [train.py:421] (5/8) Epoch 4, batch 20800, loss[loss=2.44, over 980.00 frames. , ppl: 11.471876943060687] tot_loss[loss=2.311, over 5486719.45 frames. , ppl: 10.082231546077633], batch size: 70
+2022-12-11 05:40:32,937 INFO [train.py:421] (5/8) Epoch 4, batch 21000, loss[loss=2.523, over 1190.00 frames. , ppl: 12.47009376143459] tot_loss[loss=2.311, over 5508731.58 frames. , ppl: 10.087638303716293], batch size: 70
+2022-12-11 05:40:32,938 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 05:40:33,720 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.301, over 211138.00 frames. , ppl: 9.982080879215324
+2022-12-11 05:42:16,404 INFO [train.py:421] (5/8) Epoch 4, batch 21200, loss[loss=2.274, over 1890.00 frames. , ppl: 9.721542584948754] tot_loss[loss=2.311, over 5515544.25 frames. , ppl: 10.086396627467378], batch size: 70
+2022-12-11 05:43:57,340 INFO [train.py:421] (5/8) Epoch 4, batch 21400, loss[loss=2.441, over 2170.00 frames. , ppl: 11.484153710953207] tot_loss[loss=2.309, over 5568739.57 frames. , ppl: 10.068298664373195], batch size: 70
+2022-12-11 05:45:35,240 INFO [train.py:421] (5/8) Epoch 4, batch 21600, loss[loss=2.307, over 3570.00 frames. , ppl: 10.045365654502175] tot_loss[loss=2.31, over 5562209.86 frames. , ppl: 10.073798568387305], batch size: 70
+2022-12-11 05:47:14,433 INFO [train.py:421] (5/8) Epoch 4, batch 21800, loss[loss=2.448, over 1610.00 frames. , ppl: 11.55995438681548] tot_loss[loss=2.309, over 5568201.93 frames. , ppl: 10.064985605886184], batch size: 70
+2022-12-11 05:48:55,727 INFO [train.py:421] (5/8) Epoch 4, batch 22000, loss[loss=2.352, over 1190.00 frames. , ppl: 10.502056726835105] tot_loss[loss=2.311, over 5527708.51 frames. , ppl: 10.079934728300929], batch size: 70
+2022-12-11 05:48:55,727 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 05:48:56,472 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.301, over 211138.00 frames. , ppl: 9.988403293971805
+2022-12-11 05:50:35,546 INFO [train.py:421] (5/8) Epoch 4, batch 22200, loss[loss=2.256, over 5670.00 frames. , ppl: 9.548512424872978] tot_loss[loss=2.311, over 5505461.32 frames. , ppl: 10.088409544005502], batch size: 70
+2022-12-11 05:52:14,295 INFO [train.py:421] (5/8) Epoch 4, batch 22400, loss[loss=2.386, over 2170.00 frames. , ppl: 10.874066980197835] tot_loss[loss=2.31, over 5547500.79 frames. , ppl: 10.074556436246962], batch size: 70
+2022-12-11 05:53:54,670 INFO [train.py:421] (5/8) Epoch 4, batch 22600, loss[loss=2.475, over 1610.00 frames. , ppl: 11.884957903404013] tot_loss[loss=2.31, over 5552242.70 frames. , ppl: 10.077624108991275], batch size: 70
+2022-12-11 05:55:36,838 INFO [train.py:421] (5/8) Epoch 4, batch 22800, loss[loss=2.482, over 770.00 frames. , ppl: 11.962252141395025] tot_loss[loss=2.312, over 5496953.29 frames. , ppl: 10.094374995705472], batch size: 70
+2022-12-11 05:57:12,502 INFO [train.py:421] (5/8) Epoch 4, batch 23000, loss[loss=3.339, over 490.00 frames. , ppl: 28.203979638599858] tot_loss[loss=2.312, over 5490894.61 frames. , ppl: 10.092185726129127], batch size: 70
+2022-12-11 05:57:12,503 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 05:57:13,264 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.298, over 211138.00 frames. , ppl: 9.957022795141143
+2022-12-11 05:58:56,295 INFO [train.py:421] (5/8) Epoch 4, batch 23200, loss[loss=2.204, over 5950.00 frames. , ppl: 9.060267999218091] tot_loss[loss=2.311, over 5510413.59 frames. , ppl: 10.085848811573944], batch size: 70
+2022-12-11 06:00:34,257 INFO [train.py:421] (5/8) Epoch 4, batch 23400, loss[loss=2.399, over 1190.00 frames. , ppl: 11.01680978817538] tot_loss[loss=2.31, over 5561746.25 frames. , ppl: 10.074165740579891], batch size: 70
+2022-12-11 06:02:14,833 INFO [train.py:421] (5/8) Epoch 4, batch 23600, loss[loss=2.485, over 1260.00 frames. , ppl: 12.00308395441316] tot_loss[loss=2.309, over 5608131.82 frames. , ppl: 10.061238442705346], batch size: 70
+2022-12-11 06:03:55,549 INFO [train.py:421] (5/8) Epoch 4, batch 23800, loss[loss=2.375, over 1750.00 frames. , ppl: 10.749180510529923] tot_loss[loss=2.309, over 5576667.00 frames. , ppl: 10.065725816206173], batch size: 70
+2022-12-11 06:05:34,128 INFO [train.py:421] (5/8) Epoch 4, batch 24000, loss[loss=2.504, over 910.00 frames. , ppl: 12.228186654658751] tot_loss[loss=2.31, over 5552403.48 frames. , ppl: 10.06987962565663], batch size: 70
+2022-12-11 06:05:34,129 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 06:05:34,888 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.3, over 211138.00 frames. , ppl: 9.975346105661082
+2022-12-11 06:07:17,272 INFO [train.py:421] (5/8) Epoch 4, batch 24200, loss[loss=2.479, over 1820.00 frames. , ppl: 11.926175661112557] tot_loss[loss=2.309, over 5563298.39 frames. , ppl: 10.062044934517768], batch size: 70
+2022-12-11 06:08:57,886 INFO [train.py:421] (5/8) Epoch 4, batch 24400, loss[loss=2.296, over 3220.00 frames. , ppl: 9.936172169817654] tot_loss[loss=2.307, over 5642356.26 frames. , ppl: 10.04232769055407], batch size: 70
+2022-12-11 06:10:39,363 INFO [train.py:421] (5/8) Epoch 4, batch 24600, loss[loss=2.356, over 1610.00 frames. , ppl: 10.543861179759416] tot_loss[loss=2.306, over 5659234.02 frames. , ppl: 10.030335361076922], batch size: 70
+2022-12-11 06:12:21,102 INFO [train.py:421] (5/8) Epoch 4, batch 24800, loss[loss=2.576, over 980.00 frames. , ppl: 13.147621688743786] tot_loss[loss=2.305, over 5669376.53 frames. , ppl: 10.028787612120766], batch size: 70
+2022-12-11 06:14:01,869 INFO [train.py:421] (5/8) Epoch 4, batch 25000, loss[loss=2.452, over 1750.00 frames. , ppl: 11.609816665087125] tot_loss[loss=2.306, over 5674259.83 frames. , ppl: 10.030084569379918], batch size: 70
+2022-12-11 06:14:01,869 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 06:14:02,628 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.3, over 211138.00 frames. , ppl: 9.974898758710204
+2022-12-11 06:15:42,492 INFO [train.py:421] (5/8) Epoch 4, batch 25200, loss[loss=2.267, over 3010.00 frames. , ppl: 9.653534222495441] tot_loss[loss=2.305, over 5658572.21 frames. , ppl: 10.028910624151289], batch size: 70
+2022-12-11 06:17:23,860 INFO [train.py:421] (5/8) Epoch 4, batch 25400, loss[loss=2.229, over 5460.00 frames. , ppl: 9.291607547821549] tot_loss[loss=2.306, over 5671984.23 frames. , ppl: 10.029504291168333], batch size: 70
+2022-12-11 06:19:05,299 INFO [train.py:421] (5/8) Epoch 4, batch 25600, loss[loss=2.656, over 700.00 frames. , ppl: 14.235274514548081] tot_loss[loss=2.306, over 5691211.58 frames. , ppl: 10.03086773850053], batch size: 70
+2022-12-11 06:20:48,746 INFO [train.py:421] (5/8) Epoch 4, batch 25800, loss[loss=2.463, over 1050.00 frames. , ppl: 11.741877645582866] tot_loss[loss=2.306, over 5668423.05 frames. , ppl: 10.03635780366723], batch size: 70
+2022-12-11 06:22:27,270 INFO [train.py:421] (5/8) Epoch 4, batch 26000, loss[loss=2.337, over 1470.00 frames. , ppl: 10.353986540736393] tot_loss[loss=2.307, over 5637585.00 frames. , ppl: 10.043882992352536], batch size: 70
+2022-12-11 06:22:27,271 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 06:22:28,032 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.301, over 211138.00 frames. , ppl: 9.983234812864332
+2022-12-11 06:24:08,765 INFO [train.py:421] (5/8) Epoch 4, batch 26200, loss[loss=2.535, over 910.00 frames. , ppl: 12.620203451844194] tot_loss[loss=2.307, over 5625640.25 frames. , ppl: 10.049074160263427], batch size: 70
+2022-12-11 06:25:49,617 INFO [train.py:421] (5/8) Epoch 4, batch 26400, loss[loss=2.203, over 4900.00 frames. , ppl: 9.052979313424993] tot_loss[loss=2.307, over 5612543.92 frames. , ppl: 10.048062456323931], batch size: 70
+2022-12-11 06:27:29,120 INFO [train.py:421] (5/8) Epoch 4, batch 26600, loss[loss=2.403, over 1890.00 frames. , ppl: 11.05896329908017] tot_loss[loss=2.309, over 5544800.72 frames. , ppl: 10.060936353059668], batch size: 70
+2022-12-11 06:29:04,076 INFO [train.py:421] (5/8) Epoch 4, batch 26800, loss[loss=2.411, over 1470.00 frames. , ppl: 11.14111373931709] tot_loss[loss=2.31, over 5490502.19 frames. , ppl: 10.074880953037635], batch size: 70
+2022-12-11 06:30:43,417 INFO [train.py:421] (5/8) Epoch 4, batch 27000, loss[loss=2.243, over 4200.00 frames. , ppl: 9.423642830394993] tot_loss[loss=2.31, over 5492336.90 frames. , ppl: 10.071012017774109], batch size: 70
+2022-12-11 06:30:43,417 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 06:30:44,146 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.299, over 211138.00 frames. , ppl: 9.968636532001739
+2022-12-11 06:32:23,960 INFO [train.py:421] (5/8) Epoch 4, batch 27200, loss[loss=2.515, over 910.00 frames. , ppl: 12.367257756443765] tot_loss[loss=2.309, over 5506741.84 frames. , ppl: 10.064548100720499], batch size: 70
+2022-12-11 06:34:06,636 INFO [train.py:421] (5/8) Epoch 4, batch 27400, loss[loss=2.305, over 5810.00 frames. , ppl: 10.02817359611557] tot_loss[loss=2.309, over 5503822.34 frames. , ppl: 10.069003451671614], batch size: 70
+2022-12-11 06:35:42,978 INFO [train.py:421] (5/8) Epoch 4, batch 27600, loss[loss=2.295, over 2450.00 frames. , ppl: 9.925816699902013] tot_loss[loss=2.31, over 5498508.54 frames. , ppl: 10.071958023343281], batch size: 70
+2022-12-11 06:37:23,604 INFO [train.py:421] (5/8) Epoch 4, batch 27800, loss[loss=2.346, over 2590.00 frames. , ppl: 10.440225418027568] tot_loss[loss=2.309, over 5530654.19 frames. , ppl: 10.06691900911691], batch size: 70
+2022-12-11 06:39:01,492 INFO [train.py:421] (5/8) Epoch 4, batch 28000, loss[loss=2.485, over 1330.00 frames. , ppl: 12.000774832821326] tot_loss[loss=2.308, over 5546265.81 frames. , ppl: 10.058691414706047], batch size: 70
+2022-12-11 06:39:01,493 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 06:39:02,238 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.299, over 211138.00 frames. , ppl: 9.969008347954837
+2022-12-11 06:40:41,649 INFO [train.py:421] (5/8) Epoch 4, batch 28200, loss[loss=2.347, over 2450.00 frames. , ppl: 10.44935212909328] tot_loss[loss=2.309, over 5494705.19 frames. , ppl: 10.062823019129006], batch size: 70
+2022-12-11 06:42:23,115 INFO [train.py:421] (5/8) Epoch 4, batch 28400, loss[loss=2.335, over 4200.00 frames. , ppl: 10.324595123430736] tot_loss[loss=2.31, over 5485030.28 frames. , ppl: 10.06993205898918], batch size: 70
+2022-12-11 06:44:01,461 INFO [train.py:421] (5/8) Epoch 4, batch 28600, loss[loss=2.686, over 630.00 frames. , ppl: 14.666094739479918] tot_loss[loss=2.308, over 5547206.69 frames. , ppl: 10.05010216920541], batch size: 70
+2022-12-11 06:45:43,568 INFO [train.py:421] (5/8) Epoch 4, batch 28800, loss[loss=2.335, over 4270.00 frames. , ppl: 10.32705341685239] tot_loss[loss=2.31, over 5500369.06 frames. , ppl: 10.070826120695552], batch size: 70
+2022-12-11 06:47:27,938 INFO [train.py:421] (5/8) Epoch 4, batch 29000, loss[loss=2.283, over 2800.00 frames. , ppl: 9.801974840807091] tot_loss[loss=2.31, over 5512517.45 frames. , ppl: 10.07070413478575], batch size: 70
+2022-12-11 06:47:27,938 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 06:47:28,683 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.302, over 211138.00 frames. , ppl: 9.996510942600489
+2022-12-11 06:49:06,743 INFO [train.py:421] (5/8) Epoch 4, batch 29200, loss[loss=2.296, over 3920.00 frames. , ppl: 9.935103745740514] tot_loss[loss=2.31, over 5514857.29 frames. , ppl: 10.075197025886593], batch size: 70
+2022-12-11 06:50:43,147 INFO [train.py:421] (5/8) Epoch 4, batch 29400, loss[loss=2.516, over 1610.00 frames. , ppl: 12.375612621718322] tot_loss[loss=2.311, over 5490789.08 frames. , ppl: 10.082879942718344], batch size: 70
+2022-12-11 06:52:19,513 INFO [train.py:421] (5/8) Epoch 4, batch 29600, loss[loss=2.355, over 2240.00 frames. , ppl: 10.541373827315546] tot_loss[loss=2.312, over 5431588.47 frames. , ppl: 10.098030117216824], batch size: 70
+2022-12-11 06:53:56,648 INFO [train.py:421] (5/8) Epoch 4, batch 29800, loss[loss=2.466, over 1400.00 frames. , ppl: 11.77908714863324] tot_loss[loss=2.313, over 5421525.11 frames. , ppl: 10.100521018874998], batch size: 70
+2022-12-11 06:55:37,409 INFO [train.py:421] (5/8) Epoch 4, batch 30000, loss[loss=2.252, over 2800.00 frames. , ppl: 9.51104531210182] tot_loss[loss=2.312, over 5425123.48 frames. , ppl: 10.09629095526707], batch size: 70
+2022-12-11 06:55:37,409 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 06:55:38,158 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.3, over 211138.00 frames. , ppl: 9.970644798055698
+2022-12-11 06:57:17,485 INFO [train.py:421] (5/8) Epoch 4, batch 30200, loss[loss=2.206, over 3570.00 frames. , ppl: 9.078296754540517] tot_loss[loss=2.311, over 5470598.23 frames. , ppl: 10.08262615231197], batch size: 70
+2022-12-11 06:58:56,929 INFO [train.py:421] (5/8) Epoch 4, batch 30400, loss[loss=2.289, over 3780.00 frames. , ppl: 9.863089929890775] tot_loss[loss=2.31, over 5490700.17 frames. , ppl: 10.07388742477869], batch size: 70
+2022-12-11 07:00:39,366 INFO [train.py:421] (5/8) Epoch 4, batch 30600, loss[loss=2.537, over 1400.00 frames. , ppl: 12.640014064237244] tot_loss[loss=2.31, over 5500877.13 frames. , ppl: 10.07102800246099], batch size: 70
+2022-12-11 07:02:17,072 INFO [train.py:421] (5/8) Epoch 4, batch 30800, loss[loss=3.1, over 560.00 frames. , ppl: 22.19933521262157] tot_loss[loss=2.311, over 5471966.77 frames. , ppl: 10.082394657766493], batch size: 70
+2022-12-11 07:04:02,218 INFO [train.py:421] (5/8) Epoch 4, batch 31000, loss[loss=2.451, over 1890.00 frames. , ppl: 11.598913829418944] tot_loss[loss=2.31, over 5502461.14 frames. , ppl: 10.077117505247042], batch size: 70
+2022-12-11 07:04:02,219 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 07:04:02,984 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.3, over 211138.00 frames. , ppl: 9.976530269338536
+2022-12-11 07:05:48,983 INFO [train.py:421] (5/8) Epoch 4, batch 31200, loss[loss=2.401, over 2520.00 frames. , ppl: 11.038984652268798] tot_loss[loss=2.31, over 5517287.25 frames. , ppl: 10.077358597568992], batch size: 70
+2022-12-11 07:07:34,367 INFO [train.py:421] (5/8) Epoch 4, batch 31400, loss[loss=3.068, over 560.00 frames. , ppl: 21.5055065510881] tot_loss[loss=2.309, over 5569578.36 frames. , ppl: 10.067418271384458], batch size: 70
+2022-12-11 07:09:13,736 INFO [train.py:421] (5/8) Epoch 4, batch 31600, loss[loss=2.584, over 980.00 frames. , ppl: 13.248617348759378] tot_loss[loss=2.309, over 5535791.81 frames. , ppl: 10.066421415297965], batch size: 70
+2022-12-11 07:10:54,271 INFO [train.py:421] (5/8) Epoch 4, batch 31800, loss[loss=2.298, over 3710.00 frames. , ppl: 9.95669132123615] tot_loss[loss=2.308, over 5586091.10 frames. , ppl: 10.057343469557532], batch size: 70
+2022-12-11 07:12:34,096 INFO [train.py:421] (5/8) Epoch 4, batch 32000, loss[loss=2.322, over 1400.00 frames. , ppl: 10.193294326249443] tot_loss[loss=2.308, over 5579700.86 frames. , ppl: 10.057317149911484], batch size: 70
+2022-12-11 07:12:34,097 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 07:12:34,843 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.299, over 211138.00 frames. , ppl: 9.965888179968418
+2022-12-11 07:14:15,382 INFO [train.py:421] (5/8) Epoch 4, batch 32200, loss[loss=2.2, over 6440.00 frames. , ppl: 9.021792497512406] tot_loss[loss=2.309, over 5598729.00 frames. , ppl: 10.060132427322454], batch size: 70
+2022-12-11 07:15:52,097 INFO [train.py:421] (5/8) Epoch 4, batch 32400, loss[loss=2.219, over 5740.00 frames. , ppl: 9.197402926785873] tot_loss[loss=2.309, over 5577425.89 frames. , ppl: 10.063746295788379], batch size: 70
+2022-12-11 07:17:32,126 INFO [train.py:421] (5/8) Epoch 4, batch 32600, loss[loss=2.536, over 1190.00 frames. , ppl: 12.634092390110531] tot_loss[loss=2.31, over 5534457.27 frames. , ppl: 10.070062985163375], batch size: 70
+2022-12-11 07:19:13,615 INFO [train.py:421] (5/8) Epoch 4, batch 32800, loss[loss=2.492, over 1050.00 frames. , ppl: 12.082996303854827] tot_loss[loss=2.309, over 5560397.29 frames. , ppl: 10.06213234472516], batch size: 70
+2022-12-11 07:20:53,531 INFO [train.py:421] (5/8) Epoch 4, batch 33000, loss[loss=2.364, over 3360.00 frames. , ppl: 10.6360522277254] tot_loss[loss=2.31, over 5533387.03 frames. , ppl: 10.070397884276513], batch size: 70
+2022-12-11 07:20:53,531 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 07:20:54,297 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.301, over 211138.00 frames. , ppl: 9.98494896820417
+2022-12-11 07:22:34,966 INFO [train.py:421] (5/8) Epoch 4, batch 33200, loss[loss=2.222, over 7000.00 frames. , ppl: 9.228283107032885] tot_loss[loss=2.309, over 5547651.13 frames. , ppl: 10.063171575289612], batch size: 70
+2022-12-11 07:24:11,958 INFO [train.py:421] (5/8) Epoch 4, batch 33400, loss[loss=2.231, over 4270.00 frames. , ppl: 9.310003343143268] tot_loss[loss=2.308, over 5552552.96 frames. , ppl: 10.058118457265], batch size: 70
+2022-12-11 07:25:47,474 INFO [train.py:421] (5/8) Epoch 4, batch 33600, loss[loss=2.504, over 2030.00 frames. , ppl: 12.236112798938983] tot_loss[loss=2.309, over 5505310.80 frames. , ppl: 10.06763662317883], batch size: 70
+2022-12-11 07:27:27,208 INFO [train.py:421] (5/8) Epoch 4, batch 33800, loss[loss=2.3, over 3500.00 frames. , ppl: 9.977565734626639] tot_loss[loss=2.311, over 5496448.49 frames. , ppl: 10.080327191035881], batch size: 70
+2022-12-11 07:29:07,735 INFO [train.py:421] (5/8) Epoch 4, batch 34000, loss[loss=2.296, over 4760.00 frames. , ppl: 9.931632565843273] tot_loss[loss=2.311, over 5483244.32 frames. , ppl: 10.084504249030777], batch size: 70
+2022-12-11 07:29:07,736 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 07:29:08,495 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.299, over 211138.00 frames. , ppl: 9.963913316067512
+2022-12-11 07:30:52,336 INFO [train.py:421] (5/8) Epoch 4, batch 34200, loss[loss=2.184, over 5810.00 frames. , ppl: 8.881486231038286] tot_loss[loss=2.31, over 5513196.89 frames. , ppl: 10.07611922515373], batch size: 70
+2022-12-11 07:32:31,251 INFO [train.py:421] (5/8) Epoch 4, batch 34400, loss[loss=2.205, over 2870.00 frames. , ppl: 9.070295084724707] tot_loss[loss=2.31, over 5508100.50 frames. , ppl: 10.07615163501459], batch size: 70
+2022-12-11 07:34:12,813 INFO [train.py:421] (5/8) Epoch 4, batch 34600, loss[loss=2.224, over 4410.00 frames. , ppl: 9.24248558760182] tot_loss[loss=2.311, over 5487513.41 frames. , ppl: 10.08197461180827], batch size: 70
+2022-12-11 07:35:50,515 INFO [train.py:421] (5/8) Epoch 4, batch 34800, loss[loss=2.257, over 4270.00 frames. , ppl: 9.556580674449126] tot_loss[loss=2.311, over 5460967.92 frames. , ppl: 10.086723017096718], batch size: 70
+2022-12-11 07:37:31,269 INFO [train.py:421] (5/8) Epoch 4, batch 35000, loss[loss=2.378, over 2310.00 frames. , ppl: 10.78181348934678] tot_loss[loss=2.311, over 5453181.66 frames. , ppl: 10.088634715927183], batch size: 70
+2022-12-11 07:37:31,270 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 07:37:32,028 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.3, over 211138.00 frames. , ppl: 9.97713864747801
+2022-12-11 07:39:12,241 INFO [train.py:421] (5/8) Epoch 4, batch 35200, loss[loss=3.225, over 490.00 frames. , ppl: 25.15427599524565] tot_loss[loss=2.311, over 5462448.42 frames. , ppl: 10.08729766900486], batch size: 70
+2022-12-11 07:40:54,303 INFO [train.py:421] (5/8) Epoch 4, batch 35400, loss[loss=2.138, over 6580.00 frames. , ppl: 8.479487793800528] tot_loss[loss=2.309, over 5527046.37 frames. , ppl: 10.061032911696163], batch size: 70
+2022-12-11 07:42:36,911 INFO [train.py:421] (5/8) Epoch 4, batch 35600, loss[loss=2.338, over 3780.00 frames. , ppl: 10.358965630382691] tot_loss[loss=2.308, over 5581209.32 frames. , ppl: 10.05140248386908], batch size: 70
+2022-12-11 07:44:17,399 INFO [train.py:421] (5/8) Epoch 4, batch 35800, loss[loss=2.417, over 910.00 frames. , ppl: 11.210917635382886] tot_loss[loss=2.307, over 5620314.24 frames. , ppl: 10.041917086003595], batch size: 70
+2022-12-11 07:46:03,633 INFO [train.py:421] (5/8) Epoch 4, batch 36000, loss[loss=2.219, over 5180.00 frames. , ppl: 9.199182380952484] tot_loss[loss=2.305, over 5655028.80 frames. , ppl: 10.023736813040609], batch size: 70
+2022-12-11 07:46:03,633 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 07:46:04,385 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.299, over 211138.00 frames. , ppl: 9.962801429018057
+2022-12-11 07:47:46,400 INFO [train.py:421] (5/8) Epoch 4, batch 36200, loss[loss=2.317, over 1890.00 frames. , ppl: 10.142891413866165] tot_loss[loss=2.305, over 5649115.59 frames. , ppl: 10.024576444449668], batch size: 70
+2022-12-11 07:49:21,365 INFO [train.py:421] (5/8) Epoch 4, batch 36400, loss[loss=2.317, over 2450.00 frames. , ppl: 10.147944831766806] tot_loss[loss=2.306, over 5608920.72 frames. , ppl: 10.032449791617516], batch size: 70
+2022-12-11 07:51:00,542 INFO [train.py:421] (5/8) Epoch 4, batch 36600, loss[loss=2.344, over 2450.00 frames. , ppl: 10.424316717962947] tot_loss[loss=2.307, over 5595362.34 frames. , ppl: 10.039247495872313], batch size: 70
+2022-12-11 07:52:42,134 INFO [train.py:421] (5/8) Epoch 4, batch 36800, loss[loss=2.21, over 5110.00 frames. , ppl: 9.117118184360946] tot_loss[loss=2.307, over 5564191.24 frames. , ppl: 10.046940355802066], batch size: 70
+2022-12-11 07:54:20,219 INFO [train.py:421] (5/8) Epoch 4, batch 37000, loss[loss=2.414, over 1190.00 frames. , ppl: 11.175652479925326] tot_loss[loss=2.306, over 5588958.35 frames. , ppl: 10.038640130766687], batch size: 70
+2022-12-11 07:54:20,220 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 07:54:20,987 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.297, over 211138.00 frames. , ppl: 9.946271967014194
+2022-12-11 07:56:00,678 INFO [train.py:421] (5/8) Epoch 4, batch 37200, loss[loss=2.353, over 2800.00 frames. , ppl: 10.518703880249003] tot_loss[loss=2.306, over 5602358.83 frames. , ppl: 10.03296467038195], batch size: 70
+2022-12-11 07:57:39,427 INFO [train.py:421] (5/8) Epoch 4, batch 37400, loss[loss=2.317, over 3290.00 frames. , ppl: 10.143671435368903] tot_loss[loss=2.306, over 5590947.72 frames. , ppl: 10.03137715089043], batch size: 70
+2022-12-11 07:59:17,911 INFO [train.py:421] (5/8) Epoch 4, batch 37600, loss[loss=2.54, over 1750.00 frames. , ppl: 12.679264124079458] tot_loss[loss=2.306, over 5567126.99 frames. , ppl: 10.038082293195883], batch size: 70
+2022-12-11 08:00:55,523 INFO [train.py:421] (5/8) Epoch 4, batch 37800, loss[loss=2.539, over 1470.00 frames. , ppl: 12.660713831960683] tot_loss[loss=2.307, over 5564354.64 frames. , ppl: 10.045284436540419], batch size: 70
+2022-12-11 08:02:37,239 INFO [train.py:421] (5/8) Epoch 4, batch 38000, loss[loss=2.523, over 980.00 frames. , ppl: 12.467327194682294] tot_loss[loss=2.307, over 5576745.60 frames. , ppl: 10.045743256602298], batch size: 70
+2022-12-11 08:02:37,239 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 08:02:37,986 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.297, over 211138.00 frames. , ppl: 9.945532988975891
+2022-12-11 08:04:17,112 INFO [train.py:421] (5/8) Epoch 4, batch 38200, loss[loss=2.318, over 3220.00 frames. , ppl: 10.152017408206051] tot_loss[loss=2.307, over 5572666.69 frames. , ppl: 10.040643774474544], batch size: 70
+2022-12-11 08:05:57,529 INFO [train.py:421] (5/8) Epoch 4, batch 38400, loss[loss=2.468, over 2240.00 frames. , ppl: 11.793453102323781] tot_loss[loss=2.306, over 5557181.15 frames. , ppl: 10.038119749292788], batch size: 70
+2022-12-11 08:07:37,194 INFO [train.py:421] (5/8) Epoch 4, batch 38600, loss[loss=2.311, over 3290.00 frames. , ppl: 10.08895754236561] tot_loss[loss=2.306, over 5556176.21 frames. , ppl: 10.037036473813973], batch size: 70
+2022-12-11 08:09:17,865 INFO [train.py:421] (5/8) Epoch 4, batch 38800, loss[loss=2.269, over 3990.00 frames. , ppl: 9.666180559005614] tot_loss[loss=2.307, over 5524819.35 frames. , ppl: 10.044800037603164], batch size: 70
+2022-12-11 08:10:56,256 INFO [train.py:421] (5/8) Epoch 4, batch 39000, loss[loss=2.294, over 1680.00 frames. , ppl: 9.914406575942637] tot_loss[loss=2.307, over 5514202.59 frames. , ppl: 10.045122867492939], batch size: 70
+2022-12-11 08:10:56,256 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 08:10:57,005 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.298, over 211138.00 frames. , ppl: 9.949938227560008
+2022-12-11 08:12:35,869 INFO [train.py:421] (5/8) Epoch 4, batch 39200, loss[loss=2.267, over 1820.00 frames. , ppl: 9.649352119683199] tot_loss[loss=2.306, over 5540558.75 frames. , ppl: 10.0336658206963], batch size: 70
+2022-12-11 08:14:16,246 INFO [train.py:421] (5/8) Epoch 4, batch 39400, loss[loss=2.266, over 2100.00 frames. , ppl: 9.638382925309601] tot_loss[loss=2.306, over 5541592.00 frames. , ppl: 10.03168025438866], batch size: 70
+2022-12-11 08:15:56,358 INFO [train.py:421] (5/8) Epoch 4, batch 39600, loss[loss=2.17, over 7910.00 frames. , ppl: 8.757416023347902] tot_loss[loss=2.306, over 5511876.08 frames. , ppl: 10.037737880602533], batch size: 70
+2022-12-11 08:17:35,797 INFO [train.py:421] (5/8) Epoch 4, batch 39800, loss[loss=2.187, over 5250.00 frames. , ppl: 8.912868165239468] tot_loss[loss=2.306, over 5529914.16 frames. , ppl: 10.038314389481581], batch size: 70
+2022-12-11 08:19:17,000 INFO [train.py:421] (5/8) Epoch 4, batch 40000, loss[loss=2.398, over 1890.00 frames. , ppl: 10.99803023446397] tot_loss[loss=2.306, over 5553231.78 frames. , ppl: 10.031829291157958], batch size: 70
+2022-12-11 08:19:17,000 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 08:19:17,760 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.296, over 211138.00 frames. , ppl: 9.93333160275137
+2022-12-11 08:20:58,979 INFO [train.py:421] (5/8) Epoch 4, batch 40200, loss[loss=2.342, over 1400.00 frames. , ppl: 10.399767470229861] tot_loss[loss=2.306, over 5531050.96 frames. , ppl: 10.031333041919334], batch size: 70
+2022-12-11 08:22:39,534 INFO [train.py:421] (5/8) Epoch 4, batch 40400, loss[loss=2.355, over 2800.00 frames. , ppl: 10.533811142464454] tot_loss[loss=2.305, over 5580444.22 frames. , ppl: 10.024076796416677], batch size: 70
+2022-12-11 08:24:20,707 INFO [train.py:421] (5/8) Epoch 4, batch 40600, loss[loss=2.494, over 1820.00 frames. , ppl: 12.114070100729684] tot_loss[loss=2.306, over 5570965.50 frames. , ppl: 10.030794554982934], batch size: 70
+2022-12-11 08:26:03,834 INFO [train.py:421] (5/8) Epoch 4, batch 40800, loss[loss=2.342, over 1680.00 frames. , ppl: 10.40046783622553] tot_loss[loss=2.307, over 5557160.60 frames. , ppl: 10.039341111117144], batch size: 70
+2022-12-11 08:27:44,166 INFO [train.py:421] (5/8) Epoch 4, batch 41000, loss[loss=2.214, over 7910.00 frames. , ppl: 9.148567463147693] tot_loss[loss=2.307, over 5566610.17 frames. , ppl: 10.042069693500983], batch size: 70
+2022-12-11 08:27:44,167 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 08:27:44,909 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.296, over 211138.00 frames. , ppl: 9.931032464771791
+2022-12-11 08:29:23,713 INFO [train.py:421] (5/8) Epoch 4, batch 41200, loss[loss=2.444, over 1750.00 frames. , ppl: 11.515541348652626] tot_loss[loss=2.307, over 5532235.69 frames. , ppl: 10.0449105963364], batch size: 70
+2022-12-11 08:31:02,541 INFO [train.py:421] (5/8) Epoch 4, batch 41400, loss[loss=2.336, over 3710.00 frames. , ppl: 10.33885986210874] tot_loss[loss=2.306, over 5559167.11 frames. , ppl: 10.037483224083894], batch size: 70
+2022-12-11 08:32:46,787 INFO [train.py:421] (5/8) Epoch 4, batch 41600, loss[loss=2.481, over 1540.00 frames. , ppl: 11.957405898852858] tot_loss[loss=2.307, over 5547871.47 frames. , ppl: 10.048799628242428], batch size: 70
+2022-12-11 08:34:26,739 INFO [train.py:421] (5/8) Epoch 4, batch 41800, loss[loss=2.255, over 3640.00 frames. , ppl: 9.5333165462114] tot_loss[loss=2.306, over 5606623.66 frames. , ppl: 10.037454417050776], batch size: 70
+2022-12-11 08:36:06,542 INFO [train.py:421] (5/8) Epoch 4, batch 42000, loss[loss=2.268, over 11900.00 frames. , ppl: 9.658155155660042] tot_loss[loss=2.306, over 5577639.94 frames. , ppl: 10.038944718750113], batch size: 70
+2022-12-11 08:36:06,542 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 08:36:07,272 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.298, over 211138.00 frames. , ppl: 9.952867789688982
+2022-12-11 08:37:48,693 INFO [train.py:421] (5/8) Epoch 4, batch 42200, loss[loss=2.493, over 1750.00 frames. , ppl: 12.094625262214855] tot_loss[loss=2.306, over 5584373.41 frames. , ppl: 10.034593928937939], batch size: 70
+2022-12-11 08:39:25,695 INFO [train.py:421] (5/8) Epoch 4, batch 42400, loss[loss=2.365, over 3570.00 frames. , ppl: 10.642329237446246] tot_loss[loss=2.306, over 5573581.50 frames. , ppl: 10.035598934084339], batch size: 70
+2022-12-11 08:41:06,767 INFO [train.py:421] (5/8) Epoch 4, batch 42600, loss[loss=2.436, over 1120.00 frames. , ppl: 11.424947013113524] tot_loss[loss=2.306, over 5582681.53 frames. , ppl: 10.037120580152841], batch size: 70
+2022-12-11 08:42:47,990 INFO [train.py:421] (5/8) Epoch 4, batch 42800, loss[loss=2.442, over 700.00 frames. , ppl: 11.497944928405301] tot_loss[loss=2.306, over 5584950.49 frames. , ppl: 10.029862293629263], batch size: 70
+2022-12-11 08:44:22,401 INFO [train.py:421] (5/8) Epoch 4, batch 43000, loss[loss=2.223, over 5320.00 frames. , ppl: 9.23810958033033] tot_loss[loss=2.305, over 5603923.66 frames. , ppl: 10.027843828772097], batch size: 70
+2022-12-11 08:44:22,401 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 08:44:23,129 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.296, over 211138.00 frames. , ppl: 9.932262819602796
+2022-12-11 08:46:01,985 INFO [train.py:421] (5/8) Epoch 4, batch 43200, loss[loss=2.733, over 910.00 frames. , ppl: 15.379083148557102] tot_loss[loss=2.305, over 5596349.62 frames. , ppl: 10.028158296564909], batch size: 70
+2022-12-11 08:47:41,483 INFO [train.py:421] (5/8) Epoch 4, batch 43400, loss[loss=2.279, over 3360.00 frames. , ppl: 9.769317370523252] tot_loss[loss=2.306, over 5593103.71 frames. , ppl: 10.03719179823785], batch size: 70
+2022-12-11 08:49:23,909 INFO [train.py:421] (5/8) Epoch 4, batch 43600, loss[loss=2.232, over 4830.00 frames. , ppl: 9.321981093944263] tot_loss[loss=2.306, over 5606745.61 frames. , ppl: 10.034224224034471], batch size: 70
+2022-12-11 08:51:04,765 INFO [train.py:421] (5/8) Epoch 4, batch 43800, loss[loss=2.27, over 4410.00 frames. , ppl: 9.682245996526389] tot_loss[loss=2.306, over 5581486.12 frames. , ppl: 10.036019301200039], batch size: 70
+2022-12-11 08:52:41,190 INFO [train.py:421] (5/8) Epoch 4, batch 44000, loss[loss=2.463, over 1050.00 frames. , ppl: 11.744083823132497] tot_loss[loss=2.307, over 5531193.72 frames. , ppl: 10.046712056744727], batch size: 70
+2022-12-11 08:52:41,191 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 08:52:41,937 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.296, over 211138.00 frames. , ppl: 9.938447770769514
+2022-12-11 08:54:20,560 INFO [train.py:421] (5/8) Epoch 4, batch 44200, loss[loss=2.266, over 2660.00 frames. , ppl: 9.644592917755842] tot_loss[loss=2.307, over 5540933.13 frames. , ppl: 10.047103008031062], batch size: 70
+2022-12-11 08:55:59,749 INFO [train.py:421] (5/8) Epoch 4, batch 44400, loss[loss=2.258, over 9590.00 frames. , ppl: 9.562293183385748] tot_loss[loss=2.307, over 5536284.37 frames. , ppl: 10.046803361873982], batch size: 70
+2022-12-11 08:57:40,777 INFO [train.py:421] (5/8) Epoch 4, batch 44600, loss[loss=2.907, over 630.00 frames. , ppl: 18.298443104524956] tot_loss[loss=2.307, over 5540086.23 frames. , ppl: 10.046709752132006], batch size: 70
+2022-12-11 08:59:24,320 INFO [train.py:421] (5/8) Epoch 4, batch 44800, loss[loss=2.438, over 1120.00 frames. , ppl: 11.451046713870113] tot_loss[loss=2.307, over 5555551.89 frames. , ppl: 10.049211626828404], batch size: 70
+2022-12-11 09:01:09,940 INFO [train.py:421] (5/8) Epoch 4, batch 45000, loss[loss=2.321, over 2940.00 frames. , ppl: 10.184756344134932] tot_loss[loss=2.307, over 5579241.80 frames. , ppl: 10.040667764967344], batch size: 70
+2022-12-11 09:01:09,941 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 09:01:10,691 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.293, over 211138.00 frames. , ppl: 9.909560376874897
+2022-12-11 09:02:55,711 INFO [train.py:421] (5/8) Epoch 4, batch 45200, loss[loss=2.205, over 5180.00 frames. , ppl: 9.066009087800328] tot_loss[loss=2.306, over 5611608.10 frames. , ppl: 10.037681067081515], batch size: 70
+2022-12-11 09:04:33,451 INFO [train.py:421] (5/8) Epoch 4, batch 45400, loss[loss=2.249, over 5320.00 frames. , ppl: 9.481106101576156] tot_loss[loss=2.307, over 5586478.03 frames. , ppl: 10.045116267276134], batch size: 70
+2022-12-11 09:06:16,175 INFO [train.py:421] (5/8) Epoch 4, batch 45600, loss[loss=2.236, over 5950.00 frames. , ppl: 9.351365916017258] tot_loss[loss=2.308, over 5578397.36 frames. , ppl: 10.04958363181509], batch size: 70
+2022-12-11 09:07:57,232 INFO [train.py:421] (5/8) Epoch 4, batch 45800, loss[loss=2.388, over 1820.00 frames. , ppl: 10.894854788348846] tot_loss[loss=2.307, over 5570374.61 frames. , ppl: 10.045469519463605], batch size: 70
+2022-12-11 09:09:38,462 INFO [train.py:421] (5/8) Epoch 4, batch 46000, loss[loss=2.159, over 6580.00 frames. , ppl: 8.662356074659394] tot_loss[loss=2.307, over 5548540.61 frames. , ppl: 10.045976935067586], batch size: 70
+2022-12-11 09:09:38,462 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 09:09:39,192 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.295, over 211138.00 frames. , ppl: 9.921327269150712
+2022-12-11 09:11:18,479 INFO [train.py:421] (5/8) Epoch 4, batch 46200, loss[loss=2.615, over 910.00 frames. , ppl: 13.66891637885018] tot_loss[loss=2.308, over 5522915.10 frames. , ppl: 10.051032785834296], batch size: 70
+2022-12-11 09:12:54,786 INFO [train.py:421] (5/8) Epoch 4, batch 46400, loss[loss=2.391, over 1820.00 frames. , ppl: 10.929669915896172] tot_loss[loss=2.308, over 5516290.90 frames. , ppl: 10.053841748323892], batch size: 70
+2022-12-11 09:14:38,877 INFO [train.py:421] (5/8) Epoch 4, batch 46600, loss[loss=2.244, over 4760.00 frames. , ppl: 9.430510551183605] tot_loss[loss=2.306, over 5582425.05 frames. , ppl: 10.038187748716785], batch size: 70
+2022-12-11 09:16:20,806 INFO [train.py:421] (5/8) Epoch 4, batch 46800, loss[loss=2.314, over 1400.00 frames. , ppl: 10.114352218491444] tot_loss[loss=2.306, over 5578583.77 frames. , ppl: 10.036679133602298], batch size: 70
+2022-12-11 09:18:01,883 INFO [train.py:421] (5/8) Epoch 4, batch 47000, loss[loss=2.594, over 1050.00 frames. , ppl: 13.384824414144353] tot_loss[loss=2.306, over 5591079.49 frames. , ppl: 10.030295614936916], batch size: 70
+2022-12-11 09:18:01,884 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 09:18:02,643 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.912724529894838
+2022-12-11 09:19:43,173 INFO [train.py:421] (5/8) Epoch 4, batch 47200, loss[loss=2.192, over 3010.00 frames. , ppl: 8.952790389270259] tot_loss[loss=2.305, over 5569360.44 frames. , ppl: 10.028437275267624], batch size: 70
+2022-12-11 09:21:18,453 INFO [train.py:421] (5/8) Epoch 4, batch 47400, loss[loss=2.497, over 980.00 frames. , ppl: 12.15207334250006] tot_loss[loss=2.307, over 5496722.42 frames. , ppl: 10.044207537202718], batch size: 70
+2022-12-11 09:22:56,929 INFO [train.py:421] (5/8) Epoch 4, batch 47600, loss[loss=2.293, over 4060.00 frames. , ppl: 9.900065007239098] tot_loss[loss=2.309, over 5426002.33 frames. , ppl: 10.060994299680836], batch size: 70
+2022-12-11 09:24:37,051 INFO [train.py:421] (5/8) Epoch 4, batch 47800, loss[loss=2.391, over 1750.00 frames. , ppl: 10.926702263216566] tot_loss[loss=2.308, over 5444926.51 frames. , ppl: 10.053814675485372], batch size: 70
+2022-12-11 09:26:19,532 INFO [train.py:421] (5/8) Epoch 4, batch 48000, loss[loss=2.305, over 3290.00 frames. , ppl: 10.023082453408254] tot_loss[loss=2.307, over 5474775.10 frames. , ppl: 10.046405436263251], batch size: 70
+2022-12-11 09:26:19,533 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 09:26:20,321 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.295, over 211138.00 frames. , ppl: 9.92101743515498
+2022-12-11 09:28:03,643 INFO [train.py:421] (5/8) Epoch 4, batch 48200, loss[loss=2.497, over 1400.00 frames. , ppl: 12.143602104762396] tot_loss[loss=2.307, over 5473582.22 frames. , ppl: 10.045632988490086], batch size: 70
+2022-12-11 09:29:44,757 INFO [train.py:421] (5/8) Epoch 4, batch 48400, loss[loss=2.337, over 2450.00 frames. , ppl: 10.34681652534592] tot_loss[loss=2.307, over 5482150.05 frames. , ppl: 10.041454663377914], batch size: 70
+2022-12-11 09:31:30,701 INFO [train.py:421] (5/8) Epoch 4, batch 48600, loss[loss=2.188, over 6020.00 frames. , ppl: 8.913649070151344] tot_loss[loss=2.306, over 5523946.27 frames. , ppl: 10.030839503934324], batch size: 70
+2022-12-11 09:33:10,106 INFO [train.py:421] (5/8) Epoch 4, batch 48800, loss[loss=2.35, over 2590.00 frames. , ppl: 10.484591255321734] tot_loss[loss=2.304, over 5567136.86 frames. , ppl: 10.017977719875988], batch size: 70
+2022-12-11 09:34:51,409 INFO [train.py:421] (5/8) Epoch 4, batch 49000, loss[loss=2.479, over 1750.00 frames. , ppl: 11.927378782955282] tot_loss[loss=2.304, over 5576641.09 frames. , ppl: 10.017736787168332], batch size: 70
+2022-12-11 09:34:51,409 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 09:34:52,168 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.912736267159561
+2022-12-11 09:36:32,383 INFO [train.py:421] (5/8) Epoch 4, batch 49200, loss[loss=2.356, over 1330.00 frames. , ppl: 10.553684597853383] tot_loss[loss=2.306, over 5520511.30 frames. , ppl: 10.035640045750036], batch size: 70
+2022-12-11 09:38:12,918 INFO [train.py:421] (5/8) Epoch 4, batch 49400, loss[loss=2.196, over 2940.00 frames. , ppl: 8.989543138401396] tot_loss[loss=2.307, over 5512247.75 frames. , ppl: 10.042489852023834], batch size: 70
+2022-12-11 09:39:55,521 INFO [train.py:421] (5/8) Epoch 4, batch 49600, loss[loss=2.329, over 1540.00 frames. , ppl: 10.269545124976384] tot_loss[loss=2.305, over 5571858.20 frames. , ppl: 10.0218025167115], batch size: 70
+2022-12-11 09:41:34,985 INFO [train.py:421] (5/8) Epoch 4, batch 49800, loss[loss=2.645, over 840.00 frames. , ppl: 14.086643103375987] tot_loss[loss=2.304, over 5595218.60 frames. , ppl: 10.01699584429696], batch size: 70
+2022-12-11 09:43:18,273 INFO [train.py:421] (5/8) Epoch 4, batch 50000, loss[loss=2.33, over 2100.00 frames. , ppl: 10.282064720626238] tot_loss[loss=2.304, over 5586911.48 frames. , ppl: 10.016384361538368], batch size: 70
+2022-12-11 09:43:18,274 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 09:43:19,051 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.911439383482614
+2022-12-11 09:44:57,846 INFO [train.py:421] (5/8) Epoch 4, batch 50200, loss[loss=2.32, over 3080.00 frames. , ppl: 10.175540735738211] tot_loss[loss=2.306, over 5543338.47 frames. , ppl: 10.033191355552713], batch size: 70
+2022-12-11 09:46:40,422 INFO [train.py:421] (5/8) Epoch 4, batch 50400, loss[loss=2.418, over 2030.00 frames. , ppl: 11.225512368576533] tot_loss[loss=2.307, over 5545106.48 frames. , ppl: 10.039292678485094], batch size: 70
+2022-12-11 09:48:16,334 INFO [train.py:421] (5/8) Epoch 4, batch 50600, loss[loss=3.593, over 420.00 frames. , ppl: 36.35335233254622] tot_loss[loss=2.307, over 5509345.54 frames. , ppl: 10.04339883572311], batch size: 70
+2022-12-11 09:49:58,440 INFO [train.py:421] (5/8) Epoch 4, batch 50800, loss[loss=2.276, over 4060.00 frames. , ppl: 9.739930016067461] tot_loss[loss=2.305, over 5553556.56 frames. , ppl: 10.024171779090574], batch size: 70
+2022-12-11 09:51:36,250 INFO [train.py:421] (5/8) Epoch 4, batch 51000, loss[loss=2.837, over 630.00 frames. , ppl: 17.057194369099733] tot_loss[loss=2.305, over 5587443.34 frames. , ppl: 10.022347844762372], batch size: 70
+2022-12-11 09:51:36,250 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 09:51:36,999 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.293, over 211138.00 frames. , ppl: 9.909437175772835
+2022-12-11 09:53:14,568 INFO [train.py:421] (5/8) Epoch 4, batch 51200, loss[loss=2.325, over 2380.00 frames. , ppl: 10.229920077090911] tot_loss[loss=2.305, over 5554627.64 frames. , ppl: 10.028247407491726], batch size: 70
+2022-12-11 09:54:54,289 INFO [train.py:421] (5/8) Epoch 4, batch 51400, loss[loss=2.265, over 2800.00 frames. , ppl: 9.626402906925412] tot_loss[loss=2.306, over 5539857.81 frames. , ppl: 10.038320925030067], batch size: 70
+2022-12-11 09:56:32,680 INFO [train.py:421] (5/8) Epoch 4, batch 51600, loss[loss=2.409, over 2800.00 frames. , ppl: 11.127757492466616] tot_loss[loss=2.307, over 5512073.65 frames. , ppl: 10.048306544215437], batch size: 70
+2022-12-11 09:58:13,022 INFO [train.py:421] (5/8) Epoch 4, batch 51800, loss[loss=2.546, over 1260.00 frames. , ppl: 12.752124605084834] tot_loss[loss=2.308, over 5496002.41 frames. , ppl: 10.053495243425978], batch size: 70
+2022-12-11 09:59:53,302 INFO [train.py:421] (5/8) Epoch 4, batch 52000, loss[loss=2.284, over 4270.00 frames. , ppl: 9.812655020918603] tot_loss[loss=2.309, over 5455115.50 frames. , ppl: 10.062252219850793], batch size: 70
+2022-12-11 09:59:53,303 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 09:59:54,056 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.914090547381605
+2022-12-11 10:01:33,675 INFO [train.py:421] (5/8) Epoch 4, batch 52200, loss[loss=2.433, over 2030.00 frames. , ppl: 11.397769519930453] tot_loss[loss=2.311, over 5422423.07 frames. , ppl: 10.08112816914543], batch size: 70
+2022-12-11 10:03:14,832 INFO [train.py:421] (5/8) Epoch 4, batch 52400, loss[loss=2.442, over 1610.00 frames. , ppl: 11.49913496284367] tot_loss[loss=2.31, over 5438632.77 frames. , ppl: 10.077196036987488], batch size: 70
+2022-12-11 10:04:57,998 INFO [train.py:421] (5/8) Epoch 4, batch 52600, loss[loss=2.363, over 1890.00 frames. , ppl: 10.617878902596084] tot_loss[loss=2.31, over 5445660.13 frames. , ppl: 10.072924506648002], batch size: 70
+2022-12-11 10:06:38,642 INFO [train.py:421] (5/8) Epoch 4, batch 52800, loss[loss=2.303, over 2520.00 frames. , ppl: 10.000624784020928] tot_loss[loss=2.307, over 5503957.98 frames. , ppl: 10.044323628447037], batch size: 70
+2022-12-11 10:08:18,166 INFO [train.py:421] (5/8) Epoch 4, batch 53000, loss[loss=2.222, over 6790.00 frames. , ppl: 9.228883824509944] tot_loss[loss=2.306, over 5522024.48 frames. , ppl: 10.03655964224228], batch size: 70
+2022-12-11 10:08:18,167 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 10:08:18,896 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.296, over 211138.00 frames. , ppl: 9.929699384943623
+2022-12-11 10:09:58,043 INFO [train.py:421] (5/8) Epoch 4, batch 53200, loss[loss=2.375, over 1470.00 frames. , ppl: 10.747192787188897] tot_loss[loss=2.307, over 5507325.48 frames. , ppl: 10.04497417783594], batch size: 70
+2022-12-11 10:11:38,324 INFO [train.py:421] (5/8) Epoch 4, batch 53400, loss[loss=2.269, over 3920.00 frames. , ppl: 9.673799394227158] tot_loss[loss=2.308, over 5476779.40 frames. , ppl: 10.052860759148947], batch size: 70
+2022-12-11 10:13:16,442 INFO [train.py:421] (5/8) Epoch 4, batch 53600, loss[loss=2.523, over 840.00 frames. , ppl: 12.470356849595108] tot_loss[loss=2.308, over 5491998.88 frames. , ppl: 10.05086302671076], batch size: 70
+2022-12-11 10:14:54,533 INFO [train.py:421] (5/8) Epoch 4, batch 53800, loss[loss=2.184, over 4900.00 frames. , ppl: 8.879032532632374] tot_loss[loss=2.308, over 5472032.70 frames. , ppl: 10.057460822569016], batch size: 70
+2022-12-11 10:16:36,543 INFO [train.py:421] (5/8) Epoch 4, batch 54000, loss[loss=2.496, over 1540.00 frames. , ppl: 12.134244813121514] tot_loss[loss=2.308, over 5457471.10 frames. , ppl: 10.057240215231538], batch size: 70
+2022-12-11 10:16:36,543 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 10:16:37,305 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.293, over 211138.00 frames. , ppl: 9.903191128032574
+2022-12-11 10:18:18,565 INFO [train.py:421] (5/8) Epoch 4, batch 54200, loss[loss=2.348, over 2240.00 frames. , ppl: 10.460065256261306] tot_loss[loss=2.309, over 5438204.87 frames. , ppl: 10.06314532503589], batch size: 70
+2022-12-11 10:19:56,516 INFO [train.py:421] (5/8) Epoch 4, batch 54400, loss[loss=2.382, over 980.00 frames. , ppl: 10.82494158825565] tot_loss[loss=2.308, over 5442296.49 frames. , ppl: 10.056820766833553], batch size: 70
+2022-12-11 10:21:35,656 INFO [train.py:421] (5/8) Epoch 4, batch 54600, loss[loss=2.922, over 560.00 frames. , ppl: 18.575149629514097] tot_loss[loss=2.31, over 5403612.95 frames. , ppl: 10.071858759272141], batch size: 70
+2022-12-11 10:23:17,361 INFO [train.py:421] (5/8) Epoch 4, batch 54800, loss[loss=2.212, over 5670.00 frames. , ppl: 9.129787168943023] tot_loss[loss=2.31, over 5412871.27 frames. , ppl: 10.072263003087865], batch size: 70
+2022-12-11 10:24:58,185 INFO [train.py:421] (5/8) Epoch 4, batch 55000, loss[loss=2.509, over 980.00 frames. , ppl: 12.298406077338342] tot_loss[loss=2.309, over 5448655.44 frames. , ppl: 10.062579260973747], batch size: 70
+2022-12-11 10:24:58,185 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 10:24:58,943 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.91692148777633
+2022-12-11 10:26:38,690 INFO [train.py:421] (5/8) Epoch 4, batch 55200, loss[loss=2.331, over 1890.00 frames. , ppl: 10.28323332421939] tot_loss[loss=2.309, over 5414584.24 frames. , ppl: 10.0656535170689], batch size: 70
+2022-12-11 10:28:18,411 INFO [train.py:421] (5/8) Epoch 4, batch 55400, loss[loss=2.247, over 8120.00 frames. , ppl: 9.459284155300326] tot_loss[loss=2.309, over 5391770.68 frames. , ppl: 10.068843714980469], batch size: 70
+2022-12-11 10:29:59,489 INFO [train.py:421] (5/8) Epoch 4, batch 55600, loss[loss=2.246, over 4130.00 frames. , ppl: 9.45372372761463] tot_loss[loss=2.311, over 5327263.38 frames. , ppl: 10.083469217254569], batch size: 70
+2022-12-11 10:31:38,226 INFO [train.py:421] (5/8) Epoch 4, batch 55800, loss[loss=2.593, over 840.00 frames. , ppl: 13.364710021923974] tot_loss[loss=2.311, over 5300717.40 frames. , ppl: 10.089210593271998], batch size: 70
+2022-12-11 10:33:18,394 INFO [train.py:421] (5/8) Epoch 4, batch 56000, loss[loss=2.262, over 4690.00 frames. , ppl: 9.598436988515068] tot_loss[loss=2.311, over 5289241.73 frames. , ppl: 10.088255862614899], batch size: 70
+2022-12-11 10:33:18,394 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 10:33:19,123 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.91445299177917
+2022-12-11 10:34:53,562 INFO [train.py:421] (5/8) Epoch 4, batch 56200, loss[loss=2.828, over 700.00 frames. , ppl: 16.91707768886012] tot_loss[loss=2.311, over 5328146.50 frames. , ppl: 10.079870064530734], batch size: 70
+2022-12-11 10:36:32,168 INFO [train.py:421] (5/8) Epoch 4, batch 56400, loss[loss=2.317, over 2450.00 frames. , ppl: 10.14254223090681] tot_loss[loss=2.311, over 5325771.92 frames. , ppl: 10.079574593661073], batch size: 70
+2022-12-11 10:38:13,265 INFO [train.py:421] (5/8) Epoch 4, batch 56600, loss[loss=3.222, over 560.00 frames. , ppl: 25.07580797680408] tot_loss[loss=2.31, over 5370694.07 frames. , ppl: 10.07343680336641], batch size: 70
+2022-12-11 10:39:51,813 INFO [train.py:421] (5/8) Epoch 4, batch 56800, loss[loss=2.297, over 4410.00 frames. , ppl: 9.942185500714851] tot_loss[loss=2.311, over 5353490.96 frames. , ppl: 10.079540924448803], batch size: 70
+2022-12-11 10:41:31,980 INFO [train.py:421] (5/8) Epoch 4, batch 57000, loss[loss=2.389, over 1960.00 frames. , ppl: 10.905090096801787] tot_loss[loss=2.309, over 5398959.29 frames. , ppl: 10.062951886980914], batch size: 70
+2022-12-11 10:41:31,981 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 10:41:32,726 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.293, over 211138.00 frames. , ppl: 9.90575504999001
+2022-12-11 10:43:12,819 INFO [train.py:421] (5/8) Epoch 4, batch 57200, loss[loss=2.24, over 2800.00 frames. , ppl: 9.392622031218666] tot_loss[loss=2.308, over 5423340.24 frames. , ppl: 10.050649771193047], batch size: 70
+2022-12-11 10:44:55,968 INFO [train.py:421] (5/8) Epoch 4, batch 57400, loss[loss=2.417, over 1330.00 frames. , ppl: 11.216951195281545] tot_loss[loss=2.308, over 5425575.52 frames. , ppl: 10.054320493235428], batch size: 70
+2022-12-11 10:46:33,832 INFO [train.py:421] (5/8) Epoch 4, batch 57600, loss[loss=2.68, over 840.00 frames. , ppl: 14.583858932668727] tot_loss[loss=2.308, over 5421073.01 frames. , ppl: 10.04949624945697], batch size: 70
+2022-12-11 10:48:10,496 INFO [train.py:421] (5/8) Epoch 4, batch 57800, loss[loss=2.399, over 3220.00 frames. , ppl: 11.009207687439027] tot_loss[loss=2.307, over 5458723.67 frames. , ppl: 10.04766278058551], batch size: 70
+2022-12-11 10:49:55,349 INFO [train.py:421] (5/8) Epoch 4, batch 58000, loss[loss=2.277, over 1610.00 frames. , ppl: 9.744233006933381] tot_loss[loss=2.306, over 5491309.23 frames. , ppl: 10.03800908291007], batch size: 70
+2022-12-11 10:49:55,350 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 10:49:56,110 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.915101609561932
+2022-12-11 10:51:34,570 INFO [train.py:421] (5/8) Epoch 4, batch 58200, loss[loss=2.474, over 1400.00 frames. , ppl: 11.872904353080042] tot_loss[loss=2.305, over 5506287.62 frames. , ppl: 10.028856441924837], batch size: 70
+2022-12-11 10:53:13,740 INFO [train.py:421] (5/8) Epoch 4, batch 58400, loss[loss=2.524, over 770.00 frames. , ppl: 12.484148505041823] tot_loss[loss=2.305, over 5527214.96 frames. , ppl: 10.028049259976292], batch size: 70
+2022-12-11 10:54:51,409 INFO [train.py:421] (5/8) Epoch 4, batch 58600, loss[loss=2.284, over 4200.00 frames. , ppl: 9.817232206213838] tot_loss[loss=2.305, over 5533217.80 frames. , ppl: 10.02729924569803], batch size: 70
+2022-12-11 10:56:31,643 INFO [train.py:421] (5/8) Epoch 4, batch 58800, loss[loss=2.184, over 6930.00 frames. , ppl: 8.885236371313415] tot_loss[loss=2.306, over 5508873.31 frames. , ppl: 10.036237430656918], batch size: 70
+2022-12-11 10:58:12,932 INFO [train.py:421] (5/8) Epoch 4, batch 59000, loss[loss=2.534, over 910.00 frames. , ppl: 12.607333710860363] tot_loss[loss=2.305, over 5565243.90 frames. , ppl: 10.019870750707293], batch size: 70
+2022-12-11 10:58:12,932 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 10:58:13,677 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.911222274742915
+2022-12-11 10:59:52,976 INFO [train.py:421] (5/8) Epoch 4, batch 59200, loss[loss=2.802, over 700.00 frames. , ppl: 16.471395641751137] tot_loss[loss=2.305, over 5522791.95 frames. , ppl: 10.027687673232155], batch size: 70
+2022-12-11 11:01:31,053 INFO [train.py:421] (5/8) Epoch 4, batch 59400, loss[loss=2.221, over 8050.00 frames. , ppl: 9.213335370009629] tot_loss[loss=2.305, over 5531926.07 frames. , ppl: 10.024077745264492], batch size: 70
+2022-12-11 11:03:11,584 INFO [train.py:421] (5/8) Epoch 4, batch 59600, loss[loss=2.376, over 2100.00 frames. , ppl: 10.763969798612173] tot_loss[loss=2.306, over 5490022.26 frames. , ppl: 10.034723490510398], batch size: 70
+2022-12-11 11:04:50,775 INFO [train.py:421] (5/8) Epoch 4, batch 59800, loss[loss=2.262, over 4410.00 frames. , ppl: 9.60528999733443] tot_loss[loss=2.306, over 5505462.36 frames. , ppl: 10.032917337801702], batch size: 70
+2022-12-11 11:06:31,798 INFO [train.py:421] (5/8) Epoch 4, batch 60000, loss[loss=2.437, over 1540.00 frames. , ppl: 11.435855367216083] tot_loss[loss=2.306, over 5489398.49 frames. , ppl: 10.039160218211636], batch size: 70
+2022-12-11 11:06:31,798 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 11:06:32,558 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.293, over 211138.00 frames. , ppl: 9.908699467886942
+2022-12-11 11:08:12,354 INFO [train.py:421] (5/8) Epoch 4, batch 60200, loss[loss=2.37, over 2590.00 frames. , ppl: 10.695314445203845] tot_loss[loss=2.306, over 5511650.25 frames. , ppl: 10.034542506112368], batch size: 70
+2022-12-11 11:09:54,641 INFO [train.py:421] (5/8) Epoch 4, batch 60400, loss[loss=2.358, over 2170.00 frames. , ppl: 10.57240180776672] tot_loss[loss=2.306, over 5514178.84 frames. , ppl: 10.033944936952809], batch size: 70
+2022-12-11 11:11:33,948 INFO [train.py:421] (5/8) Epoch 4, batch 60600, loss[loss=2.245, over 6090.00 frames. , ppl: 9.435956597212618] tot_loss[loss=2.305, over 5545087.73 frames. , ppl: 10.021431337695047], batch size: 70
+2022-12-11 11:13:09,650 INFO [train.py:421] (5/8) Epoch 4, batch 60800, loss[loss=2.45, over 1330.00 frames. , ppl: 11.592567867862119] tot_loss[loss=2.306, over 5516540.70 frames. , ppl: 10.030560950854854], batch size: 70
+2022-12-11 11:14:48,151 INFO [train.py:421] (5/8) Epoch 4, batch 61000, loss[loss=2.566, over 980.00 frames. , ppl: 13.017714516186052] tot_loss[loss=2.305, over 5514367.29 frames. , ppl: 10.026158406330541], batch size: 70
+2022-12-11 11:14:48,152 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 11:14:48,898 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.292, over 211138.00 frames. , ppl: 9.890057963547763
+2022-12-11 11:16:30,418 INFO [train.py:421] (5/8) Epoch 4, batch 61200, loss[loss=2.88, over 700.00 frames. , ppl: 17.81455277244978] tot_loss[loss=2.304, over 5569317.32 frames. , ppl: 10.013783607823797], batch size: 70
+2022-12-11 11:18:08,920 INFO [train.py:421] (5/8) Epoch 4, batch 61400, loss[loss=2.262, over 3220.00 frames. , ppl: 9.605076858245814] tot_loss[loss=2.304, over 5537474.37 frames. , ppl: 10.017919636765544], batch size: 70
+2022-12-11 11:19:47,063 INFO [train.py:421] (5/8) Epoch 4, batch 61600, loss[loss=2.258, over 4480.00 frames. , ppl: 9.56729262673641] tot_loss[loss=2.304, over 5556502.34 frames. , ppl: 10.011963992060485], batch size: 70
+2022-12-11 11:21:28,011 INFO [train.py:421] (5/8) Epoch 4, batch 61800, loss[loss=2.427, over 1820.00 frames. , ppl: 11.326242900311854] tot_loss[loss=2.304, over 5557421.23 frames. , ppl: 10.011963212327073], batch size: 70
+2022-12-11 11:23:08,299 INFO [train.py:421] (5/8) Epoch 4, batch 62000, loss[loss=2.186, over 4760.00 frames. , ppl: 8.897441195884044] tot_loss[loss=2.303, over 5556576.07 frames. , ppl: 10.005774476686284], batch size: 70
+2022-12-11 11:23:08,299 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 11:23:09,028 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.916335861485702
+2022-12-11 11:24:49,074 INFO [train.py:421] (5/8) Epoch 4, batch 62200, loss[loss=2.148, over 7350.00 frames. , ppl: 8.57021604871399] tot_loss[loss=2.304, over 5517722.53 frames. , ppl: 10.012760910628556], batch size: 70
+2022-12-11 11:26:31,323 INFO [train.py:421] (5/8) Epoch 4, batch 62400, loss[loss=2.523, over 1540.00 frames. , ppl: 12.469282087841437] tot_loss[loss=2.304, over 5497691.92 frames. , ppl: 10.015231168227036], batch size: 70
+2022-12-11 11:28:12,734 INFO [train.py:421] (5/8) Epoch 4, batch 62600, loss[loss=2.356, over 2170.00 frames. , ppl: 10.545703396621176] tot_loss[loss=2.304, over 5521069.11 frames. , ppl: 10.018292799929876], batch size: 70
+2022-12-11 11:29:56,768 INFO [train.py:421] (5/8) Epoch 4, batch 62800, loss[loss=2.242, over 4270.00 frames. , ppl: 9.41403087054057] tot_loss[loss=2.303, over 5558213.85 frames. , ppl: 10.008310434037309], batch size: 70
+2022-12-11 11:31:39,087 INFO [train.py:421] (5/8) Epoch 4, batch 63000, loss[loss=2.507, over 1190.00 frames. , ppl: 12.264996102057996] tot_loss[loss=2.304, over 5553776.78 frames. , ppl: 10.009879951730232], batch size: 70
+2022-12-11 11:31:39,088 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 11:31:39,834 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.912125947821703
+2022-12-11 11:33:18,823 INFO [train.py:421] (5/8) Epoch 4, batch 63200, loss[loss=2.269, over 5320.00 frames. , ppl: 9.66590021317383] tot_loss[loss=2.304, over 5555609.59 frames. , ppl: 10.01163906577024], batch size: 70
+2022-12-11 11:34:57,559 INFO [train.py:421] (5/8) Epoch 4, batch 63400, loss[loss=2.314, over 2590.00 frames. , ppl: 10.11005888924745] tot_loss[loss=2.303, over 5570644.47 frames. , ppl: 10.00828718426682], batch size: 70
+2022-12-11 11:36:40,497 INFO [train.py:421] (5/8) Epoch 4, batch 63600, loss[loss=2.219, over 4340.00 frames. , ppl: 9.194500565975165] tot_loss[loss=2.302, over 5597857.15 frames. , ppl: 9.997158412344174], batch size: 70
+2022-12-11 11:38:20,356 INFO [train.py:421] (5/8) Epoch 4, batch 63800, loss[loss=2.518, over 1050.00 frames. , ppl: 12.400043278389708] tot_loss[loss=2.303, over 5560144.58 frames. , ppl: 10.004225496494085], batch size: 70
+2022-12-11 11:39:59,621 INFO [train.py:421] (5/8) Epoch 4, batch 64000, loss[loss=2.322, over 1260.00 frames. , ppl: 10.198691467933388] tot_loss[loss=2.306, over 5482529.82 frames. , ppl: 10.029254937587014], batch size: 70
+2022-12-11 11:39:59,622 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 11:40:00,404 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.292, over 211138.00 frames. , ppl: 9.899215356849181
+2022-12-11 11:41:39,829 INFO [train.py:421] (5/8) Epoch 4, batch 64200, loss[loss=3.199, over 490.00 frames. , ppl: 24.499952467710667] tot_loss[loss=2.305, over 5505590.06 frames. , ppl: 10.024490333887563], batch size: 70
+2022-12-11 11:43:17,850 INFO [train.py:421] (5/8) Epoch 4, batch 64400, loss[loss=2.538, over 1050.00 frames. , ppl: 12.651587964620512] tot_loss[loss=2.305, over 5499899.34 frames. , ppl: 10.023096539313656], batch size: 70
+2022-12-11 11:44:57,183 INFO [train.py:421] (5/8) Epoch 4, batch 64600, loss[loss=3.647, over 420.00 frames. , ppl: 38.34998329660897] tot_loss[loss=2.305, over 5504167.49 frames. , ppl: 10.02852068155978], batch size: 70
+2022-12-11 11:46:37,646 INFO [train.py:421] (5/8) Epoch 4, batch 64800, loss[loss=2.444, over 1050.00 frames. , ppl: 11.520259059942667] tot_loss[loss=2.305, over 5530446.81 frames. , ppl: 10.025231904334168], batch size: 70
+2022-12-11 11:48:19,715 INFO [train.py:421] (5/8) Epoch 4, batch 65000, loss[loss=2.468, over 1540.00 frames. , ppl: 11.795599188360516] tot_loss[loss=2.304, over 5545299.87 frames. , ppl: 10.017590043121631], batch size: 70
+2022-12-11 11:48:19,716 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 11:48:20,473 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.916086357031197
+2022-12-11 11:50:01,972 INFO [train.py:421] (5/8) Epoch 4, batch 65200, loss[loss=2.389, over 1400.00 frames. , ppl: 10.906427886711507] tot_loss[loss=2.305, over 5508340.33 frames. , ppl: 10.028840982615499], batch size: 70
+2022-12-11 11:51:41,818 INFO [train.py:421] (5/8) Epoch 4, batch 65400, loss[loss=2.254, over 4340.00 frames. , ppl: 9.528749826055735] tot_loss[loss=2.306, over 5477389.06 frames. , ppl: 10.03324113571649], batch size: 70
+2022-12-11 11:53:19,725 INFO [train.py:421] (5/8) Epoch 4, batch 65600, loss[loss=2.506, over 980.00 frames. , ppl: 12.253016871808557] tot_loss[loss=2.305, over 5488304.10 frames. , ppl: 10.027781113986219], batch size: 70
+2022-12-11 11:55:00,755 INFO [train.py:421] (5/8) Epoch 4, batch 65800, loss[loss=2.343, over 1260.00 frames. , ppl: 10.413618902318213] tot_loss[loss=2.305, over 5505021.59 frames. , ppl: 10.024692938097651], batch size: 70
+2022-12-11 11:56:40,731 INFO [train.py:421] (5/8) Epoch 4, batch 66000, loss[loss=2.325, over 2100.00 frames. , ppl: 10.224525975785504] tot_loss[loss=2.305, over 5507966.78 frames. , ppl: 10.02119975469406], batch size: 70
+2022-12-11 11:56:40,732 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 11:56:41,491 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.919058803417657
+2022-12-11 11:58:22,192 INFO [train.py:421] (5/8) Epoch 4, batch 66200, loss[loss=2.398, over 1260.00 frames. , ppl: 11.001211574244838] tot_loss[loss=2.305, over 5517610.70 frames. , ppl: 10.025137436967999], batch size: 70
+2022-12-11 12:00:04,132 INFO [train.py:421] (5/8) Epoch 4, batch 66400, loss[loss=2.202, over 7000.00 frames. , ppl: 9.04445556505644] tot_loss[loss=2.306, over 5494258.49 frames. , ppl: 10.035021025616844], batch size: 70
+2022-12-11 12:01:41,417 INFO [train.py:421] (5/8) Epoch 4, batch 66600, loss[loss=2.821, over 630.00 frames. , ppl: 16.788182616541725] tot_loss[loss=2.307, over 5446468.88 frames. , ppl: 10.042507922636526], batch size: 70
+2022-12-11 12:03:22,240 INFO [train.py:421] (5/8) Epoch 4, batch 66800, loss[loss=2.499, over 840.00 frames. , ppl: 12.166447174050461] tot_loss[loss=2.307, over 5441045.53 frames. , ppl: 10.044328006119082], batch size: 70
+2022-12-11 12:05:01,128 INFO [train.py:421] (5/8) Epoch 4, batch 67000, loss[loss=2.427, over 1260.00 frames. , ppl: 11.325256048560574] tot_loss[loss=2.308, over 5405966.20 frames. , ppl: 10.054280910714775], batch size: 70
+2022-12-11 12:05:01,128 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 12:05:01,878 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.293, over 211138.00 frames. , ppl: 9.901741611125681
+2022-12-11 12:06:40,973 INFO [train.py:421] (5/8) Epoch 4, batch 67200, loss[loss=2.317, over 1890.00 frames. , ppl: 10.148841499689162] tot_loss[loss=2.309, over 5395252.48 frames. , ppl: 10.060591379330527], batch size: 70
+2022-12-11 12:08:24,020 INFO [train.py:421] (5/8) Epoch 4, batch 67400, loss[loss=2.324, over 2310.00 frames. , ppl: 10.220689645611388] tot_loss[loss=2.309, over 5378855.18 frames. , ppl: 10.068606451330432], batch size: 70
+2022-12-11 12:10:06,522 INFO [train.py:421] (5/8) Epoch 4, batch 67600, loss[loss=2.342, over 2170.00 frames. , ppl: 10.402322409898517] tot_loss[loss=2.309, over 5378610.65 frames. , ppl: 10.065176625742362], batch size: 70
+2022-12-11 12:11:44,333 INFO [train.py:421] (5/8) Epoch 4, batch 67800, loss[loss=2.237, over 2100.00 frames. , ppl: 9.365860745456931] tot_loss[loss=2.308, over 5411562.72 frames. , ppl: 10.054911649369414], batch size: 70
+2022-12-11 12:13:27,419 INFO [train.py:421] (5/8) Epoch 4, batch 68000, loss[loss=2.223, over 3640.00 frames. , ppl: 9.238382410535792] tot_loss[loss=2.307, over 5420365.10 frames. , ppl: 10.048535891450317], batch size: 70
+2022-12-11 12:13:27,420 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 12:13:28,169 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.910991968643046
+2022-12-11 12:15:09,925 INFO [train.py:421] (5/8) Epoch 4, batch 68200, loss[loss=2.367, over 2940.00 frames. , ppl: 10.667971778532333] tot_loss[loss=2.308, over 5399619.14 frames. , ppl: 10.053993947654163], batch size: 70
+2022-12-11 12:16:49,666 INFO [train.py:421] (5/8) Epoch 4, batch 68400, loss[loss=2.199, over 3290.00 frames. , ppl: 9.013481630965975] tot_loss[loss=2.307, over 5423476.15 frames. , ppl: 10.045118663936602], batch size: 70
+2022-12-11 12:18:30,823 INFO [train.py:421] (5/8) Epoch 4, batch 68600, loss[loss=2.349, over 2660.00 frames. , ppl: 10.47662332995163] tot_loss[loss=2.306, over 5448413.69 frames. , ppl: 10.033956952997599], batch size: 70
+2022-12-11 12:20:12,550 INFO [train.py:421] (5/8) Epoch 4, batch 68800, loss[loss=2.393, over 2800.00 frames. , ppl: 10.941178138963195] tot_loss[loss=2.306, over 5452245.93 frames. , ppl: 10.029505882485289], batch size: 70
+2022-12-11 12:21:53,589 INFO [train.py:421] (5/8) Epoch 4, batch 69000, loss[loss=2.186, over 6580.00 frames. , ppl: 8.898081468512679] tot_loss[loss=2.306, over 5465535.70 frames. , ppl: 10.032169223705457], batch size: 70
+2022-12-11 12:21:53,590 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 12:21:54,350 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.294, over 211138.00 frames. , ppl: 9.910044396033411
+2022-12-11 12:23:33,504 INFO [train.py:421] (5/8) Epoch 4, batch 69200, loss[loss=2.261, over 3920.00 frames. , ppl: 9.596341533564372] tot_loss[loss=2.306, over 5460197.26 frames. , ppl: 10.029658704544822], batch size: 70
+2022-12-11 12:25:12,763 INFO [train.py:421] (5/8) Epoch 4, batch 69400, loss[loss=2.213, over 6720.00 frames. , ppl: 9.141795085380503] tot_loss[loss=2.305, over 5494005.41 frames. , ppl: 10.02165019415281], batch size: 70
+2022-12-11 12:26:51,729 INFO [train.py:421] (5/8) Epoch 4, batch 69600, loss[loss=2.261, over 2940.00 frames. , ppl: 9.591475363188561] tot_loss[loss=2.304, over 5520917.32 frames. , ppl: 10.011740088123716], batch size: 70
+2022-12-11 12:28:30,365 INFO [train.py:421] (5/8) Epoch 4, batch 69800, loss[loss=2.497, over 1120.00 frames. , ppl: 12.151563999514591] tot_loss[loss=2.305, over 5514960.99 frames. , ppl: 10.020063037013081], batch size: 70
+2022-12-11 12:30:08,269 INFO [train.py:421] (5/8) Epoch 4, batch 70000, loss[loss=2.252, over 3920.00 frames. , ppl: 9.505474776759494] tot_loss[loss=2.304, over 5524776.18 frames. , ppl: 10.01601989010522], batch size: 70
+2022-12-11 12:30:08,270 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 12:30:09,018 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.292, over 211138.00 frames. , ppl: 9.896056980662939
+2022-12-11 12:31:48,774 INFO [train.py:421] (5/8) Epoch 4, batch 70200, loss[loss=2.53, over 910.00 frames. , ppl: 12.553806560063402] tot_loss[loss=2.305, over 5511381.52 frames. , ppl: 10.021578578799588], batch size: 70
+2022-12-11 12:33:26,587 INFO [train.py:421] (5/8) Epoch 4, batch 70400, loss[loss=2.189, over 2660.00 frames. , ppl: 8.92598658201039] tot_loss[loss=2.304, over 5508749.37 frames. , ppl: 10.01815482678063], batch size: 70
+2022-12-11 12:35:03,418 INFO [train.py:421] (5/8) Epoch 4, batch 70600, loss[loss=2.406, over 1960.00 frames. , ppl: 11.09298336341146] tot_loss[loss=2.305, over 5495275.96 frames. , ppl: 10.023245136704173], batch size: 70
+2022-12-11 12:36:42,005 INFO [train.py:421] (5/8) Epoch 4, batch 70800, loss[loss=2.325, over 1470.00 frames. , ppl: 10.229990935888086] tot_loss[loss=2.304, over 5552134.78 frames. , ppl: 10.009514582964485], batch size: 70
+2022-12-11 12:38:21,443 INFO [train.py:421] (5/8) Epoch 4, batch 71000, loss[loss=2.738, over 700.00 frames. , ppl: 15.46197507781521] tot_loss[loss=2.305, over 5515876.26 frames. , ppl: 10.023311536098506], batch size: 70
+2022-12-11 12:38:21,443 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 12:38:22,205 INFO [train.py:452] (5/8) Epoch 4, validation: loss=2.295, over 211138.00 frames. , ppl: 9.920506450462009
+2022-12-11 12:40:01,638 INFO [train.py:421] (5/8) Epoch 4, batch 71200, loss[loss=2.246, over 3220.00 frames. , ppl: 9.44720753461622] tot_loss[loss=2.305, over 5527903.31 frames. , ppl: 10.027180677137904], batch size: 70
+2022-12-11 12:41:40,328 INFO [train.py:421] (5/8) Epoch 4, batch 71400, loss[loss=2.453, over 1120.00 frames. , ppl: 11.617630369225546] tot_loss[loss=2.305, over 5503088.89 frames. , ppl: 10.026616768928793], batch size: 70
+2022-12-11 12:43:20,029 INFO [train.py:421] (5/8) Epoch 4, batch 71600, loss[loss=2.326, over 3080.00 frames. , ppl: 10.238215401873903] tot_loss[loss=2.305, over 5493440.30 frames. , ppl: 10.029082023532224], batch size: 70
+2022-12-11 12:45:06,135 INFO [train.py:421] (5/8) Epoch 4, batch 71800, loss[loss=2.333, over 2800.00 frames. , ppl: 10.307667226202883] tot_loss[loss=2.305, over 5489851.49 frames. , ppl: 10.023478248913278], batch size: 70
+2022-12-11 12:46:20,322 INFO [train.py:421] (5/8) Epoch 5, batch 0, loss[loss=2.434, over 1890.00 frames. , ppl: 11.40933450944399] tot_loss[loss=2.434, over 1890.00 frames. , ppl: 11.40933450944399], batch size: 70
+2022-12-11 12:48:02,165 INFO [train.py:421] (5/8) Epoch 5, batch 200, loss[loss=2.415, over 1330.00 frames. , ppl: 11.194141804452379] tot_loss[loss=2.304, over 508667.50 frames. , ppl: 10.009319116810445], batch size: 70
+2022-12-11 12:49:43,355 INFO [train.py:421] (5/8) Epoch 5, batch 400, loss[loss=2.651, over 630.00 frames. , ppl: 14.173521194707467] tot_loss[loss=2.299, over 987228.86 frames. , ppl: 9.962688333820221], batch size: 70
+2022-12-11 12:51:24,413 INFO [train.py:421] (5/8) Epoch 5, batch 600, loss[loss=2.276, over 2870.00 frames. , ppl: 9.737571017072112] tot_loss[loss=2.295, over 1406255.07 frames. , ppl: 9.927003245041416], batch size: 70
+2022-12-11 12:53:04,446 INFO [train.py:421] (5/8) Epoch 5, batch 800, loss[loss=2.351, over 3150.00 frames. , ppl: 10.501017205822931] tot_loss[loss=2.296, over 1781290.16 frames. , ppl: 9.937835322230615], batch size: 70
+2022-12-11 12:54:44,630 INFO [train.py:421] (5/8) Epoch 5, batch 1000, loss[loss=2.397, over 1540.00 frames. , ppl: 10.99182830611043] tot_loss[loss=2.298, over 2133883.02 frames. , ppl: 9.9552624017928], batch size: 70
+2022-12-11 12:54:44,630 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 12:54:45,418 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.293, over 211138.00 frames. , ppl: 9.90861440768603
+2022-12-11 12:56:25,143 INFO [train.py:421] (5/8) Epoch 5, batch 1200, loss[loss=2.232, over 6930.00 frames. , ppl: 9.319914231264042] tot_loss[loss=2.298, over 2459034.86 frames. , ppl: 9.9530318104265], batch size: 70
+2022-12-11 12:58:03,061 INFO [train.py:421] (5/8) Epoch 5, batch 1400, loss[loss=2.236, over 2800.00 frames. , ppl: 9.354005223666439] tot_loss[loss=2.297, over 2746944.14 frames. , ppl: 9.941459726507956], batch size: 70
+2022-12-11 12:59:45,898 INFO [train.py:421] (5/8) Epoch 5, batch 1600, loss[loss=2.307, over 2730.00 frames. , ppl: 10.045448091205111] tot_loss[loss=2.295, over 3029149.53 frames. , ppl: 9.92881035892717], batch size: 70
+2022-12-11 13:01:25,814 INFO [train.py:421] (5/8) Epoch 5, batch 1800, loss[loss=2.678, over 630.00 frames. , ppl: 14.563197178475523] tot_loss[loss=2.297, over 3270060.66 frames. , ppl: 9.940956147687976], batch size: 70
+2022-12-11 13:03:08,689 INFO [train.py:421] (5/8) Epoch 5, batch 2000, loss[loss=2.286, over 1750.00 frames. , ppl: 9.831768845941701] tot_loss[loss=2.299, over 3424075.50 frames. , ppl: 9.960114952990798], batch size: 70
+2022-12-11 13:03:08,690 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 13:03:09,438 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.292, over 211138.00 frames. , ppl: 9.893647858407924
+2022-12-11 13:04:49,709 INFO [train.py:421] (5/8) Epoch 5, batch 2200, loss[loss=2.166, over 3500.00 frames. , ppl: 8.725901259124807] tot_loss[loss=2.3, over 3567208.25 frames. , ppl: 9.974791079716057], batch size: 70
+2022-12-11 13:06:31,486 INFO [train.py:421] (5/8) Epoch 5, batch 2400, loss[loss=2.182, over 3920.00 frames. , ppl: 8.868295923704395] tot_loss[loss=2.297, over 3816776.43 frames. , ppl: 9.946498774424985], batch size: 70
+2022-12-11 13:08:14,048 INFO [train.py:421] (5/8) Epoch 5, batch 2600, loss[loss=2.287, over 3360.00 frames. , ppl: 9.849511074754542] tot_loss[loss=2.297, over 3957430.99 frames. , ppl: 9.945799350964826], batch size: 70
+2022-12-11 13:09:54,954 INFO [train.py:421] (5/8) Epoch 5, batch 2800, loss[loss=2.333, over 2100.00 frames. , ppl: 10.306035811313384] tot_loss[loss=2.298, over 4081526.08 frames. , ppl: 9.957920394144807], batch size: 70
+2022-12-11 13:11:38,193 INFO [train.py:421] (5/8) Epoch 5, batch 3000, loss[loss=2.352, over 1960.00 frames. , ppl: 10.509011627857944] tot_loss[loss=2.297, over 4248717.14 frames. , ppl: 9.94164473780271], batch size: 70
+2022-12-11 13:11:38,194 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 13:11:38,954 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.29, over 211138.00 frames. , ppl: 9.878481675447484
+2022-12-11 13:13:17,498 INFO [train.py:421] (5/8) Epoch 5, batch 3200, loss[loss=2.299, over 5320.00 frames. , ppl: 9.96245325814973] tot_loss[loss=2.296, over 4382319.85 frames. , ppl: 9.932540966527656], batch size: 70
+2022-12-11 13:14:57,406 INFO [train.py:421] (5/8) Epoch 5, batch 3400, loss[loss=2.482, over 1330.00 frames. , ppl: 11.970722910155295] tot_loss[loss=2.295, over 4488199.55 frames. , ppl: 9.928347274842334], batch size: 70
+2022-12-11 13:16:37,748 INFO [train.py:421] (5/8) Epoch 5, batch 3600, loss[loss=2.524, over 980.00 frames. , ppl: 12.481417296959604] tot_loss[loss=2.294, over 4609724.74 frames. , ppl: 9.912952942464472], batch size: 70
+2022-12-11 13:18:14,023 INFO [train.py:421] (5/8) Epoch 5, batch 3800, loss[loss=2.212, over 3220.00 frames. , ppl: 9.137040457929412] tot_loss[loss=2.295, over 4668822.20 frames. , ppl: 9.921948859887948], batch size: 70
+2022-12-11 13:19:56,337 INFO [train.py:421] (5/8) Epoch 5, batch 4000, loss[loss=2.352, over 1470.00 frames. , ppl: 10.507831831892755] tot_loss[loss=2.294, over 4805249.40 frames. , ppl: 9.914871522525218], batch size: 70
+2022-12-11 13:19:56,338 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 13:19:57,099 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.293, over 211138.00 frames. , ppl: 9.905227258803862
+2022-12-11 13:21:39,464 INFO [train.py:421] (5/8) Epoch 5, batch 4200, loss[loss=2.798, over 700.00 frames. , ppl: 16.417617816720405] tot_loss[loss=2.295, over 4846399.58 frames. , ppl: 9.920579882369825], batch size: 70
+2022-12-11 13:23:20,068 INFO [train.py:421] (5/8) Epoch 5, batch 4400, loss[loss=2.45, over 1750.00 frames. , ppl: 11.5897888855831] tot_loss[loss=2.297, over 4869280.53 frames. , ppl: 9.943231415671725], batch size: 70
+2022-12-11 13:25:01,536 INFO [train.py:421] (5/8) Epoch 5, batch 4600, loss[loss=2.319, over 3500.00 frames. , ppl: 10.164177807459724] tot_loss[loss=2.298, over 4899689.46 frames. , ppl: 9.949706515624182], batch size: 70
+2022-12-11 13:26:39,713 INFO [train.py:421] (5/8) Epoch 5, batch 4800, loss[loss=2.442, over 2450.00 frames. , ppl: 11.49092870844505] tot_loss[loss=2.299, over 4888635.78 frames. , ppl: 9.964383880921956], batch size: 70
+2022-12-11 13:28:19,899 INFO [train.py:421] (5/8) Epoch 5, batch 5000, loss[loss=2.227, over 2450.00 frames. , ppl: 9.269605964334959] tot_loss[loss=2.299, over 4930403.68 frames. , ppl: 9.964382605663502], batch size: 70
+2022-12-11 13:28:19,899 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 13:28:20,645 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.292, over 211138.00 frames. , ppl: 9.89573182475821
+2022-12-11 13:30:01,901 INFO [train.py:421] (5/8) Epoch 5, batch 5200, loss[loss=2.227, over 2100.00 frames. , ppl: 9.268159548681336] tot_loss[loss=2.298, over 4997523.99 frames. , ppl: 9.953560393099826], batch size: 70
+2022-12-11 13:31:43,063 INFO [train.py:421] (5/8) Epoch 5, batch 5400, loss[loss=2.381, over 1540.00 frames. , ppl: 10.816404126904683] tot_loss[loss=2.299, over 5036798.06 frames. , ppl: 9.9592383764986], batch size: 70
+2022-12-11 13:33:23,011 INFO [train.py:421] (5/8) Epoch 5, batch 5600, loss[loss=2.433, over 1330.00 frames. , ppl: 11.394479286526007] tot_loss[loss=2.299, over 5078247.57 frames. , ppl: 9.964511567803616], batch size: 70
+2022-12-11 13:35:01,443 INFO [train.py:421] (5/8) Epoch 5, batch 5800, loss[loss=2.396, over 1540.00 frames. , ppl: 10.980676316283752] tot_loss[loss=2.299, over 5092608.22 frames. , ppl: 9.967599625809566], batch size: 70
+2022-12-11 13:36:38,820 INFO [train.py:421] (5/8) Epoch 5, batch 6000, loss[loss=2.209, over 6930.00 frames. , ppl: 9.106942897352278] tot_loss[loss=2.299, over 5144803.56 frames. , ppl: 9.96514742252555], batch size: 70
+2022-12-11 13:36:38,820 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 13:36:39,553 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.292, over 211138.00 frames. , ppl: 9.896351387792528
+2022-12-11 13:38:21,623 INFO [train.py:421] (5/8) Epoch 5, batch 6200, loss[loss=2.558, over 910.00 frames. , ppl: 12.905636289608111] tot_loss[loss=2.299, over 5188947.24 frames. , ppl: 9.962418402808769], batch size: 70
+2022-12-11 13:39:57,535 INFO [train.py:421] (5/8) Epoch 5, batch 6400, loss[loss=2.192, over 4270.00 frames. , ppl: 8.950071156734943] tot_loss[loss=2.299, over 5212745.75 frames. , ppl: 9.965770997731813], batch size: 70
+2022-12-11 13:41:36,935 INFO [train.py:421] (5/8) Epoch 5, batch 6600, loss[loss=2.362, over 1190.00 frames. , ppl: 10.609401022505526] tot_loss[loss=2.3, over 5228470.67 frames. , ppl: 9.969681103788783], batch size: 70
+2022-12-11 13:43:15,780 INFO [train.py:421] (5/8) Epoch 5, batch 6800, loss[loss=2.557, over 1050.00 frames. , ppl: 12.899135527614693] tot_loss[loss=2.301, over 5186241.50 frames. , ppl: 9.985366226253495], batch size: 70
+2022-12-11 13:44:54,863 INFO [train.py:421] (5/8) Epoch 5, batch 7000, loss[loss=2.91, over 630.00 frames. , ppl: 18.35455219336735] tot_loss[loss=2.301, over 5207102.65 frames. , ppl: 9.980069068965129], batch size: 70
+2022-12-11 13:44:54,864 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 13:44:55,610 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.291, over 211138.00 frames. , ppl: 9.88711469496257
+2022-12-11 13:46:38,710 INFO [train.py:421] (5/8) Epoch 5, batch 7200, loss[loss=2.973, over 700.00 frames. , ppl: 19.550840457673328] tot_loss[loss=2.3, over 5230983.62 frames. , ppl: 9.978559346536132], batch size: 70
+2022-12-11 13:48:18,508 INFO [train.py:421] (5/8) Epoch 5, batch 7400, loss[loss=2.572, over 840.00 frames. , ppl: 13.09182973134075] tot_loss[loss=2.301, over 5242123.93 frames. , ppl: 9.980759788844779], batch size: 70
+2022-12-11 13:49:58,377 INFO [train.py:421] (5/8) Epoch 5, batch 7600, loss[loss=2.328, over 2940.00 frames. , ppl: 10.256572566472471] tot_loss[loss=2.301, over 5266830.55 frames. , ppl: 9.985456713087403], batch size: 70
+2022-12-11 13:51:37,159 INFO [train.py:421] (5/8) Epoch 5, batch 7800, loss[loss=2.535, over 1190.00 frames. , ppl: 12.611753460403484] tot_loss[loss=2.302, over 5265758.53 frames. , ppl: 9.99028452625143], batch size: 70
+2022-12-11 13:53:18,459 INFO [train.py:421] (5/8) Epoch 5, batch 8000, loss[loss=2.312, over 2100.00 frames. , ppl: 10.094460819178721] tot_loss[loss=2.302, over 5264882.80 frames. , ppl: 9.993286093145622], batch size: 70
+2022-12-11 13:53:18,459 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 13:53:19,219 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.293, over 211138.00 frames. , ppl: 9.903350895657939
+2022-12-11 13:54:59,219 INFO [train.py:421] (5/8) Epoch 5, batch 8200, loss[loss=2.672, over 1120.00 frames. , ppl: 14.473195064785434] tot_loss[loss=2.301, over 5306110.60 frames. , ppl: 9.98272831200966], batch size: 70
+2022-12-11 13:56:40,077 INFO [train.py:421] (5/8) Epoch 5, batch 8400, loss[loss=2.223, over 3640.00 frames. , ppl: 9.236170585213053] tot_loss[loss=2.301, over 5312843.23 frames. , ppl: 9.97942360465173], batch size: 70
+2022-12-11 13:58:20,775 INFO [train.py:421] (5/8) Epoch 5, batch 8600, loss[loss=2.325, over 1610.00 frames. , ppl: 10.229962052623774] tot_loss[loss=2.3, over 5346632.00 frames. , ppl: 9.974622975857544], batch size: 70
+2022-12-11 14:00:00,777 INFO [train.py:421] (5/8) Epoch 5, batch 8800, loss[loss=2.281, over 3920.00 frames. , ppl: 9.786119490939074] tot_loss[loss=2.299, over 5390531.79 frames. , ppl: 9.962634139294687], batch size: 70
+2022-12-11 14:01:38,184 INFO [train.py:421] (5/8) Epoch 5, batch 9000, loss[loss=2.401, over 2030.00 frames. , ppl: 11.03457017261699] tot_loss[loss=2.299, over 5416935.92 frames. , ppl: 9.960482283569318], batch size: 70
+2022-12-11 14:01:38,185 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 14:01:38,944 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.292, over 211138.00 frames. , ppl: 9.898942841256256
+2022-12-11 14:03:18,628 INFO [train.py:421] (5/8) Epoch 5, batch 9200, loss[loss=2.424, over 1120.00 frames. , ppl: 11.286494935941567] tot_loss[loss=2.299, over 5412108.09 frames. , ppl: 9.961385410114675], batch size: 70
+2022-12-11 14:04:53,956 INFO [train.py:421] (5/8) Epoch 5, batch 9400, loss[loss=2.539, over 1050.00 frames. , ppl: 12.67256776982869] tot_loss[loss=2.3, over 5397006.89 frames. , ppl: 9.970003806892157], batch size: 70
+2022-12-11 14:06:31,800 INFO [train.py:421] (5/8) Epoch 5, batch 9600, loss[loss=2.367, over 3290.00 frames. , ppl: 10.668470672212644] tot_loss[loss=2.301, over 5373057.01 frames. , ppl: 9.980151312668026], batch size: 70
+2022-12-11 14:08:10,022 INFO [train.py:421] (5/8) Epoch 5, batch 9800, loss[loss=2.499, over 1190.00 frames. , ppl: 12.176352075785122] tot_loss[loss=2.3, over 5392338.66 frames. , ppl: 9.976628911104171], batch size: 70
+2022-12-11 14:09:51,391 INFO [train.py:421] (5/8) Epoch 5, batch 10000, loss[loss=2.405, over 2030.00 frames. , ppl: 11.083864453888525] tot_loss[loss=2.302, over 5381005.45 frames. , ppl: 9.992397469117863], batch size: 70
+2022-12-11 14:09:51,392 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 14:09:52,185 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.293, over 211138.00 frames. , ppl: 9.904837297826665
+2022-12-11 14:11:34,862 INFO [train.py:421] (5/8) Epoch 5, batch 10200, loss[loss=2.191, over 5530.00 frames. , ppl: 8.940275193043439] tot_loss[loss=2.302, over 5372242.00 frames. , ppl: 9.994288669873225], batch size: 70
+2022-12-11 14:13:12,042 INFO [train.py:421] (5/8) Epoch 5, batch 10400, loss[loss=2.335, over 2730.00 frames. , ppl: 10.327994224947949] tot_loss[loss=2.302, over 5376054.16 frames. , ppl: 9.995730761887348], batch size: 70
+2022-12-11 14:14:53,715 INFO [train.py:421] (5/8) Epoch 5, batch 10600, loss[loss=2.376, over 1890.00 frames. , ppl: 10.760627711610354] tot_loss[loss=2.301, over 5420128.63 frames. , ppl: 9.983275296137093], batch size: 70
+2022-12-11 14:16:33,812 INFO [train.py:421] (5/8) Epoch 5, batch 10800, loss[loss=2.584, over 840.00 frames. , ppl: 13.245497609006934] tot_loss[loss=2.302, over 5405212.79 frames. , ppl: 9.991927551801085], batch size: 70
+2022-12-11 14:18:14,047 INFO [train.py:421] (5/8) Epoch 5, batch 11000, loss[loss=2.433, over 2030.00 frames. , ppl: 11.395280477476996] tot_loss[loss=2.302, over 5400472.80 frames. , ppl: 9.991843659569799], batch size: 70
+2022-12-11 14:18:14,047 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 14:18:14,795 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.292, over 211138.00 frames. , ppl: 9.8904824753545
+2022-12-11 14:19:51,222 INFO [train.py:421] (5/8) Epoch 5, batch 11200, loss[loss=2.327, over 2660.00 frames. , ppl: 10.245808400242364] tot_loss[loss=2.302, over 5401026.28 frames. , ppl: 9.990627855786158], batch size: 70
+2022-12-11 14:21:32,934 INFO [train.py:421] (5/8) Epoch 5, batch 11400, loss[loss=2.436, over 1190.00 frames. , ppl: 11.424819561853683] tot_loss[loss=2.302, over 5385858.50 frames. , ppl: 9.989596035331068], batch size: 70
+2022-12-11 14:23:12,623 INFO [train.py:421] (5/8) Epoch 5, batch 11600, loss[loss=2.218, over 3150.00 frames. , ppl: 9.189251393230121] tot_loss[loss=2.301, over 5406806.85 frames. , ppl: 9.984422334296188], batch size: 70
+2022-12-11 14:24:53,833 INFO [train.py:421] (5/8) Epoch 5, batch 11800, loss[loss=2.36, over 2310.00 frames. , ppl: 10.588236722903718] tot_loss[loss=2.301, over 5432586.09 frames. , ppl: 9.982928789490366], batch size: 70
+2022-12-11 14:26:33,014 INFO [train.py:421] (5/8) Epoch 5, batch 12000, loss[loss=2.317, over 2380.00 frames. , ppl: 10.141122479960046] tot_loss[loss=2.302, over 5421457.80 frames. , ppl: 9.990060585985553], batch size: 70
+2022-12-11 14:26:33,014 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 14:26:33,774 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.291, over 211138.00 frames. , ppl: 9.883088329655019
+2022-12-11 14:28:11,315 INFO [train.py:421] (5/8) Epoch 5, batch 12200, loss[loss=2.473, over 1190.00 frames. , ppl: 11.855278261340878] tot_loss[loss=2.301, over 5447926.86 frames. , ppl: 9.981698058207636], batch size: 70
+2022-12-11 14:29:52,535 INFO [train.py:421] (5/8) Epoch 5, batch 12400, loss[loss=2.188, over 3290.00 frames. , ppl: 8.91714752557929] tot_loss[loss=2.302, over 5393462.88 frames. , ppl: 9.995753188618597], batch size: 70
+2022-12-11 14:31:29,830 INFO [train.py:421] (5/8) Epoch 5, batch 12600, loss[loss=2.209, over 4970.00 frames. , ppl: 9.10298512662583] tot_loss[loss=2.302, over 5381241.90 frames. , ppl: 9.995968847959576], batch size: 70
+2022-12-11 14:33:09,806 INFO [train.py:421] (5/8) Epoch 5, batch 12800, loss[loss=2.692, over 770.00 frames. , ppl: 14.762476290905687] tot_loss[loss=2.302, over 5412443.25 frames. , ppl: 9.989490483865698], batch size: 70
+2022-12-11 14:34:47,876 INFO [train.py:421] (5/8) Epoch 5, batch 13000, loss[loss=2.269, over 3080.00 frames. , ppl: 9.670905238069391] tot_loss[loss=2.304, over 5365437.20 frames. , ppl: 10.009496924841862], batch size: 70
+2022-12-11 14:34:47,876 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 14:34:48,636 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.291, over 211138.00 frames. , ppl: 9.882500313269402
+2022-12-11 14:36:27,618 INFO [train.py:421] (5/8) Epoch 5, batch 13200, loss[loss=2.468, over 1260.00 frames. , ppl: 11.80378011463778] tot_loss[loss=2.303, over 5370845.52 frames. , ppl: 10.007862251153927], batch size: 70
+2022-12-11 14:38:12,162 INFO [train.py:421] (5/8) Epoch 5, batch 13400, loss[loss=2.219, over 3850.00 frames. , ppl: 9.20117664206134] tot_loss[loss=2.303, over 5387855.81 frames. , ppl: 10.008081353109327], batch size: 70
+2022-12-11 14:39:51,888 INFO [train.py:421] (5/8) Epoch 5, batch 13600, loss[loss=2.205, over 7770.00 frames. , ppl: 9.071938443476247] tot_loss[loss=2.303, over 5416259.09 frames. , ppl: 10.003695095677461], batch size: 70
+2022-12-11 14:41:35,232 INFO [train.py:421] (5/8) Epoch 5, batch 13800, loss[loss=2.483, over 910.00 frames. , ppl: 11.981458631553055] tot_loss[loss=2.302, over 5438587.48 frames. , ppl: 9.992742235614992], batch size: 70
+2022-12-11 14:43:18,792 INFO [train.py:421] (5/8) Epoch 5, batch 14000, loss[loss=2.505, over 1120.00 frames. , ppl: 12.246103744280875] tot_loss[loss=2.3, over 5458667.91 frames. , ppl: 9.97681692469591], batch size: 70
+2022-12-11 14:43:18,792 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 14:43:19,550 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.291, over 211138.00 frames. , ppl: 9.88920899459524
+2022-12-11 14:45:00,093 INFO [train.py:421] (5/8) Epoch 5, batch 14200, loss[loss=2.342, over 2380.00 frames. , ppl: 10.40486689420022] tot_loss[loss=2.299, over 5512517.11 frames. , ppl: 9.960443704066034], batch size: 70
+2022-12-11 14:46:37,397 INFO [train.py:421] (5/8) Epoch 5, batch 14400, loss[loss=3.052, over 630.00 frames. , ppl: 21.15678960515071] tot_loss[loss=2.298, over 5525214.10 frames. , ppl: 9.958557455301452], batch size: 70
+2022-12-11 14:48:13,461 INFO [train.py:421] (5/8) Epoch 5, batch 14600, loss[loss=2.215, over 2800.00 frames. , ppl: 9.16356455850534] tot_loss[loss=2.3, over 5449353.11 frames. , ppl: 9.977693726309846], batch size: 70
+2022-12-11 14:49:54,267 INFO [train.py:421] (5/8) Epoch 5, batch 14800, loss[loss=2.816, over 630.00 frames. , ppl: 16.704576790646044] tot_loss[loss=2.299, over 5474272.67 frames. , ppl: 9.968190220799517], batch size: 70
+2022-12-11 14:51:36,211 INFO [train.py:421] (5/8) Epoch 5, batch 15000, loss[loss=2.312, over 1960.00 frames. , ppl: 10.091858039536541] tot_loss[loss=2.3, over 5463211.28 frames. , ppl: 9.971827439568424], batch size: 70
+2022-12-11 14:51:36,212 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 14:51:36,973 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.291, over 211138.00 frames. , ppl: 9.882491537170422
+2022-12-11 14:53:18,611 INFO [train.py:421] (5/8) Epoch 5, batch 15200, loss[loss=2.181, over 3430.00 frames. , ppl: 8.856952446576685] tot_loss[loss=2.3, over 5473375.05 frames. , ppl: 9.975097067501586], batch size: 70
+2022-12-11 14:55:01,722 INFO [train.py:421] (5/8) Epoch 5, batch 15400, loss[loss=2.264, over 2590.00 frames. , ppl: 9.620396446478956] tot_loss[loss=2.3, over 5465203.70 frames. , ppl: 9.973555812279358], batch size: 70
+2022-12-11 14:56:39,993 INFO [train.py:421] (5/8) Epoch 5, batch 15600, loss[loss=2.205, over 4550.00 frames. , ppl: 9.070473498945171] tot_loss[loss=2.3, over 5516796.88 frames. , ppl: 9.97165342841315], batch size: 70
+2022-12-11 14:58:22,489 INFO [train.py:421] (5/8) Epoch 5, batch 15800, loss[loss=2.18, over 4480.00 frames. , ppl: 8.848015705975687] tot_loss[loss=2.3, over 5505025.82 frames. , ppl: 9.97149362499596], batch size: 70
+2022-12-11 15:00:04,356 INFO [train.py:421] (5/8) Epoch 5, batch 16000, loss[loss=2.389, over 1750.00 frames. , ppl: 10.907639863897996] tot_loss[loss=2.3, over 5485879.76 frames. , ppl: 9.972523846327583], batch size: 70
+2022-12-11 15:00:04,357 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 15:00:05,117 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.29, over 211138.00 frames. , ppl: 9.876905669210421
+2022-12-11 15:01:47,812 INFO [train.py:421] (5/8) Epoch 5, batch 16200, loss[loss=2.339, over 1890.00 frames. , ppl: 10.369134968355699] tot_loss[loss=2.299, over 5525021.36 frames. , ppl: 9.963480416174772], batch size: 70
+2022-12-11 15:03:30,620 INFO [train.py:421] (5/8) Epoch 5, batch 16400, loss[loss=2.835, over 700.00 frames. , ppl: 17.02724659461188] tot_loss[loss=2.297, over 5582780.01 frames. , ppl: 9.949166855907826], batch size: 70
+2022-12-11 15:05:08,591 INFO [train.py:421] (5/8) Epoch 5, batch 16600, loss[loss=2.362, over 2520.00 frames. , ppl: 10.607316260811498] tot_loss[loss=2.299, over 5537497.43 frames. , ppl: 9.964838247859664], batch size: 70
+2022-12-11 15:06:50,339 INFO [train.py:421] (5/8) Epoch 5, batch 16800, loss[loss=2.301, over 2380.00 frames. , ppl: 9.980403115223616] tot_loss[loss=2.3, over 5515626.68 frames. , ppl: 9.97141404324684], batch size: 70
+2022-12-11 15:08:31,261 INFO [train.py:421] (5/8) Epoch 5, batch 17000, loss[loss=2.345, over 2100.00 frames. , ppl: 10.429915834328996] tot_loss[loss=2.3, over 5488956.00 frames. , ppl: 9.977800806993196], batch size: 70
+2022-12-11 15:08:31,262 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 15:08:32,009 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.292, over 211138.00 frames. , ppl: 9.891302273858114
+2022-12-11 15:10:09,655 INFO [train.py:421] (5/8) Epoch 5, batch 17200, loss[loss=2.589, over 840.00 frames. , ppl: 13.322162497286525] tot_loss[loss=2.299, over 5518845.57 frames. , ppl: 9.968174353058789], batch size: 70
+2022-12-11 15:11:47,999 INFO [train.py:421] (5/8) Epoch 5, batch 17400, loss[loss=2.373, over 1610.00 frames. , ppl: 10.733275659381272] tot_loss[loss=2.299, over 5549721.85 frames. , ppl: 9.96299038383526], batch size: 70
+2022-12-11 15:13:29,126 INFO [train.py:421] (5/8) Epoch 5, batch 17600, loss[loss=2.245, over 5530.00 frames. , ppl: 9.435824766581932] tot_loss[loss=2.299, over 5564007.45 frames. , ppl: 9.961489009941875], batch size: 70
+2022-12-11 15:15:10,238 INFO [train.py:421] (5/8) Epoch 5, batch 17800, loss[loss=2.352, over 1470.00 frames. , ppl: 10.504753816438939] tot_loss[loss=2.299, over 5522316.62 frames. , ppl: 9.961815323687894], batch size: 70
+2022-12-11 15:16:49,288 INFO [train.py:421] (5/8) Epoch 5, batch 18000, loss[loss=2.473, over 1960.00 frames. , ppl: 11.855333160399198] tot_loss[loss=2.299, over 5507622.03 frames. , ppl: 9.95944562034776], batch size: 70
+2022-12-11 15:16:49,289 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 15:16:50,060 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.291, over 211138.00 frames. , ppl: 9.884169376428208
+2022-12-11 15:18:27,178 INFO [train.py:421] (5/8) Epoch 5, batch 18200, loss[loss=2.229, over 2590.00 frames. , ppl: 9.292358835358332] tot_loss[loss=2.297, over 5541964.30 frames. , ppl: 9.94917344100317], batch size: 70
+2022-12-11 15:20:07,225 INFO [train.py:421] (5/8) Epoch 5, batch 18400, loss[loss=2.203, over 6650.00 frames. , ppl: 9.051443556474748] tot_loss[loss=2.298, over 5530970.69 frames. , ppl: 9.954764054183677], batch size: 70
+2022-12-11 15:21:46,262 INFO [train.py:421] (5/8) Epoch 5, batch 18600, loss[loss=2.408, over 1680.00 frames. , ppl: 11.108626454142692] tot_loss[loss=2.299, over 5511658.78 frames. , ppl: 9.959727264155703], batch size: 70
+2022-12-11 15:23:25,453 INFO [train.py:421] (5/8) Epoch 5, batch 18800, loss[loss=2.251, over 5600.00 frames. , ppl: 9.49529673969819] tot_loss[loss=2.299, over 5526551.95 frames. , ppl: 9.959736059872721], batch size: 70
+2022-12-11 15:25:02,768 INFO [train.py:421] (5/8) Epoch 5, batch 19000, loss[loss=2.483, over 1540.00 frames. , ppl: 11.976574813204465] tot_loss[loss=2.299, over 5491120.73 frames. , ppl: 9.96454408944266], batch size: 70
+2022-12-11 15:25:02,769 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 15:25:03,528 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.292, over 211138.00 frames. , ppl: 9.892699015651312
+2022-12-11 15:26:43,246 INFO [train.py:421] (5/8) Epoch 5, batch 19200, loss[loss=2.34, over 3150.00 frames. , ppl: 10.377807935234408] tot_loss[loss=2.299, over 5518061.52 frames. , ppl: 9.959722602152917], batch size: 70
+2022-12-11 15:28:24,622 INFO [train.py:421] (5/8) Epoch 5, batch 19400, loss[loss=2.214, over 3990.00 frames. , ppl: 9.149471385098005] tot_loss[loss=2.299, over 5493264.93 frames. , ppl: 9.965980568580095], batch size: 70
+2022-12-11 15:30:03,434 INFO [train.py:421] (5/8) Epoch 5, batch 19600, loss[loss=2.459, over 1050.00 frames. , ppl: 11.698705990763553] tot_loss[loss=2.299, over 5492629.98 frames. , ppl: 9.967124900764441], batch size: 70
+2022-12-11 15:31:37,928 INFO [train.py:421] (5/8) Epoch 5, batch 19800, loss[loss=2.25, over 3920.00 frames. , ppl: 9.487870563315235] tot_loss[loss=2.3, over 5503388.84 frames. , ppl: 9.970930287366432], batch size: 70
+2022-12-11 15:33:17,500 INFO [train.py:421] (5/8) Epoch 5, batch 20000, loss[loss=2.297, over 3010.00 frames. , ppl: 9.939862187603659] tot_loss[loss=2.3, over 5478636.77 frames. , ppl: 9.971168494421512], batch size: 70
+2022-12-11 15:33:17,501 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 15:33:18,262 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.292, over 211138.00 frames. , ppl: 9.897619928623687
+2022-12-11 15:34:59,004 INFO [train.py:421] (5/8) Epoch 5, batch 20200, loss[loss=2.499, over 910.00 frames. , ppl: 12.173323022688185] tot_loss[loss=2.3, over 5441694.43 frames. , ppl: 9.975942128930996], batch size: 70
+2022-12-11 15:36:41,049 INFO [train.py:421] (5/8) Epoch 5, batch 20400, loss[loss=2.243, over 3920.00 frames. , ppl: 9.422526668611885] tot_loss[loss=2.3, over 5435488.05 frames. , ppl: 9.970853991365207], batch size: 70
+2022-12-11 15:38:20,076 INFO [train.py:421] (5/8) Epoch 5, batch 20600, loss[loss=2.493, over 1190.00 frames. , ppl: 12.096302364337536] tot_loss[loss=2.299, over 5468557.86 frames. , ppl: 9.966508919820628], batch size: 70
+2022-12-11 15:40:02,095 INFO [train.py:421] (5/8) Epoch 5, batch 20800, loss[loss=2.349, over 2030.00 frames. , ppl: 10.47954358611427] tot_loss[loss=2.299, over 5462356.78 frames. , ppl: 9.966741495288575], batch size: 70
+2022-12-11 15:41:40,027 INFO [train.py:421] (5/8) Epoch 5, batch 21000, loss[loss=2.985, over 560.00 frames. , ppl: 19.783430674157824] tot_loss[loss=2.299, over 5481773.41 frames. , ppl: 9.96859377105716], batch size: 70
+2022-12-11 15:41:40,028 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 15:41:40,786 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.8614396619666
+2022-12-11 15:43:22,037 INFO [train.py:421] (5/8) Epoch 5, batch 21200, loss[loss=2.207, over 13300.00 frames. , ppl: 9.086492271749835] tot_loss[loss=2.299, over 5520116.72 frames. , ppl: 9.96061187808084], batch size: 70
+2022-12-11 15:45:08,222 INFO [train.py:421] (5/8) Epoch 5, batch 21400, loss[loss=2.291, over 2870.00 frames. , ppl: 9.888736501290206] tot_loss[loss=2.297, over 5551747.79 frames. , ppl: 9.947812619908746], batch size: 70
+2022-12-11 15:46:45,918 INFO [train.py:421] (5/8) Epoch 5, batch 21600, loss[loss=2.953, over 700.00 frames. , ppl: 19.172278388823422] tot_loss[loss=2.299, over 5498824.72 frames. , ppl: 9.959893960647022], batch size: 70
+2022-12-11 15:48:20,272 INFO [train.py:421] (5/8) Epoch 5, batch 21800, loss[loss=2.252, over 4830.00 frames. , ppl: 9.504789444606324] tot_loss[loss=2.299, over 5479156.27 frames. , ppl: 9.960858341749825], batch size: 70
+2022-12-11 15:49:59,709 INFO [train.py:421] (5/8) Epoch 5, batch 22000, loss[loss=2.528, over 1540.00 frames. , ppl: 12.530363817242629] tot_loss[loss=2.3, over 5465033.68 frames. , ppl: 9.971645697945885], batch size: 70
+2022-12-11 15:49:59,710 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 15:50:00,468 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.291, over 211138.00 frames. , ppl: 9.884660933415132
+2022-12-11 15:51:45,362 INFO [train.py:421] (5/8) Epoch 5, batch 22200, loss[loss=2.256, over 3780.00 frames. , ppl: 9.548130069705689] tot_loss[loss=2.3, over 5464631.62 frames. , ppl: 9.972761892432382], batch size: 70
+2022-12-11 15:53:28,191 INFO [train.py:421] (5/8) Epoch 5, batch 22400, loss[loss=2.369, over 1680.00 frames. , ppl: 10.690050110389915] tot_loss[loss=2.299, over 5469652.55 frames. , ppl: 9.962760054215948], batch size: 70
+2022-12-11 15:55:07,005 INFO [train.py:421] (5/8) Epoch 5, batch 22600, loss[loss=2.504, over 1680.00 frames. , ppl: 12.235436866245024] tot_loss[loss=2.301, over 5430754.90 frames. , ppl: 9.979198543139388], batch size: 70
+2022-12-11 15:56:53,258 INFO [train.py:421] (5/8) Epoch 5, batch 22800, loss[loss=2.346, over 2800.00 frames. , ppl: 10.447758066411556] tot_loss[loss=2.301, over 5440276.11 frames. , ppl: 9.980230954452804], batch size: 70
+2022-12-11 15:58:37,014 INFO [train.py:421] (5/8) Epoch 5, batch 23000, loss[loss=2.165, over 7630.00 frames. , ppl: 8.710528607109513] tot_loss[loss=2.299, over 5492946.83 frames. , ppl: 9.966682361904732], batch size: 70
+2022-12-11 15:58:37,015 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 15:58:37,746 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.29, over 211138.00 frames. , ppl: 9.87439159731824
+2022-12-11 16:00:16,678 INFO [train.py:421] (5/8) Epoch 5, batch 23200, loss[loss=2.395, over 2310.00 frames. , ppl: 10.964782157663974] tot_loss[loss=2.301, over 5443116.79 frames. , ppl: 9.980994235612274], batch size: 70
+2022-12-11 16:01:54,246 INFO [train.py:421] (5/8) Epoch 5, batch 23400, loss[loss=2.223, over 6090.00 frames. , ppl: 9.23500665384897] tot_loss[loss=2.3, over 5457448.98 frames. , ppl: 9.97727276316203], batch size: 70
+2022-12-11 16:03:33,381 INFO [train.py:421] (5/8) Epoch 5, batch 23600, loss[loss=2.214, over 5040.00 frames. , ppl: 9.156187438254584] tot_loss[loss=2.301, over 5434375.69 frames. , ppl: 9.98218808416139], batch size: 70
+2022-12-11 16:05:15,481 INFO [train.py:421] (5/8) Epoch 5, batch 23800, loss[loss=2.122, over 9380.00 frames. , ppl: 8.350030641990042] tot_loss[loss=2.3, over 5427790.83 frames. , ppl: 9.978614906447204], batch size: 70
+2022-12-11 16:06:54,574 INFO [train.py:421] (5/8) Epoch 5, batch 24000, loss[loss=2.49, over 1330.00 frames. , ppl: 12.066251608040533] tot_loss[loss=2.302, over 5408209.08 frames. , ppl: 9.991191234436554], batch size: 70
+2022-12-11 16:06:54,575 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 16:06:55,332 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.869746633606848
+2022-12-11 16:08:33,925 INFO [train.py:421] (5/8) Epoch 5, batch 24200, loss[loss=2.281, over 2520.00 frames. , ppl: 9.783042200594052] tot_loss[loss=2.302, over 5408856.14 frames. , ppl: 9.991830997162802], batch size: 70
+2022-12-11 16:10:16,217 INFO [train.py:421] (5/8) Epoch 5, batch 24400, loss[loss=2.426, over 1330.00 frames. , ppl: 11.31904688141744] tot_loss[loss=2.301, over 5455907.87 frames. , ppl: 9.983993039174639], batch size: 70
+2022-12-11 16:11:57,882 INFO [train.py:421] (5/8) Epoch 5, batch 24600, loss[loss=2.346, over 1120.00 frames. , ppl: 10.43861317710341] tot_loss[loss=2.301, over 5452274.99 frames. , ppl: 9.988254242605056], batch size: 70
+2022-12-11 16:13:31,672 INFO [train.py:421] (5/8) Epoch 5, batch 24800, loss[loss=2.552, over 700.00 frames. , ppl: 12.836484066293222] tot_loss[loss=2.303, over 5397215.59 frames. , ppl: 10.000406740714821], batch size: 70
+2022-12-11 16:15:09,415 INFO [train.py:421] (5/8) Epoch 5, batch 25000, loss[loss=2.439, over 1680.00 frames. , ppl: 11.463622461325828] tot_loss[loss=2.303, over 5370326.43 frames. , ppl: 10.008078402175357], batch size: 70
+2022-12-11 16:15:09,416 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 16:15:10,164 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.867343916412231
+2022-12-11 16:16:49,415 INFO [train.py:421] (5/8) Epoch 5, batch 25200, loss[loss=2.344, over 980.00 frames. , ppl: 10.421610021296349] tot_loss[loss=2.302, over 5394391.74 frames. , ppl: 9.995713044002605], batch size: 70
+2022-12-11 16:18:26,295 INFO [train.py:421] (5/8) Epoch 5, batch 25400, loss[loss=2.301, over 2590.00 frames. , ppl: 9.979612614094501] tot_loss[loss=2.303, over 5374233.88 frames. , ppl: 10.004662343113491], batch size: 70
+2022-12-11 16:20:05,519 INFO [train.py:421] (5/8) Epoch 5, batch 25600, loss[loss=2.426, over 1120.00 frames. , ppl: 11.317618242360531] tot_loss[loss=2.302, over 5371984.95 frames. , ppl: 9.99740377509885], batch size: 70
+2022-12-11 16:21:46,569 INFO [train.py:421] (5/8) Epoch 5, batch 25800, loss[loss=2.136, over 11970.00 frames. , ppl: 8.461339982402926] tot_loss[loss=2.301, over 5411426.28 frames. , ppl: 9.987021104805077], batch size: 70
+2022-12-11 16:23:25,042 INFO [train.py:421] (5/8) Epoch 5, batch 26000, loss[loss=2.534, over 1330.00 frames. , ppl: 12.600248734173665] tot_loss[loss=2.302, over 5376015.48 frames. , ppl: 9.999060118938393], batch size: 70
+2022-12-11 16:23:25,043 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 16:23:25,801 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.864416167537827
+2022-12-11 16:25:05,333 INFO [train.py:421] (5/8) Epoch 5, batch 26200, loss[loss=2.399, over 2030.00 frames. , ppl: 11.016947272199253] tot_loss[loss=2.302, over 5404896.13 frames. , ppl: 9.993068468863092], batch size: 70
+2022-12-11 16:26:45,398 INFO [train.py:421] (5/8) Epoch 5, batch 26400, loss[loss=2.287, over 2940.00 frames. , ppl: 9.844622392311967] tot_loss[loss=2.3, over 5495795.02 frames. , ppl: 9.977995619683135], batch size: 70
+2022-12-11 16:28:28,141 INFO [train.py:421] (5/8) Epoch 5, batch 26600, loss[loss=2.351, over 1890.00 frames. , ppl: 10.4975446789184] tot_loss[loss=2.301, over 5484076.78 frames. , ppl: 9.979446092906434], batch size: 70
+2022-12-11 16:30:12,530 INFO [train.py:421] (5/8) Epoch 5, batch 26800, loss[loss=2.246, over 4410.00 frames. , ppl: 9.454393019349776] tot_loss[loss=2.3, over 5513783.16 frames. , ppl: 9.974640546941147], batch size: 70
+2022-12-11 16:31:49,995 INFO [train.py:421] (5/8) Epoch 5, batch 27000, loss[loss=2.485, over 1120.00 frames. , ppl: 11.995538407995154] tot_loss[loss=2.301, over 5499761.31 frames. , ppl: 9.979452230262584], batch size: 70
+2022-12-11 16:31:49,995 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 16:31:50,754 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.29, over 211138.00 frames. , ppl: 9.875375224811005
+2022-12-11 16:33:32,082 INFO [train.py:421] (5/8) Epoch 5, batch 27200, loss[loss=2.262, over 2170.00 frames. , ppl: 9.602335195305146] tot_loss[loss=2.3, over 5518286.60 frames. , ppl: 9.971229869384816], batch size: 70
+2022-12-11 16:35:10,658 INFO [train.py:421] (5/8) Epoch 5, batch 27400, loss[loss=2.224, over 3500.00 frames. , ppl: 9.246893593674491] tot_loss[loss=2.3, over 5487840.38 frames. , ppl: 9.971456437417975], batch size: 70
+2022-12-11 16:36:49,937 INFO [train.py:421] (5/8) Epoch 5, batch 27600, loss[loss=2.38, over 2100.00 frames. , ppl: 10.802099500080006] tot_loss[loss=2.3, over 5503335.99 frames. , ppl: 9.969833395413886], batch size: 70
+2022-12-11 16:38:28,348 INFO [train.py:421] (5/8) Epoch 5, batch 27800, loss[loss=2.392, over 1470.00 frames. , ppl: 10.935507972020785] tot_loss[loss=2.3, over 5500360.72 frames. , ppl: 9.972676764393126], batch size: 70
+2022-12-11 16:40:08,913 INFO [train.py:421] (5/8) Epoch 5, batch 28000, loss[loss=2.205, over 6510.00 frames. , ppl: 9.0730205898163] tot_loss[loss=2.3, over 5492093.48 frames. , ppl: 9.972168670904924], batch size: 70
+2022-12-11 16:40:08,914 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 16:40:09,677 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.288, over 211138.00 frames. , ppl: 9.858846353025392
+2022-12-11 16:41:47,874 INFO [train.py:421] (5/8) Epoch 5, batch 28200, loss[loss=2.197, over 8330.00 frames. , ppl: 8.994555312756313] tot_loss[loss=2.3, over 5483138.19 frames. , ppl: 9.972060255546504], batch size: 70
+2022-12-11 16:43:28,937 INFO [train.py:421] (5/8) Epoch 5, batch 28400, loss[loss=2.461, over 980.00 frames. , ppl: 11.71537550178081] tot_loss[loss=2.299, over 5498087.31 frames. , ppl: 9.965874778745127], batch size: 70
+2022-12-11 16:45:09,674 INFO [train.py:421] (5/8) Epoch 5, batch 28600, loss[loss=2.221, over 4550.00 frames. , ppl: 9.214127415794117] tot_loss[loss=2.3, over 5485453.75 frames. , ppl: 9.974118311440225], batch size: 70
+2022-12-11 16:46:50,672 INFO [train.py:421] (5/8) Epoch 5, batch 28800, loss[loss=2.236, over 5250.00 frames. , ppl: 9.357580429517306] tot_loss[loss=2.3, over 5445698.42 frames. , ppl: 9.97477031713469], batch size: 70
+2022-12-11 16:48:28,318 INFO [train.py:421] (5/8) Epoch 5, batch 29000, loss[loss=2.543, over 980.00 frames. , ppl: 12.713843968268456] tot_loss[loss=2.3, over 5460631.59 frames. , ppl: 9.9734292274013], batch size: 70
+2022-12-11 16:48:28,318 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 16:48:29,078 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.86421176864417
+2022-12-11 16:50:05,458 INFO [train.py:421] (5/8) Epoch 5, batch 29200, loss[loss=2.292, over 4130.00 frames. , ppl: 9.89855671699997] tot_loss[loss=2.303, over 5384224.94 frames. , ppl: 10.00100006611809], batch size: 70
+2022-12-11 16:51:46,726 INFO [train.py:421] (5/8) Epoch 5, batch 29400, loss[loss=2.711, over 840.00 frames. , ppl: 15.03683258848387] tot_loss[loss=2.302, over 5426532.98 frames. , ppl: 9.993067773263109], batch size: 70
+2022-12-11 16:53:26,108 INFO [train.py:421] (5/8) Epoch 5, batch 29600, loss[loss=2.237, over 3990.00 frames. , ppl: 9.361427993508403] tot_loss[loss=2.301, over 5457284.86 frames. , ppl: 9.984626238291584], batch size: 70
+2022-12-11 16:55:06,801 INFO [train.py:421] (5/8) Epoch 5, batch 29800, loss[loss=2.235, over 5320.00 frames. , ppl: 9.347698692468414] tot_loss[loss=2.3, over 5469794.92 frames. , ppl: 9.97045477416423], batch size: 70
+2022-12-11 16:56:46,812 INFO [train.py:421] (5/8) Epoch 5, batch 30000, loss[loss=2.356, over 3360.00 frames. , ppl: 10.548619098735374] tot_loss[loss=2.3, over 5429352.19 frames. , ppl: 9.976668615474775], batch size: 70
+2022-12-11 16:56:46,812 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 16:56:47,572 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.86519146183926
+2022-12-11 16:58:29,840 INFO [train.py:421] (5/8) Epoch 5, batch 30200, loss[loss=2.185, over 6090.00 frames. , ppl: 8.889481584073788] tot_loss[loss=2.3, over 5486081.49 frames. , ppl: 9.970273027142357], batch size: 70
+2022-12-11 17:00:10,447 INFO [train.py:421] (5/8) Epoch 5, batch 30400, loss[loss=2.415, over 1610.00 frames. , ppl: 11.193917471154853] tot_loss[loss=2.299, over 5500162.38 frames. , ppl: 9.961619284349261], batch size: 70
+2022-12-11 17:01:51,075 INFO [train.py:421] (5/8) Epoch 5, batch 30600, loss[loss=2.424, over 1400.00 frames. , ppl: 11.295685781385133] tot_loss[loss=2.298, over 5541925.93 frames. , ppl: 9.952293107415736], batch size: 70
+2022-12-11 17:03:31,209 INFO [train.py:421] (5/8) Epoch 5, batch 30800, loss[loss=2.473, over 980.00 frames. , ppl: 11.858428766659959] tot_loss[loss=2.299, over 5498691.27 frames. , ppl: 9.968565917482453], batch size: 70
+2022-12-11 17:05:13,227 INFO [train.py:421] (5/8) Epoch 5, batch 31000, loss[loss=2.275, over 2940.00 frames. , ppl: 9.723351075797403] tot_loss[loss=2.299, over 5518162.70 frames. , ppl: 9.966954508488742], batch size: 70
+2022-12-11 17:05:13,228 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 17:05:13,971 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.862906635580453
+2022-12-11 17:06:53,714 INFO [train.py:421] (5/8) Epoch 5, batch 31200, loss[loss=2.342, over 3080.00 frames. , ppl: 10.400711726459232] tot_loss[loss=2.298, over 5555706.03 frames. , ppl: 9.952380385530759], batch size: 70
+2022-12-11 17:08:34,075 INFO [train.py:421] (5/8) Epoch 5, batch 31400, loss[loss=3.054, over 490.00 frames. , ppl: 21.20751644321429] tot_loss[loss=2.297, over 5590273.85 frames. , ppl: 9.940350533592772], batch size: 70
+2022-12-11 17:10:13,743 INFO [train.py:421] (5/8) Epoch 5, batch 31600, loss[loss=2.619, over 770.00 frames. , ppl: 13.717586012506166] tot_loss[loss=2.296, over 5587221.62 frames. , ppl: 9.93544213212399], batch size: 70
+2022-12-11 17:11:57,205 INFO [train.py:421] (5/8) Epoch 5, batch 31800, loss[loss=2.249, over 3570.00 frames. , ppl: 9.475680357405961] tot_loss[loss=2.295, over 5601250.01 frames. , ppl: 9.92197997681755], batch size: 70
+2022-12-11 17:13:39,189 INFO [train.py:421] (5/8) Epoch 5, batch 32000, loss[loss=2.358, over 1890.00 frames. , ppl: 10.565630598635748] tot_loss[loss=2.296, over 5551147.46 frames. , ppl: 9.937361392519321], batch size: 70
+2022-12-11 17:13:39,190 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 17:13:39,950 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.866756836775595
+2022-12-11 17:15:20,454 INFO [train.py:421] (5/8) Epoch 5, batch 32200, loss[loss=2.685, over 840.00 frames. , ppl: 14.658182634899969] tot_loss[loss=2.295, over 5579046.47 frames. , ppl: 9.925358301416557], batch size: 70
+2022-12-11 17:17:01,705 INFO [train.py:421] (5/8) Epoch 5, batch 32400, loss[loss=2.219, over 4410.00 frames. , ppl: 9.194737473928727] tot_loss[loss=2.294, over 5621558.56 frames. , ppl: 9.913544689970415], batch size: 70
+2022-12-11 17:18:38,647 INFO [train.py:421] (5/8) Epoch 5, batch 32600, loss[loss=2.596, over 840.00 frames. , ppl: 13.404817912678071] tot_loss[loss=2.295, over 5623652.96 frames. , ppl: 9.919597669593152], batch size: 70
+2022-12-11 17:20:17,508 INFO [train.py:421] (5/8) Epoch 5, batch 32800, loss[loss=2.711, over 700.00 frames. , ppl: 15.048556711882991] tot_loss[loss=2.296, over 5562262.37 frames. , ppl: 9.932655071999884], batch size: 70
+2022-12-11 17:21:54,567 INFO [train.py:421] (5/8) Epoch 5, batch 33000, loss[loss=2.24, over 6440.00 frames. , ppl: 9.392014658754574] tot_loss[loss=2.295, over 5575202.51 frames. , ppl: 9.925271649459312], batch size: 70
+2022-12-11 17:21:54,567 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 17:21:55,298 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.29, over 211138.00 frames. , ppl: 9.874875360260138
+2022-12-11 17:23:34,426 INFO [train.py:421] (5/8) Epoch 5, batch 33200, loss[loss=2.394, over 2100.00 frames. , ppl: 10.954697598280374] tot_loss[loss=2.295, over 5575381.90 frames. , ppl: 9.920257718224077], batch size: 70
+2022-12-11 17:25:20,346 INFO [train.py:421] (5/8) Epoch 5, batch 33400, loss[loss=2.272, over 3570.00 frames. , ppl: 9.696830608539333] tot_loss[loss=2.295, over 5572278.15 frames. , ppl: 9.926491874416097], batch size: 70
+2022-12-11 17:27:01,180 INFO [train.py:421] (5/8) Epoch 5, batch 33600, loss[loss=2.232, over 4480.00 frames. , ppl: 9.322648153946943] tot_loss[loss=2.295, over 5571671.18 frames. , ppl: 9.926606521709788], batch size: 70
+2022-12-11 17:28:43,676 INFO [train.py:421] (5/8) Epoch 5, batch 33800, loss[loss=2.278, over 3780.00 frames. , ppl: 9.760852402855772] tot_loss[loss=2.297, over 5521506.57 frames. , ppl: 9.943844684286434], batch size: 70
+2022-12-11 17:30:21,119 INFO [train.py:421] (5/8) Epoch 5, batch 34000, loss[loss=2.543, over 1120.00 frames. , ppl: 12.716865151043642] tot_loss[loss=2.297, over 5519222.84 frames. , ppl: 9.946883430971411], batch size: 70
+2022-12-11 17:30:21,120 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 17:30:21,882 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.288, over 211138.00 frames. , ppl: 9.858973302750663
+2022-12-11 17:32:01,818 INFO [train.py:421] (5/8) Epoch 5, batch 34200, loss[loss=2.281, over 3360.00 frames. , ppl: 9.788182009351283] tot_loss[loss=2.295, over 5595436.36 frames. , ppl: 9.928480048055523], batch size: 70
+2022-12-11 17:33:39,848 INFO [train.py:421] (5/8) Epoch 5, batch 34400, loss[loss=2.227, over 4480.00 frames. , ppl: 9.275034344372322] tot_loss[loss=2.295, over 5618540.58 frames. , ppl: 9.924722238435148], batch size: 70
+2022-12-11 17:35:26,817 INFO [train.py:421] (5/8) Epoch 5, batch 34600, loss[loss=2.467, over 1260.00 frames. , ppl: 11.78282119901848] tot_loss[loss=2.295, over 5620530.02 frames. , ppl: 9.924491762426177], batch size: 70
+2022-12-11 17:37:05,657 INFO [train.py:421] (5/8) Epoch 5, batch 34800, loss[loss=2.33, over 2240.00 frames. , ppl: 10.276797641593197] tot_loss[loss=2.296, over 5605222.13 frames. , ppl: 9.930947297948007], batch size: 70
+2022-12-11 17:38:42,833 INFO [train.py:421] (5/8) Epoch 5, batch 35000, loss[loss=2.774, over 700.00 frames. , ppl: 16.01518921185871] tot_loss[loss=2.296, over 5602488.24 frames. , ppl: 9.932590028934749], batch size: 70
+2022-12-11 17:38:42,834 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 17:38:43,595 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.860366938807065
+2022-12-11 17:40:23,355 INFO [train.py:421] (5/8) Epoch 5, batch 35200, loss[loss=2.228, over 7140.00 frames. , ppl: 9.285464748923827] tot_loss[loss=2.298, over 5545416.09 frames. , ppl: 9.952508686925412], batch size: 70
+2022-12-11 17:42:03,912 INFO [train.py:421] (5/8) Epoch 5, batch 35400, loss[loss=2.371, over 1820.00 frames. , ppl: 10.707286181642338] tot_loss[loss=2.299, over 5510800.49 frames. , ppl: 9.964477573652134], batch size: 70
+2022-12-11 17:44:04,292 INFO [train.py:421] (5/8) Epoch 5, batch 35600, loss[loss=2.446, over 1190.00 frames. , ppl: 11.545413307462367] tot_loss[loss=2.3, over 5484101.16 frames. , ppl: 9.975818030668037], batch size: 70
+2022-12-11 17:45:45,025 INFO [train.py:421] (5/8) Epoch 5, batch 35800, loss[loss=2.416, over 1260.00 frames. , ppl: 11.198293590811751] tot_loss[loss=2.3, over 5492976.79 frames. , ppl: 9.972689621340072], batch size: 70
+2022-12-11 17:47:22,413 INFO [train.py:421] (5/8) Epoch 5, batch 36000, loss[loss=2.288, over 3640.00 frames. , ppl: 9.855425200504946] tot_loss[loss=2.3, over 5463984.43 frames. , ppl: 9.979022476546485], batch size: 70
+2022-12-11 17:47:22,413 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 17:47:23,172 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.84657788756117
+2022-12-11 17:49:05,414 INFO [train.py:421] (5/8) Epoch 5, batch 36200, loss[loss=2.246, over 9520.00 frames. , ppl: 9.452410396956497] tot_loss[loss=2.3, over 5463691.31 frames. , ppl: 9.97616533265994], batch size: 70
+2022-12-11 17:50:49,468 INFO [train.py:421] (5/8) Epoch 5, batch 36400, loss[loss=2.266, over 2870.00 frames. , ppl: 9.642587647463689] tot_loss[loss=2.3, over 5476214.22 frames. , ppl: 9.976901432495152], batch size: 70
+2022-12-11 17:52:28,479 INFO [train.py:421] (5/8) Epoch 5, batch 36600, loss[loss=3.206, over 490.00 frames. , ppl: 24.67948243426387] tot_loss[loss=2.299, over 5509744.01 frames. , ppl: 9.96520022610099], batch size: 70
+2022-12-11 17:54:09,081 INFO [train.py:421] (5/8) Epoch 5, batch 36800, loss[loss=2.283, over 2590.00 frames. , ppl: 9.807003728260977] tot_loss[loss=2.3, over 5460051.30 frames. , ppl: 9.976837374191371], batch size: 70
+2022-12-11 17:55:52,424 INFO [train.py:421] (5/8) Epoch 5, batch 37000, loss[loss=2.365, over 1680.00 frames. , ppl: 10.642092797966844] tot_loss[loss=2.301, over 5425971.82 frames. , ppl: 9.98224672017184], batch size: 70
+2022-12-11 17:55:52,424 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 17:55:53,182 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.841108451353056
+2022-12-11 17:57:35,867 INFO [train.py:421] (5/8) Epoch 5, batch 37200, loss[loss=3.101, over 560.00 frames. , ppl: 22.216147669535367] tot_loss[loss=2.301, over 5433049.37 frames. , ppl: 9.981179794070536], batch size: 70
+2022-12-11 17:59:14,079 INFO [train.py:421] (5/8) Epoch 5, batch 37400, loss[loss=3.716, over 420.00 frames. , ppl: 41.09576933975473] tot_loss[loss=2.3, over 5458147.85 frames. , ppl: 9.971939855668433], batch size: 70
+2022-12-11 18:00:54,462 INFO [train.py:421] (5/8) Epoch 5, batch 37600, loss[loss=2.514, over 1260.00 frames. , ppl: 12.348773916555878] tot_loss[loss=2.298, over 5531179.96 frames. , ppl: 9.95161172852178], batch size: 70
+2022-12-11 18:02:36,711 INFO [train.py:421] (5/8) Epoch 5, batch 37800, loss[loss=2.235, over 3640.00 frames. , ppl: 9.34638255437854] tot_loss[loss=2.299, over 5517627.25 frames. , ppl: 9.960219399745526], batch size: 70
+2022-12-11 18:04:14,228 INFO [train.py:421] (5/8) Epoch 5, batch 38000, loss[loss=2.405, over 910.00 frames. , ppl: 11.081786990037614] tot_loss[loss=2.299, over 5513623.98 frames. , ppl: 9.96125789492338], batch size: 70
+2022-12-11 18:04:14,229 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 18:04:14,993 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.84971901441481
+2022-12-11 18:05:55,158 INFO [train.py:421] (5/8) Epoch 5, batch 38200, loss[loss=2.194, over 5040.00 frames. , ppl: 8.970034380800952] tot_loss[loss=2.298, over 5544888.12 frames. , ppl: 9.949880046006795], batch size: 70
+2022-12-11 18:07:31,753 INFO [train.py:421] (5/8) Epoch 5, batch 38400, loss[loss=2.519, over 1050.00 frames. , ppl: 12.412227874851592] tot_loss[loss=2.298, over 5515931.16 frames. , ppl: 9.953317951989693], batch size: 70
+2022-12-11 18:09:11,040 INFO [train.py:421] (5/8) Epoch 5, batch 38600, loss[loss=2.293, over 1610.00 frames. , ppl: 9.900223674813818] tot_loss[loss=2.299, over 5511941.62 frames. , ppl: 9.96413943569133], batch size: 70
+2022-12-11 18:10:51,067 INFO [train.py:421] (5/8) Epoch 5, batch 38800, loss[loss=2.366, over 2170.00 frames. , ppl: 10.652495786271974] tot_loss[loss=2.299, over 5506464.34 frames. , ppl: 9.959934890073612], batch size: 70
+2022-12-11 18:12:31,786 INFO [train.py:421] (5/8) Epoch 5, batch 39000, loss[loss=2.372, over 1750.00 frames. , ppl: 10.723825120575723] tot_loss[loss=2.298, over 5501484.36 frames. , ppl: 9.952629630902873], batch size: 70
+2022-12-11 18:12:31,787 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 18:12:32,547 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.839695691985794
+2022-12-11 18:14:11,128 INFO [train.py:421] (5/8) Epoch 5, batch 39200, loss[loss=2.869, over 630.00 frames. , ppl: 17.61261032758614] tot_loss[loss=2.298, over 5517659.10 frames. , ppl: 9.951121352760167], batch size: 70
+2022-12-11 18:15:49,020 INFO [train.py:421] (5/8) Epoch 5, batch 39400, loss[loss=2.263, over 4620.00 frames. , ppl: 9.610452505817818] tot_loss[loss=2.299, over 5477676.91 frames. , ppl: 9.964199368734283], batch size: 70
+2022-12-11 18:17:29,911 INFO [train.py:421] (5/8) Epoch 5, batch 39600, loss[loss=2.269, over 2590.00 frames. , ppl: 9.674025673212565] tot_loss[loss=2.299, over 5479657.39 frames. , ppl: 9.959697378725194], batch size: 70
+2022-12-11 18:19:09,420 INFO [train.py:421] (5/8) Epoch 5, batch 39800, loss[loss=2.377, over 1890.00 frames. , ppl: 10.777432059977519] tot_loss[loss=2.298, over 5482796.00 frames. , ppl: 9.957560990737269], batch size: 70
+2022-12-11 18:20:46,138 INFO [train.py:421] (5/8) Epoch 5, batch 40000, loss[loss=2.323, over 1960.00 frames. , ppl: 10.202993547783041] tot_loss[loss=2.298, over 5470152.72 frames. , ppl: 9.958851135038142], batch size: 70
+2022-12-11 18:20:46,138 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 18:20:46,898 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.289, over 211138.00 frames. , ppl: 9.863040936604593
+2022-12-11 18:22:26,580 INFO [train.py:421] (5/8) Epoch 5, batch 40200, loss[loss=2.358, over 1960.00 frames. , ppl: 10.571379591542028] tot_loss[loss=2.299, over 5447759.37 frames. , ppl: 9.96008581003017], batch size: 70
+2022-12-11 18:24:04,890 INFO [train.py:421] (5/8) Epoch 5, batch 40400, loss[loss=2.19, over 11340.00 frames. , ppl: 8.936513306433444] tot_loss[loss=2.298, over 5447414.23 frames. , ppl: 9.955882100205658], batch size: 70
+2022-12-11 18:25:44,573 INFO [train.py:421] (5/8) Epoch 5, batch 40600, loss[loss=2.359, over 840.00 frames. , ppl: 10.5778421638153] tot_loss[loss=2.298, over 5444390.60 frames. , ppl: 9.957058196290074], batch size: 70
+2022-12-11 18:27:25,885 INFO [train.py:421] (5/8) Epoch 5, batch 40800, loss[loss=2.561, over 700.00 frames. , ppl: 12.954888590998154] tot_loss[loss=2.3, over 5398904.93 frames. , ppl: 9.971098868532112], batch size: 70
+2022-12-11 18:29:07,897 INFO [train.py:421] (5/8) Epoch 5, batch 41000, loss[loss=2.395, over 1820.00 frames. , ppl: 10.971002144255975] tot_loss[loss=2.3, over 5388927.70 frames. , ppl: 9.97171613202969], batch size: 70
+2022-12-11 18:29:07,898 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 18:29:08,657 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.844880201345818
+2022-12-11 18:30:49,865 INFO [train.py:421] (5/8) Epoch 5, batch 41200, loss[loss=2.232, over 2730.00 frames. , ppl: 9.314083290222724] tot_loss[loss=2.3, over 5415530.57 frames. , ppl: 9.97185842714471], batch size: 70
+2022-12-11 18:32:28,300 INFO [train.py:421] (5/8) Epoch 5, batch 41400, loss[loss=2.322, over 1680.00 frames. , ppl: 10.19428566568919] tot_loss[loss=2.299, over 5433391.22 frames. , ppl: 9.968718835582445], batch size: 70
+2022-12-11 18:34:08,975 INFO [train.py:421] (5/8) Epoch 5, batch 41600, loss[loss=3.228, over 490.00 frames. , ppl: 25.240965547926308] tot_loss[loss=2.3, over 5434548.96 frames. , ppl: 9.973314319599428], batch size: 70
+2022-12-11 18:35:50,122 INFO [train.py:421] (5/8) Epoch 5, batch 41800, loss[loss=3.623, over 420.00 frames. , ppl: 37.461911754955295] tot_loss[loss=2.3, over 5448289.91 frames. , ppl: 9.978588766471823], batch size: 70
+2022-12-11 18:37:33,725 INFO [train.py:421] (5/8) Epoch 5, batch 42000, loss[loss=2.332, over 1680.00 frames. , ppl: 10.298531514968825] tot_loss[loss=2.299, over 5489486.82 frames. , ppl: 9.966542993961273], batch size: 70
+2022-12-11 18:37:33,726 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 18:37:34,472 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.285, over 211138.00 frames. , ppl: 9.829064270772456
+2022-12-11 18:39:16,852 INFO [train.py:421] (5/8) Epoch 5, batch 42200, loss[loss=2.363, over 1540.00 frames. , ppl: 10.624085795412306] tot_loss[loss=2.298, over 5512801.20 frames. , ppl: 9.952970623501137], batch size: 70
+2022-12-11 18:40:58,476 INFO [train.py:421] (5/8) Epoch 5, batch 42400, loss[loss=2.326, over 1330.00 frames. , ppl: 10.236198436802164] tot_loss[loss=2.298, over 5522855.44 frames. , ppl: 9.951317185124278], batch size: 70
+2022-12-11 18:42:36,706 INFO [train.py:421] (5/8) Epoch 5, batch 42600, loss[loss=3.605, over 420.00 frames. , ppl: 36.79523959433037] tot_loss[loss=2.298, over 5518343.64 frames. , ppl: 9.95493277994161], batch size: 70
+2022-12-11 18:44:18,978 INFO [train.py:421] (5/8) Epoch 5, batch 42800, loss[loss=2.242, over 3640.00 frames. , ppl: 9.410079937674576] tot_loss[loss=2.297, over 5565614.73 frames. , ppl: 9.940515029285988], batch size: 70
+2022-12-11 18:45:56,835 INFO [train.py:421] (5/8) Epoch 5, batch 43000, loss[loss=2.361, over 1260.00 frames. , ppl: 10.606720240828407] tot_loss[loss=2.297, over 5542052.26 frames. , ppl: 9.947049477614172], batch size: 70
+2022-12-11 18:45:56,835 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 18:45:57,597 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.842544721854633
+2022-12-11 18:47:37,840 INFO [train.py:421] (5/8) Epoch 5, batch 43200, loss[loss=2.277, over 3570.00 frames. , ppl: 9.743312002622234] tot_loss[loss=2.297, over 5545856.49 frames. , ppl: 9.942325447930013], batch size: 70
+2022-12-11 18:49:19,128 INFO [train.py:421] (5/8) Epoch 5, batch 43400, loss[loss=2.221, over 2800.00 frames. , ppl: 9.216211074095582] tot_loss[loss=2.296, over 5561158.60 frames. , ppl: 9.932100635876928], batch size: 70
+2022-12-11 18:51:01,658 INFO [train.py:421] (5/8) Epoch 5, batch 43600, loss[loss=2.254, over 3640.00 frames. , ppl: 9.52207970402198] tot_loss[loss=2.295, over 5598503.26 frames. , ppl: 9.923415774443555], batch size: 70
+2022-12-11 18:52:42,485 INFO [train.py:421] (5/8) Epoch 5, batch 43800, loss[loss=2.693, over 770.00 frames. , ppl: 14.772875732574036] tot_loss[loss=2.297, over 5563892.33 frames. , ppl: 9.939729097698313], batch size: 70
+2022-12-11 18:54:22,154 INFO [train.py:421] (5/8) Epoch 5, batch 44000, loss[loss=2.341, over 2170.00 frames. , ppl: 10.38890029870889] tot_loss[loss=2.296, over 5554384.07 frames. , ppl: 9.93554893078721], batch size: 70
+2022-12-11 18:54:22,154 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 18:54:22,915 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.849284590078504
+2022-12-11 18:56:01,422 INFO [train.py:421] (5/8) Epoch 5, batch 44200, loss[loss=2.286, over 2450.00 frames. , ppl: 9.831123814027903] tot_loss[loss=2.296, over 5516445.69 frames. , ppl: 9.937059757510323], batch size: 70
+2022-12-11 18:57:41,930 INFO [train.py:421] (5/8) Epoch 5, batch 44400, loss[loss=2.445, over 1400.00 frames. , ppl: 11.533610394327786] tot_loss[loss=2.296, over 5510502.05 frames. , ppl: 9.934941369488042], batch size: 70
+2022-12-11 18:59:21,034 INFO [train.py:421] (5/8) Epoch 5, batch 44600, loss[loss=2.256, over 4620.00 frames. , ppl: 9.541891658978306] tot_loss[loss=2.298, over 5469576.24 frames. , ppl: 9.95136524553748], batch size: 70
+2022-12-11 19:01:02,202 INFO [train.py:421] (5/8) Epoch 5, batch 44800, loss[loss=2.293, over 2870.00 frames. , ppl: 9.907964801219968] tot_loss[loss=2.297, over 5494842.90 frames. , ppl: 9.94284551381694], batch size: 70
+2022-12-11 19:02:40,862 INFO [train.py:421] (5/8) Epoch 5, batch 45000, loss[loss=2.207, over 4620.00 frames. , ppl: 9.089093847259658] tot_loss[loss=2.297, over 5481573.51 frames. , ppl: 9.945842005429084], batch size: 70
+2022-12-11 19:02:40,863 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 19:02:41,620 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.835643508977666
+2022-12-11 19:04:22,576 INFO [train.py:421] (5/8) Epoch 5, batch 45200, loss[loss=2.323, over 2450.00 frames. , ppl: 10.203272133002875] tot_loss[loss=2.296, over 5498126.81 frames. , ppl: 9.939175510441393], batch size: 70
+2022-12-11 19:06:01,368 INFO [train.py:421] (5/8) Epoch 5, batch 45400, loss[loss=2.288, over 4060.00 frames. , ppl: 9.850842713895506] tot_loss[loss=2.297, over 5490534.90 frames. , ppl: 9.940920910948586], batch size: 70
+2022-12-11 19:07:43,219 INFO [train.py:421] (5/8) Epoch 5, batch 45600, loss[loss=2.324, over 2870.00 frames. , ppl: 10.213277771594269] tot_loss[loss=2.295, over 5532457.29 frames. , ppl: 9.927514801178763], batch size: 70
+2022-12-11 19:09:21,551 INFO [train.py:421] (5/8) Epoch 5, batch 45800, loss[loss=2.479, over 1540.00 frames. , ppl: 11.923641095734757] tot_loss[loss=2.294, over 5560081.46 frames. , ppl: 9.918677170504779], batch size: 70
+2022-12-11 19:11:03,159 INFO [train.py:421] (5/8) Epoch 5, batch 46000, loss[loss=2.244, over 2170.00 frames. , ppl: 9.434755757411939] tot_loss[loss=2.294, over 5547399.30 frames. , ppl: 9.919361806074157], batch size: 70
+2022-12-11 19:11:03,159 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 19:11:03,906 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.285, over 211138.00 frames. , ppl: 9.824885591047718
+2022-12-11 19:12:44,299 INFO [train.py:421] (5/8) Epoch 5, batch 46200, loss[loss=2.314, over 4130.00 frames. , ppl: 10.118904033665581] tot_loss[loss=2.295, over 5510631.58 frames. , ppl: 9.92695669902225], batch size: 70
+2022-12-11 19:14:23,630 INFO [train.py:421] (5/8) Epoch 5, batch 46400, loss[loss=2.698, over 700.00 frames. , ppl: 14.851166864490553] tot_loss[loss=2.296, over 5492657.55 frames. , ppl: 9.936507162342107], batch size: 70
+2022-12-11 19:15:58,262 INFO [train.py:421] (5/8) Epoch 5, batch 46600, loss[loss=2.55, over 980.00 frames. , ppl: 12.80332673418466] tot_loss[loss=2.296, over 5470813.97 frames. , ppl: 9.93832008255725], batch size: 70
+2022-12-11 19:17:38,469 INFO [train.py:421] (5/8) Epoch 5, batch 46800, loss[loss=2.189, over 5530.00 frames. , ppl: 8.923911764896665] tot_loss[loss=2.297, over 5456535.69 frames. , ppl: 9.94049227009337], batch size: 70
+2022-12-11 19:19:20,572 INFO [train.py:421] (5/8) Epoch 5, batch 47000, loss[loss=2.362, over 1540.00 frames. , ppl: 10.60686130209383] tot_loss[loss=2.298, over 5417872.61 frames. , ppl: 9.958049429390236], batch size: 70
+2022-12-11 19:19:20,572 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 19:19:21,331 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.837504131725721
+2022-12-11 19:21:03,345 INFO [train.py:421] (5/8) Epoch 5, batch 47200, loss[loss=2.242, over 4060.00 frames. , ppl: 9.416696313655375] tot_loss[loss=2.297, over 5455428.32 frames. , ppl: 9.946601040419209], batch size: 70
+2022-12-11 19:22:43,309 INFO [train.py:421] (5/8) Epoch 5, batch 47400, loss[loss=3.013, over 560.00 frames. , ppl: 20.349800627707058] tot_loss[loss=2.297, over 5430098.96 frames. , ppl: 9.949147317736276], batch size: 70
+2022-12-11 19:24:25,849 INFO [train.py:421] (5/8) Epoch 5, batch 47600, loss[loss=2.255, over 3640.00 frames. , ppl: 9.536534624495363] tot_loss[loss=2.297, over 5461185.84 frames. , ppl: 9.943008315477806], batch size: 70
+2022-12-11 19:26:04,799 INFO [train.py:421] (5/8) Epoch 5, batch 47800, loss[loss=2.53, over 1120.00 frames. , ppl: 12.550585045574874] tot_loss[loss=2.298, over 5444333.73 frames. , ppl: 9.957834524341939], batch size: 70
+2022-12-11 19:27:44,266 INFO [train.py:421] (5/8) Epoch 5, batch 48000, loss[loss=2.324, over 1890.00 frames. , ppl: 10.219184663549248] tot_loss[loss=2.3, over 5390500.94 frames. , ppl: 9.971587252387605], batch size: 70
+2022-12-11 19:27:44,267 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 19:27:45,017 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.847490241584369
+2022-12-11 19:29:31,063 INFO [train.py:421] (5/8) Epoch 5, batch 48200, loss[loss=2.335, over 2870.00 frames. , ppl: 10.3295203834305] tot_loss[loss=2.298, over 5445484.42 frames. , ppl: 9.95307418631389], batch size: 70
+2022-12-11 19:31:11,713 INFO [train.py:421] (5/8) Epoch 5, batch 48400, loss[loss=2.333, over 2310.00 frames. , ppl: 10.303839285723097] tot_loss[loss=2.298, over 5432402.05 frames. , ppl: 9.955271588179404], batch size: 70
+2022-12-11 19:32:50,746 INFO [train.py:421] (5/8) Epoch 5, batch 48600, loss[loss=2.448, over 1890.00 frames. , ppl: 11.564838951051737] tot_loss[loss=2.298, over 5445268.57 frames. , ppl: 9.952978500008696], batch size: 70
+2022-12-11 19:34:30,309 INFO [train.py:421] (5/8) Epoch 5, batch 48800, loss[loss=2.707, over 840.00 frames. , ppl: 14.989798201194741] tot_loss[loss=2.298, over 5456444.78 frames. , ppl: 9.953347076529745], batch size: 70
+2022-12-11 19:36:12,832 INFO [train.py:421] (5/8) Epoch 5, batch 49000, loss[loss=2.219, over 4900.00 frames. , ppl: 9.19748031667321] tot_loss[loss=2.298, over 5452373.64 frames. , ppl: 9.949708566281], batch size: 70
+2022-12-11 19:36:12,833 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 19:36:13,593 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.288, over 211138.00 frames. , ppl: 9.8508430669708
+2022-12-11 19:37:58,713 INFO [train.py:421] (5/8) Epoch 5, batch 49200, loss[loss=2.364, over 1260.00 frames. , ppl: 10.638326637620892] tot_loss[loss=2.297, over 5453347.16 frames. , ppl: 9.943933699730431], batch size: 70
+2022-12-11 19:39:40,049 INFO [train.py:421] (5/8) Epoch 5, batch 49400, loss[loss=2.22, over 13440.00 frames. , ppl: 9.209665214127774] tot_loss[loss=2.295, over 5545618.63 frames. , ppl: 9.92381090427789], batch size: 70
+2022-12-11 19:41:17,703 INFO [train.py:421] (5/8) Epoch 5, batch 49600, loss[loss=2.352, over 1750.00 frames. , ppl: 10.510748891623011] tot_loss[loss=2.295, over 5537642.57 frames. , ppl: 9.924391444749373], batch size: 70
+2022-12-11 19:42:57,230 INFO [train.py:421] (5/8) Epoch 5, batch 49800, loss[loss=2.264, over 3360.00 frames. , ppl: 9.618561204235665] tot_loss[loss=2.294, over 5549627.21 frames. , ppl: 9.918170341495227], batch size: 70
+2022-12-11 19:44:39,906 INFO [train.py:421] (5/8) Epoch 5, batch 50000, loss[loss=2.353, over 1610.00 frames. , ppl: 10.516834957656334] tot_loss[loss=2.294, over 5559762.79 frames. , ppl: 9.913326245593094], batch size: 70
+2022-12-11 19:44:39,906 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 19:44:40,663 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.837191091537502
+2022-12-11 19:46:20,685 INFO [train.py:421] (5/8) Epoch 5, batch 50200, loss[loss=2.297, over 2660.00 frames. , ppl: 9.942167482170019] tot_loss[loss=2.294, over 5563784.88 frames. , ppl: 9.913711590159037], batch size: 70
+2022-12-11 19:48:00,103 INFO [train.py:421] (5/8) Epoch 5, batch 50400, loss[loss=2.317, over 3360.00 frames. , ppl: 10.149282488798137] tot_loss[loss=2.295, over 5562441.67 frames. , ppl: 9.921127717403674], batch size: 70
+2022-12-11 19:49:38,962 INFO [train.py:421] (5/8) Epoch 5, batch 50600, loss[loss=2.215, over 3010.00 frames. , ppl: 9.161743206669595] tot_loss[loss=2.294, over 5554701.73 frames. , ppl: 9.917159490581144], batch size: 70
+2022-12-11 19:51:17,597 INFO [train.py:421] (5/8) Epoch 5, batch 50800, loss[loss=2.235, over 3780.00 frames. , ppl: 9.350403603467411] tot_loss[loss=2.295, over 5551662.31 frames. , ppl: 9.924639098559762], batch size: 70
+2022-12-11 19:52:59,276 INFO [train.py:421] (5/8) Epoch 5, batch 51000, loss[loss=2.202, over 6370.00 frames. , ppl: 9.046896611225852] tot_loss[loss=2.295, over 5558398.41 frames. , ppl: 9.923778779136061], batch size: 70
+2022-12-11 19:52:59,277 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 19:53:00,024 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.288, over 211138.00 frames. , ppl: 9.852051821719042
+2022-12-11 19:54:41,943 INFO [train.py:421] (5/8) Epoch 5, batch 51200, loss[loss=2.341, over 1470.00 frames. , ppl: 10.393609786930245] tot_loss[loss=2.296, over 5534632.81 frames. , ppl: 9.931353314926092], batch size: 70
+2022-12-11 19:56:21,663 INFO [train.py:421] (5/8) Epoch 5, batch 51400, loss[loss=2.238, over 6160.00 frames. , ppl: 9.378291589638582] tot_loss[loss=2.296, over 5535775.16 frames. , ppl: 9.933214552097283], batch size: 70
+2022-12-11 19:58:02,206 INFO [train.py:421] (5/8) Epoch 5, batch 51600, loss[loss=2.54, over 770.00 frames. , ppl: 12.682783860893252] tot_loss[loss=2.297, over 5499107.59 frames. , ppl: 9.943316332292904], batch size: 70
+2022-12-11 19:59:40,709 INFO [train.py:421] (5/8) Epoch 5, batch 51800, loss[loss=2.648, over 1050.00 frames. , ppl: 14.119351037068338] tot_loss[loss=2.297, over 5498401.87 frames. , ppl: 9.94620381896544], batch size: 70
+2022-12-11 20:01:19,072 INFO [train.py:421] (5/8) Epoch 5, batch 52000, loss[loss=2.466, over 1610.00 frames. , ppl: 11.77852505043733] tot_loss[loss=2.299, over 5442064.81 frames. , ppl: 9.96138431539866], batch size: 70
+2022-12-11 20:01:19,072 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 20:01:19,832 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.840256402098536
+2022-12-11 20:03:00,929 INFO [train.py:421] (5/8) Epoch 5, batch 52200, loss[loss=2.3, over 2520.00 frames. , ppl: 9.970543992253178] tot_loss[loss=2.298, over 5485748.75 frames. , ppl: 9.952223692186221], batch size: 70
+2022-12-11 20:04:40,908 INFO [train.py:421] (5/8) Epoch 5, batch 52400, loss[loss=2.557, over 1050.00 frames. , ppl: 12.900716224445086] tot_loss[loss=2.298, over 5485403.37 frames. , ppl: 9.953200671378912], batch size: 70
+2022-12-11 20:06:20,202 INFO [train.py:421] (5/8) Epoch 5, batch 52600, loss[loss=2.617, over 980.00 frames. , ppl: 13.695375281050083] tot_loss[loss=2.299, over 5442522.75 frames. , ppl: 9.960437428832774], batch size: 70
+2022-12-11 20:07:55,988 INFO [train.py:421] (5/8) Epoch 5, batch 52800, loss[loss=2.28, over 3010.00 frames. , ppl: 9.776406992122421] tot_loss[loss=2.299, over 5445600.51 frames. , ppl: 9.964416545773823], batch size: 70
+2022-12-11 20:09:33,171 INFO [train.py:421] (5/8) Epoch 5, batch 53000, loss[loss=2.322, over 1890.00 frames. , ppl: 10.193891936285986] tot_loss[loss=2.299, over 5429908.59 frames. , ppl: 9.962076910563711], batch size: 70
+2022-12-11 20:09:33,172 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 20:09:33,930 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.842675832035669
+2022-12-11 20:11:15,344 INFO [train.py:421] (5/8) Epoch 5, batch 53200, loss[loss=2.629, over 910.00 frames. , ppl: 13.860663728387522] tot_loss[loss=2.299, over 5416291.42 frames. , ppl: 9.962099050544465], batch size: 70
+2022-12-11 20:12:50,615 INFO [train.py:421] (5/8) Epoch 5, batch 53400, loss[loss=2.211, over 5250.00 frames. , ppl: 9.12314119542636] tot_loss[loss=2.297, over 5449355.99 frames. , ppl: 9.949259268657217], batch size: 70
+2022-12-11 20:14:31,569 INFO [train.py:421] (5/8) Epoch 5, batch 53600, loss[loss=2.515, over 1050.00 frames. , ppl: 12.361244408143584] tot_loss[loss=2.298, over 5433986.43 frames. , ppl: 9.955053701476933], batch size: 70
+2022-12-11 20:16:11,473 INFO [train.py:421] (5/8) Epoch 5, batch 53800, loss[loss=2.153, over 6230.00 frames. , ppl: 8.613464190825646] tot_loss[loss=2.296, over 5487953.90 frames. , ppl: 9.937784596913037], batch size: 70
+2022-12-11 20:17:55,393 INFO [train.py:421] (5/8) Epoch 5, batch 54000, loss[loss=2.324, over 2240.00 frames. , ppl: 10.216007602060296] tot_loss[loss=2.295, over 5539459.87 frames. , ppl: 9.92067206420777], batch size: 70
+2022-12-11 20:17:55,394 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 20:17:56,155 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.832751361696745
+2022-12-11 20:19:35,191 INFO [train.py:421] (5/8) Epoch 5, batch 54200, loss[loss=2.198, over 13790.00 frames. , ppl: 9.0036486990237] tot_loss[loss=2.295, over 5513533.01 frames. , ppl: 9.92838100987687], batch size: 70
+2022-12-11 20:21:11,753 INFO [train.py:421] (5/8) Epoch 5, batch 54400, loss[loss=2.338, over 1120.00 frames. , ppl: 10.365056111091898] tot_loss[loss=2.294, over 5600271.20 frames. , ppl: 9.909788642998757], batch size: 70
+2022-12-11 20:22:53,113 INFO [train.py:421] (5/8) Epoch 5, batch 54600, loss[loss=2.281, over 3430.00 frames. , ppl: 9.790291361184662] tot_loss[loss=2.294, over 5580958.58 frames. , ppl: 9.912677079493431], batch size: 70
+2022-12-11 20:24:36,518 INFO [train.py:421] (5/8) Epoch 5, batch 54800, loss[loss=2.45, over 1260.00 frames. , ppl: 11.583416900726283] tot_loss[loss=2.294, over 5597728.65 frames. , ppl: 9.910306844624074], batch size: 70
+2022-12-11 20:26:21,050 INFO [train.py:421] (5/8) Epoch 5, batch 55000, loss[loss=2.438, over 1470.00 frames. , ppl: 11.446342356380447] tot_loss[loss=2.293, over 5596786.90 frames. , ppl: 9.903479808970944], batch size: 70
+2022-12-11 20:26:21,051 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 20:26:21,809 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.833813803126963
+2022-12-11 20:27:59,954 INFO [train.py:421] (5/8) Epoch 5, batch 55200, loss[loss=2.371, over 2030.00 frames. , ppl: 10.712432675431339] tot_loss[loss=2.293, over 5602350.57 frames. , ppl: 9.908017859072658], batch size: 70
+2022-12-11 20:29:36,219 INFO [train.py:421] (5/8) Epoch 5, batch 55400, loss[loss=2.218, over 4620.00 frames. , ppl: 9.190762832578587] tot_loss[loss=2.294, over 5554707.42 frames. , ppl: 9.915029622845097], batch size: 70
+2022-12-11 20:31:18,002 INFO [train.py:421] (5/8) Epoch 5, batch 55600, loss[loss=2.401, over 1540.00 frames. , ppl: 11.03329925828139] tot_loss[loss=2.295, over 5539049.42 frames. , ppl: 9.919887330929358], batch size: 70
+2022-12-11 20:32:57,849 INFO [train.py:421] (5/8) Epoch 5, batch 55800, loss[loss=2.219, over 6930.00 frames. , ppl: 9.19631226057578] tot_loss[loss=2.294, over 5528014.19 frames. , ppl: 9.917815108187753], batch size: 70
+2022-12-11 20:34:40,861 INFO [train.py:421] (5/8) Epoch 5, batch 56000, loss[loss=2.204, over 11060.00 frames. , ppl: 9.064394905277817] tot_loss[loss=2.294, over 5553467.38 frames. , ppl: 9.915400774802402], batch size: 70
+2022-12-11 20:34:40,862 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 20:34:41,592 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.835016101318848
+2022-12-11 20:36:19,925 INFO [train.py:421] (5/8) Epoch 5, batch 56200, loss[loss=2.26, over 5110.00 frames. , ppl: 9.58467822648875] tot_loss[loss=2.295, over 5538323.52 frames. , ppl: 9.919875946285666], batch size: 70
+2022-12-11 20:38:00,524 INFO [train.py:421] (5/8) Epoch 5, batch 56400, loss[loss=2.237, over 3640.00 frames. , ppl: 9.360555691262963] tot_loss[loss=2.295, over 5516776.08 frames. , ppl: 9.923993941211743], batch size: 70
+2022-12-11 20:39:43,937 INFO [train.py:421] (5/8) Epoch 5, batch 56600, loss[loss=2.47, over 1050.00 frames. , ppl: 11.816703057196818] tot_loss[loss=2.294, over 5551593.78 frames. , ppl: 9.914713652931388], batch size: 70
+2022-12-11 20:41:26,617 INFO [train.py:421] (5/8) Epoch 5, batch 56800, loss[loss=2.803, over 770.00 frames. , ppl: 16.49617420608012] tot_loss[loss=2.294, over 5547237.48 frames. , ppl: 9.912363278772135], batch size: 70
+2022-12-11 20:43:05,303 INFO [train.py:421] (5/8) Epoch 5, batch 57000, loss[loss=2.253, over 4970.00 frames. , ppl: 9.519331591027784] tot_loss[loss=2.295, over 5501046.19 frames. , ppl: 9.928647485511952], batch size: 70
+2022-12-11 20:43:05,304 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 20:43:06,065 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.842181992787078
+2022-12-11 20:44:43,547 INFO [train.py:421] (5/8) Epoch 5, batch 57200, loss[loss=2.285, over 3710.00 frames. , ppl: 9.826352493018428] tot_loss[loss=2.297, over 5465967.36 frames. , ppl: 9.942145131502091], batch size: 70
+2022-12-11 20:46:22,290 INFO [train.py:421] (5/8) Epoch 5, batch 57400, loss[loss=2.336, over 1540.00 frames. , ppl: 10.33774209100815] tot_loss[loss=2.298, over 5430078.45 frames. , ppl: 9.95244462361182], batch size: 70
+2022-12-11 20:48:03,932 INFO [train.py:421] (5/8) Epoch 5, batch 57600, loss[loss=2.602, over 840.00 frames. , ppl: 13.491448129600084] tot_loss[loss=2.299, over 5418102.59 frames. , ppl: 9.967877168279415], batch size: 70
+2022-12-11 20:49:40,763 INFO [train.py:421] (5/8) Epoch 5, batch 57800, loss[loss=2.33, over 2030.00 frames. , ppl: 10.277355148589127] tot_loss[loss=2.3, over 5387677.67 frames. , ppl: 9.974963108505236], batch size: 70
+2022-12-11 20:51:21,984 INFO [train.py:421] (5/8) Epoch 5, batch 58000, loss[loss=2.353, over 1750.00 frames. , ppl: 10.514925867133973] tot_loss[loss=2.3, over 5406300.02 frames. , ppl: 9.972149414357734], batch size: 70
+2022-12-11 20:51:21,985 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 20:51:22,729 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.846384059654563
+2022-12-11 20:53:06,173 INFO [train.py:421] (5/8) Epoch 5, batch 58200, loss[loss=2.241, over 4690.00 frames. , ppl: 9.40161544019812] tot_loss[loss=2.301, over 5377457.32 frames. , ppl: 9.97934230268113], batch size: 70
+2022-12-11 20:54:48,411 INFO [train.py:421] (5/8) Epoch 5, batch 58400, loss[loss=2.742, over 700.00 frames. , ppl: 15.525534399687304] tot_loss[loss=2.3, over 5384306.64 frames. , ppl: 9.971969530689831], batch size: 70
+2022-12-11 20:56:30,697 INFO [train.py:421] (5/8) Epoch 5, batch 58600, loss[loss=2.343, over 1750.00 frames. , ppl: 10.415373378133319] tot_loss[loss=2.302, over 5347594.99 frames. , ppl: 9.989430207726], batch size: 70
+2022-12-11 20:58:11,550 INFO [train.py:421] (5/8) Epoch 5, batch 58800, loss[loss=2.285, over 2450.00 frames. , ppl: 9.83048509387268] tot_loss[loss=2.3, over 5384353.51 frames. , ppl: 9.978193381077542], batch size: 70
+2022-12-11 20:59:49,673 INFO [train.py:421] (5/8) Epoch 5, batch 59000, loss[loss=2.407, over 1610.00 frames. , ppl: 11.099439618611177] tot_loss[loss=2.301, over 5362438.82 frames. , ppl: 9.98617135538866], batch size: 70
+2022-12-11 20:59:49,674 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 20:59:50,433 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.833901132185455
+2022-12-11 21:01:27,627 INFO [train.py:421] (5/8) Epoch 5, batch 59200, loss[loss=2.284, over 3010.00 frames. , ppl: 9.81128965968289] tot_loss[loss=2.3, over 5390366.93 frames. , ppl: 9.975891014351522], batch size: 70
+2022-12-11 21:03:06,546 INFO [train.py:421] (5/8) Epoch 5, batch 59400, loss[loss=2.465, over 2870.00 frames. , ppl: 11.759043184619733] tot_loss[loss=2.3, over 5371986.26 frames. , ppl: 9.974283888477219], batch size: 70
+2022-12-11 21:04:48,215 INFO [train.py:421] (5/8) Epoch 5, batch 59600, loss[loss=2.757, over 770.00 frames. , ppl: 15.75735274361315] tot_loss[loss=2.299, over 5392204.61 frames. , ppl: 9.962797462713008], batch size: 70
+2022-12-11 21:06:24,260 INFO [train.py:421] (5/8) Epoch 5, batch 59800, loss[loss=2.522, over 910.00 frames. , ppl: 12.455898188629652] tot_loss[loss=2.299, over 5375861.88 frames. , ppl: 9.968711984827499], batch size: 70
+2022-12-11 21:08:10,094 INFO [train.py:421] (5/8) Epoch 5, batch 60000, loss[loss=2.612, over 910.00 frames. , ppl: 13.620525727410122] tot_loss[loss=2.299, over 5391559.07 frames. , ppl: 9.965979582376853], batch size: 70
+2022-12-11 21:08:10,095 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 21:08:10,853 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.831021137755421
+2022-12-11 21:09:53,969 INFO [train.py:421] (5/8) Epoch 5, batch 60200, loss[loss=3.095, over 560.00 frames. , ppl: 22.085941512714975] tot_loss[loss=2.298, over 5437425.73 frames. , ppl: 9.956605341072692], batch size: 70
+2022-12-11 21:11:35,809 INFO [train.py:421] (5/8) Epoch 5, batch 60400, loss[loss=2.616, over 910.00 frames. , ppl: 13.674968572552034] tot_loss[loss=2.297, over 5439590.36 frames. , ppl: 9.948709313423551], batch size: 70
+2022-12-11 21:13:18,739 INFO [train.py:421] (5/8) Epoch 5, batch 60600, loss[loss=2.338, over 3500.00 frames. , ppl: 10.361810218935117] tot_loss[loss=2.297, over 5457120.00 frames. , ppl: 9.942497431409725], batch size: 70
+2022-12-11 21:14:58,853 INFO [train.py:421] (5/8) Epoch 5, batch 60800, loss[loss=2.19, over 5460.00 frames. , ppl: 8.931441363474054] tot_loss[loss=2.295, over 5527926.68 frames. , ppl: 9.926492939675528], batch size: 70
+2022-12-11 21:16:39,557 INFO [train.py:421] (5/8) Epoch 5, batch 61000, loss[loss=2.292, over 1750.00 frames. , ppl: 9.890393129919936] tot_loss[loss=2.295, over 5496000.56 frames. , ppl: 9.928514801179283], batch size: 70
+2022-12-11 21:16:39,558 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 21:16:40,311 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.284, over 211138.00 frames. , ppl: 9.816234263345795
+2022-12-11 21:18:20,760 INFO [train.py:421] (5/8) Epoch 5, batch 61200, loss[loss=2.373, over 1890.00 frames. , ppl: 10.732749838378021] tot_loss[loss=2.295, over 5494122.55 frames. , ppl: 9.927853361917773], batch size: 70
+2022-12-11 21:20:01,590 INFO [train.py:421] (5/8) Epoch 5, batch 61400, loss[loss=2.373, over 2450.00 frames. , ppl: 10.726392255355846] tot_loss[loss=2.296, over 5498937.67 frames. , ppl: 9.930882277648765], batch size: 70
+2022-12-11 21:21:39,916 INFO [train.py:421] (5/8) Epoch 5, batch 61600, loss[loss=2.227, over 6930.00 frames. , ppl: 9.268143353656592] tot_loss[loss=2.296, over 5504499.63 frames. , ppl: 9.931408357357672], batch size: 70
+2022-12-11 21:23:20,543 INFO [train.py:421] (5/8) Epoch 5, batch 61800, loss[loss=2.697, over 840.00 frames. , ppl: 14.83134798450624] tot_loss[loss=2.297, over 5467801.12 frames. , ppl: 9.940782655765343], batch size: 70
+2022-12-11 21:24:59,391 INFO [train.py:421] (5/8) Epoch 5, batch 62000, loss[loss=2.33, over 2590.00 frames. , ppl: 10.273451805006438] tot_loss[loss=2.297, over 5447195.31 frames. , ppl: 9.947098251536072], batch size: 70
+2022-12-11 21:24:59,392 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 21:25:00,150 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.287, over 211138.00 frames. , ppl: 9.841092429233514
+2022-12-11 21:26:42,942 INFO [train.py:421] (5/8) Epoch 5, batch 62200, loss[loss=2.38, over 1470.00 frames. , ppl: 10.806716179380054] tot_loss[loss=2.296, over 5508600.14 frames. , ppl: 9.92965001439748], batch size: 70
+2022-12-11 21:28:21,527 INFO [train.py:421] (5/8) Epoch 5, batch 62400, loss[loss=2.328, over 2170.00 frames. , ppl: 10.256841932170444] tot_loss[loss=2.297, over 5473074.70 frames. , ppl: 9.943795043648851], batch size: 70
+2022-12-11 21:30:03,693 INFO [train.py:421] (5/8) Epoch 5, batch 62600, loss[loss=2.276, over 4200.00 frames. , ppl: 9.735048826654712] tot_loss[loss=2.296, over 5507430.21 frames. , ppl: 9.938081425799453], batch size: 70
+2022-12-11 21:31:48,069 INFO [train.py:421] (5/8) Epoch 5, batch 62800, loss[loss=2.381, over 1610.00 frames. , ppl: 10.817374125643576] tot_loss[loss=2.297, over 5502150.92 frames. , ppl: 9.942295927628692], batch size: 70
+2022-12-11 21:33:23,566 INFO [train.py:421] (5/8) Epoch 5, batch 63000, loss[loss=2.296, over 2800.00 frames. , ppl: 9.936059166270018] tot_loss[loss=2.297, over 5468837.47 frames. , ppl: 9.948394668203434], batch size: 70
+2022-12-11 21:33:23,567 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 21:33:24,313 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.831210297961553
+2022-12-11 21:35:04,615 INFO [train.py:421] (5/8) Epoch 5, batch 63200, loss[loss=2.385, over 2940.00 frames. , ppl: 10.856085141483579] tot_loss[loss=2.297, over 5493340.49 frames. , ppl: 9.9452917286083], batch size: 70
+2022-12-11 21:36:45,647 INFO [train.py:421] (5/8) Epoch 5, batch 63400, loss[loss=3.156, over 490.00 frames. , ppl: 23.475522366138502] tot_loss[loss=2.297, over 5489373.17 frames. , ppl: 9.943093252689447], batch size: 70
+2022-12-11 21:38:24,053 INFO [train.py:421] (5/8) Epoch 5, batch 63600, loss[loss=2.231, over 6370.00 frames. , ppl: 9.310812999441472] tot_loss[loss=2.297, over 5494631.85 frames. , ppl: 9.945353940225223], batch size: 70
+2022-12-11 21:40:04,571 INFO [train.py:421] (5/8) Epoch 5, batch 63800, loss[loss=2.257, over 5670.00 frames. , ppl: 9.555106142163831] tot_loss[loss=2.296, over 5534204.23 frames. , ppl: 9.938006018951059], batch size: 70
+2022-12-11 21:41:46,269 INFO [train.py:421] (5/8) Epoch 5, batch 64000, loss[loss=2.283, over 3080.00 frames. , ppl: 9.807416833583433] tot_loss[loss=2.296, over 5555375.21 frames. , ppl: 9.930218977026522], batch size: 70
+2022-12-11 21:41:46,269 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 21:41:47,016 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.285, over 211138.00 frames. , ppl: 9.828019797907846
+2022-12-11 21:43:27,812 INFO [train.py:421] (5/8) Epoch 5, batch 64200, loss[loss=2.327, over 3640.00 frames. , ppl: 10.248012460375556] tot_loss[loss=2.295, over 5589968.95 frames. , ppl: 9.921299797014434], batch size: 70
+2022-12-11 21:45:08,808 INFO [train.py:421] (5/8) Epoch 5, batch 64400, loss[loss=2.606, over 910.00 frames. , ppl: 13.545653043291503] tot_loss[loss=2.294, over 5600157.92 frames. , ppl: 9.9186361164846], batch size: 70
+2022-12-11 21:46:48,251 INFO [train.py:421] (5/8) Epoch 5, batch 64600, loss[loss=2.41, over 2170.00 frames. , ppl: 11.138500173348898] tot_loss[loss=2.294, over 5599092.43 frames. , ppl: 9.911851641917261], batch size: 70
+2022-12-11 21:48:30,043 INFO [train.py:421] (5/8) Epoch 5, batch 64800, loss[loss=2.381, over 1750.00 frames. , ppl: 10.814853138766216] tot_loss[loss=2.293, over 5617522.82 frames. , ppl: 9.904692673920701], batch size: 70
+2022-12-11 21:50:08,871 INFO [train.py:421] (5/8) Epoch 5, batch 65000, loss[loss=2.445, over 1820.00 frames. , ppl: 11.526778633017244] tot_loss[loss=2.293, over 5589369.26 frames. , ppl: 9.907890473949486], batch size: 70
+2022-12-11 21:50:08,872 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 21:50:09,631 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.286, over 211138.00 frames. , ppl: 9.835764336858247
+2022-12-11 21:51:51,527 INFO [train.py:421] (5/8) Epoch 5, batch 65200, loss[loss=2.285, over 2310.00 frames. , ppl: 9.828700720909152] tot_loss[loss=2.293, over 5615552.67 frames. , ppl: 9.900158588845454], batch size: 70
+2022-12-11 21:53:29,533 INFO [train.py:421] (5/8) Epoch 5, batch 65400, loss[loss=2.901, over 560.00 frames. , ppl: 18.200981567384744] tot_loss[loss=2.294, over 5598776.22 frames. , ppl: 9.917089835020846], batch size: 70
+2022-12-11 21:55:05,986 INFO [train.py:421] (5/8) Epoch 5, batch 65600, loss[loss=2.325, over 2310.00 frames. , ppl: 10.227449675189957] tot_loss[loss=2.294, over 5589132.18 frames. , ppl: 9.913881831601413], batch size: 70
+2022-12-11 21:56:43,518 INFO [train.py:421] (5/8) Epoch 5, batch 65800, loss[loss=2.576, over 910.00 frames. , ppl: 13.151014660257855] tot_loss[loss=2.293, over 5632170.81 frames. , ppl: 9.907417700997966], batch size: 70
+2022-12-11 21:58:21,357 INFO [train.py:421] (5/8) Epoch 5, batch 66000, loss[loss=2.281, over 2450.00 frames. , ppl: 9.79080342260147] tot_loss[loss=2.293, over 5634074.02 frames. , ppl: 9.902634171971515], batch size: 70
+2022-12-11 21:58:21,358 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 21:58:22,104 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.284, over 211138.00 frames. , ppl: 9.816399892600549
+2022-12-11 22:00:01,630 INFO [train.py:421] (5/8) Epoch 5, batch 66200, loss[loss=2.254, over 2030.00 frames. , ppl: 9.528469691643199] tot_loss[loss=2.294, over 5602887.08 frames. , ppl: 9.9109146983184], batch size: 70
+2022-12-11 22:01:40,816 INFO [train.py:421] (5/8) Epoch 5, batch 66400, loss[loss=2.333, over 1400.00 frames. , ppl: 10.311361780485392] tot_loss[loss=2.294, over 5585614.13 frames. , ppl: 9.913994344775015], batch size: 70
+2022-12-11 22:03:20,546 INFO [train.py:421] (5/8) Epoch 5, batch 66600, loss[loss=2.546, over 770.00 frames. , ppl: 12.761757269641588] tot_loss[loss=2.295, over 5547437.72 frames. , ppl: 9.923937984350188], batch size: 70
+2022-12-11 22:05:00,222 INFO [train.py:421] (5/8) Epoch 5, batch 66800, loss[loss=2.368, over 2100.00 frames. , ppl: 10.671456333819982] tot_loss[loss=2.295, over 5556165.66 frames. , ppl: 9.925843182705371], batch size: 70
+2022-12-11 22:06:37,496 INFO [train.py:421] (5/8) Epoch 5, batch 67000, loss[loss=2.449, over 840.00 frames. , ppl: 11.576665377211256] tot_loss[loss=2.296, over 5538227.25 frames. , ppl: 9.929571983350462], batch size: 70
+2022-12-11 22:06:37,497 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 22:06:38,242 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.284, over 211138.00 frames. , ppl: 9.812880145960932
+2022-12-11 22:08:16,524 INFO [train.py:421] (5/8) Epoch 5, batch 67200, loss[loss=2.269, over 2100.00 frames. , ppl: 9.669660599595968] tot_loss[loss=2.296, over 5502015.86 frames. , ppl: 9.93736804923494], batch size: 70
+2022-12-11 22:09:55,865 INFO [train.py:421] (5/8) Epoch 5, batch 67400, loss[loss=2.552, over 1120.00 frames. , ppl: 12.834785712854632] tot_loss[loss=2.294, over 5573441.88 frames. , ppl: 9.917846349813493], batch size: 70
+2022-12-11 22:11:35,230 INFO [train.py:421] (5/8) Epoch 5, batch 67600, loss[loss=2.363, over 1890.00 frames. , ppl: 10.625208324994677] tot_loss[loss=2.294, over 5549108.87 frames. , ppl: 9.913486006425703], batch size: 70
+2022-12-11 22:13:14,515 INFO [train.py:421] (5/8) Epoch 5, batch 67800, loss[loss=2.255, over 5530.00 frames. , ppl: 9.532191113102913] tot_loss[loss=2.294, over 5547670.98 frames. , ppl: 9.915848491975007], batch size: 70
+2022-12-11 22:14:52,339 INFO [train.py:421] (5/8) Epoch 5, batch 68000, loss[loss=2.569, over 770.00 frames. , ppl: 13.05768279479072] tot_loss[loss=2.294, over 5519954.99 frames. , ppl: 9.918330868702162], batch size: 70
+2022-12-11 22:14:52,340 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 22:14:53,100 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.284, over 211138.00 frames. , ppl: 9.814543261416496
+2022-12-11 22:16:33,228 INFO [train.py:421] (5/8) Epoch 5, batch 68200, loss[loss=2.435, over 1330.00 frames. , ppl: 11.420535066249663] tot_loss[loss=2.293, over 5537935.27 frames. , ppl: 9.908690045304516], batch size: 70
+2022-12-11 22:18:14,672 INFO [train.py:421] (5/8) Epoch 5, batch 68400, loss[loss=2.279, over 2240.00 frames. , ppl: 9.768456363576723] tot_loss[loss=2.293, over 5571541.54 frames. , ppl: 9.902663950802506], batch size: 70
+2022-12-11 22:19:55,895 INFO [train.py:421] (5/8) Epoch 5, batch 68600, loss[loss=2.659, over 1330.00 frames. , ppl: 14.281236756193321] tot_loss[loss=2.293, over 5591941.47 frames. , ppl: 9.900598642663809], batch size: 70
+2022-12-11 22:21:39,031 INFO [train.py:421] (5/8) Epoch 5, batch 68800, loss[loss=2.29, over 3640.00 frames. , ppl: 9.87349020347089] tot_loss[loss=2.292, over 5599776.04 frames. , ppl: 9.898604943198187], batch size: 70
+2022-12-11 22:23:22,190 INFO [train.py:421] (5/8) Epoch 5, batch 69000, loss[loss=2.5, over 980.00 frames. , ppl: 12.18740548423664] tot_loss[loss=2.293, over 5556949.21 frames. , ppl: 9.909042426099541], batch size: 70
+2022-12-11 22:23:22,190 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 22:23:22,952 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.284, over 211138.00 frames. , ppl: 9.816581506837656
+2022-12-11 22:25:00,966 INFO [train.py:421] (5/8) Epoch 5, batch 69200, loss[loss=2.531, over 770.00 frames. , ppl: 12.560122718386936] tot_loss[loss=2.294, over 5524391.22 frames. , ppl: 9.91069250180126], batch size: 70
+2022-12-11 22:26:43,991 INFO [train.py:421] (5/8) Epoch 5, batch 69400, loss[loss=2.668, over 910.00 frames. , ppl: 14.418266833450554] tot_loss[loss=2.293, over 5550247.62 frames. , ppl: 9.906614594185035], batch size: 70
+2022-12-11 22:28:25,073 INFO [train.py:421] (5/8) Epoch 5, batch 69600, loss[loss=2.259, over 2660.00 frames. , ppl: 9.574382344695278] tot_loss[loss=2.294, over 5546273.53 frames. , ppl: 9.914502162020886], batch size: 70
+2022-12-11 22:30:05,465 INFO [train.py:421] (5/8) Epoch 5, batch 69800, loss[loss=2.172, over 3220.00 frames. , ppl: 8.77904940651095] tot_loss[loss=2.295, over 5482750.45 frames. , ppl: 9.924541661883815], batch size: 70
+2022-12-11 22:31:48,577 INFO [train.py:421] (5/8) Epoch 5, batch 70000, loss[loss=2.411, over 2170.00 frames. , ppl: 11.141485603992935] tot_loss[loss=2.295, over 5475599.76 frames. , ppl: 9.921354257593627], batch size: 70
+2022-12-11 22:31:48,578 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 22:31:49,328 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.283, over 211138.00 frames. , ppl: 9.80977109406468
+2022-12-11 22:33:34,154 INFO [train.py:421] (5/8) Epoch 5, batch 70200, loss[loss=2.662, over 840.00 frames. , ppl: 14.328167803980364] tot_loss[loss=2.293, over 5527378.63 frames. , ppl: 9.909149916452364], batch size: 70
+2022-12-11 22:35:14,439 INFO [train.py:421] (5/8) Epoch 5, batch 70400, loss[loss=2.478, over 1050.00 frames. , ppl: 11.911595331301736] tot_loss[loss=2.293, over 5562463.78 frames. , ppl: 9.907395147590028], batch size: 70
+2022-12-11 22:36:54,989 INFO [train.py:421] (5/8) Epoch 5, batch 70600, loss[loss=2.389, over 2170.00 frames. , ppl: 10.898994365288429] tot_loss[loss=2.293, over 5563403.13 frames. , ppl: 9.906646943489097], batch size: 70
+2022-12-11 22:38:32,952 INFO [train.py:421] (5/8) Epoch 5, batch 70800, loss[loss=2.248, over 2590.00 frames. , ppl: 9.469744477004925] tot_loss[loss=2.294, over 5523406.46 frames. , ppl: 9.913596881828223], batch size: 70
+2022-12-11 22:40:16,509 INFO [train.py:421] (5/8) Epoch 5, batch 71000, loss[loss=2.388, over 1470.00 frames. , ppl: 10.890913566968234] tot_loss[loss=2.294, over 5521790.03 frames. , ppl: 9.916366537090461], batch size: 70
+2022-12-11 22:40:16,509 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 22:40:17,255 INFO [train.py:452] (5/8) Epoch 5, validation: loss=2.283, over 211138.00 frames. , ppl: 9.803261396653303
+2022-12-11 22:41:56,396 INFO [train.py:421] (5/8) Epoch 5, batch 71200, loss[loss=2.758, over 630.00 frames. , ppl: 15.766937654521097] tot_loss[loss=2.294, over 5536955.47 frames. , ppl: 9.912957220844344], batch size: 70
+2022-12-11 22:43:36,034 INFO [train.py:421] (5/8) Epoch 5, batch 71400, loss[loss=2.359, over 2310.00 frames. , ppl: 10.583182036030694] tot_loss[loss=2.293, over 5543565.74 frames. , ppl: 9.909415013018014], batch size: 70
+2022-12-11 22:45:15,377 INFO [train.py:421] (5/8) Epoch 5, batch 71600, loss[loss=2.577, over 1050.00 frames. , ppl: 13.151624589211695] tot_loss[loss=2.291, over 5623258.13 frames. , ppl: 9.882864054576979], batch size: 70
+2022-12-11 22:46:51,074 INFO [train.py:421] (5/8) Epoch 5, batch 71800, loss[loss=2.266, over 2870.00 frames. , ppl: 9.643990394785545] tot_loss[loss=2.291, over 5602498.59 frames. , ppl: 9.886398819387784], batch size: 70
+2022-12-11 22:48:06,680 INFO [train.py:421] (5/8) Epoch 6, batch 0, loss[loss=2.377, over 2030.00 frames. , ppl: 10.76991770891089] tot_loss[loss=2.377, over 2030.00 frames. , ppl: 10.76991770891089], batch size: 70
+2022-12-11 22:49:46,706 INFO [train.py:421] (5/8) Epoch 6, batch 200, loss[loss=2.187, over 5040.00 frames. , ppl: 8.904192871208618] tot_loss[loss=2.286, over 548467.30 frames. , ppl: 9.833152639188576], batch size: 70
+2022-12-11 22:51:26,944 INFO [train.py:421] (5/8) Epoch 6, batch 400, loss[loss=2.149, over 8470.00 frames. , ppl: 8.574368915473267] tot_loss[loss=2.288, over 1000621.68 frames. , ppl: 9.857854696793307], batch size: 70
+2022-12-11 22:53:08,635 INFO [train.py:421] (5/8) Epoch 6, batch 600, loss[loss=3.252, over 490.00 frames. , ppl: 25.849652656616282] tot_loss[loss=2.29, over 1425620.16 frames. , ppl: 9.875123035465933], batch size: 70
+2022-12-11 22:54:51,865 INFO [train.py:421] (5/8) Epoch 6, batch 800, loss[loss=2.179, over 7210.00 frames. , ppl: 8.833567625198949] tot_loss[loss=2.284, over 1872431.14 frames. , ppl: 9.820431201488303], batch size: 70
+2022-12-11 22:56:35,370 INFO [train.py:421] (5/8) Epoch 6, batch 1000, loss[loss=2.477, over 1470.00 frames. , ppl: 11.900626741861743] tot_loss[loss=2.287, over 2206095.37 frames. , ppl: 9.84502414753925], batch size: 70
+2022-12-11 22:56:35,371 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 22:56:36,121 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.284, over 211138.00 frames. , ppl: 9.816725347697853
+2022-12-11 22:58:21,105 INFO [train.py:421] (5/8) Epoch 6, batch 1200, loss[loss=2.215, over 8260.00 frames. , ppl: 9.157903027349139] tot_loss[loss=2.286, over 2552614.18 frames. , ppl: 9.838261207987431], batch size: 70
+2022-12-11 22:59:58,474 INFO [train.py:421] (5/8) Epoch 6, batch 1400, loss[loss=3.055, over 560.00 frames. , ppl: 21.21800502989271] tot_loss[loss=2.288, over 2803234.59 frames. , ppl: 9.857221697172564], batch size: 70
+2022-12-11 23:01:38,948 INFO [train.py:421] (5/8) Epoch 6, batch 1600, loss[loss=2.481, over 1050.00 frames. , ppl: 11.958324435826887] tot_loss[loss=2.289, over 3043557.22 frames. , ppl: 9.864157857234432], batch size: 70
+2022-12-11 23:03:19,723 INFO [train.py:421] (5/8) Epoch 6, batch 1800, loss[loss=2.179, over 6580.00 frames. , ppl: 8.833129703481868] tot_loss[loss=2.289, over 3269191.53 frames. , ppl: 9.8630756647565], batch size: 70
+2022-12-11 23:04:57,206 INFO [train.py:421] (5/8) Epoch 6, batch 2000, loss[loss=2.431, over 1680.00 frames. , ppl: 11.371175827861233] tot_loss[loss=2.289, over 3483819.94 frames. , ppl: 9.862875305157036], batch size: 70
+2022-12-11 23:04:57,206 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 23:04:57,952 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.286, over 211138.00 frames. , ppl: 9.840409328583648
+2022-12-11 23:06:35,144 INFO [train.py:421] (5/8) Epoch 6, batch 2200, loss[loss=2.477, over 1260.00 frames. , ppl: 11.903930558248517] tot_loss[loss=2.289, over 3685491.18 frames. , ppl: 9.865711223706592], batch size: 70
+2022-12-11 23:08:19,874 INFO [train.py:421] (5/8) Epoch 6, batch 2400, loss[loss=2.534, over 980.00 frames. , ppl: 12.605683702460247] tot_loss[loss=2.29, over 3854678.48 frames. , ppl: 9.877552974104404], batch size: 70
+2022-12-11 23:10:02,623 INFO [train.py:421] (5/8) Epoch 6, batch 2600, loss[loss=2.477, over 1750.00 frames. , ppl: 11.907062320919254] tot_loss[loss=2.29, over 3995530.39 frames. , ppl: 9.870331498029525], batch size: 70
+2022-12-11 23:11:44,591 INFO [train.py:421] (5/8) Epoch 6, batch 2800, loss[loss=2.252, over 2800.00 frames. , ppl: 9.503266368757329] tot_loss[loss=2.288, over 4134355.52 frames. , ppl: 9.859547053486775], batch size: 70
+2022-12-11 23:13:24,433 INFO [train.py:421] (5/8) Epoch 6, batch 3000, loss[loss=2.313, over 1680.00 frames. , ppl: 10.102460599181727] tot_loss[loss=2.29, over 4206900.08 frames. , ppl: 9.870120266088549], batch size: 70
+2022-12-11 23:13:24,433 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 23:13:25,195 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.285, over 211138.00 frames. , ppl: 9.824575860616017
+2022-12-11 23:15:01,076 INFO [train.py:421] (5/8) Epoch 6, batch 3200, loss[loss=3.279, over 490.00 frames. , ppl: 26.53914346288486] tot_loss[loss=2.289, over 4328584.75 frames. , ppl: 9.865956412587016], batch size: 70
+2022-12-11 23:16:39,611 INFO [train.py:421] (5/8) Epoch 6, batch 3400, loss[loss=2.236, over 4620.00 frames. , ppl: 9.359299055317445] tot_loss[loss=2.291, over 4375650.38 frames. , ppl: 9.888992653679303], batch size: 70
+2022-12-11 23:18:19,909 INFO [train.py:421] (5/8) Epoch 6, batch 3600, loss[loss=2.371, over 1820.00 frames. , ppl: 10.70814512925654] tot_loss[loss=2.292, over 4449567.75 frames. , ppl: 9.899257319848818], batch size: 70
+2022-12-11 23:19:59,208 INFO [train.py:421] (5/8) Epoch 6, batch 3800, loss[loss=2.273, over 4410.00 frames. , ppl: 9.70474616932288] tot_loss[loss=2.292, over 4585506.41 frames. , ppl: 9.893476928440023], batch size: 70
+2022-12-11 23:21:41,563 INFO [train.py:421] (5/8) Epoch 6, batch 4000, loss[loss=2.288, over 2240.00 frames. , ppl: 9.853594843241387] tot_loss[loss=2.29, over 4737686.21 frames. , ppl: 9.874357803199498], batch size: 70
+2022-12-11 23:21:41,563 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 23:21:42,322 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.808357026550453
+2022-12-11 23:23:26,050 INFO [train.py:421] (5/8) Epoch 6, batch 4200, loss[loss=2.648, over 630.00 frames. , ppl: 14.123711480093448] tot_loss[loss=2.291, over 4796725.54 frames. , ppl: 9.88182470745029], batch size: 70
+2022-12-11 23:25:03,905 INFO [train.py:421] (5/8) Epoch 6, batch 4400, loss[loss=2.402, over 1960.00 frames. , ppl: 11.048689926472992] tot_loss[loss=2.291, over 4856955.01 frames. , ppl: 9.880663950097615], batch size: 70
+2022-12-11 23:26:43,142 INFO [train.py:421] (5/8) Epoch 6, batch 4600, loss[loss=2.413, over 1750.00 frames. , ppl: 11.16264997470716] tot_loss[loss=2.289, over 4949822.18 frames. , ppl: 9.863878249985389], batch size: 70
+2022-12-11 23:28:17,781 INFO [train.py:421] (5/8) Epoch 6, batch 4800, loss[loss=3.532, over 420.00 frames. , ppl: 34.187201251397745] tot_loss[loss=2.289, over 5007963.43 frames. , ppl: 9.866739110768894], batch size: 70
+2022-12-11 23:29:55,659 INFO [train.py:421] (5/8) Epoch 6, batch 5000, loss[loss=2.261, over 4550.00 frames. , ppl: 9.589798771955811] tot_loss[loss=2.289, over 5035538.10 frames. , ppl: 9.864422480626299], batch size: 70
+2022-12-11 23:29:55,660 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 23:29:56,411 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.285, over 211138.00 frames. , ppl: 9.82324543954202
+2022-12-11 23:31:38,998 INFO [train.py:421] (5/8) Epoch 6, batch 5200, loss[loss=2.359, over 1470.00 frames. , ppl: 10.5811423689579] tot_loss[loss=2.288, over 5103630.45 frames. , ppl: 9.853695835175541], batch size: 70
+2022-12-11 23:33:16,592 INFO [train.py:421] (5/8) Epoch 6, batch 5400, loss[loss=2.203, over 6510.00 frames. , ppl: 9.056451462733106] tot_loss[loss=2.289, over 5110263.55 frames. , ppl: 9.862802734612353], batch size: 70
+2022-12-11 23:34:57,715 INFO [train.py:421] (5/8) Epoch 6, batch 5600, loss[loss=2.29, over 2590.00 frames. , ppl: 9.871217259514193] tot_loss[loss=2.287, over 5212339.97 frames. , ppl: 9.843878261462603], batch size: 70
+2022-12-11 23:36:40,264 INFO [train.py:421] (5/8) Epoch 6, batch 5800, loss[loss=2.25, over 3990.00 frames. , ppl: 9.48605707252544] tot_loss[loss=2.285, over 5263370.07 frames. , ppl: 9.829938426517444], batch size: 70
+2022-12-11 23:38:22,677 INFO [train.py:421] (5/8) Epoch 6, batch 6000, loss[loss=2.411, over 1680.00 frames. , ppl: 11.149461431829334] tot_loss[loss=2.286, over 5312610.81 frames. , ppl: 9.832221163896557], batch size: 70
+2022-12-11 23:38:22,678 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 23:38:23,436 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.284, over 211138.00 frames. , ppl: 9.813196769801683
+2022-12-11 23:40:02,703 INFO [train.py:421] (5/8) Epoch 6, batch 6200, loss[loss=2.386, over 1960.00 frames. , ppl: 10.873911608131982] tot_loss[loss=2.286, over 5313999.50 frames. , ppl: 9.839820848589321], batch size: 70
+2022-12-11 23:41:42,417 INFO [train.py:421] (5/8) Epoch 6, batch 6400, loss[loss=2.352, over 1820.00 frames. , ppl: 10.50693190364022] tot_loss[loss=2.286, over 5349443.42 frames. , ppl: 9.839180194807415], batch size: 70
+2022-12-11 23:43:22,494 INFO [train.py:421] (5/8) Epoch 6, batch 6600, loss[loss=2.409, over 1050.00 frames. , ppl: 11.120440231829917] tot_loss[loss=2.287, over 5358854.86 frames. , ppl: 9.841483246544287], batch size: 70
+2022-12-11 23:45:02,938 INFO [train.py:421] (5/8) Epoch 6, batch 6800, loss[loss=2.321, over 2450.00 frames. , ppl: 10.182433833918292] tot_loss[loss=2.287, over 5390450.63 frames. , ppl: 9.845522264935706], batch size: 70
+2022-12-11 23:46:38,670 INFO [train.py:421] (5/8) Epoch 6, batch 7000, loss[loss=2.245, over 5320.00 frames. , ppl: 9.4419205476344] tot_loss[loss=2.286, over 5432436.77 frames. , ppl: 9.839103283223954], batch size: 70
+2022-12-11 23:46:38,671 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 23:46:39,429 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.285, over 211138.00 frames. , ppl: 9.822367314949352
+2022-12-11 23:48:19,181 INFO [train.py:421] (5/8) Epoch 6, batch 7200, loss[loss=2.357, over 1820.00 frames. , ppl: 10.559105755649641] tot_loss[loss=2.286, over 5461800.55 frames. , ppl: 9.832123515483849], batch size: 70
+2022-12-11 23:50:03,958 INFO [train.py:421] (5/8) Epoch 6, batch 7400, loss[loss=2.214, over 4620.00 frames. , ppl: 9.155606833982246] tot_loss[loss=2.285, over 5481051.92 frames. , ppl: 9.829484362587928], batch size: 70
+2022-12-11 23:51:43,015 INFO [train.py:421] (5/8) Epoch 6, batch 7600, loss[loss=2.285, over 2380.00 frames. , ppl: 9.83030882501964] tot_loss[loss=2.284, over 5519366.43 frames. , ppl: 9.820765170184295], batch size: 70
+2022-12-11 23:53:23,567 INFO [train.py:421] (5/8) Epoch 6, batch 7800, loss[loss=2.418, over 1750.00 frames. , ppl: 11.222500754765015] tot_loss[loss=2.285, over 5547873.17 frames. , ppl: 9.821739354694907], batch size: 70
+2022-12-11 23:55:07,598 INFO [train.py:421] (5/8) Epoch 6, batch 8000, loss[loss=2.287, over 1120.00 frames. , ppl: 9.842283985289901] tot_loss[loss=2.286, over 5519801.16 frames. , ppl: 9.830950400748499], batch size: 70
+2022-12-11 23:55:07,598 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-11 23:55:08,354 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.793563654653626
+2022-12-11 23:56:47,538 INFO [train.py:421] (5/8) Epoch 6, batch 8200, loss[loss=2.275, over 2100.00 frames. , ppl: 9.731095218816652] tot_loss[loss=2.286, over 5492448.32 frames. , ppl: 9.838457987802379], batch size: 70
+2022-12-11 23:58:28,579 INFO [train.py:421] (5/8) Epoch 6, batch 8400, loss[loss=2.403, over 2800.00 frames. , ppl: 11.05501308930997] tot_loss[loss=2.287, over 5486649.97 frames. , ppl: 9.84154250898645], batch size: 70
+2022-12-12 00:00:05,798 INFO [train.py:421] (5/8) Epoch 6, batch 8600, loss[loss=2.477, over 1820.00 frames. , ppl: 11.908181727350133] tot_loss[loss=2.286, over 5522700.54 frames. , ppl: 9.835426080684268], batch size: 70
+2022-12-12 00:01:46,887 INFO [train.py:421] (5/8) Epoch 6, batch 8800, loss[loss=2.37, over 1680.00 frames. , ppl: 10.698122021486306] tot_loss[loss=2.287, over 5498249.92 frames. , ppl: 9.840621892597454], batch size: 70
+2022-12-12 00:03:27,014 INFO [train.py:421] (5/8) Epoch 6, batch 9000, loss[loss=4.279, over 350.00 frames. , ppl: 72.18958826781676] tot_loss[loss=2.285, over 5569277.73 frames. , ppl: 9.826443718068624], batch size: 70
+2022-12-12 00:03:27,015 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 00:03:27,760 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.285, over 211138.00 frames. , ppl: 9.82595299982713
+2022-12-12 00:05:06,483 INFO [train.py:421] (5/8) Epoch 6, batch 9200, loss[loss=2.511, over 1400.00 frames. , ppl: 12.314416047280302] tot_loss[loss=2.285, over 5596098.03 frames. , ppl: 9.827020022846643], batch size: 70
+2022-12-12 00:06:43,079 INFO [train.py:421] (5/8) Epoch 6, batch 9400, loss[loss=2.237, over 3850.00 frames. , ppl: 9.36343153700521] tot_loss[loss=2.285, over 5588462.46 frames. , ppl: 9.824132363410927], batch size: 70
+2022-12-12 00:08:21,540 INFO [train.py:421] (5/8) Epoch 6, batch 9600, loss[loss=2.72, over 700.00 frames. , ppl: 15.174814341900044] tot_loss[loss=2.286, over 5588465.89 frames. , ppl: 9.834167004122978], batch size: 70
+2022-12-12 00:10:01,616 INFO [train.py:421] (5/8) Epoch 6, batch 9800, loss[loss=2.345, over 2380.00 frames. , ppl: 10.435771409210057] tot_loss[loss=2.287, over 5565879.22 frames. , ppl: 9.844367422684055], batch size: 70
+2022-12-12 00:11:46,145 INFO [train.py:421] (5/8) Epoch 6, batch 10000, loss[loss=2.155, over 6720.00 frames. , ppl: 8.62620972159501] tot_loss[loss=2.286, over 5579686.38 frames. , ppl: 9.839253837433821], batch size: 70
+2022-12-12 00:11:46,146 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 00:11:46,874 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.804028982352971
+2022-12-12 00:13:23,522 INFO [train.py:421] (5/8) Epoch 6, batch 10200, loss[loss=2.352, over 2030.00 frames. , ppl: 10.510775791472753] tot_loss[loss=2.287, over 5571850.78 frames. , ppl: 9.846072182199979], batch size: 70
+2022-12-12 00:15:02,150 INFO [train.py:421] (5/8) Epoch 6, batch 10400, loss[loss=2.215, over 5600.00 frames. , ppl: 9.162450820999776] tot_loss[loss=2.288, over 5557886.12 frames. , ppl: 9.854195019948396], batch size: 70
+2022-12-12 00:16:43,020 INFO [train.py:421] (5/8) Epoch 6, batch 10600, loss[loss=2.22, over 6370.00 frames. , ppl: 9.211061484817595] tot_loss[loss=2.287, over 5581748.83 frames. , ppl: 9.849066573024649], batch size: 70
+2022-12-12 00:18:24,429 INFO [train.py:421] (5/8) Epoch 6, batch 10800, loss[loss=2.352, over 2240.00 frames. , ppl: 10.510577330855154] tot_loss[loss=2.286, over 5589298.01 frames. , ppl: 9.837321704169476], batch size: 70
+2022-12-12 00:20:06,164 INFO [train.py:421] (5/8) Epoch 6, batch 11000, loss[loss=2.335, over 2100.00 frames. , ppl: 10.32458552094718] tot_loss[loss=2.286, over 5617890.87 frames. , ppl: 9.832490559273062], batch size: 70
+2022-12-12 00:20:06,165 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 00:20:06,914 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.804163932742426
+2022-12-12 00:21:45,909 INFO [train.py:421] (5/8) Epoch 6, batch 11200, loss[loss=2.291, over 3500.00 frames. , ppl: 9.889133454435344] tot_loss[loss=2.286, over 5598962.28 frames. , ppl: 9.839440847590115], batch size: 70
+2022-12-12 00:23:23,935 INFO [train.py:421] (5/8) Epoch 6, batch 11400, loss[loss=2.233, over 2800.00 frames. , ppl: 9.326924019701977] tot_loss[loss=2.288, over 5544453.92 frames. , ppl: 9.85259718042156], batch size: 70
+2022-12-12 00:25:02,700 INFO [train.py:421] (5/8) Epoch 6, batch 11600, loss[loss=2.724, over 700.00 frames. , ppl: 15.237793765209563] tot_loss[loss=2.288, over 5532996.32 frames. , ppl: 9.852236197356598], batch size: 70
+2022-12-12 00:26:41,492 INFO [train.py:421] (5/8) Epoch 6, batch 11800, loss[loss=2.189, over 8470.00 frames. , ppl: 8.925399465566374] tot_loss[loss=2.288, over 5493336.29 frames. , ppl: 9.85738069078967], batch size: 70
+2022-12-12 00:28:20,379 INFO [train.py:421] (5/8) Epoch 6, batch 12000, loss[loss=2.328, over 2730.00 frames. , ppl: 10.25237302353895] tot_loss[loss=2.289, over 5451011.04 frames. , ppl: 9.866739303995159], batch size: 70
+2022-12-12 00:28:20,379 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 00:28:21,131 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.80987272894533
+2022-12-12 00:30:01,072 INFO [train.py:421] (5/8) Epoch 6, batch 12200, loss[loss=2.275, over 6650.00 frames. , ppl: 9.726470561036214] tot_loss[loss=2.288, over 5475377.83 frames. , ppl: 9.858244045910949], batch size: 70
+2022-12-12 00:31:40,378 INFO [train.py:421] (5/8) Epoch 6, batch 12400, loss[loss=2.168, over 7000.00 frames. , ppl: 8.74018748337988] tot_loss[loss=2.288, over 5468100.05 frames. , ppl: 9.856282360839618], batch size: 70
+2022-12-12 00:33:20,724 INFO [train.py:421] (5/8) Epoch 6, batch 12600, loss[loss=2.935, over 700.00 frames. , ppl: 18.825330580772814] tot_loss[loss=2.287, over 5523133.13 frames. , ppl: 9.845962357413503], batch size: 70
+2022-12-12 00:35:01,335 INFO [train.py:421] (5/8) Epoch 6, batch 12800, loss[loss=2.378, over 1820.00 frames. , ppl: 10.781503545835577] tot_loss[loss=2.287, over 5557336.28 frames. , ppl: 9.840826892831824], batch size: 70
+2022-12-12 00:36:42,459 INFO [train.py:421] (5/8) Epoch 6, batch 13000, loss[loss=2.878, over 630.00 frames. , ppl: 17.78725003270781] tot_loss[loss=2.287, over 5568688.54 frames. , ppl: 9.84292714398428], batch size: 70
+2022-12-12 00:36:42,460 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 00:36:43,214 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.805265371495329
+2022-12-12 00:38:26,661 INFO [train.py:421] (5/8) Epoch 6, batch 13200, loss[loss=2.567, over 1260.00 frames. , ppl: 13.02237952554298] tot_loss[loss=2.287, over 5581846.67 frames. , ppl: 9.840823358549263], batch size: 70
+2022-12-12 00:40:05,803 INFO [train.py:421] (5/8) Epoch 6, batch 13400, loss[loss=2.329, over 2450.00 frames. , ppl: 10.263355151549192] tot_loss[loss=2.287, over 5577462.79 frames. , ppl: 9.844136134719928], batch size: 70
+2022-12-12 00:41:44,769 INFO [train.py:421] (5/8) Epoch 6, batch 13600, loss[loss=2.314, over 2660.00 frames. , ppl: 10.116335855929824] tot_loss[loss=2.288, over 5525364.29 frames. , ppl: 9.853886395489164], batch size: 70
+2022-12-12 00:43:25,849 INFO [train.py:421] (5/8) Epoch 6, batch 13800, loss[loss=2.304, over 1890.00 frames. , ppl: 10.009304479045513] tot_loss[loss=2.287, over 5530500.83 frames. , ppl: 9.847400587938084], batch size: 70
+2022-12-12 00:45:05,054 INFO [train.py:421] (5/8) Epoch 6, batch 14000, loss[loss=2.366, over 2100.00 frames. , ppl: 10.652067936619392] tot_loss[loss=2.287, over 5557036.67 frames. , ppl: 9.845486945849768], batch size: 70
+2022-12-12 00:45:05,055 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 00:45:05,804 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.285, over 211138.00 frames. , ppl: 9.822797644344924
+2022-12-12 00:46:45,803 INFO [train.py:421] (5/8) Epoch 6, batch 14200, loss[loss=2.47, over 1470.00 frames. , ppl: 11.824729832846534] tot_loss[loss=2.286, over 5598794.48 frames. , ppl: 9.833762824757217], batch size: 70
+2022-12-12 00:48:26,187 INFO [train.py:421] (5/8) Epoch 6, batch 14400, loss[loss=2.268, over 2940.00 frames. , ppl: 9.664840075870716] tot_loss[loss=2.285, over 5609160.82 frames. , ppl: 9.826723349136474], batch size: 70
+2022-12-12 00:50:08,143 INFO [train.py:421] (5/8) Epoch 6, batch 14600, loss[loss=2.272, over 2380.00 frames. , ppl: 9.6972961401564] tot_loss[loss=2.286, over 5585347.88 frames. , ppl: 9.832562218750518], batch size: 70
+2022-12-12 00:51:47,347 INFO [train.py:421] (5/8) Epoch 6, batch 14800, loss[loss=2.726, over 630.00 frames. , ppl: 15.270603982719003] tot_loss[loss=2.286, over 5580036.03 frames. , ppl: 9.838858904210124], batch size: 70
+2022-12-12 00:53:30,909 INFO [train.py:421] (5/8) Epoch 6, batch 15000, loss[loss=2.229, over 10220.00 frames. , ppl: 9.287067459733219] tot_loss[loss=2.286, over 5611373.65 frames. , ppl: 9.831876885356452], batch size: 70
+2022-12-12 00:53:30,909 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 00:53:31,639 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.800679035400885
+2022-12-12 00:55:13,337 INFO [train.py:421] (5/8) Epoch 6, batch 15200, loss[loss=2.995, over 560.00 frames. , ppl: 19.98210413483633] tot_loss[loss=2.286, over 5596064.02 frames. , ppl: 9.839681038978147], batch size: 70
+2022-12-12 00:56:51,707 INFO [train.py:421] (5/8) Epoch 6, batch 15400, loss[loss=2.611, over 980.00 frames. , ppl: 13.615156675295735] tot_loss[loss=2.288, over 5559289.19 frames. , ppl: 9.853949394259462], batch size: 70
+2022-12-12 00:58:34,899 INFO [train.py:421] (5/8) Epoch 6, batch 15600, loss[loss=2.239, over 5810.00 frames. , ppl: 9.385059180730952] tot_loss[loss=2.287, over 5598682.32 frames. , ppl: 9.848738585960424], batch size: 70
+2022-12-12 01:00:14,988 INFO [train.py:421] (5/8) Epoch 6, batch 15800, loss[loss=2.261, over 7630.00 frames. , ppl: 9.593779252679164] tot_loss[loss=2.287, over 5585663.93 frames. , ppl: 9.840500894242165], batch size: 70
+2022-12-12 01:01:54,276 INFO [train.py:421] (5/8) Epoch 6, batch 16000, loss[loss=2.235, over 7420.00 frames. , ppl: 9.34504912871623] tot_loss[loss=2.287, over 5596735.22 frames. , ppl: 9.84417827683614], batch size: 70
+2022-12-12 01:01:54,276 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 01:01:55,007 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.805294396587579
+2022-12-12 01:03:38,892 INFO [train.py:421] (5/8) Epoch 6, batch 16200, loss[loss=2.545, over 1120.00 frames. , ppl: 12.745657784411774] tot_loss[loss=2.287, over 5615055.76 frames. , ppl: 9.844484743505816], batch size: 70
+2022-12-12 01:05:19,365 INFO [train.py:421] (5/8) Epoch 6, batch 16400, loss[loss=2.276, over 3920.00 frames. , ppl: 9.739655581041024] tot_loss[loss=2.286, over 5627827.85 frames. , ppl: 9.832515120641265], batch size: 70
+2022-12-12 01:07:02,722 INFO [train.py:421] (5/8) Epoch 6, batch 16600, loss[loss=2.355, over 2450.00 frames. , ppl: 10.5353485201087] tot_loss[loss=2.288, over 5574091.29 frames. , ppl: 9.850756932126446], batch size: 70
+2022-12-12 01:08:42,211 INFO [train.py:421] (5/8) Epoch 6, batch 16800, loss[loss=2.31, over 1470.00 frames. , ppl: 10.076854072324204] tot_loss[loss=2.288, over 5564360.29 frames. , ppl: 9.853413556297761], batch size: 70
+2022-12-12 01:10:19,698 INFO [train.py:421] (5/8) Epoch 6, batch 17000, loss[loss=2.346, over 2100.00 frames. , ppl: 10.440844254023789] tot_loss[loss=2.287, over 5572616.59 frames. , ppl: 9.845594338098755], batch size: 70
+2022-12-12 01:10:19,698 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 01:10:20,459 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.80433371167732
+2022-12-12 01:12:00,281 INFO [train.py:421] (5/8) Epoch 6, batch 17200, loss[loss=2.462, over 1330.00 frames. , ppl: 11.731876728101224] tot_loss[loss=2.287, over 5567380.19 frames. , ppl: 9.8438462720446], batch size: 70
+2022-12-12 01:13:40,302 INFO [train.py:421] (5/8) Epoch 6, batch 17400, loss[loss=2.357, over 2170.00 frames. , ppl: 10.555137658299948] tot_loss[loss=2.288, over 5550267.46 frames. , ppl: 9.850737266827425], batch size: 70
+2022-12-12 01:15:20,953 INFO [train.py:421] (5/8) Epoch 6, batch 17600, loss[loss=2.24, over 7700.00 frames. , ppl: 9.397292074725327] tot_loss[loss=2.289, over 5503920.20 frames. , ppl: 9.868252390414279], batch size: 70
+2022-12-12 01:17:03,065 INFO [train.py:421] (5/8) Epoch 6, batch 17800, loss[loss=2.325, over 1890.00 frames. , ppl: 10.228006486647297] tot_loss[loss=2.289, over 5509391.29 frames. , ppl: 9.864792935611053], batch size: 70
+2022-12-12 01:18:44,985 INFO [train.py:421] (5/8) Epoch 6, batch 18000, loss[loss=2.276, over 3640.00 frames. , ppl: 9.739064310538218] tot_loss[loss=2.288, over 5531803.50 frames. , ppl: 9.857140697407283], batch size: 70
+2022-12-12 01:18:44,985 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 01:18:45,748 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.802139871930427
+2022-12-12 01:20:22,555 INFO [train.py:421] (5/8) Epoch 6, batch 18200, loss[loss=2.781, over 630.00 frames. , ppl: 16.14005482554294] tot_loss[loss=2.289, over 5523872.37 frames. , ppl: 9.863402275909838], batch size: 70
+2022-12-12 01:22:05,359 INFO [train.py:421] (5/8) Epoch 6, batch 18400, loss[loss=2.239, over 3990.00 frames. , ppl: 9.381647015442553] tot_loss[loss=2.291, over 5480500.52 frames. , ppl: 9.881729512932749], batch size: 70
+2022-12-12 01:23:45,394 INFO [train.py:421] (5/8) Epoch 6, batch 18600, loss[loss=2.163, over 4970.00 frames. , ppl: 8.697756417929913] tot_loss[loss=2.29, over 5483149.94 frames. , ppl: 9.878929834591002], batch size: 70
+2022-12-12 01:25:24,440 INFO [train.py:421] (5/8) Epoch 6, batch 18800, loss[loss=2.393, over 1120.00 frames. , ppl: 10.94206968363565] tot_loss[loss=2.29, over 5488861.57 frames. , ppl: 9.876638100547483], batch size: 70
+2022-12-12 01:27:04,203 INFO [train.py:421] (5/8) Epoch 6, batch 19000, loss[loss=2.909, over 630.00 frames. , ppl: 18.341483440529974] tot_loss[loss=2.291, over 5477940.47 frames. , ppl: 9.88109821513888], batch size: 70
+2022-12-12 01:27:04,204 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 01:27:04,968 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.798580280075496
+2022-12-12 01:28:43,251 INFO [train.py:421] (5/8) Epoch 6, batch 19200, loss[loss=2.313, over 2730.00 frames. , ppl: 10.101672820685547] tot_loss[loss=2.291, over 5477738.59 frames. , ppl: 9.88418837280673], batch size: 70
+2022-12-12 01:30:21,610 INFO [train.py:421] (5/8) Epoch 6, batch 19400, loss[loss=2.382, over 910.00 frames. , ppl: 10.82488018566924] tot_loss[loss=2.291, over 5465553.80 frames. , ppl: 9.888053684329122], batch size: 70
+2022-12-12 01:32:02,250 INFO [train.py:421] (5/8) Epoch 6, batch 19600, loss[loss=2.433, over 2170.00 frames. , ppl: 11.387570532698238] tot_loss[loss=2.291, over 5490636.14 frames. , ppl: 9.884872192455655], batch size: 70
+2022-12-12 01:33:40,022 INFO [train.py:421] (5/8) Epoch 6, batch 19800, loss[loss=2.357, over 2100.00 frames. , ppl: 10.554872605157707] tot_loss[loss=2.29, over 5494597.34 frames. , ppl: 9.877357585133312], batch size: 70
+2022-12-12 01:35:20,136 INFO [train.py:421] (5/8) Epoch 6, batch 20000, loss[loss=2.202, over 5670.00 frames. , ppl: 9.045489584660034] tot_loss[loss=2.289, over 5547398.61 frames. , ppl: 9.862080408613169], batch size: 70
+2022-12-12 01:35:20,137 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 01:35:20,866 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.284, over 211138.00 frames. , ppl: 9.811063387394697
+2022-12-12 01:36:57,458 INFO [train.py:421] (5/8) Epoch 6, batch 20200, loss[loss=2.676, over 1190.00 frames. , ppl: 14.528609352736513] tot_loss[loss=2.288, over 5559503.37 frames. , ppl: 9.858012919679943], batch size: 70
+2022-12-12 01:38:35,707 INFO [train.py:421] (5/8) Epoch 6, batch 20400, loss[loss=2.231, over 5670.00 frames. , ppl: 9.31009981603595] tot_loss[loss=2.289, over 5528633.79 frames. , ppl: 9.867523029906476], batch size: 70
+2022-12-12 01:40:16,777 INFO [train.py:421] (5/8) Epoch 6, batch 20600, loss[loss=2.388, over 1260.00 frames. , ppl: 10.889632419380863] tot_loss[loss=2.289, over 5509307.49 frames. , ppl: 9.86885672415522], batch size: 70
+2022-12-12 01:41:54,647 INFO [train.py:421] (5/8) Epoch 6, batch 20800, loss[loss=2.172, over 7140.00 frames. , ppl: 8.77275560453226] tot_loss[loss=2.289, over 5520942.22 frames. , ppl: 9.869402572263406], batch size: 70
+2022-12-12 01:43:37,329 INFO [train.py:421] (5/8) Epoch 6, batch 21000, loss[loss=2.346, over 1400.00 frames. , ppl: 10.440130548830613] tot_loss[loss=2.29, over 5520633.07 frames. , ppl: 9.873910547397903], batch size: 70
+2022-12-12 01:43:37,330 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 01:43:38,089 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.790676635383708
+2022-12-12 01:45:21,893 INFO [train.py:421] (5/8) Epoch 6, batch 21200, loss[loss=2.741, over 910.00 frames. , ppl: 15.49634320232861] tot_loss[loss=2.289, over 5534156.50 frames. , ppl: 9.864330962099732], batch size: 70
+2022-12-12 01:47:02,204 INFO [train.py:421] (5/8) Epoch 6, batch 21400, loss[loss=2.491, over 910.00 frames. , ppl: 12.07283243742563] tot_loss[loss=2.289, over 5504431.45 frames. , ppl: 9.869309947379927], batch size: 70
+2022-12-12 01:48:39,309 INFO [train.py:421] (5/8) Epoch 6, batch 21600, loss[loss=2.246, over 2940.00 frames. , ppl: 9.44617709124112] tot_loss[loss=2.288, over 5566764.33 frames. , ppl: 9.854668658335385], batch size: 70
+2022-12-12 01:50:16,059 INFO [train.py:421] (5/8) Epoch 6, batch 21800, loss[loss=3.114, over 560.00 frames. , ppl: 22.5060003586589] tot_loss[loss=2.289, over 5558907.54 frames. , ppl: 9.860554549636648], batch size: 70
+2022-12-12 01:51:58,032 INFO [train.py:421] (5/8) Epoch 6, batch 22000, loss[loss=2.275, over 2380.00 frames. , ppl: 9.724873926373899] tot_loss[loss=2.289, over 5545596.50 frames. , ppl: 9.86348674912244], batch size: 70
+2022-12-12 01:51:58,033 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 01:51:58,792 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.776022789235363
+2022-12-12 01:53:38,880 INFO [train.py:421] (5/8) Epoch 6, batch 22200, loss[loss=2.182, over 4620.00 frames. , ppl: 8.867546936616963] tot_loss[loss=2.29, over 5508057.19 frames. , ppl: 9.879269472955727], batch size: 70
+2022-12-12 01:55:20,543 INFO [train.py:421] (5/8) Epoch 6, batch 22400, loss[loss=2.304, over 4270.00 frames. , ppl: 10.011486879498415] tot_loss[loss=2.29, over 5517514.91 frames. , ppl: 9.8778992842996], batch size: 70
+2022-12-12 01:57:04,222 INFO [train.py:421] (5/8) Epoch 6, batch 22600, loss[loss=2.139, over 1680.00 frames. , ppl: 8.492309830665478] tot_loss[loss=2.288, over 5584818.04 frames. , ppl: 9.853956053202076], batch size: 70
+2022-12-12 01:58:41,919 INFO [train.py:421] (5/8) Epoch 6, batch 22800, loss[loss=2.982, over 560.00 frames. , ppl: 19.73546551957395] tot_loss[loss=2.287, over 5601213.77 frames. , ppl: 9.844406166188266], batch size: 70
+2022-12-12 02:00:22,243 INFO [train.py:421] (5/8) Epoch 6, batch 23000, loss[loss=2.386, over 1890.00 frames. , ppl: 10.872379257348735] tot_loss[loss=2.287, over 5596095.20 frames. , ppl: 9.84107008844518], batch size: 70
+2022-12-12 02:00:22,244 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 02:00:23,004 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.793408557185776
+2022-12-12 02:01:58,413 INFO [train.py:421] (5/8) Epoch 6, batch 23200, loss[loss=2.217, over 4410.00 frames. , ppl: 9.180304663851453] tot_loss[loss=2.288, over 5532690.31 frames. , ppl: 9.853555567227662], batch size: 70
+2022-12-12 02:03:43,836 INFO [train.py:421] (5/8) Epoch 6, batch 23400, loss[loss=3.222, over 490.00 frames. , ppl: 25.075754096808804] tot_loss[loss=2.288, over 5523886.25 frames. , ppl: 9.858095151920281], batch size: 70
+2022-12-12 02:05:24,908 INFO [train.py:421] (5/8) Epoch 6, batch 23600, loss[loss=2.22, over 1540.00 frames. , ppl: 9.208675607119567] tot_loss[loss=2.287, over 5534085.94 frames. , ppl: 9.84883375784109], batch size: 70
+2022-12-12 02:07:03,693 INFO [train.py:421] (5/8) Epoch 6, batch 23800, loss[loss=2.392, over 1750.00 frames. , ppl: 10.933609853241093] tot_loss[loss=2.287, over 5548237.80 frames. , ppl: 9.846997934745758], batch size: 70
+2022-12-12 02:08:44,621 INFO [train.py:421] (5/8) Epoch 6, batch 24000, loss[loss=2.274, over 2450.00 frames. , ppl: 9.716330201098556] tot_loss[loss=2.289, over 5502584.39 frames. , ppl: 9.860707734968237], batch size: 70
+2022-12-12 02:08:44,621 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 02:08:45,409 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.790136138531322
+2022-12-12 02:10:29,023 INFO [train.py:421] (5/8) Epoch 6, batch 24200, loss[loss=2.122, over 7490.00 frames. , ppl: 8.349903476863224] tot_loss[loss=2.288, over 5539038.06 frames. , ppl: 9.856475211837836], batch size: 70
+2022-12-12 02:12:08,531 INFO [train.py:421] (5/8) Epoch 6, batch 24400, loss[loss=2.444, over 1330.00 frames. , ppl: 11.520092844997984] tot_loss[loss=2.288, over 5545281.09 frames. , ppl: 9.856469600426042], batch size: 70
+2022-12-12 02:13:52,337 INFO [train.py:421] (5/8) Epoch 6, batch 24600, loss[loss=2.39, over 2310.00 frames. , ppl: 10.916894327957] tot_loss[loss=2.288, over 5578905.91 frames. , ppl: 9.858207078989714], batch size: 70
+2022-12-12 02:15:29,118 INFO [train.py:421] (5/8) Epoch 6, batch 24800, loss[loss=2.217, over 4340.00 frames. , ppl: 9.175284089135918] tot_loss[loss=2.289, over 5536212.87 frames. , ppl: 9.864337935045464], batch size: 70
+2022-12-12 02:17:09,491 INFO [train.py:421] (5/8) Epoch 6, batch 25000, loss[loss=2.369, over 1400.00 frames. , ppl: 10.682639080518367] tot_loss[loss=2.289, over 5512253.27 frames. , ppl: 9.864783317315196], batch size: 70
+2022-12-12 02:17:09,491 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 02:17:10,254 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.788068615004098
+2022-12-12 02:18:49,737 INFO [train.py:421] (5/8) Epoch 6, batch 25200, loss[loss=2.198, over 5530.00 frames. , ppl: 9.002559906552307] tot_loss[loss=2.291, over 5453854.53 frames. , ppl: 9.883234225912695], batch size: 70
+2022-12-12 02:20:33,642 INFO [train.py:421] (5/8) Epoch 6, batch 25400, loss[loss=2.348, over 1330.00 frames. , ppl: 10.465678720776644] tot_loss[loss=2.29, over 5487140.54 frames. , ppl: 9.875693907530605], batch size: 70
+2022-12-12 02:22:12,266 INFO [train.py:421] (5/8) Epoch 6, batch 25600, loss[loss=2.522, over 980.00 frames. , ppl: 12.451456842627017] tot_loss[loss=2.29, over 5486656.43 frames. , ppl: 9.87783182056771], batch size: 70
+2022-12-12 02:23:50,898 INFO [train.py:421] (5/8) Epoch 6, batch 25800, loss[loss=2.199, over 4130.00 frames. , ppl: 9.013838175645157] tot_loss[loss=2.29, over 5505380.73 frames. , ppl: 9.87136506939551], batch size: 70
+2022-12-12 02:25:30,634 INFO [train.py:421] (5/8) Epoch 6, batch 26000, loss[loss=2.509, over 1680.00 frames. , ppl: 12.28767412693828] tot_loss[loss=2.291, over 5467208.69 frames. , ppl: 9.881832471966394], batch size: 70
+2022-12-12 02:25:30,634 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 02:25:31,395 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.806832849432856
+2022-12-12 02:27:11,365 INFO [train.py:421] (5/8) Epoch 6, batch 26200, loss[loss=3.278, over 490.00 frames. , ppl: 26.53345817046044] tot_loss[loss=2.291, over 5469353.67 frames. , ppl: 9.883386126526695], batch size: 70
+2022-12-12 02:28:48,580 INFO [train.py:421] (5/8) Epoch 6, batch 26400, loss[loss=2.553, over 1330.00 frames. , ppl: 12.84720508987554] tot_loss[loss=2.291, over 5447474.06 frames. , ppl: 9.884366614708757], batch size: 70
+2022-12-12 02:30:30,700 INFO [train.py:421] (5/8) Epoch 6, batch 26600, loss[loss=2.572, over 840.00 frames. , ppl: 13.085518664192373] tot_loss[loss=2.291, over 5436977.21 frames. , ppl: 9.889084101424734], batch size: 70
+2022-12-12 02:32:10,389 INFO [train.py:421] (5/8) Epoch 6, batch 26800, loss[loss=2.584, over 770.00 frames. , ppl: 13.249102836561127] tot_loss[loss=2.29, over 5466775.17 frames. , ppl: 9.878541827678983], batch size: 70
+2022-12-12 02:33:50,916 INFO [train.py:421] (5/8) Epoch 6, batch 27000, loss[loss=2.209, over 12950.00 frames. , ppl: 9.109281680395963] tot_loss[loss=2.288, over 5540367.20 frames. , ppl: 9.85217291967327], batch size: 70
+2022-12-12 02:33:50,916 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 02:33:51,645 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.810623407744394
+2022-12-12 02:35:28,251 INFO [train.py:421] (5/8) Epoch 6, batch 27200, loss[loss=2.211, over 8890.00 frames. , ppl: 9.128998450747865] tot_loss[loss=2.288, over 5536644.85 frames. , ppl: 9.856506101054729], batch size: 70
+2022-12-12 02:37:07,166 INFO [train.py:421] (5/8) Epoch 6, batch 27400, loss[loss=2.159, over 6860.00 frames. , ppl: 8.664718251016149] tot_loss[loss=2.288, over 5532788.35 frames. , ppl: 9.856321756059414], batch size: 70
+2022-12-12 02:38:49,954 INFO [train.py:421] (5/8) Epoch 6, batch 27600, loss[loss=2.64, over 700.00 frames. , ppl: 14.014968076547204] tot_loss[loss=2.287, over 5560726.94 frames. , ppl: 9.844597155724113], batch size: 70
+2022-12-12 02:40:23,826 INFO [train.py:421] (5/8) Epoch 6, batch 27800, loss[loss=2.28, over 1820.00 frames. , ppl: 9.775427244148565] tot_loss[loss=2.288, over 5533333.40 frames. , ppl: 9.856245625449024], batch size: 70
+2022-12-12 02:42:02,134 INFO [train.py:421] (5/8) Epoch 6, batch 28000, loss[loss=2.558, over 1050.00 frames. , ppl: 12.909946320543952] tot_loss[loss=2.288, over 5546253.56 frames. , ppl: 9.853978588154026], batch size: 70
+2022-12-12 02:42:02,135 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 02:42:02,898 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.284, over 211138.00 frames. , ppl: 9.819234910695574
+2022-12-12 02:43:40,809 INFO [train.py:421] (5/8) Epoch 6, batch 28200, loss[loss=2.964, over 700.00 frames. , ppl: 19.380030186309185] tot_loss[loss=2.289, over 5513183.83 frames. , ppl: 9.861096298653226], batch size: 70
+2022-12-12 02:45:19,070 INFO [train.py:421] (5/8) Epoch 6, batch 28400, loss[loss=2.341, over 1820.00 frames. , ppl: 10.393884919035926] tot_loss[loss=2.289, over 5515802.46 frames. , ppl: 9.862285158763589], batch size: 70
+2022-12-12 02:46:59,636 INFO [train.py:421] (5/8) Epoch 6, batch 28600, loss[loss=2.371, over 2590.00 frames. , ppl: 10.712051053597715] tot_loss[loss=2.289, over 5523007.04 frames. , ppl: 9.863637534552197], batch size: 70
+2022-12-12 02:48:38,764 INFO [train.py:421] (5/8) Epoch 6, batch 28800, loss[loss=2.404, over 1680.00 frames. , ppl: 11.06540646601362] tot_loss[loss=2.289, over 5503870.04 frames. , ppl: 9.866066344389237], batch size: 70
+2022-12-12 02:50:20,987 INFO [train.py:421] (5/8) Epoch 6, batch 29000, loss[loss=2.977, over 560.00 frames. , ppl: 19.636801614994233] tot_loss[loss=2.29, over 5488394.49 frames. , ppl: 9.876036853215702], batch size: 70
+2022-12-12 02:50:20,988 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 02:50:21,747 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.796046998105865
+2022-12-12 02:51:59,584 INFO [train.py:421] (5/8) Epoch 6, batch 29200, loss[loss=2.585, over 910.00 frames. , ppl: 13.25871525105219] tot_loss[loss=2.291, over 5452176.84 frames. , ppl: 9.881845390095181], batch size: 70
+2022-12-12 02:53:38,991 INFO [train.py:421] (5/8) Epoch 6, batch 29400, loss[loss=2.247, over 2870.00 frames. , ppl: 9.46096222996507] tot_loss[loss=2.291, over 5419743.06 frames. , ppl: 9.883712060239892], batch size: 70
+2022-12-12 02:55:17,412 INFO [train.py:421] (5/8) Epoch 6, batch 29600, loss[loss=2.776, over 630.00 frames. , ppl: 16.05015753422784] tot_loss[loss=2.291, over 5438353.76 frames. , ppl: 9.881673600646492], batch size: 70
+2022-12-12 02:57:01,348 INFO [train.py:421] (5/8) Epoch 6, batch 29800, loss[loss=2.379, over 1120.00 frames. , ppl: 10.799238483042087] tot_loss[loss=2.291, over 5436475.98 frames. , ppl: 9.886543019510398], batch size: 70
+2022-12-12 02:58:45,207 INFO [train.py:421] (5/8) Epoch 6, batch 30000, loss[loss=2.24, over 3920.00 frames. , ppl: 9.393583553144971] tot_loss[loss=2.292, over 5471739.33 frames. , ppl: 9.891227148496538], batch size: 70
+2022-12-12 02:58:45,207 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 02:58:45,956 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.793146201640242
+2022-12-12 03:00:24,701 INFO [train.py:421] (5/8) Epoch 6, batch 30200, loss[loss=2.413, over 1260.00 frames. , ppl: 11.163725594362646] tot_loss[loss=2.291, over 5489224.81 frames. , ppl: 9.884461187877568], batch size: 70
+2022-12-12 03:02:01,496 INFO [train.py:421] (5/8) Epoch 6, batch 30400, loss[loss=2.475, over 1260.00 frames. , ppl: 11.876951661751846] tot_loss[loss=2.293, over 5428960.78 frames. , ppl: 9.903199356248363], batch size: 70
+2022-12-12 03:03:40,077 INFO [train.py:421] (5/8) Epoch 6, batch 30600, loss[loss=2.291, over 2940.00 frames. , ppl: 9.889077152861912] tot_loss[loss=2.293, over 5422139.27 frames. , ppl: 9.907158964538743], batch size: 70
+2022-12-12 03:05:19,244 INFO [train.py:421] (5/8) Epoch 6, batch 30800, loss[loss=2.203, over 3780.00 frames. , ppl: 9.055426868575898] tot_loss[loss=2.293, over 5436533.76 frames. , ppl: 9.906321054132144], batch size: 70
+2022-12-12 03:07:01,537 INFO [train.py:421] (5/8) Epoch 6, batch 31000, loss[loss=2.211, over 3570.00 frames. , ppl: 9.124273872088922] tot_loss[loss=2.293, over 5438335.84 frames. , ppl: 9.904624615660362], batch size: 70
+2022-12-12 03:07:01,537 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 03:07:02,295 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.792647600527868
+2022-12-12 03:08:42,301 INFO [train.py:421] (5/8) Epoch 6, batch 31200, loss[loss=2.239, over 2170.00 frames. , ppl: 9.38621864142445] tot_loss[loss=2.293, over 5445385.14 frames. , ppl: 9.905707222674723], batch size: 70
+2022-12-12 03:10:23,242 INFO [train.py:421] (5/8) Epoch 6, batch 31400, loss[loss=2.55, over 910.00 frames. , ppl: 12.801037293741258] tot_loss[loss=2.292, over 5470798.65 frames. , ppl: 9.89573189359468], batch size: 70
+2022-12-12 03:12:05,123 INFO [train.py:421] (5/8) Epoch 6, batch 31600, loss[loss=2.521, over 1190.00 frames. , ppl: 12.440513247221794] tot_loss[loss=2.291, over 5497747.94 frames. , ppl: 9.884865819190612], batch size: 70
+2022-12-12 03:13:47,714 INFO [train.py:421] (5/8) Epoch 6, batch 31800, loss[loss=2.269, over 2730.00 frames. , ppl: 9.669678240433463] tot_loss[loss=2.291, over 5488997.73 frames. , ppl: 9.884439984425043], batch size: 70
+2022-12-12 03:15:28,300 INFO [train.py:421] (5/8) Epoch 6, batch 32000, loss[loss=2.296, over 2730.00 frames. , ppl: 9.93301894545387] tot_loss[loss=2.293, over 5440402.34 frames. , ppl: 9.903834278150013], batch size: 70
+2022-12-12 03:15:28,301 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 03:15:29,061 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.786621463485243
+2022-12-12 03:17:08,925 INFO [train.py:421] (5/8) Epoch 6, batch 32200, loss[loss=2.23, over 12600.00 frames. , ppl: 9.296582759324961] tot_loss[loss=2.291, over 5504692.91 frames. , ppl: 9.882532449296866], batch size: 70
+2022-12-12 03:18:47,922 INFO [train.py:421] (5/8) Epoch 6, batch 32400, loss[loss=2.215, over 2520.00 frames. , ppl: 9.157347789431778] tot_loss[loss=2.291, over 5488338.25 frames. , ppl: 9.889356191720138], batch size: 70
+2022-12-12 03:20:28,895 INFO [train.py:421] (5/8) Epoch 6, batch 32600, loss[loss=2.313, over 2660.00 frames. , ppl: 10.102605247466505] tot_loss[loss=2.291, over 5501302.84 frames. , ppl: 9.885798808430925], batch size: 70
+2022-12-12 03:22:12,184 INFO [train.py:421] (5/8) Epoch 6, batch 32800, loss[loss=2.383, over 1820.00 frames. , ppl: 10.84152804929741] tot_loss[loss=2.29, over 5537562.83 frames. , ppl: 9.871888566584223], batch size: 70
+2022-12-12 03:23:51,246 INFO [train.py:421] (5/8) Epoch 6, batch 33000, loss[loss=2.247, over 2590.00 frames. , ppl: 9.45747678245071] tot_loss[loss=2.29, over 5510983.59 frames. , ppl: 9.875097208728135], batch size: 70
+2022-12-12 03:23:51,247 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 03:23:52,003 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.283, over 211138.00 frames. , ppl: 9.805240700234473
+2022-12-12 03:25:32,458 INFO [train.py:421] (5/8) Epoch 6, batch 33200, loss[loss=2.528, over 1050.00 frames. , ppl: 12.532991536442836] tot_loss[loss=2.29, over 5533281.65 frames. , ppl: 9.874177871008108], batch size: 70
+2022-12-12 03:27:11,802 INFO [train.py:421] (5/8) Epoch 6, batch 33400, loss[loss=2.358, over 3360.00 frames. , ppl: 10.574715979228296] tot_loss[loss=2.29, over 5539301.67 frames. , ppl: 9.870471939839875], batch size: 70
+2022-12-12 03:28:50,124 INFO [train.py:421] (5/8) Epoch 6, batch 33600, loss[loss=2.849, over 630.00 frames. , ppl: 17.27140945705736] tot_loss[loss=2.29, over 5510652.44 frames. , ppl: 9.87746101874624], batch size: 70
+2022-12-12 03:30:30,181 INFO [train.py:421] (5/8) Epoch 6, batch 33800, loss[loss=2.238, over 7420.00 frames. , ppl: 9.3733925400531] tot_loss[loss=2.29, over 5540160.41 frames. , ppl: 9.875872224768036], batch size: 70
+2022-12-12 03:32:08,720 INFO [train.py:421] (5/8) Epoch 6, batch 34000, loss[loss=2.417, over 1470.00 frames. , ppl: 11.213898555506763] tot_loss[loss=2.29, over 5550412.80 frames. , ppl: 9.877085159514541], batch size: 70
+2022-12-12 03:32:08,721 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 03:32:09,480 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.78966811844687
+2022-12-12 03:33:48,818 INFO [train.py:421] (5/8) Epoch 6, batch 34200, loss[loss=2.463, over 1260.00 frames. , ppl: 11.73920876126155] tot_loss[loss=2.29, over 5532973.23 frames. , ppl: 9.877768089682563], batch size: 70
+2022-12-12 03:35:27,527 INFO [train.py:421] (5/8) Epoch 6, batch 34400, loss[loss=2.321, over 1400.00 frames. , ppl: 10.18131521977462] tot_loss[loss=2.291, over 5482841.19 frames. , ppl: 9.8873954593648], batch size: 70
+2022-12-12 03:37:07,526 INFO [train.py:421] (5/8) Epoch 6, batch 34600, loss[loss=2.286, over 2450.00 frames. , ppl: 9.83801914666712] tot_loss[loss=2.291, over 5472033.15 frames. , ppl: 9.888888558012082], batch size: 70
+2022-12-12 03:38:50,962 INFO [train.py:421] (5/8) Epoch 6, batch 34800, loss[loss=2.204, over 4340.00 frames. , ppl: 9.059879744874998] tot_loss[loss=2.289, over 5570039.35 frames. , ppl: 9.86106697914099], batch size: 70
+2022-12-12 03:40:30,319 INFO [train.py:421] (5/8) Epoch 6, batch 35000, loss[loss=2.441, over 1050.00 frames. , ppl: 11.489698977392804] tot_loss[loss=2.289, over 5550712.65 frames. , ppl: 9.865852363722574], batch size: 70
+2022-12-12 03:40:30,320 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 03:40:31,078 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.778649309689856
+2022-12-12 03:42:08,421 INFO [train.py:421] (5/8) Epoch 6, batch 35200, loss[loss=2.44, over 1470.00 frames. , ppl: 11.47072240576583] tot_loss[loss=2.29, over 5529446.12 frames. , ppl: 9.871174881739131], batch size: 70
+2022-12-12 03:43:48,388 INFO [train.py:421] (5/8) Epoch 6, batch 35400, loss[loss=2.343, over 2520.00 frames. , ppl: 10.408706796611492] tot_loss[loss=2.289, over 5516584.25 frames. , ppl: 9.864749906300407], batch size: 70
+2022-12-12 03:45:30,989 INFO [train.py:421] (5/8) Epoch 6, batch 35600, loss[loss=2.386, over 2240.00 frames. , ppl: 10.8691056521595] tot_loss[loss=2.289, over 5532777.13 frames. , ppl: 9.861790241569008], batch size: 70
+2022-12-12 03:47:10,358 INFO [train.py:421] (5/8) Epoch 6, batch 35800, loss[loss=2.475, over 1260.00 frames. , ppl: 11.877264643582489] tot_loss[loss=2.289, over 5516879.10 frames. , ppl: 9.867905251834314], batch size: 70
+2022-12-12 03:48:50,030 INFO [train.py:421] (5/8) Epoch 6, batch 36000, loss[loss=2.618, over 840.00 frames. , ppl: 13.709707929986395] tot_loss[loss=2.29, over 5512759.47 frames. , ppl: 9.870434901477623], batch size: 70
+2022-12-12 03:48:50,031 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 03:48:50,790 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.788203345702467
+2022-12-12 03:50:31,023 INFO [train.py:421] (5/8) Epoch 6, batch 36200, loss[loss=2.138, over 4200.00 frames. , ppl: 8.483776494185646] tot_loss[loss=2.29, over 5477337.71 frames. , ppl: 9.879444535970958], batch size: 70
+2022-12-12 03:52:13,165 INFO [train.py:421] (5/8) Epoch 6, batch 36400, loss[loss=2.217, over 8960.00 frames. , ppl: 9.178874177124857] tot_loss[loss=2.29, over 5494984.19 frames. , ppl: 9.877617497381847], batch size: 70
+2022-12-12 03:53:51,605 INFO [train.py:421] (5/8) Epoch 6, batch 36600, loss[loss=2.168, over 8050.00 frames. , ppl: 8.740791763881932] tot_loss[loss=2.291, over 5480454.25 frames. , ppl: 9.885481323692316], batch size: 70
+2022-12-12 03:55:34,328 INFO [train.py:421] (5/8) Epoch 6, batch 36800, loss[loss=2.284, over 1890.00 frames. , ppl: 9.819536941297427] tot_loss[loss=2.29, over 5497562.59 frames. , ppl: 9.877443033575647], batch size: 70
+2022-12-12 03:57:15,030 INFO [train.py:421] (5/8) Epoch 6, batch 37000, loss[loss=2.2, over 9450.00 frames. , ppl: 9.028214902750651] tot_loss[loss=2.288, over 5580118.52 frames. , ppl: 9.85400050326647], batch size: 70
+2022-12-12 03:57:15,031 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 03:57:15,781 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.776730360844793
+2022-12-12 03:58:58,765 INFO [train.py:421] (5/8) Epoch 6, batch 37200, loss[loss=2.216, over 11200.00 frames. , ppl: 9.174880577596785] tot_loss[loss=2.288, over 5563277.13 frames. , ppl: 9.853989100883254], batch size: 70
+2022-12-12 04:00:39,203 INFO [train.py:421] (5/8) Epoch 6, batch 37400, loss[loss=2.219, over 3080.00 frames. , ppl: 9.199324001867632] tot_loss[loss=2.289, over 5505250.72 frames. , ppl: 9.865867155856082], batch size: 70
+2022-12-12 04:02:19,601 INFO [train.py:421] (5/8) Epoch 6, batch 37600, loss[loss=2.28, over 4480.00 frames. , ppl: 9.781230188453824] tot_loss[loss=2.289, over 5505140.86 frames. , ppl: 9.866022216341374], batch size: 70
+2022-12-12 04:04:01,661 INFO [train.py:421] (5/8) Epoch 6, batch 37800, loss[loss=2.603, over 630.00 frames. , ppl: 13.50929617061648] tot_loss[loss=2.289, over 5512703.88 frames. , ppl: 9.86698391266652], batch size: 70
+2022-12-12 04:05:40,692 INFO [train.py:421] (5/8) Epoch 6, batch 38000, loss[loss=2.34, over 1540.00 frames. , ppl: 10.385614238941445] tot_loss[loss=2.288, over 5589378.13 frames. , ppl: 9.851484381720855], batch size: 70
+2022-12-12 04:05:40,693 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 04:05:41,465 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.785526464019718
+2022-12-12 04:07:23,630 INFO [train.py:421] (5/8) Epoch 6, batch 38200, loss[loss=2.224, over 2170.00 frames. , ppl: 9.24685555884513] tot_loss[loss=2.287, over 5614585.01 frames. , ppl: 9.845711313285047], batch size: 70
+2022-12-12 04:08:58,703 INFO [train.py:421] (5/8) Epoch 6, batch 38400, loss[loss=2.296, over 2590.00 frames. , ppl: 9.934055043014649] tot_loss[loss=2.288, over 5590534.43 frames. , ppl: 9.859698351601573], batch size: 70
+2022-12-12 04:10:38,732 INFO [train.py:421] (5/8) Epoch 6, batch 38600, loss[loss=2.297, over 4060.00 frames. , ppl: 9.942770011331136] tot_loss[loss=2.29, over 5540315.83 frames. , ppl: 9.876341025086045], batch size: 70
+2022-12-12 04:12:18,446 INFO [train.py:421] (5/8) Epoch 6, batch 38800, loss[loss=2.753, over 700.00 frames. , ppl: 15.683865887505863] tot_loss[loss=2.292, over 5492858.38 frames. , ppl: 9.88985092682389], batch size: 70
+2022-12-12 04:13:53,332 INFO [train.py:421] (5/8) Epoch 6, batch 39000, loss[loss=2.406, over 1400.00 frames. , ppl: 11.09011299207289] tot_loss[loss=2.291, over 5482864.43 frames. , ppl: 9.886965748857191], batch size: 70
+2022-12-12 04:13:53,332 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 04:13:54,080 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.790575199383582
+2022-12-12 04:15:37,674 INFO [train.py:421] (5/8) Epoch 6, batch 39200, loss[loss=2.199, over 3150.00 frames. , ppl: 9.016187540552068] tot_loss[loss=2.29, over 5518377.88 frames. , ppl: 9.871813526930877], batch size: 70
+2022-12-12 04:17:15,043 INFO [train.py:421] (5/8) Epoch 6, batch 39400, loss[loss=2.206, over 4410.00 frames. , ppl: 9.080582956927772] tot_loss[loss=2.289, over 5510628.90 frames. , ppl: 9.866126134243686], batch size: 70
+2022-12-12 04:18:56,787 INFO [train.py:421] (5/8) Epoch 6, batch 39600, loss[loss=2.177, over 3430.00 frames. , ppl: 8.819511707005578] tot_loss[loss=2.29, over 5499787.60 frames. , ppl: 9.870866901110253], batch size: 70
+2022-12-12 04:20:36,868 INFO [train.py:421] (5/8) Epoch 6, batch 39800, loss[loss=2.169, over 2940.00 frames. , ppl: 8.748798828662617] tot_loss[loss=2.288, over 5552857.46 frames. , ppl: 9.856443765697868], batch size: 70
+2022-12-12 04:22:14,683 INFO [train.py:421] (5/8) Epoch 6, batch 40000, loss[loss=2.183, over 5950.00 frames. , ppl: 8.870225655874783] tot_loss[loss=2.289, over 5509663.35 frames. , ppl: 9.864872082545775], batch size: 70
+2022-12-12 04:22:14,684 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 04:22:15,413 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.796226785877156
+2022-12-12 04:23:58,048 INFO [train.py:421] (5/8) Epoch 6, batch 40200, loss[loss=2.37, over 1260.00 frames. , ppl: 10.701579648763982] tot_loss[loss=2.289, over 5501457.38 frames. , ppl: 9.870001146472596], batch size: 70
+2022-12-12 04:25:35,491 INFO [train.py:421] (5/8) Epoch 6, batch 40400, loss[loss=2.366, over 1540.00 frames. , ppl: 10.651115550740958] tot_loss[loss=2.291, over 5463420.77 frames. , ppl: 9.88009720718998], batch size: 70
+2022-12-12 04:27:15,629 INFO [train.py:421] (5/8) Epoch 6, batch 40600, loss[loss=2.175, over 8119.00 frames. , ppl: 8.798164111629681] tot_loss[loss=2.292, over 5429225.60 frames. , ppl: 9.891503245608009], batch size: 70
+2022-12-12 04:28:54,809 INFO [train.py:421] (5/8) Epoch 6, batch 40800, loss[loss=2.588, over 980.00 frames. , ppl: 13.29946371895781] tot_loss[loss=2.289, over 5482246.05 frames. , ppl: 9.86709077171095], batch size: 70
+2022-12-12 04:30:38,979 INFO [train.py:421] (5/8) Epoch 6, batch 41000, loss[loss=2.443, over 1470.00 frames. , ppl: 11.51133043312547] tot_loss[loss=2.289, over 5472260.37 frames. , ppl: 9.869641267090596], batch size: 70
+2022-12-12 04:30:38,979 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 04:30:39,735 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.77442840863681
+2022-12-12 04:32:21,608 INFO [train.py:421] (5/8) Epoch 6, batch 41200, loss[loss=2.21, over 8330.00 frames. , ppl: 9.119779119900512] tot_loss[loss=2.29, over 5484036.07 frames. , ppl: 9.87236736052575], batch size: 70
+2022-12-12 04:34:01,967 INFO [train.py:421] (5/8) Epoch 6, batch 41400, loss[loss=2.318, over 1820.00 frames. , ppl: 10.154756228230521] tot_loss[loss=2.29, over 5517518.54 frames. , ppl: 9.871614964508776], batch size: 70
+2022-12-12 04:35:43,802 INFO [train.py:421] (5/8) Epoch 6, batch 41600, loss[loss=2.498, over 1400.00 frames. , ppl: 12.155435087462504] tot_loss[loss=2.289, over 5519568.99 frames. , ppl: 9.866092179233984], batch size: 70
+2022-12-12 04:37:22,140 INFO [train.py:421] (5/8) Epoch 6, batch 41800, loss[loss=2.25, over 5460.00 frames. , ppl: 9.48300251671614] tot_loss[loss=2.29, over 5483955.85 frames. , ppl: 9.873941633621728], batch size: 70
+2022-12-12 04:39:01,064 INFO [train.py:421] (5/8) Epoch 6, batch 42000, loss[loss=2.384, over 2940.00 frames. , ppl: 10.843204937529135] tot_loss[loss=2.29, over 5504228.42 frames. , ppl: 9.876164645764572], batch size: 70
+2022-12-12 04:39:01,065 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 04:39:01,825 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.774656988070976
+2022-12-12 04:40:40,482 INFO [train.py:421] (5/8) Epoch 6, batch 42200, loss[loss=2.171, over 3150.00 frames. , ppl: 8.770640310936502] tot_loss[loss=2.288, over 5567669.75 frames. , ppl: 9.855339192346777], batch size: 70
+2022-12-12 04:42:19,271 INFO [train.py:421] (5/8) Epoch 6, batch 42400, loss[loss=2.153, over 7000.00 frames. , ppl: 8.606381011833879] tot_loss[loss=2.288, over 5578781.33 frames. , ppl: 9.850375646861538], batch size: 70
+2022-12-12 04:43:58,779 INFO [train.py:421] (5/8) Epoch 6, batch 42600, loss[loss=2.291, over 3710.00 frames. , ppl: 9.885778651747104] tot_loss[loss=2.286, over 5623747.77 frames. , ppl: 9.839205364797406], batch size: 70
+2022-12-12 04:45:37,700 INFO [train.py:421] (5/8) Epoch 6, batch 42800, loss[loss=2.307, over 2590.00 frames. , ppl: 10.041175270871555] tot_loss[loss=2.286, over 5602272.89 frames. , ppl: 9.83740673760893], batch size: 70
+2022-12-12 04:47:16,068 INFO [train.py:421] (5/8) Epoch 6, batch 43000, loss[loss=2.527, over 910.00 frames. , ppl: 12.517854053698109] tot_loss[loss=2.286, over 5598943.82 frames. , ppl: 9.832843445392028], batch size: 70
+2022-12-12 04:47:16,068 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 04:47:16,826 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.774992633382912
+2022-12-12 04:48:53,986 INFO [train.py:421] (5/8) Epoch 6, batch 43200, loss[loss=2.508, over 1260.00 frames. , ppl: 12.285605308367815] tot_loss[loss=2.286, over 5571068.66 frames. , ppl: 9.838496500244403], batch size: 70
+2022-12-12 04:50:34,705 INFO [train.py:421] (5/8) Epoch 6, batch 43400, loss[loss=2.356, over 3010.00 frames. , ppl: 10.545219825602432] tot_loss[loss=2.286, over 5577073.28 frames. , ppl: 9.835562025853196], batch size: 70
+2022-12-12 04:52:11,350 INFO [train.py:421] (5/8) Epoch 6, batch 43600, loss[loss=2.234, over 7490.00 frames. , ppl: 9.335845016603942] tot_loss[loss=2.287, over 5537990.21 frames. , ppl: 9.84809755865821], batch size: 70
+2022-12-12 04:53:48,621 INFO [train.py:421] (5/8) Epoch 6, batch 43800, loss[loss=2.462, over 1400.00 frames. , ppl: 11.733465620647346] tot_loss[loss=2.288, over 5533063.22 frames. , ppl: 9.852315766299608], batch size: 70
+2022-12-12 04:55:27,842 INFO [train.py:421] (5/8) Epoch 6, batch 44000, loss[loss=2.189, over 5180.00 frames. , ppl: 8.925929854899392] tot_loss[loss=2.286, over 5586626.06 frames. , ppl: 9.838698842854436], batch size: 70
+2022-12-12 04:55:27,843 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 04:55:28,598 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.775564124978258
+2022-12-12 04:57:06,997 INFO [train.py:421] (5/8) Epoch 6, batch 44200, loss[loss=2.114, over 4340.00 frames. , ppl: 8.284317289723605] tot_loss[loss=2.287, over 5556705.02 frames. , ppl: 9.850237777664102], batch size: 70
+2022-12-12 04:58:47,001 INFO [train.py:421] (5/8) Epoch 6, batch 44400, loss[loss=2.303, over 3430.00 frames. , ppl: 10.008359297155527] tot_loss[loss=2.288, over 5543047.94 frames. , ppl: 9.850613697996131], batch size: 70
+2022-12-12 05:00:25,090 INFO [train.py:421] (5/8) Epoch 6, batch 44600, loss[loss=2.48, over 1260.00 frames. , ppl: 11.93862747620741] tot_loss[loss=2.288, over 5513424.15 frames. , ppl: 9.855373772384059], batch size: 70
+2022-12-12 05:02:08,397 INFO [train.py:421] (5/8) Epoch 6, batch 44800, loss[loss=2.293, over 6790.00 frames. , ppl: 9.906452922558145] tot_loss[loss=2.287, over 5557013.53 frames. , ppl: 9.841870699583673], batch size: 70
+2022-12-12 05:03:45,269 INFO [train.py:421] (5/8) Epoch 6, batch 45000, loss[loss=2.452, over 1190.00 frames. , ppl: 11.61247933182694] tot_loss[loss=2.288, over 5497942.04 frames. , ppl: 9.856869758974339], batch size: 70
+2022-12-12 05:03:45,269 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 05:03:46,014 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.282, over 211138.00 frames. , ppl: 9.792734564010422
+2022-12-12 05:05:26,496 INFO [train.py:421] (5/8) Epoch 6, batch 45200, loss[loss=2.204, over 9660.00 frames. , ppl: 9.059491945806771] tot_loss[loss=2.289, over 5492348.88 frames. , ppl: 9.861381033059562], batch size: 70
+2022-12-12 05:07:06,357 INFO [train.py:421] (5/8) Epoch 6, batch 45400, loss[loss=2.344, over 1820.00 frames. , ppl: 10.424042005499711] tot_loss[loss=2.289, over 5483248.16 frames. , ppl: 9.863937284975627], batch size: 70
+2022-12-12 05:08:44,594 INFO [train.py:421] (5/8) Epoch 6, batch 45600, loss[loss=2.246, over 2520.00 frames. , ppl: 9.4476927127693] tot_loss[loss=2.289, over 5479171.01 frames. , ppl: 9.862212643053626], batch size: 70
+2022-12-12 05:10:22,563 INFO [train.py:421] (5/8) Epoch 6, batch 45800, loss[loss=3.516, over 420.00 frames. , ppl: 33.64482511718567] tot_loss[loss=2.289, over 5453631.01 frames. , ppl: 9.868348269199995], batch size: 70
+2022-12-12 05:12:00,167 INFO [train.py:421] (5/8) Epoch 6, batch 46000, loss[loss=2.393, over 1750.00 frames. , ppl: 10.950615309382439] tot_loss[loss=2.289, over 5468565.98 frames. , ppl: 9.865664490127438], batch size: 70
+2022-12-12 05:12:00,167 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 05:12:00,928 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.7772122333999
+2022-12-12 05:13:38,802 INFO [train.py:421] (5/8) Epoch 6, batch 46200, loss[loss=2.173, over 3150.00 frames. , ppl: 8.787810364690857] tot_loss[loss=2.29, over 5447011.43 frames. , ppl: 9.870540105740412], batch size: 70
+2022-12-12 05:15:18,381 INFO [train.py:421] (5/8) Epoch 6, batch 46400, loss[loss=2.209, over 4970.00 frames. , ppl: 9.105901107305568] tot_loss[loss=2.289, over 5454731.09 frames. , ppl: 9.869695038279382], batch size: 70
+2022-12-12 05:16:59,461 INFO [train.py:421] (5/8) Epoch 6, batch 46600, loss[loss=2.327, over 2380.00 frames. , ppl: 10.251633901303709] tot_loss[loss=2.288, over 5512868.71 frames. , ppl: 9.859828219423791], batch size: 70
+2022-12-12 05:18:44,137 INFO [train.py:421] (5/8) Epoch 6, batch 46800, loss[loss=2.319, over 2240.00 frames. , ppl: 10.163029120343735] tot_loss[loss=2.29, over 5485916.97 frames. , ppl: 9.870147846558519], batch size: 70
+2022-12-12 05:20:24,469 INFO [train.py:421] (5/8) Epoch 6, batch 47000, loss[loss=2.245, over 4760.00 frames. , ppl: 9.439728405737903] tot_loss[loss=2.288, over 5522084.56 frames. , ppl: 9.857509784024892], batch size: 70
+2022-12-12 05:20:24,470 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 05:20:25,228 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.782431870183903
+2022-12-12 05:22:04,499 INFO [train.py:421] (5/8) Epoch 6, batch 47200, loss[loss=2.27, over 2800.00 frames. , ppl: 9.679373806853784] tot_loss[loss=2.289, over 5511496.65 frames. , ppl: 9.861048427107088], batch size: 70
+2022-12-12 05:23:42,961 INFO [train.py:421] (5/8) Epoch 6, batch 47400, loss[loss=2.684, over 770.00 frames. , ppl: 14.647175812418212] tot_loss[loss=2.289, over 5507012.94 frames. , ppl: 9.860623971642637], batch size: 70
+2022-12-12 05:25:19,137 INFO [train.py:421] (5/8) Epoch 6, batch 47600, loss[loss=2.312, over 2800.00 frames. , ppl: 10.097534243812753] tot_loss[loss=2.29, over 5480501.83 frames. , ppl: 9.870028410690018], batch size: 70
+2022-12-12 05:26:59,790 INFO [train.py:421] (5/8) Epoch 6, batch 47800, loss[loss=2.226, over 3500.00 frames. , ppl: 9.259321000281345] tot_loss[loss=2.291, over 5448298.14 frames. , ppl: 9.884650588297708], batch size: 70
+2022-12-12 05:28:40,353 INFO [train.py:421] (5/8) Epoch 6, batch 48000, loss[loss=2.471, over 980.00 frames. , ppl: 11.838085831480564] tot_loss[loss=2.29, over 5476223.25 frames. , ppl: 9.879045342875653], batch size: 70
+2022-12-12 05:28:40,353 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 05:28:41,113 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.767148491772168
+2022-12-12 05:30:22,489 INFO [train.py:421] (5/8) Epoch 6, batch 48200, loss[loss=2.161, over 4410.00 frames. , ppl: 8.67552748815325] tot_loss[loss=2.292, over 5455018.95 frames. , ppl: 9.889902797084444], batch size: 70
+2022-12-12 05:31:59,735 INFO [train.py:421] (5/8) Epoch 6, batch 48400, loss[loss=2.264, over 5670.00 frames. , ppl: 9.623669192338129] tot_loss[loss=2.292, over 5447601.24 frames. , ppl: 9.89545117697504], batch size: 70
+2022-12-12 05:33:38,677 INFO [train.py:421] (5/8) Epoch 6, batch 48600, loss[loss=2.361, over 2730.00 frames. , ppl: 10.604089932040692] tot_loss[loss=2.292, over 5419991.07 frames. , ppl: 9.898618928078548], batch size: 70
+2022-12-12 05:35:20,096 INFO [train.py:421] (5/8) Epoch 6, batch 48800, loss[loss=2.342, over 2310.00 frames. , ppl: 10.404691723815503] tot_loss[loss=2.291, over 5437838.64 frames. , ppl: 9.88792291914866], batch size: 70
+2022-12-12 05:36:57,681 INFO [train.py:421] (5/8) Epoch 6, batch 49000, loss[loss=2.383, over 2240.00 frames. , ppl: 10.83514075222548] tot_loss[loss=2.293, over 5400754.68 frames. , ppl: 9.900908911055652], batch size: 70
+2022-12-12 05:36:57,681 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 05:36:58,441 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.767533031836354
+2022-12-12 05:38:37,512 INFO [train.py:421] (5/8) Epoch 6, batch 49200, loss[loss=2.361, over 2170.00 frames. , ppl: 10.60505554018657] tot_loss[loss=2.293, over 5411925.60 frames. , ppl: 9.901771290879372], batch size: 70
+2022-12-12 05:40:16,673 INFO [train.py:421] (5/8) Epoch 6, batch 49400, loss[loss=2.288, over 2310.00 frames. , ppl: 9.855742014042807] tot_loss[loss=2.293, over 5393541.18 frames. , ppl: 9.90649391467864], batch size: 70
+2022-12-12 05:41:57,410 INFO [train.py:421] (5/8) Epoch 6, batch 49600, loss[loss=2.255, over 3080.00 frames. , ppl: 9.535882572426452] tot_loss[loss=2.294, over 5392728.25 frames. , ppl: 9.911368198267116], batch size: 70
+2022-12-12 05:43:39,729 INFO [train.py:421] (5/8) Epoch 6, batch 49800, loss[loss=3.095, over 560.00 frames. , ppl: 22.093713245813113] tot_loss[loss=2.292, over 5457842.45 frames. , ppl: 9.89957561969985], batch size: 70
+2022-12-12 05:45:15,853 INFO [train.py:421] (5/8) Epoch 6, batch 50000, loss[loss=2.627, over 700.00 frames. , ppl: 13.83861234899187] tot_loss[loss=2.294, over 5404363.15 frames. , ppl: 9.91369472999919], batch size: 70
+2022-12-12 05:45:15,853 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 05:45:16,606 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.785662608076919
+2022-12-12 05:46:53,747 INFO [train.py:421] (5/8) Epoch 6, batch 50200, loss[loss=2.171, over 5950.00 frames. , ppl: 8.7696662173178] tot_loss[loss=2.294, over 5406489.85 frames. , ppl: 9.91154383067024], batch size: 70
+2022-12-12 05:48:35,502 INFO [train.py:421] (5/8) Epoch 6, batch 50400, loss[loss=2.336, over 2940.00 frames. , ppl: 10.338847842421876] tot_loss[loss=2.293, over 5428805.16 frames. , ppl: 9.905503162413591], batch size: 70
+2022-12-12 05:50:15,397 INFO [train.py:421] (5/8) Epoch 6, batch 50600, loss[loss=2.273, over 1960.00 frames. , ppl: 9.706391134509687] tot_loss[loss=2.292, over 5434185.04 frames. , ppl: 9.899518740091208], batch size: 70
+2022-12-12 05:51:57,550 INFO [train.py:421] (5/8) Epoch 6, batch 50800, loss[loss=2.215, over 5810.00 frames. , ppl: 9.157930264964605] tot_loss[loss=2.292, over 5456485.43 frames. , ppl: 9.89471671567966], batch size: 70
+2022-12-12 05:53:42,724 INFO [train.py:421] (5/8) Epoch 6, batch 51000, loss[loss=2.952, over 560.00 frames. , ppl: 19.13482903478405] tot_loss[loss=2.291, over 5485548.78 frames. , ppl: 9.887729186329631], batch size: 70
+2022-12-12 05:53:42,724 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 05:53:43,469 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.76539078711788
+2022-12-12 05:55:23,427 INFO [train.py:421] (5/8) Epoch 6, batch 51200, loss[loss=2.458, over 1540.00 frames. , ppl: 11.680598063953523] tot_loss[loss=2.29, over 5552535.66 frames. , ppl: 9.870283262960234], batch size: 70
+2022-12-12 05:57:03,660 INFO [train.py:421] (5/8) Epoch 6, batch 51400, loss[loss=2.229, over 4830.00 frames. , ppl: 9.286747763613148] tot_loss[loss=2.289, over 5553685.37 frames. , ppl: 9.861921187782658], batch size: 70
+2022-12-12 05:58:44,078 INFO [train.py:421] (5/8) Epoch 6, batch 51600, loss[loss=2.742, over 630.00 frames. , ppl: 15.52498149045918] tot_loss[loss=2.288, over 5557356.76 frames. , ppl: 9.857256565276414], batch size: 70
+2022-12-12 06:00:26,239 INFO [train.py:421] (5/8) Epoch 6, batch 51800, loss[loss=2.178, over 4900.00 frames. , ppl: 8.831994786471537] tot_loss[loss=2.288, over 5564915.79 frames. , ppl: 9.852926477807587], batch size: 70
+2022-12-12 06:02:05,611 INFO [train.py:421] (5/8) Epoch 6, batch 52000, loss[loss=3.364, over 490.00 frames. , ppl: 28.890647214152963] tot_loss[loss=2.289, over 5569885.76 frames. , ppl: 9.86298442524109], batch size: 70
+2022-12-12 06:02:05,611 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 06:02:06,357 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.278, over 211138.00 frames. , ppl: 9.758165262568566
+2022-12-12 06:03:45,494 INFO [train.py:421] (5/8) Epoch 6, batch 52200, loss[loss=2.431, over 1400.00 frames. , ppl: 11.375398366915398] tot_loss[loss=2.29, over 5530250.48 frames. , ppl: 9.87239351669436], batch size: 70
+2022-12-12 06:05:23,974 INFO [train.py:421] (5/8) Epoch 6, batch 52400, loss[loss=2.217, over 2590.00 frames. , ppl: 9.181277765933107] tot_loss[loss=2.289, over 5550534.16 frames. , ppl: 9.863073512920321], batch size: 70
+2022-12-12 06:07:01,650 INFO [train.py:421] (5/8) Epoch 6, batch 52600, loss[loss=2.106, over 4340.00 frames. , ppl: 8.218289169473834] tot_loss[loss=2.289, over 5505647.31 frames. , ppl: 9.867220793107881], batch size: 70
+2022-12-12 06:08:40,079 INFO [train.py:421] (5/8) Epoch 6, batch 52800, loss[loss=2.192, over 3150.00 frames. , ppl: 8.949729101938862] tot_loss[loss=2.29, over 5489345.14 frames. , ppl: 9.876268591899162], batch size: 70
+2022-12-12 06:10:20,336 INFO [train.py:421] (5/8) Epoch 6, batch 53000, loss[loss=2.216, over 6440.00 frames. , ppl: 9.167348080546462] tot_loss[loss=2.29, over 5499931.73 frames. , ppl: 9.874529120359439], batch size: 70
+2022-12-12 06:10:20,337 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 06:10:21,067 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.764269259359716
+2022-12-12 06:11:59,976 INFO [train.py:421] (5/8) Epoch 6, batch 53200, loss[loss=2.574, over 1260.00 frames. , ppl: 13.11470285883581] tot_loss[loss=2.289, over 5534372.19 frames. , ppl: 9.860794912700753], batch size: 70
+2022-12-12 06:13:40,687 INFO [train.py:421] (5/8) Epoch 6, batch 53400, loss[loss=2.157, over 3150.00 frames. , ppl: 8.646358129922781] tot_loss[loss=2.288, over 5528082.33 frames. , ppl: 9.856135544951462], batch size: 70
+2022-12-12 06:15:21,184 INFO [train.py:421] (5/8) Epoch 6, batch 53600, loss[loss=2.878, over 630.00 frames. , ppl: 17.776276315819374] tot_loss[loss=2.288, over 5514690.38 frames. , ppl: 9.855530813444156], batch size: 70
+2022-12-12 06:17:02,491 INFO [train.py:421] (5/8) Epoch 6, batch 53800, loss[loss=2.279, over 2100.00 frames. , ppl: 9.766224629161666] tot_loss[loss=2.287, over 5527155.15 frames. , ppl: 9.846767285389703], batch size: 70
+2022-12-12 06:18:44,476 INFO [train.py:421] (5/8) Epoch 6, batch 54000, loss[loss=2.41, over 1820.00 frames. , ppl: 11.136077400975456] tot_loss[loss=2.288, over 5519813.79 frames. , ppl: 9.853341841461111], batch size: 70
+2022-12-12 06:18:44,477 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 06:18:45,222 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.773898935036751
+2022-12-12 06:20:25,788 INFO [train.py:421] (5/8) Epoch 6, batch 54200, loss[loss=3.351, over 490.00 frames. , ppl: 28.517599261052677] tot_loss[loss=2.289, over 5485528.23 frames. , ppl: 9.863099824398637], batch size: 70
+2022-12-12 06:22:07,243 INFO [train.py:421] (5/8) Epoch 6, batch 54400, loss[loss=2.247, over 3780.00 frames. , ppl: 9.46208980641502] tot_loss[loss=2.29, over 5462131.32 frames. , ppl: 9.871112201376738], batch size: 70
+2022-12-12 06:23:46,090 INFO [train.py:421] (5/8) Epoch 6, batch 54600, loss[loss=2.388, over 1470.00 frames. , ppl: 10.892527123005973] tot_loss[loss=2.29, over 5471151.89 frames. , ppl: 9.8730425788762], batch size: 70
+2022-12-12 06:25:25,861 INFO [train.py:421] (5/8) Epoch 6, batch 54800, loss[loss=2.284, over 1680.00 frames. , ppl: 9.818914950695982] tot_loss[loss=2.29, over 5465068.10 frames. , ppl: 9.879004184756042], batch size: 70
+2022-12-12 06:27:05,727 INFO [train.py:421] (5/8) Epoch 6, batch 55000, loss[loss=2.296, over 3500.00 frames. , ppl: 9.930351176457659] tot_loss[loss=2.29, over 5472824.82 frames. , ppl: 9.874352166763211], batch size: 70
+2022-12-12 06:27:05,728 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 06:27:06,473 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.278, over 211138.00 frames. , ppl: 9.758666441048378
+2022-12-12 06:28:46,951 INFO [train.py:421] (5/8) Epoch 6, batch 55200, loss[loss=2.193, over 8750.00 frames. , ppl: 8.964565929179477] tot_loss[loss=2.291, over 5413077.97 frames. , ppl: 9.888552418570452], batch size: 70
+2022-12-12 06:30:30,149 INFO [train.py:421] (5/8) Epoch 6, batch 55400, loss[loss=2.383, over 1960.00 frames. , ppl: 10.834044223707245] tot_loss[loss=2.291, over 5418252.66 frames. , ppl: 9.8844680341557], batch size: 70
+2022-12-12 06:32:09,476 INFO [train.py:421] (5/8) Epoch 6, batch 55600, loss[loss=2.152, over 7210.00 frames. , ppl: 8.603326917140365] tot_loss[loss=2.291, over 5412546.75 frames. , ppl: 9.884752020242395], batch size: 70
+2022-12-12 06:33:49,119 INFO [train.py:421] (5/8) Epoch 6, batch 55800, loss[loss=2.191, over 3990.00 frames. , ppl: 8.941578431125013] tot_loss[loss=2.29, over 5413249.18 frames. , ppl: 9.879787791512424], batch size: 70
+2022-12-12 06:35:30,415 INFO [train.py:421] (5/8) Epoch 6, batch 56000, loss[loss=2.596, over 840.00 frames. , ppl: 13.408979512761258] tot_loss[loss=2.29, over 5415034.46 frames. , ppl: 9.878271548793025], batch size: 70
+2022-12-12 06:35:30,415 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 06:35:31,174 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.763342939887206
+2022-12-12 06:37:13,842 INFO [train.py:421] (5/8) Epoch 6, batch 56200, loss[loss=2.478, over 1260.00 frames. , ppl: 11.921537809576249] tot_loss[loss=2.291, over 5383710.51 frames. , ppl: 9.887754133725574], batch size: 70
+2022-12-12 06:38:54,618 INFO [train.py:421] (5/8) Epoch 6, batch 56400, loss[loss=2.279, over 1610.00 frames. , ppl: 9.765679061806312] tot_loss[loss=2.289, over 5451295.80 frames. , ppl: 9.86855271113265], batch size: 70
+2022-12-12 06:40:33,674 INFO [train.py:421] (5/8) Epoch 6, batch 56600, loss[loss=2.363, over 1050.00 frames. , ppl: 10.624406260084427] tot_loss[loss=2.289, over 5466850.61 frames. , ppl: 9.865633518721195], batch size: 70
+2022-12-12 06:42:12,585 INFO [train.py:421] (5/8) Epoch 6, batch 56800, loss[loss=2.179, over 3220.00 frames. , ppl: 8.840812292106841] tot_loss[loss=2.289, over 5472226.20 frames. , ppl: 9.86015926482302], batch size: 70
+2022-12-12 06:43:53,604 INFO [train.py:421] (5/8) Epoch 6, batch 57000, loss[loss=2.248, over 5040.00 frames. , ppl: 9.46574508409069] tot_loss[loss=2.29, over 5470873.22 frames. , ppl: 9.871169476179656], batch size: 70
+2022-12-12 06:43:53,604 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 06:43:54,329 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.763007694592458
+2022-12-12 06:45:35,860 INFO [train.py:421] (5/8) Epoch 6, batch 57200, loss[loss=2.501, over 1260.00 frames. , ppl: 12.18964131811121] tot_loss[loss=2.289, over 5490514.41 frames. , ppl: 9.867972659160998], batch size: 70
+2022-12-12 06:47:18,363 INFO [train.py:421] (5/8) Epoch 6, batch 57400, loss[loss=2.876, over 630.00 frames. , ppl: 17.73563445666883] tot_loss[loss=2.289, over 5514017.33 frames. , ppl: 9.869147135319599], batch size: 70
+2022-12-12 06:48:59,494 INFO [train.py:421] (5/8) Epoch 6, batch 57600, loss[loss=2.27, over 2660.00 frames. , ppl: 9.678790327179893] tot_loss[loss=2.29, over 5507844.15 frames. , ppl: 9.870026223465674], batch size: 70
+2022-12-12 06:50:45,868 INFO [train.py:421] (5/8) Epoch 6, batch 57800, loss[loss=2.293, over 2380.00 frames. , ppl: 9.908874509939102] tot_loss[loss=2.29, over 5494666.58 frames. , ppl: 9.873400496738677], batch size: 70
+2022-12-12 06:52:25,579 INFO [train.py:421] (5/8) Epoch 6, batch 58000, loss[loss=2.342, over 2590.00 frames. , ppl: 10.399613023364067] tot_loss[loss=2.289, over 5502873.01 frames. , ppl: 9.862416248812941], batch size: 70
+2022-12-12 06:52:25,580 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 06:52:26,339 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.278, over 211138.00 frames. , ppl: 9.758448345808779
+2022-12-12 06:54:02,193 INFO [train.py:421] (5/8) Epoch 6, batch 58200, loss[loss=2.318, over 1890.00 frames. , ppl: 10.159975165944314] tot_loss[loss=2.288, over 5510071.58 frames. , ppl: 9.85456696469151], batch size: 70
+2022-12-12 06:55:40,804 INFO [train.py:421] (5/8) Epoch 6, batch 58400, loss[loss=2.23, over 3850.00 frames. , ppl: 9.297361227707917] tot_loss[loss=2.288, over 5489775.89 frames. , ppl: 9.858290994614785], batch size: 70
+2022-12-12 06:57:22,182 INFO [train.py:421] (5/8) Epoch 6, batch 58600, loss[loss=2.215, over 5670.00 frames. , ppl: 9.159282034846804] tot_loss[loss=2.29, over 5463933.11 frames. , ppl: 9.870499557522482], batch size: 70
+2022-12-12 06:59:04,023 INFO [train.py:421] (5/8) Epoch 6, batch 58800, loss[loss=2.915, over 630.00 frames. , ppl: 18.448267628616108] tot_loss[loss=2.289, over 5493777.72 frames. , ppl: 9.862592333524626], batch size: 70
+2022-12-12 07:00:45,054 INFO [train.py:421] (5/8) Epoch 6, batch 59000, loss[loss=2.297, over 2170.00 frames. , ppl: 9.945485111928031] tot_loss[loss=2.287, over 5550743.15 frames. , ppl: 9.841624936308403], batch size: 70
+2022-12-12 07:00:45,054 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 07:00:45,814 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.762743263537212
+2022-12-12 07:02:25,687 INFO [train.py:421] (5/8) Epoch 6, batch 59200, loss[loss=2.281, over 3780.00 frames. , ppl: 9.789935714214023] tot_loss[loss=2.287, over 5544531.32 frames. , ppl: 9.840475241284958], batch size: 70
+2022-12-12 07:04:06,627 INFO [train.py:421] (5/8) Epoch 6, batch 59400, loss[loss=2.248, over 4550.00 frames. , ppl: 9.469304482138227] tot_loss[loss=2.285, over 5580655.75 frames. , ppl: 9.829197444585116], batch size: 70
+2022-12-12 07:05:46,487 INFO [train.py:421] (5/8) Epoch 6, batch 59600, loss[loss=2.27, over 2730.00 frames. , ppl: 9.676352936493508] tot_loss[loss=2.286, over 5560660.41 frames. , ppl: 9.838483556867153], batch size: 70
+2022-12-12 07:07:27,936 INFO [train.py:421] (5/8) Epoch 6, batch 59800, loss[loss=2.801, over 700.00 frames. , ppl: 16.461267860247414] tot_loss[loss=2.288, over 5533491.70 frames. , ppl: 9.850339555341233], batch size: 70
+2022-12-12 07:09:08,960 INFO [train.py:421] (5/8) Epoch 6, batch 60000, loss[loss=2.528, over 1050.00 frames. , ppl: 12.524415340971464] tot_loss[loss=2.287, over 5544771.35 frames. , ppl: 9.843476289188946], batch size: 70
+2022-12-12 07:09:08,961 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 07:09:09,719 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.763438313496728
+2022-12-12 07:10:50,288 INFO [train.py:421] (5/8) Epoch 6, batch 60200, loss[loss=2.211, over 6370.00 frames. , ppl: 9.127716500643775] tot_loss[loss=2.288, over 5504678.16 frames. , ppl: 9.854646428623397], batch size: 70
+2022-12-12 07:12:28,642 INFO [train.py:421] (5/8) Epoch 6, batch 60400, loss[loss=2.525, over 1260.00 frames. , ppl: 12.490033678678818] tot_loss[loss=2.288, over 5476621.44 frames. , ppl: 9.858340386977341], batch size: 70
+2022-12-12 07:14:10,681 INFO [train.py:421] (5/8) Epoch 6, batch 60600, loss[loss=2.469, over 1190.00 frames. , ppl: 11.81269117045582] tot_loss[loss=2.287, over 5496468.67 frames. , ppl: 9.845488976826525], batch size: 70
+2022-12-12 07:15:51,738 INFO [train.py:421] (5/8) Epoch 6, batch 60800, loss[loss=2.58, over 1260.00 frames. , ppl: 13.190836901782385] tot_loss[loss=2.287, over 5512119.83 frames. , ppl: 9.841263020840854], batch size: 70
+2022-12-12 07:17:31,916 INFO [train.py:421] (5/8) Epoch 6, batch 61000, loss[loss=2.177, over 5460.00 frames. , ppl: 8.821111662663409] tot_loss[loss=2.287, over 5509037.69 frames. , ppl: 9.844272685491447], batch size: 70
+2022-12-12 07:17:31,916 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 07:17:32,674 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.767613989568469
+2022-12-12 07:19:10,838 INFO [train.py:421] (5/8) Epoch 6, batch 61200, loss[loss=2.247, over 5810.00 frames. , ppl: 9.460225285254483] tot_loss[loss=2.287, over 5497619.10 frames. , ppl: 9.843484711055876], batch size: 70
+2022-12-12 07:20:50,819 INFO [train.py:421] (5/8) Epoch 6, batch 61400, loss[loss=2.169, over 5880.00 frames. , ppl: 8.746921730143134] tot_loss[loss=2.287, over 5518511.43 frames. , ppl: 9.843757826123534], batch size: 70
+2022-12-12 07:22:30,434 INFO [train.py:421] (5/8) Epoch 6, batch 61600, loss[loss=2.362, over 2380.00 frames. , ppl: 10.615612756291924] tot_loss[loss=2.288, over 5482868.36 frames. , ppl: 9.857432964818058], batch size: 70
+2022-12-12 07:24:09,143 INFO [train.py:421] (5/8) Epoch 6, batch 61800, loss[loss=2.47, over 1330.00 frames. , ppl: 11.824622440354895] tot_loss[loss=2.288, over 5485973.37 frames. , ppl: 9.856067962635727], batch size: 70
+2022-12-12 07:25:49,776 INFO [train.py:421] (5/8) Epoch 6, batch 62000, loss[loss=2.695, over 630.00 frames. , ppl: 14.804548572859463] tot_loss[loss=2.288, over 5429142.58 frames. , ppl: 9.85781274402936], batch size: 70
+2022-12-12 07:25:49,777 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 07:25:50,536 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.776349799606294
+2022-12-12 07:27:29,140 INFO [train.py:421] (5/8) Epoch 6, batch 62200, loss[loss=2.223, over 3990.00 frames. , ppl: 9.239094320291942] tot_loss[loss=2.289, over 5405212.48 frames. , ppl: 9.868494274148288], batch size: 70
+2022-12-12 07:29:06,005 INFO [train.py:421] (5/8) Epoch 6, batch 62400, loss[loss=2.198, over 6720.00 frames. , ppl: 9.010315347087532] tot_loss[loss=2.29, over 5390284.08 frames. , ppl: 9.87446512594974], batch size: 70
+2022-12-12 07:30:49,310 INFO [train.py:421] (5/8) Epoch 6, batch 62600, loss[loss=2.509, over 1050.00 frames. , ppl: 12.293899824216412] tot_loss[loss=2.29, over 5406804.52 frames. , ppl: 9.872321199538874], batch size: 70
+2022-12-12 07:32:26,314 INFO [train.py:421] (5/8) Epoch 6, batch 62800, loss[loss=2.311, over 3430.00 frames. , ppl: 10.08458014770232] tot_loss[loss=2.289, over 5446933.28 frames. , ppl: 9.862571020654205], batch size: 70
+2022-12-12 07:34:06,411 INFO [train.py:421] (5/8) Epoch 6, batch 63000, loss[loss=2.772, over 700.00 frames. , ppl: 15.996543987110899] tot_loss[loss=2.288, over 5474975.97 frames. , ppl: 9.857905165728239], batch size: 70
+2022-12-12 07:34:06,411 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 07:34:07,172 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.784635781017636
+2022-12-12 07:35:46,470 INFO [train.py:421] (5/8) Epoch 6, batch 63200, loss[loss=2.397, over 1540.00 frames. , ppl: 10.99502812317294] tot_loss[loss=2.288, over 5463330.26 frames. , ppl: 9.858690737413028], batch size: 70
+2022-12-12 07:37:31,525 INFO [train.py:421] (5/8) Epoch 6, batch 63400, loss[loss=2.358, over 1820.00 frames. , ppl: 10.566156275317926] tot_loss[loss=2.287, over 5530227.70 frames. , ppl: 9.844633903668818], batch size: 70
+2022-12-12 07:39:13,150 INFO [train.py:421] (5/8) Epoch 6, batch 63600, loss[loss=2.689, over 770.00 frames. , ppl: 14.710192336971312] tot_loss[loss=2.286, over 5547731.90 frames. , ppl: 9.836110780224509], batch size: 70
+2022-12-12 07:40:49,344 INFO [train.py:421] (5/8) Epoch 6, batch 63800, loss[loss=2.381, over 2870.00 frames. , ppl: 10.812177871040683] tot_loss[loss=2.286, over 5524648.37 frames. , ppl: 9.839464914104582], batch size: 70
+2022-12-12 07:42:29,285 INFO [train.py:421] (5/8) Epoch 6, batch 64000, loss[loss=2.698, over 700.00 frames. , ppl: 14.845532446162165] tot_loss[loss=2.286, over 5518864.64 frames. , ppl: 9.838416703095604], batch size: 70
+2022-12-12 07:42:29,286 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 07:42:30,050 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.768704092947166
+2022-12-12 07:44:07,949 INFO [train.py:421] (5/8) Epoch 6, batch 64200, loss[loss=2.34, over 1680.00 frames. , ppl: 10.377111005312253] tot_loss[loss=2.287, over 5492978.23 frames. , ppl: 9.847202474261442], batch size: 70
+2022-12-12 07:45:49,265 INFO [train.py:421] (5/8) Epoch 6, batch 64400, loss[loss=2.33, over 1330.00 frames. , ppl: 10.273851305001267] tot_loss[loss=2.287, over 5529291.56 frames. , ppl: 9.844515674737181], batch size: 70
+2022-12-12 07:47:28,533 INFO [train.py:421] (5/8) Epoch 6, batch 64600, loss[loss=2.451, over 1050.00 frames. , ppl: 11.605258720192351] tot_loss[loss=2.286, over 5545449.52 frames. , ppl: 9.838489288123238], batch size: 70
+2022-12-12 07:49:11,548 INFO [train.py:421] (5/8) Epoch 6, batch 64800, loss[loss=2.38, over 1610.00 frames. , ppl: 10.800355792095386] tot_loss[loss=2.286, over 5568391.20 frames. , ppl: 9.834105910845407], batch size: 70
+2022-12-12 07:50:54,147 INFO [train.py:421] (5/8) Epoch 6, batch 65000, loss[loss=2.785, over 630.00 frames. , ppl: 16.202233138458634] tot_loss[loss=2.285, over 5610613.58 frames. , ppl: 9.823125697847006], batch size: 70
+2022-12-12 07:50:54,148 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 07:50:54,909 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.281, over 211138.00 frames. , ppl: 9.786995181821556
+2022-12-12 07:52:32,871 INFO [train.py:421] (5/8) Epoch 6, batch 65200, loss[loss=2.423, over 770.00 frames. , ppl: 11.27770516517511] tot_loss[loss=2.285, over 5623940.90 frames. , ppl: 9.820990931141303], batch size: 70
+2022-12-12 07:54:12,266 INFO [train.py:421] (5/8) Epoch 6, batch 65400, loss[loss=2.217, over 9940.00 frames. , ppl: 9.183551970068446] tot_loss[loss=2.285, over 5606174.56 frames. , ppl: 9.826733421915128], batch size: 70
+2022-12-12 07:55:56,355 INFO [train.py:421] (5/8) Epoch 6, batch 65600, loss[loss=2.363, over 2170.00 frames. , ppl: 10.62035791744929] tot_loss[loss=2.285, over 5595915.02 frames. , ppl: 9.828857829066669], batch size: 70
+2022-12-12 07:57:38,642 INFO [train.py:421] (5/8) Epoch 6, batch 65800, loss[loss=2.297, over 2800.00 frames. , ppl: 9.946214574808529] tot_loss[loss=2.285, over 5605956.88 frames. , ppl: 9.82649293000418], batch size: 70
+2022-12-12 07:59:17,514 INFO [train.py:421] (5/8) Epoch 6, batch 66000, loss[loss=2.27, over 1540.00 frames. , ppl: 9.675946036401319] tot_loss[loss=2.285, over 5585130.05 frames. , ppl: 9.827889451086103], batch size: 70
+2022-12-12 07:59:17,514 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 07:59:18,273 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.278, over 211138.00 frames. , ppl: 9.756986800195627
+2022-12-12 08:00:55,080 INFO [train.py:421] (5/8) Epoch 6, batch 66200, loss[loss=2.226, over 2660.00 frames. , ppl: 9.263618860291919] tot_loss[loss=2.286, over 5577942.10 frames. , ppl: 9.835386668859242], batch size: 70
+2022-12-12 08:02:34,946 INFO [train.py:421] (5/8) Epoch 6, batch 66400, loss[loss=2.179, over 3640.00 frames. , ppl: 8.837803571835092] tot_loss[loss=2.286, over 5593863.98 frames. , ppl: 9.83081874207353], batch size: 70
+2022-12-12 08:04:12,905 INFO [train.py:421] (5/8) Epoch 6, batch 66600, loss[loss=2.276, over 4760.00 frames. , ppl: 9.736454721960843] tot_loss[loss=2.286, over 5604271.64 frames. , ppl: 9.831824517513741], batch size: 70
+2022-12-12 08:05:56,265 INFO [train.py:421] (5/8) Epoch 6, batch 66800, loss[loss=2.433, over 2380.00 frames. , ppl: 11.390621515045572] tot_loss[loss=2.286, over 5618787.65 frames. , ppl: 9.832913091912319], batch size: 70
+2022-12-12 08:07:41,371 INFO [train.py:421] (5/8) Epoch 6, batch 67000, loss[loss=2.391, over 1120.00 frames. , ppl: 10.921997728845355] tot_loss[loss=2.287, over 5594309.99 frames. , ppl: 9.84151350141511], batch size: 70
+2022-12-12 08:07:41,371 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 08:07:42,129 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.766049889291654
+2022-12-12 08:09:21,214 INFO [train.py:421] (5/8) Epoch 6, batch 67200, loss[loss=2.389, over 910.00 frames. , ppl: 10.905581414370811] tot_loss[loss=2.286, over 5605424.79 frames. , ppl: 9.83719535780625], batch size: 70
+2022-12-12 08:11:01,750 INFO [train.py:421] (5/8) Epoch 6, batch 67400, loss[loss=2.314, over 2240.00 frames. , ppl: 10.11324152450891] tot_loss[loss=2.286, over 5611750.72 frames. , ppl: 9.83502310046445], batch size: 70
+2022-12-12 08:12:41,351 INFO [train.py:421] (5/8) Epoch 6, batch 67600, loss[loss=2.424, over 1050.00 frames. , ppl: 11.295955648753626] tot_loss[loss=2.286, over 5571030.85 frames. , ppl: 9.839094454165933], batch size: 70
+2022-12-12 08:14:15,982 INFO [train.py:421] (5/8) Epoch 6, batch 67800, loss[loss=2.331, over 2380.00 frames. , ppl: 10.287132494973912] tot_loss[loss=2.286, over 5567442.94 frames. , ppl: 9.83629099525833], batch size: 70
+2022-12-12 08:15:54,493 INFO [train.py:421] (5/8) Epoch 6, batch 68000, loss[loss=2.357, over 2030.00 frames. , ppl: 10.555939966995492] tot_loss[loss=2.288, over 5517139.02 frames. , ppl: 9.850311777402212], batch size: 70
+2022-12-12 08:15:54,493 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 08:15:55,254 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.28, over 211138.00 frames. , ppl: 9.779122592522853
+2022-12-12 08:17:39,587 INFO [train.py:421] (5/8) Epoch 6, batch 68200, loss[loss=2.27, over 3150.00 frames. , ppl: 9.675050622773416] tot_loss[loss=2.288, over 5519124.60 frames. , ppl: 9.851820923028937], batch size: 70
+2022-12-12 08:19:22,896 INFO [train.py:421] (5/8) Epoch 6, batch 68400, loss[loss=2.428, over 1050.00 frames. , ppl: 11.333719093895862] tot_loss[loss=2.286, over 5570316.31 frames. , ppl: 9.831980707870104], batch size: 70
+2022-12-12 08:21:04,973 INFO [train.py:421] (5/8) Epoch 6, batch 68600, loss[loss=2.283, over 1400.00 frames. , ppl: 9.807085356257875] tot_loss[loss=2.285, over 5562188.38 frames. , ppl: 9.83041404663754], batch size: 70
+2022-12-12 08:22:43,487 INFO [train.py:421] (5/8) Epoch 6, batch 68800, loss[loss=2.345, over 1680.00 frames. , ppl: 10.431205466634339] tot_loss[loss=2.285, over 5549968.05 frames. , ppl: 9.829726759567466], batch size: 70
+2022-12-12 08:24:23,648 INFO [train.py:421] (5/8) Epoch 6, batch 69000, loss[loss=2.201, over 4200.00 frames. , ppl: 9.033520425360765] tot_loss[loss=2.286, over 5518852.43 frames. , ppl: 9.835790636721764], batch size: 70
+2022-12-12 08:24:23,648 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 08:24:24,395 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.279, over 211138.00 frames. , ppl: 9.762691245156288
+2022-12-12 08:26:05,608 INFO [train.py:421] (5/8) Epoch 6, batch 69200, loss[loss=2.24, over 8400.00 frames. , ppl: 9.394587222748687] tot_loss[loss=2.285, over 5535323.26 frames. , ppl: 9.830060874442115], batch size: 70
+2022-12-12 08:27:44,996 INFO [train.py:421] (5/8) Epoch 6, batch 69400, loss[loss=3.163, over 560.00 frames. , ppl: 23.64135328655156] tot_loss[loss=2.285, over 5546254.41 frames. , ppl: 9.823519099704892], batch size: 70
+2022-12-12 08:29:30,567 INFO [train.py:421] (5/8) Epoch 6, batch 69600, loss[loss=2.455, over 1330.00 frames. , ppl: 11.649518049031713] tot_loss[loss=2.286, over 5533302.98 frames. , ppl: 9.835115927581091], batch size: 70
+2022-12-12 08:31:11,122 INFO [train.py:421] (5/8) Epoch 6, batch 69800, loss[loss=3.623, over 420.00 frames. , ppl: 37.43129614534657] tot_loss[loss=2.287, over 5520456.47 frames. , ppl: 9.846834010100025], batch size: 70
+2022-12-12 08:32:55,437 INFO [train.py:421] (5/8) Epoch 6, batch 70000, loss[loss=2.209, over 8400.00 frames. , ppl: 9.110033559638497] tot_loss[loss=2.287, over 5534009.56 frames. , ppl: 9.848004363066257], batch size: 70
+2022-12-12 08:32:55,437 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 08:32:56,166 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.278, over 211138.00 frames. , ppl: 9.761897999189102
+2022-12-12 08:34:38,124 INFO [train.py:421] (5/8) Epoch 6, batch 70200, loss[loss=2.466, over 1330.00 frames. , ppl: 11.779281533889396] tot_loss[loss=2.287, over 5575463.87 frames. , ppl: 9.841043679821542], batch size: 70
+2022-12-12 08:36:17,125 INFO [train.py:421] (5/8) Epoch 6, batch 70400, loss[loss=2.999, over 560.00 frames. , ppl: 20.07418282443173] tot_loss[loss=2.287, over 5556846.28 frames. , ppl: 9.844835896652278], batch size: 70
+2022-12-12 08:37:57,455 INFO [train.py:421] (5/8) Epoch 6, batch 70600, loss[loss=2.61, over 910.00 frames. , ppl: 13.596644553453187] tot_loss[loss=2.287, over 5539067.76 frames. , ppl: 9.847468043428167], batch size: 70
+2022-12-12 08:39:34,984 INFO [train.py:421] (5/8) Epoch 6, batch 70800, loss[loss=2.22, over 2660.00 frames. , ppl: 9.210577179014003] tot_loss[loss=2.286, over 5577741.62 frames. , ppl: 9.833493193724909], batch size: 70
+2022-12-12 08:41:17,309 INFO [train.py:421] (5/8) Epoch 6, batch 71000, loss[loss=2.491, over 1120.00 frames. , ppl: 12.077664742981835] tot_loss[loss=2.286, over 5575513.59 frames. , ppl: 9.836944979248287], batch size: 70
+2022-12-12 08:41:17,310 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 08:41:18,069 INFO [train.py:452] (5/8) Epoch 6, validation: loss=2.278, over 211138.00 frames. , ppl: 9.760633851500042
+2022-12-12 08:42:55,817 INFO [train.py:421] (5/8) Epoch 6, batch 71200, loss[loss=2.347, over 1960.00 frames. , ppl: 10.45745061890089] tot_loss[loss=2.287, over 5549603.71 frames. , ppl: 9.842662197077582], batch size: 70
+2022-12-12 08:44:37,451 INFO [train.py:421] (5/8) Epoch 6, batch 71400, loss[loss=2.24, over 9940.00 frames. , ppl: 9.39221433060462] tot_loss[loss=2.285, over 5603853.25 frames. , ppl: 9.826858649994877], batch size: 70
+2022-12-12 08:46:15,621 INFO [train.py:421] (5/8) Epoch 6, batch 71600, loss[loss=2.29, over 2240.00 frames. , ppl: 9.870816808659473] tot_loss[loss=2.286, over 5576988.58 frames. , ppl: 9.838764053620057], batch size: 70
+2022-12-12 08:47:54,854 INFO [train.py:421] (5/8) Epoch 6, batch 71800, loss[loss=2.734, over 700.00 frames. , ppl: 15.396141229089416] tot_loss[loss=2.287, over 5558805.27 frames. , ppl: 9.847837071315288], batch size: 70
+2022-12-12 08:49:08,129 INFO [train.py:421] (5/8) Epoch 7, batch 0, loss[loss=2.272, over 2240.00 frames. , ppl: 9.70356782513179] tot_loss[loss=2.272, over 2240.00 frames. , ppl: 9.70356782513179], batch size: 70
+2022-12-12 08:50:48,795 INFO [train.py:421] (5/8) Epoch 7, batch 200, loss[loss=2.937, over 560.00 frames. , ppl: 18.86716802760524] tot_loss[loss=2.282, over 510294.18 frames. , ppl: 9.797867199287591], batch size: 70
+2022-12-12 08:52:29,295 INFO [train.py:421] (5/8) Epoch 7, batch 400, loss[loss=2.227, over 5110.00 frames. , ppl: 9.26758073187742] tot_loss[loss=2.285, over 938492.45 frames. , ppl: 9.830303081047186], batch size: 70
+2022-12-12 08:54:08,470 INFO [train.py:421] (5/8) Epoch 7, batch 600, loss[loss=2.37, over 1750.00 frames. , ppl: 10.695929849983068] tot_loss[loss=2.283, over 1384643.20 frames. , ppl: 9.801546457865395], batch size: 70
+2022-12-12 08:55:44,635 INFO [train.py:421] (5/8) Epoch 7, batch 800, loss[loss=2.42, over 3430.00 frames. , ppl: 11.243437709759922] tot_loss[loss=2.286, over 1753704.32 frames. , ppl: 9.833405829923663], batch size: 70
+2022-12-12 08:57:20,593 INFO [train.py:421] (5/8) Epoch 7, batch 1000, loss[loss=2.476, over 980.00 frames. , ppl: 11.897906427305259] tot_loss[loss=2.287, over 2078668.01 frames. , ppl: 9.844947577132176], batch size: 70
+2022-12-12 08:57:20,593 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 08:57:21,340 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.766622303585429
+2022-12-12 08:58:56,601 INFO [train.py:421] (5/8) Epoch 7, batch 1200, loss[loss=2.19, over 5320.00 frames. , ppl: 8.930985385453148] tot_loss[loss=2.29, over 2358121.90 frames. , ppl: 9.87056704766409], batch size: 70
+2022-12-12 09:00:32,854 INFO [train.py:421] (5/8) Epoch 7, batch 1400, loss[loss=2.416, over 980.00 frames. , ppl: 11.197390986957108] tot_loss[loss=2.288, over 2651622.94 frames. , ppl: 9.85287414451139], batch size: 70
+2022-12-12 09:02:12,540 INFO [train.py:421] (5/8) Epoch 7, batch 1600, loss[loss=2.201, over 4270.00 frames. , ppl: 9.029949990467657] tot_loss[loss=2.286, over 2940243.78 frames. , ppl: 9.83076376078132], batch size: 70
+2022-12-12 09:03:50,387 INFO [train.py:421] (5/8) Epoch 7, batch 1800, loss[loss=2.182, over 3010.00 frames. , ppl: 8.867697892240443] tot_loss[loss=2.283, over 3206301.50 frames. , ppl: 9.80907063611224], batch size: 70
+2022-12-12 09:05:32,160 INFO [train.py:421] (5/8) Epoch 7, batch 2000, loss[loss=2.32, over 2660.00 frames. , ppl: 10.179717973328938] tot_loss[loss=2.282, over 3442633.75 frames. , ppl: 9.798065731914752], batch size: 70
+2022-12-12 09:05:32,160 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 09:05:32,901 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.764543848320063
+2022-12-12 09:07:08,130 INFO [train.py:421] (5/8) Epoch 7, batch 2200, loss[loss=2.46, over 1820.00 frames. , ppl: 11.701823667338784] tot_loss[loss=2.282, over 3636496.33 frames. , ppl: 9.792557472685248], batch size: 70
+2022-12-12 09:08:47,616 INFO [train.py:421] (5/8) Epoch 7, batch 2400, loss[loss=2.317, over 2660.00 frames. , ppl: 10.149416551422204] tot_loss[loss=2.281, over 3805047.37 frames. , ppl: 9.788185881573241], batch size: 70
+2022-12-12 09:10:26,448 INFO [train.py:421] (5/8) Epoch 7, batch 2600, loss[loss=2.202, over 11690.00 frames. , ppl: 9.045013373147267] tot_loss[loss=2.281, over 3970095.17 frames. , ppl: 9.789563092086837], batch size: 70
+2022-12-12 09:12:07,859 INFO [train.py:421] (5/8) Epoch 7, batch 2800, loss[loss=2.292, over 2310.00 frames. , ppl: 9.891447268253534] tot_loss[loss=2.279, over 4148992.77 frames. , ppl: 9.766027591750982], batch size: 70
+2022-12-12 09:13:50,922 INFO [train.py:421] (5/8) Epoch 7, batch 3000, loss[loss=2.201, over 3500.00 frames. , ppl: 9.037080932101869] tot_loss[loss=2.28, over 4262438.67 frames. , ppl: 9.77403128036625], batch size: 70
+2022-12-12 09:13:50,923 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 09:13:51,668 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.743588919093323
+2022-12-12 09:15:28,748 INFO [train.py:421] (5/8) Epoch 7, batch 3200, loss[loss=2.639, over 910.00 frames. , ppl: 13.998697746437491] tot_loss[loss=2.281, over 4361479.66 frames. , ppl: 9.783380916915148], batch size: 70
+2022-12-12 09:17:10,259 INFO [train.py:421] (5/8) Epoch 7, batch 3400, loss[loss=2.192, over 4690.00 frames. , ppl: 8.957383625882764] tot_loss[loss=2.281, over 4460596.72 frames. , ppl: 9.78767280224644], batch size: 70
+2022-12-12 09:18:51,745 INFO [train.py:421] (5/8) Epoch 7, batch 3600, loss[loss=2.262, over 3500.00 frames. , ppl: 9.601184149105901] tot_loss[loss=2.281, over 4578332.48 frames. , ppl: 9.783309757825823], batch size: 70
+2022-12-12 09:20:35,630 INFO [train.py:421] (5/8) Epoch 7, batch 3800, loss[loss=2.253, over 3780.00 frames. , ppl: 9.512824659175397] tot_loss[loss=2.28, over 4688183.83 frames. , ppl: 9.77769718348744], batch size: 70
+2022-12-12 09:22:13,513 INFO [train.py:421] (5/8) Epoch 7, batch 4000, loss[loss=2.166, over 2660.00 frames. , ppl: 8.727121760996965] tot_loss[loss=2.28, over 4729798.00 frames. , ppl: 9.776444579126945], batch size: 70
+2022-12-12 09:22:13,513 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 09:22:14,273 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.764824226068129
+2022-12-12 09:23:52,034 INFO [train.py:421] (5/8) Epoch 7, batch 4200, loss[loss=2.203, over 7910.00 frames. , ppl: 9.051012229597925] tot_loss[loss=2.28, over 4832444.04 frames. , ppl: 9.774171759657161], batch size: 70
+2022-12-12 09:25:30,245 INFO [train.py:421] (5/8) Epoch 7, batch 4400, loss[loss=2.324, over 2100.00 frames. , ppl: 10.218313459435219] tot_loss[loss=2.281, over 4857793.99 frames. , ppl: 9.784396629569763], batch size: 70
+2022-12-12 09:27:09,449 INFO [train.py:421] (5/8) Epoch 7, batch 4600, loss[loss=2.292, over 1470.00 frames. , ppl: 9.889780124311669] tot_loss[loss=2.281, over 4911857.70 frames. , ppl: 9.784044751408734], batch size: 70
+2022-12-12 09:28:50,537 INFO [train.py:421] (5/8) Epoch 7, batch 4800, loss[loss=2.393, over 1610.00 frames. , ppl: 10.945073454681642] tot_loss[loss=2.281, over 4961544.64 frames. , ppl: 9.790890039565484], batch size: 70
+2022-12-12 09:30:30,393 INFO [train.py:421] (5/8) Epoch 7, batch 5000, loss[loss=2.239, over 3640.00 frames. , ppl: 9.38463239397145] tot_loss[loss=2.283, over 4996191.67 frames. , ppl: 9.801252543971101], batch size: 70
+2022-12-12 09:30:30,394 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 09:30:31,152 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.770388643082995
+2022-12-12 09:32:12,050 INFO [train.py:421] (5/8) Epoch 7, batch 5200, loss[loss=2.429, over 1820.00 frames. , ppl: 11.347722378400805] tot_loss[loss=2.283, over 5043853.52 frames. , ppl: 9.80293375863843], batch size: 70
+2022-12-12 09:33:52,020 INFO [train.py:421] (5/8) Epoch 7, batch 5400, loss[loss=2.453, over 980.00 frames. , ppl: 11.62331383627178] tot_loss[loss=2.282, over 5083732.17 frames. , ppl: 9.798412438788846], batch size: 70
+2022-12-12 09:35:30,666 INFO [train.py:421] (5/8) Epoch 7, batch 5600, loss[loss=2.566, over 910.00 frames. , ppl: 13.009260980474439] tot_loss[loss=2.283, over 5080597.62 frames. , ppl: 9.808360976495349], batch size: 70
+2022-12-12 09:37:13,218 INFO [train.py:421] (5/8) Epoch 7, batch 5800, loss[loss=2.135, over 5950.00 frames. , ppl: 8.454864790375556] tot_loss[loss=2.283, over 5159326.50 frames. , ppl: 9.801192607136711], batch size: 70
+2022-12-12 09:38:55,653 INFO [train.py:421] (5/8) Epoch 7, batch 6000, loss[loss=2.219, over 2940.00 frames. , ppl: 9.201481499214324] tot_loss[loss=2.281, over 5249676.21 frames. , ppl: 9.785836687799378], batch size: 70
+2022-12-12 09:38:55,653 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 09:38:56,400 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.766977911177703
+2022-12-12 09:40:36,648 INFO [train.py:421] (5/8) Epoch 7, batch 6200, loss[loss=2.8, over 630.00 frames. , ppl: 16.448929788854088] tot_loss[loss=2.281, over 5249366.87 frames. , ppl: 9.78831323755466], batch size: 70
+2022-12-12 09:42:17,027 INFO [train.py:421] (5/8) Epoch 7, batch 6400, loss[loss=2.266, over 4550.00 frames. , ppl: 9.640617356135788] tot_loss[loss=2.281, over 5299162.01 frames. , ppl: 9.786136790817888], batch size: 70
+2022-12-12 09:43:57,321 INFO [train.py:421] (5/8) Epoch 7, batch 6600, loss[loss=2.612, over 700.00 frames. , ppl: 13.631077483365646] tot_loss[loss=2.282, over 5261078.48 frames. , ppl: 9.795414251035524], batch size: 70
+2022-12-12 09:45:38,137 INFO [train.py:421] (5/8) Epoch 7, batch 6800, loss[loss=2.194, over 5180.00 frames. , ppl: 8.97450332947396] tot_loss[loss=2.283, over 5262167.65 frames. , ppl: 9.802826229954267], batch size: 70
+2022-12-12 09:47:22,573 INFO [train.py:421] (5/8) Epoch 7, batch 7000, loss[loss=2.239, over 4620.00 frames. , ppl: 9.388001239677047] tot_loss[loss=2.282, over 5274944.86 frames. , ppl: 9.800903654108083], batch size: 70
+2022-12-12 09:47:22,573 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 09:47:23,333 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.278, over 211138.00 frames. , ppl: 9.761162606501355
+2022-12-12 09:49:03,423 INFO [train.py:421] (5/8) Epoch 7, batch 7200, loss[loss=3.569, over 420.00 frames. , ppl: 35.49790577781239] tot_loss[loss=2.282, over 5296363.03 frames. , ppl: 9.799050191762154], batch size: 70
+2022-12-12 09:50:43,930 INFO [train.py:421] (5/8) Epoch 7, batch 7400, loss[loss=2.258, over 2170.00 frames. , ppl: 9.567648043633074] tot_loss[loss=2.281, over 5337255.88 frames. , ppl: 9.788480216004467], batch size: 70
+2022-12-12 09:52:24,698 INFO [train.py:421] (5/8) Epoch 7, batch 7600, loss[loss=2.733, over 700.00 frames. , ppl: 15.384589867096986] tot_loss[loss=2.281, over 5381430.69 frames. , ppl: 9.78437480906808], batch size: 70
+2022-12-12 09:54:01,291 INFO [train.py:421] (5/8) Epoch 7, batch 7800, loss[loss=2.296, over 2520.00 frames. , ppl: 9.936399938651633] tot_loss[loss=2.282, over 5353314.90 frames. , ppl: 9.79854292684087], batch size: 70
+2022-12-12 09:55:38,126 INFO [train.py:421] (5/8) Epoch 7, batch 8000, loss[loss=2.487, over 1330.00 frames. , ppl: 12.027079988208863] tot_loss[loss=2.283, over 5344653.47 frames. , ppl: 9.804851395647407], batch size: 70
+2022-12-12 09:55:38,127 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 09:55:38,887 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.762243320567698
+2022-12-12 09:57:20,362 INFO [train.py:421] (5/8) Epoch 7, batch 8200, loss[loss=2.33, over 2870.00 frames. , ppl: 10.274642775035339] tot_loss[loss=2.283, over 5360622.63 frames. , ppl: 9.808279411188632], batch size: 70
+2022-12-12 09:59:00,364 INFO [train.py:421] (5/8) Epoch 7, batch 8400, loss[loss=2.979, over 560.00 frames. , ppl: 19.671281177937015] tot_loss[loss=2.285, over 5322140.35 frames. , ppl: 9.824802328316267], batch size: 70
+2022-12-12 10:00:40,167 INFO [train.py:421] (5/8) Epoch 7, batch 8600, loss[loss=2.171, over 8330.00 frames. , ppl: 8.76580899338051] tot_loss[loss=2.284, over 5365367.58 frames. , ppl: 9.81421047053673], batch size: 70
+2022-12-12 10:02:20,359 INFO [train.py:421] (5/8) Epoch 7, batch 8800, loss[loss=2.564, over 1050.00 frames. , ppl: 12.987521674496733] tot_loss[loss=2.283, over 5357316.41 frames. , ppl: 9.810646041527065], batch size: 70
+2022-12-12 10:04:02,209 INFO [train.py:421] (5/8) Epoch 7, batch 9000, loss[loss=2.45, over 1050.00 frames. , ppl: 11.585385882635912] tot_loss[loss=2.284, over 5362364.80 frames. , ppl: 9.811867544036888], batch size: 70
+2022-12-12 10:04:02,210 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 10:04:02,970 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.278, over 211138.00 frames. , ppl: 9.758306803162158
+2022-12-12 10:05:41,185 INFO [train.py:421] (5/8) Epoch 7, batch 9200, loss[loss=2.335, over 1400.00 frames. , ppl: 10.332243335772255] tot_loss[loss=2.284, over 5355244.77 frames. , ppl: 9.8153758918827], batch size: 70
+2022-12-12 10:07:18,496 INFO [train.py:421] (5/8) Epoch 7, batch 9400, loss[loss=2.393, over 1470.00 frames. , ppl: 10.944625420519477] tot_loss[loss=2.284, over 5368189.28 frames. , ppl: 9.815098033488683], batch size: 70
+2022-12-12 10:08:57,707 INFO [train.py:421] (5/8) Epoch 7, batch 9600, loss[loss=2.349, over 3220.00 frames. , ppl: 10.474435867020311] tot_loss[loss=2.285, over 5351158.07 frames. , ppl: 9.82202645777347], batch size: 70
+2022-12-12 10:10:41,235 INFO [train.py:421] (5/8) Epoch 7, batch 9800, loss[loss=2.382, over 2170.00 frames. , ppl: 10.82650720418994] tot_loss[loss=2.285, over 5354467.91 frames. , ppl: 9.824025557969534], batch size: 70
+2022-12-12 10:12:17,286 INFO [train.py:421] (5/8) Epoch 7, batch 10000, loss[loss=2.135, over 10920.00 frames. , ppl: 8.459210719374417] tot_loss[loss=2.284, over 5355393.63 frames. , ppl: 9.820481555881281], batch size: 70
+2022-12-12 10:12:17,287 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 10:12:18,035 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.743937919324686
+2022-12-12 10:13:56,304 INFO [train.py:421] (5/8) Epoch 7, batch 10200, loss[loss=2.172, over 5880.00 frames. , ppl: 8.772920161065171] tot_loss[loss=2.284, over 5382825.66 frames. , ppl: 9.81245169826915], batch size: 70
+2022-12-12 10:15:40,296 INFO [train.py:421] (5/8) Epoch 7, batch 10400, loss[loss=2.778, over 770.00 frames. , ppl: 16.088188460944114] tot_loss[loss=2.284, over 5386756.29 frames. , ppl: 9.817235325099277], batch size: 70
+2022-12-12 10:17:21,282 INFO [train.py:421] (5/8) Epoch 7, batch 10600, loss[loss=2.283, over 2730.00 frames. , ppl: 9.801936665872459] tot_loss[loss=2.284, over 5391686.91 frames. , ppl: 9.814864494133856], batch size: 70
+2022-12-12 10:18:59,590 INFO [train.py:421] (5/8) Epoch 7, batch 10800, loss[loss=2.269, over 1680.00 frames. , ppl: 9.672455127841063] tot_loss[loss=2.285, over 5391818.89 frames. , ppl: 9.821765186549234], batch size: 70
+2022-12-12 10:20:44,636 INFO [train.py:421] (5/8) Epoch 7, batch 11000, loss[loss=2.34, over 1820.00 frames. , ppl: 10.383814812680106] tot_loss[loss=2.285, over 5394236.96 frames. , ppl: 9.822330863755216], batch size: 70
+2022-12-12 10:20:44,637 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 10:20:45,398 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.764081387151732
+2022-12-12 10:22:26,516 INFO [train.py:421] (5/8) Epoch 7, batch 11200, loss[loss=2.596, over 980.00 frames. , ppl: 13.413713817515315] tot_loss[loss=2.284, over 5417987.26 frames. , ppl: 9.816810044446488], batch size: 70
+2022-12-12 10:24:09,620 INFO [train.py:421] (5/8) Epoch 7, batch 11400, loss[loss=3.268, over 490.00 frames. , ppl: 26.25329157112412] tot_loss[loss=2.285, over 5412194.43 frames. , ppl: 9.823736131786708], batch size: 70
+2022-12-12 10:25:51,780 INFO [train.py:421] (5/8) Epoch 7, batch 11600, loss[loss=2.391, over 1190.00 frames. , ppl: 10.922010893313109] tot_loss[loss=2.286, over 5381881.21 frames. , ppl: 9.83283772841991], batch size: 70
+2022-12-12 10:27:33,049 INFO [train.py:421] (5/8) Epoch 7, batch 11800, loss[loss=2.292, over 4550.00 frames. , ppl: 9.892052210985105] tot_loss[loss=2.285, over 5398541.30 frames. , ppl: 9.82487364390566], batch size: 70
+2022-12-12 10:29:14,367 INFO [train.py:421] (5/8) Epoch 7, batch 12000, loss[loss=2.201, over 4550.00 frames. , ppl: 9.037267719509053] tot_loss[loss=2.285, over 5401716.62 frames. , ppl: 9.82426965209854], batch size: 70
+2022-12-12 10:29:14,368 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 10:29:15,131 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.770513007653056
+2022-12-12 10:30:57,657 INFO [train.py:421] (5/8) Epoch 7, batch 12200, loss[loss=2.228, over 4690.00 frames. , ppl: 9.281853746190219] tot_loss[loss=2.283, over 5436260.59 frames. , ppl: 9.808420092200654], batch size: 70
+2022-12-12 10:32:39,419 INFO [train.py:421] (5/8) Epoch 7, batch 12400, loss[loss=2.262, over 4340.00 frames. , ppl: 9.601073450607265] tot_loss[loss=2.283, over 5434988.27 frames. , ppl: 9.808563371249097], batch size: 70
+2022-12-12 10:34:19,730 INFO [train.py:421] (5/8) Epoch 7, batch 12600, loss[loss=2.176, over 2870.00 frames. , ppl: 8.812945534105495] tot_loss[loss=2.283, over 5423946.21 frames. , ppl: 9.803072652441452], batch size: 70
+2022-12-12 10:35:58,810 INFO [train.py:421] (5/8) Epoch 7, batch 12800, loss[loss=2.251, over 2450.00 frames. , ppl: 9.499326510135466] tot_loss[loss=2.282, over 5450558.15 frames. , ppl: 9.796564080008817], batch size: 70
+2022-12-12 10:37:35,216 INFO [train.py:421] (5/8) Epoch 7, batch 13000, loss[loss=2.2, over 6230.00 frames. , ppl: 9.028572134510217] tot_loss[loss=2.283, over 5442927.17 frames. , ppl: 9.801955947183751], batch size: 70
+2022-12-12 10:37:35,217 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 10:37:35,946 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.279, over 211138.00 frames. , ppl: 9.769163881244294
+2022-12-12 10:39:15,584 INFO [train.py:421] (5/8) Epoch 7, batch 13200, loss[loss=2.749, over 840.00 frames. , ppl: 15.629643579957357] tot_loss[loss=2.282, over 5464779.11 frames. , ppl: 9.796603725468373], batch size: 70
+2022-12-12 10:40:55,457 INFO [train.py:421] (5/8) Epoch 7, batch 13400, loss[loss=2.428, over 980.00 frames. , ppl: 11.340369612937568] tot_loss[loss=2.283, over 5435502.85 frames. , ppl: 9.806533611202449], batch size: 70
+2022-12-12 10:42:34,882 INFO [train.py:421] (5/8) Epoch 7, batch 13600, loss[loss=2.427, over 1050.00 frames. , ppl: 11.32375429475351] tot_loss[loss=2.283, over 5434537.17 frames. , ppl: 9.805495434865879], batch size: 70
+2022-12-12 10:44:12,329 INFO [train.py:421] (5/8) Epoch 7, batch 13800, loss[loss=2.163, over 3920.00 frames. , ppl: 8.69897339684635] tot_loss[loss=2.283, over 5441626.07 frames. , ppl: 9.804345065681396], batch size: 70
+2022-12-12 10:45:49,784 INFO [train.py:421] (5/8) Epoch 7, batch 14000, loss[loss=2.356, over 1820.00 frames. , ppl: 10.552201490267363] tot_loss[loss=2.284, over 5406917.36 frames. , ppl: 9.818016480749332], batch size: 70
+2022-12-12 10:45:49,785 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 10:45:50,534 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.282, over 211138.00 frames. , ppl: 9.796524022897163
+2022-12-12 10:47:32,824 INFO [train.py:421] (5/8) Epoch 7, batch 14200, loss[loss=2.376, over 1470.00 frames. , ppl: 10.76408421216453] tot_loss[loss=2.283, over 5453493.05 frames. , ppl: 9.807144206614115], batch size: 70
+2022-12-12 10:49:07,720 INFO [train.py:421] (5/8) Epoch 7, batch 14400, loss[loss=2.634, over 700.00 frames. , ppl: 13.930115554782887] tot_loss[loss=2.284, over 5429297.51 frames. , ppl: 9.81917766087905], batch size: 70
+2022-12-12 10:50:46,721 INFO [train.py:421] (5/8) Epoch 7, batch 14600, loss[loss=2.245, over 2730.00 frames. , ppl: 9.43700576014691] tot_loss[loss=2.285, over 5407275.68 frames. , ppl: 9.82440767197647], batch size: 70
+2022-12-12 10:52:28,728 INFO [train.py:421] (5/8) Epoch 7, batch 14800, loss[loss=2.416, over 980.00 frames. , ppl: 11.199511229407879] tot_loss[loss=2.284, over 5415438.76 frames. , ppl: 9.82028249220266], batch size: 70
+2022-12-12 10:54:11,258 INFO [train.py:421] (5/8) Epoch 7, batch 15000, loss[loss=2.351, over 3360.00 frames. , ppl: 10.500489010389076] tot_loss[loss=2.283, over 5471586.04 frames. , ppl: 9.80323161472668], batch size: 70
+2022-12-12 10:54:11,259 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 10:54:12,022 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.747404083511375
+2022-12-12 10:55:53,710 INFO [train.py:421] (5/8) Epoch 7, batch 15200, loss[loss=2.257, over 1820.00 frames. , ppl: 9.55365113154063] tot_loss[loss=2.283, over 5470942.34 frames. , ppl: 9.806191156525768], batch size: 70
+2022-12-12 10:57:32,864 INFO [train.py:421] (5/8) Epoch 7, batch 15400, loss[loss=2.283, over 3080.00 frames. , ppl: 9.808534797039004] tot_loss[loss=2.283, over 5464843.16 frames. , ppl: 9.809389796580398], batch size: 70
+2022-12-12 10:59:08,552 INFO [train.py:421] (5/8) Epoch 7, batch 15600, loss[loss=2.951, over 490.00 frames. , ppl: 19.13011512405212] tot_loss[loss=2.282, over 5484938.79 frames. , ppl: 9.799310150623512], batch size: 70
+2022-12-12 11:00:48,237 INFO [train.py:421] (5/8) Epoch 7, batch 15800, loss[loss=2.25, over 3500.00 frames. , ppl: 9.491278515056527] tot_loss[loss=2.282, over 5508149.66 frames. , ppl: 9.792582910977359], batch size: 70
+2022-12-12 11:02:25,848 INFO [train.py:421] (5/8) Epoch 7, batch 16000, loss[loss=2.369, over 1330.00 frames. , ppl: 10.682289448495524] tot_loss[loss=2.283, over 5441921.17 frames. , ppl: 9.809497103732921], batch size: 70
+2022-12-12 11:02:25,849 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 11:02:26,595 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.278, over 211138.00 frames. , ppl: 9.759092535215544
+2022-12-12 11:04:07,771 INFO [train.py:421] (5/8) Epoch 7, batch 16200, loss[loss=2.23, over 5250.00 frames. , ppl: 9.3033853329175] tot_loss[loss=2.283, over 5478884.24 frames. , ppl: 9.80145920847872], batch size: 70
+2022-12-12 11:05:46,866 INFO [train.py:421] (5/8) Epoch 7, batch 16400, loss[loss=2.835, over 630.00 frames. , ppl: 17.030573213479432] tot_loss[loss=2.283, over 5471131.57 frames. , ppl: 9.808993678321725], batch size: 70
+2022-12-12 11:07:27,129 INFO [train.py:421] (5/8) Epoch 7, batch 16600, loss[loss=2.252, over 4270.00 frames. , ppl: 9.50464553546375] tot_loss[loss=2.283, over 5487568.42 frames. , ppl: 9.809156228612512], batch size: 70
+2022-12-12 11:09:03,975 INFO [train.py:421] (5/8) Epoch 7, batch 16800, loss[loss=2.449, over 1960.00 frames. , ppl: 11.575717536546216] tot_loss[loss=2.284, over 5481420.03 frames. , ppl: 9.812919198326997], batch size: 70
+2022-12-12 11:10:44,190 INFO [train.py:421] (5/8) Epoch 7, batch 17000, loss[loss=2.329, over 1750.00 frames. , ppl: 10.271479703531647] tot_loss[loss=2.282, over 5513789.09 frames. , ppl: 9.80016762835776], batch size: 70
+2022-12-12 11:10:44,191 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 11:10:44,952 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.278, over 211138.00 frames. , ppl: 9.755632321970248
+2022-12-12 11:12:25,660 INFO [train.py:421] (5/8) Epoch 7, batch 17200, loss[loss=2.242, over 1120.00 frames. , ppl: 9.40987639428472] tot_loss[loss=2.283, over 5515307.66 frames. , ppl: 9.801604029094744], batch size: 70
+2022-12-12 11:14:04,100 INFO [train.py:421] (5/8) Epoch 7, batch 17400, loss[loss=2.242, over 2660.00 frames. , ppl: 9.41358027718999] tot_loss[loss=2.282, over 5529739.25 frames. , ppl: 9.793541905246537], batch size: 70
+2022-12-12 11:15:49,431 INFO [train.py:421] (5/8) Epoch 7, batch 17600, loss[loss=2.262, over 1960.00 frames. , ppl: 9.605588316770383] tot_loss[loss=2.282, over 5546354.02 frames. , ppl: 9.793469397842264], batch size: 70
+2022-12-12 11:17:26,212 INFO [train.py:421] (5/8) Epoch 7, batch 17800, loss[loss=2.305, over 2100.00 frames. , ppl: 10.019282499285783] tot_loss[loss=2.281, over 5544754.03 frames. , ppl: 9.788524588859346], batch size: 70
+2022-12-12 11:19:03,110 INFO [train.py:421] (5/8) Epoch 7, batch 18000, loss[loss=2.307, over 2590.00 frames. , ppl: 10.047829541383933] tot_loss[loss=2.28, over 5601857.34 frames. , ppl: 9.776465535426928], batch size: 70
+2022-12-12 11:19:03,111 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 11:19:03,871 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.278, over 211138.00 frames. , ppl: 9.756920371505483
+2022-12-12 11:20:45,119 INFO [train.py:421] (5/8) Epoch 7, batch 18200, loss[loss=3.476, over 420.00 frames. , ppl: 32.3142226724386] tot_loss[loss=2.28, over 5592097.69 frames. , ppl: 9.777999348593893], batch size: 70
+2022-12-12 11:22:26,377 INFO [train.py:421] (5/8) Epoch 7, batch 18400, loss[loss=3.298, over 490.00 frames. , ppl: 27.066222043747004] tot_loss[loss=2.28, over 5604120.11 frames. , ppl: 9.780364832974586], batch size: 70
+2022-12-12 11:24:05,729 INFO [train.py:421] (5/8) Epoch 7, batch 18600, loss[loss=2.658, over 700.00 frames. , ppl: 14.270476693427181] tot_loss[loss=2.282, over 5565566.29 frames. , ppl: 9.792379025869739], batch size: 70
+2022-12-12 11:25:43,136 INFO [train.py:421] (5/8) Epoch 7, batch 18800, loss[loss=2.769, over 700.00 frames. , ppl: 15.939309678945476] tot_loss[loss=2.283, over 5523682.29 frames. , ppl: 9.804239320689895], batch size: 70
+2022-12-12 11:27:23,641 INFO [train.py:421] (5/8) Epoch 7, batch 19000, loss[loss=2.401, over 1190.00 frames. , ppl: 11.029487819085068] tot_loss[loss=2.283, over 5530945.81 frames. , ppl: 9.807519942636954], batch size: 70
+2022-12-12 11:27:23,642 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 11:27:24,402 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.278, over 211138.00 frames. , ppl: 9.753232844662795
+2022-12-12 11:29:02,259 INFO [train.py:421] (5/8) Epoch 7, batch 19200, loss[loss=2.375, over 1120.00 frames. , ppl: 10.756334326677138] tot_loss[loss=2.284, over 5480729.04 frames. , ppl: 9.81627027598892], batch size: 70
+2022-12-12 11:30:43,927 INFO [train.py:421] (5/8) Epoch 7, batch 19400, loss[loss=2.235, over 4620.00 frames. , ppl: 9.348491681975203] tot_loss[loss=2.284, over 5474689.37 frames. , ppl: 9.820555787650497], batch size: 70
+2022-12-12 11:32:24,645 INFO [train.py:421] (5/8) Epoch 7, batch 19600, loss[loss=3.246, over 490.00 frames. , ppl: 25.676842736684655] tot_loss[loss=2.284, over 5477692.30 frames. , ppl: 9.818006151518706], batch size: 70
+2022-12-12 11:34:05,672 INFO [train.py:421] (5/8) Epoch 7, batch 19800, loss[loss=2.692, over 770.00 frames. , ppl: 14.756916695473146] tot_loss[loss=2.285, over 5438669.54 frames. , ppl: 9.824802501623326], batch size: 70
+2022-12-12 11:35:44,818 INFO [train.py:421] (5/8) Epoch 7, batch 20000, loss[loss=3.603, over 420.00 frames. , ppl: 36.69541561800524] tot_loss[loss=2.284, over 5471961.75 frames. , ppl: 9.820001402820825], batch size: 70
+2022-12-12 11:35:44,819 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 11:35:45,578 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.73576853706518
+2022-12-12 11:37:27,967 INFO [train.py:421] (5/8) Epoch 7, batch 20200, loss[loss=2.251, over 3220.00 frames. , ppl: 9.4998936231458] tot_loss[loss=2.284, over 5503367.00 frames. , ppl: 9.811970637900087], batch size: 70
+2022-12-12 11:39:05,108 INFO [train.py:421] (5/8) Epoch 7, batch 20400, loss[loss=2.239, over 3150.00 frames. , ppl: 9.386711462380102] tot_loss[loss=2.286, over 5433303.98 frames. , ppl: 9.832316017304287], batch size: 70
+2022-12-12 11:40:47,469 INFO [train.py:421] (5/8) Epoch 7, batch 20600, loss[loss=2.355, over 1540.00 frames. , ppl: 10.537233102681414] tot_loss[loss=2.286, over 5431055.26 frames. , ppl: 9.835009714179007], batch size: 70
+2022-12-12 11:42:27,079 INFO [train.py:421] (5/8) Epoch 7, batch 20800, loss[loss=2.418, over 1190.00 frames. , ppl: 11.222641018481694] tot_loss[loss=2.285, over 5452164.85 frames. , ppl: 9.82895993835946], batch size: 70
+2022-12-12 11:44:06,627 INFO [train.py:421] (5/8) Epoch 7, batch 21000, loss[loss=3.377, over 420.00 frames. , ppl: 29.277639838659695] tot_loss[loss=2.286, over 5427483.41 frames. , ppl: 9.831057214746542], batch size: 70
+2022-12-12 11:44:06,628 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 11:44:07,376 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.750929208941372
+2022-12-12 11:45:45,792 INFO [train.py:421] (5/8) Epoch 7, batch 21200, loss[loss=2.937, over 700.00 frames. , ppl: 18.85828665154568] tot_loss[loss=2.286, over 5423597.18 frames. , ppl: 9.835462137752305], batch size: 70
+2022-12-12 11:47:26,537 INFO [train.py:421] (5/8) Epoch 7, batch 21400, loss[loss=2.311, over 2100.00 frames. , ppl: 10.080169988989647] tot_loss[loss=2.284, over 5468686.12 frames. , ppl: 9.820272370706116], batch size: 70
+2022-12-12 11:49:07,973 INFO [train.py:421] (5/8) Epoch 7, batch 21600, loss[loss=2.405, over 1750.00 frames. , ppl: 11.075339633501718] tot_loss[loss=2.284, over 5449764.29 frames. , ppl: 9.819944035139642], batch size: 70
+2022-12-12 11:50:45,599 INFO [train.py:421] (5/8) Epoch 7, batch 21800, loss[loss=2.234, over 3010.00 frames. , ppl: 9.336594292836375] tot_loss[loss=2.284, over 5418316.74 frames. , ppl: 9.817556420303342], batch size: 70
+2022-12-12 11:52:26,454 INFO [train.py:421] (5/8) Epoch 7, batch 22000, loss[loss=2.215, over 5040.00 frames. , ppl: 9.159949355558428] tot_loss[loss=2.283, over 5449065.92 frames. , ppl: 9.810358105725877], batch size: 70
+2022-12-12 11:52:26,454 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 11:52:27,187 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.739122243408765
+2022-12-12 11:54:04,986 INFO [train.py:421] (5/8) Epoch 7, batch 22200, loss[loss=2.403, over 910.00 frames. , ppl: 11.053092236510201] tot_loss[loss=2.283, over 5463175.01 frames. , ppl: 9.810299715163996], batch size: 70
+2022-12-12 11:55:47,110 INFO [train.py:421] (5/8) Epoch 7, batch 22400, loss[loss=2.201, over 1750.00 frames. , ppl: 9.031119535261086] tot_loss[loss=2.284, over 5441035.84 frames. , ppl: 9.817510715639084], batch size: 70
+2022-12-12 11:57:29,545 INFO [train.py:421] (5/8) Epoch 7, batch 22600, loss[loss=2.314, over 2800.00 frames. , ppl: 10.11386365684625] tot_loss[loss=2.284, over 5448887.57 frames. , ppl: 9.81709338134787], batch size: 70
+2022-12-12 11:59:10,045 INFO [train.py:421] (5/8) Epoch 7, batch 22800, loss[loss=2.416, over 1260.00 frames. , ppl: 11.197325898145477] tot_loss[loss=2.285, over 5425590.37 frames. , ppl: 9.823656906334024], batch size: 70
+2022-12-12 12:00:50,387 INFO [train.py:421] (5/8) Epoch 7, batch 23000, loss[loss=2.385, over 1190.00 frames. , ppl: 10.855829648648127] tot_loss[loss=2.285, over 5419308.10 frames. , ppl: 9.823802509466917], batch size: 70
+2022-12-12 12:00:50,387 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 12:00:51,133 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.741689827028287
+2022-12-12 12:02:30,418 INFO [train.py:421] (5/8) Epoch 7, batch 23200, loss[loss=2.281, over 5600.00 frames. , ppl: 9.786403272235095] tot_loss[loss=2.283, over 5453370.43 frames. , ppl: 9.810236302469221], batch size: 70
+2022-12-12 12:04:11,331 INFO [train.py:421] (5/8) Epoch 7, batch 23400, loss[loss=2.663, over 770.00 frames. , ppl: 14.336701286929262] tot_loss[loss=2.283, over 5470478.36 frames. , ppl: 9.80619475987667], batch size: 70
+2022-12-12 12:05:50,808 INFO [train.py:421] (5/8) Epoch 7, batch 23600, loss[loss=2.485, over 910.00 frames. , ppl: 12.000136343428446] tot_loss[loss=2.283, over 5472674.57 frames. , ppl: 9.805887737819965], batch size: 70
+2022-12-12 12:07:34,132 INFO [train.py:421] (5/8) Epoch 7, batch 23800, loss[loss=2.4, over 2800.00 frames. , ppl: 11.025444915539559] tot_loss[loss=2.283, over 5478672.76 frames. , ppl: 9.80406090349807], batch size: 70
+2022-12-12 12:09:14,693 INFO [train.py:421] (5/8) Epoch 7, batch 24000, loss[loss=2.189, over 8050.00 frames. , ppl: 8.927717497297694] tot_loss[loss=2.282, over 5495137.85 frames. , ppl: 9.799091349923621], batch size: 70
+2022-12-12 12:09:14,694 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 12:09:15,425 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.74665824206101
+2022-12-12 12:10:54,504 INFO [train.py:421] (5/8) Epoch 7, batch 24200, loss[loss=2.375, over 1750.00 frames. , ppl: 10.749804365001584] tot_loss[loss=2.283, over 5475512.10 frames. , ppl: 9.805942210984725], batch size: 70
+2022-12-12 12:12:37,633 INFO [train.py:421] (5/8) Epoch 7, batch 24400, loss[loss=2.742, over 910.00 frames. , ppl: 15.52217004320567] tot_loss[loss=2.283, over 5467055.76 frames. , ppl: 9.810646853610942], batch size: 70
+2022-12-12 12:14:18,797 INFO [train.py:421] (5/8) Epoch 7, batch 24600, loss[loss=2.205, over 4130.00 frames. , ppl: 9.069124376385844] tot_loss[loss=2.283, over 5479630.74 frames. , ppl: 9.810204851235953], batch size: 70
+2022-12-12 12:15:58,739 INFO [train.py:421] (5/8) Epoch 7, batch 24800, loss[loss=2.516, over 1050.00 frames. , ppl: 12.382531598582165] tot_loss[loss=2.285, over 5446757.65 frames. , ppl: 9.823185016517025], batch size: 70
+2022-12-12 12:17:36,493 INFO [train.py:421] (5/8) Epoch 7, batch 25000, loss[loss=2.61, over 910.00 frames. , ppl: 13.597523700429452] tot_loss[loss=2.284, over 5446559.11 frames. , ppl: 9.820482374282541], batch size: 70
+2022-12-12 12:17:36,493 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 12:17:37,246 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.73495874762003
+2022-12-12 12:19:18,780 INFO [train.py:421] (5/8) Epoch 7, batch 25200, loss[loss=3.581, over 420.00 frames. , ppl: 35.917387758862326] tot_loss[loss=2.285, over 5430527.93 frames. , ppl: 9.8241076642745], batch size: 70
+2022-12-12 12:20:57,119 INFO [train.py:421] (5/8) Epoch 7, batch 25400, loss[loss=2.309, over 1540.00 frames. , ppl: 10.061838534481318] tot_loss[loss=2.284, over 5420533.14 frames. , ppl: 9.814959702876765], batch size: 70
+2022-12-12 12:22:35,818 INFO [train.py:421] (5/8) Epoch 7, batch 25600, loss[loss=2.229, over 5180.00 frames. , ppl: 9.292516502955728] tot_loss[loss=2.284, over 5450234.64 frames. , ppl: 9.81460042188652], batch size: 70
+2022-12-12 12:24:16,860 INFO [train.py:421] (5/8) Epoch 7, batch 25800, loss[loss=2.553, over 770.00 frames. , ppl: 12.845860720966932] tot_loss[loss=2.285, over 5407598.97 frames. , ppl: 9.8224240303674], batch size: 70
+2022-12-12 12:25:56,646 INFO [train.py:421] (5/8) Epoch 7, batch 26000, loss[loss=2.593, over 770.00 frames. , ppl: 13.36962782944662] tot_loss[loss=2.285, over 5407105.58 frames. , ppl: 9.824387374243594], batch size: 70
+2022-12-12 12:25:56,647 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 12:25:57,404 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.75006332120515
+2022-12-12 12:27:34,575 INFO [train.py:421] (5/8) Epoch 7, batch 26200, loss[loss=2.208, over 6930.00 frames. , ppl: 9.09345479732951] tot_loss[loss=2.285, over 5386227.46 frames. , ppl: 9.830162737162897], batch size: 70
+2022-12-12 12:29:16,203 INFO [train.py:421] (5/8) Epoch 7, batch 26400, loss[loss=2.318, over 2240.00 frames. , ppl: 10.158508567001654] tot_loss[loss=2.284, over 5435943.40 frames. , ppl: 9.813066973778737], batch size: 70
+2022-12-12 12:30:55,036 INFO [train.py:421] (5/8) Epoch 7, batch 26600, loss[loss=2.293, over 1820.00 frames. , ppl: 9.907045375628595] tot_loss[loss=2.284, over 5433422.59 frames. , ppl: 9.81808871213573], batch size: 70
+2022-12-12 12:32:37,071 INFO [train.py:421] (5/8) Epoch 7, batch 26800, loss[loss=2.169, over 4830.00 frames. , ppl: 8.746384497785353] tot_loss[loss=2.283, over 5487805.35 frames. , ppl: 9.803279054392748], batch size: 70
+2022-12-12 12:34:13,518 INFO [train.py:421] (5/8) Epoch 7, batch 27000, loss[loss=2.221, over 5110.00 frames. , ppl: 9.21406589305195] tot_loss[loss=2.283, over 5485918.92 frames. , ppl: 9.806771143608103], batch size: 70
+2022-12-12 12:34:13,518 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 12:34:14,248 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.741972432318025
+2022-12-12 12:35:56,493 INFO [train.py:421] (5/8) Epoch 7, batch 27200, loss[loss=2.28, over 3780.00 frames. , ppl: 9.77857595665337] tot_loss[loss=2.283, over 5500622.35 frames. , ppl: 9.80544506503927], batch size: 70
+2022-12-12 12:37:39,587 INFO [train.py:421] (5/8) Epoch 7, batch 27400, loss[loss=2.303, over 3220.00 frames. , ppl: 9.999357747517138] tot_loss[loss=2.283, over 5512275.16 frames. , ppl: 9.802030132473648], batch size: 70
+2022-12-12 12:39:19,442 INFO [train.py:421] (5/8) Epoch 7, batch 27600, loss[loss=2.364, over 3220.00 frames. , ppl: 10.632084678874005] tot_loss[loss=2.282, over 5529370.44 frames. , ppl: 9.796298661001895], batch size: 70
+2022-12-12 12:41:01,765 INFO [train.py:421] (5/8) Epoch 7, batch 27800, loss[loss=2.199, over 2310.00 frames. , ppl: 9.013890943463215] tot_loss[loss=2.282, over 5541775.37 frames. , ppl: 9.793232626671522], batch size: 70
+2022-12-12 12:42:41,588 INFO [train.py:421] (5/8) Epoch 7, batch 28000, loss[loss=2.494, over 1190.00 frames. , ppl: 12.113790982839747] tot_loss[loss=2.282, over 5543641.96 frames. , ppl: 9.794837427658239], batch size: 70
+2022-12-12 12:42:41,588 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 12:42:42,348 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.745110478587499
+2022-12-12 12:44:19,018 INFO [train.py:421] (5/8) Epoch 7, batch 28200, loss[loss=2.442, over 1400.00 frames. , ppl: 11.498672796380271] tot_loss[loss=2.282, over 5554520.53 frames. , ppl: 9.793846818541176], batch size: 70
+2022-12-12 12:45:59,452 INFO [train.py:421] (5/8) Epoch 7, batch 28400, loss[loss=2.352, over 2100.00 frames. , ppl: 10.507777033938357] tot_loss[loss=2.282, over 5584755.91 frames. , ppl: 9.791788461340776], batch size: 70
+2022-12-12 12:47:41,304 INFO [train.py:421] (5/8) Epoch 7, batch 28600, loss[loss=2.216, over 5110.00 frames. , ppl: 9.174427643222819] tot_loss[loss=2.281, over 5572078.56 frames. , ppl: 9.789935886875401], batch size: 70
+2022-12-12 12:49:21,363 INFO [train.py:421] (5/8) Epoch 7, batch 28800, loss[loss=2.157, over 4760.00 frames. , ppl: 8.64475586581616] tot_loss[loss=2.281, over 5586709.24 frames. , ppl: 9.788614308944526], batch size: 70
+2022-12-12 12:50:59,772 INFO [train.py:421] (5/8) Epoch 7, batch 29000, loss[loss=2.31, over 2240.00 frames. , ppl: 10.072836596617437] tot_loss[loss=2.281, over 5568757.16 frames. , ppl: 9.791329515511881], batch size: 70
+2022-12-12 12:50:59,773 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 12:51:00,533 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.278, over 211138.00 frames. , ppl: 9.757074891112701
+2022-12-12 12:52:40,873 INFO [train.py:421] (5/8) Epoch 7, batch 29200, loss[loss=2.325, over 3430.00 frames. , ppl: 10.230250403490585] tot_loss[loss=2.28, over 5592316.06 frames. , ppl: 9.781059250278254], batch size: 70
+2022-12-12 12:54:23,367 INFO [train.py:421] (5/8) Epoch 7, batch 29400, loss[loss=2.577, over 980.00 frames. , ppl: 13.157303988531782] tot_loss[loss=2.281, over 5572217.42 frames. , ppl: 9.786097612638955], batch size: 70
+2022-12-12 12:56:02,439 INFO [train.py:421] (5/8) Epoch 7, batch 29600, loss[loss=2.893, over 630.00 frames. , ppl: 18.05622243084527] tot_loss[loss=2.282, over 5577733.68 frames. , ppl: 9.791912860576186], batch size: 70
+2022-12-12 12:57:39,025 INFO [train.py:421] (5/8) Epoch 7, batch 29800, loss[loss=2.233, over 4760.00 frames. , ppl: 9.332129845983319] tot_loss[loss=2.282, over 5546333.53 frames. , ppl: 9.797756311041104], batch size: 70
+2022-12-12 12:59:14,357 INFO [train.py:421] (5/8) Epoch 7, batch 30000, loss[loss=2.587, over 840.00 frames. , ppl: 13.296291704984787] tot_loss[loss=2.284, over 5485820.46 frames. , ppl: 9.816326410862734], batch size: 70
+2022-12-12 12:59:14,357 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 12:59:15,118 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.278, over 211138.00 frames. , ppl: 9.757277069995297
+2022-12-12 13:00:54,969 INFO [train.py:421] (5/8) Epoch 7, batch 30200, loss[loss=2.468, over 1330.00 frames. , ppl: 11.795862564662023] tot_loss[loss=2.286, over 5420806.77 frames. , ppl: 9.83845233016712], batch size: 70
+2022-12-12 13:02:35,774 INFO [train.py:421] (5/8) Epoch 7, batch 30400, loss[loss=2.632, over 980.00 frames. , ppl: 13.89466706375881] tot_loss[loss=2.285, over 5440771.65 frames. , ppl: 9.82940483331389], batch size: 70
+2022-12-12 13:04:16,900 INFO [train.py:421] (5/8) Epoch 7, batch 30600, loss[loss=2.302, over 2520.00 frames. , ppl: 9.994844224022614] tot_loss[loss=2.286, over 5419640.10 frames. , ppl: 9.836756986650265], batch size: 70
+2022-12-12 13:05:54,568 INFO [train.py:421] (5/8) Epoch 7, batch 30800, loss[loss=2.404, over 1050.00 frames. , ppl: 11.068435150376986] tot_loss[loss=2.287, over 5407518.44 frames. , ppl: 9.840682740566779], batch size: 70
+2022-12-12 13:07:37,486 INFO [train.py:421] (5/8) Epoch 7, batch 31000, loss[loss=2.261, over 5460.00 frames. , ppl: 9.588389660271083] tot_loss[loss=2.287, over 5432349.45 frames. , ppl: 9.844692900487184], batch size: 70
+2022-12-12 13:07:37,486 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 13:07:38,249 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.74524606034524
+2022-12-12 13:09:21,970 INFO [train.py:421] (5/8) Epoch 7, batch 31200, loss[loss=2.274, over 4760.00 frames. , ppl: 9.721323195337682] tot_loss[loss=2.286, over 5443844.51 frames. , ppl: 9.839957207353546], batch size: 70
+2022-12-12 13:10:57,985 INFO [train.py:421] (5/8) Epoch 7, batch 31400, loss[loss=2.23, over 3150.00 frames. , ppl: 9.30426391208972] tot_loss[loss=2.287, over 5418148.71 frames. , ppl: 9.847039150974034], batch size: 70
+2022-12-12 13:12:37,920 INFO [train.py:421] (5/8) Epoch 7, batch 31600, loss[loss=2.502, over 1050.00 frames. , ppl: 12.210188672877916] tot_loss[loss=2.286, over 5478381.51 frames. , ppl: 9.830626426623462], batch size: 70
+2022-12-12 13:14:17,618 INFO [train.py:421] (5/8) Epoch 7, batch 31800, loss[loss=2.204, over 5530.00 frames. , ppl: 9.063172042313822] tot_loss[loss=2.285, over 5478094.43 frames. , ppl: 9.825117786828718], batch size: 70
+2022-12-12 13:15:58,458 INFO [train.py:421] (5/8) Epoch 7, batch 32000, loss[loss=2.36, over 3150.00 frames. , ppl: 10.587428944690018] tot_loss[loss=2.285, over 5486583.99 frames. , ppl: 9.82395752058953], batch size: 70
+2022-12-12 13:15:58,458 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 13:15:59,205 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.747510843049605
+2022-12-12 13:17:36,125 INFO [train.py:421] (5/8) Epoch 7, batch 32200, loss[loss=2.39, over 1610.00 frames. , ppl: 10.916873158419373] tot_loss[loss=2.285, over 5470008.87 frames. , ppl: 9.830371821964645], batch size: 70
+2022-12-12 13:19:19,500 INFO [train.py:421] (5/8) Epoch 7, batch 32400, loss[loss=2.601, over 840.00 frames. , ppl: 13.478428179512498] tot_loss[loss=2.286, over 5455048.52 frames. , ppl: 9.833999128234899], batch size: 70
+2022-12-12 13:20:59,435 INFO [train.py:421] (5/8) Epoch 7, batch 32600, loss[loss=2.163, over 5530.00 frames. , ppl: 8.697372347889544] tot_loss[loss=2.287, over 5425467.88 frames. , ppl: 9.845043383907727], batch size: 70
+2022-12-12 13:22:38,745 INFO [train.py:421] (5/8) Epoch 7, batch 32800, loss[loss=2.266, over 1820.00 frames. , ppl: 9.640011112130452] tot_loss[loss=2.287, over 5397543.38 frames. , ppl: 9.84704719297465], batch size: 70
+2022-12-12 13:24:19,353 INFO [train.py:421] (5/8) Epoch 7, batch 33000, loss[loss=2.557, over 1050.00 frames. , ppl: 12.893345293282888] tot_loss[loss=2.287, over 5406208.56 frames. , ppl: 9.841789128400556], batch size: 70
+2022-12-12 13:24:19,353 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 13:24:20,117 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.737587206525212
+2022-12-12 13:26:00,218 INFO [train.py:421] (5/8) Epoch 7, batch 33200, loss[loss=2.418, over 1190.00 frames. , ppl: 11.227564713158682] tot_loss[loss=2.286, over 5427906.07 frames. , ppl: 9.835701053654175], batch size: 70
+2022-12-12 13:27:39,638 INFO [train.py:421] (5/8) Epoch 7, batch 33400, loss[loss=2.21, over 3290.00 frames. , ppl: 9.11433491861171] tot_loss[loss=2.286, over 5397307.27 frames. , ppl: 9.833902422664801], batch size: 70
+2022-12-12 13:29:21,157 INFO [train.py:421] (5/8) Epoch 7, batch 33600, loss[loss=2.403, over 1470.00 frames. , ppl: 11.057198375666893] tot_loss[loss=2.284, over 5438192.73 frames. , ppl: 9.81725178701464], batch size: 70
+2022-12-12 13:31:03,138 INFO [train.py:421] (5/8) Epoch 7, batch 33800, loss[loss=2.355, over 1400.00 frames. , ppl: 10.538919103971997] tot_loss[loss=2.284, over 5460650.13 frames. , ppl: 9.81591183382218], batch size: 70
+2022-12-12 13:32:47,160 INFO [train.py:421] (5/8) Epoch 7, batch 34000, loss[loss=2.322, over 3080.00 frames. , ppl: 10.195437492306684] tot_loss[loss=2.284, over 5456876.66 frames. , ppl: 9.819966668399916], batch size: 70
+2022-12-12 13:32:47,160 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 13:32:47,895 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.749546711482623
+2022-12-12 13:34:29,526 INFO [train.py:421] (5/8) Epoch 7, batch 34200, loss[loss=2.189, over 2030.00 frames. , ppl: 8.926213318447756] tot_loss[loss=2.284, over 5466371.78 frames. , ppl: 9.819643920937835], batch size: 70
+2022-12-12 13:36:09,713 INFO [train.py:421] (5/8) Epoch 7, batch 34400, loss[loss=2.12, over 4620.00 frames. , ppl: 8.32755002180982] tot_loss[loss=2.283, over 5551701.93 frames. , ppl: 9.810896289576167], batch size: 70
+2022-12-12 13:37:49,610 INFO [train.py:421] (5/8) Epoch 7, batch 34600, loss[loss=2.403, over 1120.00 frames. , ppl: 11.061303496567724] tot_loss[loss=2.284, over 5538195.91 frames. , ppl: 9.81262416873133], batch size: 70
+2022-12-12 13:39:31,008 INFO [train.py:421] (5/8) Epoch 7, batch 34800, loss[loss=2.474, over 1680.00 frames. , ppl: 11.870323447445497] tot_loss[loss=2.284, over 5541834.93 frames. , ppl: 9.812827273152084], batch size: 70
+2022-12-12 13:41:13,650 INFO [train.py:421] (5/8) Epoch 7, batch 35000, loss[loss=2.294, over 1330.00 frames. , ppl: 9.917061301966116] tot_loss[loss=2.284, over 5533811.17 frames. , ppl: 9.816361209319645], batch size: 70
+2022-12-12 13:41:13,651 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 13:41:14,435 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.743404328984035
+2022-12-12 13:42:53,064 INFO [train.py:421] (5/8) Epoch 7, batch 35200, loss[loss=2.379, over 2730.00 frames. , ppl: 10.789036397957574] tot_loss[loss=2.284, over 5543446.43 frames. , ppl: 9.813778377490694], batch size: 70
+2022-12-12 13:44:36,901 INFO [train.py:421] (5/8) Epoch 7, batch 35400, loss[loss=2.407, over 1050.00 frames. , ppl: 11.101606679560925] tot_loss[loss=2.284, over 5550379.65 frames. , ppl: 9.813843541297107], batch size: 70
+2022-12-12 13:46:17,869 INFO [train.py:421] (5/8) Epoch 7, batch 35600, loss[loss=2.288, over 2450.00 frames. , ppl: 9.858478750884224] tot_loss[loss=2.285, over 5518683.11 frames. , ppl: 9.823684304929834], batch size: 70
+2022-12-12 13:48:00,244 INFO [train.py:421] (5/8) Epoch 7, batch 35800, loss[loss=2.941, over 630.00 frames. , ppl: 18.931076130794054] tot_loss[loss=2.284, over 5512260.59 frames. , ppl: 9.819582614320884], batch size: 70
+2022-12-12 13:49:39,585 INFO [train.py:421] (5/8) Epoch 7, batch 36000, loss[loss=2.24, over 4550.00 frames. , ppl: 9.388671261150574] tot_loss[loss=2.284, over 5499683.27 frames. , ppl: 9.819040364718791], batch size: 70
+2022-12-12 13:49:39,585 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 13:49:40,314 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.730916571608322
+2022-12-12 13:51:17,512 INFO [train.py:421] (5/8) Epoch 7, batch 36200, loss[loss=2.079, over 12530.00 frames. , ppl: 7.998570946647765] tot_loss[loss=2.285, over 5478969.15 frames. , ppl: 9.825584757819664], batch size: 70
+2022-12-12 13:52:55,997 INFO [train.py:421] (5/8) Epoch 7, batch 36400, loss[loss=2.389, over 1330.00 frames. , ppl: 10.900805829484543] tot_loss[loss=2.285, over 5467392.34 frames. , ppl: 9.826688088668153], batch size: 70
+2022-12-12 13:54:35,736 INFO [train.py:421] (5/8) Epoch 7, batch 36600, loss[loss=2.173, over 3710.00 frames. , ppl: 8.78112264336607] tot_loss[loss=2.285, over 5475718.27 frames. , ppl: 9.828319541534638], batch size: 70
+2022-12-12 13:56:13,084 INFO [train.py:421] (5/8) Epoch 7, batch 36800, loss[loss=2.404, over 2170.00 frames. , ppl: 11.068157705770574] tot_loss[loss=2.286, over 5435179.21 frames. , ppl: 9.838746593495237], batch size: 70
+2022-12-12 13:57:51,835 INFO [train.py:421] (5/8) Epoch 7, batch 37000, loss[loss=2.226, over 4060.00 frames. , ppl: 9.261302991792675] tot_loss[loss=2.288, over 5378964.52 frames. , ppl: 9.85275329722523], batch size: 70
+2022-12-12 13:57:51,835 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 13:57:52,594 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.730660210790171
+2022-12-12 13:59:32,625 INFO [train.py:421] (5/8) Epoch 7, batch 37200, loss[loss=3.473, over 490.00 frames. , ppl: 32.23547493637601] tot_loss[loss=2.287, over 5376643.90 frames. , ppl: 9.849355407134022], batch size: 70
+2022-12-12 14:01:13,835 INFO [train.py:421] (5/8) Epoch 7, batch 37400, loss[loss=2.249, over 2520.00 frames. , ppl: 9.474075586584402] tot_loss[loss=2.287, over 5364888.50 frames. , ppl: 9.844888647848462], batch size: 70
+2022-12-12 14:03:00,115 INFO [train.py:421] (5/8) Epoch 7, batch 37600, loss[loss=2.353, over 1120.00 frames. , ppl: 10.517998609243822] tot_loss[loss=2.286, over 5394192.99 frames. , ppl: 9.83783797851202], batch size: 70
+2022-12-12 14:04:41,204 INFO [train.py:421] (5/8) Epoch 7, batch 37800, loss[loss=2.474, over 1190.00 frames. , ppl: 11.867781653721751] tot_loss[loss=2.287, over 5379838.16 frames. , ppl: 9.840544588335918], batch size: 70
+2022-12-12 14:06:18,527 INFO [train.py:421] (5/8) Epoch 7, batch 38000, loss[loss=2.446, over 1050.00 frames. , ppl: 11.538675964929691] tot_loss[loss=2.286, over 5383827.46 frames. , ppl: 9.838361282873006], batch size: 70
+2022-12-12 14:06:18,527 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 14:06:19,290 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.735293029761026
+2022-12-12 14:08:00,975 INFO [train.py:421] (5/8) Epoch 7, batch 38200, loss[loss=2.465, over 840.00 frames. , ppl: 11.762515987680914] tot_loss[loss=2.285, over 5398805.05 frames. , ppl: 9.827831098512965], batch size: 70
+2022-12-12 14:09:41,653 INFO [train.py:421] (5/8) Epoch 7, batch 38400, loss[loss=2.236, over 5950.00 frames. , ppl: 9.356313957215514] tot_loss[loss=2.286, over 5393524.25 frames. , ppl: 9.83199248772319], batch size: 70
+2022-12-12 14:11:19,561 INFO [train.py:421] (5/8) Epoch 7, batch 38600, loss[loss=2.17, over 4970.00 frames. , ppl: 8.757080162316267] tot_loss[loss=2.286, over 5362470.13 frames. , ppl: 9.840293132971482], batch size: 70
+2022-12-12 14:12:58,079 INFO [train.py:421] (5/8) Epoch 7, batch 38800, loss[loss=2.257, over 3850.00 frames. , ppl: 9.557207252696665] tot_loss[loss=2.286, over 5378313.34 frames. , ppl: 9.834404767691366], batch size: 70
+2022-12-12 14:14:39,697 INFO [train.py:421] (5/8) Epoch 7, batch 39000, loss[loss=2.899, over 630.00 frames. , ppl: 18.163788168692214] tot_loss[loss=2.286, over 5376484.50 frames. , ppl: 9.83401219095929], batch size: 70
+2022-12-12 14:14:39,697 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 14:14:40,458 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.733431570068024
+2022-12-12 14:16:18,515 INFO [train.py:421] (5/8) Epoch 7, batch 39200, loss[loss=2.287, over 3570.00 frames. , ppl: 9.843671051821683] tot_loss[loss=2.287, over 5333882.69 frames. , ppl: 9.845261580256947], batch size: 70
+2022-12-12 14:17:57,370 INFO [train.py:421] (5/8) Epoch 7, batch 39400, loss[loss=2.197, over 9310.00 frames. , ppl: 8.998349323250627] tot_loss[loss=2.287, over 5330774.63 frames. , ppl: 9.848493220685045], batch size: 70
+2022-12-12 14:19:39,321 INFO [train.py:421] (5/8) Epoch 7, batch 39600, loss[loss=2.304, over 2800.00 frames. , ppl: 10.014194709802155] tot_loss[loss=2.287, over 5372155.32 frames. , ppl: 9.846281061097123], batch size: 70
+2022-12-12 14:21:22,520 INFO [train.py:421] (5/8) Epoch 7, batch 39800, loss[loss=2.215, over 4830.00 frames. , ppl: 9.156829848260795] tot_loss[loss=2.286, over 5402574.87 frames. , ppl: 9.838934089867848], batch size: 70
+2022-12-12 14:23:03,815 INFO [train.py:421] (5/8) Epoch 7, batch 40000, loss[loss=2.367, over 2170.00 frames. , ppl: 10.667655381411624] tot_loss[loss=2.286, over 5435320.81 frames. , ppl: 9.832935486392302], batch size: 70
+2022-12-12 14:23:03,816 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 14:23:04,577 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.732923044306935
+2022-12-12 14:24:40,622 INFO [train.py:421] (5/8) Epoch 7, batch 40200, loss[loss=2.243, over 3010.00 frames. , ppl: 9.425378880507322] tot_loss[loss=2.287, over 5400755.79 frames. , ppl: 9.84086709659922], batch size: 70
+2022-12-12 14:26:24,155 INFO [train.py:421] (5/8) Epoch 7, batch 40400, loss[loss=2.363, over 3360.00 frames. , ppl: 10.622417519740099] tot_loss[loss=2.288, over 5352136.38 frames. , ppl: 9.855395863974058], batch size: 70
+2022-12-12 14:28:05,233 INFO [train.py:421] (5/8) Epoch 7, batch 40600, loss[loss=2.898, over 560.00 frames. , ppl: 18.1431679322906] tot_loss[loss=2.287, over 5378799.27 frames. , ppl: 9.84988651568015], batch size: 70
+2022-12-12 14:29:48,173 INFO [train.py:421] (5/8) Epoch 7, batch 40800, loss[loss=2.22, over 3430.00 frames. , ppl: 9.205903865357296] tot_loss[loss=2.285, over 5448483.43 frames. , ppl: 9.824612805675514], batch size: 70
+2022-12-12 14:31:28,831 INFO [train.py:421] (5/8) Epoch 7, batch 41000, loss[loss=2.389, over 1330.00 frames. , ppl: 10.901188027119703] tot_loss[loss=2.284, over 5455146.24 frames. , ppl: 9.819957841626614], batch size: 70
+2022-12-12 14:31:28,832 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 14:31:29,592 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.72567260083667
+2022-12-12 14:33:15,370 INFO [train.py:421] (5/8) Epoch 7, batch 41200, loss[loss=2.202, over 2940.00 frames. , ppl: 9.038684752825285] tot_loss[loss=2.284, over 5486731.68 frames. , ppl: 9.812595379926924], batch size: 70
+2022-12-12 14:34:58,546 INFO [train.py:421] (5/8) Epoch 7, batch 41400, loss[loss=2.228, over 3920.00 frames. , ppl: 9.278997913214713] tot_loss[loss=2.283, over 5496851.87 frames. , ppl: 9.810032403747783], batch size: 70
+2022-12-12 14:36:43,117 INFO [train.py:421] (5/8) Epoch 7, batch 41600, loss[loss=2.409, over 1120.00 frames. , ppl: 11.121887976841109] tot_loss[loss=2.282, over 5537989.64 frames. , ppl: 9.7986590161443], batch size: 70
+2022-12-12 14:38:19,986 INFO [train.py:421] (5/8) Epoch 7, batch 41800, loss[loss=2.338, over 980.00 frames. , ppl: 10.36261914740477] tot_loss[loss=2.284, over 5499043.48 frames. , ppl: 9.81195838283097], batch size: 70
+2022-12-12 14:39:58,260 INFO [train.py:421] (5/8) Epoch 7, batch 42000, loss[loss=2.321, over 2240.00 frames. , ppl: 10.189463939527627] tot_loss[loss=2.283, over 5503143.90 frames. , ppl: 9.808239051434755], batch size: 70
+2022-12-12 14:39:58,261 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 14:39:59,022 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.732604689011115
+2022-12-12 14:41:41,115 INFO [train.py:421] (5/8) Epoch 7, batch 42200, loss[loss=2.25, over 2520.00 frames. , ppl: 9.484750798880869] tot_loss[loss=2.284, over 5481846.14 frames. , ppl: 9.810993980462158], batch size: 70
+2022-12-12 14:43:22,130 INFO [train.py:421] (5/8) Epoch 7, batch 42400, loss[loss=2.193, over 3220.00 frames. , ppl: 8.963552568120281] tot_loss[loss=2.284, over 5489946.48 frames. , ppl: 9.812781941896773], batch size: 70
+2022-12-12 14:45:04,978 INFO [train.py:421] (5/8) Epoch 7, batch 42600, loss[loss=2.359, over 2940.00 frames. , ppl: 10.575750473203714] tot_loss[loss=2.283, over 5506553.36 frames. , ppl: 9.803149477294594], batch size: 70
+2022-12-12 14:46:47,201 INFO [train.py:421] (5/8) Epoch 7, batch 42800, loss[loss=2.231, over 3290.00 frames. , ppl: 9.3065189487518] tot_loss[loss=2.283, over 5513466.72 frames. , ppl: 9.801898382863515], batch size: 70
+2022-12-12 14:48:26,742 INFO [train.py:421] (5/8) Epoch 7, batch 43000, loss[loss=2.206, over 2730.00 frames. , ppl: 9.077260592261535] tot_loss[loss=2.284, over 5462509.73 frames. , ppl: 9.810996206388385], batch size: 70
+2022-12-12 14:48:26,743 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 14:48:27,491 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.722768179823841
+2022-12-12 14:50:02,822 INFO [train.py:421] (5/8) Epoch 7, batch 43200, loss[loss=2.402, over 3290.00 frames. , ppl: 11.03997771128414] tot_loss[loss=2.283, over 5469823.82 frames. , ppl: 9.810401153106369], batch size: 70
+2022-12-12 14:51:39,672 INFO [train.py:421] (5/8) Epoch 7, batch 43400, loss[loss=2.186, over 1680.00 frames. , ppl: 8.89771371717304] tot_loss[loss=2.285, over 5407328.48 frames. , ppl: 9.82940514252847], batch size: 70
+2022-12-12 14:53:19,395 INFO [train.py:421] (5/8) Epoch 7, batch 43600, loss[loss=2.166, over 5320.00 frames. , ppl: 8.72678534985563] tot_loss[loss=2.285, over 5402403.91 frames. , ppl: 9.830143770757873], batch size: 70
+2022-12-12 14:54:58,930 INFO [train.py:421] (5/8) Epoch 7, batch 43800, loss[loss=2.257, over 2660.00 frames. , ppl: 9.554632450155943] tot_loss[loss=2.286, over 5376469.15 frames. , ppl: 9.836581679065292], batch size: 70
+2022-12-12 14:56:40,138 INFO [train.py:421] (5/8) Epoch 7, batch 44000, loss[loss=2.266, over 3010.00 frames. , ppl: 9.6383857396738] tot_loss[loss=2.288, over 5341569.87 frames. , ppl: 9.851032899144077], batch size: 70
+2022-12-12 14:56:40,139 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 14:56:40,926 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.738247314799137
+2022-12-12 14:58:24,199 INFO [train.py:421] (5/8) Epoch 7, batch 44200, loss[loss=2.46, over 1400.00 frames. , ppl: 11.708865980121947] tot_loss[loss=2.288, over 5348461.14 frames. , ppl: 9.853363857803751], batch size: 70
+2022-12-12 15:00:05,404 INFO [train.py:421] (5/8) Epoch 7, batch 44400, loss[loss=2.372, over 1470.00 frames. , ppl: 10.719426866458114] tot_loss[loss=2.285, over 5419128.38 frames. , ppl: 9.826992828277996], batch size: 70
+2022-12-12 15:01:46,541 INFO [train.py:421] (5/8) Epoch 7, batch 44600, loss[loss=2.219, over 4760.00 frames. , ppl: 9.19721943511719] tot_loss[loss=2.284, over 5442354.39 frames. , ppl: 9.814499823020917], batch size: 70
+2022-12-12 15:03:25,945 INFO [train.py:421] (5/8) Epoch 7, batch 44800, loss[loss=2.613, over 980.00 frames. , ppl: 13.636290718147034] tot_loss[loss=2.284, over 5435943.25 frames. , ppl: 9.812825032068515], batch size: 70
+2022-12-12 15:05:07,366 INFO [train.py:421] (5/8) Epoch 7, batch 45000, loss[loss=2.272, over 2800.00 frames. , ppl: 9.69993795791186] tot_loss[loss=2.283, over 5454235.56 frames. , ppl: 9.808336033385757], batch size: 70
+2022-12-12 15:05:07,367 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 15:05:08,112 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.739091972737263
+2022-12-12 15:06:49,436 INFO [train.py:421] (5/8) Epoch 7, batch 45200, loss[loss=2.295, over 1680.00 frames. , ppl: 9.925616409165183] tot_loss[loss=2.284, over 5445221.97 frames. , ppl: 9.814175885161715], batch size: 70
+2022-12-12 15:08:26,833 INFO [train.py:421] (5/8) Epoch 7, batch 45400, loss[loss=2.295, over 2310.00 frames. , ppl: 9.922576277833732] tot_loss[loss=2.284, over 5422936.35 frames. , ppl: 9.816381128084643], batch size: 70
+2022-12-12 15:10:04,385 INFO [train.py:421] (5/8) Epoch 7, batch 45600, loss[loss=2.173, over 7490.00 frames. , ppl: 8.785613783425925] tot_loss[loss=2.285, over 5389391.81 frames. , ppl: 9.82391881289461], batch size: 70
+2022-12-12 15:11:43,470 INFO [train.py:421] (5/8) Epoch 7, batch 45800, loss[loss=2.397, over 1050.00 frames. , ppl: 10.987060750054116] tot_loss[loss=2.285, over 5390437.19 frames. , ppl: 9.822549606112867], batch size: 70
+2022-12-12 15:13:23,822 INFO [train.py:421] (5/8) Epoch 7, batch 46000, loss[loss=2.114, over 4410.00 frames. , ppl: 8.280295883845746] tot_loss[loss=2.284, over 5397200.95 frames. , ppl: 9.818653596439983], batch size: 70
+2022-12-12 15:13:23,823 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 15:13:24,592 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.725171677432357
+2022-12-12 15:15:04,859 INFO [train.py:421] (5/8) Epoch 7, batch 46200, loss[loss=2.233, over 5600.00 frames. , ppl: 9.328723087227466] tot_loss[loss=2.284, over 5406316.78 frames. , ppl: 9.82006731801561], batch size: 70
+2022-12-12 15:16:41,468 INFO [train.py:421] (5/8) Epoch 7, batch 46400, loss[loss=2.193, over 4690.00 frames. , ppl: 8.960577292257264] tot_loss[loss=2.285, over 5387563.86 frames. , ppl: 9.82290583828942], batch size: 70
+2022-12-12 15:18:27,480 INFO [train.py:421] (5/8) Epoch 7, batch 46600, loss[loss=2.593, over 1120.00 frames. , ppl: 13.369671279790298] tot_loss[loss=2.284, over 5432150.96 frames. , ppl: 9.815832488224993], batch size: 70
+2022-12-12 15:20:06,523 INFO [train.py:421] (5/8) Epoch 7, batch 46800, loss[loss=2.158, over 4200.00 frames. , ppl: 8.64973627331796] tot_loss[loss=2.285, over 5394061.86 frames. , ppl: 9.8249849513788], batch size: 70
+2022-12-12 15:21:46,495 INFO [train.py:421] (5/8) Epoch 7, batch 47000, loss[loss=2.381, over 1260.00 frames. , ppl: 10.814307565047404] tot_loss[loss=2.286, over 5379581.57 frames. , ppl: 9.83120021127917], batch size: 70
+2022-12-12 15:21:46,495 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 15:21:47,256 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.277, over 211138.00 frames. , ppl: 9.746437530099742
+2022-12-12 15:23:25,417 INFO [train.py:421] (5/8) Epoch 7, batch 47200, loss[loss=2.317, over 1750.00 frames. , ppl: 10.143329189785568] tot_loss[loss=2.286, over 5380810.36 frames. , ppl: 9.834026374234346], batch size: 70
+2022-12-12 15:25:07,548 INFO [train.py:421] (5/8) Epoch 7, batch 47400, loss[loss=2.251, over 2100.00 frames. , ppl: 9.500340689064311] tot_loss[loss=2.285, over 5427395.37 frames. , ppl: 9.825920752998005], batch size: 70
+2022-12-12 15:26:45,411 INFO [train.py:421] (5/8) Epoch 7, batch 47600, loss[loss=2.177, over 7070.00 frames. , ppl: 8.820303733725419] tot_loss[loss=2.284, over 5463516.97 frames. , ppl: 9.814385601855824], batch size: 70
+2022-12-12 15:28:25,600 INFO [train.py:421] (5/8) Epoch 7, batch 47800, loss[loss=2.428, over 1750.00 frames. , ppl: 11.334611954590995] tot_loss[loss=2.285, over 5437702.59 frames. , ppl: 9.827085149889054], batch size: 70
+2022-12-12 15:30:08,467 INFO [train.py:421] (5/8) Epoch 7, batch 48000, loss[loss=2.497, over 1190.00 frames. , ppl: 12.140133597532355] tot_loss[loss=2.285, over 5421339.79 frames. , ppl: 9.824666388229149], batch size: 70
+2022-12-12 15:30:08,468 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 15:30:09,214 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.730277122379839
+2022-12-12 15:31:50,593 INFO [train.py:421] (5/8) Epoch 7, batch 48200, loss[loss=2.229, over 4060.00 frames. , ppl: 9.287490277661039] tot_loss[loss=2.284, over 5432064.55 frames. , ppl: 9.820319491423696], batch size: 70
+2022-12-12 15:33:31,840 INFO [train.py:421] (5/8) Epoch 7, batch 48400, loss[loss=2.611, over 980.00 frames. , ppl: 13.608639100349405] tot_loss[loss=2.282, over 5513416.88 frames. , ppl: 9.797574955347411], batch size: 70
+2022-12-12 15:35:15,561 INFO [train.py:421] (5/8) Epoch 7, batch 48600, loss[loss=2.456, over 1680.00 frames. , ppl: 11.655755399220464] tot_loss[loss=2.282, over 5509144.59 frames. , ppl: 9.797543376637849], batch size: 70
+2022-12-12 15:37:00,085 INFO [train.py:421] (5/8) Epoch 7, batch 48800, loss[loss=4.829, over 280.00 frames. , ppl: 125.05761966932047] tot_loss[loss=2.282, over 5538875.19 frames. , ppl: 9.791865772886013], batch size: 70
+2022-12-12 15:38:39,954 INFO [train.py:421] (5/8) Epoch 7, batch 49000, loss[loss=2.188, over 13790.00 frames. , ppl: 8.921445709548516] tot_loss[loss=2.281, over 5552929.22 frames. , ppl: 9.785625695480528], batch size: 70
+2022-12-12 15:38:39,955 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 15:38:40,684 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.71475889319782
+2022-12-12 15:40:19,125 INFO [train.py:421] (5/8) Epoch 7, batch 49200, loss[loss=2.375, over 1820.00 frames. , ppl: 10.755386750992963] tot_loss[loss=2.283, over 5491014.30 frames. , ppl: 9.80647567711117], batch size: 70
+2022-12-12 15:42:00,517 INFO [train.py:421] (5/8) Epoch 7, batch 49400, loss[loss=2.288, over 1540.00 frames. , ppl: 9.859194090618159] tot_loss[loss=2.282, over 5498809.30 frames. , ppl: 9.80038331504405], batch size: 70
+2022-12-12 15:43:37,682 INFO [train.py:421] (5/8) Epoch 7, batch 49600, loss[loss=2.208, over 2660.00 frames. , ppl: 9.100475490195215] tot_loss[loss=2.281, over 5534185.27 frames. , ppl: 9.785115083546017], batch size: 70
+2022-12-12 15:45:24,852 INFO [train.py:421] (5/8) Epoch 7, batch 49800, loss[loss=2.434, over 1680.00 frames. , ppl: 11.400197293947889] tot_loss[loss=2.281, over 5509092.38 frames. , ppl: 9.785762133777093], batch size: 70
+2022-12-12 15:47:06,666 INFO [train.py:421] (5/8) Epoch 7, batch 50000, loss[loss=2.25, over 4690.00 frames. , ppl: 9.492181883267676] tot_loss[loss=2.281, over 5528938.34 frames. , ppl: 9.783376108738533], batch size: 70
+2022-12-12 15:47:06,666 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 15:47:07,406 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.714458385768214
+2022-12-12 15:48:44,227 INFO [train.py:421] (5/8) Epoch 7, batch 50200, loss[loss=2.458, over 1470.00 frames. , ppl: 11.67703970206561] tot_loss[loss=2.282, over 5483975.60 frames. , ppl: 9.800111295162338], batch size: 70
+2022-12-12 15:50:22,977 INFO [train.py:421] (5/8) Epoch 7, batch 50400, loss[loss=2.32, over 1120.00 frames. , ppl: 10.175487098385688] tot_loss[loss=2.283, over 5449489.43 frames. , ppl: 9.80675285861184], batch size: 70
+2022-12-12 15:52:02,527 INFO [train.py:421] (5/8) Epoch 7, batch 50600, loss[loss=2.26, over 4690.00 frames. , ppl: 9.586874014606446] tot_loss[loss=2.284, over 5408919.47 frames. , ppl: 9.818991023019255], batch size: 70
+2022-12-12 15:53:41,811 INFO [train.py:421] (5/8) Epoch 7, batch 50800, loss[loss=3.629, over 420.00 frames. , ppl: 37.68489616590935] tot_loss[loss=2.285, over 5393524.24 frames. , ppl: 9.82898850229894], batch size: 70
+2022-12-12 15:55:22,387 INFO [train.py:421] (5/8) Epoch 7, batch 51000, loss[loss=2.465, over 1260.00 frames. , ppl: 11.760684846591644] tot_loss[loss=2.286, over 5367508.47 frames. , ppl: 9.834649463432289], batch size: 70
+2022-12-12 15:55:22,387 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 15:55:23,149 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.273, over 211138.00 frames. , ppl: 9.712544846014454
+2022-12-12 15:57:03,533 INFO [train.py:421] (5/8) Epoch 7, batch 51200, loss[loss=2.458, over 1190.00 frames. , ppl: 11.68554592177857] tot_loss[loss=2.285, over 5395257.46 frames. , ppl: 9.824087207407622], batch size: 70
+2022-12-12 15:58:38,215 INFO [train.py:421] (5/8) Epoch 7, batch 51400, loss[loss=2.565, over 770.00 frames. , ppl: 12.996142197133834] tot_loss[loss=2.285, over 5399431.31 frames. , ppl: 9.823753239750099], batch size: 70
+2022-12-12 16:00:15,721 INFO [train.py:421] (5/8) Epoch 7, batch 51600, loss[loss=2.155, over 3150.00 frames. , ppl: 8.625679725805268] tot_loss[loss=2.283, over 5457950.21 frames. , ppl: 9.805851023633583], batch size: 70
+2022-12-12 16:01:57,240 INFO [train.py:421] (5/8) Epoch 7, batch 51800, loss[loss=2.499, over 980.00 frames. , ppl: 12.16592232709699] tot_loss[loss=2.283, over 5480168.91 frames. , ppl: 9.802122732472222], batch size: 70
+2022-12-12 16:03:36,215 INFO [train.py:421] (5/8) Epoch 7, batch 52000, loss[loss=2.365, over 1050.00 frames. , ppl: 10.643063752591068] tot_loss[loss=2.281, over 5526815.02 frames. , ppl: 9.789337663804018], batch size: 70
+2022-12-12 16:03:36,215 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 16:03:36,974 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.714971698339436
+2022-12-12 16:05:14,234 INFO [train.py:421] (5/8) Epoch 7, batch 52200, loss[loss=2.591, over 840.00 frames. , ppl: 13.348824638543874] tot_loss[loss=2.283, over 5501695.44 frames. , ppl: 9.801605662035355], batch size: 70
+2022-12-12 16:06:54,094 INFO [train.py:421] (5/8) Epoch 7, batch 52400, loss[loss=2.228, over 2590.00 frames. , ppl: 9.279099812156753] tot_loss[loss=2.282, over 5536960.83 frames. , ppl: 9.79314791948731], batch size: 70
+2022-12-12 16:08:37,695 INFO [train.py:421] (5/8) Epoch 7, batch 52600, loss[loss=2.443, over 1470.00 frames. , ppl: 11.508990596613048] tot_loss[loss=2.282, over 5521290.83 frames. , ppl: 9.797332310261552], batch size: 70
+2022-12-12 16:10:20,600 INFO [train.py:421] (5/8) Epoch 7, batch 52800, loss[loss=2.307, over 2380.00 frames. , ppl: 10.04415459991843] tot_loss[loss=2.282, over 5537724.94 frames. , ppl: 9.79672906631821], batch size: 70
+2022-12-12 16:12:02,791 INFO [train.py:421] (5/8) Epoch 7, batch 53000, loss[loss=2.543, over 1050.00 frames. , ppl: 12.716317034862122] tot_loss[loss=2.28, over 5598048.47 frames. , ppl: 9.78133896892234], batch size: 70
+2022-12-12 16:12:02,791 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 16:12:03,558 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.273, over 211138.00 frames. , ppl: 9.710289624670212
+2022-12-12 16:13:44,896 INFO [train.py:421] (5/8) Epoch 7, batch 53200, loss[loss=2.257, over 2450.00 frames. , ppl: 9.553234451412568] tot_loss[loss=2.28, over 5611035.34 frames. , ppl: 9.77930931845111], batch size: 70
+2022-12-12 16:15:28,420 INFO [train.py:421] (5/8) Epoch 7, batch 53400, loss[loss=2.506, over 1330.00 frames. , ppl: 12.252317385150763] tot_loss[loss=2.281, over 5562469.35 frames. , ppl: 9.787269494871504], batch size: 70
+2022-12-12 16:17:11,388 INFO [train.py:421] (5/8) Epoch 7, batch 53600, loss[loss=2.274, over 2310.00 frames. , ppl: 9.716057517593816] tot_loss[loss=2.282, over 5525537.34 frames. , ppl: 9.80065649203515], batch size: 70
+2022-12-12 16:18:49,868 INFO [train.py:421] (5/8) Epoch 7, batch 53800, loss[loss=2.252, over 2240.00 frames. , ppl: 9.511022921100045] tot_loss[loss=2.283, over 5500684.13 frames. , ppl: 9.808380418438695], batch size: 70
+2022-12-12 16:20:30,709 INFO [train.py:421] (5/8) Epoch 7, batch 54000, loss[loss=2.428, over 1400.00 frames. , ppl: 11.335105324554025] tot_loss[loss=2.283, over 5511350.15 frames. , ppl: 9.80322633194271], batch size: 70
+2022-12-12 16:20:30,710 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 16:20:31,473 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.730233917866439
+2022-12-12 16:22:14,304 INFO [train.py:421] (5/8) Epoch 7, batch 54200, loss[loss=2.418, over 1680.00 frames. , ppl: 11.224716091716175] tot_loss[loss=2.283, over 5516881.11 frames. , ppl: 9.80346713105479], batch size: 70
+2022-12-12 16:23:57,484 INFO [train.py:421] (5/8) Epoch 7, batch 54400, loss[loss=2.477, over 1050.00 frames. , ppl: 11.904797846686742] tot_loss[loss=2.283, over 5510923.72 frames. , ppl: 9.80721271534327], batch size: 70
+2022-12-12 16:25:40,588 INFO [train.py:421] (5/8) Epoch 7, batch 54600, loss[loss=2.37, over 2240.00 frames. , ppl: 10.695801150429583] tot_loss[loss=2.284, over 5510893.15 frames. , ppl: 9.811140170036564], batch size: 70
+2022-12-12 16:27:23,732 INFO [train.py:421] (5/8) Epoch 7, batch 54800, loss[loss=2.267, over 2100.00 frames. , ppl: 9.648091633897229] tot_loss[loss=2.284, over 5473074.48 frames. , ppl: 9.812141229437746], batch size: 70
+2022-12-12 16:29:03,522 INFO [train.py:421] (5/8) Epoch 7, batch 55000, loss[loss=2.269, over 3710.00 frames. , ppl: 9.665204907577953] tot_loss[loss=2.283, over 5498708.87 frames. , ppl: 9.805018625906033], batch size: 70
+2022-12-12 16:29:03,523 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 16:29:04,267 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.276, over 211138.00 frames. , ppl: 9.738142098037239
+2022-12-12 16:30:45,226 INFO [train.py:421] (5/8) Epoch 7, batch 55200, loss[loss=2.77, over 840.00 frames. , ppl: 15.961784637042816] tot_loss[loss=2.282, over 5520000.25 frames. , ppl: 9.800910121132107], batch size: 70
+2022-12-12 16:32:27,222 INFO [train.py:421] (5/8) Epoch 7, batch 55400, loss[loss=2.198, over 2870.00 frames. , ppl: 9.005430136569762] tot_loss[loss=2.282, over 5520882.02 frames. , ppl: 9.798637868322285], batch size: 70
+2022-12-12 16:34:09,958 INFO [train.py:421] (5/8) Epoch 7, batch 55600, loss[loss=2.751, over 700.00 frames. , ppl: 15.665213012822354] tot_loss[loss=2.281, over 5559376.90 frames. , ppl: 9.783853381170884], batch size: 70
+2022-12-12 16:35:52,341 INFO [train.py:421] (5/8) Epoch 7, batch 55800, loss[loss=2.351, over 1540.00 frames. , ppl: 10.495270448906691] tot_loss[loss=2.281, over 5522922.93 frames. , ppl: 9.789146143156266], batch size: 70
+2022-12-12 16:37:34,622 INFO [train.py:421] (5/8) Epoch 7, batch 56000, loss[loss=2.655, over 910.00 frames. , ppl: 14.227154686232712] tot_loss[loss=2.281, over 5519612.33 frames. , ppl: 9.790108606951218], batch size: 70
+2022-12-12 16:37:34,623 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 16:37:35,393 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.72342008782373
+2022-12-12 16:39:17,250 INFO [train.py:421] (5/8) Epoch 7, batch 56200, loss[loss=2.313, over 3220.00 frames. , ppl: 10.103506967387915] tot_loss[loss=2.283, over 5460654.41 frames. , ppl: 9.80580720950946], batch size: 70
+2022-12-12 16:40:58,446 INFO [train.py:421] (5/8) Epoch 7, batch 56400, loss[loss=2.239, over 3920.00 frames. , ppl: 9.385637397393003] tot_loss[loss=2.283, over 5447069.97 frames. , ppl: 9.809963937199742], batch size: 70
+2022-12-12 16:42:37,536 INFO [train.py:421] (5/8) Epoch 7, batch 56600, loss[loss=2.141, over 3640.00 frames. , ppl: 8.505739475248875] tot_loss[loss=2.284, over 5444491.26 frames. , ppl: 9.816159498371098], batch size: 70
+2022-12-12 16:44:14,637 INFO [train.py:421] (5/8) Epoch 7, batch 56800, loss[loss=2.505, over 1330.00 frames. , ppl: 12.241557971976786] tot_loss[loss=2.284, over 5449705.84 frames. , ppl: 9.810967262143102], batch size: 70
+2022-12-12 16:45:52,831 INFO [train.py:421] (5/8) Epoch 7, batch 57000, loss[loss=2.23, over 9520.00 frames. , ppl: 9.3038561027223] tot_loss[loss=2.283, over 5463055.37 frames. , ppl: 9.80268224022948], batch size: 70
+2022-12-12 16:45:52,832 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 16:45:53,554 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.720317799677936
+2022-12-12 16:47:33,744 INFO [train.py:421] (5/8) Epoch 7, batch 57200, loss[loss=2.279, over 2310.00 frames. , ppl: 9.767165608351943] tot_loss[loss=2.282, over 5481233.11 frames. , ppl: 9.795080691238383], batch size: 70
+2022-12-12 16:49:17,328 INFO [train.py:421] (5/8) Epoch 7, batch 57400, loss[loss=2.368, over 1960.00 frames. , ppl: 10.673424873311095] tot_loss[loss=2.283, over 5434399.83 frames. , ppl: 9.810737737490086], batch size: 70
+2022-12-12 16:50:58,689 INFO [train.py:421] (5/8) Epoch 7, batch 57600, loss[loss=2.235, over 2940.00 frames. , ppl: 9.34466709918679] tot_loss[loss=2.284, over 5423071.81 frames. , ppl: 9.814508475157666], batch size: 70
+2022-12-12 16:52:38,236 INFO [train.py:421] (5/8) Epoch 7, batch 57800, loss[loss=2.419, over 980.00 frames. , ppl: 11.236238244840758] tot_loss[loss=2.284, over 5410164.69 frames. , ppl: 9.820685126591773], batch size: 70
+2022-12-12 16:54:19,003 INFO [train.py:421] (5/8) Epoch 7, batch 58000, loss[loss=2.473, over 1610.00 frames. , ppl: 11.86105547722484] tot_loss[loss=2.284, over 5415582.75 frames. , ppl: 9.817504044419689], batch size: 70
+2022-12-12 16:54:19,004 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 16:54:19,765 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.714061557664111
+2022-12-12 16:56:00,041 INFO [train.py:421] (5/8) Epoch 7, batch 58200, loss[loss=2.547, over 770.00 frames. , ppl: 12.774631043858662] tot_loss[loss=2.285, over 5399783.55 frames. , ppl: 9.823144742978613], batch size: 70
+2022-12-12 16:57:41,618 INFO [train.py:421] (5/8) Epoch 7, batch 58400, loss[loss=3.526, over 420.00 frames. , ppl: 33.981860958915455] tot_loss[loss=2.284, over 5408855.96 frames. , ppl: 9.81811447854312], batch size: 70
+2022-12-12 16:59:24,618 INFO [train.py:421] (5/8) Epoch 7, batch 58600, loss[loss=2.236, over 4830.00 frames. , ppl: 9.353682249848834] tot_loss[loss=2.284, over 5429024.06 frames. , ppl: 9.811682811246724], batch size: 70
+2022-12-12 17:01:06,528 INFO [train.py:421] (5/8) Epoch 7, batch 58800, loss[loss=2.36, over 1680.00 frames. , ppl: 10.588567968580428] tot_loss[loss=2.283, over 5451965.45 frames. , ppl: 9.80745796313332], batch size: 70
+2022-12-12 17:02:51,778 INFO [train.py:421] (5/8) Epoch 7, batch 59000, loss[loss=2.401, over 1120.00 frames. , ppl: 11.036479435542176] tot_loss[loss=2.283, over 5467045.39 frames. , ppl: 9.805761272173815], batch size: 70
+2022-12-12 17:02:51,779 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 17:02:52,548 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.716621096251133
+2022-12-12 17:04:37,261 INFO [train.py:421] (5/8) Epoch 7, batch 59200, loss[loss=2.368, over 2450.00 frames. , ppl: 10.672060363509518] tot_loss[loss=2.283, over 5468660.87 frames. , ppl: 9.810797615752566], batch size: 70
+2022-12-12 17:06:15,608 INFO [train.py:421] (5/8) Epoch 7, batch 59400, loss[loss=2.339, over 2800.00 frames. , ppl: 10.367767297242972] tot_loss[loss=2.285, over 5439529.63 frames. , ppl: 9.821792656609826], batch size: 70
+2022-12-12 17:07:59,264 INFO [train.py:421] (5/8) Epoch 7, batch 59600, loss[loss=2.304, over 2590.00 frames. , ppl: 10.010638194686148] tot_loss[loss=2.285, over 5449120.36 frames. , ppl: 9.826927256448403], batch size: 70
+2022-12-12 17:09:44,679 INFO [train.py:421] (5/8) Epoch 7, batch 59800, loss[loss=2.914, over 560.00 frames. , ppl: 18.431645278975132] tot_loss[loss=2.285, over 5452405.53 frames. , ppl: 9.825098880482567], batch size: 70
+2022-12-12 17:11:28,849 INFO [train.py:421] (5/8) Epoch 7, batch 60000, loss[loss=2.247, over 9170.00 frames. , ppl: 9.463021367349853] tot_loss[loss=2.283, over 5503828.53 frames. , ppl: 9.808548253690054], batch size: 70
+2022-12-12 17:11:28,849 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 17:11:29,585 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.71337577339601
+2022-12-12 17:13:08,943 INFO [train.py:421] (5/8) Epoch 7, batch 60200, loss[loss=2.483, over 980.00 frames. , ppl: 11.976597054914704] tot_loss[loss=2.284, over 5493427.15 frames. , ppl: 9.813261832622283], batch size: 70
+2022-12-12 17:14:49,931 INFO [train.py:421] (5/8) Epoch 7, batch 60400, loss[loss=2.566, over 910.00 frames. , ppl: 13.015747403113442] tot_loss[loss=2.284, over 5472942.07 frames. , ppl: 9.818266028844102], batch size: 70
+2022-12-12 17:16:33,754 INFO [train.py:421] (5/8) Epoch 7, batch 60600, loss[loss=2.74, over 700.00 frames. , ppl: 15.482858952242362] tot_loss[loss=2.285, over 5433532.19 frames. , ppl: 9.827674124264991], batch size: 70
+2022-12-12 17:18:13,221 INFO [train.py:421] (5/8) Epoch 7, batch 60800, loss[loss=2.228, over 4550.00 frames. , ppl: 9.277015792671246] tot_loss[loss=2.285, over 5440035.16 frames. , ppl: 9.822346996438384], batch size: 70
+2022-12-12 17:19:56,253 INFO [train.py:421] (5/8) Epoch 7, batch 61000, loss[loss=2.456, over 1190.00 frames. , ppl: 11.65602083620777] tot_loss[loss=2.283, over 5482643.54 frames. , ppl: 9.807748849458086], batch size: 70
+2022-12-12 17:19:56,254 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 17:19:57,004 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.273, over 211138.00 frames. , ppl: 9.70608961125915
+2022-12-12 17:21:39,553 INFO [train.py:421] (5/8) Epoch 7, batch 61200, loss[loss=2.364, over 1050.00 frames. , ppl: 10.629496364084698] tot_loss[loss=2.282, over 5506265.12 frames. , ppl: 9.795831776033818], batch size: 70
+2022-12-12 17:23:24,509 INFO [train.py:421] (5/8) Epoch 7, batch 61400, loss[loss=2.216, over 4620.00 frames. , ppl: 9.166954637825992] tot_loss[loss=2.281, over 5558146.61 frames. , ppl: 9.782726661009981], batch size: 70
+2022-12-12 17:25:04,988 INFO [train.py:421] (5/8) Epoch 7, batch 61600, loss[loss=2.248, over 3010.00 frames. , ppl: 9.472369630908078] tot_loss[loss=2.282, over 5516051.12 frames. , ppl: 9.798308116501659], batch size: 70
+2022-12-12 17:26:46,420 INFO [train.py:421] (5/8) Epoch 7, batch 61800, loss[loss=2.814, over 630.00 frames. , ppl: 16.67148503012588] tot_loss[loss=2.283, over 5506921.89 frames. , ppl: 9.803729818904227], batch size: 70
+2022-12-12 17:28:30,621 INFO [train.py:421] (5/8) Epoch 7, batch 62000, loss[loss=2.768, over 770.00 frames. , ppl: 15.932594581521412] tot_loss[loss=2.283, over 5488328.68 frames. , ppl: 9.807504511303845], batch size: 70
+2022-12-12 17:28:30,622 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 17:28:31,386 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.71544765103413
+2022-12-12 17:30:11,235 INFO [train.py:421] (5/8) Epoch 7, batch 62200, loss[loss=2.408, over 1960.00 frames. , ppl: 11.11341349945178] tot_loss[loss=2.284, over 5442057.84 frames. , ppl: 9.818315713786781], batch size: 70
+2022-12-12 17:31:56,818 INFO [train.py:421] (5/8) Epoch 7, batch 62400, loss[loss=2.245, over 2310.00 frames. , ppl: 9.436890984990711] tot_loss[loss=2.284, over 5450870.05 frames. , ppl: 9.815199208036967], batch size: 70
+2022-12-12 17:33:40,849 INFO [train.py:421] (5/8) Epoch 7, batch 62600, loss[loss=2.774, over 630.00 frames. , ppl: 16.029169998195947] tot_loss[loss=2.283, over 5480323.10 frames. , ppl: 9.803858029094842], batch size: 70
+2022-12-12 17:35:24,955 INFO [train.py:421] (5/8) Epoch 7, batch 62800, loss[loss=2.154, over 3010.00 frames. , ppl: 8.621308348415814] tot_loss[loss=2.281, over 5528690.76 frames. , ppl: 9.790567551778558], batch size: 70
+2022-12-12 17:37:08,813 INFO [train.py:421] (5/8) Epoch 7, batch 63000, loss[loss=2.452, over 1820.00 frames. , ppl: 11.616585881073656] tot_loss[loss=2.282, over 5524840.78 frames. , ppl: 9.795575506674881], batch size: 70
+2022-12-12 17:37:08,814 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 17:37:09,562 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.722217042257858
+2022-12-12 17:38:51,300 INFO [train.py:421] (5/8) Epoch 7, batch 63200, loss[loss=2.219, over 3710.00 frames. , ppl: 9.198932582408744] tot_loss[loss=2.281, over 5563211.25 frames. , ppl: 9.786285996701617], batch size: 70
+2022-12-12 17:40:30,753 INFO [train.py:421] (5/8) Epoch 7, batch 63400, loss[loss=2.207, over 4480.00 frames. , ppl: 9.092637640540564] tot_loss[loss=2.28, over 5601203.01 frames. , ppl: 9.7766600676902], batch size: 70
+2022-12-12 17:42:12,622 INFO [train.py:421] (5/8) Epoch 7, batch 63600, loss[loss=2.28, over 1470.00 frames. , ppl: 9.78056254715028] tot_loss[loss=2.281, over 5560236.10 frames. , ppl: 9.790509648897185], batch size: 70
+2022-12-12 17:43:53,024 INFO [train.py:421] (5/8) Epoch 7, batch 63800, loss[loss=2.331, over 1960.00 frames. , ppl: 10.283131896779068] tot_loss[loss=2.282, over 5504903.57 frames. , ppl: 9.799801535130078], batch size: 70
+2022-12-12 17:45:29,446 INFO [train.py:421] (5/8) Epoch 7, batch 64000, loss[loss=2.401, over 1050.00 frames. , ppl: 11.035180561102344] tot_loss[loss=2.283, over 5480641.24 frames. , ppl: 9.80891943081019], batch size: 70
+2022-12-12 17:45:29,446 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 17:45:30,210 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.717427922051451
+2022-12-12 17:47:12,552 INFO [train.py:421] (5/8) Epoch 7, batch 64200, loss[loss=2.392, over 1960.00 frames. , ppl: 10.940167496056764] tot_loss[loss=2.283, over 5493767.49 frames. , ppl: 9.803541714300884], batch size: 70
+2022-12-12 17:48:56,535 INFO [train.py:421] (5/8) Epoch 7, batch 64400, loss[loss=2.424, over 2030.00 frames. , ppl: 11.291202905235643] tot_loss[loss=2.282, over 5506627.36 frames. , ppl: 9.798817397428405], batch size: 70
+2022-12-12 17:50:37,245 INFO [train.py:421] (5/8) Epoch 7, batch 64600, loss[loss=2.499, over 1470.00 frames. , ppl: 12.172480751117535] tot_loss[loss=2.282, over 5517493.50 frames. , ppl: 9.794543521920888], batch size: 70
+2022-12-12 17:52:15,807 INFO [train.py:421] (5/8) Epoch 7, batch 64800, loss[loss=2.364, over 1120.00 frames. , ppl: 10.630949714888613] tot_loss[loss=2.282, over 5482070.23 frames. , ppl: 9.797459615110345], batch size: 70
+2022-12-12 17:53:56,980 INFO [train.py:421] (5/8) Epoch 7, batch 65000, loss[loss=2.855, over 910.00 frames. , ppl: 17.375524262704907] tot_loss[loss=2.282, over 5482235.66 frames. , ppl: 9.791897419763565], batch size: 70
+2022-12-12 17:53:56,981 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 17:53:57,744 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.728521733447238
+2022-12-12 17:55:38,071 INFO [train.py:421] (5/8) Epoch 7, batch 65200, loss[loss=3.552, over 420.00 frames. , ppl: 34.87373544089] tot_loss[loss=2.283, over 5456209.01 frames. , ppl: 9.804328101510974], batch size: 70
+2022-12-12 17:57:16,314 INFO [train.py:421] (5/8) Epoch 7, batch 65400, loss[loss=2.698, over 840.00 frames. , ppl: 14.85199304578318] tot_loss[loss=2.284, over 5432182.18 frames. , ppl: 9.811635944579555], batch size: 70
+2022-12-12 17:58:55,652 INFO [train.py:421] (5/8) Epoch 7, batch 65600, loss[loss=2.329, over 3150.00 frames. , ppl: 10.267692920972934] tot_loss[loss=2.284, over 5441632.76 frames. , ppl: 9.812587317070024], batch size: 70
+2022-12-12 18:00:40,205 INFO [train.py:421] (5/8) Epoch 7, batch 65800, loss[loss=2.209, over 4060.00 frames. , ppl: 9.10859664441386] tot_loss[loss=2.283, over 5455930.23 frames. , ppl: 9.804604995634001], batch size: 70
+2022-12-12 18:02:18,980 INFO [train.py:421] (5/8) Epoch 7, batch 66000, loss[loss=2.275, over 4060.00 frames. , ppl: 9.728546121147623] tot_loss[loss=2.285, over 5395609.02 frames. , ppl: 9.822819755881152], batch size: 70
+2022-12-12 18:02:18,981 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 18:02:19,737 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.275, over 211138.00 frames. , ppl: 9.725315618297568
+2022-12-12 18:04:01,132 INFO [train.py:421] (5/8) Epoch 7, batch 66200, loss[loss=2.273, over 3780.00 frames. , ppl: 9.712823851727743] tot_loss[loss=2.285, over 5410862.03 frames. , ppl: 9.821140094523942], batch size: 70
+2022-12-12 18:05:46,669 INFO [train.py:421] (5/8) Epoch 7, batch 66400, loss[loss=2.389, over 1190.00 frames. , ppl: 10.904529030778159] tot_loss[loss=2.283, over 5443665.63 frames. , ppl: 9.806146120494331], batch size: 70
+2022-12-12 18:07:29,650 INFO [train.py:421] (5/8) Epoch 7, batch 66600, loss[loss=2.169, over 5530.00 frames. , ppl: 8.74569495934851] tot_loss[loss=2.283, over 5459439.94 frames. , ppl: 9.805613761320346], batch size: 70
+2022-12-12 18:09:07,106 INFO [train.py:421] (5/8) Epoch 7, batch 66800, loss[loss=2.207, over 10430.00 frames. , ppl: 9.091286682910052] tot_loss[loss=2.283, over 5451115.46 frames. , ppl: 9.807485223595586], batch size: 70
+2022-12-12 18:10:47,177 INFO [train.py:421] (5/8) Epoch 7, batch 67000, loss[loss=2.189, over 3850.00 frames. , ppl: 8.929437826338567] tot_loss[loss=2.284, over 5427249.05 frames. , ppl: 9.811670238812386], batch size: 70
+2022-12-12 18:10:47,178 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 18:10:47,941 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.274, over 211138.00 frames. , ppl: 9.71459066540528
+2022-12-12 18:12:28,639 INFO [train.py:421] (5/8) Epoch 7, batch 67200, loss[loss=2.157, over 8680.00 frames. , ppl: 8.64752294735989] tot_loss[loss=2.283, over 5450745.19 frames. , ppl: 9.80125341493887], batch size: 70
+2022-12-12 18:14:14,287 INFO [train.py:421] (5/8) Epoch 7, batch 67400, loss[loss=2.263, over 10010.00 frames. , ppl: 9.610549077369463] tot_loss[loss=2.283, over 5437619.36 frames. , ppl: 9.805167663504264], batch size: 70
+2022-12-12 18:15:54,875 INFO [train.py:421] (5/8) Epoch 7, batch 67600, loss[loss=2.192, over 6160.00 frames. , ppl: 8.9523927928119] tot_loss[loss=2.283, over 5463805.74 frames. , ppl: 9.805372154059723], batch size: 70
+2022-12-12 18:17:35,752 INFO [train.py:421] (5/8) Epoch 7, batch 67800, loss[loss=2.221, over 3570.00 frames. , ppl: 9.220753389535718] tot_loss[loss=2.284, over 5480180.81 frames. , ppl: 9.810963447836116], batch size: 70
+2022-12-12 18:19:18,273 INFO [train.py:421] (5/8) Epoch 7, batch 68000, loss[loss=2.23, over 3570.00 frames. , ppl: 9.3021370862379] tot_loss[loss=2.282, over 5536664.36 frames. , ppl: 9.80011854751769], batch size: 70
+2022-12-12 18:19:18,274 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 18:19:19,022 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.273, over 211138.00 frames. , ppl: 9.710605812793908
+2022-12-12 18:20:59,046 INFO [train.py:421] (5/8) Epoch 7, batch 68200, loss[loss=2.418, over 1610.00 frames. , ppl: 11.21912574455521] tot_loss[loss=2.282, over 5561021.24 frames. , ppl: 9.791533002644137], batch size: 70
+2022-12-12 18:22:38,124 INFO [train.py:421] (5/8) Epoch 7, batch 68400, loss[loss=2.166, over 4900.00 frames. , ppl: 8.72185436466713] tot_loss[loss=2.282, over 5544422.17 frames. , ppl: 9.794774069891565], batch size: 70
+2022-12-12 18:24:19,204 INFO [train.py:421] (5/8) Epoch 7, batch 68600, loss[loss=2.257, over 2310.00 frames. , ppl: 9.554580577666004] tot_loss[loss=2.282, over 5504957.66 frames. , ppl: 9.800024517451138], batch size: 70
+2022-12-12 18:25:59,199 INFO [train.py:421] (5/8) Epoch 7, batch 68800, loss[loss=2.295, over 2380.00 frames. , ppl: 9.926627504135316] tot_loss[loss=2.282, over 5500219.65 frames. , ppl: 9.801064489388734], batch size: 70
+2022-12-12 18:27:41,645 INFO [train.py:421] (5/8) Epoch 7, batch 69000, loss[loss=2.382, over 1890.00 frames. , ppl: 10.821883714073008] tot_loss[loss=2.283, over 5492179.00 frames. , ppl: 9.808951111271375], batch size: 70
+2022-12-12 18:27:41,646 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 18:27:42,411 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.273, over 211138.00 frames. , ppl: 9.71098525213279
+2022-12-12 18:29:24,643 INFO [train.py:421] (5/8) Epoch 7, batch 69200, loss[loss=2.276, over 2310.00 frames. , ppl: 9.735686549540402] tot_loss[loss=2.285, over 5475965.24 frames. , ppl: 9.822190659270662], batch size: 70
+2022-12-12 18:31:04,120 INFO [train.py:421] (5/8) Epoch 7, batch 69400, loss[loss=2.168, over 6720.00 frames. , ppl: 8.739798016394612] tot_loss[loss=2.283, over 5531384.43 frames. , ppl: 9.805418615910614], batch size: 70
+2022-12-12 18:32:46,722 INFO [train.py:421] (5/8) Epoch 7, batch 69600, loss[loss=2.308, over 2380.00 frames. , ppl: 10.059109633454641] tot_loss[loss=2.283, over 5568700.38 frames. , ppl: 9.803386073115673], batch size: 70
+2022-12-12 18:34:31,065 INFO [train.py:421] (5/8) Epoch 7, batch 69800, loss[loss=2.372, over 3150.00 frames. , ppl: 10.71743216156654] tot_loss[loss=2.281, over 5627355.55 frames. , ppl: 9.786847287458622], batch size: 70
+2022-12-12 18:36:12,716 INFO [train.py:421] (5/8) Epoch 7, batch 70000, loss[loss=2.464, over 2380.00 frames. , ppl: 11.745888134262758] tot_loss[loss=2.283, over 5576196.82 frames. , ppl: 9.80257921294065], batch size: 70
+2022-12-12 18:36:12,717 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 18:36:13,480 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.273, over 211138.00 frames. , ppl: 9.71248878254001
+2022-12-12 18:37:52,814 INFO [train.py:421] (5/8) Epoch 7, batch 70200, loss[loss=2.425, over 1470.00 frames. , ppl: 11.302537267465334] tot_loss[loss=2.283, over 5549771.27 frames. , ppl: 9.803431061565437], batch size: 70
+2022-12-12 18:39:34,550 INFO [train.py:421] (5/8) Epoch 7, batch 70400, loss[loss=2.199, over 3710.00 frames. , ppl: 9.017272353214233] tot_loss[loss=2.283, over 5533655.00 frames. , ppl: 9.802625462521735], batch size: 70
+2022-12-12 18:41:16,485 INFO [train.py:421] (5/8) Epoch 7, batch 70600, loss[loss=2.304, over 2940.00 frames. , ppl: 10.009838297436353] tot_loss[loss=2.284, over 5484129.80 frames. , ppl: 9.810983525916745], batch size: 70
+2022-12-12 18:43:01,174 INFO [train.py:421] (5/8) Epoch 7, batch 70800, loss[loss=2.633, over 840.00 frames. , ppl: 13.90922213835615] tot_loss[loss=2.283, over 5508998.97 frames. , ppl: 9.804318621499986], batch size: 70
+2022-12-12 18:44:42,633 INFO [train.py:421] (5/8) Epoch 7, batch 71000, loss[loss=2.414, over 1470.00 frames. , ppl: 11.176704378252099] tot_loss[loss=2.282, over 5554838.16 frames. , ppl: 9.794612173069527], batch size: 70
+2022-12-12 18:44:42,634 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 18:44:43,417 INFO [train.py:452] (5/8) Epoch 7, validation: loss=2.273, over 211138.00 frames. , ppl: 9.707774858509012
+2022-12-12 18:46:25,687 INFO [train.py:421] (5/8) Epoch 7, batch 71200, loss[loss=2.621, over 770.00 frames. , ppl: 13.750327988770083] tot_loss[loss=2.281, over 5581873.19 frames. , ppl: 9.788773030870468], batch size: 70
+2022-12-12 18:48:08,695 INFO [train.py:421] (5/8) Epoch 7, batch 71400, loss[loss=2.345, over 1680.00 frames. , ppl: 10.43657155123186] tot_loss[loss=2.281, over 5583806.48 frames. , ppl: 9.7878945335483], batch size: 70
+2022-12-12 18:49:50,945 INFO [train.py:421] (5/8) Epoch 7, batch 71600, loss[loss=2.308, over 3500.00 frames. , ppl: 10.056261265524116] tot_loss[loss=2.282, over 5562830.58 frames. , ppl: 9.794945363399252], batch size: 70
+2022-12-12 18:51:32,658 INFO [train.py:421] (5/8) Epoch 7, batch 71800, loss[loss=2.702, over 700.00 frames. , ppl: 14.904888982142682] tot_loss[loss=2.281, over 5593443.41 frames. , ppl: 9.785189866516673], batch size: 70
+2022-12-12 18:52:47,987 INFO [train.py:421] (5/8) Epoch 8, batch 0, loss[loss=2.278, over 2590.00 frames. , ppl: 9.75856499839812] tot_loss[loss=2.278, over 2590.00 frames. , ppl: 9.75856499839812], batch size: 70
+2022-12-12 18:54:30,707 INFO [train.py:421] (5/8) Epoch 8, batch 200, loss[loss=2.523, over 1120.00 frames. , ppl: 12.471646307023823] tot_loss[loss=2.268, over 566519.98 frames. , ppl: 9.659494278372136], batch size: 70
+2022-12-12 18:56:11,881 INFO [train.py:421] (5/8) Epoch 8, batch 400, loss[loss=2.343, over 1470.00 frames. , ppl: 10.415771794819557] tot_loss[loss=2.278, over 985692.02 frames. , ppl: 9.759540950289566], batch size: 70
+2022-12-12 18:57:54,087 INFO [train.py:421] (5/8) Epoch 8, batch 600, loss[loss=2.396, over 1400.00 frames. , ppl: 10.97510700090112] tot_loss[loss=2.274, over 1421680.61 frames. , ppl: 9.722294910324825], batch size: 70
+2022-12-12 18:59:34,618 INFO [train.py:421] (5/8) Epoch 8, batch 800, loss[loss=2.138, over 5600.00 frames. , ppl: 8.485766592133686] tot_loss[loss=2.272, over 1840431.75 frames. , ppl: 9.70088850035316], batch size: 70
+2022-12-12 19:01:16,027 INFO [train.py:421] (5/8) Epoch 8, batch 1000, loss[loss=2.302, over 2310.00 frames. , ppl: 9.994354093855625] tot_loss[loss=2.266, over 2252904.53 frames. , ppl: 9.644998445110327], batch size: 70
+2022-12-12 19:01:16,028 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 19:01:16,794 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.274, over 211138.00 frames. , ppl: 9.714002609944576
+2022-12-12 19:02:59,996 INFO [train.py:421] (5/8) Epoch 8, batch 1200, loss[loss=2.284, over 1680.00 frames. , ppl: 9.81787194044179] tot_loss[loss=2.267, over 2579400.00 frames. , ppl: 9.654246360597185], batch size: 70
+2022-12-12 19:04:36,354 INFO [train.py:421] (5/8) Epoch 8, batch 1400, loss[loss=4.789, over 280.00 frames. , ppl: 120.12499216140846] tot_loss[loss=2.268, over 2863838.12 frames. , ppl: 9.655474388840457], batch size: 70
+2022-12-12 19:06:17,625 INFO [train.py:421] (5/8) Epoch 8, batch 1600, loss[loss=2.157, over 4060.00 frames. , ppl: 8.64425647027928] tot_loss[loss=2.267, over 3126231.08 frames. , ppl: 9.649970592872341], batch size: 70
+2022-12-12 19:08:00,659 INFO [train.py:421] (5/8) Epoch 8, batch 1800, loss[loss=2.635, over 840.00 frames. , ppl: 13.944510443835966] tot_loss[loss=2.269, over 3340381.84 frames. , ppl: 9.672261524763206], batch size: 70
+2022-12-12 19:09:38,929 INFO [train.py:421] (5/8) Epoch 8, batch 2000, loss[loss=2.281, over 3990.00 frames. , ppl: 9.785926509987387] tot_loss[loss=2.27, over 3546638.30 frames. , ppl: 9.682419879215], batch size: 70
+2022-12-12 19:09:38,930 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 19:09:39,713 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.70567014079976
+2022-12-12 19:11:23,691 INFO [train.py:421] (5/8) Epoch 8, batch 2200, loss[loss=2.391, over 1400.00 frames. , ppl: 10.929372146734982] tot_loss[loss=2.269, over 3758442.68 frames. , ppl: 9.672557815542575], batch size: 70
+2022-12-12 19:13:02,999 INFO [train.py:421] (5/8) Epoch 8, batch 2400, loss[loss=2.336, over 1400.00 frames. , ppl: 10.344840186767886] tot_loss[loss=2.27, over 3928357.52 frames. , ppl: 9.67729792104743], batch size: 70
+2022-12-12 19:14:41,152 INFO [train.py:421] (5/8) Epoch 8, batch 2600, loss[loss=2.296, over 3150.00 frames. , ppl: 9.935287562172645] tot_loss[loss=2.269, over 4137010.66 frames. , ppl: 9.674041585029773], batch size: 70
+2022-12-12 19:16:22,764 INFO [train.py:421] (5/8) Epoch 8, batch 2800, loss[loss=2.148, over 3360.00 frames. , ppl: 8.565376073367668] tot_loss[loss=2.269, over 4283269.24 frames. , ppl: 9.6717950979525], batch size: 70
+2022-12-12 19:18:01,865 INFO [train.py:421] (5/8) Epoch 8, batch 3000, loss[loss=2.491, over 910.00 frames. , ppl: 12.074860207155405] tot_loss[loss=2.271, over 4397303.34 frames. , ppl: 9.687754508901525], batch size: 70
+2022-12-12 19:18:01,865 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 19:18:02,632 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.276, over 211138.00 frames. , ppl: 9.739168370327212
+2022-12-12 19:19:44,666 INFO [train.py:421] (5/8) Epoch 8, batch 3200, loss[loss=2.343, over 2730.00 frames. , ppl: 10.416702507973211] tot_loss[loss=2.272, over 4467929.37 frames. , ppl: 9.698120895583823], batch size: 70
+2022-12-12 19:21:28,589 INFO [train.py:421] (5/8) Epoch 8, batch 3400, loss[loss=2.253, over 2800.00 frames. , ppl: 9.51384045477468] tot_loss[loss=2.271, over 4599534.62 frames. , ppl: 9.688613153620809], batch size: 70
+2022-12-12 19:23:11,767 INFO [train.py:421] (5/8) Epoch 8, batch 3600, loss[loss=2.213, over 5040.00 frames. , ppl: 9.145311420925738] tot_loss[loss=2.272, over 4683586.39 frames. , ppl: 9.698626929285401], batch size: 70
+2022-12-12 19:24:50,895 INFO [train.py:421] (5/8) Epoch 8, batch 3800, loss[loss=2.425, over 1820.00 frames. , ppl: 11.307834377888527] tot_loss[loss=2.272, over 4757405.82 frames. , ppl: 9.697413939167554], batch size: 70
+2022-12-12 19:26:35,933 INFO [train.py:421] (5/8) Epoch 8, batch 4000, loss[loss=2.332, over 1890.00 frames. , ppl: 10.295580972840133] tot_loss[loss=2.273, over 4787037.28 frames. , ppl: 9.709701904937559], batch size: 70
+2022-12-12 19:26:35,934 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 19:26:36,699 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.709655842207155
+2022-12-12 19:28:16,998 INFO [train.py:421] (5/8) Epoch 8, batch 4200, loss[loss=2.581, over 980.00 frames. , ppl: 13.216637503768244] tot_loss[loss=2.274, over 4847968.46 frames. , ppl: 9.71949133610687], batch size: 70
+2022-12-12 19:29:59,662 INFO [train.py:421] (5/8) Epoch 8, batch 4400, loss[loss=2.396, over 1890.00 frames. , ppl: 10.977384573595229] tot_loss[loss=2.272, over 4958500.86 frames. , ppl: 9.702239023893053], batch size: 70
+2022-12-12 19:31:43,085 INFO [train.py:421] (5/8) Epoch 8, batch 4600, loss[loss=2.216, over 4410.00 frames. , ppl: 9.170159944324796] tot_loss[loss=2.272, over 5020183.26 frames. , ppl: 9.69774218113877], batch size: 70
+2022-12-12 19:33:23,025 INFO [train.py:421] (5/8) Epoch 8, batch 4800, loss[loss=2.201, over 6300.00 frames. , ppl: 9.034649125807137] tot_loss[loss=2.273, over 5044322.00 frames. , ppl: 9.710705426329097], batch size: 70
+2022-12-12 19:35:03,764 INFO [train.py:421] (5/8) Epoch 8, batch 5000, loss[loss=2.36, over 1750.00 frames. , ppl: 10.59609746965588] tot_loss[loss=2.274, over 5046006.76 frames. , ppl: 9.7220802216761], batch size: 70
+2022-12-12 19:35:03,764 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 19:35:04,513 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.275, over 211138.00 frames. , ppl: 9.73047586561236
+2022-12-12 19:36:45,802 INFO [train.py:421] (5/8) Epoch 8, batch 5200, loss[loss=2.362, over 1820.00 frames. , ppl: 10.609050758818476] tot_loss[loss=2.276, over 5058509.05 frames. , ppl: 9.733331585490392], batch size: 70
+2022-12-12 19:38:27,894 INFO [train.py:421] (5/8) Epoch 8, batch 5400, loss[loss=2.607, over 700.00 frames. , ppl: 13.553707908159378] tot_loss[loss=2.276, over 5081851.30 frames. , ppl: 9.733678429906803], batch size: 70
+2022-12-12 19:40:10,516 INFO [train.py:421] (5/8) Epoch 8, batch 5600, loss[loss=2.247, over 3290.00 frames. , ppl: 9.457012840639084] tot_loss[loss=2.276, over 5127050.25 frames. , ppl: 9.735164986237663], batch size: 70
+2022-12-12 19:41:53,317 INFO [train.py:421] (5/8) Epoch 8, batch 5800, loss[loss=2.284, over 3290.00 frames. , ppl: 9.815665460236914] tot_loss[loss=2.275, over 5164911.32 frames. , ppl: 9.72855414544794], batch size: 70
+2022-12-12 19:43:36,498 INFO [train.py:421] (5/8) Epoch 8, batch 6000, loss[loss=2.174, over 2380.00 frames. , ppl: 8.790164517671377] tot_loss[loss=2.275, over 5213070.85 frames. , ppl: 9.72798377661683], batch size: 70
+2022-12-12 19:43:36,498 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 19:43:37,249 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.709706140891356
+2022-12-12 19:45:21,824 INFO [train.py:421] (5/8) Epoch 8, batch 6200, loss[loss=2.284, over 2310.00 frames. , ppl: 9.814247365518842] tot_loss[loss=2.275, over 5252660.46 frames. , ppl: 9.724797951007703], batch size: 70
+2022-12-12 19:47:05,552 INFO [train.py:421] (5/8) Epoch 8, batch 6400, loss[loss=2.405, over 1750.00 frames. , ppl: 11.079536944053107] tot_loss[loss=2.275, over 5265834.22 frames. , ppl: 9.731129053011465], batch size: 70
+2022-12-12 19:48:44,525 INFO [train.py:421] (5/8) Epoch 8, batch 6600, loss[loss=2.292, over 3990.00 frames. , ppl: 9.891734045713543] tot_loss[loss=2.276, over 5284758.59 frames. , ppl: 9.738605988298607], batch size: 70
+2022-12-12 19:50:24,283 INFO [train.py:421] (5/8) Epoch 8, batch 6800, loss[loss=2.15, over 3290.00 frames. , ppl: 8.587108772208136] tot_loss[loss=2.276, over 5330959.19 frames. , ppl: 9.73442822079905], batch size: 70
+2022-12-12 19:52:06,684 INFO [train.py:421] (5/8) Epoch 8, batch 7000, loss[loss=2.337, over 1400.00 frames. , ppl: 10.348315268470603] tot_loss[loss=2.275, over 5330169.71 frames. , ppl: 9.73234727703014], batch size: 70
+2022-12-12 19:52:06,684 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 19:52:07,438 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.703498378235237
+2022-12-12 19:53:49,599 INFO [train.py:421] (5/8) Epoch 8, batch 7200, loss[loss=2.277, over 1260.00 frames. , ppl: 9.752041727219673] tot_loss[loss=2.276, over 5343112.17 frames. , ppl: 9.734939676788414], batch size: 70
+2022-12-12 19:55:33,510 INFO [train.py:421] (5/8) Epoch 8, batch 7400, loss[loss=2.471, over 1820.00 frames. , ppl: 11.83428295863413] tot_loss[loss=2.274, over 5408575.34 frames. , ppl: 9.717365193193517], batch size: 70
+2022-12-12 19:57:12,621 INFO [train.py:421] (5/8) Epoch 8, batch 7600, loss[loss=2.252, over 2380.00 frames. , ppl: 9.502572011346116] tot_loss[loss=2.275, over 5372648.30 frames. , ppl: 9.730097700711832], batch size: 70
+2022-12-12 19:58:56,550 INFO [train.py:421] (5/8) Epoch 8, batch 7800, loss[loss=2.2, over 3360.00 frames. , ppl: 9.026111317823535] tot_loss[loss=2.276, over 5392468.37 frames. , ppl: 9.73308455374703], batch size: 70
+2022-12-12 20:00:37,106 INFO [train.py:421] (5/8) Epoch 8, batch 8000, loss[loss=2.283, over 3640.00 frames. , ppl: 9.80182271172801] tot_loss[loss=2.276, over 5389281.30 frames. , ppl: 9.737818881264785], batch size: 70
+2022-12-12 20:00:37,106 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 20:00:37,873 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.274, over 211138.00 frames. , ppl: 9.715900618606277
+2022-12-12 20:02:19,441 INFO [train.py:421] (5/8) Epoch 8, batch 8200, loss[loss=2.201, over 9590.00 frames. , ppl: 9.031585111361592] tot_loss[loss=2.274, over 5478151.48 frames. , ppl: 9.717844517031432], batch size: 70
+2022-12-12 20:04:00,775 INFO [train.py:421] (5/8) Epoch 8, batch 8400, loss[loss=2.37, over 1960.00 frames. , ppl: 10.696769765843671] tot_loss[loss=2.274, over 5489773.28 frames. , ppl: 9.713967341603066], batch size: 70
+2022-12-12 20:05:43,615 INFO [train.py:421] (5/8) Epoch 8, batch 8600, loss[loss=2.373, over 2100.00 frames. , ppl: 10.734252803641974] tot_loss[loss=2.275, over 5481228.06 frames. , ppl: 9.724677965380451], batch size: 70
+2022-12-12 20:07:22,554 INFO [train.py:421] (5/8) Epoch 8, batch 8800, loss[loss=2.446, over 1610.00 frames. , ppl: 11.545023618111832] tot_loss[loss=2.275, over 5480936.18 frames. , ppl: 9.724785548622119], batch size: 70
+2022-12-12 20:09:06,528 INFO [train.py:421] (5/8) Epoch 8, batch 9000, loss[loss=2.227, over 6720.00 frames. , ppl: 9.271868750547705] tot_loss[loss=2.275, over 5465612.97 frames. , ppl: 9.731884800533441], batch size: 70
+2022-12-12 20:09:06,529 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 20:09:07,263 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.70899192399486
+2022-12-12 20:10:45,538 INFO [train.py:421] (5/8) Epoch 8, batch 9200, loss[loss=2.5, over 1120.00 frames. , ppl: 12.184947957226221] tot_loss[loss=2.276, over 5457422.06 frames. , ppl: 9.735589590493348], batch size: 70
+2022-12-12 20:12:29,544 INFO [train.py:421] (5/8) Epoch 8, batch 9400, loss[loss=2.607, over 770.00 frames. , ppl: 13.554676580456258] tot_loss[loss=2.276, over 5462852.76 frames. , ppl: 9.742168140008117], batch size: 70
+2022-12-12 20:14:10,768 INFO [train.py:421] (5/8) Epoch 8, batch 9600, loss[loss=2.188, over 7490.00 frames. , ppl: 8.917822276416299] tot_loss[loss=2.277, over 5486041.43 frames. , ppl: 9.743238560238515], batch size: 70
+2022-12-12 20:15:50,131 INFO [train.py:421] (5/8) Epoch 8, batch 9800, loss[loss=2.387, over 1960.00 frames. , ppl: 10.884636178471801] tot_loss[loss=2.278, over 5471715.90 frames. , ppl: 9.754837043330488], batch size: 70
+2022-12-12 20:17:30,549 INFO [train.py:421] (5/8) Epoch 8, batch 10000, loss[loss=2.666, over 770.00 frames. , ppl: 14.377880506572536] tot_loss[loss=2.278, over 5480566.35 frames. , ppl: 9.756680546338368], batch size: 70
+2022-12-12 20:17:30,549 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 20:17:31,332 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.276, over 211138.00 frames. , ppl: 9.734638884938837
+2022-12-12 20:19:15,341 INFO [train.py:421] (5/8) Epoch 8, batch 10200, loss[loss=2.282, over 3360.00 frames. , ppl: 9.799781732501248] tot_loss[loss=2.279, over 5459829.89 frames. , ppl: 9.763012283068484], batch size: 70
+2022-12-12 20:20:52,751 INFO [train.py:421] (5/8) Epoch 8, batch 10400, loss[loss=2.322, over 1820.00 frames. , ppl: 10.197660287909978] tot_loss[loss=2.279, over 5479491.39 frames. , ppl: 9.762526652113838], batch size: 70
+2022-12-12 20:22:35,047 INFO [train.py:421] (5/8) Epoch 8, batch 10600, loss[loss=2.304, over 4830.00 frames. , ppl: 10.018286985456688] tot_loss[loss=2.278, over 5507850.37 frames. , ppl: 9.754104987723004], batch size: 70
+2022-12-12 20:24:13,555 INFO [train.py:421] (5/8) Epoch 8, batch 10800, loss[loss=2.209, over 10780.00 frames. , ppl: 9.111146600078554] tot_loss[loss=2.277, over 5544176.36 frames. , ppl: 9.747577875760555], batch size: 70
+2022-12-12 20:25:55,050 INFO [train.py:421] (5/8) Epoch 8, batch 11000, loss[loss=2.236, over 3780.00 frames. , ppl: 9.352863084568948] tot_loss[loss=2.277, over 5527681.66 frames. , ppl: 9.749819076132027], batch size: 70
+2022-12-12 20:25:55,051 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 20:25:55,810 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.706705920947643
+2022-12-12 20:27:38,289 INFO [train.py:421] (5/8) Epoch 8, batch 11200, loss[loss=2.374, over 2100.00 frames. , ppl: 10.738319351682206] tot_loss[loss=2.276, over 5547491.95 frames. , ppl: 9.741650661578866], batch size: 70
+2022-12-12 20:29:17,857 INFO [train.py:421] (5/8) Epoch 8, batch 11400, loss[loss=2.377, over 1190.00 frames. , ppl: 10.773879489817853] tot_loss[loss=2.276, over 5535050.14 frames. , ppl: 9.740901004473985], batch size: 70
+2022-12-12 20:30:55,494 INFO [train.py:421] (5/8) Epoch 8, batch 11600, loss[loss=2.927, over 630.00 frames. , ppl: 18.673471348150596] tot_loss[loss=2.278, over 5494373.92 frames. , ppl: 9.754877778202093], batch size: 70
+2022-12-12 20:32:40,056 INFO [train.py:421] (5/8) Epoch 8, batch 11800, loss[loss=2.357, over 1540.00 frames. , ppl: 10.561340864718884] tot_loss[loss=2.275, over 5576470.94 frames. , ppl: 9.728277266373228], batch size: 70
+2022-12-12 20:34:17,560 INFO [train.py:421] (5/8) Epoch 8, batch 12000, loss[loss=2.34, over 2240.00 frames. , ppl: 10.386026479862604] tot_loss[loss=2.275, over 5598611.66 frames. , ppl: 9.727580013708005], batch size: 70
+2022-12-12 20:34:17,560 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 20:34:18,324 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.712665599212405
+2022-12-12 20:36:00,925 INFO [train.py:421] (5/8) Epoch 8, batch 12200, loss[loss=2.303, over 2590.00 frames. , ppl: 10.007900152961543] tot_loss[loss=2.274, over 5629699.26 frames. , ppl: 9.718627513680282], batch size: 70
+2022-12-12 20:37:42,065 INFO [train.py:421] (5/8) Epoch 8, batch 12400, loss[loss=2.218, over 8050.00 frames. , ppl: 9.191104571366838] tot_loss[loss=2.274, over 5625769.13 frames. , ppl: 9.71639732775821], batch size: 70
+2022-12-12 20:39:22,979 INFO [train.py:421] (5/8) Epoch 8, batch 12600, loss[loss=2.089, over 3220.00 frames. , ppl: 8.079459599787278] tot_loss[loss=2.274, over 5602368.13 frames. , ppl: 9.721798631273652], batch size: 70
+2022-12-12 20:41:03,277 INFO [train.py:421] (5/8) Epoch 8, batch 12800, loss[loss=2.231, over 5320.00 frames. , ppl: 9.31245199621816] tot_loss[loss=2.275, over 5612002.93 frames. , ppl: 9.723227576664248], batch size: 70
+2022-12-12 20:42:44,549 INFO [train.py:421] (5/8) Epoch 8, batch 13000, loss[loss=2.458, over 1120.00 frames. , ppl: 11.675875182393481] tot_loss[loss=2.275, over 5612857.18 frames. , ppl: 9.725265654716797], batch size: 70
+2022-12-12 20:42:44,549 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 20:42:45,301 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.274, over 211138.00 frames. , ppl: 9.722002639478895
+2022-12-12 20:44:28,087 INFO [train.py:421] (5/8) Epoch 8, batch 13200, loss[loss=2.235, over 3850.00 frames. , ppl: 9.343779577201571] tot_loss[loss=2.275, over 5590661.84 frames. , ppl: 9.731869607820238], batch size: 70
+2022-12-12 20:46:10,753 INFO [train.py:421] (5/8) Epoch 8, batch 13400, loss[loss=2.203, over 4410.00 frames. , ppl: 9.056057878590806] tot_loss[loss=2.276, over 5536759.65 frames. , ppl: 9.739632684056854], batch size: 70
+2022-12-12 20:47:53,569 INFO [train.py:421] (5/8) Epoch 8, batch 13600, loss[loss=2.23, over 2730.00 frames. , ppl: 9.298077815653023] tot_loss[loss=2.274, over 5598468.38 frames. , ppl: 9.71732463503901], batch size: 70
+2022-12-12 20:49:37,900 INFO [train.py:421] (5/8) Epoch 8, batch 13800, loss[loss=2.221, over 3640.00 frames. , ppl: 9.215992500252984] tot_loss[loss=2.274, over 5610780.99 frames. , ppl: 9.713701494261805], batch size: 70
+2022-12-12 20:51:19,926 INFO [train.py:421] (5/8) Epoch 8, batch 14000, loss[loss=2.435, over 1330.00 frames. , ppl: 11.419916643980653] tot_loss[loss=2.275, over 5552174.14 frames. , ppl: 9.725079804764947], batch size: 70
+2022-12-12 20:51:19,927 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 20:51:20,692 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.699218039613152
+2022-12-12 20:53:03,175 INFO [train.py:421] (5/8) Epoch 8, batch 14200, loss[loss=2.126, over 5950.00 frames. , ppl: 8.384285208760744] tot_loss[loss=2.275, over 5528926.51 frames. , ppl: 9.73068291095477], batch size: 70
+2022-12-12 20:54:47,607 INFO [train.py:421] (5/8) Epoch 8, batch 14400, loss[loss=2.258, over 3220.00 frames. , ppl: 9.562700088803897] tot_loss[loss=2.277, over 5488149.20 frames. , ppl: 9.748833602657553], batch size: 70
+2022-12-12 20:56:30,693 INFO [train.py:421] (5/8) Epoch 8, batch 14600, loss[loss=2.202, over 6440.00 frames. , ppl: 9.044972636281061] tot_loss[loss=2.277, over 5487451.37 frames. , ppl: 9.747223166719932], batch size: 70
+2022-12-12 20:58:11,810 INFO [train.py:421] (5/8) Epoch 8, batch 14800, loss[loss=2.777, over 630.00 frames. , ppl: 16.06919840220122] tot_loss[loss=2.277, over 5472392.95 frames. , ppl: 9.748340682801222], batch size: 70
+2022-12-12 20:59:52,437 INFO [train.py:421] (5/8) Epoch 8, batch 15000, loss[loss=2.297, over 2800.00 frames. , ppl: 9.941329738452167] tot_loss[loss=2.278, over 5480420.51 frames. , ppl: 9.752677592435072], batch size: 70
+2022-12-12 20:59:52,437 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 20:59:53,203 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.698958207336654
+2022-12-12 21:01:33,668 INFO [train.py:421] (5/8) Epoch 8, batch 15200, loss[loss=2.325, over 1050.00 frames. , ppl: 10.229300818077169] tot_loss[loss=2.277, over 5480034.16 frames. , ppl: 9.744710948858334], batch size: 70
+2022-12-12 21:03:13,958 INFO [train.py:421] (5/8) Epoch 8, batch 15400, loss[loss=2.253, over 3710.00 frames. , ppl: 9.517952486024035] tot_loss[loss=2.276, over 5497592.01 frames. , ppl: 9.741452424023814], batch size: 70
+2022-12-12 21:04:56,852 INFO [train.py:421] (5/8) Epoch 8, batch 15600, loss[loss=2.28, over 4130.00 frames. , ppl: 9.778320504586027] tot_loss[loss=2.276, over 5518807.36 frames. , ppl: 9.741214175710978], batch size: 70
+2022-12-12 21:06:37,415 INFO [train.py:421] (5/8) Epoch 8, batch 15800, loss[loss=2.134, over 11270.00 frames. , ppl: 8.450306701967659] tot_loss[loss=2.277, over 5502458.64 frames. , ppl: 9.74666457874234], batch size: 70
+2022-12-12 21:08:18,883 INFO [train.py:421] (5/8) Epoch 8, batch 16000, loss[loss=2.377, over 2380.00 frames. , ppl: 10.7730241101393] tot_loss[loss=2.278, over 5441532.01 frames. , ppl: 9.761380810077817], batch size: 70
+2022-12-12 21:08:18,883 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 21:08:19,648 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.704328531673617
+2022-12-12 21:09:58,159 INFO [train.py:421] (5/8) Epoch 8, batch 16200, loss[loss=2.535, over 840.00 frames. , ppl: 12.618615033724689] tot_loss[loss=2.278, over 5440281.41 frames. , ppl: 9.754188534188973], batch size: 70
+2022-12-12 21:11:38,686 INFO [train.py:421] (5/8) Epoch 8, batch 16400, loss[loss=2.237, over 2870.00 frames. , ppl: 9.36932892638687] tot_loss[loss=2.278, over 5443632.97 frames. , ppl: 9.760803634214508], batch size: 70
+2022-12-12 21:13:18,887 INFO [train.py:421] (5/8) Epoch 8, batch 16600, loss[loss=2.32, over 2660.00 frames. , ppl: 10.175529358737554] tot_loss[loss=2.279, over 5435947.68 frames. , ppl: 9.762325976117088], batch size: 70
+2022-12-12 21:14:59,021 INFO [train.py:421] (5/8) Epoch 8, batch 16800, loss[loss=2.289, over 2240.00 frames. , ppl: 9.867826444158442] tot_loss[loss=2.278, over 5478866.12 frames. , ppl: 9.75691472478989], batch size: 70
+2022-12-12 21:16:43,636 INFO [train.py:421] (5/8) Epoch 8, batch 17000, loss[loss=2.29, over 1610.00 frames. , ppl: 9.87288730393517] tot_loss[loss=2.278, over 5497163.22 frames. , ppl: 9.754728993517219], batch size: 70
+2022-12-12 21:16:43,636 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 21:16:44,419 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.701667408509223
+2022-12-12 21:18:25,209 INFO [train.py:421] (5/8) Epoch 8, batch 17200, loss[loss=2.16, over 8750.00 frames. , ppl: 8.674678450564208] tot_loss[loss=2.279, over 5467099.87 frames. , ppl: 9.764663308534281], batch size: 70
+2022-12-12 21:20:08,370 INFO [train.py:421] (5/8) Epoch 8, batch 17400, loss[loss=3.209, over 490.00 frames. , ppl: 24.765689406846207] tot_loss[loss=2.279, over 5433785.24 frames. , ppl: 9.76816039768887], batch size: 70
+2022-12-12 21:21:48,828 INFO [train.py:421] (5/8) Epoch 8, batch 17600, loss[loss=2.238, over 3290.00 frames. , ppl: 9.376668749340768] tot_loss[loss=2.28, over 5411906.34 frames. , ppl: 9.775293779533976], batch size: 70
+2022-12-12 21:23:31,881 INFO [train.py:421] (5/8) Epoch 8, batch 17800, loss[loss=2.268, over 3920.00 frames. , ppl: 9.66344208565184] tot_loss[loss=2.28, over 5393928.16 frames. , ppl: 9.77184737211058], batch size: 70
+2022-12-12 21:25:12,561 INFO [train.py:421] (5/8) Epoch 8, batch 18000, loss[loss=2.212, over 8330.00 frames. , ppl: 9.131397502684147] tot_loss[loss=2.28, over 5369329.99 frames. , ppl: 9.779720337703484], batch size: 70
+2022-12-12 21:25:12,561 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 21:25:13,280 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.271, over 211138.00 frames. , ppl: 9.693148970049764
+2022-12-12 21:26:52,972 INFO [train.py:421] (5/8) Epoch 8, batch 18200, loss[loss=2.52, over 1330.00 frames. , ppl: 12.42914969878861] tot_loss[loss=2.28, over 5368607.22 frames. , ppl: 9.778219626037282], batch size: 70
+2022-12-12 21:28:32,811 INFO [train.py:421] (5/8) Epoch 8, batch 18400, loss[loss=2.241, over 3990.00 frames. , ppl: 9.400151799537941] tot_loss[loss=2.28, over 5383818.96 frames. , ppl: 9.774247261591258], batch size: 70
+2022-12-12 21:30:13,680 INFO [train.py:421] (5/8) Epoch 8, batch 18600, loss[loss=2.524, over 1330.00 frames. , ppl: 12.473726498061723] tot_loss[loss=2.279, over 5417556.81 frames. , ppl: 9.764721605781132], batch size: 70
+2022-12-12 21:31:51,692 INFO [train.py:421] (5/8) Epoch 8, batch 18800, loss[loss=2.259, over 5950.00 frames. , ppl: 9.577230501712199] tot_loss[loss=2.278, over 5438628.68 frames. , ppl: 9.75986403385716], batch size: 70
+2022-12-12 21:33:32,792 INFO [train.py:421] (5/8) Epoch 8, batch 19000, loss[loss=2.294, over 2380.00 frames. , ppl: 9.915469419606241] tot_loss[loss=2.279, over 5438131.35 frames. , ppl: 9.763822593406992], batch size: 70
+2022-12-12 21:33:32,792 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 21:33:33,556 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.703235558989775
+2022-12-12 21:35:14,718 INFO [train.py:421] (5/8) Epoch 8, batch 19200, loss[loss=2.337, over 2170.00 frames. , ppl: 10.34771557553174] tot_loss[loss=2.278, over 5446674.61 frames. , ppl: 9.757882518459834], batch size: 70
+2022-12-12 21:36:52,519 INFO [train.py:421] (5/8) Epoch 8, batch 19400, loss[loss=2.357, over 1470.00 frames. , ppl: 10.561559166373886] tot_loss[loss=2.278, over 5435531.66 frames. , ppl: 9.761179601270412], batch size: 70
+2022-12-12 21:38:36,787 INFO [train.py:421] (5/8) Epoch 8, batch 19600, loss[loss=2.168, over 6020.00 frames. , ppl: 8.74440904355577] tot_loss[loss=2.279, over 5409621.25 frames. , ppl: 9.768712902765342], batch size: 70
+2022-12-12 21:40:17,794 INFO [train.py:421] (5/8) Epoch 8, batch 19800, loss[loss=2.306, over 3220.00 frames. , ppl: 10.029900204003166] tot_loss[loss=2.278, over 5435307.92 frames. , ppl: 9.753808535567298], batch size: 70
+2022-12-12 21:42:00,376 INFO [train.py:421] (5/8) Epoch 8, batch 20000, loss[loss=2.252, over 2730.00 frames. , ppl: 9.50584601875475] tot_loss[loss=2.278, over 5417095.77 frames. , ppl: 9.758304306420262], batch size: 70
+2022-12-12 21:42:00,377 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 21:42:01,125 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.70147643313627
+2022-12-12 21:43:41,743 INFO [train.py:421] (5/8) Epoch 8, batch 20200, loss[loss=2.573, over 700.00 frames. , ppl: 13.109841549825786] tot_loss[loss=2.278, over 5452161.28 frames. , ppl: 9.752520976832512], batch size: 70
+2022-12-12 21:45:25,099 INFO [train.py:421] (5/8) Epoch 8, batch 20400, loss[loss=2.63, over 770.00 frames. , ppl: 13.871823086827558] tot_loss[loss=2.276, over 5510031.78 frames. , ppl: 9.737591334698644], batch size: 70
+2022-12-12 21:47:09,850 INFO [train.py:421] (5/8) Epoch 8, batch 20600, loss[loss=2.176, over 10360.00 frames. , ppl: 8.814111271191459] tot_loss[loss=2.275, over 5547465.97 frames. , ppl: 9.731431693181227], batch size: 70
+2022-12-12 21:48:48,217 INFO [train.py:421] (5/8) Epoch 8, batch 20800, loss[loss=2.223, over 7700.00 frames. , ppl: 9.231663458284713] tot_loss[loss=2.276, over 5540197.70 frames. , ppl: 9.740088360126023], batch size: 70
+2022-12-12 21:50:31,019 INFO [train.py:421] (5/8) Epoch 8, batch 21000, loss[loss=2.422, over 2170.00 frames. , ppl: 11.271341737347713] tot_loss[loss=2.277, over 5542267.18 frames. , ppl: 9.74486520877694], batch size: 70
+2022-12-12 21:50:31,019 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 21:50:31,777 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.699197941840623
+2022-12-12 21:52:13,972 INFO [train.py:421] (5/8) Epoch 8, batch 21200, loss[loss=2.415, over 2170.00 frames. , ppl: 11.184553562022092] tot_loss[loss=2.277, over 5536211.06 frames. , ppl: 9.746281482889417], batch size: 70
+2022-12-12 21:53:55,500 INFO [train.py:421] (5/8) Epoch 8, batch 21400, loss[loss=2.396, over 1470.00 frames. , ppl: 10.974540225716137] tot_loss[loss=2.277, over 5545782.93 frames. , ppl: 9.74624671070798], batch size: 70
+2022-12-12 21:55:37,200 INFO [train.py:421] (5/8) Epoch 8, batch 21600, loss[loss=2.22, over 9450.00 frames. , ppl: 9.202979801510518] tot_loss[loss=2.278, over 5489170.62 frames. , ppl: 9.758706773031543], batch size: 70
+2022-12-12 21:57:17,883 INFO [train.py:421] (5/8) Epoch 8, batch 21800, loss[loss=2.264, over 2590.00 frames. , ppl: 9.624262198075762] tot_loss[loss=2.277, over 5533107.59 frames. , ppl: 9.745731966214361], batch size: 70
+2022-12-12 21:59:00,465 INFO [train.py:421] (5/8) Epoch 8, batch 22000, loss[loss=2.263, over 3290.00 frames. , ppl: 9.612268257261773] tot_loss[loss=2.278, over 5505930.04 frames. , ppl: 9.75546125665941], batch size: 70
+2022-12-12 21:59:00,466 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 21:59:01,248 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.271, over 211138.00 frames. , ppl: 9.689339266199932
+2022-12-12 22:00:41,755 INFO [train.py:421] (5/8) Epoch 8, batch 22200, loss[loss=2.298, over 2590.00 frames. , ppl: 9.952508235198001] tot_loss[loss=2.277, over 5545044.00 frames. , ppl: 9.745969141245384], batch size: 70
+2022-12-12 22:02:23,779 INFO [train.py:421] (5/8) Epoch 8, batch 22400, loss[loss=2.232, over 2240.00 frames. , ppl: 9.321051000419605] tot_loss[loss=2.276, over 5555956.41 frames. , ppl: 9.740362388127501], batch size: 70
+2022-12-12 22:04:05,302 INFO [train.py:421] (5/8) Epoch 8, batch 22600, loss[loss=2.464, over 1190.00 frames. , ppl: 11.750834061549238] tot_loss[loss=2.277, over 5537941.36 frames. , ppl: 9.747620204444473], batch size: 70
+2022-12-12 22:05:46,665 INFO [train.py:421] (5/8) Epoch 8, batch 22800, loss[loss=2.562, over 1540.00 frames. , ppl: 12.959203881570138] tot_loss[loss=2.276, over 5560303.71 frames. , ppl: 9.742059381330712], batch size: 70
+2022-12-12 22:07:29,981 INFO [train.py:421] (5/8) Epoch 8, batch 23000, loss[loss=2.373, over 1610.00 frames. , ppl: 10.730362658941974] tot_loss[loss=2.276, over 5566786.43 frames. , ppl: 9.737617733677641], batch size: 70
+2022-12-12 22:07:29,981 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 22:07:30,755 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.695581021092998
+2022-12-12 22:09:13,794 INFO [train.py:421] (5/8) Epoch 8, batch 23200, loss[loss=2.255, over 2940.00 frames. , ppl: 9.535588505423561] tot_loss[loss=2.276, over 5558490.68 frames. , ppl: 9.734424254791724], batch size: 70
+2022-12-12 22:10:53,394 INFO [train.py:421] (5/8) Epoch 8, batch 23400, loss[loss=2.245, over 2660.00 frames. , ppl: 9.438108276828359] tot_loss[loss=2.277, over 5518338.57 frames. , ppl: 9.748478019154906], batch size: 70
+2022-12-12 22:12:32,785 INFO [train.py:421] (5/8) Epoch 8, batch 23600, loss[loss=2.333, over 1890.00 frames. , ppl: 10.309301695917199] tot_loss[loss=2.278, over 5497171.59 frames. , ppl: 9.754880607009525], batch size: 70
+2022-12-12 22:14:11,811 INFO [train.py:421] (5/8) Epoch 8, batch 23800, loss[loss=2.273, over 3150.00 frames. , ppl: 9.707750355274538] tot_loss[loss=2.278, over 5483592.03 frames. , ppl: 9.756249819957649], batch size: 70
+2022-12-12 22:15:55,040 INFO [train.py:421] (5/8) Epoch 8, batch 24000, loss[loss=2.246, over 4270.00 frames. , ppl: 9.44689364132139] tot_loss[loss=2.278, over 5487285.19 frames. , ppl: 9.752416639587878], batch size: 70
+2022-12-12 22:15:55,041 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 22:15:55,805 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.708671477552679
+2022-12-12 22:17:36,360 INFO [train.py:421] (5/8) Epoch 8, batch 24200, loss[loss=2.325, over 2590.00 frames. , ppl: 10.224952751940569] tot_loss[loss=2.278, over 5457111.94 frames. , ppl: 9.753630142646404], batch size: 70
+2022-12-12 22:19:17,624 INFO [train.py:421] (5/8) Epoch 8, batch 24400, loss[loss=2.217, over 3360.00 frames. , ppl: 9.180839208173827] tot_loss[loss=2.279, over 5427170.47 frames. , ppl: 9.763841367104872], batch size: 70
+2022-12-12 22:20:57,978 INFO [train.py:421] (5/8) Epoch 8, batch 24600, loss[loss=2.652, over 700.00 frames. , ppl: 14.181297265690743] tot_loss[loss=2.277, over 5462275.42 frames. , ppl: 9.750674713750055], batch size: 70
+2022-12-12 22:22:37,693 INFO [train.py:421] (5/8) Epoch 8, batch 24800, loss[loss=2.327, over 1820.00 frames. , ppl: 10.246756060510462] tot_loss[loss=2.277, over 5495890.00 frames. , ppl: 9.745864099588324], batch size: 70
+2022-12-12 22:24:20,891 INFO [train.py:421] (5/8) Epoch 8, batch 25000, loss[loss=2.517, over 840.00 frames. , ppl: 12.392768230206622] tot_loss[loss=2.277, over 5503053.60 frames. , ppl: 9.747285328610676], batch size: 70
+2022-12-12 22:24:20,892 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 22:24:21,639 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.271, over 211138.00 frames. , ppl: 9.690235616742381
+2022-12-12 22:26:02,850 INFO [train.py:421] (5/8) Epoch 8, batch 25200, loss[loss=2.251, over 4550.00 frames. , ppl: 9.501776628451498] tot_loss[loss=2.277, over 5514587.66 frames. , ppl: 9.747515498917599], batch size: 70
+2022-12-12 22:27:41,827 INFO [train.py:421] (5/8) Epoch 8, batch 25400, loss[loss=2.706, over 770.00 frames. , ppl: 14.974977344132224] tot_loss[loss=2.276, over 5539662.74 frames. , ppl: 9.738418323800941], batch size: 70
+2022-12-12 22:29:21,120 INFO [train.py:421] (5/8) Epoch 8, batch 25600, loss[loss=2.334, over 2170.00 frames. , ppl: 10.31518804404933] tot_loss[loss=2.276, over 5513831.52 frames. , ppl: 9.74092846351868], batch size: 70
+2022-12-12 22:31:04,405 INFO [train.py:421] (5/8) Epoch 8, batch 25800, loss[loss=2.521, over 1820.00 frames. , ppl: 12.439374189118501] tot_loss[loss=2.277, over 5497741.82 frames. , ppl: 9.749178569343693], batch size: 70
+2022-12-12 22:32:42,660 INFO [train.py:421] (5/8) Epoch 8, batch 26000, loss[loss=2.473, over 1680.00 frames. , ppl: 11.852403712697807] tot_loss[loss=2.278, over 5474522.75 frames. , ppl: 9.759465064658697], batch size: 70
+2022-12-12 22:32:42,660 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 22:32:43,419 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.708037800704728
+2022-12-12 22:34:22,026 INFO [train.py:421] (5/8) Epoch 8, batch 26200, loss[loss=2.694, over 770.00 frames. , ppl: 14.795249260383313] tot_loss[loss=2.277, over 5483460.06 frames. , ppl: 9.75163382491949], batch size: 70
+2022-12-12 22:36:01,414 INFO [train.py:421] (5/8) Epoch 8, batch 26400, loss[loss=2.293, over 2310.00 frames. , ppl: 9.909365130203206] tot_loss[loss=2.278, over 5478611.66 frames. , ppl: 9.752342304218681], batch size: 70
+2022-12-12 22:37:43,186 INFO [train.py:421] (5/8) Epoch 8, batch 26600, loss[loss=2.595, over 770.00 frames. , ppl: 13.396008001435666] tot_loss[loss=2.277, over 5492906.55 frames. , ppl: 9.748041657212081], batch size: 70
+2022-12-12 22:39:21,936 INFO [train.py:421] (5/8) Epoch 8, batch 26800, loss[loss=2.353, over 2730.00 frames. , ppl: 10.514120593187632] tot_loss[loss=2.277, over 5480325.63 frames. , ppl: 9.752167098350226], batch size: 70
+2022-12-12 22:40:59,535 INFO [train.py:421] (5/8) Epoch 8, batch 27000, loss[loss=2.237, over 5390.00 frames. , ppl: 9.363425090419446] tot_loss[loss=2.278, over 5459882.15 frames. , ppl: 9.75955827305668], batch size: 70
+2022-12-12 22:40:59,536 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 22:41:00,266 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.710776846115149
+2022-12-12 22:42:41,014 INFO [train.py:421] (5/8) Epoch 8, batch 27200, loss[loss=2.184, over 4550.00 frames. , ppl: 8.883743580812888] tot_loss[loss=2.279, over 5450725.14 frames. , ppl: 9.763126819958451], batch size: 70
+2022-12-12 22:44:23,273 INFO [train.py:421] (5/8) Epoch 8, batch 27400, loss[loss=2.164, over 3920.00 frames. , ppl: 8.705377410962416] tot_loss[loss=2.279, over 5464463.71 frames. , ppl: 9.76260330111134], batch size: 70
+2022-12-12 22:46:01,804 INFO [train.py:421] (5/8) Epoch 8, batch 27600, loss[loss=2.445, over 1470.00 frames. , ppl: 11.528670353913359] tot_loss[loss=2.279, over 5456594.27 frames. , ppl: 9.764000229077539], batch size: 70
+2022-12-12 22:47:40,352 INFO [train.py:421] (5/8) Epoch 8, batch 27800, loss[loss=2.237, over 4060.00 frames. , ppl: 9.36745255559239] tot_loss[loss=2.28, over 5441706.48 frames. , ppl: 9.77304012380441], batch size: 70
+2022-12-12 22:49:20,551 INFO [train.py:421] (5/8) Epoch 8, batch 28000, loss[loss=2.855, over 560.00 frames. , ppl: 17.37999940839776] tot_loss[loss=2.28, over 5443472.99 frames. , ppl: 9.776913754675325], batch size: 70
+2022-12-12 22:49:20,552 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 22:49:21,297 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.709539437680133
+2022-12-12 22:51:00,491 INFO [train.py:421] (5/8) Epoch 8, batch 28200, loss[loss=2.445, over 770.00 frames. , ppl: 11.530124055005622] tot_loss[loss=2.282, over 5386934.72 frames. , ppl: 9.79186257654688], batch size: 70
+2022-12-12 22:52:39,727 INFO [train.py:421] (5/8) Epoch 8, batch 28400, loss[loss=2.68, over 770.00 frames. , ppl: 14.588452871241312] tot_loss[loss=2.282, over 5354794.23 frames. , ppl: 9.798190522931945], batch size: 70
+2022-12-12 22:54:17,923 INFO [train.py:421] (5/8) Epoch 8, batch 28600, loss[loss=2.47, over 840.00 frames. , ppl: 11.817950514961858] tot_loss[loss=2.281, over 5386566.17 frames. , ppl: 9.787361471089971], batch size: 70
+2022-12-12 22:55:59,819 INFO [train.py:421] (5/8) Epoch 8, batch 28800, loss[loss=2.426, over 1750.00 frames. , ppl: 11.311697107552728] tot_loss[loss=2.279, over 5427472.94 frames. , ppl: 9.771152946281273], batch size: 70
+2022-12-12 22:57:42,746 INFO [train.py:421] (5/8) Epoch 8, batch 29000, loss[loss=2.2, over 4830.00 frames. , ppl: 9.026991733371796] tot_loss[loss=2.279, over 5408314.61 frames. , ppl: 9.769019427733397], batch size: 70
+2022-12-12 22:57:42,747 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 22:57:43,506 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.696305732449282
+2022-12-12 22:59:22,325 INFO [train.py:421] (5/8) Epoch 8, batch 29200, loss[loss=2.171, over 2660.00 frames. , ppl: 8.769660175841409] tot_loss[loss=2.278, over 5465811.23 frames. , ppl: 9.757013870698948], batch size: 70
+2022-12-12 23:01:03,745 INFO [train.py:421] (5/8) Epoch 8, batch 29400, loss[loss=2.24, over 4480.00 frames. , ppl: 9.392776818135454] tot_loss[loss=2.277, over 5516117.84 frames. , ppl: 9.748963392407305], batch size: 70
+2022-12-12 23:02:44,759 INFO [train.py:421] (5/8) Epoch 8, batch 29600, loss[loss=2.222, over 2240.00 frames. , ppl: 9.22863731742847] tot_loss[loss=2.276, over 5554792.97 frames. , ppl: 9.735326956226093], batch size: 70
+2022-12-12 23:04:24,107 INFO [train.py:421] (5/8) Epoch 8, batch 29800, loss[loss=2.177, over 4200.00 frames. , ppl: 8.816937364851643] tot_loss[loss=2.276, over 5532743.54 frames. , ppl: 9.738769802766171], batch size: 70
+2022-12-12 23:06:04,137 INFO [train.py:421] (5/8) Epoch 8, batch 30000, loss[loss=2.411, over 1470.00 frames. , ppl: 11.144761354662572] tot_loss[loss=2.276, over 5558525.76 frames. , ppl: 9.734638061867516], batch size: 70
+2022-12-12 23:06:04,137 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 23:06:04,900 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.273, over 211138.00 frames. , ppl: 9.705356986317165
+2022-12-12 23:07:46,652 INFO [train.py:421] (5/8) Epoch 8, batch 30200, loss[loss=2.206, over 5250.00 frames. , ppl: 9.080461342791839] tot_loss[loss=2.276, over 5516911.73 frames. , ppl: 9.74019297451295], batch size: 70
+2022-12-12 23:09:28,063 INFO [train.py:421] (5/8) Epoch 8, batch 30400, loss[loss=2.272, over 3500.00 frames. , ppl: 9.698172832670757] tot_loss[loss=2.276, over 5548448.75 frames. , ppl: 9.737503926548596], batch size: 70
+2022-12-12 23:11:07,294 INFO [train.py:421] (5/8) Epoch 8, batch 30600, loss[loss=2.227, over 4620.00 frames. , ppl: 9.273951214541334] tot_loss[loss=2.277, over 5538089.68 frames. , ppl: 9.743529041734059], batch size: 70
+2022-12-12 23:12:43,210 INFO [train.py:421] (5/8) Epoch 8, batch 30800, loss[loss=2.262, over 2870.00 frames. , ppl: 9.60186755474916] tot_loss[loss=2.278, over 5497909.86 frames. , ppl: 9.754205352436013], batch size: 70
+2022-12-12 23:14:23,185 INFO [train.py:421] (5/8) Epoch 8, batch 31000, loss[loss=2.447, over 1190.00 frames. , ppl: 11.54847875932347] tot_loss[loss=2.279, over 5449324.45 frames. , ppl: 9.767743720006138], batch size: 70
+2022-12-12 23:14:23,186 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 23:14:23,946 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.697824214248042
+2022-12-12 23:16:06,465 INFO [train.py:421] (5/8) Epoch 8, batch 31200, loss[loss=3.227, over 490.00 frames. , ppl: 25.201808467783103] tot_loss[loss=2.279, over 5444734.52 frames. , ppl: 9.769929938034796], batch size: 70
+2022-12-12 23:17:46,227 INFO [train.py:421] (5/8) Epoch 8, batch 31400, loss[loss=2.411, over 2380.00 frames. , ppl: 11.14501689686751] tot_loss[loss=2.279, over 5448496.28 frames. , ppl: 9.771432193636883], batch size: 70
+2022-12-12 23:19:24,625 INFO [train.py:421] (5/8) Epoch 8, batch 31600, loss[loss=2.372, over 1610.00 frames. , ppl: 10.715850134985134] tot_loss[loss=2.279, over 5484294.43 frames. , ppl: 9.762418384393882], batch size: 70
+2022-12-12 23:21:02,586 INFO [train.py:421] (5/8) Epoch 8, batch 31800, loss[loss=2.343, over 2240.00 frames. , ppl: 10.416755490196717] tot_loss[loss=2.279, over 5468080.30 frames. , ppl: 9.766021891575091], batch size: 70
+2022-12-12 23:22:42,623 INFO [train.py:421] (5/8) Epoch 8, batch 32000, loss[loss=2.336, over 1750.00 frames. , ppl: 10.340478320353252] tot_loss[loss=2.278, over 5478570.23 frames. , ppl: 9.753463216097984], batch size: 70
+2022-12-12 23:22:42,624 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 23:22:43,385 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.681927833421199
+2022-12-12 23:24:20,872 INFO [train.py:421] (5/8) Epoch 8, batch 32200, loss[loss=3.253, over 560.00 frames. , ppl: 25.8783084902908] tot_loss[loss=2.279, over 5450851.86 frames. , ppl: 9.770400054508148], batch size: 70
+2022-12-12 23:26:03,836 INFO [train.py:421] (5/8) Epoch 8, batch 32400, loss[loss=2.166, over 6720.00 frames. , ppl: 8.72392244190352] tot_loss[loss=2.278, over 5463622.79 frames. , ppl: 9.759269193683684], batch size: 70
+2022-12-12 23:27:44,843 INFO [train.py:421] (5/8) Epoch 8, batch 32600, loss[loss=2.396, over 1470.00 frames. , ppl: 10.979081446519373] tot_loss[loss=2.277, over 5500571.09 frames. , ppl: 9.750626130645298], batch size: 70
+2022-12-12 23:29:26,644 INFO [train.py:421] (5/8) Epoch 8, batch 32800, loss[loss=2.185, over 2940.00 frames. , ppl: 8.886610736588635] tot_loss[loss=2.276, over 5518767.40 frames. , ppl: 9.741272308677896], batch size: 70
+2022-12-12 23:31:04,174 INFO [train.py:421] (5/8) Epoch 8, batch 33000, loss[loss=2.288, over 2730.00 frames. , ppl: 9.858737903698282] tot_loss[loss=2.276, over 5519478.90 frames. , ppl: 9.737701638361802], batch size: 70
+2022-12-12 23:31:04,175 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 23:31:04,954 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.272, over 211138.00 frames. , ppl: 9.70102126639035
+2022-12-12 23:32:46,880 INFO [train.py:421] (5/8) Epoch 8, batch 33200, loss[loss=3.696, over 420.00 frames. , ppl: 40.28270700463404] tot_loss[loss=2.278, over 5450497.94 frames. , ppl: 9.761276225169603], batch size: 70
+2022-12-12 23:34:29,598 INFO [train.py:421] (5/8) Epoch 8, batch 33400, loss[loss=2.293, over 5250.00 frames. , ppl: 9.903669251521396] tot_loss[loss=2.278, over 5457873.64 frames. , ppl: 9.761127480531586], batch size: 70
+2022-12-12 23:36:09,713 INFO [train.py:421] (5/8) Epoch 8, batch 33600, loss[loss=2.419, over 1400.00 frames. , ppl: 11.239203275430086] tot_loss[loss=2.278, over 5464626.63 frames. , ppl: 9.7598773217277], batch size: 70
+2022-12-12 23:37:51,012 INFO [train.py:421] (5/8) Epoch 8, batch 33800, loss[loss=2.294, over 1680.00 frames. , ppl: 9.909805794316703] tot_loss[loss=2.279, over 5462535.28 frames. , ppl: 9.766635718651605], batch size: 70
+2022-12-12 23:39:28,192 INFO [train.py:421] (5/8) Epoch 8, batch 34000, loss[loss=2.446, over 1400.00 frames. , ppl: 11.537818805787692] tot_loss[loss=2.28, over 5423588.86 frames. , ppl: 9.779186413540655], batch size: 70
+2022-12-12 23:39:28,192 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 23:39:28,937 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.680815891149308
+2022-12-12 23:41:01,892 INFO [train.py:421] (5/8) Epoch 8, batch 34200, loss[loss=2.395, over 1750.00 frames. , ppl: 10.964985190463327] tot_loss[loss=2.279, over 5461314.97 frames. , ppl: 9.764768677791231], batch size: 70
+2022-12-12 23:42:44,182 INFO [train.py:421] (5/8) Epoch 8, batch 34400, loss[loss=2.984, over 560.00 frames. , ppl: 19.775139579025492] tot_loss[loss=2.278, over 5490415.13 frames. , ppl: 9.75851131720673], batch size: 70
+2022-12-12 23:44:28,311 INFO [train.py:421] (5/8) Epoch 8, batch 34600, loss[loss=2.341, over 2520.00 frames. , ppl: 10.390814882371101] tot_loss[loss=2.278, over 5472128.70 frames. , ppl: 9.76189289783924], batch size: 70
+2022-12-12 23:46:06,947 INFO [train.py:421] (5/8) Epoch 8, batch 34800, loss[loss=2.334, over 2520.00 frames. , ppl: 10.321289996062957] tot_loss[loss=2.278, over 5472173.12 frames. , ppl: 9.75617296061622], batch size: 70
+2022-12-12 23:47:51,625 INFO [train.py:421] (5/8) Epoch 8, batch 35000, loss[loss=2.451, over 1470.00 frames. , ppl: 11.601268094257835] tot_loss[loss=2.276, over 5514789.27 frames. , ppl: 9.738138379400795], batch size: 70
+2022-12-12 23:47:51,625 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 23:47:52,385 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.680259967903076
+2022-12-12 23:49:30,413 INFO [train.py:421] (5/8) Epoch 8, batch 35200, loss[loss=3.024, over 560.00 frames. , ppl: 20.57394804285985] tot_loss[loss=2.277, over 5486430.89 frames. , ppl: 9.74297504397669], batch size: 70
+2022-12-12 23:51:13,489 INFO [train.py:421] (5/8) Epoch 8, batch 35400, loss[loss=4.102, over 350.00 frames. , ppl: 60.4718359025039] tot_loss[loss=2.277, over 5479947.00 frames. , ppl: 9.744365122085576], batch size: 70
+2022-12-12 23:52:48,344 INFO [train.py:421] (5/8) Epoch 8, batch 35600, loss[loss=2.193, over 4970.00 frames. , ppl: 8.963073307238798] tot_loss[loss=2.277, over 5472723.81 frames. , ppl: 9.751922417608945], batch size: 70
+2022-12-12 23:54:26,526 INFO [train.py:421] (5/8) Epoch 8, batch 35800, loss[loss=2.128, over 10360.00 frames. , ppl: 8.40156950582037] tot_loss[loss=2.279, over 5437752.99 frames. , ppl: 9.763864365031377], batch size: 70
+2022-12-12 23:56:06,756 INFO [train.py:421] (5/8) Epoch 8, batch 36000, loss[loss=2.265, over 1750.00 frames. , ppl: 9.627437048718912] tot_loss[loss=2.279, over 5408383.11 frames. , ppl: 9.76560677791196], batch size: 70
+2022-12-12 23:56:06,757 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-12 23:56:07,522 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.675932595464811
+2022-12-12 23:57:48,185 INFO [train.py:421] (5/8) Epoch 8, batch 36200, loss[loss=2.338, over 1400.00 frames. , ppl: 10.355359225533112] tot_loss[loss=2.279, over 5391101.51 frames. , ppl: 9.765644216808937], batch size: 70
+2022-12-12 23:59:26,454 INFO [train.py:421] (5/8) Epoch 8, batch 36400, loss[loss=2.412, over 1190.00 frames. , ppl: 11.155040905400945] tot_loss[loss=2.28, over 5378679.25 frames. , ppl: 9.777165977822447], batch size: 70
+2022-12-13 00:01:08,131 INFO [train.py:421] (5/8) Epoch 8, batch 36600, loss[loss=2.192, over 3430.00 frames. , ppl: 8.950681574338526] tot_loss[loss=2.28, over 5396514.07 frames. , ppl: 9.77289545320661], batch size: 70
+2022-12-13 00:02:44,810 INFO [train.py:421] (5/8) Epoch 8, batch 36800, loss[loss=2.392, over 1190.00 frames. , ppl: 10.938448033640292] tot_loss[loss=2.28, over 5373894.68 frames. , ppl: 9.780219575353382], batch size: 70
+2022-12-13 00:04:22,094 INFO [train.py:421] (5/8) Epoch 8, batch 37000, loss[loss=2.433, over 1050.00 frames. , ppl: 11.393385003217585] tot_loss[loss=2.28, over 5364471.21 frames. , ppl: 9.780955491799922], batch size: 70
+2022-12-13 00:04:22,095 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 00:04:22,851 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.678712720778323
+2022-12-13 00:06:04,170 INFO [train.py:421] (5/8) Epoch 8, batch 37200, loss[loss=2.201, over 3010.00 frames. , ppl: 9.03811439856446] tot_loss[loss=2.281, over 5376838.79 frames. , ppl: 9.78273050201692], batch size: 70
+2022-12-13 00:07:45,618 INFO [train.py:421] (5/8) Epoch 8, batch 37400, loss[loss=2.538, over 980.00 frames. , ppl: 12.660034672455065] tot_loss[loss=2.279, over 5426893.81 frames. , ppl: 9.770568740842197], batch size: 70
+2022-12-13 00:09:28,550 INFO [train.py:421] (5/8) Epoch 8, batch 37600, loss[loss=2.483, over 1330.00 frames. , ppl: 11.976648562191741] tot_loss[loss=2.278, over 5483266.83 frames. , ppl: 9.752778135314664], batch size: 70
+2022-12-13 00:11:11,010 INFO [train.py:421] (5/8) Epoch 8, batch 37800, loss[loss=2.32, over 2170.00 frames. , ppl: 10.176573729476535] tot_loss[loss=2.279, over 5417199.64 frames. , ppl: 9.769374735275578], batch size: 70
+2022-12-13 00:12:52,619 INFO [train.py:421] (5/8) Epoch 8, batch 38000, loss[loss=2.215, over 3850.00 frames. , ppl: 9.15922963222307] tot_loss[loss=2.278, over 5447552.79 frames. , ppl: 9.759116832489648], batch size: 70
+2022-12-13 00:12:52,620 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 00:12:53,370 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.666419470350528
+2022-12-13 00:14:30,995 INFO [train.py:421] (5/8) Epoch 8, batch 38200, loss[loss=2.439, over 1120.00 frames. , ppl: 11.459953872310138] tot_loss[loss=2.278, over 5453085.67 frames. , ppl: 9.758805074722124], batch size: 70
+2022-12-13 00:16:13,237 INFO [train.py:421] (5/8) Epoch 8, batch 38400, loss[loss=2.309, over 2240.00 frames. , ppl: 10.05944731675303] tot_loss[loss=2.279, over 5430903.54 frames. , ppl: 9.766699324485343], batch size: 70
+2022-12-13 00:17:51,820 INFO [train.py:421] (5/8) Epoch 8, batch 38600, loss[loss=2.538, over 910.00 frames. , ppl: 12.656718167295098] tot_loss[loss=2.278, over 5461371.99 frames. , ppl: 9.76108388767873], batch size: 70
+2022-12-13 00:19:32,517 INFO [train.py:421] (5/8) Epoch 8, batch 38800, loss[loss=2.413, over 1610.00 frames. , ppl: 11.165399402891138] tot_loss[loss=2.28, over 5434859.60 frames. , ppl: 9.77244051777591], batch size: 70
+2022-12-13 00:21:14,870 INFO [train.py:421] (5/8) Epoch 8, batch 39000, loss[loss=2.623, over 700.00 frames. , ppl: 13.779651990731049] tot_loss[loss=2.279, over 5466512.41 frames. , ppl: 9.766000725651592], batch size: 70
+2022-12-13 00:21:14,870 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 00:21:15,629 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.679752787430928
+2022-12-13 00:22:54,860 INFO [train.py:421] (5/8) Epoch 8, batch 39200, loss[loss=2.252, over 2730.00 frames. , ppl: 9.505839217985935] tot_loss[loss=2.278, over 5481979.01 frames. , ppl: 9.75822954296494], batch size: 70
+2022-12-13 00:24:36,587 INFO [train.py:421] (5/8) Epoch 8, batch 39400, loss[loss=2.693, over 910.00 frames. , ppl: 14.783231300790334] tot_loss[loss=2.278, over 5471727.97 frames. , ppl: 9.754732669430854], batch size: 70
+2022-12-13 00:26:21,911 INFO [train.py:421] (5/8) Epoch 8, batch 39600, loss[loss=2.437, over 1540.00 frames. , ppl: 11.43529698947311] tot_loss[loss=2.276, over 5569097.99 frames. , ppl: 9.733051260912617], batch size: 70
+2022-12-13 00:28:01,293 INFO [train.py:421] (5/8) Epoch 8, batch 39800, loss[loss=2.179, over 9800.00 frames. , ppl: 8.834744651624144] tot_loss[loss=2.275, over 5588484.53 frames. , ppl: 9.73124142506534], batch size: 70
+2022-12-13 00:29:43,006 INFO [train.py:421] (5/8) Epoch 8, batch 40000, loss[loss=2.326, over 1750.00 frames. , ppl: 10.236790489689739] tot_loss[loss=2.275, over 5605572.18 frames. , ppl: 9.727921532734774], batch size: 70
+2022-12-13 00:29:43,007 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 00:29:43,771 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.271, over 211138.00 frames. , ppl: 9.689682020856184
+2022-12-13 00:31:23,406 INFO [train.py:421] (5/8) Epoch 8, batch 40200, loss[loss=2.34, over 1680.00 frames. , ppl: 10.376700831854775] tot_loss[loss=2.276, over 5575445.16 frames. , ppl: 9.739834517897656], batch size: 70
+2022-12-13 00:33:03,356 INFO [train.py:421] (5/8) Epoch 8, batch 40400, loss[loss=3.05, over 490.00 frames. , ppl: 21.10793919111542] tot_loss[loss=2.278, over 5522163.16 frames. , ppl: 9.754606343353066], batch size: 70
+2022-12-13 00:34:47,611 INFO [train.py:421] (5/8) Epoch 8, batch 40600, loss[loss=2.482, over 980.00 frames. , ppl: 11.962009945172268] tot_loss[loss=2.277, over 5544080.67 frames. , ppl: 9.746900584898777], batch size: 70
+2022-12-13 00:36:29,479 INFO [train.py:421] (5/8) Epoch 8, batch 40800, loss[loss=2.37, over 1680.00 frames. , ppl: 10.69208693940919] tot_loss[loss=2.278, over 5501485.22 frames. , ppl: 9.76059936503677], batch size: 70
+2022-12-13 00:38:06,222 INFO [train.py:421] (5/8) Epoch 8, batch 41000, loss[loss=2.34, over 1120.00 frames. , ppl: 10.384742885871152] tot_loss[loss=2.278, over 5535552.16 frames. , ppl: 9.75730828374822], batch size: 70
+2022-12-13 00:38:06,222 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 00:38:06,968 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.67065241409767
+2022-12-13 00:39:49,692 INFO [train.py:421] (5/8) Epoch 8, batch 41200, loss[loss=4.017, over 350.00 frames. , ppl: 55.544860287255595] tot_loss[loss=2.279, over 5500628.04 frames. , ppl: 9.766631831830793], batch size: 70
+2022-12-13 00:41:29,203 INFO [train.py:421] (5/8) Epoch 8, batch 41400, loss[loss=2.337, over 4830.00 frames. , ppl: 10.346685029650125] tot_loss[loss=2.277, over 5544575.52 frames. , ppl: 9.749425234079563], batch size: 70
+2022-12-13 00:43:12,246 INFO [train.py:421] (5/8) Epoch 8, batch 41600, loss[loss=3.65, over 420.00 frames. , ppl: 38.48175635369423] tot_loss[loss=2.276, over 5570019.32 frames. , ppl: 9.736680714628985], batch size: 70
+2022-12-13 00:44:46,363 INFO [train.py:421] (5/8) Epoch 8, batch 41800, loss[loss=2.315, over 2800.00 frames. , ppl: 10.125957640422763] tot_loss[loss=2.278, over 5524887.22 frames. , ppl: 9.753071427014246], batch size: 70
+2022-12-13 00:46:26,562 INFO [train.py:421] (5/8) Epoch 8, batch 42000, loss[loss=2.657, over 840.00 frames. , ppl: 14.248413482302183] tot_loss[loss=2.277, over 5538027.77 frames. , ppl: 9.746226848709005], batch size: 70
+2022-12-13 00:46:26,563 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 00:46:27,308 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.675724941707049
+2022-12-13 00:48:05,872 INFO [train.py:421] (5/8) Epoch 8, batch 42200, loss[loss=2.597, over 770.00 frames. , ppl: 13.426441269726583] tot_loss[loss=2.277, over 5558001.18 frames. , ppl: 9.744477951047383], batch size: 70
+2022-12-13 00:49:47,952 INFO [train.py:421] (5/8) Epoch 8, batch 42400, loss[loss=2.246, over 3010.00 frames. , ppl: 9.44984217573923] tot_loss[loss=2.277, over 5562830.14 frames. , ppl: 9.74618615009268], batch size: 70
+2022-12-13 00:51:28,027 INFO [train.py:421] (5/8) Epoch 8, batch 42600, loss[loss=2.407, over 1400.00 frames. , ppl: 11.100654224870087] tot_loss[loss=2.277, over 5551372.76 frames. , ppl: 9.745297943930082], batch size: 70
+2022-12-13 00:53:10,398 INFO [train.py:421] (5/8) Epoch 8, batch 42800, loss[loss=2.533, over 1120.00 frames. , ppl: 12.597402528604817] tot_loss[loss=2.276, over 5578125.51 frames. , ppl: 9.732902525143588], batch size: 70
+2022-12-13 00:54:50,419 INFO [train.py:421] (5/8) Epoch 8, batch 43000, loss[loss=2.427, over 1400.00 frames. , ppl: 11.324301959804199] tot_loss[loss=2.274, over 5628729.39 frames. , ppl: 9.718511417256748], batch size: 70
+2022-12-13 00:54:50,419 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 00:54:51,176 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.675021816214633
+2022-12-13 00:56:31,399 INFO [train.py:421] (5/8) Epoch 8, batch 43200, loss[loss=2.342, over 2590.00 frames. , ppl: 10.403758543139407] tot_loss[loss=2.275, over 5580628.37 frames. , ppl: 9.726375684591444], batch size: 70
+2022-12-13 00:58:13,083 INFO [train.py:421] (5/8) Epoch 8, batch 43400, loss[loss=2.174, over 3710.00 frames. , ppl: 8.796873959203204] tot_loss[loss=2.277, over 5552623.29 frames. , ppl: 9.744458845644907], batch size: 70
+2022-12-13 00:59:53,237 INFO [train.py:421] (5/8) Epoch 8, batch 43600, loss[loss=2.284, over 2800.00 frames. , ppl: 9.817124351472005] tot_loss[loss=2.277, over 5566975.28 frames. , ppl: 9.744720003082953], batch size: 70
+2022-12-13 01:01:34,005 INFO [train.py:421] (5/8) Epoch 8, batch 43800, loss[loss=2.339, over 1610.00 frames. , ppl: 10.374220467185099] tot_loss[loss=2.278, over 5540927.77 frames. , ppl: 9.75285029961824], batch size: 70
+2022-12-13 01:03:12,893 INFO [train.py:421] (5/8) Epoch 8, batch 44000, loss[loss=2.997, over 560.00 frames. , ppl: 20.02266245516215] tot_loss[loss=2.277, over 5566125.21 frames. , ppl: 9.743564006016001], batch size: 70
+2022-12-13 01:03:12,893 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 01:03:13,638 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.674165532792536
+2022-12-13 01:04:54,833 INFO [train.py:421] (5/8) Epoch 8, batch 44200, loss[loss=2.278, over 2800.00 frames. , ppl: 9.753337308647097] tot_loss[loss=2.277, over 5525855.75 frames. , ppl: 9.750889171858523], batch size: 70
+2022-12-13 01:06:33,909 INFO [train.py:421] (5/8) Epoch 8, batch 44400, loss[loss=2.214, over 4130.00 frames. , ppl: 9.152934692147612] tot_loss[loss=2.278, over 5491072.49 frames. , ppl: 9.753570815127208], batch size: 70
+2022-12-13 01:08:12,907 INFO [train.py:421] (5/8) Epoch 8, batch 44600, loss[loss=2.297, over 3080.00 frames. , ppl: 9.943134296178012] tot_loss[loss=2.278, over 5492850.97 frames. , ppl: 9.757686850760546], batch size: 70
+2022-12-13 01:09:58,344 INFO [train.py:421] (5/8) Epoch 8, batch 44800, loss[loss=2.485, over 1330.00 frames. , ppl: 12.00192480943556] tot_loss[loss=2.278, over 5510588.23 frames. , ppl: 9.75550921798778], batch size: 70
+2022-12-13 01:11:44,374 INFO [train.py:421] (5/8) Epoch 8, batch 45000, loss[loss=2.924, over 560.00 frames. , ppl: 18.60957827214235] tot_loss[loss=2.278, over 5493913.09 frames. , ppl: 9.754432251527913], batch size: 70
+2022-12-13 01:11:44,374 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 01:11:45,133 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.674410382020383
+2022-12-13 01:13:24,902 INFO [train.py:421] (5/8) Epoch 8, batch 45200, loss[loss=2.203, over 8400.00 frames. , ppl: 9.052759799621052] tot_loss[loss=2.278, over 5483294.65 frames. , ppl: 9.754735357471032], batch size: 70
+2022-12-13 01:15:00,537 INFO [train.py:421] (5/8) Epoch 8, batch 45400, loss[loss=2.327, over 3430.00 frames. , ppl: 10.242047713189466] tot_loss[loss=2.278, over 5483270.82 frames. , ppl: 9.757739785269658], batch size: 70
+2022-12-13 01:16:39,284 INFO [train.py:421] (5/8) Epoch 8, batch 45600, loss[loss=2.28, over 3710.00 frames. , ppl: 9.778931932229506] tot_loss[loss=2.279, over 5455514.50 frames. , ppl: 9.763692497065346], batch size: 70
+2022-12-13 01:18:18,207 INFO [train.py:421] (5/8) Epoch 8, batch 45800, loss[loss=2.27, over 4830.00 frames. , ppl: 9.676865986591782] tot_loss[loss=2.278, over 5490233.12 frames. , ppl: 9.75733081772575], batch size: 70
+2022-12-13 01:19:57,220 INFO [train.py:421] (5/8) Epoch 8, batch 46000, loss[loss=2.325, over 1540.00 frames. , ppl: 10.223144718101508] tot_loss[loss=2.279, over 5443038.74 frames. , ppl: 9.767000333795936], batch size: 70
+2022-12-13 01:19:57,221 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 01:19:57,983 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.271, over 211138.00 frames. , ppl: 9.68747512264899
+2022-12-13 01:21:36,866 INFO [train.py:421] (5/8) Epoch 8, batch 46200, loss[loss=2.365, over 1820.00 frames. , ppl: 10.644882981379403] tot_loss[loss=2.278, over 5488801.31 frames. , ppl: 9.757232618985837], batch size: 70
+2022-12-13 01:23:16,620 INFO [train.py:421] (5/8) Epoch 8, batch 46400, loss[loss=2.343, over 2380.00 frames. , ppl: 10.409278045764584] tot_loss[loss=2.276, over 5558341.99 frames. , ppl: 9.735689872285164], batch size: 70
+2022-12-13 01:24:55,150 INFO [train.py:421] (5/8) Epoch 8, batch 46600, loss[loss=2.321, over 2100.00 frames. , ppl: 10.183006801346728] tot_loss[loss=2.275, over 5592895.73 frames. , ppl: 9.729082938615742], batch size: 70
+2022-12-13 01:26:38,480 INFO [train.py:421] (5/8) Epoch 8, batch 46800, loss[loss=2.353, over 2310.00 frames. , ppl: 10.513147036397212] tot_loss[loss=2.275, over 5608486.98 frames. , ppl: 9.723531715819984], batch size: 70
+2022-12-13 01:28:17,302 INFO [train.py:421] (5/8) Epoch 8, batch 47000, loss[loss=2.343, over 1470.00 frames. , ppl: 10.41740146248673] tot_loss[loss=2.275, over 5607199.64 frames. , ppl: 9.73110620626602], batch size: 70
+2022-12-13 01:28:17,303 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 01:28:18,064 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.67573496626884
+2022-12-13 01:29:59,145 INFO [train.py:421] (5/8) Epoch 8, batch 47200, loss[loss=2.291, over 1330.00 frames. , ppl: 9.883386238528521] tot_loss[loss=2.275, over 5628016.17 frames. , ppl: 9.724113925603563], batch size: 70
+2022-12-13 01:31:41,109 INFO [train.py:421] (5/8) Epoch 8, batch 47400, loss[loss=2.308, over 2800.00 frames. , ppl: 10.051573037640434] tot_loss[loss=2.273, over 5653066.26 frames. , ppl: 9.709726653320724], batch size: 70
+2022-12-13 01:33:21,451 INFO [train.py:421] (5/8) Epoch 8, batch 47600, loss[loss=2.445, over 980.00 frames. , ppl: 11.534587353834034] tot_loss[loss=2.274, over 5616666.65 frames. , ppl: 9.714172480559714], batch size: 70
+2022-12-13 01:35:04,432 INFO [train.py:421] (5/8) Epoch 8, batch 47800, loss[loss=2.569, over 1400.00 frames. , ppl: 13.051081762293318] tot_loss[loss=2.273, over 5619658.45 frames. , ppl: 9.71249287072862], batch size: 70
+2022-12-13 01:36:41,927 INFO [train.py:421] (5/8) Epoch 8, batch 48000, loss[loss=2.203, over 4550.00 frames. , ppl: 9.050270853096784] tot_loss[loss=2.273, over 5628524.77 frames. , ppl: 9.710673343974504], batch size: 70
+2022-12-13 01:36:41,928 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 01:36:42,686 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.268, over 211138.00 frames. , ppl: 9.664405251487315
+2022-12-13 01:38:24,558 INFO [train.py:421] (5/8) Epoch 8, batch 48200, loss[loss=2.375, over 2800.00 frames. , ppl: 10.746046039911539] tot_loss[loss=2.273, over 5625222.60 frames. , ppl: 9.708837215845724], batch size: 70
+2022-12-13 01:40:06,222 INFO [train.py:421] (5/8) Epoch 8, batch 48400, loss[loss=2.357, over 1050.00 frames. , ppl: 10.561876660617846] tot_loss[loss=2.274, over 5609256.43 frames. , ppl: 9.718443459946307], batch size: 70
+2022-12-13 01:41:49,537 INFO [train.py:421] (5/8) Epoch 8, batch 48600, loss[loss=2.257, over 2800.00 frames. , ppl: 9.55812122373336] tot_loss[loss=2.274, over 5633898.55 frames. , ppl: 9.716248645258533], batch size: 70
+2022-12-13 01:43:30,069 INFO [train.py:421] (5/8) Epoch 8, batch 48800, loss[loss=2.72, over 630.00 frames. , ppl: 15.183348633890903] tot_loss[loss=2.275, over 5584540.94 frames. , ppl: 9.726713823621465], batch size: 70
+2022-12-13 01:45:08,311 INFO [train.py:421] (5/8) Epoch 8, batch 49000, loss[loss=2.222, over 5390.00 frames. , ppl: 9.224574037194282] tot_loss[loss=2.274, over 5586948.02 frames. , ppl: 9.720401713573287], batch size: 70
+2022-12-13 01:45:08,312 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 01:45:09,059 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.66599885304498
+2022-12-13 01:46:48,511 INFO [train.py:421] (5/8) Epoch 8, batch 49200, loss[loss=2.217, over 5600.00 frames. , ppl: 9.183037659870548] tot_loss[loss=2.276, over 5536433.83 frames. , ppl: 9.741292132165283], batch size: 70
+2022-12-13 01:48:27,304 INFO [train.py:421] (5/8) Epoch 8, batch 49400, loss[loss=2.315, over 3080.00 frames. , ppl: 10.120871410486815] tot_loss[loss=2.276, over 5541268.66 frames. , ppl: 9.741623439855736], batch size: 70
+2022-12-13 01:50:08,163 INFO [train.py:421] (5/8) Epoch 8, batch 49600, loss[loss=2.376, over 1750.00 frames. , ppl: 10.760430444139873] tot_loss[loss=2.277, over 5553366.79 frames. , ppl: 9.7429433349321], batch size: 70
+2022-12-13 01:51:45,763 INFO [train.py:421] (5/8) Epoch 8, batch 49800, loss[loss=2.382, over 2240.00 frames. , ppl: 10.830540799851507] tot_loss[loss=2.276, over 5558535.23 frames. , ppl: 9.74138214384204], batch size: 70
+2022-12-13 01:53:28,037 INFO [train.py:421] (5/8) Epoch 8, batch 50000, loss[loss=2.361, over 1820.00 frames. , ppl: 10.603849063771785] tot_loss[loss=2.276, over 5551101.49 frames. , ppl: 9.737745045127125], batch size: 70
+2022-12-13 01:53:28,038 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 01:53:28,798 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.267, over 211138.00 frames. , ppl: 9.6543676004614
+2022-12-13 01:55:10,390 INFO [train.py:421] (5/8) Epoch 8, batch 50200, loss[loss=2.355, over 1610.00 frames. , ppl: 10.538686899234143] tot_loss[loss=2.276, over 5555314.97 frames. , ppl: 9.73882481876284], batch size: 70
+2022-12-13 01:56:54,450 INFO [train.py:421] (5/8) Epoch 8, batch 50400, loss[loss=2.199, over 5250.00 frames. , ppl: 9.011997396081696] tot_loss[loss=2.275, over 5581079.02 frames. , ppl: 9.732372097896013], batch size: 70
+2022-12-13 01:58:36,243 INFO [train.py:421] (5/8) Epoch 8, batch 50600, loss[loss=3.654, over 420.00 frames. , ppl: 38.61555343056795] tot_loss[loss=2.276, over 5553002.24 frames. , ppl: 9.735249481994009], batch size: 70
+2022-12-13 02:00:17,743 INFO [train.py:421] (5/8) Epoch 8, batch 50800, loss[loss=2.408, over 1680.00 frames. , ppl: 11.111014298304413] tot_loss[loss=2.276, over 5534055.67 frames. , ppl: 9.738468402256094], batch size: 70
+2022-12-13 02:01:59,729 INFO [train.py:421] (5/8) Epoch 8, batch 51000, loss[loss=2.41, over 1540.00 frames. , ppl: 11.131973115225918] tot_loss[loss=2.275, over 5532300.41 frames. , ppl: 9.732620427630597], batch size: 70
+2022-12-13 02:01:59,730 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 02:02:00,491 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.268, over 211138.00 frames. , ppl: 9.66433230115276
+2022-12-13 02:03:40,185 INFO [train.py:421] (5/8) Epoch 8, batch 51200, loss[loss=2.178, over 4970.00 frames. , ppl: 8.828132010546083] tot_loss[loss=2.277, over 5489383.54 frames. , ppl: 9.743066160952718], batch size: 70
+2022-12-13 02:05:20,066 INFO [train.py:421] (5/8) Epoch 8, batch 51400, loss[loss=2.338, over 1610.00 frames. , ppl: 10.3621378645744] tot_loss[loss=2.277, over 5492846.80 frames. , ppl: 9.743665895261772], batch size: 70
+2022-12-13 02:07:01,059 INFO [train.py:421] (5/8) Epoch 8, batch 51600, loss[loss=2.568, over 770.00 frames. , ppl: 13.038848385932447] tot_loss[loss=2.277, over 5493346.44 frames. , ppl: 9.747147779832085], batch size: 70
+2022-12-13 02:08:42,082 INFO [train.py:421] (5/8) Epoch 8, batch 51800, loss[loss=2.416, over 1540.00 frames. , ppl: 11.201442622520103] tot_loss[loss=2.278, over 5460867.77 frames. , ppl: 9.760629245440967], batch size: 70
+2022-12-13 02:10:19,803 INFO [train.py:421] (5/8) Epoch 8, batch 52000, loss[loss=2.225, over 5320.00 frames. , ppl: 9.25727657713872] tot_loss[loss=2.278, over 5465126.61 frames. , ppl: 9.759863955646397], batch size: 70
+2022-12-13 02:10:19,803 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 02:10:20,563 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.669551784931777
+2022-12-13 02:11:58,346 INFO [train.py:421] (5/8) Epoch 8, batch 52200, loss[loss=2.193, over 4550.00 frames. , ppl: 8.963113922557298] tot_loss[loss=2.279, over 5451420.26 frames. , ppl: 9.763492561966661], batch size: 70
+2022-12-13 02:13:38,907 INFO [train.py:421] (5/8) Epoch 8, batch 52400, loss[loss=2.167, over 6230.00 frames. , ppl: 8.735816427949452] tot_loss[loss=2.278, over 5471876.36 frames. , ppl: 9.752712590243314], batch size: 70
+2022-12-13 02:15:21,732 INFO [train.py:421] (5/8) Epoch 8, batch 52600, loss[loss=2.346, over 1260.00 frames. , ppl: 10.441133148567918] tot_loss[loss=2.277, over 5537062.25 frames. , ppl: 9.74409945007787], batch size: 70
+2022-12-13 02:17:02,402 INFO [train.py:421] (5/8) Epoch 8, batch 52800, loss[loss=2.179, over 13090.00 frames. , ppl: 8.83336005651696] tot_loss[loss=2.275, over 5562864.74 frames. , ppl: 9.732518239255631], batch size: 70
+2022-12-13 02:18:40,326 INFO [train.py:421] (5/8) Epoch 8, batch 53000, loss[loss=2.265, over 1890.00 frames. , ppl: 9.62958005055869] tot_loss[loss=2.276, over 5533701.29 frames. , ppl: 9.73549290981634], batch size: 70
+2022-12-13 02:18:40,327 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 02:18:41,091 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.271, over 211138.00 frames. , ppl: 9.68564430928278
+2022-12-13 02:20:19,613 INFO [train.py:421] (5/8) Epoch 8, batch 53200, loss[loss=2.374, over 1470.00 frames. , ppl: 10.742504835880812] tot_loss[loss=2.275, over 5589848.47 frames. , ppl: 9.72695075162579], batch size: 70
+2022-12-13 02:21:58,986 INFO [train.py:421] (5/8) Epoch 8, batch 53400, loss[loss=2.385, over 1120.00 frames. , ppl: 10.859822053240867] tot_loss[loss=2.276, over 5565470.25 frames. , ppl: 9.734722328991563], batch size: 70
+2022-12-13 02:23:40,432 INFO [train.py:421] (5/8) Epoch 8, batch 53600, loss[loss=2.612, over 770.00 frames. , ppl: 13.627857364401414] tot_loss[loss=2.275, over 5579740.01 frames. , ppl: 9.728611069876786], batch size: 70
+2022-12-13 02:25:20,304 INFO [train.py:421] (5/8) Epoch 8, batch 53800, loss[loss=2.44, over 2030.00 frames. , ppl: 11.476284883078396] tot_loss[loss=2.274, over 5599984.52 frames. , ppl: 9.721807533348112], batch size: 70
+2022-12-13 02:26:59,831 INFO [train.py:421] (5/8) Epoch 8, batch 54000, loss[loss=2.359, over 1750.00 frames. , ppl: 10.582394089334628] tot_loss[loss=2.275, over 5592989.31 frames. , ppl: 9.723185328362995], batch size: 70
+2022-12-13 02:26:59,832 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 02:27:00,592 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.268, over 211138.00 frames. , ppl: 9.66121883652427
+2022-12-13 02:28:44,769 INFO [train.py:421] (5/8) Epoch 8, batch 54200, loss[loss=2.255, over 2380.00 frames. , ppl: 9.538306039108992] tot_loss[loss=2.273, over 5611171.64 frames. , ppl: 9.712101072897454], batch size: 70
+2022-12-13 02:30:23,554 INFO [train.py:421] (5/8) Epoch 8, batch 54400, loss[loss=2.199, over 2590.00 frames. , ppl: 9.014373828993255] tot_loss[loss=2.275, over 5572255.82 frames. , ppl: 9.727579128884884], batch size: 70
+2022-12-13 02:32:02,924 INFO [train.py:421] (5/8) Epoch 8, batch 54600, loss[loss=2.276, over 4060.00 frames. , ppl: 9.733140690221497] tot_loss[loss=2.273, over 5625789.09 frames. , ppl: 9.711760315979847], batch size: 70
+2022-12-13 02:33:42,246 INFO [train.py:421] (5/8) Epoch 8, batch 54800, loss[loss=2.165, over 5740.00 frames. , ppl: 8.712087129322256] tot_loss[loss=2.273, over 5624675.04 frames. , ppl: 9.7111767771474], batch size: 70
+2022-12-13 02:35:25,399 INFO [train.py:421] (5/8) Epoch 8, batch 55000, loss[loss=2.218, over 3150.00 frames. , ppl: 9.193083895804142] tot_loss[loss=2.274, over 5603381.22 frames. , ppl: 9.719433113705096], batch size: 70
+2022-12-13 02:35:25,400 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 02:35:26,159 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.668597244433377
+2022-12-13 02:37:08,637 INFO [train.py:421] (5/8) Epoch 8, batch 55200, loss[loss=2.372, over 1960.00 frames. , ppl: 10.714067709787582] tot_loss[loss=2.275, over 5561783.12 frames. , ppl: 9.727391095532946], batch size: 70
+2022-12-13 02:38:47,759 INFO [train.py:421] (5/8) Epoch 8, batch 55400, loss[loss=2.267, over 1750.00 frames. , ppl: 9.653399454054945] tot_loss[loss=2.275, over 5529952.56 frames. , ppl: 9.732153921001684], batch size: 70
+2022-12-13 02:40:28,429 INFO [train.py:421] (5/8) Epoch 8, batch 55600, loss[loss=2.329, over 2590.00 frames. , ppl: 10.265103680611901] tot_loss[loss=2.276, over 5500290.77 frames. , ppl: 9.733935460939643], batch size: 70
+2022-12-13 02:42:10,518 INFO [train.py:421] (5/8) Epoch 8, batch 55800, loss[loss=2.846, over 770.00 frames. , ppl: 17.223015084535643] tot_loss[loss=2.275, over 5531582.64 frames. , ppl: 9.728533049140811], batch size: 70
+2022-12-13 02:43:52,137 INFO [train.py:421] (5/8) Epoch 8, batch 56000, loss[loss=2.243, over 2310.00 frames. , ppl: 9.41757573265274] tot_loss[loss=2.275, over 5558663.44 frames. , ppl: 9.73084687372982], batch size: 70
+2022-12-13 02:43:52,138 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 02:43:52,898 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.676379424179105
+2022-12-13 02:45:35,164 INFO [train.py:421] (5/8) Epoch 8, batch 56200, loss[loss=2.169, over 6160.00 frames. , ppl: 8.748161240954543] tot_loss[loss=2.275, over 5547807.86 frames. , ppl: 9.727673657643283], batch size: 70
+2022-12-13 02:47:16,351 INFO [train.py:421] (5/8) Epoch 8, batch 56400, loss[loss=2.193, over 2940.00 frames. , ppl: 8.963347843872818] tot_loss[loss=2.275, over 5539531.91 frames. , ppl: 9.728678961574055], batch size: 70
+2022-12-13 02:48:53,785 INFO [train.py:421] (5/8) Epoch 8, batch 56600, loss[loss=2.16, over 3570.00 frames. , ppl: 8.670427759001996] tot_loss[loss=2.274, over 5537507.63 frames. , ppl: 9.721839154397761], batch size: 70
+2022-12-13 02:50:34,339 INFO [train.py:421] (5/8) Epoch 8, batch 56800, loss[loss=3.743, over 420.00 frames. , ppl: 42.21314159729585] tot_loss[loss=2.275, over 5530486.04 frames. , ppl: 9.729521301371097], batch size: 70
+2022-12-13 02:52:17,207 INFO [train.py:421] (5/8) Epoch 8, batch 57000, loss[loss=2.529, over 1050.00 frames. , ppl: 12.534769267802696] tot_loss[loss=2.275, over 5514047.62 frames. , ppl: 9.732051756763488], batch size: 70
+2022-12-13 02:52:17,207 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 02:52:17,969 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.268, over 211138.00 frames. , ppl: 9.66042525657999
+2022-12-13 02:54:00,655 INFO [train.py:421] (5/8) Epoch 8, batch 57200, loss[loss=2.3, over 1610.00 frames. , ppl: 9.976514981572132] tot_loss[loss=2.276, over 5495601.55 frames. , ppl: 9.736250138804152], batch size: 70
+2022-12-13 02:55:44,459 INFO [train.py:421] (5/8) Epoch 8, batch 57400, loss[loss=2.311, over 2660.00 frames. , ppl: 10.088067334168569] tot_loss[loss=2.275, over 5536975.08 frames. , ppl: 9.730346178470095], batch size: 70
+2022-12-13 02:57:27,576 INFO [train.py:421] (5/8) Epoch 8, batch 57600, loss[loss=2.382, over 1190.00 frames. , ppl: 10.825806484104817] tot_loss[loss=2.274, over 5596482.75 frames. , ppl: 9.719090077457897], batch size: 70
+2022-12-13 02:59:04,500 INFO [train.py:421] (5/8) Epoch 8, batch 57800, loss[loss=2.197, over 6510.00 frames. , ppl: 8.995767979729354] tot_loss[loss=2.273, over 5633335.33 frames. , ppl: 9.711928730636572], batch size: 70
+2022-12-13 03:00:43,937 INFO [train.py:421] (5/8) Epoch 8, batch 58000, loss[loss=2.143, over 4760.00 frames. , ppl: 8.5279264661396] tot_loss[loss=2.273, over 5636959.58 frames. , ppl: 9.70703465941328], batch size: 70
+2022-12-13 03:00:43,938 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 03:00:44,670 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.27, over 211138.00 frames. , ppl: 9.679271420376821
+2022-12-13 03:02:24,476 INFO [train.py:421] (5/8) Epoch 8, batch 58200, loss[loss=2.186, over 5250.00 frames. , ppl: 8.902457666777618] tot_loss[loss=2.273, over 5622783.15 frames. , ppl: 9.708411449818769], batch size: 70
+2022-12-13 03:04:02,655 INFO [train.py:421] (5/8) Epoch 8, batch 58400, loss[loss=3.141, over 490.00 frames. , ppl: 23.13831048228295] tot_loss[loss=2.274, over 5611965.45 frames. , ppl: 9.715075934721241], batch size: 70
+2022-12-13 03:05:40,980 INFO [train.py:421] (5/8) Epoch 8, batch 58600, loss[loss=2.215, over 3150.00 frames. , ppl: 9.164576502754105] tot_loss[loss=2.274, over 5610048.88 frames. , ppl: 9.720249245809946], batch size: 70
+2022-12-13 03:07:18,316 INFO [train.py:421] (5/8) Epoch 8, batch 58800, loss[loss=2.288, over 2240.00 frames. , ppl: 9.850697734291938] tot_loss[loss=2.275, over 5584704.16 frames. , ppl: 9.727980567978234], batch size: 70
+2022-12-13 03:09:03,624 INFO [train.py:421] (5/8) Epoch 8, batch 59000, loss[loss=2.439, over 1400.00 frames. , ppl: 11.461118532512812] tot_loss[loss=2.275, over 5588988.68 frames. , ppl: 9.729596509056801], batch size: 70
+2022-12-13 03:09:03,625 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 03:09:04,370 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.666656969839083
+2022-12-13 03:10:48,155 INFO [train.py:421] (5/8) Epoch 8, batch 59200, loss[loss=2.237, over 2730.00 frames. , ppl: 9.368590306275447] tot_loss[loss=2.274, over 5640099.22 frames. , ppl: 9.716553214942865], batch size: 70
+2022-12-13 03:12:28,716 INFO [train.py:421] (5/8) Epoch 8, batch 59400, loss[loss=2.473, over 1540.00 frames. , ppl: 11.856914160189222] tot_loss[loss=2.274, over 5631357.14 frames. , ppl: 9.722850632932628], batch size: 70
+2022-12-13 03:14:12,138 INFO [train.py:421] (5/8) Epoch 8, batch 59600, loss[loss=2.31, over 2940.00 frames. , ppl: 10.075476806276752] tot_loss[loss=2.274, over 5621197.28 frames. , ppl: 9.719168758608442], batch size: 70
+2022-12-13 03:15:48,403 INFO [train.py:421] (5/8) Epoch 8, batch 59800, loss[loss=2.19, over 3640.00 frames. , ppl: 8.932167037419479] tot_loss[loss=2.275, over 5606191.06 frames. , ppl: 9.725754451335868], batch size: 70
+2022-12-13 03:17:30,952 INFO [train.py:421] (5/8) Epoch 8, batch 60000, loss[loss=2.212, over 5180.00 frames. , ppl: 9.136861462623253] tot_loss[loss=2.273, over 5632829.36 frames. , ppl: 9.709171328183169], batch size: 70
+2022-12-13 03:17:30,952 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 03:17:31,713 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.674444747320207
+2022-12-13 03:19:10,631 INFO [train.py:421] (5/8) Epoch 8, batch 60200, loss[loss=2.236, over 3780.00 frames. , ppl: 9.35421389992322] tot_loss[loss=2.274, over 5610056.23 frames. , ppl: 9.718774202200528], batch size: 70
+2022-12-13 03:20:51,023 INFO [train.py:421] (5/8) Epoch 8, batch 60400, loss[loss=2.263, over 4060.00 frames. , ppl: 9.615792587452292] tot_loss[loss=2.276, over 5557243.45 frames. , ppl: 9.736876949195828], batch size: 70
+2022-12-13 03:22:27,997 INFO [train.py:421] (5/8) Epoch 8, batch 60600, loss[loss=2.115, over 2940.00 frames. , ppl: 8.292917072448228] tot_loss[loss=2.275, over 5558467.43 frames. , ppl: 9.732097544926093], batch size: 70
+2022-12-13 03:24:04,437 INFO [train.py:421] (5/8) Epoch 8, batch 60800, loss[loss=2.36, over 3080.00 frames. , ppl: 10.591582108855256] tot_loss[loss=2.275, over 5555737.76 frames. , ppl: 9.727447382513853], batch size: 70
+2022-12-13 03:25:46,091 INFO [train.py:421] (5/8) Epoch 8, batch 61000, loss[loss=2.193, over 5040.00 frames. , ppl: 8.960586077031515] tot_loss[loss=2.275, over 5541247.04 frames. , ppl: 9.728629520056764], batch size: 70
+2022-12-13 03:25:46,091 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 03:25:46,839 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.268, over 211138.00 frames. , ppl: 9.66320950762563
+2022-12-13 03:27:25,712 INFO [train.py:421] (5/8) Epoch 8, batch 61200, loss[loss=2.253, over 3990.00 frames. , ppl: 9.51314622700841] tot_loss[loss=2.275, over 5523206.82 frames. , ppl: 9.732423881200733], batch size: 70
+2022-12-13 03:29:02,367 INFO [train.py:421] (5/8) Epoch 8, batch 61400, loss[loss=2.336, over 2940.00 frames. , ppl: 10.343428354827894] tot_loss[loss=2.276, over 5484183.31 frames. , ppl: 9.742433841787202], batch size: 70
+2022-12-13 03:30:39,175 INFO [train.py:421] (5/8) Epoch 8, batch 61600, loss[loss=2.222, over 3150.00 frames. , ppl: 9.229940175482994] tot_loss[loss=2.278, over 5424453.71 frames. , ppl: 9.757434193611989], batch size: 70
+2022-12-13 03:32:18,661 INFO [train.py:421] (5/8) Epoch 8, batch 61800, loss[loss=2.153, over 6720.00 frames. , ppl: 8.610784483560732] tot_loss[loss=2.277, over 5443607.74 frames. , ppl: 9.749424248782558], batch size: 70
+2022-12-13 03:33:59,898 INFO [train.py:421] (5/8) Epoch 8, batch 62000, loss[loss=2.262, over 3640.00 frames. , ppl: 9.597668631191569] tot_loss[loss=2.277, over 5481716.64 frames. , ppl: 9.747403649860681], batch size: 70
+2022-12-13 03:33:59,899 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 03:34:00,657 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.267, over 211138.00 frames. , ppl: 9.655050647650823
+2022-12-13 03:35:43,438 INFO [train.py:421] (5/8) Epoch 8, batch 62200, loss[loss=2.454, over 1400.00 frames. , ppl: 11.6361755431483] tot_loss[loss=2.278, over 5463669.39 frames. , ppl: 9.752687675575151], batch size: 70
+2022-12-13 03:37:22,806 INFO [train.py:421] (5/8) Epoch 8, batch 62400, loss[loss=2.224, over 4410.00 frames. , ppl: 9.239698427910785] tot_loss[loss=2.278, over 5446813.67 frames. , ppl: 9.759602287980751], batch size: 70
+2022-12-13 03:39:04,712 INFO [train.py:421] (5/8) Epoch 8, batch 62600, loss[loss=2.517, over 1680.00 frames. , ppl: 12.394763831712828] tot_loss[loss=2.279, over 5419066.47 frames. , ppl: 9.76905164305386], batch size: 70
+2022-12-13 03:40:42,483 INFO [train.py:421] (5/8) Epoch 8, batch 62800, loss[loss=2.417, over 1330.00 frames. , ppl: 11.207444554127763] tot_loss[loss=2.279, over 5418696.09 frames. , ppl: 9.771159562095269], batch size: 70
+2022-12-13 03:42:19,090 INFO [train.py:421] (5/8) Epoch 8, batch 63000, loss[loss=2.273, over 3920.00 frames. , ppl: 9.7131520387488] tot_loss[loss=2.28, over 5411968.87 frames. , ppl: 9.772657823108984], batch size: 70
+2022-12-13 03:42:19,090 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 03:42:19,855 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.268, over 211138.00 frames. , ppl: 9.660179331522313
+2022-12-13 03:44:04,024 INFO [train.py:421] (5/8) Epoch 8, batch 63200, loss[loss=2.358, over 1400.00 frames. , ppl: 10.5654209651877] tot_loss[loss=2.278, over 5480652.77 frames. , ppl: 9.759654728844282], batch size: 70
+2022-12-13 03:45:42,069 INFO [train.py:421] (5/8) Epoch 8, batch 63400, loss[loss=2.698, over 770.00 frames. , ppl: 14.85670285066846] tot_loss[loss=2.278, over 5467925.14 frames. , ppl: 9.761209775115127], batch size: 70
+2022-12-13 03:47:24,377 INFO [train.py:421] (5/8) Epoch 8, batch 63600, loss[loss=2.295, over 3430.00 frames. , ppl: 9.921341324967702] tot_loss[loss=2.278, over 5491089.47 frames. , ppl: 9.755191082930116], batch size: 70
+2022-12-13 03:49:06,245 INFO [train.py:421] (5/8) Epoch 8, batch 63800, loss[loss=2.185, over 8750.00 frames. , ppl: 8.888753538981428] tot_loss[loss=2.278, over 5494936.37 frames. , ppl: 9.754918910927984], batch size: 70
+2022-12-13 03:50:46,485 INFO [train.py:421] (5/8) Epoch 8, batch 64000, loss[loss=2.336, over 2170.00 frames. , ppl: 10.339206500380959] tot_loss[loss=2.276, over 5527972.86 frames. , ppl: 9.740759292798268], batch size: 70
+2022-12-13 03:50:46,485 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 03:50:47,232 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.672815393681503
+2022-12-13 03:52:29,193 INFO [train.py:421] (5/8) Epoch 8, batch 64200, loss[loss=2.194, over 3710.00 frames. , ppl: 8.967291165130433] tot_loss[loss=2.277, over 5538262.88 frames. , ppl: 9.744155459883952], batch size: 70
+2022-12-13 03:54:10,489 INFO [train.py:421] (5/8) Epoch 8, batch 64400, loss[loss=2.467, over 1680.00 frames. , ppl: 11.781816692040534] tot_loss[loss=2.277, over 5544631.50 frames. , ppl: 9.750987314764386], batch size: 70
+2022-12-13 03:55:52,715 INFO [train.py:421] (5/8) Epoch 8, batch 64600, loss[loss=2.558, over 1120.00 frames. , ppl: 12.914508292546419] tot_loss[loss=2.278, over 5533130.83 frames. , ppl: 9.761737263250001], batch size: 70
+2022-12-13 03:57:33,013 INFO [train.py:421] (5/8) Epoch 8, batch 64800, loss[loss=2.253, over 2240.00 frames. , ppl: 9.513150301650448] tot_loss[loss=2.278, over 5552802.82 frames. , ppl: 9.759481543859057], batch size: 70
+2022-12-13 03:59:15,150 INFO [train.py:421] (5/8) Epoch 8, batch 65000, loss[loss=2.444, over 1540.00 frames. , ppl: 11.520918752755392] tot_loss[loss=2.277, over 5559266.65 frames. , ppl: 9.751559772216869], batch size: 70
+2022-12-13 03:59:15,151 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 03:59:15,897 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.268, over 211138.00 frames. , ppl: 9.65988194214127
+2022-12-13 04:00:56,719 INFO [train.py:421] (5/8) Epoch 8, batch 65200, loss[loss=3.581, over 420.00 frames. , ppl: 35.91260694228104] tot_loss[loss=2.278, over 5545056.61 frames. , ppl: 9.75451875389162], batch size: 70
+2022-12-13 04:02:38,086 INFO [train.py:421] (5/8) Epoch 8, batch 65400, loss[loss=2.411, over 1400.00 frames. , ppl: 11.14011386094005] tot_loss[loss=2.277, over 5573717.76 frames. , ppl: 9.74794662621502], batch size: 70
+2022-12-13 04:04:12,154 INFO [train.py:421] (5/8) Epoch 8, batch 65600, loss[loss=2.358, over 1400.00 frames. , ppl: 10.566882139533345] tot_loss[loss=2.278, over 5529383.62 frames. , ppl: 9.753201702866429], batch size: 70
+2022-12-13 04:05:51,083 INFO [train.py:421] (5/8) Epoch 8, batch 65800, loss[loss=2.385, over 910.00 frames. , ppl: 10.854083162879336] tot_loss[loss=2.278, over 5532831.18 frames. , ppl: 9.752472249337666], batch size: 70
+2022-12-13 04:07:30,778 INFO [train.py:421] (5/8) Epoch 8, batch 66000, loss[loss=2.397, over 2520.00 frames. , ppl: 10.989506678772612] tot_loss[loss=2.278, over 5499194.92 frames. , ppl: 9.758144928360668], batch size: 70
+2022-12-13 04:07:30,778 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 04:07:31,525 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.665475252994222
+2022-12-13 04:09:13,156 INFO [train.py:421] (5/8) Epoch 8, batch 66200, loss[loss=2.381, over 840.00 frames. , ppl: 10.81241976885461] tot_loss[loss=2.279, over 5469477.06 frames. , ppl: 9.766388274737674], batch size: 70
+2022-12-13 04:10:50,316 INFO [train.py:421] (5/8) Epoch 8, batch 66400, loss[loss=2.278, over 2170.00 frames. , ppl: 9.75481638959288] tot_loss[loss=2.279, over 5469376.32 frames. , ppl: 9.7620616586882], batch size: 70
+2022-12-13 04:12:32,883 INFO [train.py:421] (5/8) Epoch 8, batch 66600, loss[loss=2.18, over 6230.00 frames. , ppl: 8.848850336883137] tot_loss[loss=2.278, over 5501015.72 frames. , ppl: 9.758704016154667], batch size: 70
+2022-12-13 04:14:11,381 INFO [train.py:421] (5/8) Epoch 8, batch 66800, loss[loss=4.129, over 350.00 frames. , ppl: 62.11736193475308] tot_loss[loss=2.278, over 5486655.30 frames. , ppl: 9.760301424639925], batch size: 70
+2022-12-13 04:15:48,896 INFO [train.py:421] (5/8) Epoch 8, batch 67000, loss[loss=2.358, over 2310.00 frames. , ppl: 10.565705043332585] tot_loss[loss=2.278, over 5504556.31 frames. , ppl: 9.757109636437091], batch size: 70
+2022-12-13 04:15:48,896 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 04:15:49,642 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.267, over 211138.00 frames. , ppl: 9.651940173203029
+2022-12-13 04:17:32,965 INFO [train.py:421] (5/8) Epoch 8, batch 67200, loss[loss=2.228, over 3850.00 frames. , ppl: 9.282511094484317] tot_loss[loss=2.278, over 5512151.82 frames. , ppl: 9.753252170026801], batch size: 70
+2022-12-13 04:19:12,960 INFO [train.py:421] (5/8) Epoch 8, batch 67400, loss[loss=2.228, over 5880.00 frames. , ppl: 9.283137393372849] tot_loss[loss=2.278, over 5510598.24 frames. , ppl: 9.753954199375428], batch size: 70
+2022-12-13 04:20:51,765 INFO [train.py:421] (5/8) Epoch 8, batch 67600, loss[loss=2.397, over 1890.00 frames. , ppl: 10.99453594190738] tot_loss[loss=2.278, over 5517219.19 frames. , ppl: 9.752804821065325], batch size: 70
+2022-12-13 04:22:29,390 INFO [train.py:421] (5/8) Epoch 8, batch 67800, loss[loss=2.388, over 1680.00 frames. , ppl: 10.895612221862342] tot_loss[loss=2.278, over 5515375.30 frames. , ppl: 9.756242227725942], batch size: 70
+2022-12-13 04:24:11,409 INFO [train.py:421] (5/8) Epoch 8, batch 68000, loss[loss=2.462, over 1610.00 frames. , ppl: 11.724062864278785] tot_loss[loss=2.278, over 5502708.00 frames. , ppl: 9.758159714716752], batch size: 70
+2022-12-13 04:24:11,410 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 04:24:12,154 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.269, over 211138.00 frames. , ppl: 9.674181283141165
+2022-12-13 04:25:51,983 INFO [train.py:421] (5/8) Epoch 8, batch 68200, loss[loss=2.389, over 1610.00 frames. , ppl: 10.897924944723483] tot_loss[loss=2.279, over 5486582.74 frames. , ppl: 9.763217463279533], batch size: 70
+2022-12-13 04:27:29,340 INFO [train.py:421] (5/8) Epoch 8, batch 68400, loss[loss=2.514, over 1050.00 frames. , ppl: 12.349902378676564] tot_loss[loss=2.279, over 5468396.89 frames. , ppl: 9.766597216713956], batch size: 70
+2022-12-13 04:29:12,441 INFO [train.py:421] (5/8) Epoch 8, batch 68600, loss[loss=2.12, over 5250.00 frames. , ppl: 8.328314429549422] tot_loss[loss=2.28, over 5460756.18 frames. , ppl: 9.773234520816262], batch size: 70
+2022-12-13 04:30:48,901 INFO [train.py:421] (5/8) Epoch 8, batch 68800, loss[loss=2.428, over 3010.00 frames. , ppl: 11.33688687190228] tot_loss[loss=2.281, over 5433993.81 frames. , ppl: 9.78483739586965], batch size: 70
+2022-12-13 04:32:30,672 INFO [train.py:421] (5/8) Epoch 8, batch 69000, loss[loss=3.07, over 560.00 frames. , ppl: 21.549777028616884] tot_loss[loss=2.28, over 5441038.97 frames. , ppl: 9.771867720524849], batch size: 70
+2022-12-13 04:32:30,673 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 04:32:31,419 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.265, over 211138.00 frames. , ppl: 9.629115969513046
+2022-12-13 04:34:10,754 INFO [train.py:421] (5/8) Epoch 8, batch 69200, loss[loss=2.315, over 2730.00 frames. , ppl: 10.120306851166712] tot_loss[loss=2.279, over 5456394.29 frames. , ppl: 9.77017602099737], batch size: 70
+2022-12-13 04:35:51,617 INFO [train.py:421] (5/8) Epoch 8, batch 69400, loss[loss=2.171, over 6020.00 frames. , ppl: 8.765573329127882] tot_loss[loss=2.279, over 5447640.65 frames. , ppl: 9.771054997825797], batch size: 70
+2022-12-13 04:37:25,625 INFO [train.py:421] (5/8) Epoch 8, batch 69600, loss[loss=2.438, over 1260.00 frames. , ppl: 11.454527957668677] tot_loss[loss=2.279, over 5428073.08 frames. , ppl: 9.767059130882618], batch size: 70
+2022-12-13 04:39:03,668 INFO [train.py:421] (5/8) Epoch 8, batch 69800, loss[loss=2.267, over 1960.00 frames. , ppl: 9.65482267351065] tot_loss[loss=2.279, over 5441366.42 frames. , ppl: 9.762500909812175], batch size: 70
+2022-12-13 04:40:44,763 INFO [train.py:421] (5/8) Epoch 8, batch 70000, loss[loss=2.159, over 2730.00 frames. , ppl: 8.666093431804596] tot_loss[loss=2.278, over 5451724.91 frames. , ppl: 9.755805770631927], batch size: 70
+2022-12-13 04:40:44,764 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 04:40:45,525 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.267, over 211138.00 frames. , ppl: 9.655207841082497
+2022-12-13 04:42:28,625 INFO [train.py:421] (5/8) Epoch 8, batch 70200, loss[loss=2.296, over 3150.00 frames. , ppl: 9.930469857424121] tot_loss[loss=2.279, over 5449301.23 frames. , ppl: 9.76455841339103], batch size: 70
+2022-12-13 04:44:05,974 INFO [train.py:421] (5/8) Epoch 8, batch 70400, loss[loss=2.252, over 4480.00 frames. , ppl: 9.50546826467038] tot_loss[loss=2.279, over 5434842.17 frames. , ppl: 9.76989076515605], batch size: 70
+2022-12-13 04:45:49,704 INFO [train.py:421] (5/8) Epoch 8, batch 70600, loss[loss=2.196, over 3500.00 frames. , ppl: 8.98886265392706] tot_loss[loss=2.277, over 5504249.82 frames. , ppl: 9.751366575126346], batch size: 70
+2022-12-13 04:47:27,338 INFO [train.py:421] (5/8) Epoch 8, batch 70800, loss[loss=2.311, over 1470.00 frames. , ppl: 10.0828137953922] tot_loss[loss=2.277, over 5503251.31 frames. , ppl: 9.74827193619471], batch size: 70
+2022-12-13 04:49:05,280 INFO [train.py:421] (5/8) Epoch 8, batch 71000, loss[loss=2.277, over 3010.00 frames. , ppl: 9.743162377708646] tot_loss[loss=2.278, over 5454760.71 frames. , ppl: 9.756921210377069], batch size: 70
+2022-12-13 04:49:05,281 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 04:49:06,043 INFO [train.py:452] (5/8) Epoch 8, validation: loss=2.268, over 211138.00 frames. , ppl: 9.664858700229336
+2022-12-13 04:50:47,506 INFO [train.py:421] (5/8) Epoch 8, batch 71200, loss[loss=2.141, over 6300.00 frames. , ppl: 8.511527668375756] tot_loss[loss=2.278, over 5468021.19 frames. , ppl: 9.753963407750133], batch size: 70
+2022-12-13 04:52:24,910 INFO [train.py:421] (5/8) Epoch 8, batch 71400, loss[loss=2.168, over 5040.00 frames. , ppl: 8.744263922766693] tot_loss[loss=2.278, over 5447051.23 frames. , ppl: 9.756047903938603], batch size: 70
+2022-12-13 04:54:03,433 INFO [train.py:421] (5/8) Epoch 8, batch 71600, loss[loss=2.339, over 2310.00 frames. , ppl: 10.37056606789151] tot_loss[loss=2.278, over 5454828.00 frames. , ppl: 9.754322482630105], batch size: 70
+2022-12-13 04:55:49,181 INFO [train.py:421] (5/8) Epoch 8, batch 71800, loss[loss=3.012, over 630.00 frames. , ppl: 20.326217263198114] tot_loss[loss=2.279, over 5444888.92 frames. , ppl: 9.766105609298274], batch size: 70
+2022-12-13 04:57:06,870 INFO [train.py:421] (5/8) Epoch 9, batch 0, loss[loss=2.257, over 3290.00 frames. , ppl: 9.556362091676226] tot_loss[loss=2.257, over 3290.00 frames. , ppl: 9.556362091676226], batch size: 70
+2022-12-13 04:58:47,672 INFO [train.py:421] (5/8) Epoch 9, batch 200, loss[loss=2.452, over 1260.00 frames. , ppl: 11.612323019756996] tot_loss[loss=2.277, over 503416.56 frames. , ppl: 9.75070177664635], batch size: 70
+2022-12-13 05:00:29,924 INFO [train.py:421] (5/8) Epoch 9, batch 400, loss[loss=2.749, over 910.00 frames. , ppl: 15.619579597131644] tot_loss[loss=2.274, over 987547.88 frames. , ppl: 9.716043528190019], batch size: 70
+2022-12-13 05:02:09,652 INFO [train.py:421] (5/8) Epoch 9, batch 600, loss[loss=2.107, over 6720.00 frames. , ppl: 8.221680361364829] tot_loss[loss=2.27, over 1431012.86 frames. , ppl: 9.681246184127195], batch size: 70
+2022-12-13 05:03:48,157 INFO [train.py:421] (5/8) Epoch 9, batch 800, loss[loss=3.048, over 560.00 frames. , ppl: 21.07129580079409] tot_loss[loss=2.272, over 1803530.07 frames. , ppl: 9.698457769588833], batch size: 70
+2022-12-13 05:05:26,179 INFO [train.py:421] (5/8) Epoch 9, batch 1000, loss[loss=2.418, over 1820.00 frames. , ppl: 11.220639845966785] tot_loss[loss=2.275, over 2102920.19 frames. , ppl: 9.726582005000534], batch size: 70
+2022-12-13 05:05:26,179 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 05:05:26,939 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.664010468014997
+2022-12-13 05:07:07,627 INFO [train.py:421] (5/8) Epoch 9, batch 1200, loss[loss=2.238, over 7630.00 frames. , ppl: 9.370352203624858] tot_loss[loss=2.272, over 2454609.33 frames. , ppl: 9.700115499502772], batch size: 70
+2022-12-13 05:08:50,658 INFO [train.py:421] (5/8) Epoch 9, batch 1400, loss[loss=2.129, over 3430.00 frames. , ppl: 8.408688837593573] tot_loss[loss=2.27, over 2763825.93 frames. , ppl: 9.675314424170944], batch size: 70
+2022-12-13 05:10:29,015 INFO [train.py:421] (5/8) Epoch 9, batch 1600, loss[loss=2.586, over 910.00 frames. , ppl: 13.272752008584183] tot_loss[loss=2.266, over 3084547.57 frames. , ppl: 9.641738011201758], batch size: 70
+2022-12-13 05:12:07,749 INFO [train.py:421] (5/8) Epoch 9, batch 1800, loss[loss=2.168, over 9660.00 frames. , ppl: 8.741367206970262] tot_loss[loss=2.264, over 3351040.54 frames. , ppl: 9.617139205211023], batch size: 70
+2022-12-13 05:13:46,772 INFO [train.py:421] (5/8) Epoch 9, batch 2000, loss[loss=2.24, over 2800.00 frames. , ppl: 9.394261755568452] tot_loss[loss=2.267, over 3520297.02 frames. , ppl: 9.646300617523158], batch size: 70
+2022-12-13 05:13:46,772 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 05:13:47,536 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.269, over 211138.00 frames. , ppl: 9.667164894955036
+2022-12-13 05:15:30,256 INFO [train.py:421] (5/8) Epoch 9, batch 2200, loss[loss=2.657, over 1120.00 frames. , ppl: 14.247057300505434] tot_loss[loss=2.266, over 3738507.53 frames. , ppl: 9.644089797715978], batch size: 70
+2022-12-13 05:17:11,207 INFO [train.py:421] (5/8) Epoch 9, batch 2400, loss[loss=2.454, over 1120.00 frames. , ppl: 11.633859455600927] tot_loss[loss=2.266, over 3911305.51 frames. , ppl: 9.642124735976747], batch size: 70
+2022-12-13 05:18:52,166 INFO [train.py:421] (5/8) Epoch 9, batch 2600, loss[loss=2.215, over 4060.00 frames. , ppl: 9.160859105096923] tot_loss[loss=2.269, over 4022842.14 frames. , ppl: 9.666455720661935], batch size: 70
+2022-12-13 05:20:30,385 INFO [train.py:421] (5/8) Epoch 9, batch 2800, loss[loss=2.423, over 1260.00 frames. , ppl: 11.275919406108509] tot_loss[loss=2.269, over 4161129.56 frames. , ppl: 9.672341226756567], batch size: 70
+2022-12-13 05:22:11,118 INFO [train.py:421] (5/8) Epoch 9, batch 3000, loss[loss=2.341, over 1820.00 frames. , ppl: 10.395622323389391] tot_loss[loss=2.269, over 4290734.67 frames. , ppl: 9.670635811705658], batch size: 70
+2022-12-13 05:22:11,118 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 05:22:11,878 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.65743883465414
+2022-12-13 05:23:52,012 INFO [train.py:421] (5/8) Epoch 9, batch 3200, loss[loss=2.474, over 1890.00 frames. , ppl: 11.873115236177592] tot_loss[loss=2.271, over 4339254.14 frames. , ppl: 9.689376470345103], batch size: 70
+2022-12-13 05:25:30,714 INFO [train.py:421] (5/8) Epoch 9, batch 3400, loss[loss=2.357, over 1680.00 frames. , ppl: 10.557395788263765] tot_loss[loss=2.27, over 4473044.37 frames. , ppl: 9.676210495316564], batch size: 70
+2022-12-13 05:27:10,001 INFO [train.py:421] (5/8) Epoch 9, batch 3600, loss[loss=2.193, over 4340.00 frames. , ppl: 8.9584280976462] tot_loss[loss=2.271, over 4540480.16 frames. , ppl: 9.689413005837192], batch size: 70
+2022-12-13 05:28:50,469 INFO [train.py:421] (5/8) Epoch 9, batch 3800, loss[loss=2.332, over 1960.00 frames. , ppl: 10.29794700191052] tot_loss[loss=2.27, over 4654126.07 frames. , ppl: 9.679679892766755], batch size: 70
+2022-12-13 05:30:30,351 INFO [train.py:421] (5/8) Epoch 9, batch 4000, loss[loss=2.58, over 980.00 frames. , ppl: 13.197314382187086] tot_loss[loss=2.271, over 4712061.28 frames. , ppl: 9.690561558958612], batch size: 70
+2022-12-13 05:30:30,351 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 05:30:31,100 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.269, over 211138.00 frames. , ppl: 9.671714518344173
+2022-12-13 05:32:07,770 INFO [train.py:421] (5/8) Epoch 9, batch 4200, loss[loss=2.164, over 5250.00 frames. , ppl: 8.704716066133185] tot_loss[loss=2.272, over 4763984.08 frames. , ppl: 9.701232330013395], batch size: 70
+2022-12-13 05:33:45,392 INFO [train.py:421] (5/8) Epoch 9, batch 4400, loss[loss=3.185, over 490.00 frames. , ppl: 24.1609446992232] tot_loss[loss=2.273, over 4801495.83 frames. , ppl: 9.708307521700712], batch size: 70
+2022-12-13 05:35:25,480 INFO [train.py:421] (5/8) Epoch 9, batch 4600, loss[loss=2.288, over 2520.00 frames. , ppl: 9.85262522746154] tot_loss[loss=2.273, over 4858395.65 frames. , ppl: 9.704256435792432], batch size: 70
+2022-12-13 05:37:03,949 INFO [train.py:421] (5/8) Epoch 9, batch 4800, loss[loss=2.323, over 2100.00 frames. , ppl: 10.20891393372886] tot_loss[loss=2.273, over 4888474.97 frames. , ppl: 9.712519112644141], batch size: 70
+2022-12-13 05:38:46,323 INFO [train.py:421] (5/8) Epoch 9, batch 5000, loss[loss=2.2, over 7070.00 frames. , ppl: 9.020879461334111] tot_loss[loss=2.272, over 4966725.16 frames. , ppl: 9.697071363736539], batch size: 70
+2022-12-13 05:38:46,324 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 05:38:47,053 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.655562250208732
+2022-12-13 05:40:30,491 INFO [train.py:421] (5/8) Epoch 9, batch 5200, loss[loss=2.353, over 2170.00 frames. , ppl: 10.5186889690425] tot_loss[loss=2.269, over 5096851.19 frames. , ppl: 9.667254005239645], batch size: 70
+2022-12-13 05:42:13,825 INFO [train.py:421] (5/8) Epoch 9, batch 5400, loss[loss=2.814, over 630.00 frames. , ppl: 16.672263552281343] tot_loss[loss=2.268, over 5166130.31 frames. , ppl: 9.658376151134815], batch size: 70
+2022-12-13 05:43:55,319 INFO [train.py:421] (5/8) Epoch 9, batch 5600, loss[loss=2.704, over 630.00 frames. , ppl: 14.936666917448836] tot_loss[loss=2.268, over 5176700.68 frames. , ppl: 9.656865475886631], batch size: 70
+2022-12-13 05:45:33,975 INFO [train.py:421] (5/8) Epoch 9, batch 5800, loss[loss=2.541, over 700.00 frames. , ppl: 12.693801512128074] tot_loss[loss=2.267, over 5209265.21 frames. , ppl: 9.65461633597556], batch size: 70
+2022-12-13 05:47:17,486 INFO [train.py:421] (5/8) Epoch 9, batch 6000, loss[loss=2.294, over 2100.00 frames. , ppl: 9.918555736001194] tot_loss[loss=2.266, over 5273815.72 frames. , ppl: 9.644913679518801], batch size: 70
+2022-12-13 05:47:17,486 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 05:47:18,250 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.269, over 211138.00 frames. , ppl: 9.66698032189641
+2022-12-13 05:48:59,167 INFO [train.py:421] (5/8) Epoch 9, batch 6200, loss[loss=2.594, over 910.00 frames. , ppl: 13.381043899952525] tot_loss[loss=2.268, over 5240760.67 frames. , ppl: 9.664529956381015], batch size: 70
+2022-12-13 05:50:38,231 INFO [train.py:421] (5/8) Epoch 9, batch 6400, loss[loss=2.26, over 3220.00 frames. , ppl: 9.581149947662967] tot_loss[loss=2.268, over 5319472.63 frames. , ppl: 9.657083809589658], batch size: 70
+2022-12-13 05:52:22,372 INFO [train.py:421] (5/8) Epoch 9, batch 6600, loss[loss=2.15, over 5110.00 frames. , ppl: 8.585980665660173] tot_loss[loss=2.267, over 5338112.49 frames. , ppl: 9.653788472445607], batch size: 70
+2022-12-13 05:54:01,417 INFO [train.py:421] (5/8) Epoch 9, batch 6800, loss[loss=2.165, over 8120.00 frames. , ppl: 8.710264812354477] tot_loss[loss=2.268, over 5351910.17 frames. , ppl: 9.662643236280458], batch size: 70
+2022-12-13 05:55:37,120 INFO [train.py:421] (5/8) Epoch 9, batch 7000, loss[loss=2.437, over 1890.00 frames. , ppl: 11.436378317127394] tot_loss[loss=2.268, over 5364017.16 frames. , ppl: 9.657854744227635], batch size: 70
+2022-12-13 05:55:37,121 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 05:55:37,868 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.271, over 211138.00 frames. , ppl: 9.689801055644402
+2022-12-13 05:57:20,777 INFO [train.py:421] (5/8) Epoch 9, batch 7200, loss[loss=2.193, over 4200.00 frames. , ppl: 8.958283097380841] tot_loss[loss=2.267, over 5435279.10 frames. , ppl: 9.646213832863772], batch size: 70
+2022-12-13 05:59:03,230 INFO [train.py:421] (5/8) Epoch 9, batch 7400, loss[loss=2.4, over 1540.00 frames. , ppl: 11.025525319024016] tot_loss[loss=2.265, over 5489468.97 frames. , ppl: 9.632803821226668], batch size: 70
+2022-12-13 06:00:38,564 INFO [train.py:421] (5/8) Epoch 9, batch 7600, loss[loss=2.476, over 1330.00 frames. , ppl: 11.89252572196226] tot_loss[loss=2.265, over 5499220.70 frames. , ppl: 9.632201475532899], batch size: 70
+2022-12-13 06:02:16,060 INFO [train.py:421] (5/8) Epoch 9, batch 7800, loss[loss=2.575, over 910.00 frames. , ppl: 13.12880201036517] tot_loss[loss=2.267, over 5485297.05 frames. , ppl: 9.6463390879848], batch size: 70
+2022-12-13 06:03:55,032 INFO [train.py:421] (5/8) Epoch 9, batch 8000, loss[loss=3.272, over 490.00 frames. , ppl: 26.375302297642833] tot_loss[loss=2.266, over 5529711.10 frames. , ppl: 9.642075186591427], batch size: 70
+2022-12-13 06:03:55,033 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 06:03:55,778 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.27, over 211138.00 frames. , ppl: 9.677170017837494
+2022-12-13 06:05:34,733 INFO [train.py:421] (5/8) Epoch 9, batch 8200, loss[loss=2.338, over 3010.00 frames. , ppl: 10.361427252129733] tot_loss[loss=2.266, over 5553594.51 frames. , ppl: 9.64510336728242], batch size: 70
+2022-12-13 06:07:15,119 INFO [train.py:421] (5/8) Epoch 9, batch 8400, loss[loss=2.515, over 910.00 frames. , ppl: 12.365811207404239] tot_loss[loss=2.268, over 5512946.90 frames. , ppl: 9.658176879953347], batch size: 70
+2022-12-13 06:08:56,774 INFO [train.py:421] (5/8) Epoch 9, batch 8600, loss[loss=2.349, over 1960.00 frames. , ppl: 10.478091059387236] tot_loss[loss=2.269, over 5478032.86 frames. , ppl: 9.668818727425432], batch size: 70
+2022-12-13 06:10:33,948 INFO [train.py:421] (5/8) Epoch 9, batch 8800, loss[loss=2.203, over 8190.00 frames. , ppl: 9.050942318800862] tot_loss[loss=2.268, over 5509877.26 frames. , ppl: 9.663191128162778], batch size: 70
+2022-12-13 06:12:16,353 INFO [train.py:421] (5/8) Epoch 9, batch 9000, loss[loss=2.474, over 1330.00 frames. , ppl: 11.867346986463836] tot_loss[loss=2.269, over 5518211.95 frames. , ppl: 9.671259340146898], batch size: 70
+2022-12-13 06:12:16,354 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 06:12:17,100 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.269, over 211138.00 frames. , ppl: 9.667984785319996
+2022-12-13 06:13:54,674 INFO [train.py:421] (5/8) Epoch 9, batch 9200, loss[loss=2.309, over 1960.00 frames. , ppl: 10.064679410104956] tot_loss[loss=2.269, over 5513706.13 frames. , ppl: 9.673235741590233], batch size: 70
+2022-12-13 06:15:37,243 INFO [train.py:421] (5/8) Epoch 9, batch 9400, loss[loss=2.262, over 2380.00 frames. , ppl: 9.598806209135732] tot_loss[loss=2.269, over 5513378.53 frames. , ppl: 9.6715129838662], batch size: 70
+2022-12-13 06:17:15,205 INFO [train.py:421] (5/8) Epoch 9, batch 9600, loss[loss=2.276, over 2170.00 frames. , ppl: 9.741004424775127] tot_loss[loss=2.27, over 5485915.69 frames. , ppl: 9.68222151569859], batch size: 70
+2022-12-13 06:18:55,627 INFO [train.py:421] (5/8) Epoch 9, batch 9800, loss[loss=2.307, over 2380.00 frames. , ppl: 10.041539961917943] tot_loss[loss=2.269, over 5533029.93 frames. , ppl: 9.671998416405035], batch size: 70
+2022-12-13 06:20:37,101 INFO [train.py:421] (5/8) Epoch 9, batch 10000, loss[loss=2.365, over 1680.00 frames. , ppl: 10.646179509650265] tot_loss[loss=2.268, over 5588059.49 frames. , ppl: 9.663487145544977], batch size: 70
+2022-12-13 06:20:37,101 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 06:20:37,860 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.658802552803898
+2022-12-13 06:22:20,006 INFO [train.py:421] (5/8) Epoch 9, batch 10200, loss[loss=2.122, over 9380.00 frames. , ppl: 8.343714732579578] tot_loss[loss=2.269, over 5560731.58 frames. , ppl: 9.669904935536762], batch size: 70
+2022-12-13 06:23:58,842 INFO [train.py:421] (5/8) Epoch 9, batch 10400, loss[loss=2.155, over 4270.00 frames. , ppl: 8.624853129882728] tot_loss[loss=2.269, over 5553459.74 frames. , ppl: 9.668323671825675], batch size: 70
+2022-12-13 06:25:43,275 INFO [train.py:421] (5/8) Epoch 9, batch 10600, loss[loss=2.191, over 3290.00 frames. , ppl: 8.944514238293545] tot_loss[loss=2.269, over 5574468.32 frames. , ppl: 9.668948197276192], batch size: 70
+2022-12-13 06:27:26,738 INFO [train.py:421] (5/8) Epoch 9, batch 10800, loss[loss=2.149, over 9590.00 frames. , ppl: 8.574639751726851] tot_loss[loss=2.27, over 5549332.79 frames. , ppl: 9.67772428779759], batch size: 70
+2022-12-13 06:29:05,721 INFO [train.py:421] (5/8) Epoch 9, batch 11000, loss[loss=2.343, over 1960.00 frames. , ppl: 10.41027830302721] tot_loss[loss=2.27, over 5536675.24 frames. , ppl: 9.681007988946584], batch size: 70
+2022-12-13 06:29:05,722 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 06:29:06,467 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.269, over 211138.00 frames. , ppl: 9.671750305558916
+2022-12-13 06:30:49,635 INFO [train.py:421] (5/8) Epoch 9, batch 11200, loss[loss=2.573, over 700.00 frames. , ppl: 13.105311134080262] tot_loss[loss=2.269, over 5579568.47 frames. , ppl: 9.667866632486191], batch size: 70
+2022-12-13 06:32:30,897 INFO [train.py:421] (5/8) Epoch 9, batch 11400, loss[loss=2.329, over 2100.00 frames. , ppl: 10.2702660554682] tot_loss[loss=2.27, over 5559316.19 frames. , ppl: 9.677304149775328], batch size: 70
+2022-12-13 06:34:12,383 INFO [train.py:421] (5/8) Epoch 9, batch 11600, loss[loss=2.436, over 1680.00 frames. , ppl: 11.424739478264533] tot_loss[loss=2.27, over 5551015.43 frames. , ppl: 9.680753905354042], batch size: 70
+2022-12-13 06:35:50,659 INFO [train.py:421] (5/8) Epoch 9, batch 11800, loss[loss=2.372, over 1680.00 frames. , ppl: 10.713715836365704] tot_loss[loss=2.271, over 5524534.86 frames. , ppl: 9.68833430935093], batch size: 70
+2022-12-13 06:37:30,279 INFO [train.py:421] (5/8) Epoch 9, batch 12000, loss[loss=2.269, over 3500.00 frames. , ppl: 9.673369272921262] tot_loss[loss=2.269, over 5553239.67 frames. , ppl: 9.673777285932944], batch size: 70
+2022-12-13 06:37:30,280 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 06:37:31,009 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.655009206172396
+2022-12-13 06:39:10,529 INFO [train.py:421] (5/8) Epoch 9, batch 12200, loss[loss=2.24, over 2730.00 frames. , ppl: 9.391584850397024] tot_loss[loss=2.271, over 5515725.75 frames. , ppl: 9.686615440928867], batch size: 70
+2022-12-13 06:40:53,668 INFO [train.py:421] (5/8) Epoch 9, batch 12400, loss[loss=2.774, over 630.00 frames. , ppl: 16.021158906572268] tot_loss[loss=2.272, over 5493684.88 frames. , ppl: 9.70099401167453], batch size: 70
+2022-12-13 06:42:37,761 INFO [train.py:421] (5/8) Epoch 9, batch 12600, loss[loss=2.084, over 7560.00 frames. , ppl: 8.035382113526643] tot_loss[loss=2.273, over 5507693.56 frames. , ppl: 9.707165970928362], batch size: 70
+2022-12-13 06:44:16,096 INFO [train.py:421] (5/8) Epoch 9, batch 12800, loss[loss=3.233, over 490.00 frames. , ppl: 25.346890269594706] tot_loss[loss=2.273, over 5510115.98 frames. , ppl: 9.70457233083008], batch size: 70
+2022-12-13 06:45:55,156 INFO [train.py:421] (5/8) Epoch 9, batch 13000, loss[loss=2.378, over 2310.00 frames. , ppl: 10.784822227569018] tot_loss[loss=2.274, over 5480866.31 frames. , ppl: 9.715642728292748], batch size: 70
+2022-12-13 06:45:55,157 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 06:45:55,900 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.66105725550274
+2022-12-13 06:47:36,600 INFO [train.py:421] (5/8) Epoch 9, batch 13200, loss[loss=2.697, over 700.00 frames. , ppl: 14.830160020696908] tot_loss[loss=2.274, over 5474563.54 frames. , ppl: 9.714683091540413], batch size: 70
+2022-12-13 06:49:17,403 INFO [train.py:421] (5/8) Epoch 9, batch 13400, loss[loss=2.702, over 770.00 frames. , ppl: 14.907659284673548] tot_loss[loss=2.274, over 5473912.29 frames. , ppl: 9.717463208764663], batch size: 70
+2022-12-13 06:50:56,808 INFO [train.py:421] (5/8) Epoch 9, batch 13600, loss[loss=2.441, over 1680.00 frames. , ppl: 11.485526811483236] tot_loss[loss=2.274, over 5468491.53 frames. , ppl: 9.71828919111088], batch size: 70
+2022-12-13 06:52:32,853 INFO [train.py:421] (5/8) Epoch 9, batch 13800, loss[loss=2.305, over 3220.00 frames. , ppl: 10.021280206162388] tot_loss[loss=2.273, over 5471218.69 frames. , ppl: 9.709191424748543], batch size: 70
+2022-12-13 06:54:17,187 INFO [train.py:421] (5/8) Epoch 9, batch 14000, loss[loss=2.289, over 3430.00 frames. , ppl: 9.866139314315213] tot_loss[loss=2.273, over 5462853.34 frames. , ppl: 9.706160443861174], batch size: 70
+2022-12-13 06:54:17,188 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 06:54:17,917 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.27, over 211138.00 frames. , ppl: 9.675671954910115
+2022-12-13 06:55:57,100 INFO [train.py:421] (5/8) Epoch 9, batch 14200, loss[loss=2.342, over 1190.00 frames. , ppl: 10.402218112815408] tot_loss[loss=2.273, over 5486799.97 frames. , ppl: 9.705363718843145], batch size: 70
+2022-12-13 06:57:38,475 INFO [train.py:421] (5/8) Epoch 9, batch 14400, loss[loss=2.225, over 6930.00 frames. , ppl: 9.24989233064181] tot_loss[loss=2.272, over 5508464.21 frames. , ppl: 9.702368945861132], batch size: 70
+2022-12-13 06:59:18,391 INFO [train.py:421] (5/8) Epoch 9, batch 14600, loss[loss=2.377, over 1610.00 frames. , ppl: 10.7777210307657] tot_loss[loss=2.272, over 5503473.54 frames. , ppl: 9.700945901802886], batch size: 70
+2022-12-13 07:00:59,417 INFO [train.py:421] (5/8) Epoch 9, batch 14800, loss[loss=2.244, over 9310.00 frames. , ppl: 9.429607509975083] tot_loss[loss=2.274, over 5453251.40 frames. , ppl: 9.716283900881324], batch size: 70
+2022-12-13 07:02:36,592 INFO [train.py:421] (5/8) Epoch 9, batch 15000, loss[loss=2.753, over 630.00 frames. , ppl: 15.69306321067447] tot_loss[loss=2.273, over 5477238.42 frames. , ppl: 9.706943251661514], batch size: 70
+2022-12-13 07:02:36,592 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 07:02:37,339 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.661317502370732
+2022-12-13 07:04:17,421 INFO [train.py:421] (5/8) Epoch 9, batch 15200, loss[loss=2.219, over 3500.00 frames. , ppl: 9.196213770655765] tot_loss[loss=2.274, over 5417516.28 frames. , ppl: 9.715923332874864], batch size: 70
+2022-12-13 07:05:53,691 INFO [train.py:421] (5/8) Epoch 9, batch 15400, loss[loss=2.16, over 4550.00 frames. , ppl: 8.674298345921546] tot_loss[loss=2.275, over 5385654.98 frames. , ppl: 9.726034734028193], batch size: 70
+2022-12-13 07:07:28,823 INFO [train.py:421] (5/8) Epoch 9, batch 15600, loss[loss=2.416, over 1400.00 frames. , ppl: 11.199316485135913] tot_loss[loss=2.274, over 5397014.06 frames. , ppl: 9.720645586045496], batch size: 70
+2022-12-13 07:09:10,482 INFO [train.py:421] (5/8) Epoch 9, batch 15800, loss[loss=2.417, over 2100.00 frames. , ppl: 11.217119899961798] tot_loss[loss=2.274, over 5410857.79 frames. , ppl: 9.717480429304931], batch size: 70
+2022-12-13 07:10:50,253 INFO [train.py:421] (5/8) Epoch 9, batch 16000, loss[loss=2.156, over 6090.00 frames. , ppl: 8.638501703560397] tot_loss[loss=2.275, over 5398889.29 frames. , ppl: 9.723250716109625], batch size: 70
+2022-12-13 07:10:50,254 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 07:10:51,000 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.65729732776054
+2022-12-13 07:12:29,286 INFO [train.py:421] (5/8) Epoch 9, batch 16200, loss[loss=2.25, over 4270.00 frames. , ppl: 9.484698497383809] tot_loss[loss=2.276, over 5349715.24 frames. , ppl: 9.733669867613049], batch size: 70
+2022-12-13 07:14:07,469 INFO [train.py:421] (5/8) Epoch 9, batch 16400, loss[loss=2.28, over 2240.00 frames. , ppl: 9.773687458381797] tot_loss[loss=2.275, over 5356954.62 frames. , ppl: 9.724810548145882], batch size: 70
+2022-12-13 07:15:49,883 INFO [train.py:421] (5/8) Epoch 9, batch 16600, loss[loss=2.573, over 840.00 frames. , ppl: 13.106491207607279] tot_loss[loss=2.274, over 5411571.40 frames. , ppl: 9.71532963283394], batch size: 70
+2022-12-13 07:17:30,258 INFO [train.py:421] (5/8) Epoch 9, batch 16800, loss[loss=2.228, over 5530.00 frames. , ppl: 9.284817422541824] tot_loss[loss=2.273, over 5419744.62 frames. , ppl: 9.709059122865213], batch size: 70
+2022-12-13 07:19:11,498 INFO [train.py:421] (5/8) Epoch 9, batch 17000, loss[loss=3.172, over 490.00 frames. , ppl: 23.84434671002195] tot_loss[loss=2.273, over 5416212.43 frames. , ppl: 9.711647766114416], batch size: 70
+2022-12-13 07:19:11,498 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 07:19:12,256 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.27, over 211138.00 frames. , ppl: 9.679668259935742
+2022-12-13 07:20:56,033 INFO [train.py:421] (5/8) Epoch 9, batch 17200, loss[loss=2.287, over 3080.00 frames. , ppl: 9.84130765999773] tot_loss[loss=2.272, over 5466594.72 frames. , ppl: 9.699485634505603], batch size: 70
+2022-12-13 07:22:38,913 INFO [train.py:421] (5/8) Epoch 9, batch 17400, loss[loss=2.306, over 2800.00 frames. , ppl: 10.03239408769835] tot_loss[loss=2.273, over 5451491.97 frames. , ppl: 9.707229983915632], batch size: 70
+2022-12-13 07:24:16,738 INFO [train.py:421] (5/8) Epoch 9, batch 17600, loss[loss=2.209, over 3710.00 frames. , ppl: 9.105012847731691] tot_loss[loss=2.273, over 5448633.22 frames. , ppl: 9.706618980018922], batch size: 70
+2022-12-13 07:25:57,052 INFO [train.py:421] (5/8) Epoch 9, batch 17800, loss[loss=2.21, over 6860.00 frames. , ppl: 9.119926511300717] tot_loss[loss=2.272, over 5486959.03 frames. , ppl: 9.69800922444851], batch size: 70
+2022-12-13 07:27:36,348 INFO [train.py:421] (5/8) Epoch 9, batch 18000, loss[loss=2.258, over 2730.00 frames. , ppl: 9.566411996239859] tot_loss[loss=2.272, over 5497034.49 frames. , ppl: 9.700215597913923], batch size: 70
+2022-12-13 07:27:36,348 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 07:27:37,108 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.635843748058024
+2022-12-13 07:29:17,364 INFO [train.py:421] (5/8) Epoch 9, batch 18200, loss[loss=2.177, over 3850.00 frames. , ppl: 8.817501338271999] tot_loss[loss=2.273, over 5498748.66 frames. , ppl: 9.70709385133237], batch size: 70
+2022-12-13 07:30:55,921 INFO [train.py:421] (5/8) Epoch 9, batch 18400, loss[loss=2.531, over 840.00 frames. , ppl: 12.567246237108224] tot_loss[loss=2.274, over 5426439.05 frames. , ppl: 9.722812123467204], batch size: 70
+2022-12-13 07:32:36,191 INFO [train.py:421] (5/8) Epoch 9, batch 18600, loss[loss=2.646, over 840.00 frames. , ppl: 14.094878931477545] tot_loss[loss=2.274, over 5440459.84 frames. , ppl: 9.721249978695711], batch size: 70
+2022-12-13 07:34:20,274 INFO [train.py:421] (5/8) Epoch 9, batch 18800, loss[loss=2.445, over 1050.00 frames. , ppl: 11.529477238960633] tot_loss[loss=2.273, over 5477400.30 frames. , ppl: 9.706026250059875], batch size: 70
+2022-12-13 07:36:01,309 INFO [train.py:421] (5/8) Epoch 9, batch 19000, loss[loss=2.242, over 3220.00 frames. , ppl: 9.41177919939543] tot_loss[loss=2.273, over 5482272.94 frames. , ppl: 9.70528948583854], batch size: 70
+2022-12-13 07:36:01,310 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 07:36:02,070 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.654813433317269
+2022-12-13 07:37:39,895 INFO [train.py:421] (5/8) Epoch 9, batch 19200, loss[loss=2.557, over 980.00 frames. , ppl: 12.894251974156088] tot_loss[loss=2.272, over 5490999.41 frames. , ppl: 9.699105497995589], batch size: 70
+2022-12-13 07:39:21,549 INFO [train.py:421] (5/8) Epoch 9, batch 19400, loss[loss=2.245, over 3640.00 frames. , ppl: 9.44147505900856] tot_loss[loss=2.271, over 5491650.50 frames. , ppl: 9.693577587675481], batch size: 70
+2022-12-13 07:41:01,750 INFO [train.py:421] (5/8) Epoch 9, batch 19600, loss[loss=2.346, over 910.00 frames. , ppl: 10.447223829907898] tot_loss[loss=2.271, over 5492283.78 frames. , ppl: 9.687584062801285], batch size: 70
+2022-12-13 07:42:40,816 INFO [train.py:421] (5/8) Epoch 9, batch 19800, loss[loss=2.193, over 6720.00 frames. , ppl: 8.958880398653928] tot_loss[loss=2.271, over 5494294.24 frames. , ppl: 9.69368331914619], batch size: 70
+2022-12-13 07:44:22,224 INFO [train.py:421] (5/8) Epoch 9, batch 20000, loss[loss=2.246, over 3080.00 frames. , ppl: 9.447770097862977] tot_loss[loss=2.27, over 5540919.60 frames. , ppl: 9.675146640663154], batch size: 70
+2022-12-13 07:44:22,225 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 07:44:22,986 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.269, over 211138.00 frames. , ppl: 9.671224962542615
+2022-12-13 07:46:03,948 INFO [train.py:421] (5/8) Epoch 9, batch 20200, loss[loss=2.339, over 2660.00 frames. , ppl: 10.373778339153537] tot_loss[loss=2.27, over 5531858.50 frames. , ppl: 9.680008967300019], batch size: 70
+2022-12-13 07:47:42,245 INFO [train.py:421] (5/8) Epoch 9, batch 20400, loss[loss=2.354, over 2030.00 frames. , ppl: 10.522376443119807] tot_loss[loss=2.269, over 5556086.12 frames. , ppl: 9.674547463949558], batch size: 70
+2022-12-13 07:49:23,897 INFO [train.py:421] (5/8) Epoch 9, batch 20600, loss[loss=2.513, over 840.00 frames. , ppl: 12.342838984834977] tot_loss[loss=2.269, over 5595704.88 frames. , ppl: 9.668190078571854], batch size: 70
+2022-12-13 07:51:06,183 INFO [train.py:421] (5/8) Epoch 9, batch 20800, loss[loss=2.311, over 2030.00 frames. , ppl: 10.080003500487956] tot_loss[loss=2.271, over 5530037.70 frames. , ppl: 9.688293066236751], batch size: 70
+2022-12-13 07:52:48,694 INFO [train.py:421] (5/8) Epoch 9, batch 21000, loss[loss=2.51, over 980.00 frames. , ppl: 12.310949608452008] tot_loss[loss=2.271, over 5552735.88 frames. , ppl: 9.686858393307944], batch size: 70
+2022-12-13 07:52:48,695 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 07:52:49,455 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.652867352648064
+2022-12-13 07:54:33,301 INFO [train.py:421] (5/8) Epoch 9, batch 21200, loss[loss=2.823, over 700.00 frames. , ppl: 16.82143528551299] tot_loss[loss=2.271, over 5542224.97 frames. , ppl: 9.69034966854401], batch size: 70
+2022-12-13 07:56:15,662 INFO [train.py:421] (5/8) Epoch 9, batch 21400, loss[loss=2.509, over 840.00 frames. , ppl: 12.298156889422986] tot_loss[loss=2.272, over 5491659.75 frames. , ppl: 9.702861839229438], batch size: 70
+2022-12-13 07:57:54,060 INFO [train.py:421] (5/8) Epoch 9, batch 21600, loss[loss=2.899, over 560.00 frames. , ppl: 18.15545204339222] tot_loss[loss=2.272, over 5493010.71 frames. , ppl: 9.698579867466957], batch size: 70
+2022-12-13 07:59:33,285 INFO [train.py:421] (5/8) Epoch 9, batch 21800, loss[loss=2.536, over 1050.00 frames. , ppl: 12.631045241802175] tot_loss[loss=2.271, over 5520494.27 frames. , ppl: 9.692084224327129], batch size: 70
+2022-12-13 08:01:12,352 INFO [train.py:421] (5/8) Epoch 9, batch 22000, loss[loss=2.182, over 5880.00 frames. , ppl: 8.862034041346451] tot_loss[loss=2.272, over 5483573.99 frames. , ppl: 9.699783814899801], batch size: 70
+2022-12-13 08:01:12,353 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 08:01:13,114 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.65536646613963
+2022-12-13 08:02:53,702 INFO [train.py:421] (5/8) Epoch 9, batch 22200, loss[loss=2.685, over 840.00 frames. , ppl: 14.656363596694167] tot_loss[loss=2.272, over 5515762.73 frames. , ppl: 9.695269155792017], batch size: 70
+2022-12-13 08:04:32,702 INFO [train.py:421] (5/8) Epoch 9, batch 22400, loss[loss=2.167, over 9170.00 frames. , ppl: 8.733604125962602] tot_loss[loss=2.271, over 5486918.93 frames. , ppl: 9.693234329217823], batch size: 70
+2022-12-13 08:06:15,182 INFO [train.py:421] (5/8) Epoch 9, batch 22600, loss[loss=2.363, over 2240.00 frames. , ppl: 10.619961826512288] tot_loss[loss=2.272, over 5501876.39 frames. , ppl: 9.697622501213605], batch size: 70
+2022-12-13 08:07:55,314 INFO [train.py:421] (5/8) Epoch 9, batch 22800, loss[loss=2.372, over 1890.00 frames. , ppl: 10.720110525175732] tot_loss[loss=2.271, over 5535321.40 frames. , ppl: 9.690341395590732], batch size: 70
+2022-12-13 08:09:36,111 INFO [train.py:421] (5/8) Epoch 9, batch 23000, loss[loss=2.145, over 9380.00 frames. , ppl: 8.539380445662616] tot_loss[loss=2.27, over 5557274.09 frames. , ppl: 9.6814722745501], batch size: 70
+2022-12-13 08:09:36,111 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 08:09:36,840 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.639090273243687
+2022-12-13 08:11:16,442 INFO [train.py:421] (5/8) Epoch 9, batch 23200, loss[loss=2.233, over 2590.00 frames. , ppl: 9.324464217396383] tot_loss[loss=2.271, over 5547558.45 frames. , ppl: 9.685905467912267], batch size: 70
+2022-12-13 08:12:57,319 INFO [train.py:421] (5/8) Epoch 9, batch 23400, loss[loss=2.211, over 3150.00 frames. , ppl: 9.124034433592755] tot_loss[loss=2.27, over 5566811.79 frames. , ppl: 9.679236875483518], batch size: 70
+2022-12-13 08:14:44,681 INFO [train.py:421] (5/8) Epoch 9, batch 23600, loss[loss=2.284, over 3150.00 frames. , ppl: 9.8206583045437] tot_loss[loss=2.269, over 5592438.41 frames. , ppl: 9.670826343824716], batch size: 70
+2022-12-13 08:16:23,460 INFO [train.py:421] (5/8) Epoch 9, batch 23800, loss[loss=2.279, over 2380.00 frames. , ppl: 9.763941547360876] tot_loss[loss=2.27, over 5575432.74 frames. , ppl: 9.679395551777318], batch size: 70
+2022-12-13 08:18:03,862 INFO [train.py:421] (5/8) Epoch 9, batch 24000, loss[loss=2.393, over 1260.00 frames. , ppl: 10.947142020691949] tot_loss[loss=2.27, over 5590277.81 frames. , ppl: 9.674573455447403], batch size: 70
+2022-12-13 08:18:03,863 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 08:18:04,611 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.650770254134942
+2022-12-13 08:19:39,211 INFO [train.py:421] (5/8) Epoch 9, batch 24200, loss[loss=2.35, over 3570.00 frames. , ppl: 10.488487186833323] tot_loss[loss=2.271, over 5529539.72 frames. , ppl: 9.690004043536977], batch size: 70
+2022-12-13 08:21:19,435 INFO [train.py:421] (5/8) Epoch 9, batch 24400, loss[loss=2.276, over 2940.00 frames. , ppl: 9.7327929817008] tot_loss[loss=2.272, over 5503616.00 frames. , ppl: 9.694514403821854], batch size: 70
+2022-12-13 08:23:01,200 INFO [train.py:421] (5/8) Epoch 9, batch 24600, loss[loss=2.256, over 3290.00 frames. , ppl: 9.543511054287718] tot_loss[loss=2.272, over 5485213.33 frames. , ppl: 9.70183108313721], batch size: 70
+2022-12-13 08:24:43,333 INFO [train.py:421] (5/8) Epoch 9, batch 24800, loss[loss=2.332, over 3570.00 frames. , ppl: 10.299380739076454] tot_loss[loss=2.272, over 5493806.39 frames. , ppl: 9.700784887299793], batch size: 70
+2022-12-13 08:26:25,431 INFO [train.py:421] (5/8) Epoch 9, batch 25000, loss[loss=2.232, over 2660.00 frames. , ppl: 9.315645767353889] tot_loss[loss=2.272, over 5490658.33 frames. , ppl: 9.699866915639637], batch size: 70
+2022-12-13 08:26:25,431 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 08:26:26,196 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.640315849779926
+2022-12-13 08:28:06,125 INFO [train.py:421] (5/8) Epoch 9, batch 25200, loss[loss=2.348, over 2940.00 frames. , ppl: 10.462105396535138] tot_loss[loss=2.272, over 5503290.63 frames. , ppl: 9.70049331206918], batch size: 70
+2022-12-13 08:29:44,792 INFO [train.py:421] (5/8) Epoch 9, batch 25400, loss[loss=2.324, over 1750.00 frames. , ppl: 10.212455670018551] tot_loss[loss=2.273, over 5493325.71 frames. , ppl: 9.704601369976967], batch size: 70
+2022-12-13 08:31:25,694 INFO [train.py:421] (5/8) Epoch 9, batch 25600, loss[loss=2.33, over 2730.00 frames. , ppl: 10.275983575225304] tot_loss[loss=2.273, over 5480626.49 frames. , ppl: 9.704623131360322], batch size: 70
+2022-12-13 08:33:08,240 INFO [train.py:421] (5/8) Epoch 9, batch 25800, loss[loss=2.235, over 4130.00 frames. , ppl: 9.351015280707617] tot_loss[loss=2.273, over 5461382.20 frames. , ppl: 9.711834656759393], batch size: 70
+2022-12-13 08:34:47,565 INFO [train.py:421] (5/8) Epoch 9, batch 26000, loss[loss=2.271, over 2100.00 frames. , ppl: 9.692765120064367] tot_loss[loss=2.273, over 5490905.12 frames. , ppl: 9.707851630244667], batch size: 70
+2022-12-13 08:34:47,565 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 08:34:48,310 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.645219714400634
+2022-12-13 08:36:28,484 INFO [train.py:421] (5/8) Epoch 9, batch 26200, loss[loss=2.158, over 6440.00 frames. , ppl: 8.657538407136416] tot_loss[loss=2.274, over 5472588.26 frames. , ppl: 9.717389541212873], batch size: 70
+2022-12-13 08:38:06,093 INFO [train.py:421] (5/8) Epoch 9, batch 26400, loss[loss=2.318, over 2520.00 frames. , ppl: 10.158381364018895] tot_loss[loss=2.275, over 5432288.94 frames. , ppl: 9.730630928774607], batch size: 70
+2022-12-13 08:39:46,906 INFO [train.py:421] (5/8) Epoch 9, batch 26600, loss[loss=3.13, over 560.00 frames. , ppl: 22.879816055641573] tot_loss[loss=2.276, over 5441119.48 frames. , ppl: 9.740405009975168], batch size: 70
+2022-12-13 08:41:30,300 INFO [train.py:421] (5/8) Epoch 9, batch 26800, loss[loss=2.259, over 1960.00 frames. , ppl: 9.571796886315335] tot_loss[loss=2.276, over 5430669.15 frames. , ppl: 9.740822888784969], batch size: 70
+2022-12-13 08:43:11,277 INFO [train.py:421] (5/8) Epoch 9, batch 27000, loss[loss=2.196, over 4200.00 frames. , ppl: 8.986024821780807] tot_loss[loss=2.276, over 5438177.20 frames. , ppl: 9.73370334656564], batch size: 70
+2022-12-13 08:43:11,278 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 08:43:12,039 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.652885925719394
+2022-12-13 08:44:53,159 INFO [train.py:421] (5/8) Epoch 9, batch 27200, loss[loss=2.54, over 980.00 frames. , ppl: 12.675150903884203] tot_loss[loss=2.277, over 5438361.75 frames. , ppl: 9.743783022750643], batch size: 70
+2022-12-13 08:46:33,215 INFO [train.py:421] (5/8) Epoch 9, batch 27400, loss[loss=3.01, over 560.00 frames. , ppl: 20.29571024828348] tot_loss[loss=2.276, over 5416691.30 frames. , ppl: 9.737140188013067], batch size: 70
+2022-12-13 08:48:13,489 INFO [train.py:421] (5/8) Epoch 9, batch 27600, loss[loss=4.996, over 280.00 frames. , ppl: 147.8932128470895] tot_loss[loss=2.277, over 5420047.55 frames. , ppl: 9.744571639840046], batch size: 70
+2022-12-13 08:49:50,164 INFO [train.py:421] (5/8) Epoch 9, batch 27800, loss[loss=2.315, over 1820.00 frames. , ppl: 10.127494569763273] tot_loss[loss=2.277, over 5428063.82 frames. , ppl: 9.744151838201308], batch size: 70
+2022-12-13 08:51:34,738 INFO [train.py:421] (5/8) Epoch 9, batch 28000, loss[loss=2.2, over 9100.00 frames. , ppl: 9.0212719318953] tot_loss[loss=2.276, over 5449812.55 frames. , ppl: 9.739476844951184], batch size: 70
+2022-12-13 08:51:34,738 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 08:51:35,469 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.631405082310689
+2022-12-13 08:53:13,314 INFO [train.py:421] (5/8) Epoch 9, batch 28200, loss[loss=3.518, over 420.00 frames. , ppl: 33.716510453950185] tot_loss[loss=2.275, over 5467757.85 frames. , ppl: 9.730481865681154], batch size: 70
+2022-12-13 08:54:53,462 INFO [train.py:421] (5/8) Epoch 9, batch 28400, loss[loss=2.231, over 1960.00 frames. , ppl: 9.311108477646302] tot_loss[loss=2.275, over 5487159.84 frames. , ppl: 9.727329017544177], batch size: 70
+2022-12-13 08:56:32,697 INFO [train.py:421] (5/8) Epoch 9, batch 28600, loss[loss=2.571, over 700.00 frames. , ppl: 13.083718342627275] tot_loss[loss=2.274, over 5514457.73 frames. , ppl: 9.716663451314169], batch size: 70
+2022-12-13 08:58:14,669 INFO [train.py:421] (5/8) Epoch 9, batch 28800, loss[loss=2.327, over 1330.00 frames. , ppl: 10.248212300465523] tot_loss[loss=2.273, over 5529236.84 frames. , ppl: 9.710004714396037], batch size: 70
+2022-12-13 08:59:57,440 INFO [train.py:421] (5/8) Epoch 9, batch 29000, loss[loss=2.354, over 2100.00 frames. , ppl: 10.528550204683896] tot_loss[loss=2.273, over 5558928.46 frames. , ppl: 9.707269156011522], batch size: 70
+2022-12-13 08:59:57,441 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 08:59:58,199 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.650904523356159
+2022-12-13 09:01:42,270 INFO [train.py:421] (5/8) Epoch 9, batch 29200, loss[loss=2.271, over 2590.00 frames. , ppl: 9.690676702141142] tot_loss[loss=2.274, over 5538567.85 frames. , ppl: 9.71714821372806], batch size: 70
+2022-12-13 09:03:21,199 INFO [train.py:421] (5/8) Epoch 9, batch 29400, loss[loss=2.219, over 5600.00 frames. , ppl: 9.195899452314585] tot_loss[loss=2.275, over 5490743.56 frames. , ppl: 9.732017234035675], batch size: 70
+2022-12-13 09:05:00,162 INFO [train.py:421] (5/8) Epoch 9, batch 29600, loss[loss=2.197, over 3990.00 frames. , ppl: 8.999096271813059] tot_loss[loss=2.275, over 5490955.03 frames. , ppl: 9.724983923776758], batch size: 70
+2022-12-13 09:06:41,422 INFO [train.py:421] (5/8) Epoch 9, batch 29800, loss[loss=2.201, over 3640.00 frames. , ppl: 9.030741663540596] tot_loss[loss=2.274, over 5509089.20 frames. , ppl: 9.718591134575025], batch size: 70
+2022-12-13 09:08:24,043 INFO [train.py:421] (5/8) Epoch 9, batch 30000, loss[loss=2.198, over 4270.00 frames. , ppl: 9.007490243916058] tot_loss[loss=2.272, over 5564423.72 frames. , ppl: 9.702829738226216], batch size: 70
+2022-12-13 09:08:24,043 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 09:08:24,805 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.637924764691014
+2022-12-13 09:10:04,498 INFO [train.py:421] (5/8) Epoch 9, batch 30200, loss[loss=2.424, over 1470.00 frames. , ppl: 11.293035383002215] tot_loss[loss=2.274, over 5511158.20 frames. , ppl: 9.721726668427024], batch size: 70
+2022-12-13 09:11:45,358 INFO [train.py:421] (5/8) Epoch 9, batch 30400, loss[loss=2.589, over 840.00 frames. , ppl: 13.32052086989549] tot_loss[loss=2.273, over 5535592.33 frames. , ppl: 9.711792924012354], batch size: 70
+2022-12-13 09:13:24,107 INFO [train.py:421] (5/8) Epoch 9, batch 30600, loss[loss=2.693, over 700.00 frames. , ppl: 14.782787298093805] tot_loss[loss=2.275, over 5477946.63 frames. , ppl: 9.727969616789263], batch size: 70
+2022-12-13 09:15:01,163 INFO [train.py:421] (5/8) Epoch 9, batch 30800, loss[loss=2.278, over 3150.00 frames. , ppl: 9.760257296275139] tot_loss[loss=2.275, over 5468826.60 frames. , ppl: 9.727835464661958], batch size: 70
+2022-12-13 09:16:34,884 INFO [train.py:421] (5/8) Epoch 9, batch 31000, loss[loss=2.437, over 980.00 frames. , ppl: 11.441075808411359] tot_loss[loss=2.276, over 5451006.37 frames. , ppl: 9.733827305281013], batch size: 70
+2022-12-13 09:16:34,885 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 09:16:35,648 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.66225130295525
+2022-12-13 09:18:16,658 INFO [train.py:421] (5/8) Epoch 9, batch 31200, loss[loss=2.291, over 2520.00 frames. , ppl: 9.887535061122227] tot_loss[loss=2.276, over 5419759.94 frames. , ppl: 9.741614242845502], batch size: 70
+2022-12-13 09:19:54,595 INFO [train.py:421] (5/8) Epoch 9, batch 31400, loss[loss=2.273, over 2380.00 frames. , ppl: 9.70685840533935] tot_loss[loss=2.277, over 5392667.73 frames. , ppl: 9.748705594392558], batch size: 70
+2022-12-13 09:21:34,364 INFO [train.py:421] (5/8) Epoch 9, batch 31600, loss[loss=2.449, over 980.00 frames. , ppl: 11.577840192103727] tot_loss[loss=2.277, over 5411452.77 frames. , ppl: 9.742869985224276], batch size: 70
+2022-12-13 09:23:15,108 INFO [train.py:421] (5/8) Epoch 9, batch 31800, loss[loss=2.246, over 4130.00 frames. , ppl: 9.452568102701633] tot_loss[loss=2.275, over 5427559.52 frames. , ppl: 9.731219513665572], batch size: 70
+2022-12-13 09:24:55,717 INFO [train.py:421] (5/8) Epoch 9, batch 32000, loss[loss=2.263, over 2100.00 frames. , ppl: 9.608660748285143] tot_loss[loss=2.275, over 5438700.03 frames. , ppl: 9.727875629552743], batch size: 70
+2022-12-13 09:24:55,718 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 09:24:56,478 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.664727097772253
+2022-12-13 09:26:38,798 INFO [train.py:421] (5/8) Epoch 9, batch 32200, loss[loss=2.513, over 1330.00 frames. , ppl: 12.347398048629778] tot_loss[loss=2.274, over 5471103.40 frames. , ppl: 9.720544385967004], batch size: 70
+2022-12-13 09:28:15,359 INFO [train.py:421] (5/8) Epoch 9, batch 32400, loss[loss=2.19, over 3360.00 frames. , ppl: 8.934028718599734] tot_loss[loss=2.277, over 5393519.44 frames. , ppl: 9.745498048285862], batch size: 70
+2022-12-13 09:29:54,372 INFO [train.py:421] (5/8) Epoch 9, batch 32600, loss[loss=2.193, over 5320.00 frames. , ppl: 8.960196531511176] tot_loss[loss=2.277, over 5369383.44 frames. , ppl: 9.751557065533097], batch size: 70
+2022-12-13 09:31:30,579 INFO [train.py:421] (5/8) Epoch 9, batch 32800, loss[loss=2.19, over 4690.00 frames. , ppl: 8.938223183316563] tot_loss[loss=2.278, over 5343672.47 frames. , ppl: 9.760352367587563], batch size: 70
+2022-12-13 09:33:11,503 INFO [train.py:421] (5/8) Epoch 9, batch 33000, loss[loss=2.305, over 3150.00 frames. , ppl: 10.022198846706537] tot_loss[loss=2.277, over 5386730.51 frames. , ppl: 9.750229184260467], batch size: 70
+2022-12-13 09:33:11,503 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 09:33:12,263 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.644468844695679
+2022-12-13 09:34:49,445 INFO [train.py:421] (5/8) Epoch 9, batch 33200, loss[loss=2.241, over 2940.00 frames. , ppl: 9.40058245314689] tot_loss[loss=2.276, over 5443759.14 frames. , ppl: 9.734074883094102], batch size: 70
+2022-12-13 09:36:29,258 INFO [train.py:421] (5/8) Epoch 9, batch 33400, loss[loss=3.255, over 490.00 frames. , ppl: 25.911206074256327] tot_loss[loss=2.277, over 5386696.35 frames. , ppl: 9.747327752110758], batch size: 70
+2022-12-13 09:38:10,170 INFO [train.py:421] (5/8) Epoch 9, batch 33600, loss[loss=4.136, over 350.00 frames. , ppl: 62.53254126421028] tot_loss[loss=2.276, over 5419576.20 frames. , ppl: 9.741797631061154], batch size: 70
+2022-12-13 09:39:46,992 INFO [train.py:421] (5/8) Epoch 9, batch 33800, loss[loss=2.348, over 1680.00 frames. , ppl: 10.460756913488144] tot_loss[loss=2.277, over 5406034.03 frames. , ppl: 9.746141940310315], batch size: 70
+2022-12-13 09:41:27,519 INFO [train.py:421] (5/8) Epoch 9, batch 34000, loss[loss=2.172, over 4620.00 frames. , ppl: 8.77178746377026] tot_loss[loss=2.276, over 5425360.18 frames. , ppl: 9.74218985823476], batch size: 70
+2022-12-13 09:41:27,520 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 09:41:28,282 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.6631322756689
+2022-12-13 09:43:09,902 INFO [train.py:421] (5/8) Epoch 9, batch 34200, loss[loss=2.259, over 2170.00 frames. , ppl: 9.571620892346857] tot_loss[loss=2.276, over 5446020.63 frames. , ppl: 9.734587507011911], batch size: 70
+2022-12-13 09:44:49,447 INFO [train.py:421] (5/8) Epoch 9, batch 34400, loss[loss=2.202, over 4690.00 frames. , ppl: 9.046262259925001] tot_loss[loss=2.275, over 5468430.61 frames. , ppl: 9.728017636887019], batch size: 70
+2022-12-13 09:46:28,568 INFO [train.py:421] (5/8) Epoch 9, batch 34600, loss[loss=2.33, over 2450.00 frames. , ppl: 10.283006371032808] tot_loss[loss=2.275, over 5477211.17 frames. , ppl: 9.726330613750342], batch size: 70
+2022-12-13 09:48:11,553 INFO [train.py:421] (5/8) Epoch 9, batch 34800, loss[loss=2.488, over 1260.00 frames. , ppl: 12.03740512784178] tot_loss[loss=2.274, over 5460938.55 frames. , ppl: 9.721899956681828], batch size: 70
+2022-12-13 09:49:50,504 INFO [train.py:421] (5/8) Epoch 9, batch 35000, loss[loss=2.203, over 7210.00 frames. , ppl: 9.049070516426518] tot_loss[loss=2.274, over 5485957.69 frames. , ppl: 9.716849546453], batch size: 70
+2022-12-13 09:49:50,505 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 09:49:51,240 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.644501676179262
+2022-12-13 09:51:35,362 INFO [train.py:421] (5/8) Epoch 9, batch 35200, loss[loss=2.593, over 840.00 frames. , ppl: 13.36958967804177] tot_loss[loss=2.274, over 5501044.69 frames. , ppl: 9.714044796392852], batch size: 70
+2022-12-13 09:53:15,453 INFO [train.py:421] (5/8) Epoch 9, batch 35400, loss[loss=2.227, over 3640.00 frames. , ppl: 9.267577223718005] tot_loss[loss=2.273, over 5495430.39 frames. , ppl: 9.712782322169417], batch size: 70
+2022-12-13 09:54:55,419 INFO [train.py:421] (5/8) Epoch 9, batch 35600, loss[loss=2.374, over 1610.00 frames. , ppl: 10.738871162868806] tot_loss[loss=2.275, over 5448169.33 frames. , ppl: 9.729163101172054], batch size: 70
+2022-12-13 09:56:35,170 INFO [train.py:421] (5/8) Epoch 9, batch 35800, loss[loss=2.293, over 2730.00 frames. , ppl: 9.90469073345424] tot_loss[loss=2.275, over 5447985.73 frames. , ppl: 9.726140653714936], batch size: 70
+2022-12-13 09:58:16,595 INFO [train.py:421] (5/8) Epoch 9, batch 36000, loss[loss=2.522, over 910.00 frames. , ppl: 12.450980106612608] tot_loss[loss=2.273, over 5468340.97 frames. , ppl: 9.710593150501937], batch size: 70
+2022-12-13 09:58:16,596 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 09:58:17,363 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.647108567176648
+2022-12-13 09:59:55,817 INFO [train.py:421] (5/8) Epoch 9, batch 36200, loss[loss=2.394, over 1260.00 frames. , ppl: 10.954649202913325] tot_loss[loss=2.274, over 5434758.51 frames. , ppl: 9.720892761037224], batch size: 70
+2022-12-13 10:01:36,713 INFO [train.py:421] (5/8) Epoch 9, batch 36400, loss[loss=2.391, over 1540.00 frames. , ppl: 10.921265763573723] tot_loss[loss=2.274, over 5452779.10 frames. , ppl: 9.721797487525697], batch size: 70
+2022-12-13 10:03:20,896 INFO [train.py:421] (5/8) Epoch 9, batch 36600, loss[loss=2.69, over 770.00 frames. , ppl: 14.725788014275913] tot_loss[loss=2.273, over 5496811.02 frames. , ppl: 9.711322076605127], batch size: 70
+2022-12-13 10:05:04,601 INFO [train.py:421] (5/8) Epoch 9, batch 36800, loss[loss=2.227, over 3010.00 frames. , ppl: 9.272491178387543] tot_loss[loss=2.273, over 5505440.00 frames. , ppl: 9.709210560867618], batch size: 70
+2022-12-13 10:06:43,639 INFO [train.py:421] (5/8) Epoch 9, batch 37000, loss[loss=2.17, over 4970.00 frames. , ppl: 8.757162755891397] tot_loss[loss=2.273, over 5498886.69 frames. , ppl: 9.712249983051596], batch size: 70
+2022-12-13 10:06:43,640 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 10:06:44,368 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.65039602425002
+2022-12-13 10:08:24,180 INFO [train.py:421] (5/8) Epoch 9, batch 37200, loss[loss=2.325, over 1260.00 frames. , ppl: 10.229466519124008] tot_loss[loss=2.273, over 5504997.66 frames. , ppl: 9.706466388905286], batch size: 70
+2022-12-13 10:10:01,428 INFO [train.py:421] (5/8) Epoch 9, batch 37400, loss[loss=2.351, over 2240.00 frames. , ppl: 10.49096758713473] tot_loss[loss=2.274, over 5451696.61 frames. , ppl: 9.71487670084088], batch size: 70
+2022-12-13 10:11:39,959 INFO [train.py:421] (5/8) Epoch 9, batch 37600, loss[loss=2.602, over 910.00 frames. , ppl: 13.490639179497093] tot_loss[loss=2.275, over 5438190.79 frames. , ppl: 9.72874674060322], batch size: 70
+2022-12-13 10:13:18,222 INFO [train.py:421] (5/8) Epoch 9, batch 37800, loss[loss=2.499, over 1400.00 frames. , ppl: 12.167682750133427] tot_loss[loss=2.275, over 5453056.20 frames. , ppl: 9.727052894828537], batch size: 70
+2022-12-13 10:14:59,565 INFO [train.py:421] (5/8) Epoch 9, batch 38000, loss[loss=2.447, over 1190.00 frames. , ppl: 11.559285396779512] tot_loss[loss=2.273, over 5484856.34 frames. , ppl: 9.71231705736586], batch size: 70
+2022-12-13 10:14:59,566 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 10:15:00,328 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.637123113571448
+2022-12-13 10:16:42,063 INFO [train.py:421] (5/8) Epoch 9, batch 38200, loss[loss=3.626, over 420.00 frames. , ppl: 37.57527525748518] tot_loss[loss=2.272, over 5549007.99 frames. , ppl: 9.70120227828701], batch size: 70
+2022-12-13 10:18:21,502 INFO [train.py:421] (5/8) Epoch 9, batch 38400, loss[loss=2.216, over 5460.00 frames. , ppl: 9.169719534821109] tot_loss[loss=2.273, over 5545817.55 frames. , ppl: 9.709597463688544], batch size: 70
+2022-12-13 10:20:03,168 INFO [train.py:421] (5/8) Epoch 9, batch 38600, loss[loss=2.721, over 700.00 frames. , ppl: 15.195132820713045] tot_loss[loss=2.273, over 5545089.01 frames. , ppl: 9.708727564469632], batch size: 70
+2022-12-13 10:21:45,084 INFO [train.py:421] (5/8) Epoch 9, batch 38800, loss[loss=2.536, over 980.00 frames. , ppl: 12.62464378011807] tot_loss[loss=2.273, over 5534243.01 frames. , ppl: 9.71293618653714], batch size: 70
+2022-12-13 10:23:26,884 INFO [train.py:421] (5/8) Epoch 9, batch 39000, loss[loss=2.372, over 1750.00 frames. , ppl: 10.723385285206136] tot_loss[loss=2.273, over 5538048.60 frames. , ppl: 9.711649492117255], batch size: 70
+2022-12-13 10:23:26,885 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 10:23:27,633 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.646513174570288
+2022-12-13 10:25:07,556 INFO [train.py:421] (5/8) Epoch 9, batch 39200, loss[loss=2.156, over 9240.00 frames. , ppl: 8.637122504651055] tot_loss[loss=2.272, over 5574544.15 frames. , ppl: 9.702985039221891], batch size: 70
+2022-12-13 10:26:46,247 INFO [train.py:421] (5/8) Epoch 9, batch 39400, loss[loss=2.174, over 4200.00 frames. , ppl: 8.792546640071414] tot_loss[loss=2.273, over 5564124.30 frames. , ppl: 9.7099641902594], batch size: 70
+2022-12-13 10:28:27,571 INFO [train.py:421] (5/8) Epoch 9, batch 39600, loss[loss=2.22, over 2800.00 frames. , ppl: 9.210732220616288] tot_loss[loss=2.272, over 5597366.19 frames. , ppl: 9.703185013913073], batch size: 70
+2022-12-13 10:30:04,664 INFO [train.py:421] (5/8) Epoch 9, batch 39800, loss[loss=2.367, over 1750.00 frames. , ppl: 10.664479292619408] tot_loss[loss=2.272, over 5580621.68 frames. , ppl: 9.700943856394323], batch size: 70
+2022-12-13 10:31:44,861 INFO [train.py:421] (5/8) Epoch 9, batch 40000, loss[loss=2.383, over 2380.00 frames. , ppl: 10.836174746305954] tot_loss[loss=2.274, over 5530496.55 frames. , ppl: 9.71362574631142], batch size: 70
+2022-12-13 10:31:44,861 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 10:31:45,622 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.648730734417011
+2022-12-13 10:33:26,258 INFO [train.py:421] (5/8) Epoch 9, batch 40200, loss[loss=2.705, over 700.00 frames. , ppl: 14.94903416853506] tot_loss[loss=2.273, over 5536848.19 frames. , ppl: 9.712085873374676], batch size: 70
+2022-12-13 10:35:02,356 INFO [train.py:421] (5/8) Epoch 9, batch 40400, loss[loss=2.247, over 3850.00 frames. , ppl: 9.455902059237102] tot_loss[loss=2.273, over 5554336.11 frames. , ppl: 9.712601149977363], batch size: 70
+2022-12-13 10:36:42,737 INFO [train.py:421] (5/8) Epoch 9, batch 40600, loss[loss=2.325, over 3570.00 frames. , ppl: 10.230093577075394] tot_loss[loss=2.274, over 5549920.47 frames. , ppl: 9.714761225420576], batch size: 70
+2022-12-13 10:38:21,179 INFO [train.py:421] (5/8) Epoch 9, batch 40800, loss[loss=2.31, over 1960.00 frames. , ppl: 10.076867461010384] tot_loss[loss=2.275, over 5496862.33 frames. , ppl: 9.729868136758194], batch size: 70
+2022-12-13 10:39:58,044 INFO [train.py:421] (5/8) Epoch 9, batch 41000, loss[loss=2.154, over 5250.00 frames. , ppl: 8.621880354024363] tot_loss[loss=2.276, over 5463837.11 frames. , ppl: 9.740078215069273], batch size: 70
+2022-12-13 10:39:58,044 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 10:39:58,775 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.268, over 211138.00 frames. , ppl: 9.663169461348792
+2022-12-13 10:41:39,566 INFO [train.py:421] (5/8) Epoch 9, batch 41200, loss[loss=2.163, over 5040.00 frames. , ppl: 8.693770746730822] tot_loss[loss=2.277, over 5437887.53 frames. , ppl: 9.747839742299984], batch size: 70
+2022-12-13 10:43:16,339 INFO [train.py:421] (5/8) Epoch 9, batch 41400, loss[loss=2.194, over 3150.00 frames. , ppl: 8.968624581062155] tot_loss[loss=2.277, over 5433340.58 frames. , ppl: 9.748085648526793], batch size: 70
+2022-12-13 10:44:58,173 INFO [train.py:421] (5/8) Epoch 9, batch 41600, loss[loss=2.221, over 2240.00 frames. , ppl: 9.213246706250027] tot_loss[loss=2.276, over 5456153.62 frames. , ppl: 9.737211529525437], batch size: 70
+2022-12-13 10:46:39,492 INFO [train.py:421] (5/8) Epoch 9, batch 41800, loss[loss=2.443, over 2100.00 frames. , ppl: 11.507893101910264] tot_loss[loss=2.276, over 5486112.71 frames. , ppl: 9.7377548663934], batch size: 70
+2022-12-13 10:48:20,323 INFO [train.py:421] (5/8) Epoch 9, batch 42000, loss[loss=2.231, over 3150.00 frames. , ppl: 9.306016414765988] tot_loss[loss=2.277, over 5487631.02 frames. , ppl: 9.745962236455481], batch size: 70
+2022-12-13 10:48:20,323 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 10:48:21,082 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.635082199742264
+2022-12-13 10:50:03,520 INFO [train.py:421] (5/8) Epoch 9, batch 42200, loss[loss=2.245, over 3990.00 frames. , ppl: 9.435926733371605] tot_loss[loss=2.278, over 5462867.20 frames. , ppl: 9.754219579310382], batch size: 70
+2022-12-13 10:51:44,294 INFO [train.py:421] (5/8) Epoch 9, batch 42400, loss[loss=2.242, over 3640.00 frames. , ppl: 9.414123947651955] tot_loss[loss=2.277, over 5434497.46 frames. , ppl: 9.749136870552366], batch size: 70
+2022-12-13 10:53:20,410 INFO [train.py:421] (5/8) Epoch 9, batch 42600, loss[loss=2.252, over 1890.00 frames. , ppl: 9.509389503040403] tot_loss[loss=2.278, over 5423452.03 frames. , ppl: 9.755437766277202], batch size: 70
+2022-12-13 10:54:59,747 INFO [train.py:421] (5/8) Epoch 9, batch 42800, loss[loss=2.154, over 8260.00 frames. , ppl: 8.61979146252503] tot_loss[loss=2.276, over 5446315.61 frames. , ppl: 9.73849473545099], batch size: 70
+2022-12-13 10:56:37,378 INFO [train.py:421] (5/8) Epoch 9, batch 43000, loss[loss=2.609, over 770.00 frames. , ppl: 13.586469688406448] tot_loss[loss=2.277, over 5435226.33 frames. , ppl: 9.743448918142349], batch size: 70
+2022-12-13 10:56:37,378 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 10:56:38,138 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.269, over 211138.00 frames. , ppl: 9.668268114386317
+2022-12-13 10:58:19,721 INFO [train.py:421] (5/8) Epoch 9, batch 43200, loss[loss=2.368, over 2800.00 frames. , ppl: 10.673706464760263] tot_loss[loss=2.276, over 5456077.67 frames. , ppl: 9.733530326358872], batch size: 70
+2022-12-13 11:00:01,371 INFO [train.py:421] (5/8) Epoch 9, batch 43400, loss[loss=2.244, over 2940.00 frames. , ppl: 9.426676165375419] tot_loss[loss=2.274, over 5471529.04 frames. , ppl: 9.72223311377905], batch size: 70
+2022-12-13 11:01:45,175 INFO [train.py:421] (5/8) Epoch 9, batch 43600, loss[loss=2.25, over 5250.00 frames. , ppl: 9.48698051840879] tot_loss[loss=2.275, over 5459130.53 frames. , ppl: 9.724466980488101], batch size: 70
+2022-12-13 11:03:26,870 INFO [train.py:421] (5/8) Epoch 9, batch 43800, loss[loss=2.287, over 2590.00 frames. , ppl: 9.843412555450804] tot_loss[loss=2.275, over 5459543.51 frames. , ppl: 9.72620286280544], batch size: 70
+2022-12-13 11:05:05,109 INFO [train.py:421] (5/8) Epoch 9, batch 44000, loss[loss=2.475, over 1260.00 frames. , ppl: 11.882347149747957] tot_loss[loss=2.275, over 5447684.86 frames. , ppl: 9.728143548981377], batch size: 70
+2022-12-13 11:05:05,109 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 11:05:05,860 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.264, over 211138.00 frames. , ppl: 9.623468208814646
+2022-12-13 11:06:45,137 INFO [train.py:421] (5/8) Epoch 9, batch 44200, loss[loss=2.224, over 3780.00 frames. , ppl: 9.242657169315786] tot_loss[loss=2.275, over 5457031.74 frames. , ppl: 9.727325040390513], batch size: 70
+2022-12-13 11:08:25,118 INFO [train.py:421] (5/8) Epoch 9, batch 44400, loss[loss=2.316, over 2240.00 frames. , ppl: 10.131844852145601] tot_loss[loss=2.274, over 5464537.78 frames. , ppl: 9.72252386464195], batch size: 70
+2022-12-13 11:10:01,990 INFO [train.py:421] (5/8) Epoch 9, batch 44600, loss[loss=2.27, over 4130.00 frames. , ppl: 9.683534264417908] tot_loss[loss=2.274, over 5438516.92 frames. , ppl: 9.722085954962525], batch size: 70
+2022-12-13 11:11:38,150 INFO [train.py:421] (5/8) Epoch 9, batch 44800, loss[loss=2.207, over 3500.00 frames. , ppl: 9.091404262491015] tot_loss[loss=2.275, over 5429456.05 frames. , ppl: 9.723619299106979], batch size: 70
+2022-12-13 11:13:19,789 INFO [train.py:421] (5/8) Epoch 9, batch 45000, loss[loss=2.233, over 3990.00 frames. , ppl: 9.33081458758153] tot_loss[loss=2.275, over 5420269.47 frames. , ppl: 9.724405096692236], batch size: 70
+2022-12-13 11:13:19,790 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 11:13:20,550 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.648610776092447
+2022-12-13 11:15:01,466 INFO [train.py:421] (5/8) Epoch 9, batch 45200, loss[loss=2.35, over 2030.00 frames. , ppl: 10.48446509360338] tot_loss[loss=2.275, over 5409234.98 frames. , ppl: 9.725992670970124], batch size: 70
+2022-12-13 11:16:44,206 INFO [train.py:421] (5/8) Epoch 9, batch 45400, loss[loss=2.551, over 770.00 frames. , ppl: 12.815991663552186] tot_loss[loss=2.274, over 5426252.86 frames. , ppl: 9.719879164560734], batch size: 70
+2022-12-13 11:18:27,658 INFO [train.py:421] (5/8) Epoch 9, batch 45600, loss[loss=2.165, over 4130.00 frames. , ppl: 8.71685075361261] tot_loss[loss=2.275, over 5416944.48 frames. , ppl: 9.728982139266598], batch size: 70
+2022-12-13 11:20:08,208 INFO [train.py:421] (5/8) Epoch 9, batch 45800, loss[loss=2.277, over 3010.00 frames. , ppl: 9.745555598131709] tot_loss[loss=2.274, over 5425249.76 frames. , ppl: 9.721904280509978], batch size: 70
+2022-12-13 11:21:46,359 INFO [train.py:421] (5/8) Epoch 9, batch 46000, loss[loss=2.235, over 4060.00 frames. , ppl: 9.345860488638639] tot_loss[loss=2.276, over 5410724.73 frames. , ppl: 9.736306329516196], batch size: 70
+2022-12-13 11:21:46,360 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 11:21:47,106 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.636034857644312
+2022-12-13 11:23:26,562 INFO [train.py:421] (5/8) Epoch 9, batch 46200, loss[loss=2.899, over 630.00 frames. , ppl: 18.14874871030834] tot_loss[loss=2.275, over 5428818.24 frames. , ppl: 9.731027806127514], batch size: 70
+2022-12-13 11:25:07,647 INFO [train.py:421] (5/8) Epoch 9, batch 46400, loss[loss=2.302, over 2730.00 frames. , ppl: 9.996210084888236] tot_loss[loss=2.275, over 5441362.63 frames. , ppl: 9.727215178348054], batch size: 70
+2022-12-13 11:26:44,519 INFO [train.py:421] (5/8) Epoch 9, batch 46600, loss[loss=2.455, over 1050.00 frames. , ppl: 11.649401861224998] tot_loss[loss=2.275, over 5452815.42 frames. , ppl: 9.725303044404185], batch size: 70
+2022-12-13 11:28:29,186 INFO [train.py:421] (5/8) Epoch 9, batch 46800, loss[loss=2.337, over 2450.00 frames. , ppl: 10.349208843318724] tot_loss[loss=2.275, over 5437906.34 frames. , ppl: 9.727227306409404], batch size: 70
+2022-12-13 11:30:14,119 INFO [train.py:421] (5/8) Epoch 9, batch 47000, loss[loss=2.36, over 1820.00 frames. , ppl: 10.591262307058113] tot_loss[loss=2.274, over 5473659.30 frames. , ppl: 9.722980395187275], batch size: 70
+2022-12-13 11:30:14,120 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 11:30:14,868 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.63486544043961
+2022-12-13 11:31:58,222 INFO [train.py:421] (5/8) Epoch 9, batch 47200, loss[loss=2.248, over 2380.00 frames. , ppl: 9.470041102177282] tot_loss[loss=2.273, over 5535700.20 frames. , ppl: 9.708315693572803], batch size: 70
+2022-12-13 11:33:39,674 INFO [train.py:421] (5/8) Epoch 9, batch 47400, loss[loss=2.232, over 7000.00 frames. , ppl: 9.315058222757946] tot_loss[loss=2.273, over 5510589.42 frames. , ppl: 9.713193940177256], batch size: 70
+2022-12-13 11:35:21,971 INFO [train.py:421] (5/8) Epoch 9, batch 47600, loss[loss=2.774, over 770.00 frames. , ppl: 16.028815370466177] tot_loss[loss=2.272, over 5521164.33 frames. , ppl: 9.702824435975172], batch size: 70
+2022-12-13 11:37:01,535 INFO [train.py:421] (5/8) Epoch 9, batch 47800, loss[loss=2.237, over 3430.00 frames. , ppl: 9.364166113318525] tot_loss[loss=2.272, over 5516719.95 frames. , ppl: 9.699330799580858], batch size: 70
+2022-12-13 11:38:43,295 INFO [train.py:421] (5/8) Epoch 9, batch 48000, loss[loss=2.545, over 1470.00 frames. , ppl: 12.743366924075115] tot_loss[loss=2.273, over 5465643.37 frames. , ppl: 9.709828339763428], batch size: 70
+2022-12-13 11:38:43,295 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 11:38:44,056 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.627387380087804
+2022-12-13 11:40:21,734 INFO [train.py:421] (5/8) Epoch 9, batch 48200, loss[loss=2.3, over 1890.00 frames. , ppl: 9.969694629126758] tot_loss[loss=2.274, over 5440259.24 frames. , ppl: 9.716647538134167], batch size: 70
+2022-12-13 11:42:04,578 INFO [train.py:421] (5/8) Epoch 9, batch 48400, loss[loss=2.313, over 1470.00 frames. , ppl: 10.099732803831888] tot_loss[loss=2.273, over 5484755.01 frames. , ppl: 9.705536717053901], batch size: 70
+2022-12-13 11:43:45,249 INFO [train.py:421] (5/8) Epoch 9, batch 48600, loss[loss=2.252, over 1960.00 frames. , ppl: 9.507483081053813] tot_loss[loss=2.273, over 5493929.72 frames. , ppl: 9.711988404620723], batch size: 70
+2022-12-13 11:45:22,551 INFO [train.py:421] (5/8) Epoch 9, batch 48800, loss[loss=2.158, over 7070.00 frames. , ppl: 8.657534052700264] tot_loss[loss=2.273, over 5501332.40 frames. , ppl: 9.710318283532526], batch size: 70
+2022-12-13 11:47:02,920 INFO [train.py:421] (5/8) Epoch 9, batch 49000, loss[loss=2.822, over 630.00 frames. , ppl: 16.812642848547604] tot_loss[loss=2.273, over 5503244.29 frames. , ppl: 9.705348938826788], batch size: 70
+2022-12-13 11:47:02,921 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 11:47:03,679 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.644904227803092
+2022-12-13 11:48:42,689 INFO [train.py:421] (5/8) Epoch 9, batch 49200, loss[loss=2.387, over 1400.00 frames. , ppl: 10.87906686364079] tot_loss[loss=2.273, over 5495629.10 frames. , ppl: 9.709598321451603], batch size: 70
+2022-12-13 11:50:20,797 INFO [train.py:421] (5/8) Epoch 9, batch 49400, loss[loss=2.636, over 840.00 frames. , ppl: 13.951604792650638] tot_loss[loss=2.273, over 5508205.72 frames. , ppl: 9.71293500725242], batch size: 70
+2022-12-13 11:52:05,013 INFO [train.py:421] (5/8) Epoch 9, batch 49600, loss[loss=2.356, over 2520.00 frames. , ppl: 10.55133429251107] tot_loss[loss=2.273, over 5543440.95 frames. , ppl: 9.706285331392703], batch size: 70
+2022-12-13 11:53:47,506 INFO [train.py:421] (5/8) Epoch 9, batch 49800, loss[loss=4.105, over 350.00 frames. , ppl: 60.6524322552751] tot_loss[loss=2.274, over 5488100.98 frames. , ppl: 9.721897965334767], batch size: 70
+2022-12-13 11:55:28,993 INFO [train.py:421] (5/8) Epoch 9, batch 50000, loss[loss=2.166, over 4690.00 frames. , ppl: 8.719209612842416] tot_loss[loss=2.274, over 5494930.71 frames. , ppl: 9.719699808388551], batch size: 70
+2022-12-13 11:55:28,994 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 11:55:29,755 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.643979240614769
+2022-12-13 11:57:09,485 INFO [train.py:421] (5/8) Epoch 9, batch 50200, loss[loss=2.072, over 3290.00 frames. , ppl: 7.9379110853892065] tot_loss[loss=2.274, over 5508442.00 frames. , ppl: 9.720136253582156], batch size: 70
+2022-12-13 11:58:51,076 INFO [train.py:421] (5/8) Epoch 9, batch 50400, loss[loss=2.785, over 700.00 frames. , ppl: 16.194620666382495] tot_loss[loss=2.275, over 5515268.44 frames. , ppl: 9.72571859125165], batch size: 70
+2022-12-13 12:00:30,186 INFO [train.py:421] (5/8) Epoch 9, batch 50600, loss[loss=2.177, over 5040.00 frames. , ppl: 8.818416272077467] tot_loss[loss=2.273, over 5557981.59 frames. , ppl: 9.708083649642557], batch size: 70
+2022-12-13 12:02:07,611 INFO [train.py:421] (5/8) Epoch 9, batch 50800, loss[loss=2.235, over 7280.00 frames. , ppl: 9.349891215561934] tot_loss[loss=2.274, over 5507539.62 frames. , ppl: 9.72026938223746], batch size: 70
+2022-12-13 12:03:46,572 INFO [train.py:421] (5/8) Epoch 9, batch 51000, loss[loss=2.225, over 6020.00 frames. , ppl: 9.255362368288312] tot_loss[loss=2.273, over 5526659.04 frames. , ppl: 9.71246627739128], batch size: 70
+2022-12-13 12:03:46,572 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 12:03:47,301 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.639113099787668
+2022-12-13 12:05:25,875 INFO [train.py:421] (5/8) Epoch 9, batch 51200, loss[loss=2.318, over 1680.00 frames. , ppl: 10.151222183416198] tot_loss[loss=2.274, over 5508290.57 frames. , ppl: 9.715635991999738], batch size: 70
+2022-12-13 12:07:04,769 INFO [train.py:421] (5/8) Epoch 9, batch 51400, loss[loss=2.885, over 630.00 frames. , ppl: 17.907534533260975] tot_loss[loss=2.273, over 5532671.49 frames. , ppl: 9.703980257956117], batch size: 70
+2022-12-13 12:08:46,949 INFO [train.py:421] (5/8) Epoch 9, batch 51600, loss[loss=2.189, over 4060.00 frames. , ppl: 8.92758753341209] tot_loss[loss=2.27, over 5594447.92 frames. , ppl: 9.681784179585101], batch size: 70
+2022-12-13 12:10:24,705 INFO [train.py:421] (5/8) Epoch 9, batch 51800, loss[loss=2.778, over 770.00 frames. , ppl: 16.089132176117] tot_loss[loss=2.269, over 5606996.92 frames. , ppl: 9.672615311437738], batch size: 70
+2022-12-13 12:12:05,979 INFO [train.py:421] (5/8) Epoch 9, batch 52000, loss[loss=2.182, over 4200.00 frames. , ppl: 8.867979557980654] tot_loss[loss=2.268, over 5663695.35 frames. , ppl: 9.662325163975998], batch size: 70
+2022-12-13 12:12:05,980 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 12:12:06,730 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.646610262463758
+2022-12-13 12:13:45,404 INFO [train.py:421] (5/8) Epoch 9, batch 52200, loss[loss=2.366, over 1820.00 frames. , ppl: 10.654262956964676] tot_loss[loss=2.269, over 5641363.44 frames. , ppl: 9.665932373699903], batch size: 70
+2022-12-13 12:15:24,528 INFO [train.py:421] (5/8) Epoch 9, batch 52400, loss[loss=2.218, over 3010.00 frames. , ppl: 9.186540019132636] tot_loss[loss=2.27, over 5581669.15 frames. , ppl: 9.68229616557168], batch size: 70
+2022-12-13 12:17:07,157 INFO [train.py:421] (5/8) Epoch 9, batch 52600, loss[loss=2.323, over 1400.00 frames. , ppl: 10.210729400874389] tot_loss[loss=2.271, over 5584291.88 frames. , ppl: 9.684665034427036], batch size: 70
+2022-12-13 12:18:50,844 INFO [train.py:421] (5/8) Epoch 9, batch 52800, loss[loss=2.415, over 2030.00 frames. , ppl: 11.191736405468932] tot_loss[loss=2.269, over 5607240.88 frames. , ppl: 9.673776922567338], batch size: 70
+2022-12-13 12:20:28,986 INFO [train.py:421] (5/8) Epoch 9, batch 53000, loss[loss=2.398, over 910.00 frames. , ppl: 11.00391119370434] tot_loss[loss=2.269, over 5600377.19 frames. , ppl: 9.665719984391616], batch size: 70
+2022-12-13 12:20:28,987 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 12:20:29,752 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.642737499169275
+2022-12-13 12:22:10,639 INFO [train.py:421] (5/8) Epoch 9, batch 53200, loss[loss=2.325, over 1190.00 frames. , ppl: 10.231624422188302] tot_loss[loss=2.269, over 5579226.99 frames. , ppl: 9.669990383846939], batch size: 70
+2022-12-13 12:23:51,929 INFO [train.py:421] (5/8) Epoch 9, batch 53400, loss[loss=2.139, over 5040.00 frames. , ppl: 8.494563628705583] tot_loss[loss=2.269, over 5607797.93 frames. , ppl: 9.670695112848705], batch size: 70
+2022-12-13 12:25:35,097 INFO [train.py:421] (5/8) Epoch 9, batch 53600, loss[loss=2.167, over 5390.00 frames. , ppl: 8.734502516733324] tot_loss[loss=2.268, over 5629889.60 frames. , ppl: 9.662942826932602], batch size: 70
+2022-12-13 12:27:15,425 INFO [train.py:421] (5/8) Epoch 9, batch 53800, loss[loss=2.795, over 630.00 frames. , ppl: 16.369966213335637] tot_loss[loss=2.269, over 5611061.23 frames. , ppl: 9.671157590880599], batch size: 70
+2022-12-13 12:28:55,762 INFO [train.py:421] (5/8) Epoch 9, batch 54000, loss[loss=2.399, over 910.00 frames. , ppl: 11.008585518396817] tot_loss[loss=2.27, over 5601546.83 frames. , ppl: 9.681291273817438], batch size: 70
+2022-12-13 12:28:55,763 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 12:28:56,510 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.633422404126998
+2022-12-13 12:30:35,831 INFO [train.py:421] (5/8) Epoch 9, batch 54200, loss[loss=2.185, over 13650.00 frames. , ppl: 8.887033900438285] tot_loss[loss=2.27, over 5603327.03 frames. , ppl: 9.678986521485989], batch size: 70
+2022-12-13 12:32:16,552 INFO [train.py:421] (5/8) Epoch 9, batch 54400, loss[loss=2.345, over 2520.00 frames. , ppl: 10.432433402967758] tot_loss[loss=2.27, over 5633267.12 frames. , ppl: 9.677030723025775], batch size: 70
+2022-12-13 12:33:55,484 INFO [train.py:421] (5/8) Epoch 9, batch 54600, loss[loss=2.409, over 2100.00 frames. , ppl: 11.12237447957013] tot_loss[loss=2.272, over 5569910.67 frames. , ppl: 9.696782824006826], batch size: 70
+2022-12-13 12:35:38,800 INFO [train.py:421] (5/8) Epoch 9, batch 54800, loss[loss=2.396, over 1400.00 frames. , ppl: 10.98135960082289] tot_loss[loss=2.273, over 5560036.67 frames. , ppl: 9.706603055271081], batch size: 70
+2022-12-13 12:37:17,706 INFO [train.py:421] (5/8) Epoch 9, batch 55000, loss[loss=2.458, over 1540.00 frames. , ppl: 11.686478877966259] tot_loss[loss=2.273, over 5556861.99 frames. , ppl: 9.706513868198005], batch size: 70
+2022-12-13 12:37:17,707 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 12:37:18,465 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.64445171527035
+2022-12-13 12:39:02,245 INFO [train.py:421] (5/8) Epoch 9, batch 55200, loss[loss=2.403, over 1960.00 frames. , ppl: 11.057283768701723] tot_loss[loss=2.272, over 5553165.93 frames. , ppl: 9.702431892484707], batch size: 70
+2022-12-13 12:40:39,118 INFO [train.py:421] (5/8) Epoch 9, batch 55400, loss[loss=2.263, over 1890.00 frames. , ppl: 9.615345434875591] tot_loss[loss=2.273, over 5545650.65 frames. , ppl: 9.704907081453488], batch size: 70
+2022-12-13 12:42:18,508 INFO [train.py:421] (5/8) Epoch 9, batch 55600, loss[loss=2.653, over 770.00 frames. , ppl: 14.194283650895482] tot_loss[loss=2.272, over 5550677.97 frames. , ppl: 9.6993463656403], batch size: 70
+2022-12-13 12:43:58,844 INFO [train.py:421] (5/8) Epoch 9, batch 55800, loss[loss=2.421, over 1680.00 frames. , ppl: 11.260404551048618] tot_loss[loss=2.272, over 5538988.18 frames. , ppl: 9.703548653937748], batch size: 70
+2022-12-13 12:45:38,605 INFO [train.py:421] (5/8) Epoch 9, batch 56000, loss[loss=2.17, over 6300.00 frames. , ppl: 8.75603067994936] tot_loss[loss=2.273, over 5519203.04 frames. , ppl: 9.70814638944808], batch size: 70
+2022-12-13 12:45:38,605 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 12:45:39,335 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.264, over 211138.00 frames. , ppl: 9.625552251361995
+2022-12-13 12:47:19,624 INFO [train.py:421] (5/8) Epoch 9, batch 56200, loss[loss=2.216, over 5740.00 frames. , ppl: 9.173616218401607] tot_loss[loss=2.274, over 5508294.10 frames. , ppl: 9.71473408641725], batch size: 70
+2022-12-13 12:48:56,802 INFO [train.py:421] (5/8) Epoch 9, batch 56400, loss[loss=2.353, over 2520.00 frames. , ppl: 10.518610534340471] tot_loss[loss=2.274, over 5493824.71 frames. , ppl: 9.721010405125913], batch size: 70
+2022-12-13 12:50:35,595 INFO [train.py:421] (5/8) Epoch 9, batch 56600, loss[loss=2.186, over 4340.00 frames. , ppl: 8.902792818651932] tot_loss[loss=2.273, over 5509624.74 frames. , ppl: 9.7100671140327], batch size: 70
+2022-12-13 12:52:13,282 INFO [train.py:421] (5/8) Epoch 9, batch 56800, loss[loss=2.209, over 4550.00 frames. , ppl: 9.105565866509457] tot_loss[loss=2.274, over 5487134.62 frames. , ppl: 9.71652130668753], batch size: 70
+2022-12-13 12:53:52,633 INFO [train.py:421] (5/8) Epoch 9, batch 57000, loss[loss=2.712, over 630.00 frames. , ppl: 15.06388902952954] tot_loss[loss=2.274, over 5473064.40 frames. , ppl: 9.717975224616591], batch size: 70
+2022-12-13 12:53:52,634 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 12:53:53,394 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.636780791774152
+2022-12-13 12:55:34,865 INFO [train.py:421] (5/8) Epoch 9, batch 57200, loss[loss=2.263, over 2030.00 frames. , ppl: 9.613547012210512] tot_loss[loss=2.272, over 5518799.19 frames. , ppl: 9.702642330757843], batch size: 70
+2022-12-13 12:57:12,827 INFO [train.py:421] (5/8) Epoch 9, batch 57400, loss[loss=2.176, over 5040.00 frames. , ppl: 8.807507992433095] tot_loss[loss=2.272, over 5533297.28 frames. , ppl: 9.69953937728869], batch size: 70
+2022-12-13 12:58:55,750 INFO [train.py:421] (5/8) Epoch 9, batch 57600, loss[loss=2.357, over 1330.00 frames. , ppl: 10.558073450657792] tot_loss[loss=2.272, over 5550341.55 frames. , ppl: 9.694656287973904], batch size: 70
+2022-12-13 13:00:36,185 INFO [train.py:421] (5/8) Epoch 9, batch 57800, loss[loss=2.164, over 12740.00 frames. , ppl: 8.705464159814307] tot_loss[loss=2.271, over 5561010.19 frames. , ppl: 9.693113331950215], batch size: 70
+2022-12-13 13:02:18,880 INFO [train.py:421] (5/8) Epoch 9, batch 58000, loss[loss=2.221, over 5740.00 frames. , ppl: 9.219117940450975] tot_loss[loss=2.27, over 5610727.58 frames. , ppl: 9.677403241886967], batch size: 70
+2022-12-13 13:02:18,880 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 13:02:19,612 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.635025157347723
+2022-12-13 13:04:04,620 INFO [train.py:421] (5/8) Epoch 9, batch 58200, loss[loss=2.742, over 700.00 frames. , ppl: 15.519646837921274] tot_loss[loss=2.269, over 5642566.22 frames. , ppl: 9.670141886078861], batch size: 70
+2022-12-13 13:05:43,783 INFO [train.py:421] (5/8) Epoch 9, batch 58400, loss[loss=2.191, over 6090.00 frames. , ppl: 8.941891916615987] tot_loss[loss=2.27, over 5620522.37 frames. , ppl: 9.678161657641322], batch size: 70
+2022-12-13 13:07:23,478 INFO [train.py:421] (5/8) Epoch 9, batch 58600, loss[loss=2.533, over 770.00 frames. , ppl: 12.591812631323462] tot_loss[loss=2.269, over 5638542.84 frames. , ppl: 9.671110943898508], batch size: 70
+2022-12-13 13:09:02,105 INFO [train.py:421] (5/8) Epoch 9, batch 58800, loss[loss=2.842, over 770.00 frames. , ppl: 17.154585229078922] tot_loss[loss=2.27, over 5631908.12 frames. , ppl: 9.676756379172561], batch size: 70
+2022-12-13 13:10:46,549 INFO [train.py:421] (5/8) Epoch 9, batch 59000, loss[loss=2.175, over 8750.00 frames. , ppl: 8.804712183546949] tot_loss[loss=2.27, over 5634542.78 frames. , ppl: 9.6754645502857], batch size: 70
+2022-12-13 13:10:46,549 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 13:10:47,279 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.634386305600566
+2022-12-13 13:12:26,044 INFO [train.py:421] (5/8) Epoch 9, batch 59200, loss[loss=2.436, over 1470.00 frames. , ppl: 11.430360141543083] tot_loss[loss=2.27, over 5640649.89 frames. , ppl: 9.675536555203411], batch size: 70
+2022-12-13 13:14:06,392 INFO [train.py:421] (5/8) Epoch 9, batch 59400, loss[loss=2.195, over 5040.00 frames. , ppl: 8.982567555688714] tot_loss[loss=2.269, over 5636918.18 frames. , ppl: 9.673530342662549], batch size: 70
+2022-12-13 13:15:48,493 INFO [train.py:421] (5/8) Epoch 9, batch 59600, loss[loss=2.352, over 2240.00 frames. , ppl: 10.503733681227004] tot_loss[loss=2.269, over 5667297.21 frames. , ppl: 9.671310562572783], batch size: 70
+2022-12-13 13:17:26,967 INFO [train.py:421] (5/8) Epoch 9, batch 59800, loss[loss=2.263, over 6580.00 frames. , ppl: 9.613509473141617] tot_loss[loss=2.27, over 5636046.81 frames. , ppl: 9.67729101577833], batch size: 70
+2022-12-13 13:19:05,928 INFO [train.py:421] (5/8) Epoch 9, batch 60000, loss[loss=2.47, over 1050.00 frames. , ppl: 11.819632318425123] tot_loss[loss=2.269, over 5664989.27 frames. , ppl: 9.672860567229083], batch size: 70
+2022-12-13 13:19:05,928 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 13:19:06,687 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.63870508346662
+2022-12-13 13:20:45,779 INFO [train.py:421] (5/8) Epoch 9, batch 60200, loss[loss=2.306, over 2240.00 frames. , ppl: 10.033951273047752] tot_loss[loss=2.269, over 5671374.53 frames. , ppl: 9.669317700058683], batch size: 70
+2022-12-13 13:22:24,272 INFO [train.py:421] (5/8) Epoch 9, batch 60400, loss[loss=2.248, over 2940.00 frames. , ppl: 9.470248622049729] tot_loss[loss=2.268, over 5681499.73 frames. , ppl: 9.663115960386385], batch size: 70
+2022-12-13 13:24:05,149 INFO [train.py:421] (5/8) Epoch 9, batch 60600, loss[loss=2.218, over 9380.00 frames. , ppl: 9.185204196083331] tot_loss[loss=2.268, over 5698872.95 frames. , ppl: 9.657783743123966], batch size: 70
+2022-12-13 13:25:47,347 INFO [train.py:421] (5/8) Epoch 9, batch 60800, loss[loss=2.349, over 2030.00 frames. , ppl: 10.47867651073745] tot_loss[loss=2.268, over 5687392.32 frames. , ppl: 9.656561024908024], batch size: 70
+2022-12-13 13:27:25,583 INFO [train.py:421] (5/8) Epoch 9, batch 61000, loss[loss=2.679, over 700.00 frames. , ppl: 14.576999777825689] tot_loss[loss=2.268, over 5655347.80 frames. , ppl: 9.658432212504005], batch size: 70
+2022-12-13 13:27:25,583 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 13:27:26,343 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.267, over 211138.00 frames. , ppl: 9.648386572270214
+2022-12-13 13:29:06,953 INFO [train.py:421] (5/8) Epoch 9, batch 61200, loss[loss=2.462, over 1120.00 frames. , ppl: 11.728908697679215] tot_loss[loss=2.27, over 5605384.67 frames. , ppl: 9.676229925494566], batch size: 70
+2022-12-13 13:30:44,585 INFO [train.py:421] (5/8) Epoch 9, batch 61400, loss[loss=2.435, over 1330.00 frames. , ppl: 11.410129070463222] tot_loss[loss=2.271, over 5539368.24 frames. , ppl: 9.689391139549445], batch size: 70
+2022-12-13 13:32:23,424 INFO [train.py:421] (5/8) Epoch 9, batch 61600, loss[loss=2.312, over 1960.00 frames. , ppl: 10.089914815887656] tot_loss[loss=2.272, over 5523001.01 frames. , ppl: 9.693998871787747], batch size: 70
+2022-12-13 13:34:08,412 INFO [train.py:421] (5/8) Epoch 9, batch 61800, loss[loss=2.328, over 1400.00 frames. , ppl: 10.25595883804131] tot_loss[loss=2.272, over 5527398.77 frames. , ppl: 9.697066284178995], batch size: 70
+2022-12-13 13:35:49,227 INFO [train.py:421] (5/8) Epoch 9, batch 62000, loss[loss=2.256, over 2660.00 frames. , ppl: 9.54816103406203] tot_loss[loss=2.27, over 5584734.71 frames. , ppl: 9.681112668723639], batch size: 70
+2022-12-13 13:35:49,228 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 13:35:49,994 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.263, over 211138.00 frames. , ppl: 9.615221936937656
+2022-12-13 13:37:32,296 INFO [train.py:421] (5/8) Epoch 9, batch 62200, loss[loss=2.381, over 2590.00 frames. , ppl: 10.820426579087254] tot_loss[loss=2.271, over 5568577.89 frames. , ppl: 9.68545910067154], batch size: 70
+2022-12-13 13:39:16,055 INFO [train.py:421] (5/8) Epoch 9, batch 62400, loss[loss=2.453, over 1400.00 frames. , ppl: 11.621410401153456] tot_loss[loss=2.27, over 5593664.26 frames. , ppl: 9.677518649942298], batch size: 70
+2022-12-13 13:40:53,821 INFO [train.py:421] (5/8) Epoch 9, batch 62600, loss[loss=2.273, over 2730.00 frames. , ppl: 9.712065684632018] tot_loss[loss=2.27, over 5562981.89 frames. , ppl: 9.67937020748424], batch size: 70
+2022-12-13 13:42:33,899 INFO [train.py:421] (5/8) Epoch 9, batch 62800, loss[loss=2.966, over 560.00 frames. , ppl: 19.418355623135156] tot_loss[loss=2.272, over 5507230.71 frames. , ppl: 9.696657157548056], batch size: 70
+2022-12-13 13:44:11,462 INFO [train.py:421] (5/8) Epoch 9, batch 63000, loss[loss=2.185, over 5530.00 frames. , ppl: 8.886844867033448] tot_loss[loss=2.271, over 5532653.23 frames. , ppl: 9.692153781173863], batch size: 70
+2022-12-13 13:44:11,463 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 13:44:12,223 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.266, over 211138.00 frames. , ppl: 9.639786507154147
+2022-12-13 13:45:47,947 INFO [train.py:421] (5/8) Epoch 9, batch 63200, loss[loss=2.464, over 1190.00 frames. , ppl: 11.752762862500466] tot_loss[loss=2.271, over 5520230.33 frames. , ppl: 9.692003967696028], batch size: 70
+2022-12-13 13:47:29,990 INFO [train.py:421] (5/8) Epoch 9, batch 63400, loss[loss=2.637, over 840.00 frames. , ppl: 13.976061299703595] tot_loss[loss=2.271, over 5537542.35 frames. , ppl: 9.685456117332171], batch size: 70
+2022-12-13 13:49:09,473 INFO [train.py:421] (5/8) Epoch 9, batch 63600, loss[loss=2.259, over 4130.00 frames. , ppl: 9.574140019133983] tot_loss[loss=2.269, over 5579895.26 frames. , ppl: 9.673160775305266], batch size: 70
+2022-12-13 13:50:48,800 INFO [train.py:421] (5/8) Epoch 9, batch 63800, loss[loss=2.383, over 3080.00 frames. , ppl: 10.842224821870488] tot_loss[loss=2.269, over 5572629.27 frames. , ppl: 9.674076033172561], batch size: 70
+2022-12-13 13:52:29,000 INFO [train.py:421] (5/8) Epoch 9, batch 64000, loss[loss=2.62, over 840.00 frames. , ppl: 13.730379867217586] tot_loss[loss=2.27, over 5553662.92 frames. , ppl: 9.68015987403377], batch size: 70
+2022-12-13 13:52:29,001 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 13:52:29,747 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.627401629350286
+2022-12-13 13:54:05,581 INFO [train.py:421] (5/8) Epoch 9, batch 64200, loss[loss=2.193, over 5320.00 frames. , ppl: 8.961616082976569] tot_loss[loss=2.271, over 5524893.19 frames. , ppl: 9.686226912361166], batch size: 70
+2022-12-13 13:55:48,758 INFO [train.py:421] (5/8) Epoch 9, batch 64400, loss[loss=2.451, over 1400.00 frames. , ppl: 11.600502424116673] tot_loss[loss=2.27, over 5546332.94 frames. , ppl: 9.68072398594456], batch size: 70
+2022-12-13 13:57:26,853 INFO [train.py:421] (5/8) Epoch 9, batch 64600, loss[loss=2.266, over 2100.00 frames. , ppl: 9.643148612532295] tot_loss[loss=2.271, over 5504853.54 frames. , ppl: 9.693103551757359], batch size: 70
+2022-12-13 13:59:05,983 INFO [train.py:421] (5/8) Epoch 9, batch 64800, loss[loss=2.258, over 2590.00 frames. , ppl: 9.566278167235717] tot_loss[loss=2.272, over 5521023.82 frames. , ppl: 9.700204576714693], batch size: 70
+2022-12-13 14:00:44,182 INFO [train.py:421] (5/8) Epoch 9, batch 65000, loss[loss=2.228, over 4270.00 frames. , ppl: 9.283833664290862] tot_loss[loss=2.273, over 5522829.79 frames. , ppl: 9.703933510634014], batch size: 70
+2022-12-13 14:00:44,182 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 14:00:44,934 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.633545025291745
+2022-12-13 14:02:29,184 INFO [train.py:421] (5/8) Epoch 9, batch 65200, loss[loss=2.189, over 4410.00 frames. , ppl: 8.927194129134007] tot_loss[loss=2.273, over 5487515.77 frames. , ppl: 9.707521302768674], batch size: 70
+2022-12-13 14:04:10,420 INFO [train.py:421] (5/8) Epoch 9, batch 65400, loss[loss=3.064, over 560.00 frames. , ppl: 21.421281626821663] tot_loss[loss=2.273, over 5495000.32 frames. , ppl: 9.706962669717564], batch size: 70
+2022-12-13 14:05:48,253 INFO [train.py:421] (5/8) Epoch 9, batch 65600, loss[loss=2.372, over 910.00 frames. , ppl: 10.713717034009214] tot_loss[loss=2.273, over 5465360.16 frames. , ppl: 9.713301016424507], batch size: 70
+2022-12-13 14:07:28,832 INFO [train.py:421] (5/8) Epoch 9, batch 65800, loss[loss=2.371, over 1540.00 frames. , ppl: 10.703396056250739] tot_loss[loss=2.274, over 5462532.47 frames. , ppl: 9.718342700069407], batch size: 70
+2022-12-13 14:09:11,363 INFO [train.py:421] (5/8) Epoch 9, batch 66000, loss[loss=6.208, over 210.00 frames. , ppl: 496.94060758421836] tot_loss[loss=2.273, over 5466951.62 frames. , ppl: 9.708780651893806], batch size: 70
+2022-12-13 14:09:11,364 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 14:09:12,123 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.633080213783197
+2022-12-13 14:10:51,440 INFO [train.py:421] (5/8) Epoch 9, batch 66200, loss[loss=2.097, over 6160.00 frames. , ppl: 8.145283596809694] tot_loss[loss=2.274, over 5444111.25 frames. , ppl: 9.720039217761835], batch size: 70
+2022-12-13 14:12:30,194 INFO [train.py:421] (5/8) Epoch 9, batch 66400, loss[loss=2.523, over 840.00 frames. , ppl: 12.462182816933678] tot_loss[loss=2.274, over 5464857.94 frames. , ppl: 9.719944872714063], batch size: 70
+2022-12-13 14:14:07,586 INFO [train.py:421] (5/8) Epoch 9, batch 66600, loss[loss=2.427, over 1190.00 frames. , ppl: 11.32262512896054] tot_loss[loss=2.275, over 5443704.74 frames. , ppl: 9.725960828742945], batch size: 70
+2022-12-13 14:15:45,368 INFO [train.py:421] (5/8) Epoch 9, batch 66800, loss[loss=2.214, over 4060.00 frames. , ppl: 9.150585609793064] tot_loss[loss=2.274, over 5427791.78 frames. , ppl: 9.722701756371777], batch size: 70
+2022-12-13 14:17:23,668 INFO [train.py:421] (5/8) Epoch 9, batch 67000, loss[loss=2.657, over 700.00 frames. , ppl: 14.259411809768945] tot_loss[loss=2.274, over 5419928.86 frames. , ppl: 9.717172912595945], batch size: 70
+2022-12-13 14:17:23,669 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 14:17:24,429 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.635444426828531
+2022-12-13 14:19:04,965 INFO [train.py:421] (5/8) Epoch 9, batch 67200, loss[loss=2.373, over 1190.00 frames. , ppl: 10.729245341104752] tot_loss[loss=2.274, over 5418983.34 frames. , ppl: 9.720983542586245], batch size: 70
+2022-12-13 14:20:42,270 INFO [train.py:421] (5/8) Epoch 9, batch 67400, loss[loss=2.271, over 2240.00 frames. , ppl: 9.687868759026603] tot_loss[loss=2.276, over 5367680.05 frames. , ppl: 9.7394831030767], batch size: 70
+2022-12-13 14:22:19,562 INFO [train.py:421] (5/8) Epoch 9, batch 67600, loss[loss=2.498, over 1960.00 frames. , ppl: 12.156566778297101] tot_loss[loss=2.277, over 5363619.69 frames. , ppl: 9.742993646401542], batch size: 70
+2022-12-13 14:23:58,429 INFO [train.py:421] (5/8) Epoch 9, batch 67800, loss[loss=2.376, over 1330.00 frames. , ppl: 10.763078054396404] tot_loss[loss=2.277, over 5361381.67 frames. , ppl: 9.751626654841957], batch size: 70
+2022-12-13 14:25:36,735 INFO [train.py:421] (5/8) Epoch 9, batch 68000, loss[loss=2.361, over 2590.00 frames. , ppl: 10.597324328535072] tot_loss[loss=2.278, over 5342339.27 frames. , ppl: 9.753122900665465], batch size: 70
+2022-12-13 14:25:36,735 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 14:25:37,495 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.264, over 211138.00 frames. , ppl: 9.621968490265408
+2022-12-13 14:27:17,453 INFO [train.py:421] (5/8) Epoch 9, batch 68200, loss[loss=2.195, over 3990.00 frames. , ppl: 8.977361354099113] tot_loss[loss=2.278, over 5332085.12 frames. , ppl: 9.759224714956732], batch size: 70
+2022-12-13 14:28:57,912 INFO [train.py:421] (5/8) Epoch 9, batch 68400, loss[loss=2.155, over 4340.00 frames. , ppl: 8.631302280760218] tot_loss[loss=2.278, over 5352504.78 frames. , ppl: 9.753129030084937], batch size: 70
+2022-12-13 14:30:42,246 INFO [train.py:421] (5/8) Epoch 9, batch 68600, loss[loss=2.321, over 2100.00 frames. , ppl: 10.181342443844928] tot_loss[loss=2.277, over 5361747.48 frames. , ppl: 9.751136067077654], batch size: 70
+2022-12-13 14:32:21,733 INFO [train.py:421] (5/8) Epoch 9, batch 68800, loss[loss=2.209, over 5320.00 frames. , ppl: 9.103210547487445] tot_loss[loss=2.277, over 5377124.16 frames. , ppl: 9.749571049449088], batch size: 70
+2022-12-13 14:34:01,335 INFO [train.py:421] (5/8) Epoch 9, batch 69000, loss[loss=2.925, over 630.00 frames. , ppl: 18.62675358093956] tot_loss[loss=2.278, over 5342576.10 frames. , ppl: 9.757579966560757], batch size: 70
+2022-12-13 14:34:01,335 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 14:34:02,096 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.634126784178106
+2022-12-13 14:35:35,880 INFO [train.py:421] (5/8) Epoch 9, batch 69200, loss[loss=2.53, over 1540.00 frames. , ppl: 12.554126678716717] tot_loss[loss=2.277, over 5374949.42 frames. , ppl: 9.744366876579027], batch size: 70
+2022-12-13 14:37:12,959 INFO [train.py:421] (5/8) Epoch 9, batch 69400, loss[loss=2.624, over 1050.00 frames. , ppl: 13.79185652381939] tot_loss[loss=2.276, over 5385515.41 frames. , ppl: 9.739649313797665], batch size: 70
+2022-12-13 14:38:55,420 INFO [train.py:421] (5/8) Epoch 9, batch 69600, loss[loss=2.26, over 1890.00 frames. , ppl: 9.583601173776183] tot_loss[loss=2.276, over 5378672.34 frames. , ppl: 9.740538333506134], batch size: 70
+2022-12-13 14:40:38,310 INFO [train.py:421] (5/8) Epoch 9, batch 69800, loss[loss=2.149, over 8050.00 frames. , ppl: 8.575923266616321] tot_loss[loss=2.275, over 5414634.09 frames. , ppl: 9.726128994598495], batch size: 70
+2022-12-13 14:42:20,448 INFO [train.py:421] (5/8) Epoch 9, batch 70000, loss[loss=2.254, over 3150.00 frames. , ppl: 9.525841337685748] tot_loss[loss=2.275, over 5404138.45 frames. , ppl: 9.729479340463415], batch size: 70
+2022-12-13 14:42:20,448 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 14:42:21,216 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.265, over 211138.00 frames. , ppl: 9.632411551904289
+2022-12-13 14:44:01,631 INFO [train.py:421] (5/8) Epoch 9, batch 70200, loss[loss=2.327, over 1750.00 frames. , ppl: 10.249736040689791] tot_loss[loss=2.276, over 5374688.53 frames. , ppl: 9.736120805612341], batch size: 70
+2022-12-13 14:45:43,982 INFO [train.py:421] (5/8) Epoch 9, batch 70400, loss[loss=2.241, over 2870.00 frames. , ppl: 9.406546751035163] tot_loss[loss=2.276, over 5376888.14 frames. , ppl: 9.7372659251336], batch size: 70
+2022-12-13 14:47:24,087 INFO [train.py:421] (5/8) Epoch 9, batch 70600, loss[loss=2.388, over 910.00 frames. , ppl: 10.891558209312503] tot_loss[loss=2.276, over 5374035.88 frames. , ppl: 9.740465282830737], batch size: 70
+2022-12-13 14:49:06,611 INFO [train.py:421] (5/8) Epoch 9, batch 70800, loss[loss=2.278, over 3010.00 frames. , ppl: 9.761129033038713] tot_loss[loss=2.276, over 5395843.75 frames. , ppl: 9.736162390882097], batch size: 70
+2022-12-13 14:50:45,766 INFO [train.py:421] (5/8) Epoch 9, batch 71000, loss[loss=2.445, over 1610.00 frames. , ppl: 11.53174003510517] tot_loss[loss=2.275, over 5430193.17 frames. , ppl: 9.731901957506965], batch size: 70
+2022-12-13 14:50:45,767 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 14:50:46,533 INFO [train.py:452] (5/8) Epoch 9, validation: loss=2.264, over 211138.00 frames. , ppl: 9.622304589200576
+2022-12-13 14:52:28,161 INFO [train.py:421] (5/8) Epoch 9, batch 71200, loss[loss=3.013, over 630.00 frames. , ppl: 20.34908498103113] tot_loss[loss=2.275, over 5452768.12 frames. , ppl: 9.73273194064089], batch size: 70
+2022-12-13 14:54:10,841 INFO [train.py:421] (5/8) Epoch 9, batch 71400, loss[loss=2.174, over 4270.00 frames. , ppl: 8.794188746567727] tot_loss[loss=2.275, over 5453236.04 frames. , ppl: 9.732750176397598], batch size: 70
+2022-12-13 14:55:50,291 INFO [train.py:421] (5/8) Epoch 9, batch 71600, loss[loss=2.402, over 1960.00 frames. , ppl: 11.040705105344582] tot_loss[loss=2.275, over 5449491.08 frames. , ppl: 9.72903671465471], batch size: 70
+2022-12-13 14:57:34,041 INFO [train.py:421] (5/8) Epoch 9, batch 71800, loss[loss=2.408, over 980.00 frames. , ppl: 11.109258595908944] tot_loss[loss=2.275, over 5459521.92 frames. , ppl: 9.730960321365412], batch size: 70
+2022-12-13 14:58:47,404 INFO [train.py:421] (5/8) Epoch 10, batch 0, loss[loss=2.288, over 3850.00 frames. , ppl: 9.859487627583773] tot_loss[loss=2.288, over 3850.00 frames. , ppl: 9.859487627583773], batch size: 70
+2022-12-13 15:00:28,724 INFO [train.py:421] (5/8) Epoch 10, batch 200, loss[loss=2.234, over 5250.00 frames. , ppl: 9.34081934982937] tot_loss[loss=2.267, over 517162.65 frames. , ppl: 9.65190461104696], batch size: 70
+2022-12-13 15:02:15,425 INFO [train.py:421] (5/8) Epoch 10, batch 400, loss[loss=2.514, over 1260.00 frames. , ppl: 12.35800611813839] tot_loss[loss=2.263, over 1013151.05 frames. , ppl: 9.61225913224136], batch size: 70
+2022-12-13 15:03:56,689 INFO [train.py:421] (5/8) Epoch 10, batch 600, loss[loss=2.38, over 840.00 frames. , ppl: 10.806803900667283] tot_loss[loss=2.259, over 1491599.79 frames. , ppl: 9.577577004048532], batch size: 70
+2022-12-13 15:05:40,250 INFO [train.py:421] (5/8) Epoch 10, batch 800, loss[loss=2.221, over 4060.00 frames. , ppl: 9.214219273294404] tot_loss[loss=2.263, over 1832932.14 frames. , ppl: 9.615885833627912], batch size: 70
+2022-12-13 15:07:21,172 INFO [train.py:421] (5/8) Epoch 10, batch 1000, loss[loss=2.321, over 2170.00 frames. , ppl: 10.185982595592442] tot_loss[loss=2.263, over 2179116.57 frames. , ppl: 9.610401194831162], batch size: 70
+2022-12-13 15:07:21,173 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 15:07:21,948 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.266, over 211138.00 frames. , ppl: 9.636474138823111
+2022-12-13 15:09:07,555 INFO [train.py:421] (5/8) Epoch 10, batch 1200, loss[loss=2.452, over 1050.00 frames. , ppl: 11.612611477389143] tot_loss[loss=2.262, over 2511604.85 frames. , ppl: 9.60109678714828], batch size: 70
+2022-12-13 15:10:54,475 INFO [train.py:421] (5/8) Epoch 10, batch 1400, loss[loss=2.443, over 1190.00 frames. , ppl: 11.502808871630501] tot_loss[loss=2.261, over 2817806.13 frames. , ppl: 9.591920781280312], batch size: 70
+2022-12-13 15:12:36,554 INFO [train.py:421] (5/8) Epoch 10, batch 1600, loss[loss=2.146, over 5810.00 frames. , ppl: 8.54942889425896] tot_loss[loss=2.261, over 3090842.32 frames. , ppl: 9.592187375414351], batch size: 70
+2022-12-13 15:14:19,443 INFO [train.py:421] (5/8) Epoch 10, batch 1800, loss[loss=2.195, over 11130.00 frames. , ppl: 8.98350451029431] tot_loss[loss=2.261, over 3338069.97 frames. , ppl: 9.589644889973446], batch size: 70
+2022-12-13 15:16:01,289 INFO [train.py:421] (5/8) Epoch 10, batch 2000, loss[loss=2.419, over 1260.00 frames. , ppl: 11.239669973732012] tot_loss[loss=2.261, over 3523621.93 frames. , ppl: 9.596501831388643], batch size: 70
+2022-12-13 15:16:01,289 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 15:16:02,057 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.634249414308712
+2022-12-13 15:17:44,593 INFO [train.py:421] (5/8) Epoch 10, batch 2200, loss[loss=2.381, over 1470.00 frames. , ppl: 10.811868487988805] tot_loss[loss=2.261, over 3704171.31 frames. , ppl: 9.596630282993049], batch size: 70
+2022-12-13 15:19:27,874 INFO [train.py:421] (5/8) Epoch 10, batch 2400, loss[loss=2.377, over 2450.00 frames. , ppl: 10.769829335646948] tot_loss[loss=2.262, over 3846378.11 frames. , ppl: 9.600871798958076], batch size: 70
+2022-12-13 15:21:09,658 INFO [train.py:421] (5/8) Epoch 10, batch 2600, loss[loss=2.535, over 1050.00 frames. , ppl: 12.616395645991007] tot_loss[loss=2.262, over 3989518.53 frames. , ppl: 9.602328665557733], batch size: 70
+2022-12-13 15:22:53,475 INFO [train.py:421] (5/8) Epoch 10, batch 2800, loss[loss=2.136, over 11270.00 frames. , ppl: 8.466856682009546] tot_loss[loss=2.261, over 4146035.67 frames. , ppl: 9.596065040894587], batch size: 70
+2022-12-13 15:24:36,441 INFO [train.py:421] (5/8) Epoch 10, batch 3000, loss[loss=2.204, over 3780.00 frames. , ppl: 9.057207379000054] tot_loss[loss=2.263, over 4249945.80 frames. , ppl: 9.61035530400468], batch size: 70
+2022-12-13 15:24:36,441 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 15:24:37,197 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.624782969170553
+2022-12-13 15:26:21,717 INFO [train.py:421] (5/8) Epoch 10, batch 3200, loss[loss=2.266, over 3500.00 frames. , ppl: 9.64280646261319] tot_loss[loss=2.263, over 4365042.54 frames. , ppl: 9.614926321265939], batch size: 70
+2022-12-13 15:28:06,713 INFO [train.py:421] (5/8) Epoch 10, batch 3400, loss[loss=2.4, over 1260.00 frames. , ppl: 11.025904233192309] tot_loss[loss=2.263, over 4486009.72 frames. , ppl: 9.611748907880221], batch size: 70
+2022-12-13 15:29:50,715 INFO [train.py:421] (5/8) Epoch 10, batch 3600, loss[loss=2.35, over 2310.00 frames. , ppl: 10.485186292808924] tot_loss[loss=2.261, over 4603956.29 frames. , ppl: 9.595141174674167], batch size: 70
+2022-12-13 15:31:33,968 INFO [train.py:421] (5/8) Epoch 10, batch 3800, loss[loss=2.766, over 630.00 frames. , ppl: 15.894759553593806] tot_loss[loss=2.262, over 4679082.61 frames. , ppl: 9.59962298066453], batch size: 70
+2022-12-13 15:33:14,230 INFO [train.py:421] (5/8) Epoch 10, batch 4000, loss[loss=2.162, over 7280.00 frames. , ppl: 8.689338075426413] tot_loss[loss=2.263, over 4739909.69 frames. , ppl: 9.61395027123842], batch size: 70
+2022-12-13 15:33:14,230 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 15:33:14,973 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.266, over 211138.00 frames. , ppl: 9.6403101424268
+2022-12-13 15:34:56,293 INFO [train.py:421] (5/8) Epoch 10, batch 4200, loss[loss=2.272, over 3080.00 frames. , ppl: 9.703075186017367] tot_loss[loss=2.262, over 4831280.73 frames. , ppl: 9.603462468870644], batch size: 70
+2022-12-13 15:36:36,711 INFO [train.py:421] (5/8) Epoch 10, batch 4400, loss[loss=2.307, over 2590.00 frames. , ppl: 10.043856142718724] tot_loss[loss=2.262, over 4888693.53 frames. , ppl: 9.605119715974553], batch size: 70
+2022-12-13 15:38:18,535 INFO [train.py:421] (5/8) Epoch 10, batch 4600, loss[loss=2.647, over 770.00 frames. , ppl: 14.104848705597977] tot_loss[loss=2.264, over 4918505.42 frames. , ppl: 9.617312265254471], batch size: 70
+2022-12-13 15:39:59,357 INFO [train.py:421] (5/8) Epoch 10, batch 4800, loss[loss=2.163, over 5950.00 frames. , ppl: 8.695508181780617] tot_loss[loss=2.264, over 4952889.77 frames. , ppl: 9.624306588231597], batch size: 70
+2022-12-13 15:41:41,311 INFO [train.py:421] (5/8) Epoch 10, batch 5000, loss[loss=2.361, over 1330.00 frames. , ppl: 10.597145678697022] tot_loss[loss=2.265, over 4993470.96 frames. , ppl: 9.626376077363485], batch size: 70
+2022-12-13 15:41:41,312 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 15:41:42,117 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.634593072176935
+2022-12-13 15:43:26,471 INFO [train.py:421] (5/8) Epoch 10, batch 5200, loss[loss=2.388, over 1050.00 frames. , ppl: 10.887137839589352] tot_loss[loss=2.264, over 5060220.14 frames. , ppl: 9.625063469776002], batch size: 70
+2022-12-13 15:45:07,012 INFO [train.py:421] (5/8) Epoch 10, batch 5400, loss[loss=2.219, over 4690.00 frames. , ppl: 9.194033007585423] tot_loss[loss=2.264, over 5124173.43 frames. , ppl: 9.61701256475594], batch size: 70
+2022-12-13 15:46:49,887 INFO [train.py:421] (5/8) Epoch 10, batch 5600, loss[loss=2.343, over 1470.00 frames. , ppl: 10.409973145981636] tot_loss[loss=2.265, over 5140035.94 frames. , ppl: 9.628194179305275], batch size: 70
+2022-12-13 15:48:31,083 INFO [train.py:421] (5/8) Epoch 10, batch 5800, loss[loss=2.172, over 4690.00 frames. , ppl: 8.77411282766434] tot_loss[loss=2.265, over 5163763.69 frames. , ppl: 9.634473563577062], batch size: 70
+2022-12-13 15:50:14,249 INFO [train.py:421] (5/8) Epoch 10, batch 6000, loss[loss=2.427, over 1750.00 frames. , ppl: 11.319966125880121] tot_loss[loss=2.266, over 5188939.23 frames. , ppl: 9.641165298810236], batch size: 70
+2022-12-13 15:50:14,250 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 15:50:14,999 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.622741820215266
+2022-12-13 15:51:54,468 INFO [train.py:421] (5/8) Epoch 10, batch 6200, loss[loss=2.549, over 1120.00 frames. , ppl: 12.788772475754492] tot_loss[loss=2.267, over 5216072.84 frames. , ppl: 9.653096294434045], batch size: 70
+2022-12-13 15:53:34,734 INFO [train.py:421] (5/8) Epoch 10, batch 6400, loss[loss=3.183, over 560.00 frames. , ppl: 24.118317821627546] tot_loss[loss=2.266, over 5272209.97 frames. , ppl: 9.644214328188692], batch size: 70
+2022-12-13 15:55:17,579 INFO [train.py:421] (5/8) Epoch 10, batch 6600, loss[loss=2.215, over 7280.00 frames. , ppl: 9.162627318656346] tot_loss[loss=2.267, over 5282782.07 frames. , ppl: 9.653534109350563], batch size: 70
+2022-12-13 15:56:56,905 INFO [train.py:421] (5/8) Epoch 10, batch 6800, loss[loss=2.279, over 2660.00 frames. , ppl: 9.766961110810568] tot_loss[loss=2.268, over 5285576.41 frames. , ppl: 9.65904194279564], batch size: 70
+2022-12-13 15:58:42,412 INFO [train.py:421] (5/8) Epoch 10, batch 7000, loss[loss=2.225, over 3500.00 frames. , ppl: 9.253895915038942] tot_loss[loss=2.267, over 5312953.91 frames. , ppl: 9.646879186191018], batch size: 70
+2022-12-13 15:58:42,413 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 15:58:43,176 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.266, over 211138.00 frames. , ppl: 9.637175889263702
+2022-12-13 16:00:25,417 INFO [train.py:421] (5/8) Epoch 10, batch 7200, loss[loss=2.274, over 3010.00 frames. , ppl: 9.717386434217975] tot_loss[loss=2.267, over 5316373.34 frames. , ppl: 9.652991698898258], batch size: 70
+2022-12-13 16:02:05,001 INFO [train.py:421] (5/8) Epoch 10, batch 7400, loss[loss=2.487, over 840.00 frames. , ppl: 12.028269100829865] tot_loss[loss=2.269, over 5277396.76 frames. , ppl: 9.672805392167799], batch size: 70
+2022-12-13 16:03:46,147 INFO [train.py:421] (5/8) Epoch 10, batch 7600, loss[loss=2.448, over 1330.00 frames. , ppl: 11.567807607461377] tot_loss[loss=2.27, over 5282873.15 frames. , ppl: 9.676474394667009], batch size: 70
+2022-12-13 16:05:27,166 INFO [train.py:421] (5/8) Epoch 10, batch 7800, loss[loss=2.376, over 1190.00 frames. , ppl: 10.759177299966591] tot_loss[loss=2.27, over 5287493.30 frames. , ppl: 9.681167337691123], batch size: 70
+2022-12-13 16:07:06,114 INFO [train.py:421] (5/8) Epoch 10, batch 8000, loss[loss=2.427, over 1120.00 frames. , ppl: 11.3248317127083] tot_loss[loss=2.271, over 5279898.70 frames. , ppl: 9.68533540880622], batch size: 70
+2022-12-13 16:07:06,114 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 16:07:06,898 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.623687560464123
+2022-12-13 16:08:45,826 INFO [train.py:421] (5/8) Epoch 10, batch 8200, loss[loss=2.28, over 3010.00 frames. , ppl: 9.7775600280644] tot_loss[loss=2.271, over 5304270.88 frames. , ppl: 9.684610832175158], batch size: 70
+2022-12-13 16:10:28,709 INFO [train.py:421] (5/8) Epoch 10, batch 8400, loss[loss=2.364, over 1890.00 frames. , ppl: 10.634383399857098] tot_loss[loss=2.27, over 5350002.60 frames. , ppl: 9.67561570257819], batch size: 70
+2022-12-13 16:12:10,318 INFO [train.py:421] (5/8) Epoch 10, batch 8600, loss[loss=2.381, over 1260.00 frames. , ppl: 10.811560836403718] tot_loss[loss=2.268, over 5364795.77 frames. , ppl: 9.661122532680675], batch size: 70
+2022-12-13 16:13:50,371 INFO [train.py:421] (5/8) Epoch 10, batch 8800, loss[loss=2.445, over 770.00 frames. , ppl: 11.536066272698733] tot_loss[loss=2.269, over 5353717.22 frames. , ppl: 9.66605399583695], batch size: 70
+2022-12-13 16:15:29,983 INFO [train.py:421] (5/8) Epoch 10, batch 9000, loss[loss=2.366, over 1820.00 frames. , ppl: 10.656338353022761] tot_loss[loss=2.267, over 5402249.74 frames. , ppl: 9.654767435614751], batch size: 70
+2022-12-13 16:15:29,984 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 16:15:30,747 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.634809825352011
+2022-12-13 16:17:11,733 INFO [train.py:421] (5/8) Epoch 10, batch 9200, loss[loss=2.414, over 2030.00 frames. , ppl: 11.176244037716918] tot_loss[loss=2.267, over 5445932.28 frames. , ppl: 9.649153826726465], batch size: 70
+2022-12-13 16:18:52,513 INFO [train.py:421] (5/8) Epoch 10, batch 9400, loss[loss=2.268, over 2800.00 frames. , ppl: 9.662953879998117] tot_loss[loss=2.267, over 5436256.41 frames. , ppl: 9.652277435439046], batch size: 70
+2022-12-13 16:20:35,332 INFO [train.py:421] (5/8) Epoch 10, batch 9600, loss[loss=2.251, over 1890.00 frames. , ppl: 9.494891004195187] tot_loss[loss=2.266, over 5458481.76 frames. , ppl: 9.64284805942367], batch size: 70
+2022-12-13 16:22:23,363 INFO [train.py:421] (5/8) Epoch 10, batch 9800, loss[loss=2.222, over 2520.00 frames. , ppl: 9.223416911736019] tot_loss[loss=2.266, over 5473053.08 frames. , ppl: 9.645514458006803], batch size: 70
+2022-12-13 16:24:04,254 INFO [train.py:421] (5/8) Epoch 10, batch 10000, loss[loss=2.406, over 1890.00 frames. , ppl: 11.08913516657897] tot_loss[loss=2.268, over 5443775.47 frames. , ppl: 9.655472041108396], batch size: 70
+2022-12-13 16:24:04,255 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 16:24:05,013 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.621310567860883
+2022-12-13 16:25:47,065 INFO [train.py:421] (5/8) Epoch 10, batch 10200, loss[loss=2.278, over 2590.00 frames. , ppl: 9.761810837889126] tot_loss[loss=2.267, over 5475431.45 frames. , ppl: 9.650077051194604], batch size: 70
+2022-12-13 16:27:29,192 INFO [train.py:421] (5/8) Epoch 10, batch 10400, loss[loss=2.19, over 3850.00 frames. , ppl: 8.931655512302534] tot_loss[loss=2.268, over 5434820.17 frames. , ppl: 9.658559643509431], batch size: 70
+2022-12-13 16:29:09,731 INFO [train.py:421] (5/8) Epoch 10, batch 10600, loss[loss=2.506, over 1400.00 frames. , ppl: 12.253948533090592] tot_loss[loss=2.269, over 5412519.35 frames. , ppl: 9.667658905063554], batch size: 70
+2022-12-13 16:30:51,418 INFO [train.py:421] (5/8) Epoch 10, batch 10800, loss[loss=2.446, over 1120.00 frames. , ppl: 11.54271512782819] tot_loss[loss=2.268, over 5426784.68 frames. , ppl: 9.661407309234772], batch size: 70
+2022-12-13 16:32:33,741 INFO [train.py:421] (5/8) Epoch 10, batch 11000, loss[loss=2.228, over 5180.00 frames. , ppl: 9.284919967026987] tot_loss[loss=2.268, over 5454971.07 frames. , ppl: 9.66389918428063], batch size: 70
+2022-12-13 16:32:33,742 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 16:32:34,520 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.60629313197587
+2022-12-13 16:34:15,547 INFO [train.py:421] (5/8) Epoch 10, batch 11200, loss[loss=2.266, over 1750.00 frames. , ppl: 9.644712881057735] tot_loss[loss=2.269, over 5419094.61 frames. , ppl: 9.673800214998227], batch size: 70
+2022-12-13 16:35:59,451 INFO [train.py:421] (5/8) Epoch 10, batch 11400, loss[loss=2.072, over 9380.00 frames. , ppl: 7.937091050418549] tot_loss[loss=2.269, over 5443679.10 frames. , ppl: 9.669476895894954], batch size: 70
+2022-12-13 16:37:41,285 INFO [train.py:421] (5/8) Epoch 10, batch 11600, loss[loss=2.296, over 1540.00 frames. , ppl: 9.938453746624102] tot_loss[loss=2.269, over 5446386.72 frames. , ppl: 9.669015709641878], batch size: 70
+2022-12-13 16:39:23,731 INFO [train.py:421] (5/8) Epoch 10, batch 11800, loss[loss=2.144, over 4480.00 frames. , ppl: 8.536128772928464] tot_loss[loss=2.269, over 5438289.97 frames. , ppl: 9.668910474254902], batch size: 70
+2022-12-13 16:41:07,319 INFO [train.py:421] (5/8) Epoch 10, batch 12000, loss[loss=2.553, over 1050.00 frames. , ppl: 12.84141867993685] tot_loss[loss=2.27, over 5408631.00 frames. , ppl: 9.67704240128363], batch size: 70
+2022-12-13 16:41:07,320 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 16:41:08,102 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.625352801927015
+2022-12-13 16:42:53,922 INFO [train.py:421] (5/8) Epoch 10, batch 12200, loss[loss=2.257, over 3150.00 frames. , ppl: 9.559090241612797] tot_loss[loss=2.27, over 5425742.29 frames. , ppl: 9.675028736069162], batch size: 70
+2022-12-13 16:44:34,116 INFO [train.py:421] (5/8) Epoch 10, batch 12400, loss[loss=2.501, over 980.00 frames. , ppl: 12.193187711867262] tot_loss[loss=2.27, over 5412799.61 frames. , ppl: 9.677475967329142], batch size: 70
+2022-12-13 16:46:16,871 INFO [train.py:421] (5/8) Epoch 10, batch 12600, loss[loss=2.365, over 1400.00 frames. , ppl: 10.640278875127484] tot_loss[loss=2.269, over 5459919.38 frames. , ppl: 9.666509136854668], batch size: 70
+2022-12-13 16:47:56,481 INFO [train.py:421] (5/8) Epoch 10, batch 12800, loss[loss=2.546, over 1120.00 frames. , ppl: 12.756200358877617] tot_loss[loss=2.269, over 5464741.17 frames. , ppl: 9.666676494432311], batch size: 70
+2022-12-13 16:49:37,745 INFO [train.py:421] (5/8) Epoch 10, batch 13000, loss[loss=2.257, over 3290.00 frames. , ppl: 9.555645879253044] tot_loss[loss=2.269, over 5442519.23 frames. , ppl: 9.664898901402276], batch size: 70
+2022-12-13 16:49:37,746 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 16:49:38,513 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.634478993517094
+2022-12-13 16:51:20,368 INFO [train.py:421] (5/8) Epoch 10, batch 13200, loss[loss=2.374, over 2100.00 frames. , ppl: 10.742692171931727] tot_loss[loss=2.267, over 5488124.53 frames. , ppl: 9.652878607999495], batch size: 70
+2022-12-13 16:53:01,609 INFO [train.py:421] (5/8) Epoch 10, batch 13400, loss[loss=2.209, over 3080.00 frames. , ppl: 9.107577527720307] tot_loss[loss=2.268, over 5460575.63 frames. , ppl: 9.659423037281657], batch size: 70
+2022-12-13 16:54:46,215 INFO [train.py:421] (5/8) Epoch 10, batch 13600, loss[loss=2.465, over 1120.00 frames. , ppl: 11.766970542387428] tot_loss[loss=2.266, over 5528114.46 frames. , ppl: 9.641451867261209], batch size: 70
+2022-12-13 16:56:30,070 INFO [train.py:421] (5/8) Epoch 10, batch 13800, loss[loss=2.893, over 560.00 frames. , ppl: 18.041087637658077] tot_loss[loss=2.267, over 5461218.76 frames. , ppl: 9.654479762350011], batch size: 70
+2022-12-13 16:58:09,139 INFO [train.py:421] (5/8) Epoch 10, batch 14000, loss[loss=2.34, over 1960.00 frames. , ppl: 10.38206625067267] tot_loss[loss=2.268, over 5462768.12 frames. , ppl: 9.656983429729909], batch size: 70
+2022-12-13 16:58:09,139 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 16:58:09,934 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.266, over 211138.00 frames. , ppl: 9.641104923874973
+2022-12-13 16:59:47,674 INFO [train.py:421] (5/8) Epoch 10, batch 14200, loss[loss=2.239, over 2730.00 frames. , ppl: 9.38267964748051] tot_loss[loss=2.267, over 5492195.71 frames. , ppl: 9.647387162827094], batch size: 70
+2022-12-13 17:01:29,028 INFO [train.py:421] (5/8) Epoch 10, batch 14400, loss[loss=2.397, over 1120.00 frames. , ppl: 10.99348822566742] tot_loss[loss=2.268, over 5459346.00 frames. , ppl: 9.657222432108453], batch size: 70
+2022-12-13 17:03:09,990 INFO [train.py:421] (5/8) Epoch 10, batch 14600, loss[loss=2.355, over 3500.00 frames. , ppl: 10.53526831268254] tot_loss[loss=2.266, over 5497094.83 frames. , ppl: 9.641279352179547], batch size: 70
+2022-12-13 17:04:50,630 INFO [train.py:421] (5/8) Epoch 10, batch 14800, loss[loss=3.526, over 420.00 frames. , ppl: 33.975175151435124] tot_loss[loss=2.265, over 5522357.27 frames. , ppl: 9.633814511034112], batch size: 70
+2022-12-13 17:06:32,761 INFO [train.py:421] (5/8) Epoch 10, batch 15000, loss[loss=2.249, over 3920.00 frames. , ppl: 9.477469128579138] tot_loss[loss=2.266, over 5512762.14 frames. , ppl: 9.63954961532699], batch size: 70
+2022-12-13 17:06:32,762 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 17:06:33,530 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.634267951557288
+2022-12-13 17:08:12,492 INFO [train.py:421] (5/8) Epoch 10, batch 15200, loss[loss=2.301, over 2660.00 frames. , ppl: 9.981765280866314] tot_loss[loss=2.267, over 5470172.42 frames. , ppl: 9.650826654665142], batch size: 70
+2022-12-13 17:09:58,766 INFO [train.py:421] (5/8) Epoch 10, batch 15400, loss[loss=2.19, over 3640.00 frames. , ppl: 8.938338646152369] tot_loss[loss=2.267, over 5481234.95 frames. , ppl: 9.646560137063014], batch size: 70
+2022-12-13 17:11:43,376 INFO [train.py:421] (5/8) Epoch 10, batch 15600, loss[loss=2.577, over 910.00 frames. , ppl: 13.156551631586412] tot_loss[loss=2.267, over 5461097.42 frames. , ppl: 9.653800008815445], batch size: 70
+2022-12-13 17:13:24,547 INFO [train.py:421] (5/8) Epoch 10, batch 15800, loss[loss=2.175, over 3640.00 frames. , ppl: 8.799317539663559] tot_loss[loss=2.269, over 5392895.20 frames. , ppl: 9.669255824229703], batch size: 70
+2022-12-13 17:15:07,635 INFO [train.py:421] (5/8) Epoch 10, batch 16000, loss[loss=2.241, over 2940.00 frames. , ppl: 9.39891360572032] tot_loss[loss=2.269, over 5404985.49 frames. , ppl: 9.666201618727737], batch size: 70
+2022-12-13 17:15:07,636 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 17:15:08,400 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.61156377960389
+2022-12-13 17:16:51,467 INFO [train.py:421] (5/8) Epoch 10, batch 16200, loss[loss=2.369, over 1260.00 frames. , ppl: 10.682066563236729] tot_loss[loss=2.269, over 5424022.31 frames. , ppl: 9.666057134985643], batch size: 70
+2022-12-13 17:18:33,686 INFO [train.py:421] (5/8) Epoch 10, batch 16400, loss[loss=2.509, over 1120.00 frames. , ppl: 12.287351818284737] tot_loss[loss=2.268, over 5449888.33 frames. , ppl: 9.656861370051935], batch size: 70
+2022-12-13 17:20:17,194 INFO [train.py:421] (5/8) Epoch 10, batch 16600, loss[loss=2.198, over 7700.00 frames. , ppl: 9.003173356804234] tot_loss[loss=2.267, over 5500125.82 frames. , ppl: 9.650836920744412], batch size: 70
+2022-12-13 17:21:59,444 INFO [train.py:421] (5/8) Epoch 10, batch 16800, loss[loss=2.293, over 1610.00 frames. , ppl: 9.906851002284352] tot_loss[loss=2.268, over 5485273.53 frames. , ppl: 9.656598673804252], batch size: 70
+2022-12-13 17:23:43,870 INFO [train.py:421] (5/8) Epoch 10, batch 17000, loss[loss=2.205, over 4620.00 frames. , ppl: 9.066167990638803] tot_loss[loss=2.267, over 5500025.89 frames. , ppl: 9.652321702348067], batch size: 70
+2022-12-13 17:23:43,871 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 17:23:44,636 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.629349702040027
+2022-12-13 17:25:25,466 INFO [train.py:421] (5/8) Epoch 10, batch 17200, loss[loss=2.234, over 3150.00 frames. , ppl: 9.332825332999905] tot_loss[loss=2.266, over 5563456.58 frames. , ppl: 9.638644072592676], batch size: 70
+2022-12-13 17:27:07,628 INFO [train.py:421] (5/8) Epoch 10, batch 17400, loss[loss=2.603, over 770.00 frames. , ppl: 13.498655799577119] tot_loss[loss=2.265, over 5598134.11 frames. , ppl: 9.632580136393843], batch size: 70
+2022-12-13 17:28:48,475 INFO [train.py:421] (5/8) Epoch 10, batch 17600, loss[loss=2.441, over 1260.00 frames. , ppl: 11.480654080886643] tot_loss[loss=2.266, over 5570186.55 frames. , ppl: 9.638314204083697], batch size: 70
+2022-12-13 17:30:34,236 INFO [train.py:421] (5/8) Epoch 10, batch 17800, loss[loss=2.69, over 700.00 frames. , ppl: 14.726038043649737] tot_loss[loss=2.265, over 5612399.53 frames. , ppl: 9.630080513085565], batch size: 70
+2022-12-13 17:32:15,339 INFO [train.py:421] (5/8) Epoch 10, batch 18000, loss[loss=2.199, over 6090.00 frames. , ppl: 9.015018838400714] tot_loss[loss=2.265, over 5591805.71 frames. , ppl: 9.630691038655526], batch size: 70
+2022-12-13 17:32:15,340 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 17:32:16,072 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.622539580598964
+2022-12-13 17:33:56,552 INFO [train.py:421] (5/8) Epoch 10, batch 18200, loss[loss=2.425, over 1260.00 frames. , ppl: 11.299833871053373] tot_loss[loss=2.266, over 5574968.67 frames. , ppl: 9.644490611129992], batch size: 70
+2022-12-13 17:35:35,469 INFO [train.py:421] (5/8) Epoch 10, batch 18400, loss[loss=2.773, over 700.00 frames. , ppl: 16.01135232675621] tot_loss[loss=2.267, over 5595245.62 frames. , ppl: 9.647925624962745], batch size: 70
+2022-12-13 17:37:19,303 INFO [train.py:421] (5/8) Epoch 10, batch 18600, loss[loss=2.413, over 1050.00 frames. , ppl: 11.167375790636328] tot_loss[loss=2.267, over 5570294.32 frames. , ppl: 9.645723284686053], batch size: 70
+2022-12-13 17:39:02,831 INFO [train.py:421] (5/8) Epoch 10, batch 18800, loss[loss=2.246, over 3010.00 frames. , ppl: 9.448186734309406] tot_loss[loss=2.265, over 5603279.04 frames. , ppl: 9.63169780197038], batch size: 70
+2022-12-13 17:40:48,048 INFO [train.py:421] (5/8) Epoch 10, batch 19000, loss[loss=2.248, over 3990.00 frames. , ppl: 9.465830804777648] tot_loss[loss=2.264, over 5663083.51 frames. , ppl: 9.61811251041737], batch size: 70
+2022-12-13 17:40:48,049 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 17:40:48,813 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.63137799747174
+2022-12-13 17:42:30,791 INFO [train.py:421] (5/8) Epoch 10, batch 19200, loss[loss=2.29, over 2520.00 frames. , ppl: 9.874280644345689] tot_loss[loss=2.265, over 5625513.19 frames. , ppl: 9.630488112021862], batch size: 70
+2022-12-13 17:44:15,187 INFO [train.py:421] (5/8) Epoch 10, batch 19400, loss[loss=2.459, over 1120.00 frames. , ppl: 11.693618086929527] tot_loss[loss=2.265, over 5629226.02 frames. , ppl: 9.627085007832338], batch size: 70
+2022-12-13 17:45:57,130 INFO [train.py:421] (5/8) Epoch 10, batch 19600, loss[loss=2.194, over 4270.00 frames. , ppl: 8.974483081888444] tot_loss[loss=2.264, over 5654347.70 frames. , ppl: 9.624447380083046], batch size: 70
+2022-12-13 17:47:36,313 INFO [train.py:421] (5/8) Epoch 10, batch 19800, loss[loss=2.18, over 4410.00 frames. , ppl: 8.842760877473852] tot_loss[loss=2.264, over 5648687.74 frames. , ppl: 9.618095544799207], batch size: 70
+2022-12-13 17:49:20,623 INFO [train.py:421] (5/8) Epoch 10, batch 20000, loss[loss=2.392, over 1330.00 frames. , ppl: 10.931279548462074] tot_loss[loss=2.264, over 5660479.91 frames. , ppl: 9.619690890194962], batch size: 70
+2022-12-13 17:49:20,624 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 17:49:21,384 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.266, over 211138.00 frames. , ppl: 9.641742794080622
+2022-12-13 17:51:06,026 INFO [train.py:421] (5/8) Epoch 10, batch 20200, loss[loss=2.156, over 7490.00 frames. , ppl: 8.634688741530214] tot_loss[loss=2.264, over 5638231.64 frames. , ppl: 9.62356500923139], batch size: 70
+2022-12-13 17:52:43,974 INFO [train.py:421] (5/8) Epoch 10, batch 20400, loss[loss=2.303, over 1750.00 frames. , ppl: 10.008591933119558] tot_loss[loss=2.264, over 5625296.61 frames. , ppl: 9.623221664282399], batch size: 70
+2022-12-13 17:54:27,189 INFO [train.py:421] (5/8) Epoch 10, batch 20600, loss[loss=3.671, over 420.00 frames. , ppl: 39.28248292210024] tot_loss[loss=2.265, over 5595188.47 frames. , ppl: 9.627559186879834], batch size: 70
+2022-12-13 17:56:10,937 INFO [train.py:421] (5/8) Epoch 10, batch 20800, loss[loss=2.174, over 6440.00 frames. , ppl: 8.79701709670641] tot_loss[loss=2.264, over 5608175.45 frames. , ppl: 9.624303300197193], batch size: 70
+2022-12-13 17:57:56,884 INFO [train.py:421] (5/8) Epoch 10, batch 21000, loss[loss=2.233, over 2170.00 frames. , ppl: 9.323276576396283] tot_loss[loss=2.262, over 5680396.12 frames. , ppl: 9.60254531865984], batch size: 70
+2022-12-13 17:57:56,885 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 17:57:57,648 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.622576610069983
+2022-12-13 17:59:39,845 INFO [train.py:421] (5/8) Epoch 10, batch 21200, loss[loss=2.306, over 2730.00 frames. , ppl: 10.0359408380779] tot_loss[loss=2.264, over 5624705.48 frames. , ppl: 9.618084749907359], batch size: 70
+2022-12-13 18:01:19,781 INFO [train.py:421] (5/8) Epoch 10, batch 21400, loss[loss=2.368, over 1330.00 frames. , ppl: 10.677827399950097] tot_loss[loss=2.265, over 5579387.17 frames. , ppl: 9.63108784381574], batch size: 70
+2022-12-13 18:03:00,309 INFO [train.py:421] (5/8) Epoch 10, batch 21600, loss[loss=2.179, over 3220.00 frames. , ppl: 8.83453772751757] tot_loss[loss=2.264, over 5619966.14 frames. , ppl: 9.622364063492778], batch size: 70
+2022-12-13 18:04:42,566 INFO [train.py:421] (5/8) Epoch 10, batch 21800, loss[loss=2.255, over 2030.00 frames. , ppl: 9.532343803399838] tot_loss[loss=2.265, over 5611086.71 frames. , ppl: 9.626399289956504], batch size: 70
+2022-12-13 18:06:24,589 INFO [train.py:421] (5/8) Epoch 10, batch 22000, loss[loss=2.272, over 3430.00 frames. , ppl: 9.694154702180143] tot_loss[loss=2.264, over 5612493.67 frames. , ppl: 9.62299654484112], batch size: 70
+2022-12-13 18:06:24,589 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 18:06:25,356 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.620783692748795
+2022-12-13 18:08:04,733 INFO [train.py:421] (5/8) Epoch 10, batch 22200, loss[loss=2.179, over 2030.00 frames. , ppl: 8.836399973416311] tot_loss[loss=2.266, over 5531994.57 frames. , ppl: 9.644245492525299], batch size: 70
+2022-12-13 18:09:43,513 INFO [train.py:421] (5/8) Epoch 10, batch 22400, loss[loss=2.243, over 2380.00 frames. , ppl: 9.424011979796536] tot_loss[loss=2.267, over 5507165.23 frames. , ppl: 9.652443649235218], batch size: 70
+2022-12-13 18:11:22,901 INFO [train.py:421] (5/8) Epoch 10, batch 22600, loss[loss=2.145, over 3360.00 frames. , ppl: 8.53881236503468] tot_loss[loss=2.268, over 5473349.40 frames. , ppl: 9.663625175910154], batch size: 70
+2022-12-13 18:13:02,879 INFO [train.py:421] (5/8) Epoch 10, batch 22800, loss[loss=2.352, over 1820.00 frames. , ppl: 10.503555443675879] tot_loss[loss=2.267, over 5510630.36 frames. , ppl: 9.651572067457197], batch size: 70
+2022-12-13 18:14:48,430 INFO [train.py:421] (5/8) Epoch 10, batch 23000, loss[loss=2.302, over 2240.00 frames. , ppl: 9.991565087595909] tot_loss[loss=2.267, over 5517907.94 frames. , ppl: 9.654449811402381], batch size: 70
+2022-12-13 18:14:48,430 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 18:14:49,200 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.624770148321595
+2022-12-13 18:16:34,755 INFO [train.py:421] (5/8) Epoch 10, batch 23200, loss[loss=2.391, over 1400.00 frames. , ppl: 10.927353951904426] tot_loss[loss=2.267, over 5544661.12 frames. , ppl: 9.646253291953473], batch size: 70
+2022-12-13 18:18:16,613 INFO [train.py:421] (5/8) Epoch 10, batch 23400, loss[loss=2.419, over 1050.00 frames. , ppl: 11.22916296002725] tot_loss[loss=2.268, over 5478884.83 frames. , ppl: 9.660984268120806], batch size: 70
+2022-12-13 18:19:54,568 INFO [train.py:421] (5/8) Epoch 10, batch 23600, loss[loss=2.398, over 2100.00 frames. , ppl: 11.006104306183309] tot_loss[loss=2.269, over 5445849.30 frames. , ppl: 9.667358539921784], batch size: 70
+2022-12-13 18:21:36,245 INFO [train.py:421] (5/8) Epoch 10, batch 23800, loss[loss=2.214, over 3360.00 frames. , ppl: 9.149922486502918] tot_loss[loss=2.267, over 5501580.10 frames. , ppl: 9.654272791386061], batch size: 70
+2022-12-13 18:23:16,509 INFO [train.py:421] (5/8) Epoch 10, batch 24000, loss[loss=2.266, over 1960.00 frames. , ppl: 9.639069196611384] tot_loss[loss=2.267, over 5518947.81 frames. , ppl: 9.649740291119725], batch size: 70
+2022-12-13 18:23:16,510 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 18:23:17,301 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.631834173341721
+2022-12-13 18:24:57,919 INFO [train.py:421] (5/8) Epoch 10, batch 24200, loss[loss=2.357, over 2520.00 frames. , ppl: 10.561557412288552] tot_loss[loss=2.268, over 5490775.59 frames. , ppl: 9.662237499157365], batch size: 70
+2022-12-13 18:26:41,699 INFO [train.py:421] (5/8) Epoch 10, batch 24400, loss[loss=2.345, over 1120.00 frames. , ppl: 10.43546899665886] tot_loss[loss=2.268, over 5501860.42 frames. , ppl: 9.65559711141272], batch size: 70
+2022-12-13 18:28:25,418 INFO [train.py:421] (5/8) Epoch 10, batch 24600, loss[loss=2.342, over 2660.00 frames. , ppl: 10.399895662173893] tot_loss[loss=2.267, over 5539737.43 frames. , ppl: 9.6470128101169], batch size: 70
+2022-12-13 18:30:06,404 INFO [train.py:421] (5/8) Epoch 10, batch 24800, loss[loss=2.234, over 8050.00 frames. , ppl: 9.335428903326791] tot_loss[loss=2.266, over 5547006.88 frames. , ppl: 9.644574584357372], batch size: 70
+2022-12-13 18:31:48,379 INFO [train.py:421] (5/8) Epoch 10, batch 25000, loss[loss=2.185, over 4970.00 frames. , ppl: 8.889719581123302] tot_loss[loss=2.268, over 5492630.78 frames. , ppl: 9.657770174357008], batch size: 70
+2022-12-13 18:31:48,380 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 18:31:49,120 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.617161851458512
+2022-12-13 18:33:31,057 INFO [train.py:421] (5/8) Epoch 10, batch 25200, loss[loss=2.31, over 2450.00 frames. , ppl: 10.072210275804483] tot_loss[loss=2.268, over 5476406.06 frames. , ppl: 9.664674639218404], batch size: 70
+2022-12-13 18:35:13,475 INFO [train.py:421] (5/8) Epoch 10, batch 25400, loss[loss=2.89, over 630.00 frames. , ppl: 18.002012432411004] tot_loss[loss=2.269, over 5467424.45 frames. , ppl: 9.670115975683652], batch size: 70
+2022-12-13 18:36:55,982 INFO [train.py:421] (5/8) Epoch 10, batch 25600, loss[loss=2.258, over 2170.00 frames. , ppl: 9.56428371729985] tot_loss[loss=2.27, over 5454175.62 frames. , ppl: 9.6771792361391], batch size: 70
+2022-12-13 18:38:38,921 INFO [train.py:421] (5/8) Epoch 10, batch 25800, loss[loss=2.344, over 1750.00 frames. , ppl: 10.428028299025208] tot_loss[loss=2.27, over 5445492.78 frames. , ppl: 9.678083719926402], batch size: 70
+2022-12-13 18:40:24,877 INFO [train.py:421] (5/8) Epoch 10, batch 26000, loss[loss=2.349, over 2590.00 frames. , ppl: 10.47862950760132] tot_loss[loss=2.269, over 5482107.41 frames. , ppl: 9.664946102729681], batch size: 70
+2022-12-13 18:40:24,878 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 18:40:25,674 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.266, over 211138.00 frames. , ppl: 9.637863426039562
+2022-12-13 18:42:07,000 INFO [train.py:421] (5/8) Epoch 10, batch 26200, loss[loss=2.172, over 4620.00 frames. , ppl: 8.77235856227351] tot_loss[loss=2.268, over 5504417.33 frames. , ppl: 9.656996136720151], batch size: 70
+2022-12-13 18:43:50,166 INFO [train.py:421] (5/8) Epoch 10, batch 26400, loss[loss=2.767, over 630.00 frames. , ppl: 15.909447774928216] tot_loss[loss=2.269, over 5494528.12 frames. , ppl: 9.665658218256397], batch size: 70
+2022-12-13 18:45:37,768 INFO [train.py:421] (5/8) Epoch 10, batch 26600, loss[loss=2.389, over 1190.00 frames. , ppl: 10.902104202463653] tot_loss[loss=2.266, over 5564258.59 frames. , ppl: 9.645173685441698], batch size: 70
+2022-12-13 18:47:19,961 INFO [train.py:421] (5/8) Epoch 10, batch 26800, loss[loss=2.345, over 2520.00 frames. , ppl: 10.438001860905123] tot_loss[loss=2.266, over 5592866.28 frames. , ppl: 9.639717495953366], batch size: 70
+2022-12-13 18:49:02,367 INFO [train.py:421] (5/8) Epoch 10, batch 27000, loss[loss=2.283, over 2030.00 frames. , ppl: 9.802174009102245] tot_loss[loss=2.266, over 5603616.78 frames. , ppl: 9.639263567701745], batch size: 70
+2022-12-13 18:49:02,368 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 18:49:03,131 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.621924342548795
+2022-12-13 18:50:45,266 INFO [train.py:421] (5/8) Epoch 10, batch 27200, loss[loss=2.319, over 2380.00 frames. , ppl: 10.168969335788114] tot_loss[loss=2.267, over 5551638.25 frames. , ppl: 9.648570826152126], batch size: 70
+2022-12-13 18:52:26,352 INFO [train.py:421] (5/8) Epoch 10, batch 27400, loss[loss=2.19, over 10430.00 frames. , ppl: 8.931568935734175] tot_loss[loss=2.268, over 5515420.90 frames. , ppl: 9.657272347639827], batch size: 70
+2022-12-13 18:54:07,431 INFO [train.py:421] (5/8) Epoch 10, batch 27600, loss[loss=2.341, over 1890.00 frames. , ppl: 10.389444553023282] tot_loss[loss=2.267, over 5549456.48 frames. , ppl: 9.65421083454224], batch size: 70
+2022-12-13 18:55:51,707 INFO [train.py:421] (5/8) Epoch 10, batch 27800, loss[loss=2.281, over 1540.00 frames. , ppl: 9.781763625404531] tot_loss[loss=2.268, over 5528465.23 frames. , ppl: 9.660919426806865], batch size: 70
+2022-12-13 18:57:33,165 INFO [train.py:421] (5/8) Epoch 10, batch 28000, loss[loss=2.732, over 910.00 frames. , ppl: 15.359645483901762] tot_loss[loss=2.268, over 5539187.07 frames. , ppl: 9.658690686153513], batch size: 70
+2022-12-13 18:57:33,166 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 18:57:33,932 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.617768243976228
+2022-12-13 18:59:13,885 INFO [train.py:421] (5/8) Epoch 10, batch 28200, loss[loss=2.334, over 2380.00 frames. , ppl: 10.317695782622906] tot_loss[loss=2.269, over 5518384.46 frames. , ppl: 9.672243033386636], batch size: 70
+2022-12-13 19:00:57,597 INFO [train.py:421] (5/8) Epoch 10, batch 28400, loss[loss=2.16, over 8330.00 frames. , ppl: 8.667817004109079] tot_loss[loss=2.268, over 5548880.01 frames. , ppl: 9.662483979887265], batch size: 70
+2022-12-13 19:02:36,588 INFO [train.py:421] (5/8) Epoch 10, batch 28600, loss[loss=2.554, over 980.00 frames. , ppl: 12.859407616323264] tot_loss[loss=2.269, over 5527152.41 frames. , ppl: 9.669191324991083], batch size: 70
+2022-12-13 19:04:16,653 INFO [train.py:421] (5/8) Epoch 10, batch 28800, loss[loss=2.285, over 2590.00 frames. , ppl: 9.826121175564092] tot_loss[loss=2.269, over 5506425.36 frames. , ppl: 9.673696747934287], batch size: 70
+2022-12-13 19:06:00,378 INFO [train.py:421] (5/8) Epoch 10, batch 29000, loss[loss=2.944, over 560.00 frames. , ppl: 18.992290991217747] tot_loss[loss=2.269, over 5529682.72 frames. , ppl: 9.667815477479756], batch size: 70
+2022-12-13 19:06:00,378 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 19:06:01,145 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.614629935321982
+2022-12-13 19:07:42,003 INFO [train.py:421] (5/8) Epoch 10, batch 29200, loss[loss=2.452, over 2100.00 frames. , ppl: 11.607085677633393] tot_loss[loss=2.268, over 5554232.70 frames. , ppl: 9.65884229745711], batch size: 70
+2022-12-13 19:09:24,501 INFO [train.py:421] (5/8) Epoch 10, batch 29400, loss[loss=2.137, over 6440.00 frames. , ppl: 8.477022585504994] tot_loss[loss=2.268, over 5572095.97 frames. , ppl: 9.655897755548152], batch size: 70
+2022-12-13 19:11:08,697 INFO [train.py:421] (5/8) Epoch 10, batch 29600, loss[loss=2.372, over 1750.00 frames. , ppl: 10.719971938333172] tot_loss[loss=2.268, over 5582819.00 frames. , ppl: 9.659613821292217], batch size: 70
+2022-12-13 19:12:55,139 INFO [train.py:421] (5/8) Epoch 10, batch 29800, loss[loss=2.561, over 980.00 frames. , ppl: 12.946998803864311] tot_loss[loss=2.267, over 5597689.80 frames. , ppl: 9.654323823244349], batch size: 70
+2022-12-13 19:14:36,417 INFO [train.py:421] (5/8) Epoch 10, batch 30000, loss[loss=2.36, over 1610.00 frames. , ppl: 10.59272849468061] tot_loss[loss=2.267, over 5581079.18 frames. , ppl: 9.653659171600902], batch size: 70
+2022-12-13 19:14:36,417 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 19:14:37,218 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.266, over 211138.00 frames. , ppl: 9.638813505699115
+2022-12-13 19:16:16,778 INFO [train.py:421] (5/8) Epoch 10, batch 30200, loss[loss=2.154, over 9450.00 frames. , ppl: 8.623137446816932] tot_loss[loss=2.267, over 5592143.02 frames. , ppl: 9.652304606033955], batch size: 70
+2022-12-13 19:18:02,135 INFO [train.py:421] (5/8) Epoch 10, batch 30400, loss[loss=2.911, over 700.00 frames. , ppl: 18.37723020287862] tot_loss[loss=2.268, over 5589296.66 frames. , ppl: 9.656397032581669], batch size: 70
+2022-12-13 19:19:43,254 INFO [train.py:421] (5/8) Epoch 10, batch 30600, loss[loss=2.684, over 840.00 frames. , ppl: 14.637232649716946] tot_loss[loss=2.268, over 5568233.31 frames. , ppl: 9.655695591550131], batch size: 70
+2022-12-13 19:21:25,051 INFO [train.py:421] (5/8) Epoch 10, batch 30800, loss[loss=2.351, over 2380.00 frames. , ppl: 10.491840258274324] tot_loss[loss=2.268, over 5566482.60 frames. , ppl: 9.660681008196507], batch size: 70
+2022-12-13 19:23:07,043 INFO [train.py:421] (5/8) Epoch 10, batch 31000, loss[loss=2.137, over 7560.00 frames. , ppl: 8.473562365865655] tot_loss[loss=2.268, over 5567207.56 frames. , ppl: 9.660374964885857], batch size: 70
+2022-12-13 19:23:07,044 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 19:23:07,797 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.624022295019808
+2022-12-13 19:24:49,278 INFO [train.py:421] (5/8) Epoch 10, batch 31200, loss[loss=2.809, over 630.00 frames. , ppl: 16.585599953840756] tot_loss[loss=2.268, over 5560276.48 frames. , ppl: 9.659432609946695], batch size: 70
+2022-12-13 19:26:30,756 INFO [train.py:421] (5/8) Epoch 10, batch 31400, loss[loss=2.251, over 2730.00 frames. , ppl: 9.495764125010128] tot_loss[loss=2.268, over 5532769.58 frames. , ppl: 9.661208778019224], batch size: 70
+2022-12-13 19:28:10,501 INFO [train.py:421] (5/8) Epoch 10, batch 31600, loss[loss=3.315, over 490.00 frames. , ppl: 27.522762864789875] tot_loss[loss=2.269, over 5494227.83 frames. , ppl: 9.66800475020485], batch size: 70
+2022-12-13 19:29:51,153 INFO [train.py:421] (5/8) Epoch 10, batch 31800, loss[loss=2.239, over 2590.00 frames. , ppl: 9.384670704970949] tot_loss[loss=2.268, over 5521138.03 frames. , ppl: 9.658186825030882], batch size: 70
+2022-12-13 19:31:35,513 INFO [train.py:421] (5/8) Epoch 10, batch 32000, loss[loss=2.373, over 1540.00 frames. , ppl: 10.72533149619172] tot_loss[loss=2.269, over 5506300.66 frames. , ppl: 9.66589532135384], batch size: 70
+2022-12-13 19:31:35,513 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 19:31:36,277 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.616229561960207
+2022-12-13 19:33:19,159 INFO [train.py:421] (5/8) Epoch 10, batch 32200, loss[loss=2.251, over 4200.00 frames. , ppl: 9.49798621822406] tot_loss[loss=2.27, over 5481579.25 frames. , ppl: 9.675881211307287], batch size: 70
+2022-12-13 19:35:01,660 INFO [train.py:421] (5/8) Epoch 10, batch 32400, loss[loss=2.43, over 1120.00 frames. , ppl: 11.35913810540157] tot_loss[loss=2.27, over 5450079.20 frames. , ppl: 9.68224004300827], batch size: 70
+2022-12-13 19:36:43,500 INFO [train.py:421] (5/8) Epoch 10, batch 32600, loss[loss=2.382, over 1470.00 frames. , ppl: 10.828550431633753] tot_loss[loss=2.27, over 5459807.95 frames. , ppl: 9.681432881730556], batch size: 70
+2022-12-13 19:38:26,634 INFO [train.py:421] (5/8) Epoch 10, batch 32800, loss[loss=2.318, over 1680.00 frames. , ppl: 10.15943791061932] tot_loss[loss=2.269, over 5480171.37 frames. , ppl: 9.674458519862524], batch size: 70
+2022-12-13 19:40:07,395 INFO [train.py:421] (5/8) Epoch 10, batch 33000, loss[loss=2.459, over 1120.00 frames. , ppl: 11.690682000159468] tot_loss[loss=2.27, over 5450024.72 frames. , ppl: 9.683078567473508], batch size: 70
+2022-12-13 19:40:07,395 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 19:40:08,159 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.611626373467763
+2022-12-13 19:41:48,442 INFO [train.py:421] (5/8) Epoch 10, batch 33200, loss[loss=2.294, over 3150.00 frames. , ppl: 9.916047663450835] tot_loss[loss=2.27, over 5456657.71 frames. , ppl: 9.680546286897757], batch size: 70
+2022-12-13 19:43:28,468 INFO [train.py:421] (5/8) Epoch 10, batch 33400, loss[loss=2.382, over 1470.00 frames. , ppl: 10.829518029092705] tot_loss[loss=2.271, over 5425821.95 frames. , ppl: 9.691389766611847], batch size: 70
+2022-12-13 19:45:10,998 INFO [train.py:421] (5/8) Epoch 10, batch 33600, loss[loss=3.005, over 560.00 frames. , ppl: 20.18171250673371] tot_loss[loss=2.272, over 5444239.01 frames. , ppl: 9.6996879615368], batch size: 70
+2022-12-13 19:46:54,767 INFO [train.py:421] (5/8) Epoch 10, batch 33800, loss[loss=2.134, over 3850.00 frames. , ppl: 8.449048874573005] tot_loss[loss=2.272, over 5455216.85 frames. , ppl: 9.69677948564492], batch size: 70
+2022-12-13 19:48:42,876 INFO [train.py:421] (5/8) Epoch 10, batch 34000, loss[loss=2.351, over 3010.00 frames. , ppl: 10.494370811444375] tot_loss[loss=2.27, over 5508821.94 frames. , ppl: 9.6817562453882], batch size: 70
+2022-12-13 19:48:42,877 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 19:48:43,659 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.612292169791925
+2022-12-13 19:50:27,172 INFO [train.py:421] (5/8) Epoch 10, batch 34200, loss[loss=2.355, over 1960.00 frames. , ppl: 10.538933806715379] tot_loss[loss=2.27, over 5498478.23 frames. , ppl: 9.684090005949331], batch size: 70
+2022-12-13 19:52:08,998 INFO [train.py:421] (5/8) Epoch 10, batch 34400, loss[loss=2.216, over 3570.00 frames. , ppl: 9.166986324727594] tot_loss[loss=2.27, over 5526704.25 frames. , ppl: 9.674991192107004], batch size: 70
+2022-12-13 19:53:50,199 INFO [train.py:421] (5/8) Epoch 10, batch 34600, loss[loss=8, over 140.00 frames. , ppl: 2979.949672244301] tot_loss[loss=2.269, over 5508655.20 frames. , ppl: 9.674318560463197], batch size: 70
+2022-12-13 19:55:28,936 INFO [train.py:421] (5/8) Epoch 10, batch 34800, loss[loss=2.154, over 3710.00 frames. , ppl: 8.618438165657212] tot_loss[loss=2.268, over 5541326.55 frames. , ppl: 9.66462962386912], batch size: 70
+2022-12-13 19:57:12,629 INFO [train.py:421] (5/8) Epoch 10, batch 35000, loss[loss=2.218, over 7490.00 frames. , ppl: 9.188374308707896] tot_loss[loss=2.269, over 5515187.26 frames. , ppl: 9.674551500229967], batch size: 70
+2022-12-13 19:57:12,630 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 19:57:13,388 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.616518490872702
+2022-12-13 19:58:57,927 INFO [train.py:421] (5/8) Epoch 10, batch 35200, loss[loss=2.235, over 5040.00 frames. , ppl: 9.349982077588404] tot_loss[loss=2.27, over 5478610.38 frames. , ppl: 9.68351898844358], batch size: 70
+2022-12-13 20:00:39,613 INFO [train.py:421] (5/8) Epoch 10, batch 35400, loss[loss=2.127, over 5670.00 frames. , ppl: 8.38951675723588] tot_loss[loss=2.268, over 5539570.85 frames. , ppl: 9.660356715908094], batch size: 70
+2022-12-13 20:02:22,788 INFO [train.py:421] (5/8) Epoch 10, batch 35600, loss[loss=2.612, over 1120.00 frames. , ppl: 13.63242534941036] tot_loss[loss=2.269, over 5493646.70 frames. , ppl: 9.670503943098664], batch size: 70
+2022-12-13 20:04:05,805 INFO [train.py:421] (5/8) Epoch 10, batch 35800, loss[loss=2.819, over 630.00 frames. , ppl: 16.764339406401426] tot_loss[loss=2.269, over 5512567.88 frames. , ppl: 9.668123014051174], batch size: 70
+2022-12-13 20:05:47,822 INFO [train.py:421] (5/8) Epoch 10, batch 36000, loss[loss=2.337, over 1540.00 frames. , ppl: 10.349840762573407] tot_loss[loss=2.269, over 5502445.40 frames. , ppl: 9.669203021354896], batch size: 70
+2022-12-13 20:05:47,823 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 20:05:48,581 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.601890816854365
+2022-12-13 20:07:33,622 INFO [train.py:421] (5/8) Epoch 10, batch 36200, loss[loss=2.503, over 1050.00 frames. , ppl: 12.222587419446954] tot_loss[loss=2.269, over 5488924.96 frames. , ppl: 9.670810690205201], batch size: 70
+2022-12-13 20:09:17,749 INFO [train.py:421] (5/8) Epoch 10, batch 36400, loss[loss=2.4, over 2380.00 frames. , ppl: 11.02722071468797] tot_loss[loss=2.269, over 5527017.75 frames. , ppl: 9.666427294149338], batch size: 70
+2022-12-13 20:11:03,403 INFO [train.py:421] (5/8) Epoch 10, batch 36600, loss[loss=2.775, over 630.00 frames. , ppl: 16.046510016045715] tot_loss[loss=2.269, over 5524949.07 frames. , ppl: 9.67196143624257], batch size: 70
+2022-12-13 20:12:48,801 INFO [train.py:421] (5/8) Epoch 10, batch 36800, loss[loss=2.163, over 7350.00 frames. , ppl: 8.701308126491059] tot_loss[loss=2.27, over 5499372.53 frames. , ppl: 9.681550470185165], batch size: 70
+2022-12-13 20:14:30,489 INFO [train.py:421] (5/8) Epoch 10, batch 37000, loss[loss=2.347, over 1820.00 frames. , ppl: 10.457035599950736] tot_loss[loss=2.27, over 5500034.84 frames. , ppl: 9.678757983496537], batch size: 70
+2022-12-13 20:14:30,489 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 20:14:31,281 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.600985586014557
+2022-12-13 20:16:16,937 INFO [train.py:421] (5/8) Epoch 10, batch 37200, loss[loss=2.664, over 700.00 frames. , ppl: 14.35720466372378] tot_loss[loss=2.269, over 5552195.79 frames. , ppl: 9.668485927608772], batch size: 70
+2022-12-13 20:17:58,370 INFO [train.py:421] (5/8) Epoch 10, batch 37400, loss[loss=2.205, over 5600.00 frames. , ppl: 9.06946073916049] tot_loss[loss=2.269, over 5543870.34 frames. , ppl: 9.667211723978046], batch size: 70
+2022-12-13 20:19:44,307 INFO [train.py:421] (5/8) Epoch 10, batch 37600, loss[loss=2.384, over 1470.00 frames. , ppl: 10.847275649086974] tot_loss[loss=2.268, over 5554177.30 frames. , ppl: 9.661650475448345], batch size: 70
+2022-12-13 20:21:26,441 INFO [train.py:421] (5/8) Epoch 10, batch 37800, loss[loss=2.295, over 2730.00 frames. , ppl: 9.927417142054244] tot_loss[loss=2.271, over 5481973.14 frames. , ppl: 9.685219176875835], batch size: 70
+2022-12-13 20:23:05,380 INFO [train.py:421] (5/8) Epoch 10, batch 38000, loss[loss=2.239, over 3990.00 frames. , ppl: 9.384896675121134] tot_loss[loss=2.27, over 5485211.16 frames. , ppl: 9.67874569461972], batch size: 70
+2022-12-13 20:23:05,380 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 20:23:06,141 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.261, over 211138.00 frames. , ppl: 9.589634108742406
+2022-12-13 20:24:47,954 INFO [train.py:421] (5/8) Epoch 10, batch 38200, loss[loss=2.347, over 1540.00 frames. , ppl: 10.457491117878998] tot_loss[loss=2.268, over 5554209.17 frames. , ppl: 9.656054238433295], batch size: 70
+2022-12-13 20:26:26,943 INFO [train.py:421] (5/8) Epoch 10, batch 38400, loss[loss=2.454, over 1190.00 frames. , ppl: 11.632420599638722] tot_loss[loss=2.27, over 5471006.20 frames. , ppl: 9.680714354734528], batch size: 70
+2022-12-13 20:28:07,272 INFO [train.py:421] (5/8) Epoch 10, batch 38600, loss[loss=2.748, over 630.00 frames. , ppl: 15.61092391473723] tot_loss[loss=2.271, over 5436240.83 frames. , ppl: 9.685747324556514], batch size: 70
+2022-12-13 20:29:50,034 INFO [train.py:421] (5/8) Epoch 10, batch 38800, loss[loss=2.256, over 3500.00 frames. , ppl: 9.548075011822736] tot_loss[loss=2.27, over 5480109.09 frames. , ppl: 9.675593920468701], batch size: 70
+2022-12-13 20:31:30,533 INFO [train.py:421] (5/8) Epoch 10, batch 39000, loss[loss=2.1, over 8330.00 frames. , ppl: 8.165783149911258] tot_loss[loss=2.268, over 5511702.06 frames. , ppl: 9.662613071375588], batch size: 70
+2022-12-13 20:31:30,534 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 20:31:31,277 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.59960729794989
+2022-12-13 20:33:11,828 INFO [train.py:421] (5/8) Epoch 10, batch 39200, loss[loss=2.222, over 2520.00 frames. , ppl: 9.222530527233651] tot_loss[loss=2.268, over 5497047.73 frames. , ppl: 9.660575768076457], batch size: 70
+2022-12-13 20:34:51,241 INFO [train.py:421] (5/8) Epoch 10, batch 39400, loss[loss=2.153, over 6510.00 frames. , ppl: 8.606439573597461] tot_loss[loss=2.269, over 5478796.81 frames. , ppl: 9.670210049082524], batch size: 70
+2022-12-13 20:36:33,969 INFO [train.py:421] (5/8) Epoch 10, batch 39600, loss[loss=2.28, over 3010.00 frames. , ppl: 9.78112784631846] tot_loss[loss=2.27, over 5449926.03 frames. , ppl: 9.676517583983276], batch size: 70
+2022-12-13 20:38:17,863 INFO [train.py:421] (5/8) Epoch 10, batch 39800, loss[loss=2.147, over 4690.00 frames. , ppl: 8.56125223465887] tot_loss[loss=2.271, over 5394756.28 frames. , ppl: 9.689870854411065], batch size: 70
+2022-12-13 20:40:01,117 INFO [train.py:421] (5/8) Epoch 10, batch 40000, loss[loss=2.169, over 9170.00 frames. , ppl: 8.75382156913289] tot_loss[loss=2.269, over 5460901.34 frames. , ppl: 9.667180515515565], batch size: 70
+2022-12-13 20:40:01,117 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 20:40:01,849 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.261, over 211138.00 frames. , ppl: 9.595198105641067
+2022-12-13 20:41:45,210 INFO [train.py:421] (5/8) Epoch 10, batch 40200, loss[loss=2.448, over 1260.00 frames. , ppl: 11.57088629353146] tot_loss[loss=2.27, over 5438930.87 frames. , ppl: 9.677034050014663], batch size: 70
+2022-12-13 20:43:28,830 INFO [train.py:421] (5/8) Epoch 10, batch 40400, loss[loss=2.482, over 910.00 frames. , ppl: 11.966530332394823] tot_loss[loss=2.27, over 5445319.46 frames. , ppl: 9.674895487869545], batch size: 70
+2022-12-13 20:45:07,759 INFO [train.py:421] (5/8) Epoch 10, batch 40600, loss[loss=2.237, over 4340.00 frames. , ppl: 9.363828256831997] tot_loss[loss=2.269, over 5448148.72 frames. , ppl: 9.667020470414938], batch size: 70
+2022-12-13 20:46:47,452 INFO [train.py:421] (5/8) Epoch 10, batch 40800, loss[loss=2.504, over 910.00 frames. , ppl: 12.23335477997886] tot_loss[loss=2.269, over 5452638.20 frames. , ppl: 9.668212324170865], batch size: 70
+2022-12-13 20:48:25,962 INFO [train.py:421] (5/8) Epoch 10, batch 41000, loss[loss=2.385, over 1960.00 frames. , ppl: 10.856718013266999] tot_loss[loss=2.269, over 5429587.85 frames. , ppl: 9.674274881656117], batch size: 70
+2022-12-13 20:48:25,963 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 20:48:26,717 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.60531213793815
+2022-12-13 20:50:09,272 INFO [train.py:421] (5/8) Epoch 10, batch 41200, loss[loss=2.379, over 1260.00 frames. , ppl: 10.789896435719875] tot_loss[loss=2.269, over 5433807.33 frames. , ppl: 9.674048844263977], batch size: 70
+2022-12-13 20:51:52,024 INFO [train.py:421] (5/8) Epoch 10, batch 41400, loss[loss=2.421, over 1260.00 frames. , ppl: 11.25611586458313] tot_loss[loss=2.27, over 5424995.40 frames. , ppl: 9.676947726542307], batch size: 70
+2022-12-13 20:53:33,181 INFO [train.py:421] (5/8) Epoch 10, batch 41600, loss[loss=2.311, over 2380.00 frames. , ppl: 10.08015482330413] tot_loss[loss=2.271, over 5415467.09 frames. , ppl: 9.689496802016832], batch size: 70
+2022-12-13 20:55:14,354 INFO [train.py:421] (5/8) Epoch 10, batch 41800, loss[loss=2.317, over 3010.00 frames. , ppl: 10.148856865558203] tot_loss[loss=2.27, over 5436763.04 frames. , ppl: 9.681094964249471], batch size: 70
+2022-12-13 20:56:55,461 INFO [train.py:421] (5/8) Epoch 10, batch 42000, loss[loss=2.259, over 4200.00 frames. , ppl: 9.573054106578626] tot_loss[loss=2.269, over 5455417.68 frames. , ppl: 9.673096345410528], batch size: 70
+2022-12-13 20:56:55,462 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 20:56:56,194 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.599615822834881
+2022-12-13 20:58:38,558 INFO [train.py:421] (5/8) Epoch 10, batch 42200, loss[loss=2.155, over 7980.00 frames. , ppl: 8.628948626797136] tot_loss[loss=2.27, over 5458383.36 frames. , ppl: 9.680383391908792], batch size: 70
+2022-12-13 21:00:23,424 INFO [train.py:421] (5/8) Epoch 10, batch 42400, loss[loss=2.404, over 1260.00 frames. , ppl: 11.069227840132166] tot_loss[loss=2.271, over 5449984.12 frames. , ppl: 9.689613714491049], batch size: 70
+2022-12-13 21:02:03,395 INFO [train.py:421] (5/8) Epoch 10, batch 42600, loss[loss=2.735, over 630.00 frames. , ppl: 15.405744138616118] tot_loss[loss=2.271, over 5445966.45 frames. , ppl: 9.693282519755495], batch size: 70
+2022-12-13 21:03:47,731 INFO [train.py:421] (5/8) Epoch 10, batch 42800, loss[loss=2.202, over 4620.00 frames. , ppl: 9.040368123100759] tot_loss[loss=2.272, over 5418916.54 frames. , ppl: 9.703052928538002], batch size: 70
+2022-12-13 21:05:29,126 INFO [train.py:421] (5/8) Epoch 10, batch 43000, loss[loss=2.608, over 770.00 frames. , ppl: 13.5651391267874] tot_loss[loss=2.273, over 5390564.78 frames. , ppl: 9.711248480953483], batch size: 70
+2022-12-13 21:05:29,127 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 21:05:29,892 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.605019280876657
+2022-12-13 21:07:11,583 INFO [train.py:421] (5/8) Epoch 10, batch 43200, loss[loss=2.324, over 2380.00 frames. , ppl: 10.21888229605866] tot_loss[loss=2.274, over 5374488.59 frames. , ppl: 9.715098047607873], batch size: 70
+2022-12-13 21:08:55,560 INFO [train.py:421] (5/8) Epoch 10, batch 43400, loss[loss=2.28, over 2870.00 frames. , ppl: 9.781540549606914] tot_loss[loss=2.273, over 5382138.39 frames. , ppl: 9.710155498139322], batch size: 70
+2022-12-13 21:10:37,369 INFO [train.py:421] (5/8) Epoch 10, batch 43600, loss[loss=2.34, over 2450.00 frames. , ppl: 10.376776833359893] tot_loss[loss=2.273, over 5382731.68 frames. , ppl: 9.712314635816204], batch size: 70
+2022-12-13 21:12:19,476 INFO [train.py:421] (5/8) Epoch 10, batch 43800, loss[loss=3.416, over 490.00 frames. , ppl: 30.44294248010584] tot_loss[loss=2.275, over 5356645.57 frames. , ppl: 9.72497323285801], batch size: 70
+2022-12-13 21:13:59,377 INFO [train.py:421] (5/8) Epoch 10, batch 44000, loss[loss=2.167, over 8120.00 frames. , ppl: 8.73278900569792] tot_loss[loss=2.273, over 5404359.49 frames. , ppl: 9.706100804233868], batch size: 70
+2022-12-13 21:13:59,377 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 21:14:00,109 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.620145785284024
+2022-12-13 21:15:38,898 INFO [train.py:421] (5/8) Epoch 10, batch 44200, loss[loss=2.681, over 770.00 frames. , ppl: 14.597808628319928] tot_loss[loss=2.273, over 5401424.04 frames. , ppl: 9.705185720119934], batch size: 70
+2022-12-13 21:17:20,593 INFO [train.py:421] (5/8) Epoch 10, batch 44400, loss[loss=2.288, over 2310.00 frames. , ppl: 9.8527550449624] tot_loss[loss=2.271, over 5463619.97 frames. , ppl: 9.684711586222956], batch size: 70
+2022-12-13 21:19:00,922 INFO [train.py:421] (5/8) Epoch 10, batch 44600, loss[loss=2.327, over 1680.00 frames. , ppl: 10.244621862675398] tot_loss[loss=2.27, over 5445492.51 frames. , ppl: 9.683575693592095], batch size: 70
+2022-12-13 21:20:45,693 INFO [train.py:421] (5/8) Epoch 10, batch 44800, loss[loss=2.172, over 4200.00 frames. , ppl: 8.77642908652715] tot_loss[loss=2.27, over 5449435.19 frames. , ppl: 9.67972856790092], batch size: 70
+2022-12-13 21:22:25,975 INFO [train.py:421] (5/8) Epoch 10, batch 45000, loss[loss=2.192, over 6370.00 frames. , ppl: 8.952784420401448] tot_loss[loss=2.269, over 5484444.67 frames. , ppl: 9.672515319922546], batch size: 70
+2022-12-13 21:22:25,976 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 21:22:26,698 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.598352801576091
+2022-12-13 21:24:05,961 INFO [train.py:421] (5/8) Epoch 10, batch 45200, loss[loss=2.359, over 3220.00 frames. , ppl: 10.579079073429595] tot_loss[loss=2.27, over 5461979.12 frames. , ppl: 9.681549722823029], batch size: 70
+2022-12-13 21:25:48,527 INFO [train.py:421] (5/8) Epoch 10, batch 45400, loss[loss=2.234, over 3780.00 frames. , ppl: 9.340222246844512] tot_loss[loss=2.269, over 5501745.06 frames. , ppl: 9.67173638241835], batch size: 70
+2022-12-13 21:27:30,593 INFO [train.py:421] (5/8) Epoch 10, batch 45600, loss[loss=2.193, over 3150.00 frames. , ppl: 8.964857860397652] tot_loss[loss=2.271, over 5472128.77 frames. , ppl: 9.687444707451176], batch size: 70
+2022-12-13 21:29:13,811 INFO [train.py:421] (5/8) Epoch 10, batch 45800, loss[loss=2.428, over 1120.00 frames. , ppl: 11.340540535672533] tot_loss[loss=2.27, over 5502626.37 frames. , ppl: 9.679899036708942], batch size: 70
+2022-12-13 21:30:58,371 INFO [train.py:421] (5/8) Epoch 10, batch 46000, loss[loss=2.545, over 840.00 frames. , ppl: 12.737900262327296] tot_loss[loss=2.269, over 5535870.39 frames. , ppl: 9.669933972891139], batch size: 70
+2022-12-13 21:30:58,372 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 21:30:59,139 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.603299281294177
+2022-12-13 21:32:38,608 INFO [train.py:421] (5/8) Epoch 10, batch 46200, loss[loss=2.588, over 980.00 frames. , ppl: 13.301385515799097] tot_loss[loss=2.269, over 5541154.35 frames. , ppl: 9.66698840361372], batch size: 70
+2022-12-13 21:34:24,232 INFO [train.py:421] (5/8) Epoch 10, batch 46400, loss[loss=2.553, over 840.00 frames. , ppl: 12.842421208447611] tot_loss[loss=2.268, over 5517704.87 frames. , ppl: 9.664491616871585], batch size: 70
+2022-12-13 21:36:10,344 INFO [train.py:421] (5/8) Epoch 10, batch 46600, loss[loss=2.364, over 2520.00 frames. , ppl: 10.635395861834716] tot_loss[loss=2.269, over 5516850.15 frames. , ppl: 9.669798273517568], batch size: 70
+2022-12-13 21:37:49,342 INFO [train.py:421] (5/8) Epoch 10, batch 46800, loss[loss=2.254, over 2450.00 frames. , ppl: 9.52539372748697] tot_loss[loss=2.27, over 5440079.39 frames. , ppl: 9.683509114245131], batch size: 70
+2022-12-13 21:39:32,524 INFO [train.py:421] (5/8) Epoch 10, batch 47000, loss[loss=2.145, over 8470.00 frames. , ppl: 8.544395791673493] tot_loss[loss=2.27, over 5437033.26 frames. , ppl: 9.684167012353612], batch size: 70
+2022-12-13 21:39:32,525 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 21:39:33,262 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.62268485087924
+2022-12-13 21:41:16,054 INFO [train.py:421] (5/8) Epoch 10, batch 47200, loss[loss=2.314, over 2450.00 frames. , ppl: 10.117951520933824] tot_loss[loss=2.271, over 5443409.75 frames. , ppl: 9.685322469908167], batch size: 70
+2022-12-13 21:42:58,755 INFO [train.py:421] (5/8) Epoch 10, batch 47400, loss[loss=2.196, over 8680.00 frames. , ppl: 8.993069481426529] tot_loss[loss=2.27, over 5471984.86 frames. , ppl: 9.67807551921869], batch size: 70
+2022-12-13 21:44:41,590 INFO [train.py:421] (5/8) Epoch 10, batch 47600, loss[loss=2.25, over 2170.00 frames. , ppl: 9.485592661750605] tot_loss[loss=2.269, over 5471235.62 frames. , ppl: 9.673517561114876], batch size: 70
+2022-12-13 21:46:26,615 INFO [train.py:421] (5/8) Epoch 10, batch 47800, loss[loss=2.641, over 910.00 frames. , ppl: 14.026784422689259] tot_loss[loss=2.269, over 5482015.86 frames. , ppl: 9.666389473109637], batch size: 70
+2022-12-13 21:48:08,378 INFO [train.py:421] (5/8) Epoch 10, batch 48000, loss[loss=2.252, over 3010.00 frames. , ppl: 9.510930638328643] tot_loss[loss=2.27, over 5425973.28 frames. , ppl: 9.682536779533502], batch size: 70
+2022-12-13 21:48:08,378 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 21:48:09,144 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.621020070987242
+2022-12-13 21:49:55,440 INFO [train.py:421] (5/8) Epoch 10, batch 48200, loss[loss=2.372, over 2450.00 frames. , ppl: 10.718682012946516] tot_loss[loss=2.269, over 5504953.77 frames. , ppl: 9.666855647925473], batch size: 70
+2022-12-13 21:51:37,488 INFO [train.py:421] (5/8) Epoch 10, batch 48400, loss[loss=2.445, over 1120.00 frames. , ppl: 11.52981468645539] tot_loss[loss=2.268, over 5496453.68 frames. , ppl: 9.664493881685678], batch size: 70
+2022-12-13 21:53:19,230 INFO [train.py:421] (5/8) Epoch 10, batch 48600, loss[loss=2.25, over 4340.00 frames. , ppl: 9.485208478439835] tot_loss[loss=2.268, over 5538320.48 frames. , ppl: 9.658064866694934], batch size: 70
+2022-12-13 21:55:03,419 INFO [train.py:421] (5/8) Epoch 10, batch 48800, loss[loss=2.243, over 2800.00 frames. , ppl: 9.423817575206108] tot_loss[loss=2.268, over 5526551.48 frames. , ppl: 9.658000114809923], batch size: 70
+2022-12-13 21:56:44,874 INFO [train.py:421] (5/8) Epoch 10, batch 49000, loss[loss=2.199, over 5670.00 frames. , ppl: 9.01481458220098] tot_loss[loss=2.268, over 5521864.02 frames. , ppl: 9.659855522029124], batch size: 70
+2022-12-13 21:56:44,875 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 21:56:45,606 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.608761699260915
+2022-12-13 21:58:25,302 INFO [train.py:421] (5/8) Epoch 10, batch 49200, loss[loss=2.158, over 3920.00 frames. , ppl: 8.651584612932142] tot_loss[loss=2.268, over 5496972.42 frames. , ppl: 9.66413138435069], batch size: 70
+2022-12-13 22:00:03,862 INFO [train.py:421] (5/8) Epoch 10, batch 49400, loss[loss=2.428, over 1120.00 frames. , ppl: 11.33151129187654] tot_loss[loss=2.269, over 5466505.86 frames. , ppl: 9.673731009585522], batch size: 70
+2022-12-13 22:01:44,565 INFO [train.py:421] (5/8) Epoch 10, batch 49600, loss[loss=2.395, over 1610.00 frames. , ppl: 10.972168548881726] tot_loss[loss=2.27, over 5428490.15 frames. , ppl: 9.680667605367423], batch size: 70
+2022-12-13 22:03:22,301 INFO [train.py:421] (5/8) Epoch 10, batch 49800, loss[loss=2.227, over 2240.00 frames. , ppl: 9.274260028430305] tot_loss[loss=2.27, over 5429342.71 frames. , ppl: 9.676985018663753], batch size: 70
+2022-12-13 22:05:00,295 INFO [train.py:421] (5/8) Epoch 10, batch 50000, loss[loss=2.229, over 4480.00 frames. , ppl: 9.293673799302445] tot_loss[loss=2.272, over 5400003.74 frames. , ppl: 9.696251642098733], batch size: 70
+2022-12-13 22:05:00,296 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 22:05:01,070 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.608694857576053
+2022-12-13 22:06:42,910 INFO [train.py:421] (5/8) Epoch 10, batch 50200, loss[loss=2.296, over 2100.00 frames. , ppl: 9.933933934501162] tot_loss[loss=2.271, over 5417551.07 frames. , ppl: 9.68953058706218], batch size: 70
+2022-12-13 22:08:19,153 INFO [train.py:421] (5/8) Epoch 10, batch 50400, loss[loss=2.431, over 1120.00 frames. , ppl: 11.370504114684762] tot_loss[loss=2.272, over 5384462.55 frames. , ppl: 9.697878446630002], batch size: 70
+2022-12-13 22:10:01,751 INFO [train.py:421] (5/8) Epoch 10, batch 50600, loss[loss=2.18, over 6930.00 frames. , ppl: 8.849362713509219] tot_loss[loss=2.27, over 5455636.40 frames. , ppl: 9.681597097987506], batch size: 70
+2022-12-13 22:11:42,211 INFO [train.py:421] (5/8) Epoch 10, batch 50800, loss[loss=2.308, over 1470.00 frames. , ppl: 10.050225766552117] tot_loss[loss=2.272, over 5406155.24 frames. , ppl: 9.701337656523464], batch size: 70
+2022-12-13 22:13:22,635 INFO [train.py:421] (5/8) Epoch 10, batch 51000, loss[loss=2.188, over 5320.00 frames. , ppl: 8.919236904855852] tot_loss[loss=2.272, over 5418430.66 frames. , ppl: 9.700636981154245], batch size: 70
+2022-12-13 22:13:22,636 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 22:13:23,385 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.617215941256719
+2022-12-13 22:15:05,584 INFO [train.py:421] (5/8) Epoch 10, batch 51200, loss[loss=2.393, over 1190.00 frames. , ppl: 10.949261124371295] tot_loss[loss=2.271, over 5464614.74 frames. , ppl: 9.690350052460543], batch size: 70
+2022-12-13 22:16:46,799 INFO [train.py:421] (5/8) Epoch 10, batch 51400, loss[loss=2.544, over 840.00 frames. , ppl: 12.733232657135831] tot_loss[loss=2.271, over 5456471.10 frames. , ppl: 9.68811744341817], batch size: 70
+2022-12-13 22:18:32,939 INFO [train.py:421] (5/8) Epoch 10, batch 51600, loss[loss=2.526, over 980.00 frames. , ppl: 12.500654474241088] tot_loss[loss=2.27, over 5463757.77 frames. , ppl: 9.68378330139242], batch size: 70
+2022-12-13 22:20:11,645 INFO [train.py:421] (5/8) Epoch 10, batch 51800, loss[loss=4.933, over 280.00 frames. , ppl: 138.74735377108348] tot_loss[loss=2.271, over 5445812.49 frames. , ppl: 9.690032791066928], batch size: 70
+2022-12-13 22:21:49,943 INFO [train.py:421] (5/8) Epoch 10, batch 52000, loss[loss=2.17, over 4620.00 frames. , ppl: 8.757864545924967] tot_loss[loss=2.271, over 5446314.71 frames. , ppl: 9.688227766917867], batch size: 70
+2022-12-13 22:21:49,944 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 22:21:50,672 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.599999450495101
+2022-12-13 22:23:31,973 INFO [train.py:421] (5/8) Epoch 10, batch 52200, loss[loss=2.849, over 630.00 frames. , ppl: 17.268163608826093] tot_loss[loss=2.272, over 5405395.35 frames. , ppl: 9.700885448451169], batch size: 70
+2022-12-13 22:25:13,161 INFO [train.py:421] (5/8) Epoch 10, batch 52400, loss[loss=2.481, over 910.00 frames. , ppl: 11.956399209556006] tot_loss[loss=2.271, over 5429035.30 frames. , ppl: 9.691676330644386], batch size: 70
+2022-12-13 22:26:53,088 INFO [train.py:421] (5/8) Epoch 10, batch 52600, loss[loss=2.3, over 2170.00 frames. , ppl: 9.970623575068302] tot_loss[loss=2.273, over 5396926.55 frames. , ppl: 9.705293727466316], batch size: 70
+2022-12-13 22:28:32,244 INFO [train.py:421] (5/8) Epoch 10, batch 52800, loss[loss=2.249, over 2590.00 frames. , ppl: 9.482328406874839] tot_loss[loss=2.275, over 5338902.83 frames. , ppl: 9.724175595191445], batch size: 70
+2022-12-13 22:30:12,637 INFO [train.py:421] (5/8) Epoch 10, batch 53000, loss[loss=2.835, over 560.00 frames. , ppl: 17.034019466771937] tot_loss[loss=2.273, over 5404655.72 frames. , ppl: 9.705577302579263], batch size: 70
+2022-12-13 22:30:12,637 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 22:30:13,403 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.621981307382367
+2022-12-13 22:31:58,583 INFO [train.py:421] (5/8) Epoch 10, batch 53200, loss[loss=2.259, over 3850.00 frames. , ppl: 9.571278021083597] tot_loss[loss=2.272, over 5420016.40 frames. , ppl: 9.697635654486806], batch size: 70
+2022-12-13 22:33:42,494 INFO [train.py:421] (5/8) Epoch 10, batch 53400, loss[loss=2.304, over 3150.00 frames. , ppl: 10.013065087537921] tot_loss[loss=2.271, over 5446611.08 frames. , ppl: 9.687830052651353], batch size: 70
+2022-12-13 22:35:23,717 INFO [train.py:421] (5/8) Epoch 10, batch 53600, loss[loss=2.195, over 1890.00 frames. , ppl: 8.98383355681908] tot_loss[loss=2.273, over 5407977.75 frames. , ppl: 9.705639870983573], batch size: 70
+2022-12-13 22:37:03,239 INFO [train.py:421] (5/8) Epoch 10, batch 53800, loss[loss=2.359, over 2030.00 frames. , ppl: 10.582491120852126] tot_loss[loss=2.272, over 5423929.28 frames. , ppl: 9.702490736415303], batch size: 70
+2022-12-13 22:38:45,605 INFO [train.py:421] (5/8) Epoch 10, batch 54000, loss[loss=2.243, over 10080.00 frames. , ppl: 9.422836526405144] tot_loss[loss=2.271, over 5487316.42 frames. , ppl: 9.688326672362738], batch size: 70
+2022-12-13 22:38:45,605 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 22:38:46,336 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.606318724465547
+2022-12-13 22:40:29,332 INFO [train.py:421] (5/8) Epoch 10, batch 54200, loss[loss=2.391, over 1050.00 frames. , ppl: 10.91981934938527] tot_loss[loss=2.272, over 5471932.02 frames. , ppl: 9.694468807799074], batch size: 70
+2022-12-13 22:42:07,612 INFO [train.py:421] (5/8) Epoch 10, batch 54400, loss[loss=2.229, over 5320.00 frames. , ppl: 9.288236915442772] tot_loss[loss=2.272, over 5436236.04 frames. , ppl: 9.703566463629363], batch size: 70
+2022-12-13 22:43:49,814 INFO [train.py:421] (5/8) Epoch 10, batch 54600, loss[loss=2.267, over 4200.00 frames. , ppl: 9.646391343479287] tot_loss[loss=2.271, over 5502685.58 frames. , ppl: 9.68624020977279], batch size: 70
+2022-12-13 22:45:29,927 INFO [train.py:421] (5/8) Epoch 10, batch 54800, loss[loss=2.369, over 1120.00 frames. , ppl: 10.682710802579132] tot_loss[loss=2.269, over 5549241.20 frames. , ppl: 9.672142150611716], batch size: 70
+2022-12-13 22:47:09,472 INFO [train.py:421] (5/8) Epoch 10, batch 55000, loss[loss=2.133, over 6720.00 frames. , ppl: 8.442106803424736] tot_loss[loss=2.268, over 5597074.11 frames. , ppl: 9.657095927434751], batch size: 70
+2022-12-13 22:47:09,473 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 22:47:10,234 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.265, over 211138.00 frames. , ppl: 9.6300680375108
+2022-12-13 22:48:52,900 INFO [train.py:421] (5/8) Epoch 10, batch 55200, loss[loss=2.216, over 4060.00 frames. , ppl: 9.174914894306056] tot_loss[loss=2.266, over 5657733.79 frames. , ppl: 9.637547309195538], batch size: 70
+2022-12-13 22:50:33,397 INFO [train.py:421] (5/8) Epoch 10, batch 55400, loss[loss=2.263, over 7630.00 frames. , ppl: 9.612070368625785] tot_loss[loss=2.268, over 5577884.38 frames. , ppl: 9.663265835946127], batch size: 70
+2022-12-13 22:52:12,072 INFO [train.py:421] (5/8) Epoch 10, batch 55600, loss[loss=2.368, over 1330.00 frames. , ppl: 10.672150590300838] tot_loss[loss=2.268, over 5569019.65 frames. , ppl: 9.663526433490265], batch size: 70
+2022-12-13 22:53:51,906 INFO [train.py:421] (5/8) Epoch 10, batch 55800, loss[loss=2.302, over 3150.00 frames. , ppl: 9.993553740431869] tot_loss[loss=2.267, over 5602962.67 frames. , ppl: 9.65150009853289], batch size: 70
+2022-12-13 22:55:32,380 INFO [train.py:421] (5/8) Epoch 10, batch 56000, loss[loss=2.349, over 2240.00 frames. , ppl: 10.472633945448935] tot_loss[loss=2.268, over 5586624.13 frames. , ppl: 9.65589340815708], batch size: 70
+2022-12-13 22:55:32,380 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 22:55:33,125 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.602260323357212
+2022-12-13 22:57:14,363 INFO [train.py:421] (5/8) Epoch 10, batch 56200, loss[loss=2.342, over 1960.00 frames. , ppl: 10.405898903166898] tot_loss[loss=2.267, over 5635199.85 frames. , ppl: 9.646286581750196], batch size: 70
+2022-12-13 22:58:56,428 INFO [train.py:421] (5/8) Epoch 10, batch 56400, loss[loss=2.316, over 2730.00 frames. , ppl: 10.134515191479293] tot_loss[loss=2.267, over 5620811.32 frames. , ppl: 9.653283685116982], batch size: 70
+2022-12-13 23:00:37,873 INFO [train.py:421] (5/8) Epoch 10, batch 56600, loss[loss=2.227, over 3010.00 frames. , ppl: 9.276318603524413] tot_loss[loss=2.268, over 5583749.14 frames. , ppl: 9.661387348796717], batch size: 70
+2022-12-13 23:02:17,338 INFO [train.py:421] (5/8) Epoch 10, batch 56800, loss[loss=2.455, over 1050.00 frames. , ppl: 11.647885110773236] tot_loss[loss=2.269, over 5566332.28 frames. , ppl: 9.66592790849176], batch size: 70
+2022-12-13 23:03:57,912 INFO [train.py:421] (5/8) Epoch 10, batch 57000, loss[loss=2.336, over 1680.00 frames. , ppl: 10.339765886229124] tot_loss[loss=2.27, over 5518864.11 frames. , ppl: 9.67759953066858], batch size: 70
+2022-12-13 23:03:57,913 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 23:03:58,673 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.264, over 211138.00 frames. , ppl: 9.62573033470701
+2022-12-13 23:05:40,807 INFO [train.py:421] (5/8) Epoch 10, batch 57200, loss[loss=2.285, over 2520.00 frames. , ppl: 9.822195947366268] tot_loss[loss=2.27, over 5515027.01 frames. , ppl: 9.676811855422871], batch size: 70
+2022-12-13 23:07:18,411 INFO [train.py:421] (5/8) Epoch 10, batch 57400, loss[loss=2.22, over 4550.00 frames. , ppl: 9.207342722857861] tot_loss[loss=2.27, over 5518103.55 frames. , ppl: 9.67688185927107], batch size: 70
+2022-12-13 23:08:53,700 INFO [train.py:421] (5/8) Epoch 10, batch 57600, loss[loss=2.764, over 700.00 frames. , ppl: 15.86130837171714] tot_loss[loss=2.27, over 5516870.46 frames. , ppl: 9.677698148769434], batch size: 70
+2022-12-13 23:10:36,336 INFO [train.py:421] (5/8) Epoch 10, batch 57800, loss[loss=2.453, over 1260.00 frames. , ppl: 11.624261069517486] tot_loss[loss=2.271, over 5471965.84 frames. , ppl: 9.689237417677683], batch size: 70
+2022-12-13 23:12:20,210 INFO [train.py:421] (5/8) Epoch 10, batch 58000, loss[loss=2.656, over 630.00 frames. , ppl: 14.23856770735455] tot_loss[loss=2.271, over 5462686.60 frames. , ppl: 9.69085891728332], batch size: 70
+2022-12-13 23:12:20,211 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 23:12:20,972 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.609582325499408
+2022-12-13 23:14:05,369 INFO [train.py:421] (5/8) Epoch 10, batch 58200, loss[loss=2.446, over 1190.00 frames. , ppl: 11.54621157320038] tot_loss[loss=2.27, over 5501544.96 frames. , ppl: 9.67529934885576], batch size: 70
+2022-12-13 23:15:40,222 INFO [train.py:421] (5/8) Epoch 10, batch 58400, loss[loss=2.936, over 700.00 frames. , ppl: 18.83886089451121] tot_loss[loss=2.271, over 5431708.19 frames. , ppl: 9.691366972831021], batch size: 70
+2022-12-13 23:17:18,338 INFO [train.py:421] (5/8) Epoch 10, batch 58600, loss[loss=2.223, over 3220.00 frames. , ppl: 9.234500174473464] tot_loss[loss=2.272, over 5404169.59 frames. , ppl: 9.70168773419045], batch size: 70
+2022-12-13 23:18:57,520 INFO [train.py:421] (5/8) Epoch 10, batch 58800, loss[loss=2.575, over 1400.00 frames. , ppl: 13.13139959586822] tot_loss[loss=2.272, over 5417959.01 frames. , ppl: 9.696607065833302], batch size: 70
+2022-12-13 23:20:40,942 INFO [train.py:421] (5/8) Epoch 10, batch 59000, loss[loss=2.232, over 2870.00 frames. , ppl: 9.319344817635328] tot_loss[loss=2.271, over 5449978.69 frames. , ppl: 9.684891767273673], batch size: 70
+2022-12-13 23:20:40,942 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 23:20:41,703 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.606206402379026
+2022-12-13 23:22:20,234 INFO [train.py:421] (5/8) Epoch 10, batch 59200, loss[loss=2.201, over 3290.00 frames. , ppl: 9.03270781975402] tot_loss[loss=2.27, over 5452145.20 frames. , ppl: 9.684155446535822], batch size: 70
+2022-12-13 23:24:00,428 INFO [train.py:421] (5/8) Epoch 10, batch 59400, loss[loss=2.415, over 1330.00 frames. , ppl: 11.194949387940651] tot_loss[loss=2.271, over 5425194.72 frames. , ppl: 9.687069339359104], batch size: 70
+2022-12-13 23:25:41,349 INFO [train.py:421] (5/8) Epoch 10, batch 59600, loss[loss=2.472, over 1050.00 frames. , ppl: 11.844901880367267] tot_loss[loss=2.27, over 5456750.01 frames. , ppl: 9.679258521028796], batch size: 70
+2022-12-13 23:27:19,478 INFO [train.py:421] (5/8) Epoch 10, batch 59800, loss[loss=2.297, over 2240.00 frames. , ppl: 9.942441924514222] tot_loss[loss=2.271, over 5409845.33 frames. , ppl: 9.692371681323188], batch size: 70
+2022-12-13 23:28:53,373 INFO [train.py:421] (5/8) Epoch 10, batch 60000, loss[loss=2.563, over 840.00 frames. , ppl: 12.974877924110691] tot_loss[loss=2.273, over 5369086.71 frames. , ppl: 9.704189598350178], batch size: 70
+2022-12-13 23:28:53,373 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 23:28:54,130 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.612700490778435
+2022-12-13 23:30:35,192 INFO [train.py:421] (5/8) Epoch 10, batch 60200, loss[loss=2.258, over 4760.00 frames. , ppl: 9.562415875548675] tot_loss[loss=2.273, over 5335097.92 frames. , ppl: 9.713126676907528], batch size: 70
+2022-12-13 23:32:13,575 INFO [train.py:421] (5/8) Epoch 10, batch 60400, loss[loss=2.666, over 980.00 frames. , ppl: 14.385649058168276] tot_loss[loss=2.273, over 5365696.92 frames. , ppl: 9.710417794102263], batch size: 70
+2022-12-13 23:33:56,148 INFO [train.py:421] (5/8) Epoch 10, batch 60600, loss[loss=2.229, over 2450.00 frames. , ppl: 9.291447819301187] tot_loss[loss=2.272, over 5390910.05 frames. , ppl: 9.702944810831731], batch size: 70
+2022-12-13 23:35:35,542 INFO [train.py:421] (5/8) Epoch 10, batch 60800, loss[loss=2.422, over 1400.00 frames. , ppl: 11.27067130906126] tot_loss[loss=2.273, over 5382482.09 frames. , ppl: 9.708922847009353], batch size: 70
+2022-12-13 23:37:16,253 INFO [train.py:421] (5/8) Epoch 10, batch 61000, loss[loss=2.142, over 6790.00 frames. , ppl: 8.515252837067246] tot_loss[loss=2.273, over 5361904.36 frames. , ppl: 9.70786543233044], batch size: 70
+2022-12-13 23:37:16,253 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 23:37:16,983 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.60258010007721
+2022-12-13 23:38:59,955 INFO [train.py:421] (5/8) Epoch 10, batch 61200, loss[loss=2.255, over 2940.00 frames. , ppl: 9.534522741507896] tot_loss[loss=2.271, over 5445126.84 frames. , ppl: 9.684657012370185], batch size: 70
+2022-12-13 23:40:40,175 INFO [train.py:421] (5/8) Epoch 10, batch 61400, loss[loss=2.629, over 910.00 frames. , ppl: 13.860916597370137] tot_loss[loss=2.271, over 5431397.44 frames. , ppl: 9.68554977187262], batch size: 70
+2022-12-13 23:42:19,009 INFO [train.py:421] (5/8) Epoch 10, batch 61600, loss[loss=2.256, over 2660.00 frames. , ppl: 9.541876160414658] tot_loss[loss=2.271, over 5423811.58 frames. , ppl: 9.688946332601851], batch size: 70
+2022-12-13 23:44:03,588 INFO [train.py:421] (5/8) Epoch 10, batch 61800, loss[loss=2.732, over 630.00 frames. , ppl: 15.361756615047218] tot_loss[loss=2.269, over 5488105.00 frames. , ppl: 9.668628944953078], batch size: 70
+2022-12-13 23:45:42,347 INFO [train.py:421] (5/8) Epoch 10, batch 62000, loss[loss=2.324, over 1890.00 frames. , ppl: 10.215644869127376] tot_loss[loss=2.271, over 5448858.21 frames. , ppl: 9.686804824342731], batch size: 70
+2022-12-13 23:45:42,347 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 23:45:43,104 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.61007587277309
+2022-12-13 23:47:24,544 INFO [train.py:421] (5/8) Epoch 10, batch 62200, loss[loss=2.576, over 1190.00 frames. , ppl: 13.145374009544685] tot_loss[loss=2.27, over 5451410.17 frames. , ppl: 9.682249760662536], batch size: 70
+2022-12-13 23:49:09,224 INFO [train.py:421] (5/8) Epoch 10, batch 62400, loss[loss=3.273, over 490.00 frames. , ppl: 26.39144486667393] tot_loss[loss=2.271, over 5443766.13 frames. , ppl: 9.688423772639132], batch size: 70
+2022-12-13 23:50:45,969 INFO [train.py:421] (5/8) Epoch 10, batch 62600, loss[loss=2.86, over 630.00 frames. , ppl: 17.459748718722405] tot_loss[loss=2.271, over 5437102.58 frames. , ppl: 9.684918711684919], batch size: 70
+2022-12-13 23:52:24,484 INFO [train.py:421] (5/8) Epoch 10, batch 62800, loss[loss=2.409, over 1330.00 frames. , ppl: 11.117955818688905] tot_loss[loss=2.27, over 5436467.88 frames. , ppl: 9.683524385475097], batch size: 70
+2022-12-13 23:54:03,434 INFO [train.py:421] (5/8) Epoch 10, batch 63000, loss[loss=2.207, over 2870.00 frames. , ppl: 9.089988573854086] tot_loss[loss=2.27, over 5454748.73 frames. , ppl: 9.678902807169306], batch size: 70
+2022-12-13 23:54:03,435 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-13 23:54:04,195 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.606716838640049
+2022-12-13 23:55:42,812 INFO [train.py:421] (5/8) Epoch 10, batch 63200, loss[loss=2.453, over 1400.00 frames. , ppl: 11.625541372764832] tot_loss[loss=2.271, over 5410436.64 frames. , ppl: 9.688867870398619], batch size: 70
+2022-12-13 23:57:21,549 INFO [train.py:421] (5/8) Epoch 10, batch 63400, loss[loss=2.487, over 1190.00 frames. , ppl: 12.031155653612178] tot_loss[loss=2.271, over 5395289.47 frames. , ppl: 9.692904300678299], batch size: 70
+2022-12-13 23:59:00,620 INFO [train.py:421] (5/8) Epoch 10, batch 63600, loss[loss=2.288, over 1610.00 frames. , ppl: 9.856943822158513] tot_loss[loss=2.271, over 5417988.58 frames. , ppl: 9.689791956086106], batch size: 70
+2022-12-14 00:00:37,636 INFO [train.py:421] (5/8) Epoch 10, batch 63800, loss[loss=2.342, over 1890.00 frames. , ppl: 10.401011455923975] tot_loss[loss=2.271, over 5437095.32 frames. , ppl: 9.684615681266179], batch size: 70
+2022-12-14 00:02:18,824 INFO [train.py:421] (5/8) Epoch 10, batch 64000, loss[loss=2.375, over 2100.00 frames. , ppl: 10.753013188523582] tot_loss[loss=2.27, over 5450277.66 frames. , ppl: 9.682487865822928], batch size: 70
+2022-12-14 00:02:18,825 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 00:02:19,585 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.616598196927988
+2022-12-14 00:03:57,270 INFO [train.py:421] (5/8) Epoch 10, batch 64200, loss[loss=2.633, over 840.00 frames. , ppl: 13.911583232556302] tot_loss[loss=2.27, over 5446991.89 frames. , ppl: 9.680072695825126], batch size: 70
+2022-12-14 00:05:36,743 INFO [train.py:421] (5/8) Epoch 10, batch 64400, loss[loss=2.26, over 2660.00 frames. , ppl: 9.579144638325571] tot_loss[loss=2.269, over 5456711.48 frames. , ppl: 9.672357872025247], batch size: 70
+2022-12-14 00:07:12,134 INFO [train.py:421] (5/8) Epoch 10, batch 64600, loss[loss=2.244, over 3430.00 frames. , ppl: 9.433311954594606] tot_loss[loss=2.271, over 5399225.89 frames. , ppl: 9.688043006904845], batch size: 70
+2022-12-14 00:08:51,365 INFO [train.py:421] (5/8) Epoch 10, batch 64800, loss[loss=2.158, over 5110.00 frames. , ppl: 8.653420570401655] tot_loss[loss=2.27, over 5420682.74 frames. , ppl: 9.680566639085976], batch size: 70
+2022-12-14 00:10:28,828 INFO [train.py:421] (5/8) Epoch 10, batch 65000, loss[loss=2.185, over 5110.00 frames. , ppl: 8.88740323267715] tot_loss[loss=2.27, over 5425959.84 frames. , ppl: 9.681421686642532], batch size: 70
+2022-12-14 00:10:28,828 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 00:10:29,560 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.599982400051333
+2022-12-14 00:12:08,439 INFO [train.py:421] (5/8) Epoch 10, batch 65200, loss[loss=2.551, over 700.00 frames. , ppl: 12.81967942478029] tot_loss[loss=2.27, over 5417125.00 frames. , ppl: 9.68280045327238], batch size: 70
+2022-12-14 00:13:46,654 INFO [train.py:421] (5/8) Epoch 10, batch 65400, loss[loss=2.409, over 1540.00 frames. , ppl: 11.1267929159907] tot_loss[loss=2.27, over 5421609.48 frames. , ppl: 9.680517298252665], batch size: 70
+2022-12-14 00:15:25,095 INFO [train.py:421] (5/8) Epoch 10, batch 65600, loss[loss=2.379, over 1750.00 frames. , ppl: 10.796040089146082] tot_loss[loss=2.27, over 5430658.23 frames. , ppl: 9.682621563317758], batch size: 70
+2022-12-14 00:17:06,085 INFO [train.py:421] (5/8) Epoch 10, batch 65800, loss[loss=2.518, over 910.00 frames. , ppl: 12.407150836762696] tot_loss[loss=2.27, over 5441357.51 frames. , ppl: 9.6789916594075], batch size: 70
+2022-12-14 00:18:46,934 INFO [train.py:421] (5/8) Epoch 10, batch 66000, loss[loss=2.665, over 700.00 frames. , ppl: 14.374535697110806] tot_loss[loss=2.269, over 5482842.84 frames. , ppl: 9.668250694654038], batch size: 70
+2022-12-14 00:18:46,935 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 00:18:47,681 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.599644239172868
+2022-12-14 00:20:26,482 INFO [train.py:421] (5/8) Epoch 10, batch 66200, loss[loss=2.401, over 1680.00 frames. , ppl: 11.038714614375435] tot_loss[loss=2.27, over 5463279.71 frames. , ppl: 9.682694501196902], batch size: 70
+2022-12-14 00:22:05,234 INFO [train.py:421] (5/8) Epoch 10, batch 66400, loss[loss=2.618, over 700.00 frames. , ppl: 13.712742966857114] tot_loss[loss=2.27, over 5479760.73 frames. , ppl: 9.678298186319415], batch size: 70
+2022-12-14 00:23:48,624 INFO [train.py:421] (5/8) Epoch 10, batch 66600, loss[loss=2.22, over 2590.00 frames. , ppl: 9.209668614299277] tot_loss[loss=2.27, over 5483358.12 frames. , ppl: 9.677547662919185], batch size: 70
+2022-12-14 00:25:27,853 INFO [train.py:421] (5/8) Epoch 10, batch 66800, loss[loss=2.274, over 3500.00 frames. , ppl: 9.720728873811165] tot_loss[loss=2.271, over 5436951.17 frames. , ppl: 9.68843073064957], batch size: 70
+2022-12-14 00:27:09,577 INFO [train.py:421] (5/8) Epoch 10, batch 67000, loss[loss=2.196, over 8890.00 frames. , ppl: 8.99191025759188] tot_loss[loss=2.269, over 5502680.48 frames. , ppl: 9.669005065949753], batch size: 70
+2022-12-14 00:27:09,577 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 00:27:10,307 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.608690591101315
+2022-12-14 00:28:53,508 INFO [train.py:421] (5/8) Epoch 10, batch 67200, loss[loss=2.318, over 2660.00 frames. , ppl: 10.152765042889596] tot_loss[loss=2.268, over 5526789.36 frames. , ppl: 9.664349793370889], batch size: 70
+2022-12-14 00:30:34,765 INFO [train.py:421] (5/8) Epoch 10, batch 67400, loss[loss=2.49, over 980.00 frames. , ppl: 12.055660963549137] tot_loss[loss=2.268, over 5546259.73 frames. , ppl: 9.662393186616587], batch size: 70
+2022-12-14 00:32:15,562 INFO [train.py:421] (5/8) Epoch 10, batch 67600, loss[loss=2.262, over 4200.00 frames. , ppl: 9.601650289339108] tot_loss[loss=2.268, over 5539933.19 frames. , ppl: 9.659714382002157], batch size: 70
+2022-12-14 00:33:55,294 INFO [train.py:421] (5/8) Epoch 10, batch 67800, loss[loss=2.263, over 4900.00 frames. , ppl: 9.610546880724215] tot_loss[loss=2.267, over 5579481.65 frames. , ppl: 9.653346127786499], batch size: 70
+2022-12-14 00:35:36,607 INFO [train.py:421] (5/8) Epoch 10, batch 68000, loss[loss=2.395, over 3290.00 frames. , ppl: 10.973635712433621] tot_loss[loss=2.267, over 5577458.55 frames. , ppl: 9.649805362571895], batch size: 70
+2022-12-14 00:35:36,607 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 00:35:37,368 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.606830588577454
+2022-12-14 00:37:17,019 INFO [train.py:421] (5/8) Epoch 10, batch 68200, loss[loss=2.145, over 5880.00 frames. , ppl: 8.54563068363094] tot_loss[loss=2.266, over 5561133.61 frames. , ppl: 9.642205204588839], batch size: 70
+2022-12-14 00:38:58,133 INFO [train.py:421] (5/8) Epoch 10, batch 68400, loss[loss=2.49, over 1050.00 frames. , ppl: 12.0667207401796] tot_loss[loss=2.265, over 5575794.54 frames. , ppl: 9.634874647251614], batch size: 70
+2022-12-14 00:40:37,840 INFO [train.py:421] (5/8) Epoch 10, batch 68600, loss[loss=2.267, over 1400.00 frames. , ppl: 9.648844859625514] tot_loss[loss=2.266, over 5549896.52 frames. , ppl: 9.640267489031348], batch size: 70
+2022-12-14 00:42:17,523 INFO [train.py:421] (5/8) Epoch 10, batch 68800, loss[loss=2.219, over 3640.00 frames. , ppl: 9.198882455251399] tot_loss[loss=2.266, over 5570247.08 frames. , ppl: 9.636928810395998], batch size: 70
+2022-12-14 00:44:01,099 INFO [train.py:421] (5/8) Epoch 10, batch 69000, loss[loss=2.114, over 3430.00 frames. , ppl: 8.277706705495378] tot_loss[loss=2.266, over 5567617.93 frames. , ppl: 9.641306196619896], batch size: 70
+2022-12-14 00:44:01,099 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 00:44:01,845 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.60023105535575
+2022-12-14 00:45:40,104 INFO [train.py:421] (5/8) Epoch 10, batch 69200, loss[loss=3.354, over 490.00 frames. , ppl: 28.61883852096619] tot_loss[loss=2.265, over 5608417.03 frames. , ppl: 9.628268817375126], batch size: 70
+2022-12-14 00:47:14,230 INFO [train.py:421] (5/8) Epoch 10, batch 69400, loss[loss=2.154, over 4410.00 frames. , ppl: 8.62342018937429] tot_loss[loss=2.266, over 5582018.66 frames. , ppl: 9.64184377348632], batch size: 70
+2022-12-14 00:48:53,975 INFO [train.py:421] (5/8) Epoch 10, batch 69600, loss[loss=2.174, over 5040.00 frames. , ppl: 8.790872778244522] tot_loss[loss=2.266, over 5614146.24 frames. , ppl: 9.639002155492374], batch size: 70
+2022-12-14 00:50:33,906 INFO [train.py:421] (5/8) Epoch 10, batch 69800, loss[loss=2.415, over 1330.00 frames. , ppl: 11.187044563396645] tot_loss[loss=2.267, over 5591276.35 frames. , ppl: 9.646934900912235], batch size: 70
+2022-12-14 00:52:14,676 INFO [train.py:421] (5/8) Epoch 10, batch 70000, loss[loss=2.443, over 1190.00 frames. , ppl: 11.504012493003806] tot_loss[loss=2.267, over 5565005.60 frames. , ppl: 9.648597716811384], batch size: 70
+2022-12-14 00:52:14,677 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 00:52:15,423 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.262, over 211138.00 frames. , ppl: 9.601824022965605
+2022-12-14 00:53:54,452 INFO [train.py:421] (5/8) Epoch 10, batch 70200, loss[loss=2.345, over 1960.00 frames. , ppl: 10.430111057562435] tot_loss[loss=2.268, over 5548931.21 frames. , ppl: 9.657534033145467], batch size: 70
+2022-12-14 00:55:36,203 INFO [train.py:421] (5/8) Epoch 10, batch 70400, loss[loss=2.345, over 1680.00 frames. , ppl: 10.429479020422555] tot_loss[loss=2.268, over 5528800.61 frames. , ppl: 9.658286856590077], batch size: 70
+2022-12-14 00:57:16,765 INFO [train.py:421] (5/8) Epoch 10, batch 70600, loss[loss=2.793, over 700.00 frames. , ppl: 16.331134561491723] tot_loss[loss=2.268, over 5543290.38 frames. , ppl: 9.656626829224878], batch size: 70
+2022-12-14 00:58:55,099 INFO [train.py:421] (5/8) Epoch 10, batch 70800, loss[loss=2.491, over 1260.00 frames. , ppl: 12.070668994812953] tot_loss[loss=2.267, over 5569109.02 frames. , ppl: 9.654263367942258], batch size: 70
+2022-12-14 01:00:36,976 INFO [train.py:421] (5/8) Epoch 10, batch 71000, loss[loss=2.23, over 6090.00 frames. , ppl: 9.296768012246828] tot_loss[loss=2.268, over 5576489.73 frames. , ppl: 9.657048343850137], batch size: 70
+2022-12-14 01:00:36,976 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 01:00:37,735 INFO [train.py:452] (5/8) Epoch 10, validation: loss=2.263, over 211138.00 frames. , ppl: 9.609851142052824
+2022-12-14 01:02:17,478 INFO [train.py:421] (5/8) Epoch 10, batch 71200, loss[loss=2.315, over 3290.00 frames. , ppl: 10.12136700545702] tot_loss[loss=2.268, over 5572627.83 frames. , ppl: 9.662657798177731], batch size: 70
+2022-12-14 01:03:55,204 INFO [train.py:421] (5/8) Epoch 10, batch 71400, loss[loss=2.38, over 1330.00 frames. , ppl: 10.80020244643063] tot_loss[loss=2.27, over 5520899.19 frames. , ppl: 9.67688771827748], batch size: 70
+2022-12-14 01:05:31,328 INFO [train.py:421] (5/8) Epoch 10, batch 71600, loss[loss=2.482, over 1470.00 frames. , ppl: 11.961578844416358] tot_loss[loss=2.269, over 5505029.52 frames. , ppl: 9.674230459509362], batch size: 70
+2022-12-14 01:07:12,760 INFO [train.py:421] (5/8) Epoch 10, batch 71800, loss[loss=2.172, over 3920.00 frames. , ppl: 8.777060544769549] tot_loss[loss=2.27, over 5492029.84 frames. , ppl: 9.681460518910372], batch size: 70
+2022-12-14 01:08:27,258 INFO [train.py:421] (5/8) Epoch 11, batch 0, loss[loss=2.333, over 1400.00 frames. , ppl: 10.310558034831724] tot_loss[loss=2.333, over 1400.00 frames. , ppl: 10.310558034831724], batch size: 70
+2022-12-14 01:10:06,171 INFO [train.py:421] (5/8) Epoch 11, batch 200, loss[loss=2.287, over 2100.00 frames. , ppl: 9.842108060155224] tot_loss[loss=2.253, over 532351.25 frames. , ppl: 9.515447499846234], batch size: 70
+2022-12-14 01:11:46,145 INFO [train.py:421] (5/8) Epoch 11, batch 400, loss[loss=3.472, over 420.00 frames. , ppl: 32.20130411496333] tot_loss[loss=2.254, over 1006683.95 frames. , ppl: 9.527806594051276], batch size: 70
+2022-12-14 01:13:23,774 INFO [train.py:421] (5/8) Epoch 11, batch 600, loss[loss=2.355, over 2030.00 frames. , ppl: 10.534663902248411] tot_loss[loss=2.257, over 1436771.53 frames. , ppl: 9.557751744129881], batch size: 70
+2022-12-14 01:15:02,896 INFO [train.py:421] (5/8) Epoch 11, batch 800, loss[loss=2.2, over 5740.00 frames. , ppl: 9.02142125878662] tot_loss[loss=2.255, over 1858244.41 frames. , ppl: 9.536996046714485], batch size: 70
+2022-12-14 01:16:43,480 INFO [train.py:421] (5/8) Epoch 11, batch 1000, loss[loss=2.398, over 2030.00 frames. , ppl: 11.006453229448942] tot_loss[loss=2.258, over 2181219.33 frames. , ppl: 9.564164524507769], batch size: 70
+2022-12-14 01:16:43,480 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 01:16:44,237 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.264, over 211138.00 frames. , ppl: 9.619912276163067
+2022-12-14 01:18:27,845 INFO [train.py:421] (5/8) Epoch 11, batch 1200, loss[loss=2.209, over 4270.00 frames. , ppl: 9.110245870097756] tot_loss[loss=2.257, over 2541148.27 frames. , ppl: 9.550646856196005], batch size: 70
+2022-12-14 01:20:08,976 INFO [train.py:421] (5/8) Epoch 11, batch 1400, loss[loss=2.392, over 1050.00 frames. , ppl: 10.938181732972685] tot_loss[loss=2.259, over 2813033.51 frames. , ppl: 9.577635358428209], batch size: 70
+2022-12-14 01:21:47,195 INFO [train.py:421] (5/8) Epoch 11, batch 1600, loss[loss=2.481, over 840.00 frames. , ppl: 11.957614737227537] tot_loss[loss=2.259, over 3057417.65 frames. , ppl: 9.577276429970164], batch size: 70
+2022-12-14 01:23:27,107 INFO [train.py:421] (5/8) Epoch 11, batch 1800, loss[loss=2.292, over 2730.00 frames. , ppl: 9.899239467176589] tot_loss[loss=2.262, over 3265800.28 frames. , ppl: 9.602887019191131], batch size: 70
+2022-12-14 01:25:04,642 INFO [train.py:421] (5/8) Epoch 11, batch 2000, loss[loss=2.392, over 1190.00 frames. , ppl: 10.930845306636241] tot_loss[loss=2.261, over 3492102.61 frames. , ppl: 9.596633778091062], batch size: 70
+2022-12-14 01:25:04,643 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 01:25:05,404 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.264, over 211138.00 frames. , ppl: 9.623191889826966
+2022-12-14 01:26:47,629 INFO [train.py:421] (5/8) Epoch 11, batch 2200, loss[loss=2.493, over 980.00 frames. , ppl: 12.10004478431252] tot_loss[loss=2.262, over 3678751.56 frames. , ppl: 9.60685192469297], batch size: 70
+2022-12-14 01:28:26,721 INFO [train.py:421] (5/8) Epoch 11, batch 2400, loss[loss=2.439, over 770.00 frames. , ppl: 11.460632686623004] tot_loss[loss=2.264, over 3848510.97 frames. , ppl: 9.617078151177537], batch size: 70
+2022-12-14 01:30:12,135 INFO [train.py:421] (5/8) Epoch 11, batch 2600, loss[loss=2.296, over 3010.00 frames. , ppl: 9.93386571543117] tot_loss[loss=2.263, over 4041133.31 frames. , ppl: 9.613784056642913], batch size: 70
+2022-12-14 01:31:49,623 INFO [train.py:421] (5/8) Epoch 11, batch 2800, loss[loss=2.245, over 2940.00 frames. , ppl: 9.442316963621849] tot_loss[loss=2.265, over 4154858.74 frames. , ppl: 9.626794553391743], batch size: 70
+2022-12-14 01:33:29,897 INFO [train.py:421] (5/8) Epoch 11, batch 3000, loss[loss=2.307, over 2380.00 frames. , ppl: 10.040534671614942] tot_loss[loss=2.264, over 4298904.59 frames. , ppl: 9.625415044006429], batch size: 70
+2022-12-14 01:33:29,897 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 01:33:30,660 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.612494194035516
+2022-12-14 01:35:15,619 INFO [train.py:421] (5/8) Epoch 11, batch 3200, loss[loss=2.292, over 2520.00 frames. , ppl: 9.889939723310285] tot_loss[loss=2.264, over 4421539.92 frames. , ppl: 9.620236252174962], batch size: 70
+2022-12-14 01:36:56,902 INFO [train.py:421] (5/8) Epoch 11, batch 3400, loss[loss=2.606, over 910.00 frames. , ppl: 13.542767863684649] tot_loss[loss=2.264, over 4514149.48 frames. , ppl: 9.620330811876341], batch size: 70
+2022-12-14 01:38:36,970 INFO [train.py:421] (5/8) Epoch 11, batch 3600, loss[loss=2.358, over 2240.00 frames. , ppl: 10.565970954766968] tot_loss[loss=2.264, over 4585770.82 frames. , ppl: 9.62433015781048], batch size: 70
+2022-12-14 01:40:18,049 INFO [train.py:421] (5/8) Epoch 11, batch 3800, loss[loss=2.144, over 7070.00 frames. , ppl: 8.530248576110116] tot_loss[loss=2.266, over 4619399.70 frames. , ppl: 9.644346063725845], batch size: 70
+2022-12-14 01:42:00,374 INFO [train.py:421] (5/8) Epoch 11, batch 4000, loss[loss=2.145, over 5460.00 frames. , ppl: 8.540827021006173] tot_loss[loss=2.267, over 4683694.27 frames. , ppl: 9.65214313906238], batch size: 70
+2022-12-14 01:42:00,374 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 01:42:01,137 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.606853338726559
+2022-12-14 01:43:42,448 INFO [train.py:421] (5/8) Epoch 11, batch 4200, loss[loss=2.43, over 1540.00 frames. , ppl: 11.362463988601462] tot_loss[loss=2.266, over 4780042.80 frames. , ppl: 9.63927182561042], batch size: 70
+2022-12-14 01:45:19,906 INFO [train.py:421] (5/8) Epoch 11, batch 4400, loss[loss=2.344, over 2310.00 frames. , ppl: 10.421979484031109] tot_loss[loss=2.266, over 4826579.41 frames. , ppl: 9.643405648597579], batch size: 70
+2022-12-14 01:46:59,748 INFO [train.py:421] (5/8) Epoch 11, batch 4600, loss[loss=2.347, over 1680.00 frames. , ppl: 10.450872840944765] tot_loss[loss=2.265, over 4909965.61 frames. , ppl: 9.633191575173369], batch size: 70
+2022-12-14 01:48:42,457 INFO [train.py:421] (5/8) Epoch 11, batch 4800, loss[loss=2.541, over 980.00 frames. , ppl: 12.69839371355981] tot_loss[loss=2.264, over 4992742.07 frames. , ppl: 9.619969551728424], batch size: 70
+2022-12-14 01:50:23,265 INFO [train.py:421] (5/8) Epoch 11, batch 5000, loss[loss=2.393, over 1820.00 frames. , ppl: 10.944188480705114] tot_loss[loss=2.264, over 5032522.94 frames. , ppl: 9.618591594869008], batch size: 70
+2022-12-14 01:50:23,265 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 01:50:24,027 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.615150780970064
+2022-12-14 01:52:05,450 INFO [train.py:421] (5/8) Epoch 11, batch 5200, loss[loss=2.376, over 1610.00 frames. , ppl: 10.76052521445849] tot_loss[loss=2.265, over 5029606.16 frames. , ppl: 9.633191064844812], batch size: 70
+2022-12-14 01:53:46,206 INFO [train.py:421] (5/8) Epoch 11, batch 5400, loss[loss=2.365, over 1400.00 frames. , ppl: 10.63911738464501] tot_loss[loss=2.265, over 5057113.42 frames. , ppl: 9.634067989930745], batch size: 70
+2022-12-14 01:55:25,272 INFO [train.py:421] (5/8) Epoch 11, batch 5600, loss[loss=2.187, over 4830.00 frames. , ppl: 8.910762376989233] tot_loss[loss=2.265, over 5106954.57 frames. , ppl: 9.631029359342607], batch size: 70
+2022-12-14 01:57:09,546 INFO [train.py:421] (5/8) Epoch 11, batch 5800, loss[loss=2.61, over 1120.00 frames. , ppl: 13.593419145511373] tot_loss[loss=2.265, over 5176071.37 frames. , ppl: 9.626630159325911], batch size: 70
+2022-12-14 01:58:49,576 INFO [train.py:421] (5/8) Epoch 11, batch 6000, loss[loss=2.278, over 1750.00 frames. , ppl: 9.76047667456334] tot_loss[loss=2.263, over 5220399.79 frames. , ppl: 9.616100284126821], batch size: 70
+2022-12-14 01:58:49,577 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 01:58:50,322 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.606113986457235
+2022-12-14 02:00:27,433 INFO [train.py:421] (5/8) Epoch 11, batch 6200, loss[loss=2.609, over 840.00 frames. , ppl: 13.58137343517255] tot_loss[loss=2.263, over 5241653.85 frames. , ppl: 9.616480363283692], batch size: 70
+2022-12-14 02:02:09,158 INFO [train.py:421] (5/8) Epoch 11, batch 6400, loss[loss=2.58, over 1190.00 frames. , ppl: 13.191374551344975] tot_loss[loss=2.263, over 5263369.68 frames. , ppl: 9.616242899460008], batch size: 70
+2022-12-14 02:03:48,306 INFO [train.py:421] (5/8) Epoch 11, batch 6600, loss[loss=2.207, over 3570.00 frames. , ppl: 9.089302133243205] tot_loss[loss=2.261, over 5357405.58 frames. , ppl: 9.595716845700439], batch size: 70
+2022-12-14 02:05:25,883 INFO [train.py:421] (5/8) Epoch 11, batch 6800, loss[loss=2.253, over 2870.00 frames. , ppl: 9.514925151650623] tot_loss[loss=2.261, over 5381086.34 frames. , ppl: 9.59690804559066], batch size: 70
+2022-12-14 02:07:09,115 INFO [train.py:421] (5/8) Epoch 11, batch 7000, loss[loss=2.347, over 1120.00 frames. , ppl: 10.455533373084293] tot_loss[loss=2.26, over 5412318.69 frames. , ppl: 9.587089528303649], batch size: 70
+2022-12-14 02:07:09,116 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 02:07:09,879 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.616433092260849
+2022-12-14 02:08:51,396 INFO [train.py:421] (5/8) Epoch 11, batch 7200, loss[loss=2.204, over 5600.00 frames. , ppl: 9.065267336363414] tot_loss[loss=2.26, over 5446974.00 frames. , ppl: 9.581029314188461], batch size: 70
+2022-12-14 02:10:28,062 INFO [train.py:421] (5/8) Epoch 11, batch 7400, loss[loss=2.258, over 1960.00 frames. , ppl: 9.560603179058198] tot_loss[loss=2.261, over 5410931.68 frames. , ppl: 9.594805866580247], batch size: 70
+2022-12-14 02:12:08,227 INFO [train.py:421] (5/8) Epoch 11, batch 7600, loss[loss=2.274, over 1820.00 frames. , ppl: 9.714706400258017] tot_loss[loss=2.262, over 5398650.20 frames. , ppl: 9.603014014924748], batch size: 70
+2022-12-14 02:13:46,076 INFO [train.py:421] (5/8) Epoch 11, batch 7800, loss[loss=2.361, over 1050.00 frames. , ppl: 10.604208281919115] tot_loss[loss=2.262, over 5415310.24 frames. , ppl: 9.599260128564863], batch size: 70
+2022-12-14 02:15:21,912 INFO [train.py:421] (5/8) Epoch 11, batch 8000, loss[loss=2.204, over 3780.00 frames. , ppl: 9.063028650682776] tot_loss[loss=2.262, over 5404729.21 frames. , ppl: 9.60632745131748], batch size: 70
+2022-12-14 02:15:21,913 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 02:15:22,633 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.595916733986652
+2022-12-14 02:17:02,600 INFO [train.py:421] (5/8) Epoch 11, batch 8200, loss[loss=2.565, over 770.00 frames. , ppl: 12.995738381517635] tot_loss[loss=2.264, over 5386230.10 frames. , ppl: 9.617855777412183], batch size: 70
+2022-12-14 02:18:44,666 INFO [train.py:421] (5/8) Epoch 11, batch 8400, loss[loss=2.236, over 3570.00 frames. , ppl: 9.353046441725848] tot_loss[loss=2.265, over 5347963.61 frames. , ppl: 9.626877385991806], batch size: 70
+2022-12-14 02:20:24,732 INFO [train.py:421] (5/8) Epoch 11, batch 8600, loss[loss=2.176, over 10850.00 frames. , ppl: 8.81456872898085] tot_loss[loss=2.264, over 5365676.63 frames. , ppl: 9.618686913467865], batch size: 70
+2022-12-14 02:22:03,174 INFO [train.py:421] (5/8) Epoch 11, batch 8800, loss[loss=2.202, over 2100.00 frames. , ppl: 9.045660647433488] tot_loss[loss=2.264, over 5357514.29 frames. , ppl: 9.625802696746371], batch size: 70
+2022-12-14 02:23:44,977 INFO [train.py:421] (5/8) Epoch 11, batch 9000, loss[loss=2.226, over 4060.00 frames. , ppl: 9.266751187216059] tot_loss[loss=2.264, over 5382726.07 frames. , ppl: 9.621966296360394], batch size: 70
+2022-12-14 02:23:44,978 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 02:23:45,723 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.615415483833607
+2022-12-14 02:25:22,448 INFO [train.py:421] (5/8) Epoch 11, batch 9200, loss[loss=2.301, over 1470.00 frames. , ppl: 9.979670400495129] tot_loss[loss=2.265, over 5330442.95 frames. , ppl: 9.634539834534358], batch size: 70
+2022-12-14 02:27:01,801 INFO [train.py:421] (5/8) Epoch 11, batch 9400, loss[loss=2.235, over 2800.00 frames. , ppl: 9.347267495610096] tot_loss[loss=2.267, over 5294078.40 frames. , ppl: 9.654423479635069], batch size: 70
+2022-12-14 02:28:45,756 INFO [train.py:421] (5/8) Epoch 11, batch 9600, loss[loss=2.962, over 700.00 frames. , ppl: 19.344482303969126] tot_loss[loss=2.266, over 5339534.34 frames. , ppl: 9.643044473559673], batch size: 70
+2022-12-14 02:30:24,400 INFO [train.py:421] (5/8) Epoch 11, batch 9800, loss[loss=2.273, over 3360.00 frames. , ppl: 9.708088376322916] tot_loss[loss=2.266, over 5366125.62 frames. , ppl: 9.640452742723237], batch size: 70
+2022-12-14 02:32:06,052 INFO [train.py:421] (5/8) Epoch 11, batch 10000, loss[loss=3.579, over 420.00 frames. , ppl: 35.824047463579554] tot_loss[loss=2.266, over 5383009.79 frames. , ppl: 9.637589021818583], batch size: 70
+2022-12-14 02:32:06,052 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 02:32:06,813 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.607311196929732
+2022-12-14 02:33:49,149 INFO [train.py:421] (5/8) Epoch 11, batch 10200, loss[loss=2.478, over 1260.00 frames. , ppl: 11.921380734297529] tot_loss[loss=2.265, over 5396709.81 frames. , ppl: 9.631032908212553], batch size: 70
+2022-12-14 02:35:28,712 INFO [train.py:421] (5/8) Epoch 11, batch 10400, loss[loss=2.539, over 840.00 frames. , ppl: 12.672583975922274] tot_loss[loss=2.265, over 5413234.19 frames. , ppl: 9.62891815540498], batch size: 70
+2022-12-14 02:37:09,002 INFO [train.py:421] (5/8) Epoch 11, batch 10600, loss[loss=2.364, over 1330.00 frames. , ppl: 10.631788877393566] tot_loss[loss=2.265, over 5416847.73 frames. , ppl: 9.630721803927015], batch size: 70
+2022-12-14 02:38:52,736 INFO [train.py:421] (5/8) Epoch 11, batch 10800, loss[loss=2.211, over 4550.00 frames. , ppl: 9.125339614088109] tot_loss[loss=2.264, over 5463549.31 frames. , ppl: 9.617339362060823], batch size: 70
+2022-12-14 02:40:34,254 INFO [train.py:421] (5/8) Epoch 11, batch 11000, loss[loss=2.313, over 1120.00 frames. , ppl: 10.107469556116207] tot_loss[loss=2.263, over 5489123.52 frames. , ppl: 9.608388800666626], batch size: 70
+2022-12-14 02:40:34,254 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 02:40:35,016 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.604206151775301
+2022-12-14 02:42:14,871 INFO [train.py:421] (5/8) Epoch 11, batch 11200, loss[loss=2.346, over 1680.00 frames. , ppl: 10.447740150640149] tot_loss[loss=2.263, over 5481839.49 frames. , ppl: 9.614512754216474], batch size: 70
+2022-12-14 02:43:54,215 INFO [train.py:421] (5/8) Epoch 11, batch 11400, loss[loss=2.35, over 2240.00 frames. , ppl: 10.488756436768517] tot_loss[loss=2.264, over 5466336.43 frames. , ppl: 9.620503535409656], batch size: 70
+2022-12-14 02:45:34,743 INFO [train.py:421] (5/8) Epoch 11, batch 11600, loss[loss=2.217, over 3220.00 frames. , ppl: 9.17800272402252] tot_loss[loss=2.264, over 5459107.66 frames. , ppl: 9.623398682996129], batch size: 70
+2022-12-14 02:47:16,923 INFO [train.py:421] (5/8) Epoch 11, batch 11800, loss[loss=2.417, over 1120.00 frames. , ppl: 11.207109735571022] tot_loss[loss=2.266, over 5412840.36 frames. , ppl: 9.636238659067738], batch size: 70
+2022-12-14 02:48:54,574 INFO [train.py:421] (5/8) Epoch 11, batch 12000, loss[loss=2.196, over 7910.00 frames. , ppl: 8.988103722435973] tot_loss[loss=2.266, over 5403657.09 frames. , ppl: 9.638155018470735], batch size: 70
+2022-12-14 02:48:54,574 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 02:48:55,303 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.264, over 211138.00 frames. , ppl: 9.621384617486425
+2022-12-14 02:50:34,274 INFO [train.py:421] (5/8) Epoch 11, batch 12200, loss[loss=2.354, over 1190.00 frames. , ppl: 10.530378332680476] tot_loss[loss=2.266, over 5386884.04 frames. , ppl: 9.641708381401912], batch size: 70
+2022-12-14 02:52:10,860 INFO [train.py:421] (5/8) Epoch 11, batch 12400, loss[loss=2.126, over 4410.00 frames. , ppl: 8.383721399181693] tot_loss[loss=2.266, over 5360328.14 frames. , ppl: 9.644887537255316], batch size: 70
+2022-12-14 02:53:50,842 INFO [train.py:421] (5/8) Epoch 11, batch 12600, loss[loss=2.269, over 2170.00 frames. , ppl: 9.66862960936253] tot_loss[loss=2.267, over 5338798.30 frames. , ppl: 9.654934936600371], batch size: 70
+2022-12-14 02:55:29,991 INFO [train.py:421] (5/8) Epoch 11, batch 12800, loss[loss=2.085, over 5880.00 frames. , ppl: 8.047540444377136] tot_loss[loss=2.268, over 5318189.34 frames. , ppl: 9.656218330480613], batch size: 70
+2022-12-14 02:57:14,314 INFO [train.py:421] (5/8) Epoch 11, batch 13000, loss[loss=2.348, over 1960.00 frames. , ppl: 10.469042329158539] tot_loss[loss=2.267, over 5352587.31 frames. , ppl: 9.649160801370034], batch size: 70
+2022-12-14 02:57:14,315 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 02:57:15,076 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.593847627631558
+2022-12-14 02:58:53,991 INFO [train.py:421] (5/8) Epoch 11, batch 13200, loss[loss=4.1, over 350.00 frames. , ppl: 60.317647401956066] tot_loss[loss=2.268, over 5344673.83 frames. , ppl: 9.658536763544088], batch size: 70
+2022-12-14 03:00:37,503 INFO [train.py:421] (5/8) Epoch 11, batch 13400, loss[loss=3.004, over 560.00 frames. , ppl: 20.168334381065048] tot_loss[loss=2.267, over 5364712.82 frames. , ppl: 9.646067665314627], batch size: 70
+2022-12-14 03:02:16,670 INFO [train.py:421] (5/8) Epoch 11, batch 13600, loss[loss=2.552, over 1050.00 frames. , ppl: 12.83601548109646] tot_loss[loss=2.266, over 5387850.88 frames. , ppl: 9.637939470319498], batch size: 70
+2022-12-14 03:03:48,771 INFO [train.py:421] (5/8) Epoch 11, batch 13800, loss[loss=2.493, over 910.00 frames. , ppl: 12.092908093745509] tot_loss[loss=2.266, over 5360185.39 frames. , ppl: 9.642142718384926], batch size: 70
+2022-12-14 03:05:29,080 INFO [train.py:421] (5/8) Epoch 11, batch 14000, loss[loss=2.335, over 1680.00 frames. , ppl: 10.32768758761566] tot_loss[loss=2.266, over 5365121.64 frames. , ppl: 9.640387283186303], batch size: 70
+2022-12-14 03:05:29,080 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 03:05:29,843 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.264, over 211138.00 frames. , ppl: 9.617450808382541
+2022-12-14 03:07:07,756 INFO [train.py:421] (5/8) Epoch 11, batch 14200, loss[loss=2.248, over 1680.00 frames. , ppl: 9.46453877807461] tot_loss[loss=2.267, over 5350405.12 frames. , ppl: 9.650410969170824], batch size: 70
+2022-12-14 03:08:46,879 INFO [train.py:421] (5/8) Epoch 11, batch 14400, loss[loss=2.27, over 2100.00 frames. , ppl: 9.674648703795649] tot_loss[loss=2.266, over 5380537.65 frames. , ppl: 9.642915227927494], batch size: 70
+2022-12-14 03:10:30,194 INFO [train.py:421] (5/8) Epoch 11, batch 14600, loss[loss=2.227, over 2520.00 frames. , ppl: 9.273136745030138] tot_loss[loss=2.266, over 5423294.03 frames. , ppl: 9.637432278113602], batch size: 70
+2022-12-14 03:12:09,169 INFO [train.py:421] (5/8) Epoch 11, batch 14800, loss[loss=2.194, over 2380.00 frames. , ppl: 8.970667464259312] tot_loss[loss=2.266, over 5421417.19 frames. , ppl: 9.643703019770975], batch size: 70
+2022-12-14 03:13:48,746 INFO [train.py:421] (5/8) Epoch 11, batch 15000, loss[loss=2.234, over 4270.00 frames. , ppl: 9.338363042143609] tot_loss[loss=2.267, over 5393181.04 frames. , ppl: 9.649299345977958], batch size: 70
+2022-12-14 03:13:48,746 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 03:13:49,508 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.603182730483441
+2022-12-14 03:15:28,744 INFO [train.py:421] (5/8) Epoch 11, batch 15200, loss[loss=2.302, over 1610.00 frames. , ppl: 9.990303449139226] tot_loss[loss=2.266, over 5393202.22 frames. , ppl: 9.645463308605542], batch size: 70
+2022-12-14 03:17:08,551 INFO [train.py:421] (5/8) Epoch 11, batch 15400, loss[loss=3.016, over 560.00 frames. , ppl: 20.40584102109923] tot_loss[loss=2.265, over 5423228.50 frames. , ppl: 9.632971773286116], batch size: 70
+2022-12-14 03:18:44,459 INFO [train.py:421] (5/8) Epoch 11, batch 15600, loss[loss=2.449, over 1120.00 frames. , ppl: 11.575412941451377] tot_loss[loss=2.264, over 5439026.18 frames. , ppl: 9.625856042936766], batch size: 70
+2022-12-14 03:20:25,762 INFO [train.py:421] (5/8) Epoch 11, batch 15800, loss[loss=2.291, over 2450.00 frames. , ppl: 9.880063839376799] tot_loss[loss=2.264, over 5448601.00 frames. , ppl: 9.626173561797264], batch size: 70
+2022-12-14 03:22:04,991 INFO [train.py:421] (5/8) Epoch 11, batch 16000, loss[loss=2.213, over 3150.00 frames. , ppl: 9.146854623949409] tot_loss[loss=2.266, over 5403842.92 frames. , ppl: 9.63942444357825], batch size: 70
+2022-12-14 03:22:04,991 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 03:22:05,752 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.608784453983127
+2022-12-14 03:23:48,841 INFO [train.py:421] (5/8) Epoch 11, batch 16200, loss[loss=2.199, over 6580.00 frames. , ppl: 9.01201694149998] tot_loss[loss=2.265, over 5451952.17 frames. , ppl: 9.63064555201106], batch size: 70
+2022-12-14 03:25:29,649 INFO [train.py:421] (5/8) Epoch 11, batch 16400, loss[loss=2.205, over 9240.00 frames. , ppl: 9.071146198051078] tot_loss[loss=2.264, over 5473509.82 frames. , ppl: 9.621769302280914], batch size: 70
+2022-12-14 03:27:08,106 INFO [train.py:421] (5/8) Epoch 11, batch 16600, loss[loss=2.323, over 1750.00 frames. , ppl: 10.211163522795957] tot_loss[loss=2.264, over 5459757.98 frames. , ppl: 9.62495815788432], batch size: 70
+2022-12-14 03:28:48,330 INFO [train.py:421] (5/8) Epoch 11, batch 16800, loss[loss=2.281, over 3010.00 frames. , ppl: 9.784406500282923] tot_loss[loss=2.265, over 5438816.05 frames. , ppl: 9.628017916759633], batch size: 70
+2022-12-14 03:30:30,498 INFO [train.py:421] (5/8) Epoch 11, batch 17000, loss[loss=2.17, over 1890.00 frames. , ppl: 8.762390901501613] tot_loss[loss=2.265, over 5423404.34 frames. , ppl: 9.627650132522419], batch size: 70
+2022-12-14 03:30:30,499 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 03:30:31,251 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.599325980992132
+2022-12-14 03:32:09,742 INFO [train.py:421] (5/8) Epoch 11, batch 17200, loss[loss=2.144, over 10430.00 frames. , ppl: 8.532487142860981] tot_loss[loss=2.266, over 5398294.19 frames. , ppl: 9.637621190539242], batch size: 70
+2022-12-14 03:33:49,145 INFO [train.py:421] (5/8) Epoch 11, batch 17400, loss[loss=2.156, over 5740.00 frames. , ppl: 8.635153699646155] tot_loss[loss=2.266, over 5403979.11 frames. , ppl: 9.637497146344984], batch size: 70
+2022-12-14 03:35:30,988 INFO [train.py:421] (5/8) Epoch 11, batch 17600, loss[loss=2.448, over 1050.00 frames. , ppl: 11.560371597181566] tot_loss[loss=2.267, over 5399954.21 frames. , ppl: 9.649251026482393], batch size: 70
+2022-12-14 03:37:07,977 INFO [train.py:421] (5/8) Epoch 11, batch 17800, loss[loss=2.187, over 6300.00 frames. , ppl: 8.907331395824349] tot_loss[loss=2.267, over 5415333.75 frames. , ppl: 9.646113030098482], batch size: 70
+2022-12-14 03:38:46,523 INFO [train.py:421] (5/8) Epoch 11, batch 18000, loss[loss=2.291, over 1960.00 frames. , ppl: 9.887826545653759] tot_loss[loss=2.267, over 5390591.36 frames. , ppl: 9.65196318150458], batch size: 70
+2022-12-14 03:38:46,523 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 03:38:47,290 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.611424367463018
+2022-12-14 03:40:27,911 INFO [train.py:421] (5/8) Epoch 11, batch 18200, loss[loss=2.215, over 3290.00 frames. , ppl: 9.164546700422296] tot_loss[loss=2.268, over 5377719.58 frames. , ppl: 9.657373994187477], batch size: 70
+2022-12-14 03:42:05,025 INFO [train.py:421] (5/8) Epoch 11, batch 18400, loss[loss=2.498, over 1050.00 frames. , ppl: 12.16024996958119] tot_loss[loss=2.269, over 5379142.55 frames. , ppl: 9.665667857710742], batch size: 70
+2022-12-14 03:43:44,909 INFO [train.py:421] (5/8) Epoch 11, batch 18600, loss[loss=2.524, over 980.00 frames. , ppl: 12.479051260842285] tot_loss[loss=2.267, over 5433052.92 frames. , ppl: 9.646089111878887], batch size: 70
+2022-12-14 03:45:24,640 INFO [train.py:421] (5/8) Epoch 11, batch 18800, loss[loss=2.638, over 770.00 frames. , ppl: 13.978500221265437] tot_loss[loss=2.267, over 5441555.64 frames. , ppl: 9.647922489446401], batch size: 70
+2022-12-14 03:47:03,770 INFO [train.py:421] (5/8) Epoch 11, batch 19000, loss[loss=2.332, over 1820.00 frames. , ppl: 10.293795746430236] tot_loss[loss=2.267, over 5444866.80 frames. , ppl: 9.648081126225776], batch size: 70
+2022-12-14 03:47:03,770 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 03:47:04,530 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.604402319978165
+2022-12-14 03:48:43,945 INFO [train.py:421] (5/8) Epoch 11, batch 19200, loss[loss=2.383, over 1470.00 frames. , ppl: 10.842765536155733] tot_loss[loss=2.266, over 5461610.58 frames. , ppl: 9.643593759257804], batch size: 70
+2022-12-14 03:50:21,980 INFO [train.py:421] (5/8) Epoch 11, batch 19400, loss[loss=2.209, over 3850.00 frames. , ppl: 9.103322051124417] tot_loss[loss=2.267, over 5427211.43 frames. , ppl: 9.652224706162867], batch size: 70
+2022-12-14 03:52:06,008 INFO [train.py:421] (5/8) Epoch 11, batch 19600, loss[loss=2.164, over 6090.00 frames. , ppl: 8.706233370781723] tot_loss[loss=2.266, over 5458964.93 frames. , ppl: 9.63896114036436], batch size: 70
+2022-12-14 03:53:44,134 INFO [train.py:421] (5/8) Epoch 11, batch 19800, loss[loss=2.181, over 10430.00 frames. , ppl: 8.85553347974669] tot_loss[loss=2.267, over 5424651.59 frames. , ppl: 9.650276776771854], batch size: 70
+2022-12-14 03:55:25,012 INFO [train.py:421] (5/8) Epoch 11, batch 20000, loss[loss=2.647, over 910.00 frames. , ppl: 14.118229572229774] tot_loss[loss=2.267, over 5416248.50 frames. , ppl: 9.653586815805715], batch size: 70
+2022-12-14 03:55:25,013 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 03:55:25,775 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.61202043965363
+2022-12-14 03:57:05,992 INFO [train.py:421] (5/8) Epoch 11, batch 20200, loss[loss=2.266, over 2030.00 frames. , ppl: 9.642601846464675] tot_loss[loss=2.267, over 5403417.41 frames. , ppl: 9.647601797667733], batch size: 70
+2022-12-14 03:58:46,599 INFO [train.py:421] (5/8) Epoch 11, batch 20400, loss[loss=2.198, over 3850.00 frames. , ppl: 9.008271969945405] tot_loss[loss=2.265, over 5458267.37 frames. , ppl: 9.628120868043442], batch size: 70
+2022-12-14 04:00:28,898 INFO [train.py:421] (5/8) Epoch 11, batch 20600, loss[loss=2.135, over 7070.00 frames. , ppl: 8.460243408390012] tot_loss[loss=2.263, over 5497110.28 frames. , ppl: 9.615544885959887], batch size: 70
+2022-12-14 04:02:14,438 INFO [train.py:421] (5/8) Epoch 11, batch 20800, loss[loss=2.499, over 1820.00 frames. , ppl: 12.175517930220064] tot_loss[loss=2.264, over 5484187.46 frames. , ppl: 9.625345862432171], batch size: 70
+2022-12-14 04:03:51,824 INFO [train.py:421] (5/8) Epoch 11, batch 21000, loss[loss=2.179, over 6860.00 frames. , ppl: 8.83522711059473] tot_loss[loss=2.264, over 5501592.80 frames. , ppl: 9.624175561828713], batch size: 70
+2022-12-14 04:03:51,825 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 04:03:52,554 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.59960729794989
+2022-12-14 04:05:34,812 INFO [train.py:421] (5/8) Epoch 11, batch 21200, loss[loss=2.364, over 1330.00 frames. , ppl: 10.63625121172031] tot_loss[loss=2.263, over 5540760.04 frames. , ppl: 9.615234835413482], batch size: 70
+2022-12-14 04:07:12,357 INFO [train.py:421] (5/8) Epoch 11, batch 21400, loss[loss=2.175, over 5110.00 frames. , ppl: 8.8007385756694] tot_loss[loss=2.263, over 5545340.70 frames. , ppl: 9.607432688010556], batch size: 70
+2022-12-14 04:08:53,894 INFO [train.py:421] (5/8) Epoch 11, batch 21600, loss[loss=2.14, over 6160.00 frames. , ppl: 8.501458220968141] tot_loss[loss=2.262, over 5563920.02 frames. , ppl: 9.602728599012837], batch size: 70
+2022-12-14 04:10:28,568 INFO [train.py:421] (5/8) Epoch 11, batch 21800, loss[loss=2.329, over 1820.00 frames. , ppl: 10.272638154200962] tot_loss[loss=2.263, over 5489575.27 frames. , ppl: 9.615782485459054], batch size: 70
+2022-12-14 04:12:08,370 INFO [train.py:421] (5/8) Epoch 11, batch 22000, loss[loss=2.233, over 3430.00 frames. , ppl: 9.326358816182594] tot_loss[loss=2.265, over 5440731.01 frames. , ppl: 9.632290374830742], batch size: 70
+2022-12-14 04:12:08,370 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 04:12:09,131 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.59981047643547
+2022-12-14 04:13:46,451 INFO [train.py:421] (5/8) Epoch 11, batch 22200, loss[loss=2.466, over 1470.00 frames. , ppl: 11.771139643598648] tot_loss[loss=2.265, over 5463137.00 frames. , ppl: 9.62775856049884], batch size: 70
+2022-12-14 04:15:26,607 INFO [train.py:421] (5/8) Epoch 11, batch 22400, loss[loss=4.833, over 280.00 frames. , ppl: 125.57154991786354] tot_loss[loss=2.264, over 5483867.90 frames. , ppl: 9.62316464517015], batch size: 70
+2022-12-14 04:17:04,240 INFO [train.py:421] (5/8) Epoch 11, batch 22600, loss[loss=2.24, over 3710.00 frames. , ppl: 9.389723040917286] tot_loss[loss=2.263, over 5494319.98 frames. , ppl: 9.61455311709182], batch size: 70
+2022-12-14 04:18:46,595 INFO [train.py:421] (5/8) Epoch 11, batch 22800, loss[loss=2.254, over 1960.00 frames. , ppl: 9.530473766579945] tot_loss[loss=2.264, over 5493784.15 frames. , ppl: 9.619017317520552], batch size: 70
+2022-12-14 04:20:26,417 INFO [train.py:421] (5/8) Epoch 11, batch 23000, loss[loss=2.52, over 980.00 frames. , ppl: 12.427758846468954] tot_loss[loss=2.265, over 5469726.83 frames. , ppl: 9.634950304590967], batch size: 70
+2022-12-14 04:20:26,418 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 04:20:27,166 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.589605722035854
+2022-12-14 04:22:05,159 INFO [train.py:421] (5/8) Epoch 11, batch 23200, loss[loss=2.621, over 840.00 frames. , ppl: 13.74811377330994] tot_loss[loss=2.267, over 5443246.08 frames. , ppl: 9.648838727146181], batch size: 70
+2022-12-14 04:23:43,583 INFO [train.py:421] (5/8) Epoch 11, batch 23400, loss[loss=2.182, over 7236.00 frames. , ppl: 8.861641268082742] tot_loss[loss=2.267, over 5423649.07 frames. , ppl: 9.650214235667772], batch size: 36
+2022-12-14 04:25:27,480 INFO [train.py:421] (5/8) Epoch 11, batch 23600, loss[loss=2.213, over 3290.00 frames. , ppl: 9.1450141038443] tot_loss[loss=2.267, over 5430775.23 frames. , ppl: 9.650023401829337], batch size: 70
+2022-12-14 04:27:09,144 INFO [train.py:421] (5/8) Epoch 11, batch 23800, loss[loss=2.135, over 3990.00 frames. , ppl: 8.456950050848524] tot_loss[loss=2.266, over 5477251.88 frames. , ppl: 9.640577851843341], batch size: 70
+2022-12-14 04:28:49,656 INFO [train.py:421] (5/8) Epoch 11, batch 24000, loss[loss=2.198, over 4410.00 frames. , ppl: 9.003620715124356] tot_loss[loss=2.265, over 5493338.01 frames. , ppl: 9.63408138323631], batch size: 70
+2022-12-14 04:28:49,656 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 04:28:50,387 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.589086260144834
+2022-12-14 04:30:27,504 INFO [train.py:421] (5/8) Epoch 11, batch 24200, loss[loss=2.193, over 4550.00 frames. , ppl: 8.963390946155862] tot_loss[loss=2.265, over 5496956.90 frames. , ppl: 9.63451375101307], batch size: 70
+2022-12-14 04:32:06,806 INFO [train.py:421] (5/8) Epoch 11, batch 24400, loss[loss=2.279, over 3850.00 frames. , ppl: 9.766275412472702] tot_loss[loss=2.265, over 5484642.46 frames. , ppl: 9.635032158238769], batch size: 70
+2022-12-14 04:33:49,455 INFO [train.py:421] (5/8) Epoch 11, batch 24600, loss[loss=2.241, over 2310.00 frames. , ppl: 9.404555868306185] tot_loss[loss=2.266, over 5463640.25 frames. , ppl: 9.639562408497527], batch size: 70
+2022-12-14 04:35:31,673 INFO [train.py:421] (5/8) Epoch 11, batch 24800, loss[loss=2.188, over 4340.00 frames. , ppl: 8.91727611241416] tot_loss[loss=2.266, over 5458435.73 frames. , ppl: 9.644815768981868], batch size: 70
+2022-12-14 04:37:15,799 INFO [train.py:421] (5/8) Epoch 11, batch 25000, loss[loss=2.542, over 1050.00 frames. , ppl: 12.702236004258307] tot_loss[loss=2.265, over 5497955.30 frames. , ppl: 9.63258200061579], batch size: 70
+2022-12-14 04:37:15,800 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 04:37:16,564 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.606207824169385
+2022-12-14 04:38:53,653 INFO [train.py:421] (5/8) Epoch 11, batch 25200, loss[loss=2.382, over 1540.00 frames. , ppl: 10.827792666012552] tot_loss[loss=2.264, over 5539109.76 frames. , ppl: 9.620116232684824], batch size: 70
+2022-12-14 04:40:34,697 INFO [train.py:421] (5/8) Epoch 11, batch 25400, loss[loss=2.171, over 4690.00 frames. , ppl: 8.762725437929515] tot_loss[loss=2.263, over 5553445.43 frames. , ppl: 9.614711834655195], batch size: 70
+2022-12-14 04:42:13,996 INFO [train.py:421] (5/8) Epoch 11, batch 25600, loss[loss=2.548, over 1050.00 frames. , ppl: 12.784111681206044] tot_loss[loss=2.264, over 5548630.80 frames. , ppl: 9.6174459261298], batch size: 70
+2022-12-14 04:43:56,157 INFO [train.py:421] (5/8) Epoch 11, batch 25800, loss[loss=2.221, over 3430.00 frames. , ppl: 9.22031245934101] tot_loss[loss=2.263, over 5550946.48 frames. , ppl: 9.61084312842711], batch size: 70
+2022-12-14 04:45:33,980 INFO [train.py:421] (5/8) Epoch 11, batch 26000, loss[loss=2.361, over 1190.00 frames. , ppl: 10.606804624583939] tot_loss[loss=2.263, over 5512366.01 frames. , ppl: 9.615957359841229], batch size: 70
+2022-12-14 04:45:33,981 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 04:45:34,741 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.612918174211162
+2022-12-14 04:47:14,142 INFO [train.py:421] (5/8) Epoch 11, batch 26200, loss[loss=2.427, over 1470.00 frames. , ppl: 11.323231103774356] tot_loss[loss=2.263, over 5534098.55 frames. , ppl: 9.611975821448338], batch size: 70
+2022-12-14 04:48:55,148 INFO [train.py:421] (5/8) Epoch 11, batch 26400, loss[loss=2.286, over 2170.00 frames. , ppl: 9.836533236402348] tot_loss[loss=2.263, over 5544780.00 frames. , ppl: 9.610672894112835], batch size: 70
+2022-12-14 04:50:33,171 INFO [train.py:421] (5/8) Epoch 11, batch 26600, loss[loss=2.267, over 3430.00 frames. , ppl: 9.654763573527688] tot_loss[loss=2.263, over 5538458.31 frames. , ppl: 9.614735759148243], batch size: 70
+2022-12-14 04:52:12,404 INFO [train.py:421] (5/8) Epoch 11, batch 26800, loss[loss=2.834, over 630.00 frames. , ppl: 17.017510681933864] tot_loss[loss=2.264, over 5524159.49 frames. , ppl: 9.617934779126395], batch size: 70
+2022-12-14 04:53:51,698 INFO [train.py:421] (5/8) Epoch 11, batch 27000, loss[loss=2.428, over 1890.00 frames. , ppl: 11.336977327075259] tot_loss[loss=2.264, over 5500444.70 frames. , ppl: 9.618610299650546], batch size: 70
+2022-12-14 04:53:51,699 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 04:53:52,445 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.594008084573185
+2022-12-14 04:55:35,141 INFO [train.py:421] (5/8) Epoch 11, batch 27200, loss[loss=2.364, over 1610.00 frames. , ppl: 10.630977120261843] tot_loss[loss=2.263, over 5549135.27 frames. , ppl: 9.614017233245226], batch size: 70
+2022-12-14 04:57:14,480 INFO [train.py:421] (5/8) Epoch 11, batch 27400, loss[loss=2.174, over 7980.00 frames. , ppl: 8.796251869613373] tot_loss[loss=2.262, over 5622048.92 frames. , ppl: 9.59835274096138], batch size: 70
+2022-12-14 04:58:50,881 INFO [train.py:421] (5/8) Epoch 11, batch 27600, loss[loss=2.175, over 4480.00 frames. , ppl: 8.805699007185698] tot_loss[loss=2.262, over 5592963.05 frames. , ppl: 9.60536330646484], batch size: 70
+2022-12-14 05:00:29,827 INFO [train.py:421] (5/8) Epoch 11, batch 27800, loss[loss=2.723, over 770.00 frames. , ppl: 15.222928334819363] tot_loss[loss=2.262, over 5580989.86 frames. , ppl: 9.606527989891925], batch size: 70
+2022-12-14 05:02:14,476 INFO [train.py:421] (5/8) Epoch 11, batch 28000, loss[loss=2.435, over 1890.00 frames. , ppl: 11.414794770922585] tot_loss[loss=2.263, over 5567283.21 frames. , ppl: 9.611587957214901], batch size: 70
+2022-12-14 05:02:14,476 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 05:02:15,236 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.594497991840212
+2022-12-14 05:04:00,693 INFO [train.py:421] (5/8) Epoch 11, batch 28200, loss[loss=2.302, over 2100.00 frames. , ppl: 9.997991347713619] tot_loss[loss=2.264, over 5549493.22 frames. , ppl: 9.620485763073779], batch size: 70
+2022-12-14 05:05:42,753 INFO [train.py:421] (5/8) Epoch 11, batch 28400, loss[loss=2.297, over 2240.00 frames. , ppl: 9.940201209169857] tot_loss[loss=2.265, over 5487816.41 frames. , ppl: 9.63110415295772], batch size: 70
+2022-12-14 05:07:20,266 INFO [train.py:421] (5/8) Epoch 11, batch 28600, loss[loss=2.184, over 5530.00 frames. , ppl: 8.879556611862727] tot_loss[loss=2.265, over 5489975.79 frames. , ppl: 9.633860931615803], batch size: 70
+2022-12-14 05:09:04,634 INFO [train.py:421] (5/8) Epoch 11, batch 28800, loss[loss=3.11, over 490.00 frames. , ppl: 22.41519257307544] tot_loss[loss=2.266, over 5472757.24 frames. , ppl: 9.638546751893015], batch size: 70
+2022-12-14 05:10:45,354 INFO [train.py:421] (5/8) Epoch 11, batch 29000, loss[loss=2.421, over 1120.00 frames. , ppl: 11.260011826153686] tot_loss[loss=2.266, over 5469030.43 frames. , ppl: 9.6376877978396], batch size: 70
+2022-12-14 05:10:45,355 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 05:10:46,097 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.587935312328966
+2022-12-14 05:12:24,282 INFO [train.py:421] (5/8) Epoch 11, batch 29200, loss[loss=2.156, over 2450.00 frames. , ppl: 8.636788148954995] tot_loss[loss=2.264, over 5500336.76 frames. , ppl: 9.621708218537481], batch size: 70
+2022-12-14 05:14:04,310 INFO [train.py:421] (5/8) Epoch 11, batch 29400, loss[loss=4.889, over 280.00 frames. , ppl: 132.75662215010468] tot_loss[loss=2.265, over 5475510.08 frames. , ppl: 9.63225880462876], batch size: 70
+2022-12-14 05:15:46,582 INFO [train.py:421] (5/8) Epoch 11, batch 29600, loss[loss=2.857, over 560.00 frames. , ppl: 17.412880897231886] tot_loss[loss=2.267, over 5405148.80 frames. , ppl: 9.652451058266013], batch size: 70
+2022-12-14 05:17:28,206 INFO [train.py:421] (5/8) Epoch 11, batch 29800, loss[loss=2.14, over 3290.00 frames. , ppl: 8.501793303797248] tot_loss[loss=2.265, over 5439261.68 frames. , ppl: 9.635548577595918], batch size: 70
+2022-12-14 05:19:07,020 INFO [train.py:421] (5/8) Epoch 11, batch 30000, loss[loss=2.399, over 2170.00 frames. , ppl: 11.008016850321491] tot_loss[loss=2.266, over 5429595.12 frames. , ppl: 9.641349790628107], batch size: 70
+2022-12-14 05:19:07,020 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 05:19:07,779 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.613103137864265
+2022-12-14 05:20:48,039 INFO [train.py:421] (5/8) Epoch 11, batch 30200, loss[loss=2.692, over 910.00 frames. , ppl: 14.764602190775523] tot_loss[loss=2.266, over 5443729.21 frames. , ppl: 9.642229295636499], batch size: 70
+2022-12-14 05:22:30,485 INFO [train.py:421] (5/8) Epoch 11, batch 30400, loss[loss=2.072, over 6020.00 frames. , ppl: 7.940234492060649] tot_loss[loss=2.266, over 5462722.49 frames. , ppl: 9.637211136971061], batch size: 70
+2022-12-14 05:24:12,352 INFO [train.py:421] (5/8) Epoch 11, batch 30600, loss[loss=2.298, over 3360.00 frames. , ppl: 9.952225503899259] tot_loss[loss=2.266, over 5474224.35 frames. , ppl: 9.639454579871275], batch size: 70
+2022-12-14 05:25:49,459 INFO [train.py:421] (5/8) Epoch 11, batch 30800, loss[loss=2.433, over 1540.00 frames. , ppl: 11.39622402175312] tot_loss[loss=2.266, over 5468228.69 frames. , ppl: 9.641810886677822], batch size: 70
+2022-12-14 05:27:28,471 INFO [train.py:421] (5/8) Epoch 11, batch 31000, loss[loss=2.234, over 4970.00 frames. , ppl: 9.335574247351458] tot_loss[loss=2.265, over 5514116.47 frames. , ppl: 9.632848020534416], batch size: 70
+2022-12-14 05:27:28,471 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 05:27:29,214 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.613066144848931
+2022-12-14 05:29:07,430 INFO [train.py:421] (5/8) Epoch 11, batch 31200, loss[loss=2.228, over 4130.00 frames. , ppl: 9.278547864909513] tot_loss[loss=2.266, over 5488807.73 frames. , ppl: 9.640366406882155], batch size: 70
+2022-12-14 05:30:46,537 INFO [train.py:421] (5/8) Epoch 11, batch 31400, loss[loss=2.324, over 2170.00 frames. , ppl: 10.221179515252722] tot_loss[loss=2.266, over 5472543.27 frames. , ppl: 9.644472517149904], batch size: 70
+2022-12-14 05:32:28,444 INFO [train.py:421] (5/8) Epoch 11, batch 31600, loss[loss=2.209, over 2030.00 frames. , ppl: 9.106108105256723] tot_loss[loss=2.266, over 5483814.99 frames. , ppl: 9.645384313788739], batch size: 70
+2022-12-14 05:34:09,689 INFO [train.py:421] (5/8) Epoch 11, batch 31800, loss[loss=2.463, over 910.00 frames. , ppl: 11.738508899886499] tot_loss[loss=2.266, over 5497630.34 frames. , ppl: 9.64176867228167], batch size: 70
+2022-12-14 05:35:44,348 INFO [train.py:421] (5/8) Epoch 11, batch 32000, loss[loss=2.192, over 6860.00 frames. , ppl: 8.950789880605726] tot_loss[loss=2.265, over 5501073.27 frames. , ppl: 9.634752721007326], batch size: 70
+2022-12-14 05:35:44,348 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 05:35:45,109 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.604530257921299
+2022-12-14 05:37:24,334 INFO [train.py:421] (5/8) Epoch 11, batch 32200, loss[loss=2.185, over 3780.00 frames. , ppl: 8.889028928530829] tot_loss[loss=2.264, over 5546811.90 frames. , ppl: 9.61964773789273], batch size: 70
+2022-12-14 05:39:05,741 INFO [train.py:421] (5/8) Epoch 11, batch 32400, loss[loss=2.272, over 2310.00 frames. , ppl: 9.696693521514764] tot_loss[loss=2.264, over 5524511.34 frames. , ppl: 9.621168311624563], batch size: 70
+2022-12-14 05:40:47,673 INFO [train.py:421] (5/8) Epoch 11, batch 32600, loss[loss=2.483, over 1540.00 frames. , ppl: 11.976198880302638] tot_loss[loss=2.263, over 5538259.02 frames. , ppl: 9.611893178045204], batch size: 70
+2022-12-14 05:42:26,787 INFO [train.py:421] (5/8) Epoch 11, batch 32800, loss[loss=2.425, over 1330.00 frames. , ppl: 11.305937492616238] tot_loss[loss=2.263, over 5536211.49 frames. , ppl: 9.61255056153698], batch size: 70
+2022-12-14 05:44:07,639 INFO [train.py:421] (5/8) Epoch 11, batch 33000, loss[loss=2.412, over 1470.00 frames. , ppl: 11.15633576644077] tot_loss[loss=2.263, over 5518572.68 frames. , ppl: 9.61623713007932], batch size: 70
+2022-12-14 05:44:07,640 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 05:44:08,397 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.259, over 211138.00 frames. , ppl: 9.5747582284789
+2022-12-14 05:45:47,690 INFO [train.py:421] (5/8) Epoch 11, batch 33200, loss[loss=2.108, over 7000.00 frames. , ppl: 8.229259295111056] tot_loss[loss=2.263, over 5545245.92 frames. , ppl: 9.608853600527], batch size: 70
+2022-12-14 05:47:26,463 INFO [train.py:421] (5/8) Epoch 11, batch 33400, loss[loss=2.173, over 6440.00 frames. , ppl: 8.782290282347807] tot_loss[loss=2.266, over 5463373.28 frames. , ppl: 9.63782384046163], batch size: 70
+2022-12-14 05:49:07,578 INFO [train.py:421] (5/8) Epoch 11, batch 33600, loss[loss=2.128, over 6300.00 frames. , ppl: 8.39695578512114] tot_loss[loss=2.266, over 5470205.08 frames. , ppl: 9.637427100283714], batch size: 70
+2022-12-14 05:50:49,688 INFO [train.py:421] (5/8) Epoch 11, batch 33800, loss[loss=2.478, over 1540.00 frames. , ppl: 11.921548729324975] tot_loss[loss=2.263, over 5537250.25 frames. , ppl: 9.613064959243392], batch size: 70
+2022-12-14 05:52:29,250 INFO [train.py:421] (5/8) Epoch 11, batch 34000, loss[loss=2.44, over 1120.00 frames. , ppl: 11.477606338203643] tot_loss[loss=2.265, over 5497795.40 frames. , ppl: 9.628282421452528], batch size: 70
+2022-12-14 05:52:29,251 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 05:52:30,011 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.580360386976093
+2022-12-14 05:54:12,838 INFO [train.py:421] (5/8) Epoch 11, batch 34200, loss[loss=2.39, over 1330.00 frames. , ppl: 10.912874531372193] tot_loss[loss=2.265, over 5497717.52 frames. , ppl: 9.628862342960431], batch size: 70
+2022-12-14 05:55:56,807 INFO [train.py:421] (5/8) Epoch 11, batch 34400, loss[loss=2.137, over 6930.00 frames. , ppl: 8.47254726102841] tot_loss[loss=2.265, over 5491823.87 frames. , ppl: 9.633876774756235], batch size: 70
+2022-12-14 05:57:35,021 INFO [train.py:421] (5/8) Epoch 11, batch 34600, loss[loss=2.359, over 2240.00 frames. , ppl: 10.581602801504816] tot_loss[loss=2.266, over 5482643.15 frames. , ppl: 9.643655531351266], batch size: 70
+2022-12-14 05:59:12,269 INFO [train.py:421] (5/8) Epoch 11, batch 34800, loss[loss=2.542, over 1050.00 frames. , ppl: 12.702250771561353] tot_loss[loss=2.265, over 5523683.49 frames. , ppl: 9.62728522183547], batch size: 70
+2022-12-14 06:00:51,154 INFO [train.py:421] (5/8) Epoch 11, batch 35000, loss[loss=2.388, over 1750.00 frames. , ppl: 10.887401616911157] tot_loss[loss=2.266, over 5465892.94 frames. , ppl: 9.643440041639282], batch size: 70
+2022-12-14 06:00:51,154 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 06:00:51,940 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.259, over 211138.00 frames. , ppl: 9.571281208986392
+2022-12-14 06:02:35,354 INFO [train.py:421] (5/8) Epoch 11, batch 35200, loss[loss=2.333, over 1890.00 frames. , ppl: 10.306321774545625] tot_loss[loss=2.266, over 5482837.19 frames. , ppl: 9.64168145115102], batch size: 70
+2022-12-14 06:04:14,611 INFO [train.py:421] (5/8) Epoch 11, batch 35400, loss[loss=2.182, over 4620.00 frames. , ppl: 8.860693042773358] tot_loss[loss=2.266, over 5467236.12 frames. , ppl: 9.643876959924889], batch size: 70
+2022-12-14 06:05:56,234 INFO [train.py:421] (5/8) Epoch 11, batch 35600, loss[loss=2.406, over 1190.00 frames. , ppl: 11.094053059108687] tot_loss[loss=2.266, over 5477358.76 frames. , ppl: 9.641962129987478], batch size: 70
+2022-12-14 06:07:32,112 INFO [train.py:421] (5/8) Epoch 11, batch 35800, loss[loss=2.287, over 3290.00 frames. , ppl: 9.84384791262472] tot_loss[loss=2.267, over 5460906.71 frames. , ppl: 9.645810030268603], batch size: 70
+2022-12-14 06:09:12,202 INFO [train.py:421] (5/8) Epoch 11, batch 36000, loss[loss=2.242, over 3150.00 frames. , ppl: 9.414971379975412] tot_loss[loss=2.267, over 5460490.60 frames. , ppl: 9.650624993822305], batch size: 70
+2022-12-14 06:09:12,203 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 06:09:12,963 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.593320836541546
+2022-12-14 06:10:52,350 INFO [train.py:421] (5/8) Epoch 11, batch 36200, loss[loss=2.494, over 1540.00 frames. , ppl: 12.111386160289868] tot_loss[loss=2.267, over 5480248.64 frames. , ppl: 9.645709561544725], batch size: 70
+2022-12-14 06:12:34,299 INFO [train.py:421] (5/8) Epoch 11, batch 36400, loss[loss=2.301, over 3920.00 frames. , ppl: 9.980878750462683] tot_loss[loss=2.265, over 5517796.18 frames. , ppl: 9.63131205825295], batch size: 70
+2022-12-14 06:14:15,795 INFO [train.py:421] (5/8) Epoch 11, batch 36600, loss[loss=3.619, over 420.00 frames. , ppl: 37.31525425250405] tot_loss[loss=2.266, over 5507686.54 frames. , ppl: 9.63780865095771], batch size: 70
+2022-12-14 06:15:56,234 INFO [train.py:421] (5/8) Epoch 11, batch 36800, loss[loss=3.103, over 560.00 frames. , ppl: 22.260536889122353] tot_loss[loss=2.266, over 5498098.78 frames. , ppl: 9.638450191170653], batch size: 70
+2022-12-14 06:17:38,578 INFO [train.py:421] (5/8) Epoch 11, batch 37000, loss[loss=2.105, over 6090.00 frames. , ppl: 8.206467644891566] tot_loss[loss=2.266, over 5509857.74 frames. , ppl: 9.636574675151286], batch size: 70
+2022-12-14 06:17:38,579 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 06:17:39,307 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.585279151337225
+2022-12-14 06:19:19,620 INFO [train.py:421] (5/8) Epoch 11, batch 37200, loss[loss=2.106, over 5180.00 frames. , ppl: 8.219097374192778] tot_loss[loss=2.265, over 5526205.25 frames. , ppl: 9.626903403778782], batch size: 70
+2022-12-14 06:21:00,622 INFO [train.py:421] (5/8) Epoch 11, batch 37400, loss[loss=2.334, over 2170.00 frames. , ppl: 10.315248391938171] tot_loss[loss=2.266, over 5494792.02 frames. , ppl: 9.639923574694611], batch size: 70
+2022-12-14 06:22:44,801 INFO [train.py:421] (5/8) Epoch 11, batch 37600, loss[loss=2.191, over 5810.00 frames. , ppl: 8.944484985603301] tot_loss[loss=2.267, over 5453720.16 frames. , ppl: 9.652404559025328], batch size: 70
+2022-12-14 06:24:22,292 INFO [train.py:421] (5/8) Epoch 11, batch 37800, loss[loss=2.28, over 2380.00 frames. , ppl: 9.779984095313788] tot_loss[loss=2.268, over 5419262.52 frames. , ppl: 9.661416959707523], batch size: 70
+2022-12-14 06:26:03,855 INFO [train.py:421] (5/8) Epoch 11, batch 38000, loss[loss=2.467, over 1890.00 frames. , ppl: 11.786465538572886] tot_loss[loss=2.268, over 5427673.20 frames. , ppl: 9.658199197239847], batch size: 70
+2022-12-14 06:26:03,855 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 06:26:04,620 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.584332929905047
+2022-12-14 06:27:46,868 INFO [train.py:421] (5/8) Epoch 11, batch 38200, loss[loss=2.308, over 1610.00 frames. , ppl: 10.057482200885946] tot_loss[loss=2.268, over 5412590.43 frames. , ppl: 9.66194697616955], batch size: 70
+2022-12-14 06:29:24,550 INFO [train.py:421] (5/8) Epoch 11, batch 38400, loss[loss=2.512, over 840.00 frames. , ppl: 12.329873879287058] tot_loss[loss=2.268, over 5410901.73 frames. , ppl: 9.659460999191172], batch size: 70
+2022-12-14 06:31:03,586 INFO [train.py:421] (5/8) Epoch 11, batch 38600, loss[loss=2.287, over 2100.00 frames. , ppl: 9.842481082387117] tot_loss[loss=2.268, over 5402050.52 frames. , ppl: 9.661185079394548], batch size: 70
+2022-12-14 06:32:43,274 INFO [train.py:421] (5/8) Epoch 11, batch 38800, loss[loss=2.589, over 910.00 frames. , ppl: 13.313251238514058] tot_loss[loss=2.268, over 5392128.87 frames. , ppl: 9.663556633059969], batch size: 70
+2022-12-14 06:34:25,390 INFO [train.py:421] (5/8) Epoch 11, batch 39000, loss[loss=2.344, over 980.00 frames. , ppl: 10.418326265219704] tot_loss[loss=2.268, over 5413317.91 frames. , ppl: 9.663545028547786], batch size: 70
+2022-12-14 06:34:25,390 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 06:34:26,137 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.58167067609052
+2022-12-14 06:36:07,552 INFO [train.py:421] (5/8) Epoch 11, batch 39200, loss[loss=2.635, over 1050.00 frames. , ppl: 13.940083777507681] tot_loss[loss=2.269, over 5391298.15 frames. , ppl: 9.667132375883007], batch size: 70
+2022-12-14 06:37:47,055 INFO [train.py:421] (5/8) Epoch 11, batch 39400, loss[loss=2.274, over 2520.00 frames. , ppl: 9.715147747749068] tot_loss[loss=2.268, over 5405374.17 frames. , ppl: 9.663513050186209], batch size: 70
+2022-12-14 06:39:24,956 INFO [train.py:421] (5/8) Epoch 11, batch 39600, loss[loss=2.211, over 2660.00 frames. , ppl: 9.124938040070658] tot_loss[loss=2.268, over 5419738.36 frames. , ppl: 9.656356452365351], batch size: 70
+2022-12-14 06:41:07,682 INFO [train.py:421] (5/8) Epoch 11, batch 39800, loss[loss=2.305, over 1610.00 frames. , ppl: 10.021289323936518] tot_loss[loss=2.268, over 5407354.10 frames. , ppl: 9.657342328553119], batch size: 70
+2022-12-14 06:42:42,716 INFO [train.py:421] (5/8) Epoch 11, batch 40000, loss[loss=2.204, over 6930.00 frames. , ppl: 9.057877686908672] tot_loss[loss=2.27, over 5377591.44 frames. , ppl: 9.67602828103409], batch size: 70
+2022-12-14 06:42:42,717 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 06:42:43,479 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.58547067678701
+2022-12-14 06:44:22,144 INFO [train.py:421] (5/8) Epoch 11, batch 40200, loss[loss=2.783, over 630.00 frames. , ppl: 16.166540047727665] tot_loss[loss=2.269, over 5402467.32 frames. , ppl: 9.666022639256482], batch size: 70
+2022-12-14 06:46:06,989 INFO [train.py:421] (5/8) Epoch 11, batch 40400, loss[loss=2.346, over 1470.00 frames. , ppl: 10.446546630372836] tot_loss[loss=2.268, over 5425528.10 frames. , ppl: 9.662189166677837], batch size: 70
+2022-12-14 06:47:50,955 INFO [train.py:421] (5/8) Epoch 11, batch 40600, loss[loss=2.14, over 10920.00 frames. , ppl: 8.499638904295786] tot_loss[loss=2.268, over 5439045.96 frames. , ppl: 9.661201816101903], batch size: 70
+2022-12-14 06:49:30,439 INFO [train.py:421] (5/8) Epoch 11, batch 40800, loss[loss=2.295, over 1960.00 frames. , ppl: 9.923863419543409] tot_loss[loss=2.267, over 5443299.58 frames. , ppl: 9.652364870158847], batch size: 70
+2022-12-14 06:51:11,967 INFO [train.py:421] (5/8) Epoch 11, batch 41000, loss[loss=2.511, over 1050.00 frames. , ppl: 12.311125856633831] tot_loss[loss=2.268, over 5430823.38 frames. , ppl: 9.659688571175518], batch size: 70
+2022-12-14 06:51:11,967 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 06:51:12,725 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.258, over 211138.00 frames. , ppl: 9.567495329829377
+2022-12-14 06:52:55,089 INFO [train.py:421] (5/8) Epoch 11, batch 41200, loss[loss=2.406, over 1470.00 frames. , ppl: 11.085303442376539] tot_loss[loss=2.269, over 5406144.84 frames. , ppl: 9.671347596241649], batch size: 70
+2022-12-14 06:54:38,130 INFO [train.py:421] (5/8) Epoch 11, batch 41400, loss[loss=2.344, over 1120.00 frames. , ppl: 10.420050757096128] tot_loss[loss=2.268, over 5437786.00 frames. , ppl: 9.663260376420673], batch size: 70
+2022-12-14 06:56:20,597 INFO [train.py:421] (5/8) Epoch 11, batch 41600, loss[loss=2.2, over 3570.00 frames. , ppl: 9.023400305939283] tot_loss[loss=2.267, over 5464232.76 frames. , ppl: 9.65412052567805], batch size: 70
+2022-12-14 06:58:01,897 INFO [train.py:421] (5/8) Epoch 11, batch 41800, loss[loss=2.283, over 2380.00 frames. , ppl: 9.810795625626541] tot_loss[loss=2.267, over 5486022.77 frames. , ppl: 9.64604725185067], batch size: 70
+2022-12-14 06:59:44,359 INFO [train.py:421] (5/8) Epoch 11, batch 42000, loss[loss=2.504, over 840.00 frames. , ppl: 12.23458697557122] tot_loss[loss=2.267, over 5461242.15 frames. , ppl: 9.655116221304128], batch size: 70
+2022-12-14 06:59:44,360 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 06:59:45,115 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.592491660652186
+2022-12-14 07:01:25,039 INFO [train.py:421] (5/8) Epoch 11, batch 42200, loss[loss=2.752, over 840.00 frames. , ppl: 15.66648243895923] tot_loss[loss=2.267, over 5463503.95 frames. , ppl: 9.6495517613829], batch size: 70
+2022-12-14 07:03:05,977 INFO [train.py:421] (5/8) Epoch 11, batch 42400, loss[loss=2.225, over 4410.00 frames. , ppl: 9.254195923968068] tot_loss[loss=2.266, over 5496216.49 frames. , ppl: 9.639605782023615], batch size: 70
+2022-12-14 07:04:49,210 INFO [train.py:421] (5/8) Epoch 11, batch 42600, loss[loss=2.338, over 2310.00 frames. , ppl: 10.360206992320926] tot_loss[loss=2.265, over 5501826.09 frames. , ppl: 9.634854002410298], batch size: 70
+2022-12-14 07:06:29,113 INFO [train.py:421] (5/8) Epoch 11, batch 42800, loss[loss=2.309, over 3010.00 frames. , ppl: 10.063744155938672] tot_loss[loss=2.265, over 5506929.16 frames. , ppl: 9.635732036929715], batch size: 70
+2022-12-14 07:08:09,283 INFO [train.py:421] (5/8) Epoch 11, batch 43000, loss[loss=2.801, over 630.00 frames. , ppl: 16.4588895661558] tot_loss[loss=2.267, over 5466373.55 frames. , ppl: 9.648548338935276], batch size: 70
+2022-12-14 07:08:09,284 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 07:08:10,045 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.259, over 211138.00 frames. , ppl: 9.569812285675939
+2022-12-14 07:09:48,736 INFO [train.py:421] (5/8) Epoch 11, batch 43200, loss[loss=2.295, over 4550.00 frames. , ppl: 9.921501205742691] tot_loss[loss=2.267, over 5452111.69 frames. , ppl: 9.650100777902566], batch size: 70
+2022-12-14 07:11:27,396 INFO [train.py:421] (5/8) Epoch 11, batch 43400, loss[loss=2.31, over 2520.00 frames. , ppl: 10.079255952518317] tot_loss[loss=2.268, over 5429396.25 frames. , ppl: 9.655233287703473], batch size: 70
+2022-12-14 07:13:07,427 INFO [train.py:421] (5/8) Epoch 11, batch 43600, loss[loss=2.283, over 1680.00 frames. , ppl: 9.80291815078205] tot_loss[loss=2.268, over 5426778.36 frames. , ppl: 9.663551259477996], batch size: 70
+2022-12-14 07:14:45,248 INFO [train.py:421] (5/8) Epoch 11, batch 43800, loss[loss=2.406, over 2030.00 frames. , ppl: 11.084317688446774] tot_loss[loss=2.269, over 5434947.30 frames. , ppl: 9.669180594338545], batch size: 70
+2022-12-14 07:16:27,746 INFO [train.py:421] (5/8) Epoch 11, batch 44000, loss[loss=2.609, over 1050.00 frames. , ppl: 13.580790030957102] tot_loss[loss=2.269, over 5438731.84 frames. , ppl: 9.668383886205064], batch size: 70
+2022-12-14 07:16:27,746 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 07:16:28,507 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.259, over 211138.00 frames. , ppl: 9.577322170162937
+2022-12-14 07:18:06,715 INFO [train.py:421] (5/8) Epoch 11, batch 44200, loss[loss=2.158, over 8540.00 frames. , ppl: 8.655822105424697] tot_loss[loss=2.269, over 5472574.69 frames. , ppl: 9.667652019910502], batch size: 70
+2022-12-14 07:19:47,741 INFO [train.py:421] (5/8) Epoch 11, batch 44400, loss[loss=2.25, over 4060.00 frames. , ppl: 9.486044943462824] tot_loss[loss=2.269, over 5483971.24 frames. , ppl: 9.666920252632062], batch size: 70
+2022-12-14 07:21:28,931 INFO [train.py:421] (5/8) Epoch 11, batch 44600, loss[loss=2.189, over 4270.00 frames. , ppl: 8.922331529924403] tot_loss[loss=2.267, over 5495198.95 frames. , ppl: 9.654397328454897], batch size: 70
+2022-12-14 07:23:10,233 INFO [train.py:421] (5/8) Epoch 11, batch 44800, loss[loss=2.419, over 1050.00 frames. , ppl: 11.23020738759103] tot_loss[loss=2.268, over 5486537.01 frames. , ppl: 9.657139099647054], batch size: 70
+2022-12-14 07:24:47,373 INFO [train.py:421] (5/8) Epoch 11, batch 45000, loss[loss=2.514, over 910.00 frames. , ppl: 12.356856987633705] tot_loss[loss=2.267, over 5486012.71 frames. , ppl: 9.65287128828793], batch size: 70
+2022-12-14 07:24:47,374 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 07:24:48,136 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.587963694090844
+2022-12-14 07:26:33,422 INFO [train.py:421] (5/8) Epoch 11, batch 45200, loss[loss=2.161, over 7140.00 frames. , ppl: 8.676541961984723] tot_loss[loss=2.266, over 5489591.56 frames. , ppl: 9.644734168012702], batch size: 70
+2022-12-14 07:28:14,101 INFO [train.py:421] (5/8) Epoch 11, batch 45400, loss[loss=2.405, over 1050.00 frames. , ppl: 11.078698179014907] tot_loss[loss=2.266, over 5537862.11 frames. , ppl: 9.636013031259965], batch size: 70
+2022-12-14 07:29:58,397 INFO [train.py:421] (5/8) Epoch 11, batch 45600, loss[loss=2.181, over 12040.00 frames. , ppl: 8.851108648367687] tot_loss[loss=2.264, over 5579322.46 frames. , ppl: 9.6224558000434], batch size: 70
+2022-12-14 07:31:41,339 INFO [train.py:421] (5/8) Epoch 11, batch 45800, loss[loss=2.312, over 1960.00 frames. , ppl: 10.093897192647296] tot_loss[loss=2.263, over 5604847.88 frames. , ppl: 9.615446418747272], batch size: 70
+2022-12-14 07:33:22,374 INFO [train.py:421] (5/8) Epoch 11, batch 46000, loss[loss=2.222, over 3780.00 frames. , ppl: 9.221155846680931] tot_loss[loss=2.264, over 5574678.55 frames. , ppl: 9.620463009854976], batch size: 70
+2022-12-14 07:33:22,375 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 07:33:23,121 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.604798933149176
+2022-12-14 07:35:07,447 INFO [train.py:421] (5/8) Epoch 11, batch 46200, loss[loss=2.472, over 1190.00 frames. , ppl: 11.84082989098955] tot_loss[loss=2.264, over 5565948.32 frames. , ppl: 9.62545996751278], batch size: 70
+2022-12-14 07:36:50,560 INFO [train.py:421] (5/8) Epoch 11, batch 46400, loss[loss=2.438, over 1750.00 frames. , ppl: 11.454130948109926] tot_loss[loss=2.265, over 5567281.16 frames. , ppl: 9.627130045869949], batch size: 70
+2022-12-14 07:38:31,817 INFO [train.py:421] (5/8) Epoch 11, batch 46600, loss[loss=2.389, over 2030.00 frames. , ppl: 10.900770915546385] tot_loss[loss=2.264, over 5589364.53 frames. , ppl: 9.623587494382141], batch size: 70
+2022-12-14 07:40:11,811 INFO [train.py:421] (5/8) Epoch 11, batch 46800, loss[loss=2.073, over 7000.00 frames. , ppl: 7.947936922564024] tot_loss[loss=2.264, over 5594724.62 frames. , ppl: 9.62092988996904], batch size: 70
+2022-12-14 07:41:53,103 INFO [train.py:421] (5/8) Epoch 11, batch 47000, loss[loss=2.323, over 1750.00 frames. , ppl: 10.208915832708561] tot_loss[loss=2.265, over 5561790.45 frames. , ppl: 9.630828111424037], batch size: 70
+2022-12-14 07:41:53,104 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 07:41:53,854 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.593797929123168
+2022-12-14 07:43:37,129 INFO [train.py:421] (5/8) Epoch 11, batch 47200, loss[loss=2.457, over 1470.00 frames. , ppl: 11.67191319483212] tot_loss[loss=2.264, over 5609501.45 frames. , ppl: 9.619421610403183], batch size: 70
+2022-12-14 07:45:10,242 INFO [train.py:421] (5/8) Epoch 11, batch 47400, loss[loss=2.365, over 1050.00 frames. , ppl: 10.645182281532783] tot_loss[loss=2.265, over 5554418.68 frames. , ppl: 9.63405817635763], batch size: 70
+2022-12-14 07:46:47,936 INFO [train.py:421] (5/8) Epoch 11, batch 47600, loss[loss=2.495, over 1400.00 frames. , ppl: 12.12277146954699] tot_loss[loss=2.267, over 5486793.33 frames. , ppl: 9.64987416252323], batch size: 70
+2022-12-14 07:48:29,964 INFO [train.py:421] (5/8) Epoch 11, batch 47800, loss[loss=4.974, over 280.00 frames. , ppl: 144.5472804782944] tot_loss[loss=2.266, over 5531941.87 frames. , ppl: 9.638817705451459], batch size: 70
+2022-12-14 07:50:08,692 INFO [train.py:421] (5/8) Epoch 11, batch 48000, loss[loss=2.282, over 2450.00 frames. , ppl: 9.793377515246187] tot_loss[loss=2.268, over 5462822.91 frames. , ppl: 9.656086772838513], batch size: 70
+2022-12-14 07:50:08,692 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 07:50:09,455 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.597875482483978
+2022-12-14 07:51:49,511 INFO [train.py:421] (5/8) Epoch 11, batch 48200, loss[loss=2.484, over 910.00 frames. , ppl: 11.987496913012375] tot_loss[loss=2.267, over 5487210.07 frames. , ppl: 9.651773390865324], batch size: 70
+2022-12-14 07:53:28,427 INFO [train.py:421] (5/8) Epoch 11, batch 48400, loss[loss=2.268, over 4970.00 frames. , ppl: 9.655897705907961] tot_loss[loss=2.267, over 5469290.67 frames. , ppl: 9.653314915133441], batch size: 70
+2022-12-14 07:55:09,672 INFO [train.py:421] (5/8) Epoch 11, batch 48600, loss[loss=2.411, over 1400.00 frames. , ppl: 11.140677252874834] tot_loss[loss=2.268, over 5463725.35 frames. , ppl: 9.661095332031774], batch size: 70
+2022-12-14 07:56:48,791 INFO [train.py:421] (5/8) Epoch 11, batch 48800, loss[loss=2.58, over 1260.00 frames. , ppl: 13.194448866277108] tot_loss[loss=2.268, over 5448680.30 frames. , ppl: 9.661919081081836], batch size: 70
+2022-12-14 07:58:30,715 INFO [train.py:421] (5/8) Epoch 11, batch 49000, loss[loss=2.404, over 1820.00 frames. , ppl: 11.071264267708136] tot_loss[loss=2.267, over 5491868.33 frames. , ppl: 9.651951488608981], batch size: 70
+2022-12-14 07:58:30,715 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 07:58:31,475 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.595209466935325
+2022-12-14 08:00:11,580 INFO [train.py:421] (5/8) Epoch 11, batch 49200, loss[loss=2.311, over 2730.00 frames. , ppl: 10.088688754208595] tot_loss[loss=2.267, over 5493438.58 frames. , ppl: 9.65272221753104], batch size: 70
+2022-12-14 08:01:52,649 INFO [train.py:421] (5/8) Epoch 11, batch 49400, loss[loss=2.205, over 3920.00 frames. , ppl: 9.071240426684453] tot_loss[loss=2.268, over 5474481.68 frames. , ppl: 9.65988292054353], batch size: 70
+2022-12-14 08:03:32,888 INFO [train.py:421] (5/8) Epoch 11, batch 49600, loss[loss=3.299, over 490.00 frames. , ppl: 27.080507106535944] tot_loss[loss=2.269, over 5453712.29 frames. , ppl: 9.668583147816323], batch size: 70
+2022-12-14 08:05:11,615 INFO [train.py:421] (5/8) Epoch 11, batch 49800, loss[loss=2.344, over 1820.00 frames. , ppl: 10.41869063222318] tot_loss[loss=2.27, over 5434710.92 frames. , ppl: 9.676780894155906], batch size: 70
+2022-12-14 08:06:50,133 INFO [train.py:421] (5/8) Epoch 11, batch 50000, loss[loss=2.388, over 1610.00 frames. , ppl: 10.894983779786465] tot_loss[loss=2.27, over 5418298.27 frames. , ppl: 9.680742730562061], batch size: 70
+2022-12-14 08:06:50,134 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 08:06:50,881 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.581751511481997
+2022-12-14 08:08:32,386 INFO [train.py:421] (5/8) Epoch 11, batch 50200, loss[loss=2.248, over 3010.00 frames. , ppl: 9.467893022634168] tot_loss[loss=2.268, over 5454232.74 frames. , ppl: 9.660284666839463], batch size: 70
+2022-12-14 08:10:12,783 INFO [train.py:421] (5/8) Epoch 11, batch 50400, loss[loss=2.166, over 3780.00 frames. , ppl: 8.723376047508404] tot_loss[loss=2.268, over 5457564.05 frames. , ppl: 9.658585198231277], batch size: 70
+2022-12-14 08:11:48,457 INFO [train.py:421] (5/8) Epoch 11, batch 50600, loss[loss=2.202, over 4410.00 frames. , ppl: 9.044307857619685] tot_loss[loss=2.268, over 5463636.05 frames. , ppl: 9.660166235522256], batch size: 70
+2022-12-14 08:13:30,544 INFO [train.py:421] (5/8) Epoch 11, batch 50800, loss[loss=2.19, over 4760.00 frames. , ppl: 8.939088133949655] tot_loss[loss=2.269, over 5449384.96 frames. , ppl: 9.665169269743409], batch size: 70
+2022-12-14 08:15:09,636 INFO [train.py:421] (5/8) Epoch 11, batch 51000, loss[loss=2.297, over 3010.00 frames. , ppl: 9.941058691061711] tot_loss[loss=2.268, over 5485663.10 frames. , ppl: 9.655512502480843], batch size: 70
+2022-12-14 08:15:09,636 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 08:15:10,400 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.599885781441973
+2022-12-14 08:16:50,480 INFO [train.py:421] (5/8) Epoch 11, batch 51200, loss[loss=2.337, over 3220.00 frames. , ppl: 10.34715364012717] tot_loss[loss=2.267, over 5503398.13 frames. , ppl: 9.64958758688083], batch size: 70
+2022-12-14 08:18:32,234 INFO [train.py:421] (5/8) Epoch 11, batch 51400, loss[loss=2.736, over 770.00 frames. , ppl: 15.431446245474442] tot_loss[loss=2.268, over 5494806.68 frames. , ppl: 9.655699390149083], batch size: 70
+2022-12-14 08:20:10,263 INFO [train.py:421] (5/8) Epoch 11, batch 51600, loss[loss=2.308, over 1750.00 frames. , ppl: 10.049498572284596] tot_loss[loss=2.266, over 5522464.08 frames. , ppl: 9.642146997214407], batch size: 70
+2022-12-14 08:21:49,056 INFO [train.py:421] (5/8) Epoch 11, batch 51800, loss[loss=2.234, over 4130.00 frames. , ppl: 9.34173547143758] tot_loss[loss=2.267, over 5485464.45 frames. , ppl: 9.653338149896431], batch size: 70
+2022-12-14 08:23:27,648 INFO [train.py:421] (5/8) Epoch 11, batch 52000, loss[loss=2.307, over 3220.00 frames. , ppl: 10.043259865077586] tot_loss[loss=2.267, over 5491898.65 frames. , ppl: 9.65191003346385], batch size: 70
+2022-12-14 08:23:27,649 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 08:23:28,401 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.587526624273279
+2022-12-14 08:25:07,202 INFO [train.py:421] (5/8) Epoch 11, batch 52200, loss[loss=2.214, over 3150.00 frames. , ppl: 9.155404018233673] tot_loss[loss=2.268, over 5466346.96 frames. , ppl: 9.659993634429105], batch size: 70
+2022-12-14 08:26:45,635 INFO [train.py:421] (5/8) Epoch 11, batch 52400, loss[loss=2.398, over 1890.00 frames. , ppl: 10.996416471946041] tot_loss[loss=2.268, over 5455217.79 frames. , ppl: 9.65566293441011], batch size: 70
+2022-12-14 08:28:22,624 INFO [train.py:421] (5/8) Epoch 11, batch 52600, loss[loss=2.184, over 4130.00 frames. , ppl: 8.879301152575719] tot_loss[loss=2.268, over 5415145.05 frames. , ppl: 9.662160384549347], batch size: 70
+2022-12-14 08:30:00,798 INFO [train.py:421] (5/8) Epoch 11, batch 52800, loss[loss=2.27, over 3150.00 frames. , ppl: 9.680155546281139] tot_loss[loss=2.268, over 5408951.00 frames. , ppl: 9.662263822956344], batch size: 70
+2022-12-14 08:31:42,278 INFO [train.py:421] (5/8) Epoch 11, batch 53000, loss[loss=2.261, over 4480.00 frames. , ppl: 9.59583143786289] tot_loss[loss=2.267, over 5437232.84 frames. , ppl: 9.65507218683368], batch size: 70
+2022-12-14 08:31:42,278 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 08:31:43,029 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.581084994414798
+2022-12-14 08:33:24,598 INFO [train.py:421] (5/8) Epoch 11, batch 53200, loss[loss=2.195, over 5040.00 frames. , ppl: 8.981591197312511] tot_loss[loss=2.268, over 5434154.04 frames. , ppl: 9.660717117610709], batch size: 70
+2022-12-14 08:35:06,681 INFO [train.py:421] (5/8) Epoch 11, batch 53400, loss[loss=2.488, over 1470.00 frames. , ppl: 12.03742045502865] tot_loss[loss=2.268, over 5438952.46 frames. , ppl: 9.663745445939943], batch size: 70
+2022-12-14 08:36:47,427 INFO [train.py:421] (5/8) Epoch 11, batch 53600, loss[loss=2.322, over 2170.00 frames. , ppl: 10.199264177191367] tot_loss[loss=2.267, over 5488757.55 frames. , ppl: 9.650525252007249], batch size: 70
+2022-12-14 08:38:27,754 INFO [train.py:421] (5/8) Epoch 11, batch 53800, loss[loss=2.593, over 1120.00 frames. , ppl: 13.364330330399394] tot_loss[loss=2.268, over 5454836.82 frames. , ppl: 9.658026083087778], batch size: 70
+2022-12-14 08:40:09,534 INFO [train.py:421] (5/8) Epoch 11, batch 54000, loss[loss=2.348, over 4340.00 frames. , ppl: 10.46834843457491] tot_loss[loss=2.266, over 5545689.70 frames. , ppl: 9.641182542996228], batch size: 70
+2022-12-14 08:40:09,534 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 08:40:10,265 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.59367723381749
+2022-12-14 08:41:46,178 INFO [train.py:421] (5/8) Epoch 11, batch 54200, loss[loss=2.243, over 8190.00 frames. , ppl: 9.421609674440676] tot_loss[loss=2.266, over 5519326.78 frames. , ppl: 9.6418816413297], batch size: 70
+2022-12-14 08:43:25,433 INFO [train.py:421] (5/8) Epoch 11, batch 54400, loss[loss=2.306, over 1680.00 frames. , ppl: 10.029274612057812] tot_loss[loss=2.266, over 5541156.50 frames. , ppl: 9.639053958534499], batch size: 70
+2022-12-14 08:45:04,359 INFO [train.py:421] (5/8) Epoch 11, batch 54600, loss[loss=2.82, over 630.00 frames. , ppl: 16.77232237627949] tot_loss[loss=2.264, over 5604709.41 frames. , ppl: 9.6206982793818], batch size: 70
+2022-12-14 08:46:40,893 INFO [train.py:421] (5/8) Epoch 11, batch 54800, loss[loss=2.179, over 3990.00 frames. , ppl: 8.83473933848211] tot_loss[loss=2.265, over 5561309.33 frames. , ppl: 9.631673300468993], batch size: 70
+2022-12-14 08:48:23,510 INFO [train.py:421] (5/8) Epoch 11, batch 55000, loss[loss=2.382, over 1610.00 frames. , ppl: 10.826862777635336] tot_loss[loss=2.266, over 5533100.13 frames. , ppl: 9.64203664643903], batch size: 70
+2022-12-14 08:48:23,511 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 08:48:24,275 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.600448456413142
+2022-12-14 08:50:04,126 INFO [train.py:421] (5/8) Epoch 11, batch 55200, loss[loss=2.333, over 2240.00 frames. , ppl: 10.303752982979884] tot_loss[loss=2.265, over 5572531.16 frames. , ppl: 9.63101702643299], batch size: 70
+2022-12-14 08:51:43,860 INFO [train.py:421] (5/8) Epoch 11, batch 55400, loss[loss=2.198, over 12810.00 frames. , ppl: 9.005965942198738] tot_loss[loss=2.264, over 5585491.14 frames. , ppl: 9.62069426801934], batch size: 70
+2022-12-14 08:53:24,613 INFO [train.py:421] (5/8) Epoch 11, batch 55600, loss[loss=2.196, over 3640.00 frames. , ppl: 8.989385742933242] tot_loss[loss=2.264, over 5582653.72 frames. , ppl: 9.618465609928966], batch size: 70
+2022-12-14 08:55:02,629 INFO [train.py:421] (5/8) Epoch 11, batch 55800, loss[loss=2.424, over 1330.00 frames. , ppl: 11.295092676414178] tot_loss[loss=2.264, over 5588351.44 frames. , ppl: 9.618166602955974], batch size: 70
+2022-12-14 08:56:44,614 INFO [train.py:421] (5/8) Epoch 11, batch 56000, loss[loss=2.428, over 1190.00 frames. , ppl: 11.3378041703283] tot_loss[loss=2.264, over 5559508.17 frames. , ppl: 9.623928762497139], batch size: 70
+2022-12-14 08:56:44,615 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 08:56:45,361 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.589785979049779
+2022-12-14 08:58:29,520 INFO [train.py:421] (5/8) Epoch 11, batch 56200, loss[loss=2.362, over 2450.00 frames. , ppl: 10.61117916458109] tot_loss[loss=2.264, over 5581210.87 frames. , ppl: 9.621596968263063], batch size: 70
+2022-12-14 09:00:08,834 INFO [train.py:421] (5/8) Epoch 11, batch 56400, loss[loss=2.188, over 3500.00 frames. , ppl: 8.919774333361847] tot_loss[loss=2.265, over 5530276.51 frames. , ppl: 9.632185173011441], batch size: 70
+2022-12-14 09:01:50,867 INFO [train.py:421] (5/8) Epoch 11, batch 56600, loss[loss=2.403, over 1400.00 frames. , ppl: 11.060738332099396] tot_loss[loss=2.263, over 5594135.41 frames. , ppl: 9.612064107719686], batch size: 70
+2022-12-14 09:03:29,228 INFO [train.py:421] (5/8) Epoch 11, batch 56800, loss[loss=2.306, over 3430.00 frames. , ppl: 10.0325522589931] tot_loss[loss=2.262, over 5616026.58 frames. , ppl: 9.60581365834814], batch size: 70
+2022-12-14 09:05:09,686 INFO [train.py:421] (5/8) Epoch 11, batch 57000, loss[loss=2.21, over 5530.00 frames. , ppl: 9.117379122346724] tot_loss[loss=2.261, over 5643035.32 frames. , ppl: 9.592271135900019], batch size: 70
+2022-12-14 09:05:09,687 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 09:05:10,418 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.587563519007391
+2022-12-14 09:06:48,001 INFO [train.py:421] (5/8) Epoch 11, batch 57200, loss[loss=2.24, over 2030.00 frames. , ppl: 9.39035251141119] tot_loss[loss=2.262, over 5607586.60 frames. , ppl: 9.599350430519593], batch size: 70
+2022-12-14 09:08:32,480 INFO [train.py:421] (5/8) Epoch 11, batch 57400, loss[loss=2.431, over 1610.00 frames. , ppl: 11.370778808571506] tot_loss[loss=2.262, over 5575086.18 frames. , ppl: 9.604473186041314], batch size: 70
+2022-12-14 09:10:14,461 INFO [train.py:421] (5/8) Epoch 11, batch 57600, loss[loss=2.145, over 11900.00 frames. , ppl: 8.542630093153136] tot_loss[loss=2.263, over 5559959.65 frames. , ppl: 9.60928912800849], batch size: 70
+2022-12-14 09:11:54,086 INFO [train.py:421] (5/8) Epoch 11, batch 57800, loss[loss=2.32, over 1260.00 frames. , ppl: 10.171617038192] tot_loss[loss=2.262, over 5561488.69 frames. , ppl: 9.602849558412151], batch size: 70
+2022-12-14 09:13:32,715 INFO [train.py:421] (5/8) Epoch 11, batch 58000, loss[loss=2.274, over 3290.00 frames. , ppl: 9.713843911790464] tot_loss[loss=2.261, over 5575054.41 frames. , ppl: 9.595990132000347], batch size: 70
+2022-12-14 09:13:32,715 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 09:13:33,485 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.579467110769087
+2022-12-14 09:15:15,193 INFO [train.py:421] (5/8) Epoch 11, batch 58200, loss[loss=2.539, over 1190.00 frames. , ppl: 12.669676310521515] tot_loss[loss=2.262, over 5575868.11 frames. , ppl: 9.602553038063832], batch size: 70
+2022-12-14 09:16:56,037 INFO [train.py:421] (5/8) Epoch 11, batch 58400, loss[loss=2.157, over 5950.00 frames. , ppl: 8.644532400176244] tot_loss[loss=2.262, over 5574572.79 frames. , ppl: 9.602603215536359], batch size: 70
+2022-12-14 09:18:34,148 INFO [train.py:421] (5/8) Epoch 11, batch 58600, loss[loss=2.669, over 840.00 frames. , ppl: 14.422903001527681] tot_loss[loss=2.261, over 5610193.92 frames. , ppl: 9.592722552852068], batch size: 70
+2022-12-14 09:20:16,253 INFO [train.py:421] (5/8) Epoch 11, batch 58800, loss[loss=2.193, over 4900.00 frames. , ppl: 8.964211898705353] tot_loss[loss=2.262, over 5582603.17 frames. , ppl: 9.602611612928042], batch size: 70
+2022-12-14 09:21:58,265 INFO [train.py:421] (5/8) Epoch 11, batch 59000, loss[loss=2.613, over 840.00 frames. , ppl: 13.637236281224805] tot_loss[loss=2.262, over 5580068.13 frames. , ppl: 9.602159021922448], batch size: 70
+2022-12-14 09:21:58,266 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 09:21:59,029 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.585608293732157
+2022-12-14 09:23:40,364 INFO [train.py:421] (5/8) Epoch 11, batch 59200, loss[loss=2.457, over 1330.00 frames. , ppl: 11.666717347622496] tot_loss[loss=2.262, over 5613954.09 frames. , ppl: 9.59794232099684], batch size: 70
+2022-12-14 09:25:21,848 INFO [train.py:421] (5/8) Epoch 11, batch 59400, loss[loss=2.172, over 4550.00 frames. , ppl: 8.774589397606135] tot_loss[loss=2.261, over 5623815.02 frames. , ppl: 9.589244575164102], batch size: 70
+2022-12-14 09:27:00,559 INFO [train.py:421] (5/8) Epoch 11, batch 59600, loss[loss=2.516, over 1190.00 frames. , ppl: 12.381140890982827] tot_loss[loss=2.262, over 5582213.69 frames. , ppl: 9.601118574770648], batch size: 70
+2022-12-14 09:28:41,363 INFO [train.py:421] (5/8) Epoch 11, batch 59800, loss[loss=2.234, over 1820.00 frames. , ppl: 9.334911484678234] tot_loss[loss=2.262, over 5581770.24 frames. , ppl: 9.604575881017409], batch size: 70
+2022-12-14 09:30:19,023 INFO [train.py:421] (5/8) Epoch 11, batch 60000, loss[loss=2.19, over 4200.00 frames. , ppl: 8.931225070527207] tot_loss[loss=2.261, over 5623417.88 frames. , ppl: 9.59048660268123], batch size: 70
+2022-12-14 09:30:19,023 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 09:30:19,782 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.259, over 211138.00 frames. , ppl: 9.57757590873942
+2022-12-14 09:32:01,996 INFO [train.py:421] (5/8) Epoch 11, batch 60200, loss[loss=2.384, over 1470.00 frames. , ppl: 10.852739287404853] tot_loss[loss=2.26, over 5641155.79 frames. , ppl: 9.585791678951514], batch size: 70
+2022-12-14 09:33:42,587 INFO [train.py:421] (5/8) Epoch 11, batch 60400, loss[loss=2.208, over 3710.00 frames. , ppl: 9.10164616093508] tot_loss[loss=2.261, over 5630183.87 frames. , ppl: 9.591783620303548], batch size: 70
+2022-12-14 09:35:24,852 INFO [train.py:421] (5/8) Epoch 11, batch 60600, loss[loss=2.211, over 5040.00 frames. , ppl: 9.126784277909056] tot_loss[loss=2.262, over 5588834.02 frames. , ppl: 9.603685648428707], batch size: 70
+2022-12-14 09:37:07,335 INFO [train.py:421] (5/8) Epoch 11, batch 60800, loss[loss=2.464, over 1470.00 frames. , ppl: 11.756562607761788] tot_loss[loss=2.263, over 5549148.08 frames. , ppl: 9.61435197855479], batch size: 70
+2022-12-14 09:38:49,724 INFO [train.py:421] (5/8) Epoch 11, batch 61000, loss[loss=2.504, over 980.00 frames. , ppl: 12.233155045866285] tot_loss[loss=2.263, over 5582301.48 frames. , ppl: 9.614178739012445], batch size: 70
+2022-12-14 09:38:49,724 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 09:38:50,484 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.587292487923966
+2022-12-14 09:40:29,790 INFO [train.py:421] (5/8) Epoch 11, batch 61200, loss[loss=2.577, over 840.00 frames. , ppl: 13.15292886888532] tot_loss[loss=2.263, over 5563988.53 frames. , ppl: 9.614072927323312], batch size: 70
+2022-12-14 09:42:09,350 INFO [train.py:421] (5/8) Epoch 11, batch 61400, loss[loss=2.242, over 4200.00 frames. , ppl: 9.414941462036637] tot_loss[loss=2.264, over 5533050.16 frames. , ppl: 9.625200276010517], batch size: 70
+2022-12-14 09:43:48,639 INFO [train.py:421] (5/8) Epoch 11, batch 61600, loss[loss=2.166, over 4060.00 frames. , ppl: 8.720486527375012] tot_loss[loss=2.265, over 5528910.41 frames. , ppl: 9.629841629306014], batch size: 70
+2022-12-14 09:45:23,760 INFO [train.py:421] (5/8) Epoch 11, batch 61800, loss[loss=2.281, over 2310.00 frames. , ppl: 9.78608077935181] tot_loss[loss=2.265, over 5520789.76 frames. , ppl: 9.63007203019927], batch size: 70
+2022-12-14 09:47:03,482 INFO [train.py:421] (5/8) Epoch 11, batch 62000, loss[loss=2.319, over 1400.00 frames. , ppl: 10.164986804976834] tot_loss[loss=2.265, over 5523449.81 frames. , ppl: 9.632664686037533], batch size: 70
+2022-12-14 09:47:03,483 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 09:47:04,244 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.582718752746382
+2022-12-14 09:48:44,734 INFO [train.py:421] (5/8) Epoch 11, batch 62200, loss[loss=2.367, over 2310.00 frames. , ppl: 10.670338466373197] tot_loss[loss=2.266, over 5525671.38 frames. , ppl: 9.6427385389263], batch size: 70
+2022-12-14 09:50:24,715 INFO [train.py:421] (5/8) Epoch 11, batch 62400, loss[loss=2.505, over 840.00 frames. , ppl: 12.238523995080786] tot_loss[loss=2.265, over 5535948.57 frames. , ppl: 9.631133999917038], batch size: 70
+2022-12-14 09:52:03,661 INFO [train.py:421] (5/8) Epoch 11, batch 62600, loss[loss=2.246, over 2940.00 frames. , ppl: 9.452099751349948] tot_loss[loss=2.265, over 5501695.06 frames. , ppl: 9.634545949165704], batch size: 70
+2022-12-14 09:53:42,469 INFO [train.py:421] (5/8) Epoch 11, batch 62800, loss[loss=2.388, over 980.00 frames. , ppl: 10.892966731833402] tot_loss[loss=2.265, over 5521460.29 frames. , ppl: 9.630392547380277], batch size: 70
+2022-12-14 09:55:20,749 INFO [train.py:421] (5/8) Epoch 11, batch 63000, loss[loss=2.227, over 4060.00 frames. , ppl: 9.267529125541731] tot_loss[loss=2.264, over 5573344.26 frames. , ppl: 9.619545529380114], batch size: 70
+2022-12-14 09:55:20,750 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 09:55:21,511 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.60109500498798
+2022-12-14 09:57:02,142 INFO [train.py:421] (5/8) Epoch 11, batch 63200, loss[loss=2.788, over 630.00 frames. , ppl: 16.24789061655444] tot_loss[loss=2.265, over 5570392.45 frames. , ppl: 9.62643873204434], batch size: 70
+2022-12-14 09:58:42,040 INFO [train.py:421] (5/8) Epoch 11, batch 63400, loss[loss=2.709, over 770.00 frames. , ppl: 15.015594941024526] tot_loss[loss=2.265, over 5566118.88 frames. , ppl: 9.63374157373298], batch size: 70
+2022-12-14 10:00:19,149 INFO [train.py:421] (5/8) Epoch 11, batch 63600, loss[loss=2.353, over 2030.00 frames. , ppl: 10.522130941414888] tot_loss[loss=2.266, over 5515412.18 frames. , ppl: 9.641062730118925], batch size: 70
+2022-12-14 10:02:00,367 INFO [train.py:421] (5/8) Epoch 11, batch 63800, loss[loss=2.19, over 3290.00 frames. , ppl: 8.93372428410526] tot_loss[loss=2.264, over 5566394.79 frames. , ppl: 9.623707942380367], batch size: 70
+2022-12-14 10:03:38,191 INFO [train.py:421] (5/8) Epoch 11, batch 64000, loss[loss=2.227, over 3150.00 frames. , ppl: 9.268728482186571] tot_loss[loss=2.265, over 5534256.15 frames. , ppl: 9.629495096954887], batch size: 70
+2022-12-14 10:03:38,191 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 10:03:38,952 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.263, over 211138.00 frames. , ppl: 9.61195926586509
+2022-12-14 10:05:19,744 INFO [train.py:421] (5/8) Epoch 11, batch 64200, loss[loss=3.219, over 490.00 frames. , ppl: 24.995795958695258] tot_loss[loss=2.265, over 5524055.12 frames. , ppl: 9.634313243964442], batch size: 70
+2022-12-14 10:07:02,734 INFO [train.py:421] (5/8) Epoch 11, batch 64400, loss[loss=2.441, over 1750.00 frames. , ppl: 11.481616715145872] tot_loss[loss=2.266, over 5504936.95 frames. , ppl: 9.639985877744499], batch size: 70
+2022-12-14 10:08:42,830 INFO [train.py:421] (5/8) Epoch 11, batch 64600, loss[loss=2.285, over 2100.00 frames. , ppl: 9.824674190249718] tot_loss[loss=2.267, over 5479661.75 frames. , ppl: 9.652069130300752], batch size: 70
+2022-12-14 10:10:23,668 INFO [train.py:421] (5/8) Epoch 11, batch 64800, loss[loss=2.117, over 8610.00 frames. , ppl: 8.304575510885488] tot_loss[loss=2.266, over 5526310.61 frames. , ppl: 9.641023837251444], batch size: 70
+2022-12-14 10:12:04,659 INFO [train.py:421] (5/8) Epoch 11, batch 65000, loss[loss=2.626, over 770.00 frames. , ppl: 13.814813497009968] tot_loss[loss=2.266, over 5517294.04 frames. , ppl: 9.639034795378596], batch size: 70
+2022-12-14 10:12:04,659 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 10:12:05,419 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.588153854062595
+2022-12-14 10:13:48,151 INFO [train.py:421] (5/8) Epoch 11, batch 65200, loss[loss=2.251, over 3570.00 frames. , ppl: 9.499453244098738] tot_loss[loss=2.264, over 5543575.38 frames. , ppl: 9.626046331873965], batch size: 70
+2022-12-14 10:15:30,903 INFO [train.py:421] (5/8) Epoch 11, batch 65400, loss[loss=2.303, over 2520.00 frames. , ppl: 10.005160135574696] tot_loss[loss=2.264, over 5566170.40 frames. , ppl: 9.619769998840432], batch size: 70
+2022-12-14 10:17:13,918 INFO [train.py:421] (5/8) Epoch 11, batch 65600, loss[loss=2.3, over 2310.00 frames. , ppl: 9.97749938477749] tot_loss[loss=2.266, over 5537308.16 frames. , ppl: 9.63972340803574], batch size: 70
+2022-12-14 10:18:52,834 INFO [train.py:421] (5/8) Epoch 11, batch 65800, loss[loss=2.662, over 700.00 frames. , ppl: 14.324098939630467] tot_loss[loss=2.266, over 5545155.11 frames. , ppl: 9.642192173648883], batch size: 70
+2022-12-14 10:20:34,406 INFO [train.py:421] (5/8) Epoch 11, batch 66000, loss[loss=2.293, over 2450.00 frames. , ppl: 9.90044118996814] tot_loss[loss=2.266, over 5539356.82 frames. , ppl: 9.640492586314329], batch size: 70
+2022-12-14 10:20:34,407 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 10:20:35,171 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.583815172113377
+2022-12-14 10:22:18,347 INFO [train.py:421] (5/8) Epoch 11, batch 66200, loss[loss=2.359, over 1750.00 frames. , ppl: 10.58031560892452] tot_loss[loss=2.266, over 5545115.70 frames. , ppl: 9.636571580900895], batch size: 70
+2022-12-14 10:24:06,433 INFO [train.py:421] (5/8) Epoch 11, batch 66400, loss[loss=2.287, over 2240.00 frames. , ppl: 9.847631893218658] tot_loss[loss=2.265, over 5575867.04 frames. , ppl: 9.632429162114688], batch size: 70
+2022-12-14 10:25:47,258 INFO [train.py:421] (5/8) Epoch 11, batch 66600, loss[loss=2.16, over 4340.00 frames. , ppl: 8.668957348789865] tot_loss[loss=2.264, over 5586035.72 frames. , ppl: 9.620764038460791], batch size: 70
+2022-12-14 10:27:31,529 INFO [train.py:421] (5/8) Epoch 11, batch 66800, loss[loss=2.291, over 2170.00 frames. , ppl: 9.885941834842532] tot_loss[loss=2.262, over 5626243.84 frames. , ppl: 9.605402508823877], batch size: 70
+2022-12-14 10:29:12,953 INFO [train.py:421] (5/8) Epoch 11, batch 67000, loss[loss=2.613, over 840.00 frames. , ppl: 13.642080630665015] tot_loss[loss=2.262, over 5614956.92 frames. , ppl: 9.604506747855705], batch size: 70
+2022-12-14 10:29:12,953 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 10:29:13,717 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.262, over 211138.00 frames. , ppl: 9.59918248410303
+2022-12-14 10:30:53,795 INFO [train.py:421] (5/8) Epoch 11, batch 67200, loss[loss=2.766, over 700.00 frames. , ppl: 15.9021426406667] tot_loss[loss=2.263, over 5631873.60 frames. , ppl: 9.607588548256297], batch size: 70
+2022-12-14 10:32:32,049 INFO [train.py:421] (5/8) Epoch 11, batch 67400, loss[loss=2.352, over 1260.00 frames. , ppl: 10.510754512607223] tot_loss[loss=2.263, over 5620435.67 frames. , ppl: 9.615575232494436], batch size: 70
+2022-12-14 10:34:15,945 INFO [train.py:421] (5/8) Epoch 11, batch 67600, loss[loss=2.254, over 2450.00 frames. , ppl: 9.521053084240764] tot_loss[loss=2.262, over 5638011.47 frames. , ppl: 9.607030399381028], batch size: 70
+2022-12-14 10:35:56,743 INFO [train.py:421] (5/8) Epoch 11, batch 67800, loss[loss=2.203, over 3780.00 frames. , ppl: 9.053384743019162] tot_loss[loss=2.265, over 5555223.01 frames. , ppl: 9.632694991759797], batch size: 70
+2022-12-14 10:37:37,856 INFO [train.py:421] (5/8) Epoch 11, batch 68000, loss[loss=2.585, over 910.00 frames. , ppl: 13.262408069270522] tot_loss[loss=2.265, over 5551544.38 frames. , ppl: 9.631416475513001], batch size: 70
+2022-12-14 10:37:37,856 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 10:37:38,622 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.26, over 211138.00 frames. , ppl: 9.586750448742256
+2022-12-14 10:39:17,198 INFO [train.py:421] (5/8) Epoch 11, batch 68200, loss[loss=2.434, over 1120.00 frames. , ppl: 11.40196098452864] tot_loss[loss=2.265, over 5578456.11 frames. , ppl: 9.628647046871023], batch size: 70
+2022-12-14 10:40:56,989 INFO [train.py:421] (5/8) Epoch 11, batch 68400, loss[loss=2.451, over 1190.00 frames. , ppl: 11.60487714161732] tot_loss[loss=2.264, over 5572591.62 frames. , ppl: 9.622737165083361], batch size: 70
+2022-12-14 10:42:35,925 INFO [train.py:421] (5/8) Epoch 11, batch 68600, loss[loss=2.316, over 1610.00 frames. , ppl: 10.138377785798106] tot_loss[loss=2.264, over 5566446.26 frames. , ppl: 9.62227802241054], batch size: 70
+2022-12-14 10:44:17,967 INFO [train.py:421] (5/8) Epoch 11, batch 68800, loss[loss=2.232, over 3430.00 frames. , ppl: 9.314009107212105] tot_loss[loss=2.265, over 5547875.30 frames. , ppl: 9.629887107602512], batch size: 70
+2022-12-14 10:46:02,172 INFO [train.py:421] (5/8) Epoch 11, batch 69000, loss[loss=2.333, over 2030.00 frames. , ppl: 10.305689096356966] tot_loss[loss=2.264, over 5579078.74 frames. , ppl: 9.618222914056538], batch size: 70
+2022-12-14 10:46:02,172 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 10:46:02,935 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.261, over 211138.00 frames. , ppl: 9.590407678839963
+2022-12-14 10:47:43,618 INFO [train.py:421] (5/8) Epoch 11, batch 69200, loss[loss=2.474, over 980.00 frames. , ppl: 11.870449620771756] tot_loss[loss=2.263, over 5607362.58 frames. , ppl: 9.615579917822236], batch size: 70
+2022-12-14 10:49:26,938 INFO [train.py:421] (5/8) Epoch 11, batch 69400, loss[loss=2.501, over 1190.00 frames. , ppl: 12.200122211347521] tot_loss[loss=2.263, over 5629721.12 frames. , ppl: 9.608238558687605], batch size: 70
+2022-12-14 10:51:09,376 INFO [train.py:421] (5/8) Epoch 11, batch 69600, loss[loss=2.147, over 12740.00 frames. , ppl: 8.556245156908068] tot_loss[loss=2.263, over 5619751.37 frames. , ppl: 9.611214557802446], batch size: 70
+2022-12-14 10:52:54,763 INFO [train.py:421] (5/8) Epoch 11, batch 69800, loss[loss=2.187, over 4550.00 frames. , ppl: 8.910850602302084] tot_loss[loss=2.262, over 5658701.75 frames. , ppl: 9.603465535356271], batch size: 70
+2022-12-14 10:54:39,529 INFO [train.py:421] (5/8) Epoch 11, batch 70000, loss[loss=2.492, over 1190.00 frames. , ppl: 12.089563999859637] tot_loss[loss=2.262, over 5651076.20 frames. , ppl: 9.604613695259665], batch size: 70
+2022-12-14 10:54:39,529 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 10:54:40,301 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.259, over 211138.00 frames. , ppl: 9.575361947313867
+2022-12-14 10:56:19,552 INFO [train.py:421] (5/8) Epoch 11, batch 70200, loss[loss=2.155, over 5460.00 frames. , ppl: 8.630272844081656] tot_loss[loss=2.263, over 5636496.86 frames. , ppl: 9.60832466391471], batch size: 70
+2022-12-14 10:58:03,412 INFO [train.py:421] (5/8) Epoch 11, batch 70400, loss[loss=2.494, over 1330.00 frames. , ppl: 12.107884288144493] tot_loss[loss=2.264, over 5627605.34 frames. , ppl: 9.61737248582195], batch size: 70
+2022-12-14 10:59:41,663 INFO [train.py:421] (5/8) Epoch 11, batch 70600, loss[loss=2.769, over 630.00 frames. , ppl: 15.94177753266176] tot_loss[loss=2.262, over 5657068.62 frames. , ppl: 9.605980476333988], batch size: 70
+2022-12-14 11:01:20,613 INFO [train.py:421] (5/8) Epoch 11, batch 70800, loss[loss=4.065, over 350.00 frames. , ppl: 58.24782091011647] tot_loss[loss=2.262, over 5629243.93 frames. , ppl: 9.605623223899364], batch size: 70
+2022-12-14 11:03:04,108 INFO [train.py:421] (5/8) Epoch 11, batch 71000, loss[loss=3.191, over 490.00 frames. , ppl: 24.307723870048456] tot_loss[loss=2.262, over 5674344.73 frames. , ppl: 9.598519993181363], batch size: 70
+2022-12-14 11:03:04,109 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 11:03:04,885 INFO [train.py:452] (5/8) Epoch 11, validation: loss=2.259, over 211138.00 frames. , ppl: 9.57697205031597
+2022-12-14 11:04:44,389 INFO [train.py:421] (5/8) Epoch 11, batch 71200, loss[loss=2.174, over 6300.00 frames. , ppl: 8.79404462988408] tot_loss[loss=2.261, over 5673079.32 frames. , ppl: 9.594824433849988], batch size: 70
+2022-12-14 11:06:25,521 INFO [train.py:421] (5/8) Epoch 11, batch 71400, loss[loss=2.329, over 3080.00 frames. , ppl: 10.270079142703393] tot_loss[loss=2.262, over 5669825.44 frames. , ppl: 9.600881204876611], batch size: 70
+2022-12-14 11:08:07,114 INFO [train.py:421] (5/8) Epoch 11, batch 71600, loss[loss=2.316, over 1540.00 frames. , ppl: 10.133243172259718] tot_loss[loss=2.263, over 5626535.68 frames. , ppl: 9.614396281728787], batch size: 70
+2022-12-14 11:09:48,375 INFO [train.py:421] (5/8) Epoch 11, batch 71800, loss[loss=2.747, over 630.00 frames. , ppl: 15.602057690274679] tot_loss[loss=2.264, over 5583473.58 frames. , ppl: 9.623832119124337], batch size: 70
+2022-12-14 11:11:04,470 INFO [train.py:421] (5/8) Epoch 12, batch 0, loss[loss=2.239, over 6580.00 frames. , ppl: 9.386048826410322] tot_loss[loss=2.239, over 6580.00 frames. , ppl: 9.386048826410322], batch size: 70
+2022-12-14 11:12:45,932 INFO [train.py:421] (5/8) Epoch 12, batch 200, loss[loss=2.199, over 2800.00 frames. , ppl: 9.015776655951306] tot_loss[loss=2.251, over 523545.41 frames. , ppl: 9.494058956430669], batch size: 70
+2022-12-14 11:14:25,216 INFO [train.py:421] (5/8) Epoch 12, batch 400, loss[loss=2.325, over 2030.00 frames. , ppl: 10.2226909867951] tot_loss[loss=2.249, over 1001072.44 frames. , ppl: 9.481820502691583], batch size: 70
+2022-12-14 11:16:08,524 INFO [train.py:421] (5/8) Epoch 12, batch 600, loss[loss=2.324, over 2170.00 frames. , ppl: 10.217668152016376] tot_loss[loss=2.249, over 1454970.36 frames. , ppl: 9.474122542435973], batch size: 70
+2022-12-14 11:17:49,309 INFO [train.py:421] (5/8) Epoch 12, batch 800, loss[loss=2.331, over 1890.00 frames. , ppl: 10.292647563045387] tot_loss[loss=2.252, over 1821652.61 frames. , ppl: 9.511012923963031], batch size: 70
+2022-12-14 11:19:30,205 INFO [train.py:421] (5/8) Epoch 12, batch 1000, loss[loss=2.334, over 1820.00 frames. , ppl: 10.318476089955757] tot_loss[loss=2.255, over 2138846.51 frames. , ppl: 9.539279717818], batch size: 70
+2022-12-14 11:19:30,206 INFO [train.py:441] (5/8) Computing validation loss
+2022-12-14 11:19:30,968 INFO [train.py:452] (5/8) Epoch 12, validation: loss=2.26, over 211138.00 frames. , ppl: 9.584870576501038
+2022-12-14 11:21:11,025 INFO [train.py:421] (5/8) Epoch 12, batch 1200, loss[loss=2.151, over 5250.00 frames. , ppl: 8.593489110108163] tot_loss[loss=2.254, over 2499459.29 frames. , ppl: 9.529075898150603], batch size: 70