INFO:Pipeline: Preparing Repository... WARNING:huggingface_hub.repository:Cloning https://huggingface.co/benlipkin/gpt2_1024_wikitext_100M_20_e12e6d4615e6a1e5 into local empty directory. INFO:Pipeline: Preparing Dataset... INFO:Pipeline: Preparing Optimizer... INFO:Pipeline: Preparing Accelerator... INFO:Pipeline: Initializing Training Loop... INFO:Pipeline: Total Steps: 122080 INFO:Pipeline: Eval Loss: 11.04, Perplexity: 62102.11 INFO:Pipeline: Time: 00:00:44, Epoch: 1, Step: 100, Step Loss: 10.31 INFO:Pipeline: Time: 00:01:21, Epoch: 1, Step: 200, Step Loss: 9.58 INFO:Pipeline: Time: 00:01:59, Epoch: 1, Step: 300, Step Loss: 9.17 INFO:Pipeline: Time: 00:02:36, Epoch: 1, Step: 400, Step Loss: 8.69 INFO:Pipeline: Time: 00:03:13, Epoch: 1, Step: 500, Step Loss: 8.56 INFO:Pipeline: Time: 00:03:50, Epoch: 1, Step: 600, Step Loss: 8.42 INFO:Pipeline: Time: 00:04:27, Epoch: 1, Step: 700, Step Loss: 8.02 INFO:Pipeline: Time: 00:05:05, Epoch: 1, Step: 800, Step Loss: 7.96 INFO:Pipeline: Time: 00:05:42, Epoch: 1, Step: 900, Step Loss: 7.71 INFO:Pipeline: Time: 00:06:19, Epoch: 1, Step: 1000, Step Loss: 7.48 INFO:Pipeline: Eval Loss: 7.49, Perplexity: 1788.48 INFO:Pipeline: Time: 00:07:00, Epoch: 1, Step: 1100, Step Loss: 7.39 INFO:Pipeline: Time: 00:07:37, Epoch: 1, Step: 1200, Step Loss: 7.27 INFO:Pipeline: Time: 00:08:14, Epoch: 1, Step: 1300, Step Loss: 7.07 INFO:Pipeline: Time: 00:08:52, Epoch: 1, Step: 1400, Step Loss: 6.99 INFO:Pipeline: Time: 00:09:29, Epoch: 1, Step: 1500, Step Loss: 6.76 INFO:Pipeline: Time: 00:10:06, Epoch: 1, Step: 1600, Step Loss: 6.75 INFO:Pipeline: Time: 00:10:44, Epoch: 1, Step: 1700, Step Loss: 6.71 INFO:Pipeline: Time: 00:11:21, Epoch: 1, Step: 1800, Step Loss: 6.57 INFO:Pipeline: Time: 00:11:58, Epoch: 1, Step: 1900, Step Loss: 6.61 INFO:Pipeline: Time: 00:12:36, Epoch: 1, Step: 2000, Step Loss: 6.59 INFO:Pipeline: Eval Loss: 6.42, Perplexity: 614.90 INFO:Pipeline: Time: 00:13:16, Epoch: 1, Step: 2100, Step Loss: 6.58 INFO:Pipeline: Time: 00:13:54, Epoch: 1, Step: 2200, Step Loss: 6.53 INFO:Pipeline: Time: 00:14:31, Epoch: 1, Step: 2300, Step Loss: 6.37 INFO:Pipeline: Time: 00:15:08, Epoch: 1, Step: 2400, Step Loss: 6.42 INFO:Pipeline: Time: 00:15:46, Epoch: 1, Step: 2500, Step Loss: 6.34 INFO:Pipeline: Time: 00:16:23, Epoch: 1, Step: 2600, Step Loss: 6.35 INFO:Pipeline: Time: 00:17:00, Epoch: 1, Step: 2700, Step Loss: 6.28 INFO:Pipeline: Time: 00:17:38, Epoch: 1, Step: 2800, Step Loss: 6.25 INFO:Pipeline: Time: 00:18:15, Epoch: 1, Step: 2900, Step Loss: 6.37 INFO:Pipeline: Time: 00:18:53, Epoch: 1, Step: 3000, Step Loss: 6.26 INFO:Pipeline: Eval Loss: 6.06, Perplexity: 430.52 INFO:Pipeline: Time: 00:19:33, Epoch: 1, Step: 3100, Step Loss: 6.23 INFO:Pipeline: Time: 00:20:10, Epoch: 1, Step: 3200, Step Loss: 6.18 INFO:Pipeline: Time: 00:20:48, Epoch: 1, Step: 3300, Step Loss: 6.02 INFO:Pipeline: Time: 00:21:25, Epoch: 1, Step: 3400, Step Loss: 6.06 INFO:Pipeline: Time: 00:22:02, Epoch: 1, Step: 3500, Step Loss: 6.00 INFO:Pipeline: Time: 00:22:40, Epoch: 1, Step: 3600, Step Loss: 5.99 INFO:Pipeline: Time: 00:23:17, Epoch: 1, Step: 3700, Step Loss: 6.04 INFO:Pipeline: Time: 00:23:55, Epoch: 1, Step: 3800, Step Loss: 6.08 INFO:Pipeline: Time: 00:24:32, Epoch: 1, Step: 3900, Step Loss: 5.90 INFO:Pipeline: Time: 00:25:09, Epoch: 1, Step: 4000, Step Loss: 5.96 INFO:Pipeline: Eval Loss: 5.79, Perplexity: 326.31 INFO:Pipeline: Time: 00:25:50, Epoch: 1, Step: 4100, Step Loss: 5.82 INFO:Pipeline: Time: 00:26:27, Epoch: 1, Step: 4200, Step Loss: 5.82 INFO:Pipeline: Time: 00:27:04, Epoch: 1, Step: 4300, Step Loss: 5.99 INFO:Pipeline: Time: 00:27:42, Epoch: 1, Step: 4400, Step Loss: 5.93 INFO:Pipeline: Time: 00:28:19, Epoch: 1, Step: 4500, Step Loss: 5.85 INFO:Pipeline: Time: 00:28:57, Epoch: 1, Step: 4600, Step Loss: 5.89 INFO:Pipeline: Time: 00:29:34, Epoch: 1, Step: 4700, Step Loss: 5.74 INFO:Pipeline: Time: 00:30:11, Epoch: 1, Step: 4800, Step Loss: 5.76 INFO:Pipeline: Time: 00:30:49, Epoch: 1, Step: 4900, Step Loss: 5.65 INFO:Pipeline: Time: 00:31:26, Epoch: 1, Step: 5000, Step Loss: 5.87 INFO:Pipeline: Eval Loss: 5.57, Perplexity: 262.09 INFO:Pipeline: Time: 00:32:06, Epoch: 1, Step: 5100, Step Loss: 5.71 INFO:Pipeline: Time: 00:32:44, Epoch: 1, Step: 5200, Step Loss: 5.62 INFO:Pipeline: Time: 00:33:21, Epoch: 1, Step: 5300, Step Loss: 5.73 INFO:Pipeline: Time: 00:33:59, Epoch: 1, Step: 5400, Step Loss: 5.70 INFO:Pipeline: Time: 00:34:36, Epoch: 1, Step: 5500, Step Loss: 5.74 INFO:Pipeline: Time: 00:35:13, Epoch: 1, Step: 5600, Step Loss: 5.64 INFO:Pipeline: Time: 00:35:51, Epoch: 1, Step: 5700, Step Loss: 5.73 INFO:Pipeline: Time: 00:36:28, Epoch: 1, Step: 5800, Step Loss: 5.64 INFO:Pipeline: Time: 00:37:05, Epoch: 1, Step: 5900, Step Loss: 5.60 INFO:Pipeline: Time: 00:37:43, Epoch: 1, Step: 6000, Step Loss: 5.75 INFO:Pipeline: Eval Loss: 5.40, Perplexity: 221.36 INFO:Pipeline: Time: 00:38:23, Epoch: 1, Step: 6100, Step Loss: 5.64 INFO:Pipeline: Time: 00:39:00, Epoch: 2, Step: 6200, Step Loss: 5.57 INFO:Pipeline: Time: 00:39:38, Epoch: 2, Step: 6300, Step Loss: 5.49 INFO:Pipeline: Time: 00:40:15, Epoch: 2, Step: 6400, Step Loss: 5.52 INFO:Pipeline: Time: 00:40:53, Epoch: 2, Step: 6500, Step Loss: 5.58 INFO:Pipeline: Time: 00:41:30, Epoch: 2, Step: 6600, Step Loss: 5.50 INFO:Pipeline: Time: 00:42:08, Epoch: 2, Step: 6700, Step Loss: 5.68 INFO:Pipeline: Time: 00:42:45, Epoch: 2, Step: 6800, Step Loss: 5.34 INFO:Pipeline: Time: 00:43:22, Epoch: 2, Step: 6900, Step Loss: 5.42 INFO:Pipeline: Time: 00:44:00, Epoch: 2, Step: 7000, Step Loss: 5.35 INFO:Pipeline: Eval Loss: 5.25, Perplexity: 190.40 INFO:Pipeline: Time: 00:44:40, Epoch: 2, Step: 7100, Step Loss: 5.31 INFO:Pipeline: Time: 00:45:18, Epoch: 2, Step: 7200, Step Loss: 5.59 INFO:Pipeline: Time: 00:45:55, Epoch: 2, Step: 7300, Step Loss: 5.40 INFO:Pipeline: Time: 00:46:32, Epoch: 2, Step: 7400, Step Loss: 5.32 INFO:Pipeline: Time: 00:47:10, Epoch: 2, Step: 7500, Step Loss: 5.34 INFO:Pipeline: Time: 00:47:47, Epoch: 2, Step: 7600, Step Loss: 5.33 INFO:Pipeline: Time: 00:48:25, Epoch: 2, Step: 7700, Step Loss: 5.45 INFO:Pipeline: Time: 00:49:02, Epoch: 2, Step: 7800, Step Loss: 5.38 INFO:Pipeline: Time: 00:49:39, Epoch: 2, Step: 7900, Step Loss: 5.17 INFO:Pipeline: Time: 00:50:17, Epoch: 2, Step: 8000, Step Loss: 5.24 INFO:Pipeline: Eval Loss: 5.11, Perplexity: 166.31 INFO:Pipeline: Time: 00:50:57, Epoch: 2, Step: 8100, Step Loss: 5.36 INFO:Pipeline: Time: 00:51:35, Epoch: 2, Step: 8200, Step Loss: 5.23 INFO:Pipeline: Time: 00:52:12, Epoch: 2, Step: 8300, Step Loss: 5.36 INFO:Pipeline: Time: 00:52:49, Epoch: 2, Step: 8400, Step Loss: 5.31 INFO:Pipeline: Time: 00:53:27, Epoch: 2, Step: 8500, Step Loss: 5.16 INFO:Pipeline: Time: 00:54:04, Epoch: 2, Step: 8600, Step Loss: 5.15 INFO:Pipeline: Time: 00:54:41, Epoch: 2, Step: 8700, Step Loss: 5.23 INFO:Pipeline: Time: 00:55:19, Epoch: 2, Step: 8800, Step Loss: 5.21 INFO:Pipeline: Time: 00:55:56, Epoch: 2, Step: 8900, Step Loss: 5.15 INFO:Pipeline: Time: 00:56:34, Epoch: 2, Step: 9000, Step Loss: 5.16 INFO:Pipeline: Eval Loss: 5.00, Perplexity: 148.88 INFO:Pipeline: Time: 00:57:14, Epoch: 2, Step: 9100, Step Loss: 5.11 INFO:Pipeline: Time: 00:57:51, Epoch: 2, Step: 9200, Step Loss: 5.17 INFO:Pipeline: Time: 00:58:29, Epoch: 2, Step: 9300, Step Loss: 5.22 INFO:Pipeline: Time: 00:59:06, Epoch: 2, Step: 9400, Step Loss: 5.09 INFO:Pipeline: Time: 00:59:44, Epoch: 2, Step: 9500, Step Loss: 5.15 INFO:Pipeline: Time: 01:00:21, Epoch: 2, Step: 9600, Step Loss: 5.06 INFO:Pipeline: Time: 01:00:58, Epoch: 2, Step: 9700, Step Loss: 5.17 INFO:Pipeline: Time: 01:01:36, Epoch: 2, Step: 9800, Step Loss: 5.26 INFO:Pipeline: Time: 01:02:13, Epoch: 2, Step: 9900, Step Loss: 5.21 INFO:Pipeline: Time: 01:02:50, Epoch: 2, Step: 10000, Step Loss: 5.12 INFO:Pipeline: Eval Loss: 4.90, Perplexity: 134.11 INFO:Pipeline: Time: 01:03:31, Epoch: 2, Step: 10100, Step Loss: 5.06 INFO:Pipeline: Time: 01:04:08, Epoch: 2, Step: 10200, Step Loss: 5.15 INFO:Pipeline: Time: 01:04:45, Epoch: 2, Step: 10300, Step Loss: 5.01 INFO:Pipeline: Time: 01:05:23, Epoch: 2, Step: 10400, Step Loss: 5.11 INFO:Pipeline: Time: 01:06:00, Epoch: 2, Step: 10500, Step Loss: 5.08 INFO:Pipeline: Time: 01:06:37, Epoch: 2, Step: 10600, Step Loss: 5.17 INFO:Pipeline: Time: 01:07:15, Epoch: 2, Step: 10700, Step Loss: 5.12 INFO:Pipeline: Time: 01:07:52, Epoch: 2, Step: 10800, Step Loss: 5.08 INFO:Pipeline: Time: 01:08:29, Epoch: 2, Step: 10900, Step Loss: 5.17 INFO:Pipeline: Time: 01:09:07, Epoch: 2, Step: 11000, Step Loss: 5.15 INFO:Pipeline: Eval Loss: 4.80, Perplexity: 121.93 INFO:Pipeline: Time: 01:09:47, Epoch: 2, Step: 11100, Step Loss: 4.99 INFO:Pipeline: Time: 01:10:24, Epoch: 2, Step: 11200, Step Loss: 4.96 INFO:Pipeline: Time: 01:11:02, Epoch: 2, Step: 11300, Step Loss: 5.02 INFO:Pipeline: Time: 01:11:39, Epoch: 2, Step: 11400, Step Loss: 5.04 INFO:Pipeline: Time: 01:12:16, Epoch: 2, Step: 11500, Step Loss: 4.99 INFO:Pipeline: Time: 01:12:54, Epoch: 2, Step: 11600, Step Loss: 4.96 INFO:Pipeline: Time: 01:13:31, Epoch: 2, Step: 11700, Step Loss: 4.95 INFO:Pipeline: Time: 01:14:08, Epoch: 2, Step: 11800, Step Loss: 4.95 INFO:Pipeline: Time: 01:14:46, Epoch: 2, Step: 11900, Step Loss: 4.95 INFO:Pipeline: Time: 01:15:23, Epoch: 2, Step: 12000, Step Loss: 5.02 INFO:Pipeline: Eval Loss: 4.72, Perplexity: 112.62 INFO:Pipeline: Time: 01:16:03, Epoch: 2, Step: 12100, Step Loss: 5.10 INFO:Pipeline: Time: 01:16:41, Epoch: 2, Step: 12200, Step Loss: 4.90 INFO:Pipeline: Time: 01:17:18, Epoch: 3, Step: 12300, Step Loss: 4.91 INFO:Pipeline: Time: 01:17:55, Epoch: 3, Step: 12400, Step Loss: 4.92 INFO:Pipeline: Time: 01:18:33, Epoch: 3, Step: 12500, Step Loss: 5.05 INFO:Pipeline: Time: 01:19:10, Epoch: 3, Step: 12600, Step Loss: 4.89 INFO:Pipeline: Time: 01:19:47, Epoch: 3, Step: 12700, Step Loss: 4.94 INFO:Pipeline: Time: 01:20:25, Epoch: 3, Step: 12800, Step Loss: 4.87 INFO:Pipeline: Time: 01:21:02, Epoch: 3, Step: 12900, Step Loss: 5.08 INFO:Pipeline: Time: 01:21:39, Epoch: 3, Step: 13000, Step Loss: 4.60 INFO:Pipeline: Eval Loss: 4.65, Perplexity: 104.63 INFO:Pipeline: Time: 01:22:20, Epoch: 3, Step: 13100, Step Loss: 4.78 INFO:Pipeline: Time: 01:22:57, Epoch: 3, Step: 13200, Step Loss: 4.87 INFO:Pipeline: Time: 01:23:34, Epoch: 3, Step: 13300, Step Loss: 4.85 INFO:Pipeline: Time: 01:24:12, Epoch: 3, Step: 13400, Step Loss: 5.01 INFO:Pipeline: Time: 01:24:49, Epoch: 3, Step: 13500, Step Loss: 4.79 INFO:Pipeline: Time: 01:25:26, Epoch: 3, Step: 13600, Step Loss: 4.71 INFO:Pipeline: Time: 01:26:04, Epoch: 3, Step: 13700, Step Loss: 4.79 INFO:Pipeline: Time: 01:26:41, Epoch: 3, Step: 13800, Step Loss: 4.76 INFO:Pipeline: Time: 01:27:18, Epoch: 3, Step: 13900, Step Loss: 4.87 INFO:Pipeline: Time: 01:27:56, Epoch: 3, Step: 14000, Step Loss: 4.88 INFO:Pipeline: Eval Loss: 4.58, Perplexity: 97.66 INFO:Pipeline: Time: 01:28:36, Epoch: 3, Step: 14100, Step Loss: 4.83 INFO:Pipeline: Time: 01:29:13, Epoch: 3, Step: 14200, Step Loss: 4.69 INFO:Pipeline: Time: 01:29:51, Epoch: 3, Step: 14300, Step Loss: 4.70 INFO:Pipeline: Time: 01:30:28, Epoch: 3, Step: 14400, Step Loss: 4.81 INFO:Pipeline: Time: 01:31:05, Epoch: 3, Step: 14500, Step Loss: 4.69 INFO:Pipeline: Time: 01:31:43, Epoch: 3, Step: 14600, Step Loss: 4.90 INFO:Pipeline: Time: 01:32:20, Epoch: 3, Step: 14700, Step Loss: 4.78 INFO:Pipeline: Time: 01:32:57, Epoch: 3, Step: 14800, Step Loss: 4.81 INFO:Pipeline: Time: 01:33:35, Epoch: 3, Step: 14900, Step Loss: 4.77 INFO:Pipeline: Time: 01:34:12, Epoch: 3, Step: 15000, Step Loss: 4.76 INFO:Pipeline: Eval Loss: 4.52, Perplexity: 91.90 INFO:Pipeline: Time: 01:34:52, Epoch: 3, Step: 15100, Step Loss: 4.78 INFO:Pipeline: Time: 01:35:29, Epoch: 3, Step: 15200, Step Loss: 4.76 INFO:Pipeline: Time: 01:36:07, Epoch: 3, Step: 15300, Step Loss: 4.69 INFO:Pipeline: Time: 01:36:44, Epoch: 3, Step: 15400, Step Loss: 4.73 INFO:Pipeline: Time: 01:37:21, Epoch: 3, Step: 15500, Step Loss: 4.63 INFO:Pipeline: Time: 01:37:59, Epoch: 3, Step: 15600, Step Loss: 4.76 INFO:Pipeline: Time: 01:38:36, Epoch: 3, Step: 15700, Step Loss: 4.81 INFO:Pipeline: Time: 01:39:13, Epoch: 3, Step: 15800, Step Loss: 4.78 INFO:Pipeline: Time: 01:39:50, Epoch: 3, Step: 15900, Step Loss: 4.79 INFO:Pipeline: Time: 01:40:27, Epoch: 3, Step: 16000, Step Loss: 4.79 INFO:Pipeline: Eval Loss: 4.46, Perplexity: 86.53 INFO:Pipeline: Time: 01:41:08, Epoch: 3, Step: 16100, Step Loss: 4.62 INFO:Pipeline: Time: 01:41:45, Epoch: 3, Step: 16200, Step Loss: 4.54 INFO:Pipeline: Time: 01:42:22, Epoch: 3, Step: 16300, Step Loss: 4.48 INFO:Pipeline: Time: 01:43:00, Epoch: 3, Step: 16400, Step Loss: 4.72 INFO:Pipeline: Time: 01:43:37, Epoch: 3, Step: 16500, Step Loss: 4.75 INFO:Pipeline: Time: 01:44:14, Epoch: 3, Step: 16600, Step Loss: 4.74 INFO:Pipeline: Time: 01:44:52, Epoch: 3, Step: 16700, Step Loss: 4.64 INFO:Pipeline: Time: 01:45:29, Epoch: 3, Step: 16800, Step Loss: 4.62 INFO:Pipeline: Time: 01:46:06, Epoch: 3, Step: 16900, Step Loss: 4.57 INFO:Pipeline: Time: 01:46:44, Epoch: 3, Step: 17000, Step Loss: 4.63 INFO:Pipeline: Eval Loss: 4.40, Perplexity: 81.76 INFO:Pipeline: Time: 01:47:24, Epoch: 3, Step: 17100, Step Loss: 4.52 INFO:Pipeline: Time: 01:48:02, Epoch: 3, Step: 17200, Step Loss: 4.52 INFO:Pipeline: Time: 01:48:39, Epoch: 3, Step: 17300, Step Loss: 4.63 INFO:Pipeline: Time: 01:49:16, Epoch: 3, Step: 17400, Step Loss: 4.44 INFO:Pipeline: Time: 01:49:54, Epoch: 3, Step: 17500, Step Loss: 4.65 INFO:Pipeline: Time: 01:50:31, Epoch: 3, Step: 17600, Step Loss: 4.57 INFO:Pipeline: Time: 01:51:09, Epoch: 3, Step: 17700, Step Loss: 4.64 INFO:Pipeline: Time: 01:51:46, Epoch: 3, Step: 17800, Step Loss: 4.55 INFO:Pipeline: Time: 01:52:23, Epoch: 3, Step: 17900, Step Loss: 4.55 INFO:Pipeline: Time: 01:53:01, Epoch: 3, Step: 18000, Step Loss: 4.63 INFO:Pipeline: Eval Loss: 4.35, Perplexity: 77.74 INFO:Pipeline: Time: 01:53:41, Epoch: 3, Step: 18100, Step Loss: 4.50 INFO:Pipeline: Time: 01:54:19, Epoch: 3, Step: 18200, Step Loss: 4.58 INFO:Pipeline: Time: 01:54:56, Epoch: 3, Step: 18300, Step Loss: 4.73 INFO:Pipeline: Time: 01:55:33, Epoch: 4, Step: 18400, Step Loss: 4.42 INFO:Pipeline: Time: 01:56:11, Epoch: 4, Step: 18500, Step Loss: 4.41 INFO:Pipeline: Time: 01:56:48, Epoch: 4, Step: 18600, Step Loss: 4.54 INFO:Pipeline: Time: 01:57:26, Epoch: 4, Step: 18700, Step Loss: 4.59 INFO:Pipeline: Time: 01:58:03, Epoch: 4, Step: 18800, Step Loss: 4.57 INFO:Pipeline: Time: 01:58:40, Epoch: 4, Step: 18900, Step Loss: 4.40 INFO:Pipeline: Time: 01:59:18, Epoch: 4, Step: 19000, Step Loss: 4.59 INFO:Pipeline: Eval Loss: 4.31, Perplexity: 74.25 INFO:Pipeline: Time: 01:59:58, Epoch: 4, Step: 19100, Step Loss: 4.53 INFO:Pipeline: Time: 02:00:36, Epoch: 4, Step: 19200, Step Loss: 4.30 INFO:Pipeline: Time: 02:01:13, Epoch: 4, Step: 19300, Step Loss: 4.52 INFO:Pipeline: Time: 02:01:50, Epoch: 4, Step: 19400, Step Loss: 4.50 INFO:Pipeline: Time: 02:02:28, Epoch: 4, Step: 19500, Step Loss: 4.71 INFO:Pipeline: Time: 02:03:05, Epoch: 4, Step: 19600, Step Loss: 4.32 INFO:Pipeline: Time: 02:03:43, Epoch: 4, Step: 19700, Step Loss: 4.39 INFO:Pipeline: Time: 02:04:20, Epoch: 4, Step: 19800, Step Loss: 4.38 INFO:Pipeline: Time: 02:04:57, Epoch: 4, Step: 19900, Step Loss: 4.44 INFO:Pipeline: Time: 02:05:35, Epoch: 4, Step: 20000, Step Loss: 4.61 INFO:Pipeline: Eval Loss: 4.26, Perplexity: 70.71 INFO:Pipeline: Time: 02:06:15, Epoch: 4, Step: 20100, Step Loss: 4.66 INFO:Pipeline: Time: 02:06:52, Epoch: 4, Step: 20200, Step Loss: 4.54 INFO:Pipeline: Time: 02:07:30, Epoch: 4, Step: 20300, Step Loss: 4.48 INFO:Pipeline: Time: 02:08:07, Epoch: 4, Step: 20400, Step Loss: 4.47 INFO:Pipeline: Time: 02:08:44, Epoch: 4, Step: 20500, Step Loss: 4.46 INFO:Pipeline: Time: 02:09:22, Epoch: 4, Step: 20600, Step Loss: 4.50 INFO:Pipeline: Time: 02:09:59, Epoch: 4, Step: 20700, Step Loss: 4.50 INFO:Pipeline: Time: 02:10:36, Epoch: 4, Step: 20800, Step Loss: 4.52 INFO:Pipeline: Time: 02:11:14, Epoch: 4, Step: 20900, Step Loss: 4.41 INFO:Pipeline: Time: 02:11:51, Epoch: 4, Step: 21000, Step Loss: 4.49 INFO:Pipeline: Eval Loss: 4.22, Perplexity: 67.70 INFO:Pipeline: Time: 02:12:32, Epoch: 4, Step: 21100, Step Loss: 4.37 INFO:Pipeline: Time: 02:13:09, Epoch: 4, Step: 21200, Step Loss: 4.35 INFO:Pipeline: Time: 02:13:46, Epoch: 4, Step: 21300, Step Loss: 4.15 INFO:Pipeline: Time: 02:14:24, Epoch: 4, Step: 21400, Step Loss: 4.50 INFO:Pipeline: Time: 02:15:01, Epoch: 4, Step: 21500, Step Loss: 4.47 INFO:Pipeline: Time: 02:15:38, Epoch: 4, Step: 21600, Step Loss: 4.40 INFO:Pipeline: Time: 02:16:16, Epoch: 4, Step: 21700, Step Loss: 4.38 INFO:Pipeline: Time: 02:16:53, Epoch: 4, Step: 21800, Step Loss: 4.37 INFO:Pipeline: Time: 02:17:30, Epoch: 4, Step: 21900, Step Loss: 4.29 INFO:Pipeline: Time: 02:18:07, Epoch: 4, Step: 22000, Step Loss: 4.56 INFO:Pipeline: Eval Loss: 4.17, Perplexity: 64.83 INFO:Pipeline: Time: 02:18:47, Epoch: 4, Step: 22100, Step Loss: 4.40 INFO:Pipeline: Time: 02:19:25, Epoch: 4, Step: 22200, Step Loss: 4.41 INFO:Pipeline: Time: 02:20:02, Epoch: 4, Step: 22300, Step Loss: 4.37 INFO:Pipeline: Time: 02:20:39, Epoch: 4, Step: 22400, Step Loss: 4.50 INFO:Pipeline: Time: 02:21:17, Epoch: 4, Step: 22500, Step Loss: 4.36 INFO:Pipeline: Time: 02:21:54, Epoch: 4, Step: 22600, Step Loss: 4.43 INFO:Pipeline: Time: 02:22:32, Epoch: 4, Step: 22700, Step Loss: 4.47 INFO:Pipeline: Time: 02:23:09, Epoch: 4, Step: 22800, Step Loss: 4.14 INFO:Pipeline: Time: 02:23:46, Epoch: 4, Step: 22900, Step Loss: 4.28 INFO:Pipeline: Time: 02:24:24, Epoch: 4, Step: 23000, Step Loss: 4.34 INFO:Pipeline: Eval Loss: 4.13, Perplexity: 62.21 INFO:Pipeline: Time: 02:25:04, Epoch: 4, Step: 23100, Step Loss: 4.32 INFO:Pipeline: Time: 02:25:42, Epoch: 4, Step: 23200, Step Loss: 4.40 INFO:Pipeline: Time: 02:26:19, Epoch: 4, Step: 23300, Step Loss: 4.40 INFO:Pipeline: Time: 02:26:56, Epoch: 4, Step: 23400, Step Loss: 4.24 INFO:Pipeline: Time: 02:27:34, Epoch: 4, Step: 23500, Step Loss: 4.24 INFO:Pipeline: Time: 02:28:11, Epoch: 4, Step: 23600, Step Loss: 4.17 INFO:Pipeline: Time: 02:28:48, Epoch: 4, Step: 23700, Step Loss: 4.31 INFO:Pipeline: Time: 02:29:26, Epoch: 4, Step: 23800, Step Loss: 4.19 INFO:Pipeline: Time: 02:30:03, Epoch: 4, Step: 23900, Step Loss: 4.20 INFO:Pipeline: Time: 02:30:41, Epoch: 4, Step: 24000, Step Loss: 4.51 INFO:Pipeline: Eval Loss: 4.09, Perplexity: 59.76 INFO:Pipeline: Time: 02:31:21, Epoch: 4, Step: 24100, Step Loss: 4.10 INFO:Pipeline: Time: 02:31:59, Epoch: 4, Step: 24200, Step Loss: 4.12 INFO:Pipeline: Time: 02:32:36, Epoch: 4, Step: 24300, Step Loss: 4.19 INFO:Pipeline: Time: 02:33:13, Epoch: 4, Step: 24400, Step Loss: 4.30 INFO:Pipeline: Time: 02:33:51, Epoch: 5, Step: 24500, Step Loss: 4.27 INFO:Pipeline: Time: 02:34:28, Epoch: 5, Step: 24600, Step Loss: 4.33 INFO:Pipeline: Time: 02:35:05, Epoch: 5, Step: 24700, Step Loss: 4.27 INFO:Pipeline: Time: 02:35:43, Epoch: 5, Step: 24800, Step Loss: 4.33 INFO:Pipeline: Time: 02:36:20, Epoch: 5, Step: 24900, Step Loss: 4.28 INFO:Pipeline: Time: 02:36:58, Epoch: 5, Step: 25000, Step Loss: 4.23 INFO:Pipeline: Eval Loss: 4.05, Perplexity: 57.55 INFO:Pipeline: Time: 02:37:38, Epoch: 5, Step: 25100, Step Loss: 4.25 INFO:Pipeline: Time: 02:38:15, Epoch: 5, Step: 25200, Step Loss: 4.31 INFO:Pipeline: Time: 02:38:53, Epoch: 5, Step: 25300, Step Loss: 4.19 INFO:Pipeline: Time: 02:39:30, Epoch: 5, Step: 25400, Step Loss: 4.20 INFO:Pipeline: Time: 02:40:08, Epoch: 5, Step: 25500, Step Loss: 4.29 INFO:Pipeline: Time: 02:40:45, Epoch: 5, Step: 25600, Step Loss: 4.09 INFO:Pipeline: Time: 02:41:22, Epoch: 5, Step: 25700, Step Loss: 4.20 INFO:Pipeline: Time: 02:42:00, Epoch: 5, Step: 25800, Step Loss: 4.19 INFO:Pipeline: Time: 02:42:37, Epoch: 5, Step: 25900, Step Loss: 4.43 INFO:Pipeline: Time: 02:43:14, Epoch: 5, Step: 26000, Step Loss: 4.36 INFO:Pipeline: Eval Loss: 4.02, Perplexity: 55.66 INFO:Pipeline: Time: 02:43:55, Epoch: 5, Step: 26100, Step Loss: 4.10 INFO:Pipeline: Time: 02:44:32, Epoch: 5, Step: 26200, Step Loss: 4.25 INFO:Pipeline: Time: 02:45:10, Epoch: 5, Step: 26300, Step Loss: 4.15 INFO:Pipeline: Time: 02:45:47, Epoch: 5, Step: 26400, Step Loss: 4.10 INFO:Pipeline: Time: 02:46:24, Epoch: 5, Step: 26500, Step Loss: 4.12 INFO:Pipeline: Time: 02:47:02, Epoch: 5, Step: 26600, Step Loss: 4.18 INFO:Pipeline: Time: 02:47:39, Epoch: 5, Step: 26700, Step Loss: 4.25 INFO:Pipeline: Time: 02:48:16, Epoch: 5, Step: 26800, Step Loss: 4.25 INFO:Pipeline: Time: 02:48:54, Epoch: 5, Step: 26900, Step Loss: 4.30 INFO:Pipeline: Time: 02:49:31, Epoch: 5, Step: 27000, Step Loss: 4.02 INFO:Pipeline: Eval Loss: 3.99, Perplexity: 53.98 INFO:Pipeline: Time: 02:50:12, Epoch: 5, Step: 27100, Step Loss: 4.25 INFO:Pipeline: Time: 02:50:49, Epoch: 5, Step: 27200, Step Loss: 4.18 INFO:Pipeline: Time: 02:51:26, Epoch: 5, Step: 27300, Step Loss: 4.06 INFO:Pipeline: Time: 02:52:04, Epoch: 5, Step: 27400, Step Loss: 4.22 INFO:Pipeline: Time: 02:52:41, Epoch: 5, Step: 27500, Step Loss: 4.36 INFO:Pipeline: Time: 02:53:19, Epoch: 5, Step: 27600, Step Loss: 4.18 INFO:Pipeline: Time: 02:53:56, Epoch: 5, Step: 27700, Step Loss: 4.19 INFO:Pipeline: Time: 02:54:33, Epoch: 5, Step: 27800, Step Loss: 4.12 INFO:Pipeline: Time: 02:55:11, Epoch: 5, Step: 27900, Step Loss: 4.05 INFO:Pipeline: Time: 02:55:48, Epoch: 5, Step: 28000, Step Loss: 4.16 INFO:Pipeline: Eval Loss: 3.96, Perplexity: 52.42 INFO:Pipeline: Time: 02:56:28, Epoch: 5, Step: 28100, Step Loss: 4.15 INFO:Pipeline: Time: 02:57:06, Epoch: 5, Step: 28200, Step Loss: 4.12 INFO:Pipeline: Time: 02:57:43, Epoch: 5, Step: 28300, Step Loss: 4.03 INFO:Pipeline: Time: 02:58:21, Epoch: 5, Step: 28400, Step Loss: 4.00 INFO:Pipeline: Time: 02:58:58, Epoch: 5, Step: 28500, Step Loss: 4.09 INFO:Pipeline: Time: 02:59:35, Epoch: 5, Step: 28600, Step Loss: 4.03 INFO:Pipeline: Time: 03:00:13, Epoch: 5, Step: 28700, Step Loss: 4.07 INFO:Pipeline: Time: 03:00:50, Epoch: 5, Step: 28800, Step Loss: 4.04 INFO:Pipeline: Time: 03:01:28, Epoch: 5, Step: 28900, Step Loss: 4.17 INFO:Pipeline: Time: 03:02:05, Epoch: 5, Step: 29000, Step Loss: 4.16 INFO:Pipeline: Eval Loss: 3.93, Perplexity: 50.86 INFO:Pipeline: Time: 03:02:45, Epoch: 5, Step: 29100, Step Loss: 4.14 INFO:Pipeline: Time: 03:03:23, Epoch: 5, Step: 29200, Step Loss: 4.04 INFO:Pipeline: Time: 03:04:00, Epoch: 5, Step: 29300, Step Loss: 3.90 INFO:Pipeline: Time: 03:04:38, Epoch: 5, Step: 29400, Step Loss: 3.90 INFO:Pipeline: Time: 03:05:15, Epoch: 5, Step: 29500, Step Loss: 4.10 INFO:Pipeline: Time: 03:05:52, Epoch: 5, Step: 29600, Step Loss: 4.05 INFO:Pipeline: Time: 03:06:30, Epoch: 5, Step: 29700, Step Loss: 3.91 INFO:Pipeline: Time: 03:07:07, Epoch: 5, Step: 29800, Step Loss: 3.97 INFO:Pipeline: Time: 03:07:44, Epoch: 5, Step: 29900, Step Loss: 4.04 INFO:Pipeline: Time: 03:08:22, Epoch: 5, Step: 30000, Step Loss: 4.14 INFO:Pipeline: Eval Loss: 3.90, Perplexity: 49.58 INFO:Pipeline: Time: 03:09:02, Epoch: 5, Step: 30100, Step Loss: 3.92 INFO:Pipeline: Time: 03:09:40, Epoch: 5, Step: 30200, Step Loss: 4.12 INFO:Pipeline: Time: 03:10:17, Epoch: 5, Step: 30300, Step Loss: 4.20 INFO:Pipeline: Time: 03:10:54, Epoch: 5, Step: 30400, Step Loss: 4.16 INFO:Pipeline: Time: 03:11:32, Epoch: 5, Step: 30500, Step Loss: 3.96 INFO:Pipeline: Time: 03:12:09, Epoch: 6, Step: 30600, Step Loss: 4.11 INFO:Pipeline: Time: 03:12:46, Epoch: 6, Step: 30700, Step Loss: 4.23 INFO:Pipeline: Time: 03:13:24, Epoch: 6, Step: 30800, Step Loss: 3.89 INFO:Pipeline: Time: 03:14:01, Epoch: 6, Step: 30900, Step Loss: 4.05 INFO:Pipeline: Time: 03:14:38, Epoch: 6, Step: 31000, Step Loss: 4.12 INFO:Pipeline: Eval Loss: 3.88, Perplexity: 48.57 INFO:Pipeline: Time: 03:15:19, Epoch: 6, Step: 31100, Step Loss: 3.98 INFO:Pipeline: Time: 03:15:56, Epoch: 6, Step: 31200, Step Loss: 4.04 INFO:Pipeline: Time: 03:16:34, Epoch: 6, Step: 31300, Step Loss: 4.13 INFO:Pipeline: Time: 03:17:11, Epoch: 6, Step: 31400, Step Loss: 3.99 INFO:Pipeline: Time: 03:17:48, Epoch: 6, Step: 31500, Step Loss: 3.91 INFO:Pipeline: Time: 03:18:26, Epoch: 6, Step: 31600, Step Loss: 4.06 INFO:Pipeline: Time: 03:19:03, Epoch: 6, Step: 31700, Step Loss: 4.11 INFO:Pipeline: Time: 03:19:41, Epoch: 6, Step: 31800, Step Loss: 4.22 INFO:Pipeline: Time: 03:20:18, Epoch: 6, Step: 31900, Step Loss: 4.02 INFO:Pipeline: Time: 03:20:55, Epoch: 6, Step: 32000, Step Loss: 4.13 INFO:Pipeline: Eval Loss: 3.86, Perplexity: 47.28 INFO:Pipeline: Time: 03:21:36, Epoch: 6, Step: 32100, Step Loss: 3.99 INFO:Pipeline: Time: 03:22:13, Epoch: 6, Step: 32200, Step Loss: 4.09 INFO:Pipeline: Time: 03:22:50, Epoch: 6, Step: 32300, Step Loss: 4.10 INFO:Pipeline: Time: 03:23:28, Epoch: 6, Step: 32400, Step Loss: 4.04 INFO:Pipeline: Time: 03:24:05, Epoch: 6, Step: 32500, Step Loss: 4.01 INFO:Pipeline: Time: 03:24:43, Epoch: 6, Step: 32600, Step Loss: 4.14 INFO:Pipeline: Time: 03:25:20, Epoch: 6, Step: 32700, Step Loss: 3.89 INFO:Pipeline: Time: 03:25:57, Epoch: 6, Step: 32800, Step Loss: 4.15 INFO:Pipeline: Time: 03:26:35, Epoch: 6, Step: 32900, Step Loss: 4.23 INFO:Pipeline: Time: 03:27:12, Epoch: 6, Step: 33000, Step Loss: 3.97 INFO:Pipeline: Eval Loss: 3.83, Perplexity: 46.19 INFO:Pipeline: Time: 03:27:53, Epoch: 6, Step: 33100, Step Loss: 3.96 INFO:Pipeline: Time: 03:28:30, Epoch: 6, Step: 33200, Step Loss: 3.96 INFO:Pipeline: Time: 03:29:07, Epoch: 6, Step: 33300, Step Loss: 4.08 INFO:Pipeline: Time: 03:29:45, Epoch: 6, Step: 33400, Step Loss: 3.99 INFO:Pipeline: Time: 03:30:22, Epoch: 6, Step: 33500, Step Loss: 4.00 INFO:Pipeline: Time: 03:31:00, Epoch: 6, Step: 33600, Step Loss: 3.94 INFO:Pipeline: Time: 03:31:37, Epoch: 6, Step: 33700, Step Loss: 3.99 INFO:Pipeline: Time: 03:32:14, Epoch: 6, Step: 33800, Step Loss: 4.14 INFO:Pipeline: Time: 03:32:52, Epoch: 6, Step: 33900, Step Loss: 3.97 INFO:Pipeline: Time: 03:33:29, Epoch: 6, Step: 34000, Step Loss: 4.00 INFO:Pipeline: Eval Loss: 3.81, Perplexity: 45.21 INFO:Pipeline: Time: 03:34:09, Epoch: 6, Step: 34100, Step Loss: 4.03 INFO:Pipeline: Time: 03:34:47, Epoch: 6, Step: 34200, Step Loss: 4.13 INFO:Pipeline: Time: 03:35:24, Epoch: 6, Step: 34300, Step Loss: 3.94 INFO:Pipeline: Time: 03:36:01, Epoch: 6, Step: 34400, Step Loss: 3.90 INFO:Pipeline: Time: 03:36:39, Epoch: 6, Step: 34500, Step Loss: 3.94 INFO:Pipeline: Time: 03:37:16, Epoch: 6, Step: 34600, Step Loss: 4.04 INFO:Pipeline: Time: 03:37:54, Epoch: 6, Step: 34700, Step Loss: 4.13 INFO:Pipeline: Time: 03:38:31, Epoch: 6, Step: 34800, Step Loss: 3.82 INFO:Pipeline: Time: 03:39:08, Epoch: 6, Step: 34900, Step Loss: 4.01 INFO:Pipeline: Time: 03:39:46, Epoch: 6, Step: 35000, Step Loss: 4.01 INFO:Pipeline: Eval Loss: 3.79, Perplexity: 44.34 INFO:Pipeline: Time: 03:40:26, Epoch: 6, Step: 35100, Step Loss: 3.88 INFO:Pipeline: Time: 03:41:03, Epoch: 6, Step: 35200, Step Loss: 3.97 INFO:Pipeline: Time: 03:41:41, Epoch: 6, Step: 35300, Step Loss: 4.19 INFO:Pipeline: Time: 03:42:18, Epoch: 6, Step: 35400, Step Loss: 3.93 INFO:Pipeline: Time: 03:42:55, Epoch: 6, Step: 35500, Step Loss: 3.96 INFO:Pipeline: Time: 03:43:33, Epoch: 6, Step: 35600, Step Loss: 3.97 INFO:Pipeline: Time: 03:44:10, Epoch: 6, Step: 35700, Step Loss: 4.11 INFO:Pipeline: Time: 03:44:47, Epoch: 6, Step: 35800, Step Loss: 3.97 INFO:Pipeline: Time: 03:45:24, Epoch: 6, Step: 35900, Step Loss: 3.88 INFO:Pipeline: Time: 03:46:02, Epoch: 6, Step: 36000, Step Loss: 4.09 INFO:Pipeline: Eval Loss: 3.77, Perplexity: 43.53 INFO:Pipeline: Time: 03:46:42, Epoch: 6, Step: 36100, Step Loss: 3.85 INFO:Pipeline: Time: 03:47:19, Epoch: 6, Step: 36200, Step Loss: 3.71 INFO:Pipeline: Time: 03:47:57, Epoch: 6, Step: 36300, Step Loss: 4.05 INFO:Pipeline: Time: 03:48:34, Epoch: 6, Step: 36400, Step Loss: 3.89 INFO:Pipeline: Time: 03:49:11, Epoch: 6, Step: 36500, Step Loss: 3.91 INFO:Pipeline: Time: 03:49:48, Epoch: 6, Step: 36600, Step Loss: 3.89 INFO:Pipeline: Time: 03:50:26, Epoch: 7, Step: 36700, Step Loss: 4.02 INFO:Pipeline: Time: 03:51:03, Epoch: 7, Step: 36800, Step Loss: 3.82 INFO:Pipeline: Time: 03:51:40, Epoch: 7, Step: 36900, Step Loss: 3.94 INFO:Pipeline: Time: 03:52:17, Epoch: 7, Step: 37000, Step Loss: 3.77 INFO:Pipeline: Eval Loss: 3.76, Perplexity: 42.75 INFO:Pipeline: Time: 03:52:58, Epoch: 7, Step: 37100, Step Loss: 3.97 INFO:Pipeline: Time: 03:53:35, Epoch: 7, Step: 37200, Step Loss: 4.12 INFO:Pipeline: Time: 03:54:12, Epoch: 7, Step: 37300, Step Loss: 3.89 INFO:Pipeline: Time: 03:54:50, Epoch: 7, Step: 37400, Step Loss: 3.92 INFO:Pipeline: Time: 03:55:27, Epoch: 7, Step: 37500, Step Loss: 3.93 INFO:Pipeline: Time: 03:56:04, Epoch: 7, Step: 37600, Step Loss: 4.02 INFO:Pipeline: Time: 03:56:42, Epoch: 7, Step: 37700, Step Loss: 4.04 INFO:Pipeline: Time: 03:57:19, Epoch: 7, Step: 37800, Step Loss: 3.99 INFO:Pipeline: Time: 03:57:56, Epoch: 7, Step: 37900, Step Loss: 4.06 INFO:Pipeline: Time: 03:58:33, Epoch: 7, Step: 38000, Step Loss: 3.82 INFO:Pipeline: Eval Loss: 3.74, Perplexity: 41.96 INFO:Pipeline: Time: 03:59:14, Epoch: 7, Step: 38100, Step Loss: 4.10 INFO:Pipeline: Time: 03:59:51, Epoch: 7, Step: 38200, Step Loss: 3.92 INFO:Pipeline: Time: 04:00:28, Epoch: 7, Step: 38300, Step Loss: 3.94 INFO:Pipeline: Time: 04:01:06, Epoch: 7, Step: 38400, Step Loss: 3.89 INFO:Pipeline: Time: 04:01:43, Epoch: 7, Step: 38500, Step Loss: 3.98 INFO:Pipeline: Time: 04:02:20, Epoch: 7, Step: 38600, Step Loss: 3.89 INFO:Pipeline: Time: 04:02:57, Epoch: 7, Step: 38700, Step Loss: 3.98 INFO:Pipeline: Time: 04:03:35, Epoch: 7, Step: 38800, Step Loss: 3.89 INFO:Pipeline: Time: 04:04:12, Epoch: 7, Step: 38900, Step Loss: 3.84 INFO:Pipeline: Time: 04:04:49, Epoch: 7, Step: 39000, Step Loss: 3.83 INFO:Pipeline: Eval Loss: 3.72, Perplexity: 41.36 INFO:Pipeline: Time: 04:05:29, Epoch: 7, Step: 39100, Step Loss: 3.57 INFO:Pipeline: Time: 04:06:07, Epoch: 7, Step: 39200, Step Loss: 3.92 INFO:Pipeline: Time: 04:06:44, Epoch: 7, Step: 39300, Step Loss: 3.75 INFO:Pipeline: Time: 04:07:21, Epoch: 7, Step: 39400, Step Loss: 3.88 INFO:Pipeline: Time: 04:07:59, Epoch: 7, Step: 39500, Step Loss: 3.78 INFO:Pipeline: Time: 04:08:36, Epoch: 7, Step: 39600, Step Loss: 3.79 INFO:Pipeline: Time: 04:09:13, Epoch: 7, Step: 39700, Step Loss: 3.84 INFO:Pipeline: Time: 04:09:50, Epoch: 7, Step: 39800, Step Loss: 3.81 INFO:Pipeline: Time: 04:10:28, Epoch: 7, Step: 39900, Step Loss: 3.75 INFO:Pipeline: Time: 04:11:05, Epoch: 7, Step: 40000, Step Loss: 3.72 INFO:Pipeline: Eval Loss: 3.71, Perplexity: 40.69 INFO:Pipeline: Time: 04:11:45, Epoch: 7, Step: 40100, Step Loss: 3.86 INFO:Pipeline: Time: 04:12:23, Epoch: 7, Step: 40200, Step Loss: 3.98 INFO:Pipeline: Time: 04:13:00, Epoch: 7, Step: 40300, Step Loss: 4.05 INFO:Pipeline: Time: 04:13:37, Epoch: 7, Step: 40400, Step Loss: 3.85 INFO:Pipeline: Time: 04:14:15, Epoch: 7, Step: 40500, Step Loss: 3.73 INFO:Pipeline: Time: 04:14:52, Epoch: 7, Step: 40600, Step Loss: 3.75 INFO:Pipeline: Time: 04:15:29, Epoch: 7, Step: 40700, Step Loss: 3.85 INFO:Pipeline: Time: 04:16:06, Epoch: 7, Step: 40800, Step Loss: 3.81 INFO:Pipeline: Time: 04:16:44, Epoch: 7, Step: 40900, Step Loss: 3.88 INFO:Pipeline: Time: 04:17:21, Epoch: 7, Step: 41000, Step Loss: 3.79 INFO:Pipeline: Eval Loss: 3.69, Perplexity: 40.10 INFO:Pipeline: Time: 04:18:02, Epoch: 7, Step: 41100, Step Loss: 3.72 INFO:Pipeline: Time: 04:18:39, Epoch: 7, Step: 41200, Step Loss: 3.66 INFO:Pipeline: Time: 04:19:16, Epoch: 7, Step: 41300, Step Loss: 3.74 INFO:Pipeline: Time: 04:19:54, Epoch: 7, Step: 41400, Step Loss: 4.12 INFO:Pipeline: Time: 04:20:31, Epoch: 7, Step: 41500, Step Loss: 3.89 INFO:Pipeline: Time: 04:21:08, Epoch: 7, Step: 41600, Step Loss: 3.80 INFO:Pipeline: Time: 04:21:45, Epoch: 7, Step: 41700, Step Loss: 3.88 INFO:Pipeline: Time: 04:22:23, Epoch: 7, Step: 41800, Step Loss: 3.93 INFO:Pipeline: Time: 04:23:00, Epoch: 7, Step: 41900, Step Loss: 3.89 INFO:Pipeline: Time: 04:23:37, Epoch: 7, Step: 42000, Step Loss: 3.63 INFO:Pipeline: Eval Loss: 3.68, Perplexity: 39.62 INFO:Pipeline: Time: 04:24:18, Epoch: 7, Step: 42100, Step Loss: 3.75 INFO:Pipeline: Time: 04:24:55, Epoch: 7, Step: 42200, Step Loss: 3.92 INFO:Pipeline: Time: 04:25:32, Epoch: 7, Step: 42300, Step Loss: 3.90 INFO:Pipeline: Time: 04:26:10, Epoch: 7, Step: 42400, Step Loss: 3.73 INFO:Pipeline: Time: 04:26:47, Epoch: 7, Step: 42500, Step Loss: 3.85 INFO:Pipeline: Time: 04:27:24, Epoch: 7, Step: 42600, Step Loss: 3.89 INFO:Pipeline: Time: 04:28:02, Epoch: 7, Step: 42700, Step Loss: 3.94 INFO:Pipeline: Time: 04:28:39, Epoch: 8, Step: 42800, Step Loss: 3.77 INFO:Pipeline: Time: 04:29:16, Epoch: 8, Step: 42900, Step Loss: 3.81 INFO:Pipeline: Time: 04:29:53, Epoch: 8, Step: 43000, Step Loss: 3.96 INFO:Pipeline: Eval Loss: 3.66, Perplexity: 39.05 INFO:Pipeline: Time: 04:30:34, Epoch: 8, Step: 43100, Step Loss: 3.63 INFO:Pipeline: Time: 04:31:11, Epoch: 8, Step: 43200, Step Loss: 3.62 INFO:Pipeline: Time: 04:31:48, Epoch: 8, Step: 43300, Step Loss: 3.74 INFO:Pipeline: Time: 04:32:25, Epoch: 8, Step: 43400, Step Loss: 3.93 INFO:Pipeline: Time: 04:33:03, Epoch: 8, Step: 43500, Step Loss: 3.75 INFO:Pipeline: Time: 04:33:40, Epoch: 8, Step: 43600, Step Loss: 3.62 INFO:Pipeline: Time: 04:34:17, Epoch: 8, Step: 43700, Step Loss: 4.00 INFO:Pipeline: Time: 04:34:55, Epoch: 8, Step: 43800, Step Loss: 3.73 INFO:Pipeline: Time: 04:35:32, Epoch: 8, Step: 43900, Step Loss: 3.69 INFO:Pipeline: Time: 04:36:09, Epoch: 8, Step: 44000, Step Loss: 3.98 INFO:Pipeline: Eval Loss: 3.65, Perplexity: 38.54 INFO:Pipeline: Time: 04:36:50, Epoch: 8, Step: 44100, Step Loss: 3.77 INFO:Pipeline: Time: 04:37:27, Epoch: 8, Step: 44200, Step Loss: 3.74 INFO:Pipeline: Time: 04:38:04, Epoch: 8, Step: 44300, Step Loss: 3.66 INFO:Pipeline: Time: 04:38:42, Epoch: 8, Step: 44400, Step Loss: 3.76 INFO:Pipeline: Time: 04:39:19, Epoch: 8, Step: 44500, Step Loss: 3.87 INFO:Pipeline: Time: 04:39:56, Epoch: 8, Step: 44600, Step Loss: 3.82 INFO:Pipeline: Time: 04:40:34, Epoch: 8, Step: 44700, Step Loss: 3.82 INFO:Pipeline: Time: 04:41:11, Epoch: 8, Step: 44800, Step Loss: 3.75 INFO:Pipeline: Time: 04:41:49, Epoch: 8, Step: 44900, Step Loss: 3.86 INFO:Pipeline: Time: 04:42:26, Epoch: 8, Step: 45000, Step Loss: 3.64 INFO:Pipeline: Eval Loss: 3.64, Perplexity: 38.08 INFO:Pipeline: Time: 04:43:07, Epoch: 8, Step: 45100, Step Loss: 3.71 INFO:Pipeline: Time: 04:43:44, Epoch: 8, Step: 45200, Step Loss: 3.84 INFO:Pipeline: Time: 04:44:21, Epoch: 8, Step: 45300, Step Loss: 3.73 INFO:Pipeline: Time: 04:44:59, Epoch: 8, Step: 45400, Step Loss: 3.61 INFO:Pipeline: Time: 04:45:36, Epoch: 8, Step: 45500, Step Loss: 3.76 INFO:Pipeline: Time: 04:46:13, Epoch: 8, Step: 45600, Step Loss: 3.87 INFO:Pipeline: Time: 04:46:51, Epoch: 8, Step: 45700, Step Loss: 3.91 INFO:Pipeline: Time: 04:47:28, Epoch: 8, Step: 45800, Step Loss: 3.81 INFO:Pipeline: Time: 04:48:05, Epoch: 8, Step: 45900, Step Loss: 3.66 INFO:Pipeline: Time: 04:48:43, Epoch: 8, Step: 46000, Step Loss: 3.82 INFO:Pipeline: Eval Loss: 3.63, Perplexity: 37.62 INFO:Pipeline: Time: 04:49:23, Epoch: 8, Step: 46100, Step Loss: 3.81 INFO:Pipeline: Time: 04:50:01, Epoch: 8, Step: 46200, Step Loss: 3.47 INFO:Pipeline: Time: 04:50:38, Epoch: 8, Step: 46300, Step Loss: 3.87 INFO:Pipeline: Time: 04:51:16, Epoch: 8, Step: 46400, Step Loss: 3.87 INFO:Pipeline: Time: 04:51:53, Epoch: 8, Step: 46500, Step Loss: 3.97 INFO:Pipeline: Time: 04:52:30, Epoch: 8, Step: 46600, Step Loss: 3.88 INFO:Pipeline: Time: 04:53:08, Epoch: 8, Step: 46700, Step Loss: 3.85 INFO:Pipeline: Time: 04:53:45, Epoch: 8, Step: 46800, Step Loss: 3.81 INFO:Pipeline: Time: 04:54:22, Epoch: 8, Step: 46900, Step Loss: 3.74 INFO:Pipeline: Time: 04:54:59, Epoch: 8, Step: 47000, Step Loss: 3.52 INFO:Pipeline: Eval Loss: 3.62, Perplexity: 37.28 INFO:Pipeline: Time: 04:55:40, Epoch: 8, Step: 47100, Step Loss: 3.72 INFO:Pipeline: Time: 04:56:18, Epoch: 8, Step: 47200, Step Loss: 3.72 INFO:Pipeline: Time: 04:56:55, Epoch: 8, Step: 47300, Step Loss: 3.88 INFO:Pipeline: Time: 04:57:32, Epoch: 8, Step: 47400, Step Loss: 3.71 INFO:Pipeline: Time: 04:58:10, Epoch: 8, Step: 47500, Step Loss: 3.69 INFO:Pipeline: Time: 04:58:47, Epoch: 8, Step: 47600, Step Loss: 3.85 INFO:Pipeline: Time: 04:59:24, Epoch: 8, Step: 47700, Step Loss: 3.83 INFO:Pipeline: Time: 05:00:02, Epoch: 8, Step: 47800, Step Loss: 3.75 INFO:Pipeline: Time: 05:00:39, Epoch: 8, Step: 47900, Step Loss: 3.80 INFO:Pipeline: Time: 05:01:16, Epoch: 8, Step: 48000, Step Loss: 3.66 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 36.80 INFO:Pipeline: Time: 05:01:57, Epoch: 8, Step: 48100, Step Loss: 3.74 INFO:Pipeline: Time: 05:02:34, Epoch: 8, Step: 48200, Step Loss: 3.90 INFO:Pipeline: Time: 05:03:11, Epoch: 8, Step: 48300, Step Loss: 3.85 INFO:Pipeline: Time: 05:03:49, Epoch: 8, Step: 48400, Step Loss: 3.85 INFO:Pipeline: Time: 05:04:26, Epoch: 8, Step: 48500, Step Loss: 3.72 INFO:Pipeline: Time: 05:05:04, Epoch: 8, Step: 48600, Step Loss: 3.75 INFO:Pipeline: Time: 05:05:41, Epoch: 8, Step: 48700, Step Loss: 3.57 INFO:Pipeline: Time: 05:06:18, Epoch: 8, Step: 48800, Step Loss: 3.83 INFO:Pipeline: Time: 05:06:56, Epoch: 9, Step: 48900, Step Loss: 3.59 INFO:Pipeline: Time: 05:07:33, Epoch: 9, Step: 49000, Step Loss: 3.84 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.43 INFO:Pipeline: Time: 05:08:14, Epoch: 9, Step: 49100, Step Loss: 3.69 INFO:Pipeline: Time: 05:08:51, Epoch: 9, Step: 49200, Step Loss: 4.02 INFO:Pipeline: Time: 05:09:28, Epoch: 9, Step: 49300, Step Loss: 3.72 INFO:Pipeline: Time: 05:10:06, Epoch: 9, Step: 49400, Step Loss: 3.85 INFO:Pipeline: Time: 05:10:43, Epoch: 9, Step: 49500, Step Loss: 3.84 INFO:Pipeline: Time: 05:11:20, Epoch: 9, Step: 49600, Step Loss: 3.64 INFO:Pipeline: Time: 05:11:58, Epoch: 9, Step: 49700, Step Loss: 3.76 INFO:Pipeline: Time: 05:12:35, Epoch: 9, Step: 49800, Step Loss: 3.55 INFO:Pipeline: Time: 05:13:12, Epoch: 9, Step: 49900, Step Loss: 3.75 INFO:Pipeline: Time: 05:13:50, Epoch: 9, Step: 50000, Step Loss: 3.74 INFO:Pipeline: Eval Loss: 3.58, Perplexity: 36.04 INFO:Pipeline: Time: 05:14:31, Epoch: 9, Step: 50100, Step Loss: 3.75 INFO:Pipeline: Time: 05:15:08, Epoch: 9, Step: 50200, Step Loss: 3.80 INFO:Pipeline: Time: 05:15:45, Epoch: 9, Step: 50300, Step Loss: 3.72 INFO:Pipeline: Time: 05:16:23, Epoch: 9, Step: 50400, Step Loss: 3.72 INFO:Pipeline: Time: 05:17:00, Epoch: 9, Step: 50500, Step Loss: 3.83 INFO:Pipeline: Time: 05:17:37, Epoch: 9, Step: 50600, Step Loss: 3.88 INFO:Pipeline: Time: 05:18:15, Epoch: 9, Step: 50700, Step Loss: 3.77 INFO:Pipeline: Time: 05:18:52, Epoch: 9, Step: 50800, Step Loss: 3.61 INFO:Pipeline: Time: 05:19:30, Epoch: 9, Step: 50900, Step Loss: 3.78 INFO:Pipeline: Time: 05:20:07, Epoch: 9, Step: 51000, Step Loss: 3.65 INFO:Pipeline: Eval Loss: 3.58, Perplexity: 35.82 INFO:Pipeline: Time: 05:20:48, Epoch: 9, Step: 51100, Step Loss: 3.89 INFO:Pipeline: Time: 05:21:25, Epoch: 9, Step: 51200, Step Loss: 3.76 INFO:Pipeline: Time: 05:22:02, Epoch: 9, Step: 51300, Step Loss: 3.84 INFO:Pipeline: Time: 05:22:39, Epoch: 9, Step: 51400, Step Loss: 3.99 INFO:Pipeline: Time: 05:23:17, Epoch: 9, Step: 51500, Step Loss: 3.85 INFO:Pipeline: Time: 05:23:54, Epoch: 9, Step: 51600, Step Loss: 3.75 INFO:Pipeline: Time: 05:24:32, Epoch: 9, Step: 51700, Step Loss: 3.74 INFO:Pipeline: Time: 05:25:09, Epoch: 9, Step: 51800, Step Loss: 3.76 INFO:Pipeline: Time: 05:25:46, Epoch: 9, Step: 51900, Step Loss: 3.57 INFO:Pipeline: Time: 05:26:23, Epoch: 9, Step: 52000, Step Loss: 3.78 INFO:Pipeline: Eval Loss: 3.57, Perplexity: 35.41 INFO:Pipeline: Time: 05:27:04, Epoch: 9, Step: 52100, Step Loss: 3.67 INFO:Pipeline: Time: 05:27:41, Epoch: 9, Step: 52200, Step Loss: 3.63 INFO:Pipeline: Time: 05:28:18, Epoch: 9, Step: 52300, Step Loss: 3.76 INFO:Pipeline: Time: 05:28:56, Epoch: 9, Step: 52400, Step Loss: 3.77 INFO:Pipeline: Time: 05:29:33, Epoch: 9, Step: 52500, Step Loss: 3.57 INFO:Pipeline: Time: 05:30:10, Epoch: 9, Step: 52600, Step Loss: 3.63 INFO:Pipeline: Time: 05:30:48, Epoch: 9, Step: 52700, Step Loss: 3.69 INFO:Pipeline: Time: 05:31:25, Epoch: 9, Step: 52800, Step Loss: 3.78 INFO:Pipeline: Time: 05:32:02, Epoch: 9, Step: 52900, Step Loss: 3.72 INFO:Pipeline: Time: 05:32:39, Epoch: 9, Step: 53000, Step Loss: 3.44 INFO:Pipeline: Eval Loss: 3.56, Perplexity: 35.24 INFO:Pipeline: Time: 05:33:20, Epoch: 9, Step: 53100, Step Loss: 3.70 INFO:Pipeline: Time: 05:33:57, Epoch: 9, Step: 53200, Step Loss: 3.80 INFO:Pipeline: Time: 05:34:35, Epoch: 9, Step: 53300, Step Loss: 3.60 INFO:Pipeline: Time: 05:35:12, Epoch: 9, Step: 53400, Step Loss: 3.48 INFO:Pipeline: Time: 05:35:49, Epoch: 9, Step: 53500, Step Loss: 3.80 INFO:Pipeline: Time: 05:36:27, Epoch: 9, Step: 53600, Step Loss: 3.67 INFO:Pipeline: Time: 05:37:04, Epoch: 9, Step: 53700, Step Loss: 3.82 INFO:Pipeline: Time: 05:37:41, Epoch: 9, Step: 53800, Step Loss: 3.75 INFO:Pipeline: Time: 05:38:18, Epoch: 9, Step: 53900, Step Loss: 3.64 INFO:Pipeline: Time: 05:38:56, Epoch: 9, Step: 54000, Step Loss: 3.68 INFO:Pipeline: Eval Loss: 3.55, Perplexity: 34.74 INFO:Pipeline: Time: 05:39:37, Epoch: 9, Step: 54100, Step Loss: 3.72 INFO:Pipeline: Time: 05:40:14, Epoch: 9, Step: 54200, Step Loss: 3.64 INFO:Pipeline: Time: 05:40:51, Epoch: 9, Step: 54300, Step Loss: 3.59 INFO:Pipeline: Time: 05:41:28, Epoch: 9, Step: 54400, Step Loss: 3.45 INFO:Pipeline: Time: 05:42:06, Epoch: 9, Step: 54500, Step Loss: 3.64 INFO:Pipeline: Time: 05:42:43, Epoch: 9, Step: 54600, Step Loss: 3.82 INFO:Pipeline: Time: 05:43:20, Epoch: 9, Step: 54700, Step Loss: 3.59 INFO:Pipeline: Time: 05:43:58, Epoch: 9, Step: 54800, Step Loss: 3.73 INFO:Pipeline: Time: 05:44:35, Epoch: 9, Step: 54900, Step Loss: 3.57 INFO:Pipeline: Time: 05:45:12, Epoch: 10, Step: 55000, Step Loss: 3.64 INFO:Pipeline: Eval Loss: 3.54, Perplexity: 34.56 INFO:Pipeline: Time: 05:45:53, Epoch: 10, Step: 55100, Step Loss: 3.71 INFO:Pipeline: Time: 05:46:30, Epoch: 10, Step: 55200, Step Loss: 3.79 INFO:Pipeline: Time: 05:47:07, Epoch: 10, Step: 55300, Step Loss: 3.57 INFO:Pipeline: Time: 05:47:44, Epoch: 10, Step: 55400, Step Loss: 3.52 INFO:Pipeline: Time: 05:48:22, Epoch: 10, Step: 55500, Step Loss: 3.59 INFO:Pipeline: Time: 05:48:59, Epoch: 10, Step: 55600, Step Loss: 3.58 INFO:Pipeline: Time: 05:49:36, Epoch: 10, Step: 55700, Step Loss: 3.73 INFO:Pipeline: Time: 05:50:14, Epoch: 10, Step: 55800, Step Loss: 3.63 INFO:Pipeline: Time: 05:50:51, Epoch: 10, Step: 55900, Step Loss: 3.64 INFO:Pipeline: Time: 05:51:28, Epoch: 10, Step: 56000, Step Loss: 3.67 INFO:Pipeline: Eval Loss: 3.54, Perplexity: 34.33 INFO:Pipeline: Time: 05:52:09, Epoch: 10, Step: 56100, Step Loss: 3.54 INFO:Pipeline: Time: 05:52:46, Epoch: 10, Step: 56200, Step Loss: 3.73 INFO:Pipeline: Time: 05:53:23, Epoch: 10, Step: 56300, Step Loss: 3.50 INFO:Pipeline: Time: 05:54:00, Epoch: 10, Step: 56400, Step Loss: 3.58 INFO:Pipeline: Time: 05:54:38, Epoch: 10, Step: 56500, Step Loss: 3.57 INFO:Pipeline: Time: 05:55:15, Epoch: 10, Step: 56600, Step Loss: 3.74 INFO:Pipeline: Time: 05:55:52, Epoch: 10, Step: 56700, Step Loss: 3.63 INFO:Pipeline: Time: 05:56:29, Epoch: 10, Step: 56800, Step Loss: 3.44 INFO:Pipeline: Time: 05:57:07, Epoch: 10, Step: 56900, Step Loss: 3.44 INFO:Pipeline: Time: 05:57:44, Epoch: 10, Step: 57000, Step Loss: 3.83 INFO:Pipeline: Eval Loss: 3.53, Perplexity: 34.04 INFO:Pipeline: Time: 05:58:24, Epoch: 10, Step: 57100, Step Loss: 3.54 INFO:Pipeline: Time: 05:59:02, Epoch: 10, Step: 57200, Step Loss: 3.56 INFO:Pipeline: Time: 05:59:39, Epoch: 10, Step: 57300, Step Loss: 3.68 INFO:Pipeline: Time: 06:00:16, Epoch: 10, Step: 57400, Step Loss: 3.73 INFO:Pipeline: Time: 06:00:54, Epoch: 10, Step: 57500, Step Loss: 3.67 INFO:Pipeline: Time: 06:01:31, Epoch: 10, Step: 57600, Step Loss: 3.66 INFO:Pipeline: Time: 06:02:08, Epoch: 10, Step: 57700, Step Loss: 3.62 INFO:Pipeline: Time: 06:02:45, Epoch: 10, Step: 57800, Step Loss: 3.66 INFO:Pipeline: Time: 06:03:23, Epoch: 10, Step: 57900, Step Loss: 3.71 INFO:Pipeline: Time: 06:04:00, Epoch: 10, Step: 58000, Step Loss: 3.71 INFO:Pipeline: Eval Loss: 3.52, Perplexity: 33.82 INFO:Pipeline: Time: 06:04:40, Epoch: 10, Step: 58100, Step Loss: 3.62 INFO:Pipeline: Time: 06:05:18, Epoch: 10, Step: 58200, Step Loss: 3.64 INFO:Pipeline: Time: 06:05:55, Epoch: 10, Step: 58300, Step Loss: 3.80 INFO:Pipeline: Time: 06:06:32, Epoch: 10, Step: 58400, Step Loss: 3.66 INFO:Pipeline: Time: 06:07:10, Epoch: 10, Step: 58500, Step Loss: 3.62 INFO:Pipeline: Time: 06:07:47, Epoch: 10, Step: 58600, Step Loss: 3.76 INFO:Pipeline: Time: 06:08:24, Epoch: 10, Step: 58700, Step Loss: 3.65 INFO:Pipeline: Time: 06:09:01, Epoch: 10, Step: 58800, Step Loss: 3.50 INFO:Pipeline: Time: 06:09:39, Epoch: 10, Step: 58900, Step Loss: 3.78 INFO:Pipeline: Time: 06:10:16, Epoch: 10, Step: 59000, Step Loss: 3.69 INFO:Pipeline: Eval Loss: 3.51, Perplexity: 33.54 INFO:Pipeline: Time: 06:10:56, Epoch: 10, Step: 59100, Step Loss: 3.46 INFO:Pipeline: Time: 06:11:33, Epoch: 10, Step: 59200, Step Loss: 3.65 INFO:Pipeline: Time: 06:12:11, Epoch: 10, Step: 59300, Step Loss: 3.64 INFO:Pipeline: Time: 06:12:48, Epoch: 10, Step: 59400, Step Loss: 3.50 INFO:Pipeline: Time: 06:13:25, Epoch: 10, Step: 59500, Step Loss: 3.69 INFO:Pipeline: Time: 06:14:02, Epoch: 10, Step: 59600, Step Loss: 3.45 INFO:Pipeline: Time: 06:14:40, Epoch: 10, Step: 59700, Step Loss: 3.60 INFO:Pipeline: Time: 06:15:17, Epoch: 10, Step: 59800, Step Loss: 3.63 INFO:Pipeline: Time: 06:15:54, Epoch: 10, Step: 59900, Step Loss: 3.59 INFO:Pipeline: Time: 06:16:31, Epoch: 10, Step: 60000, Step Loss: 3.65 INFO:Pipeline: Eval Loss: 3.51, Perplexity: 33.33 INFO:Pipeline: Time: 06:17:12, Epoch: 10, Step: 60100, Step Loss: 3.57 INFO:Pipeline: Time: 06:17:49, Epoch: 10, Step: 60200, Step Loss: 3.74 INFO:Pipeline: Time: 06:18:26, Epoch: 10, Step: 60300, Step Loss: 3.71 INFO:Pipeline: Time: 06:19:03, Epoch: 10, Step: 60400, Step Loss: 3.80 INFO:Pipeline: Time: 06:19:40, Epoch: 10, Step: 60500, Step Loss: 3.74 INFO:Pipeline: Time: 06:20:18, Epoch: 10, Step: 60600, Step Loss: 3.62 INFO:Pipeline: Time: 06:20:55, Epoch: 10, Step: 60700, Step Loss: 3.56 INFO:Pipeline: Time: 06:21:32, Epoch: 10, Step: 60800, Step Loss: 3.69 INFO:Pipeline: Time: 06:22:09, Epoch: 10, Step: 60900, Step Loss: 3.48 INFO:Pipeline: Time: 06:22:47, Epoch: 10, Step: 61000, Step Loss: 3.69 INFO:Pipeline: Eval Loss: 3.50, Perplexity: 33.13 INFO:Pipeline: Time: 06:23:27, Epoch: 11, Step: 61100, Step Loss: 3.66 INFO:Pipeline: Time: 06:24:04, Epoch: 11, Step: 61200, Step Loss: 3.44 INFO:Pipeline: Time: 06:24:41, Epoch: 11, Step: 61300, Step Loss: 3.83 INFO:Pipeline: Time: 06:25:18, Epoch: 11, Step: 61400, Step Loss: 3.68 INFO:Pipeline: Time: 06:25:56, Epoch: 11, Step: 61500, Step Loss: 3.54 INFO:Pipeline: Time: 06:26:33, Epoch: 11, Step: 61600, Step Loss: 3.71 INFO:Pipeline: Time: 06:27:10, Epoch: 11, Step: 61700, Step Loss: 3.54 INFO:Pipeline: Time: 06:27:47, Epoch: 11, Step: 61800, Step Loss: 3.57 INFO:Pipeline: Time: 06:28:25, Epoch: 11, Step: 61900, Step Loss: 3.67 INFO:Pipeline: Time: 06:29:02, Epoch: 11, Step: 62000, Step Loss: 3.48 INFO:Pipeline: Eval Loss: 3.49, Perplexity: 32.89 INFO:Pipeline: Time: 06:29:42, Epoch: 11, Step: 62100, Step Loss: 3.72 INFO:Pipeline: Time: 06:30:19, Epoch: 11, Step: 62200, Step Loss: 3.60 INFO:Pipeline: Time: 06:30:56, Epoch: 11, Step: 62300, Step Loss: 3.54 INFO:Pipeline: Time: 06:31:34, Epoch: 11, Step: 62400, Step Loss: 3.64 INFO:Pipeline: Time: 06:32:11, Epoch: 11, Step: 62500, Step Loss: 3.56 INFO:Pipeline: Time: 06:32:48, Epoch: 11, Step: 62600, Step Loss: 3.43 INFO:Pipeline: Time: 06:33:25, Epoch: 11, Step: 62700, Step Loss: 3.59 INFO:Pipeline: Time: 06:34:03, Epoch: 11, Step: 62800, Step Loss: 3.60 INFO:Pipeline: Time: 06:34:40, Epoch: 11, Step: 62900, Step Loss: 3.51 INFO:Pipeline: Time: 06:35:17, Epoch: 11, Step: 63000, Step Loss: 3.58 INFO:Pipeline: Eval Loss: 3.49, Perplexity: 32.68 INFO:Pipeline: Time: 06:35:57, Epoch: 11, Step: 63100, Step Loss: 3.54 INFO:Pipeline: Time: 06:36:35, Epoch: 11, Step: 63200, Step Loss: 3.59 INFO:Pipeline: Time: 06:37:12, Epoch: 11, Step: 63300, Step Loss: 3.44 INFO:Pipeline: Time: 06:37:49, Epoch: 11, Step: 63400, Step Loss: 3.53 INFO:Pipeline: Time: 06:38:26, Epoch: 11, Step: 63500, Step Loss: 3.68 INFO:Pipeline: Time: 06:39:04, Epoch: 11, Step: 63600, Step Loss: 3.62 INFO:Pipeline: Time: 06:39:41, Epoch: 11, Step: 63700, Step Loss: 3.50 INFO:Pipeline: Time: 06:40:18, Epoch: 11, Step: 63800, Step Loss: 3.62 INFO:Pipeline: Time: 06:40:55, Epoch: 11, Step: 63900, Step Loss: 3.63 INFO:Pipeline: Time: 06:41:32, Epoch: 11, Step: 64000, Step Loss: 3.58 INFO:Pipeline: Eval Loss: 3.48, Perplexity: 32.53 INFO:Pipeline: Time: 06:42:13, Epoch: 11, Step: 64100, Step Loss: 3.55 INFO:Pipeline: Time: 06:42:50, Epoch: 11, Step: 64200, Step Loss: 3.52 INFO:Pipeline: Time: 06:43:27, Epoch: 11, Step: 64300, Step Loss: 3.68 INFO:Pipeline: Time: 06:44:04, Epoch: 11, Step: 64400, Step Loss: 3.61 INFO:Pipeline: Time: 06:44:41, Epoch: 11, Step: 64500, Step Loss: 3.63 INFO:Pipeline: Time: 06:45:19, Epoch: 11, Step: 64600, Step Loss: 3.46 INFO:Pipeline: Time: 06:45:56, Epoch: 11, Step: 64700, Step Loss: 3.59 INFO:Pipeline: Time: 06:46:33, Epoch: 11, Step: 64800, Step Loss: 3.61 INFO:Pipeline: Time: 06:47:10, Epoch: 11, Step: 64900, Step Loss: 3.55 INFO:Pipeline: Time: 06:47:47, Epoch: 11, Step: 65000, Step Loss: 3.82 INFO:Pipeline: Eval Loss: 3.48, Perplexity: 32.36 INFO:Pipeline: Time: 06:48:28, Epoch: 11, Step: 65100, Step Loss: 3.56 INFO:Pipeline: Time: 06:49:05, Epoch: 11, Step: 65200, Step Loss: 3.69 INFO:Pipeline: Time: 06:49:42, Epoch: 11, Step: 65300, Step Loss: 3.54 INFO:Pipeline: Time: 06:50:19, Epoch: 11, Step: 65400, Step Loss: 3.74 INFO:Pipeline: Time: 06:50:56, Epoch: 11, Step: 65500, Step Loss: 3.80 INFO:Pipeline: Time: 06:51:34, Epoch: 11, Step: 65600, Step Loss: 3.61 INFO:Pipeline: Time: 06:52:11, Epoch: 11, Step: 65700, Step Loss: 3.42 INFO:Pipeline: Time: 06:52:48, Epoch: 11, Step: 65800, Step Loss: 3.79 INFO:Pipeline: Time: 06:53:25, Epoch: 11, Step: 65900, Step Loss: 3.62 INFO:Pipeline: Time: 06:54:03, Epoch: 11, Step: 66000, Step Loss: 3.44 INFO:Pipeline: Eval Loss: 3.47, Perplexity: 32.17 INFO:Pipeline: Time: 06:54:43, Epoch: 11, Step: 66100, Step Loss: 3.51 INFO:Pipeline: Time: 06:55:20, Epoch: 11, Step: 66200, Step Loss: 3.54 INFO:Pipeline: Time: 06:55:57, Epoch: 11, Step: 66300, Step Loss: 3.57 INFO:Pipeline: Time: 06:56:35, Epoch: 11, Step: 66400, Step Loss: 3.48 INFO:Pipeline: Time: 06:57:12, Epoch: 11, Step: 66500, Step Loss: 3.82 INFO:Pipeline: Time: 06:57:49, Epoch: 11, Step: 66600, Step Loss: 3.52 INFO:Pipeline: Time: 06:58:26, Epoch: 11, Step: 66700, Step Loss: 3.59 INFO:Pipeline: Time: 06:59:03, Epoch: 11, Step: 66800, Step Loss: 3.52 INFO:Pipeline: Time: 06:59:41, Epoch: 11, Step: 66900, Step Loss: 3.54 INFO:Pipeline: Time: 07:00:18, Epoch: 11, Step: 67000, Step Loss: 3.53 INFO:Pipeline: Eval Loss: 3.47, Perplexity: 32.00 INFO:Pipeline: Time: 07:00:59, Epoch: 11, Step: 67100, Step Loss: 3.73 INFO:Pipeline: Time: 07:01:36, Epoch: 12, Step: 67200, Step Loss: 3.68 INFO:Pipeline: Time: 07:02:13, Epoch: 12, Step: 67300, Step Loss: 3.65 INFO:Pipeline: Time: 07:02:50, Epoch: 12, Step: 67400, Step Loss: 3.51 INFO:Pipeline: Time: 07:03:27, Epoch: 12, Step: 67500, Step Loss: 3.85 INFO:Pipeline: Time: 07:04:05, Epoch: 12, Step: 67600, Step Loss: 3.67 INFO:Pipeline: Time: 07:04:42, Epoch: 12, Step: 67700, Step Loss: 3.66 INFO:Pipeline: Time: 07:05:19, Epoch: 12, Step: 67800, Step Loss: 3.68 INFO:Pipeline: Time: 07:05:57, Epoch: 12, Step: 67900, Step Loss: 3.68 INFO:Pipeline: Time: 07:06:34, Epoch: 12, Step: 68000, Step Loss: 3.65 INFO:Pipeline: Eval Loss: 3.46, Perplexity: 31.84 INFO:Pipeline: Time: 07:07:14, Epoch: 12, Step: 68100, Step Loss: 3.54 INFO:Pipeline: Time: 07:07:52, Epoch: 12, Step: 68200, Step Loss: 3.50 INFO:Pipeline: Time: 07:08:29, Epoch: 12, Step: 68300, Step Loss: 3.44 INFO:Pipeline: Time: 07:09:06, Epoch: 12, Step: 68400, Step Loss: 3.59 INFO:Pipeline: Time: 07:09:43, Epoch: 12, Step: 68500, Step Loss: 3.61 INFO:Pipeline: Time: 07:10:21, Epoch: 12, Step: 68600, Step Loss: 3.79 INFO:Pipeline: Time: 07:10:58, Epoch: 12, Step: 68700, Step Loss: 3.66 INFO:Pipeline: Time: 07:11:35, Epoch: 12, Step: 68800, Step Loss: 3.56 INFO:Pipeline: Time: 07:12:13, Epoch: 12, Step: 68900, Step Loss: 3.62 INFO:Pipeline: Time: 07:12:50, Epoch: 12, Step: 69000, Step Loss: 3.64 INFO:Pipeline: Eval Loss: 3.46, Perplexity: 31.70 INFO:Pipeline: Time: 07:13:30, Epoch: 12, Step: 69100, Step Loss: 3.54 INFO:Pipeline: Time: 07:14:08, Epoch: 12, Step: 69200, Step Loss: 3.51 INFO:Pipeline: Time: 07:14:45, Epoch: 12, Step: 69300, Step Loss: 3.50 INFO:Pipeline: Time: 07:15:22, Epoch: 12, Step: 69400, Step Loss: 3.63 INFO:Pipeline: Time: 07:15:59, Epoch: 12, Step: 69500, Step Loss: 3.51 INFO:Pipeline: Time: 07:16:37, Epoch: 12, Step: 69600, Step Loss: 3.56 INFO:Pipeline: Time: 07:17:14, Epoch: 12, Step: 69700, Step Loss: 3.62 INFO:Pipeline: Time: 07:17:51, Epoch: 12, Step: 69800, Step Loss: 3.80 INFO:Pipeline: Time: 07:18:29, Epoch: 12, Step: 69900, Step Loss: 3.69 INFO:Pipeline: Time: 07:19:06, Epoch: 12, Step: 70000, Step Loss: 3.60 INFO:Pipeline: Eval Loss: 3.45, Perplexity: 31.53 INFO:Pipeline: Time: 07:19:46, Epoch: 12, Step: 70100, Step Loss: 3.61 INFO:Pipeline: Time: 07:20:23, Epoch: 12, Step: 70200, Step Loss: 3.35 INFO:Pipeline: Time: 07:21:01, Epoch: 12, Step: 70300, Step Loss: 3.70 INFO:Pipeline: Time: 07:21:38, Epoch: 12, Step: 70400, Step Loss: 3.60 INFO:Pipeline: Time: 07:22:15, Epoch: 12, Step: 70500, Step Loss: 3.67 INFO:Pipeline: Time: 07:22:53, Epoch: 12, Step: 70600, Step Loss: 3.51 INFO:Pipeline: Time: 07:23:30, Epoch: 12, Step: 70700, Step Loss: 3.55 INFO:Pipeline: Time: 07:24:07, Epoch: 12, Step: 70800, Step Loss: 3.69 INFO:Pipeline: Time: 07:24:44, Epoch: 12, Step: 70900, Step Loss: 3.58 INFO:Pipeline: Time: 07:25:22, Epoch: 12, Step: 71000, Step Loss: 3.58 INFO:Pipeline: Eval Loss: 3.45, Perplexity: 31.36 INFO:Pipeline: Time: 07:26:02, Epoch: 12, Step: 71100, Step Loss: 3.34 INFO:Pipeline: Time: 07:26:39, Epoch: 12, Step: 71200, Step Loss: 3.50 INFO:Pipeline: Time: 07:27:17, Epoch: 12, Step: 71300, Step Loss: 3.58 INFO:Pipeline: Time: 07:27:54, Epoch: 12, Step: 71400, Step Loss: 3.62 INFO:Pipeline: Time: 07:28:31, Epoch: 12, Step: 71500, Step Loss: 3.60 INFO:Pipeline: Time: 07:29:09, Epoch: 12, Step: 71600, Step Loss: 3.64 INFO:Pipeline: Time: 07:29:46, Epoch: 12, Step: 71700, Step Loss: 3.50 INFO:Pipeline: Time: 07:30:23, Epoch: 12, Step: 71800, Step Loss: 3.44 INFO:Pipeline: Time: 07:31:00, Epoch: 12, Step: 71900, Step Loss: 3.66 INFO:Pipeline: Time: 07:31:38, Epoch: 12, Step: 72000, Step Loss: 3.73 INFO:Pipeline: Eval Loss: 3.44, Perplexity: 31.31 INFO:Pipeline: Time: 07:32:18, Epoch: 12, Step: 72100, Step Loss: 3.37 INFO:Pipeline: Time: 07:32:55, Epoch: 12, Step: 72200, Step Loss: 3.46 INFO:Pipeline: Time: 07:33:33, Epoch: 12, Step: 72300, Step Loss: 3.63 INFO:Pipeline: Time: 07:34:10, Epoch: 12, Step: 72400, Step Loss: 3.46 INFO:Pipeline: Time: 07:34:47, Epoch: 12, Step: 72500, Step Loss: 3.50 INFO:Pipeline: Time: 07:35:25, Epoch: 12, Step: 72600, Step Loss: 3.64 INFO:Pipeline: Time: 07:36:02, Epoch: 12, Step: 72700, Step Loss: 3.64 INFO:Pipeline: Time: 07:36:39, Epoch: 12, Step: 72800, Step Loss: 3.35 INFO:Pipeline: Time: 07:37:16, Epoch: 12, Step: 72900, Step Loss: 3.55 INFO:Pipeline: Time: 07:37:54, Epoch: 12, Step: 73000, Step Loss: 3.81 INFO:Pipeline: Eval Loss: 3.44, Perplexity: 31.18 INFO:Pipeline: Time: 07:38:34, Epoch: 12, Step: 73100, Step Loss: 3.66 INFO:Pipeline: Time: 07:39:11, Epoch: 12, Step: 73200, Step Loss: 3.45 INFO:Pipeline: Time: 07:39:48, Epoch: 13, Step: 73300, Step Loss: 3.59 INFO:Pipeline: Time: 07:40:26, Epoch: 13, Step: 73400, Step Loss: 3.55 INFO:Pipeline: Time: 07:41:03, Epoch: 13, Step: 73500, Step Loss: 3.65 INFO:Pipeline: Time: 07:41:40, Epoch: 13, Step: 73600, Step Loss: 3.59 INFO:Pipeline: Time: 07:42:18, Epoch: 13, Step: 73700, Step Loss: 3.50 INFO:Pipeline: Time: 07:42:55, Epoch: 13, Step: 73800, Step Loss: 3.52 INFO:Pipeline: Time: 07:43:32, Epoch: 13, Step: 73900, Step Loss: 3.62 INFO:Pipeline: Time: 07:44:09, Epoch: 13, Step: 74000, Step Loss: 3.59 INFO:Pipeline: Eval Loss: 3.43, Perplexity: 31.02 INFO:Pipeline: Time: 07:44:50, Epoch: 13, Step: 74100, Step Loss: 3.45 INFO:Pipeline: Time: 07:45:27, Epoch: 13, Step: 74200, Step Loss: 3.42 INFO:Pipeline: Time: 07:46:04, Epoch: 13, Step: 74300, Step Loss: 3.70 INFO:Pipeline: Time: 07:46:41, Epoch: 13, Step: 74400, Step Loss: 3.52 INFO:Pipeline: Time: 07:47:19, Epoch: 13, Step: 74500, Step Loss: 3.40 INFO:Pipeline: Time: 07:47:56, Epoch: 13, Step: 74600, Step Loss: 3.64 INFO:Pipeline: Time: 07:48:33, Epoch: 13, Step: 74700, Step Loss: 3.72 INFO:Pipeline: Time: 07:49:10, Epoch: 13, Step: 74800, Step Loss: 3.48 INFO:Pipeline: Time: 07:49:47, Epoch: 13, Step: 74900, Step Loss: 3.55 INFO:Pipeline: Time: 07:50:25, Epoch: 13, Step: 75000, Step Loss: 3.54 INFO:Pipeline: Eval Loss: 3.43, Perplexity: 30.88 INFO:Pipeline: Time: 07:51:05, Epoch: 13, Step: 75100, Step Loss: 3.52 INFO:Pipeline: Time: 07:51:42, Epoch: 13, Step: 75200, Step Loss: 3.56 INFO:Pipeline: Time: 07:52:19, Epoch: 13, Step: 75300, Step Loss: 3.56 INFO:Pipeline: Time: 07:52:56, Epoch: 13, Step: 75400, Step Loss: 3.66 INFO:Pipeline: Time: 07:53:34, Epoch: 13, Step: 75500, Step Loss: 3.65 INFO:Pipeline: Time: 07:54:11, Epoch: 13, Step: 75600, Step Loss: 3.57 INFO:Pipeline: Time: 07:54:48, Epoch: 13, Step: 75700, Step Loss: 3.58 INFO:Pipeline: Time: 07:55:25, Epoch: 13, Step: 75800, Step Loss: 3.60 INFO:Pipeline: Time: 07:56:02, Epoch: 13, Step: 75900, Step Loss: 3.44 INFO:Pipeline: Time: 07:56:40, Epoch: 13, Step: 76000, Step Loss: 3.52 INFO:Pipeline: Eval Loss: 3.43, Perplexity: 30.76 INFO:Pipeline: Time: 07:57:20, Epoch: 13, Step: 76100, Step Loss: 3.60 INFO:Pipeline: Time: 07:57:57, Epoch: 13, Step: 76200, Step Loss: 3.67 INFO:Pipeline: Time: 07:58:34, Epoch: 13, Step: 76300, Step Loss: 3.44 INFO:Pipeline: Time: 07:59:11, Epoch: 13, Step: 76400, Step Loss: 3.65 INFO:Pipeline: Time: 07:59:49, Epoch: 13, Step: 76500, Step Loss: 3.61 INFO:Pipeline: Time: 08:00:26, Epoch: 13, Step: 76600, Step Loss: 3.58 INFO:Pipeline: Time: 08:01:03, Epoch: 13, Step: 76700, Step Loss: 3.42 INFO:Pipeline: Time: 08:01:40, Epoch: 13, Step: 76800, Step Loss: 3.63 INFO:Pipeline: Time: 08:02:17, Epoch: 13, Step: 76900, Step Loss: 3.59 INFO:Pipeline: Time: 08:02:55, Epoch: 13, Step: 77000, Step Loss: 3.81 INFO:Pipeline: Eval Loss: 3.43, Perplexity: 30.73 INFO:Pipeline: Time: 08:03:35, Epoch: 13, Step: 77100, Step Loss: 3.43 INFO:Pipeline: Time: 08:04:12, Epoch: 13, Step: 77200, Step Loss: 3.68 INFO:Pipeline: Time: 08:04:49, Epoch: 13, Step: 77300, Step Loss: 3.63 INFO:Pipeline: Time: 08:05:26, Epoch: 13, Step: 77400, Step Loss: 3.55 INFO:Pipeline: Time: 08:06:04, Epoch: 13, Step: 77500, Step Loss: 3.43 INFO:Pipeline: Time: 08:06:41, Epoch: 13, Step: 77600, Step Loss: 3.64 INFO:Pipeline: Time: 08:07:18, Epoch: 13, Step: 77700, Step Loss: 3.55 INFO:Pipeline: Time: 08:07:55, Epoch: 13, Step: 77800, Step Loss: 3.40 INFO:Pipeline: Time: 08:08:32, Epoch: 13, Step: 77900, Step Loss: 3.68 INFO:Pipeline: Time: 08:09:10, Epoch: 13, Step: 78000, Step Loss: 3.67 INFO:Pipeline: Eval Loss: 3.42, Perplexity: 30.62 INFO:Pipeline: Time: 08:09:50, Epoch: 13, Step: 78100, Step Loss: 3.41 INFO:Pipeline: Time: 08:10:27, Epoch: 13, Step: 78200, Step Loss: 3.47 INFO:Pipeline: Time: 08:11:04, Epoch: 13, Step: 78300, Step Loss: 3.66 INFO:Pipeline: Time: 08:11:41, Epoch: 13, Step: 78400, Step Loss: 3.56 INFO:Pipeline: Time: 08:12:18, Epoch: 13, Step: 78500, Step Loss: 3.56 INFO:Pipeline: Time: 08:12:56, Epoch: 13, Step: 78600, Step Loss: 3.52 INFO:Pipeline: Time: 08:13:33, Epoch: 13, Step: 78700, Step Loss: 3.44 INFO:Pipeline: Time: 08:14:10, Epoch: 13, Step: 78800, Step Loss: 3.59 INFO:Pipeline: Time: 08:14:47, Epoch: 13, Step: 78900, Step Loss: 3.67 INFO:Pipeline: Time: 08:15:24, Epoch: 13, Step: 79000, Step Loss: 3.40 INFO:Pipeline: Eval Loss: 3.42, Perplexity: 30.51 INFO:Pipeline: Time: 08:16:05, Epoch: 13, Step: 79100, Step Loss: 3.43 INFO:Pipeline: Time: 08:16:42, Epoch: 13, Step: 79200, Step Loss: 3.67 INFO:Pipeline: Time: 08:17:19, Epoch: 13, Step: 79300, Step Loss: 3.66 INFO:Pipeline: Time: 08:17:56, Epoch: 14, Step: 79400, Step Loss: 3.74 INFO:Pipeline: Time: 08:18:33, Epoch: 14, Step: 79500, Step Loss: 3.52 INFO:Pipeline: Time: 08:19:10, Epoch: 14, Step: 79600, Step Loss: 3.63 INFO:Pipeline: Time: 08:19:48, Epoch: 14, Step: 79700, Step Loss: 3.45 INFO:Pipeline: Time: 08:20:25, Epoch: 14, Step: 79800, Step Loss: 3.56 INFO:Pipeline: Time: 08:21:02, Epoch: 14, Step: 79900, Step Loss: 3.43 INFO:Pipeline: Time: 08:21:39, Epoch: 14, Step: 80000, Step Loss: 3.53 INFO:Pipeline: Eval Loss: 3.41, Perplexity: 30.40 INFO:Pipeline: Time: 08:22:20, Epoch: 14, Step: 80100, Step Loss: 3.50 INFO:Pipeline: Time: 08:22:57, Epoch: 14, Step: 80200, Step Loss: 3.40 INFO:Pipeline: Time: 08:23:34, Epoch: 14, Step: 80300, Step Loss: 3.63 INFO:Pipeline: Time: 08:24:11, Epoch: 14, Step: 80400, Step Loss: 3.50 INFO:Pipeline: Time: 08:24:49, Epoch: 14, Step: 80500, Step Loss: 3.57 INFO:Pipeline: Time: 08:25:26, Epoch: 14, Step: 80600, Step Loss: 3.54 INFO:Pipeline: Time: 08:26:03, Epoch: 14, Step: 80700, Step Loss: 3.71 INFO:Pipeline: Time: 08:26:40, Epoch: 14, Step: 80800, Step Loss: 3.35 INFO:Pipeline: Time: 08:27:18, Epoch: 14, Step: 80900, Step Loss: 3.65 INFO:Pipeline: Time: 08:27:55, Epoch: 14, Step: 81000, Step Loss: 3.48 INFO:Pipeline: Eval Loss: 3.41, Perplexity: 30.31 INFO:Pipeline: Time: 08:28:35, Epoch: 14, Step: 81100, Step Loss: 3.45 INFO:Pipeline: Time: 08:29:12, Epoch: 14, Step: 81200, Step Loss: 3.40 INFO:Pipeline: Time: 08:29:50, Epoch: 14, Step: 81300, Step Loss: 3.43 INFO:Pipeline: Time: 08:30:27, Epoch: 14, Step: 81400, Step Loss: 3.79 INFO:Pipeline: Time: 08:31:04, Epoch: 14, Step: 81500, Step Loss: 3.54 INFO:Pipeline: Time: 08:31:41, Epoch: 14, Step: 81600, Step Loss: 3.54 INFO:Pipeline: Time: 08:32:19, Epoch: 14, Step: 81700, Step Loss: 3.69 INFO:Pipeline: Time: 08:32:56, Epoch: 14, Step: 81800, Step Loss: 3.48 INFO:Pipeline: Time: 08:33:33, Epoch: 14, Step: 81900, Step Loss: 3.53 INFO:Pipeline: Time: 08:34:11, Epoch: 14, Step: 82000, Step Loss: 3.58 INFO:Pipeline: Eval Loss: 3.41, Perplexity: 30.25 INFO:Pipeline: Time: 08:34:51, Epoch: 14, Step: 82100, Step Loss: 3.73 INFO:Pipeline: Time: 08:35:28, Epoch: 14, Step: 82200, Step Loss: 3.47 INFO:Pipeline: Time: 08:36:05, Epoch: 14, Step: 82300, Step Loss: 3.71 INFO:Pipeline: Time: 08:36:43, Epoch: 14, Step: 82400, Step Loss: 3.64 INFO:Pipeline: Time: 08:37:20, Epoch: 14, Step: 82500, Step Loss: 3.50 INFO:Pipeline: Time: 08:37:57, Epoch: 14, Step: 82600, Step Loss: 3.61 INFO:Pipeline: Time: 08:38:35, Epoch: 14, Step: 82700, Step Loss: 3.56 INFO:Pipeline: Time: 08:39:12, Epoch: 14, Step: 82800, Step Loss: 3.59 INFO:Pipeline: Time: 08:39:49, Epoch: 14, Step: 82900, Step Loss: 3.41 INFO:Pipeline: Time: 08:40:26, Epoch: 14, Step: 83000, Step Loss: 3.44 INFO:Pipeline: Eval Loss: 3.41, Perplexity: 30.14 INFO:Pipeline: Time: 08:41:07, Epoch: 14, Step: 83100, Step Loss: 3.71 INFO:Pipeline: Time: 08:41:44, Epoch: 14, Step: 83200, Step Loss: 3.55 INFO:Pipeline: Time: 08:42:21, Epoch: 14, Step: 83300, Step Loss: 3.69 INFO:Pipeline: Time: 08:42:59, Epoch: 14, Step: 83400, Step Loss: 3.46 INFO:Pipeline: Time: 08:43:36, Epoch: 14, Step: 83500, Step Loss: 3.53 INFO:Pipeline: Time: 08:44:14, Epoch: 14, Step: 83600, Step Loss: 3.59 INFO:Pipeline: Time: 08:44:51, Epoch: 14, Step: 83700, Step Loss: 3.71 INFO:Pipeline: Time: 08:45:28, Epoch: 14, Step: 83800, Step Loss: 3.59 INFO:Pipeline: Time: 08:46:06, Epoch: 14, Step: 83900, Step Loss: 3.55 INFO:Pipeline: Time: 08:46:43, Epoch: 14, Step: 84000, Step Loss: 3.18 INFO:Pipeline: Eval Loss: 3.40, Perplexity: 30.08 INFO:Pipeline: Time: 08:47:23, Epoch: 14, Step: 84100, Step Loss: 3.45 INFO:Pipeline: Time: 08:48:01, Epoch: 14, Step: 84200, Step Loss: 3.53 INFO:Pipeline: Time: 08:48:38, Epoch: 14, Step: 84300, Step Loss: 3.70 INFO:Pipeline: Time: 08:49:15, Epoch: 14, Step: 84400, Step Loss: 3.39 INFO:Pipeline: Time: 08:49:52, Epoch: 14, Step: 84500, Step Loss: 3.38 INFO:Pipeline: Time: 08:50:29, Epoch: 14, Step: 84600, Step Loss: 3.37 INFO:Pipeline: Time: 08:51:07, Epoch: 14, Step: 84700, Step Loss: 3.64 INFO:Pipeline: Time: 08:51:44, Epoch: 14, Step: 84800, Step Loss: 3.47 INFO:Pipeline: Time: 08:52:21, Epoch: 14, Step: 84900, Step Loss: 3.78 INFO:Pipeline: Time: 08:52:58, Epoch: 14, Step: 85000, Step Loss: 3.36 INFO:Pipeline: Eval Loss: 3.40, Perplexity: 30.03 INFO:Pipeline: Time: 08:53:39, Epoch: 14, Step: 85100, Step Loss: 3.44 INFO:Pipeline: Time: 08:54:16, Epoch: 14, Step: 85200, Step Loss: 3.60 INFO:Pipeline: Time: 08:54:53, Epoch: 14, Step: 85300, Step Loss: 3.52 INFO:Pipeline: Time: 08:55:30, Epoch: 14, Step: 85400, Step Loss: 3.56 INFO:Pipeline: Time: 08:56:07, Epoch: 15, Step: 85500, Step Loss: 3.59 INFO:Pipeline: Time: 08:56:44, Epoch: 15, Step: 85600, Step Loss: 3.57 INFO:Pipeline: Time: 08:57:21, Epoch: 15, Step: 85700, Step Loss: 3.59 INFO:Pipeline: Time: 08:57:59, Epoch: 15, Step: 85800, Step Loss: 3.50 INFO:Pipeline: Time: 08:58:36, Epoch: 15, Step: 85900, Step Loss: 3.20 INFO:Pipeline: Time: 08:59:13, Epoch: 15, Step: 86000, Step Loss: 3.51 INFO:Pipeline: Eval Loss: 3.40, Perplexity: 29.95 INFO:Pipeline: Time: 08:59:54, Epoch: 15, Step: 86100, Step Loss: 3.36 INFO:Pipeline: Time: 09:00:31, Epoch: 15, Step: 86200, Step Loss: 3.52 INFO:Pipeline: Time: 09:01:08, Epoch: 15, Step: 86300, Step Loss: 3.41 INFO:Pipeline: Time: 09:01:45, Epoch: 15, Step: 86400, Step Loss: 3.36 INFO:Pipeline: Time: 09:02:22, Epoch: 15, Step: 86500, Step Loss: 3.56 INFO:Pipeline: Time: 09:03:00, Epoch: 15, Step: 86600, Step Loss: 3.61 INFO:Pipeline: Time: 09:03:37, Epoch: 15, Step: 86700, Step Loss: 3.52 INFO:Pipeline: Time: 09:04:14, Epoch: 15, Step: 86800, Step Loss: 3.58 INFO:Pipeline: Time: 09:04:51, Epoch: 15, Step: 86900, Step Loss: 3.58 INFO:Pipeline: Time: 09:05:28, Epoch: 15, Step: 87000, Step Loss: 3.56 INFO:Pipeline: Eval Loss: 3.40, Perplexity: 29.89 INFO:Pipeline: Time: 09:06:09, Epoch: 15, Step: 87100, Step Loss: 3.45 INFO:Pipeline: Time: 09:06:46, Epoch: 15, Step: 87200, Step Loss: 3.47 INFO:Pipeline: Time: 09:07:25, Epoch: 15, Step: 87300, Step Loss: 3.45 INFO:Pipeline: Time: 09:08:03, Epoch: 15, Step: 87400, Step Loss: 3.53 INFO:Pipeline: Time: 09:08:41, Epoch: 15, Step: 87500, Step Loss: 3.49 INFO:Pipeline: Time: 09:09:19, Epoch: 15, Step: 87600, Step Loss: 3.59 INFO:Pipeline: Time: 09:09:58, Epoch: 15, Step: 87700, Step Loss: 3.48 INFO:Pipeline: Time: 09:10:36, Epoch: 15, Step: 87800, Step Loss: 3.44 INFO:Pipeline: Time: 09:11:14, Epoch: 15, Step: 87900, Step Loss: 3.54 INFO:Pipeline: Time: 09:11:53, Epoch: 15, Step: 88000, Step Loss: 3.49 INFO:Pipeline: Eval Loss: 3.39, Perplexity: 29.81 INFO:Pipeline: Time: 09:12:35, Epoch: 15, Step: 88100, Step Loss: 3.65 INFO:Pipeline: Time: 09:13:13, Epoch: 15, Step: 88200, Step Loss: 3.41 INFO:Pipeline: Time: 09:13:51, Epoch: 15, Step: 88300, Step Loss: 3.50 INFO:Pipeline: Time: 09:14:30, Epoch: 15, Step: 88400, Step Loss: 3.58 INFO:Pipeline: Time: 09:15:08, Epoch: 15, Step: 88500, Step Loss: 3.57 INFO:Pipeline: Time: 09:15:46, Epoch: 15, Step: 88600, Step Loss: 3.50 INFO:Pipeline: Time: 09:16:24, Epoch: 15, Step: 88700, Step Loss: 3.54 INFO:Pipeline: Time: 09:17:03, Epoch: 15, Step: 88800, Step Loss: 3.45 INFO:Pipeline: Time: 09:17:41, Epoch: 15, Step: 88900, Step Loss: 3.52 INFO:Pipeline: Time: 09:18:19, Epoch: 15, Step: 89000, Step Loss: 3.40 INFO:Pipeline: Eval Loss: 3.39, Perplexity: 29.77 INFO:Pipeline: Time: 09:19:01, Epoch: 15, Step: 89100, Step Loss: 3.58 INFO:Pipeline: Time: 09:19:39, Epoch: 15, Step: 89200, Step Loss: 3.57 INFO:Pipeline: Time: 09:20:18, Epoch: 15, Step: 89300, Step Loss: 3.55 INFO:Pipeline: Time: 09:20:56, Epoch: 15, Step: 89400, Step Loss: 3.48 INFO:Pipeline: Time: 09:21:34, Epoch: 15, Step: 89500, Step Loss: 3.56 INFO:Pipeline: Time: 09:22:13, Epoch: 15, Step: 89600, Step Loss: 3.47 INFO:Pipeline: Time: 09:22:51, Epoch: 15, Step: 89700, Step Loss: 3.19 INFO:Pipeline: Time: 09:23:29, Epoch: 15, Step: 89800, Step Loss: 3.63 INFO:Pipeline: Time: 09:24:08, Epoch: 15, Step: 89900, Step Loss: 3.67 INFO:Pipeline: Time: 09:24:46, Epoch: 15, Step: 90000, Step Loss: 3.67 INFO:Pipeline: Eval Loss: 3.39, Perplexity: 29.68 INFO:Pipeline: Time: 09:25:28, Epoch: 15, Step: 90100, Step Loss: 3.47 INFO:Pipeline: Time: 09:26:06, Epoch: 15, Step: 90200, Step Loss: 3.44 INFO:Pipeline: Time: 09:26:45, Epoch: 15, Step: 90300, Step Loss: 3.46 INFO:Pipeline: Time: 09:27:23, Epoch: 15, Step: 90400, Step Loss: 3.52 INFO:Pipeline: Time: 09:28:01, Epoch: 15, Step: 90500, Step Loss: 3.50 INFO:Pipeline: Time: 09:28:40, Epoch: 15, Step: 90600, Step Loss: 3.46 INFO:Pipeline: Time: 09:29:18, Epoch: 15, Step: 90700, Step Loss: 3.46 INFO:Pipeline: Time: 09:29:56, Epoch: 15, Step: 90800, Step Loss: 3.51 INFO:Pipeline: Time: 09:30:34, Epoch: 15, Step: 90900, Step Loss: 3.58 INFO:Pipeline: Time: 09:31:13, Epoch: 15, Step: 91000, Step Loss: 3.53 INFO:Pipeline: Eval Loss: 3.39, Perplexity: 29.67 INFO:Pipeline: Time: 09:31:55, Epoch: 15, Step: 91100, Step Loss: 3.61 INFO:Pipeline: Time: 09:32:34, Epoch: 15, Step: 91200, Step Loss: 3.57 INFO:Pipeline: Time: 09:33:12, Epoch: 15, Step: 91300, Step Loss: 3.52 INFO:Pipeline: Time: 09:33:50, Epoch: 15, Step: 91400, Step Loss: 3.63 INFO:Pipeline: Time: 09:34:29, Epoch: 15, Step: 91500, Step Loss: 3.57 INFO:Pipeline: Time: 09:35:07, Epoch: 16, Step: 91600, Step Loss: 3.60 INFO:Pipeline: Time: 09:35:45, Epoch: 16, Step: 91700, Step Loss: 3.39 INFO:Pipeline: Time: 09:36:23, Epoch: 16, Step: 91800, Step Loss: 3.40 INFO:Pipeline: Time: 09:37:02, Epoch: 16, Step: 91900, Step Loss: 3.46 INFO:Pipeline: Time: 09:37:40, Epoch: 16, Step: 92000, Step Loss: 3.60 INFO:Pipeline: Eval Loss: 3.39, Perplexity: 29.60 INFO:Pipeline: Time: 09:38:22, Epoch: 16, Step: 92100, Step Loss: 3.54 INFO:Pipeline: Time: 09:39:01, Epoch: 16, Step: 92200, Step Loss: 3.36 INFO:Pipeline: Time: 09:39:39, Epoch: 16, Step: 92300, Step Loss: 3.62 INFO:Pipeline: Time: 09:40:17, Epoch: 16, Step: 92400, Step Loss: 3.35 INFO:Pipeline: Time: 09:40:55, Epoch: 16, Step: 92500, Step Loss: 3.49 INFO:Pipeline: Time: 09:41:34, Epoch: 16, Step: 92600, Step Loss: 3.50 INFO:Pipeline: Time: 09:42:12, Epoch: 16, Step: 92700, Step Loss: 3.21 INFO:Pipeline: Time: 09:42:50, Epoch: 16, Step: 92800, Step Loss: 3.54 INFO:Pipeline: Time: 09:43:28, Epoch: 16, Step: 92900, Step Loss: 3.48 INFO:Pipeline: Time: 09:44:07, Epoch: 16, Step: 93000, Step Loss: 3.42 INFO:Pipeline: Eval Loss: 3.39, Perplexity: 29.55 INFO:Pipeline: Time: 09:44:50, Epoch: 16, Step: 93100, Step Loss: 3.36 INFO:Pipeline: Time: 09:45:28, Epoch: 16, Step: 93200, Step Loss: 3.48 INFO:Pipeline: Time: 09:46:06, Epoch: 16, Step: 93300, Step Loss: 3.45 INFO:Pipeline: Time: 09:46:44, Epoch: 16, Step: 93400, Step Loss: 3.71 INFO:Pipeline: Time: 09:47:23, Epoch: 16, Step: 93500, Step Loss: 3.60 INFO:Pipeline: Time: 09:48:01, Epoch: 16, Step: 93600, Step Loss: 3.43 INFO:Pipeline: Time: 09:48:39, Epoch: 16, Step: 93700, Step Loss: 3.52 INFO:Pipeline: Time: 09:49:17, Epoch: 16, Step: 93800, Step Loss: 3.42 INFO:Pipeline: Time: 09:49:56, Epoch: 16, Step: 93900, Step Loss: 3.51 INFO:Pipeline: Time: 09:50:34, Epoch: 16, Step: 94000, Step Loss: 3.53 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.49 INFO:Pipeline: Time: 09:51:17, Epoch: 16, Step: 94100, Step Loss: 3.48 INFO:Pipeline: Time: 09:51:55, Epoch: 16, Step: 94200, Step Loss: 3.35 INFO:Pipeline: Time: 09:52:33, Epoch: 16, Step: 94300, Step Loss: 3.65 INFO:Pipeline: Time: 09:53:11, Epoch: 16, Step: 94400, Step Loss: 3.52 INFO:Pipeline: Time: 09:53:50, Epoch: 16, Step: 94500, Step Loss: 3.58 INFO:Pipeline: Time: 09:54:28, Epoch: 16, Step: 94600, Step Loss: 3.31 INFO:Pipeline: Time: 09:55:06, Epoch: 16, Step: 94700, Step Loss: 3.44 INFO:Pipeline: Time: 09:55:44, Epoch: 16, Step: 94800, Step Loss: 3.46 INFO:Pipeline: Time: 09:56:23, Epoch: 16, Step: 94900, Step Loss: 3.47 INFO:Pipeline: Time: 09:57:01, Epoch: 16, Step: 95000, Step Loss: 3.49 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.45 INFO:Pipeline: Time: 09:57:44, Epoch: 16, Step: 95100, Step Loss: 3.48 INFO:Pipeline: Time: 09:58:22, Epoch: 16, Step: 95200, Step Loss: 3.50 INFO:Pipeline: Time: 09:59:00, Epoch: 16, Step: 95300, Step Loss: 3.59 INFO:Pipeline: Time: 09:59:38, Epoch: 16, Step: 95400, Step Loss: 3.48 INFO:Pipeline: Time: 10:00:16, Epoch: 16, Step: 95500, Step Loss: 3.58 INFO:Pipeline: Time: 10:00:55, Epoch: 16, Step: 95600, Step Loss: 3.54 INFO:Pipeline: Time: 10:01:33, Epoch: 16, Step: 95700, Step Loss: 3.53 INFO:Pipeline: Time: 10:02:11, Epoch: 16, Step: 95800, Step Loss: 3.54 INFO:Pipeline: Time: 10:02:49, Epoch: 16, Step: 95900, Step Loss: 3.30 INFO:Pipeline: Time: 10:03:28, Epoch: 16, Step: 96000, Step Loss: 3.47 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.42 INFO:Pipeline: Time: 10:04:10, Epoch: 16, Step: 96100, Step Loss: 3.45 INFO:Pipeline: Time: 10:04:49, Epoch: 16, Step: 96200, Step Loss: 3.58 INFO:Pipeline: Time: 10:05:27, Epoch: 16, Step: 96300, Step Loss: 3.38 INFO:Pipeline: Time: 10:06:05, Epoch: 16, Step: 96400, Step Loss: 3.62 INFO:Pipeline: Time: 10:06:43, Epoch: 16, Step: 96500, Step Loss: 3.47 INFO:Pipeline: Time: 10:07:21, Epoch: 16, Step: 96600, Step Loss: 3.52 INFO:Pipeline: Time: 10:07:58, Epoch: 16, Step: 96700, Step Loss: 3.56 INFO:Pipeline: Time: 10:08:36, Epoch: 16, Step: 96800, Step Loss: 3.42 INFO:Pipeline: Time: 10:09:13, Epoch: 16, Step: 96900, Step Loss: 3.20 INFO:Pipeline: Time: 10:09:50, Epoch: 16, Step: 97000, Step Loss: 3.57 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.37 INFO:Pipeline: Time: 10:10:31, Epoch: 16, Step: 97100, Step Loss: 3.63 INFO:Pipeline: Time: 10:11:08, Epoch: 16, Step: 97200, Step Loss: 3.57 INFO:Pipeline: Time: 10:11:45, Epoch: 16, Step: 97300, Step Loss: 3.50 INFO:Pipeline: Time: 10:12:23, Epoch: 16, Step: 97400, Step Loss: 3.60 INFO:Pipeline: Time: 10:13:00, Epoch: 16, Step: 97500, Step Loss: 3.51 INFO:Pipeline: Time: 10:13:37, Epoch: 16, Step: 97600, Step Loss: 3.32 INFO:Pipeline: Time: 10:14:14, Epoch: 17, Step: 97700, Step Loss: 3.56 INFO:Pipeline: Time: 10:14:51, Epoch: 17, Step: 97800, Step Loss: 3.42 INFO:Pipeline: Time: 10:15:28, Epoch: 17, Step: 97900, Step Loss: 3.56 INFO:Pipeline: Time: 10:16:06, Epoch: 17, Step: 98000, Step Loss: 3.61 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.36 INFO:Pipeline: Time: 10:16:46, Epoch: 17, Step: 98100, Step Loss: 3.51 INFO:Pipeline: Time: 10:17:24, Epoch: 17, Step: 98200, Step Loss: 3.56 INFO:Pipeline: Time: 10:18:01, Epoch: 17, Step: 98300, Step Loss: 3.42 INFO:Pipeline: Time: 10:18:38, Epoch: 17, Step: 98400, Step Loss: 3.62 INFO:Pipeline: Time: 10:19:15, Epoch: 17, Step: 98500, Step Loss: 3.37 INFO:Pipeline: Time: 10:19:52, Epoch: 17, Step: 98600, Step Loss: 3.49 INFO:Pipeline: Time: 10:20:29, Epoch: 17, Step: 98700, Step Loss: 3.51 INFO:Pipeline: Time: 10:21:06, Epoch: 17, Step: 98800, Step Loss: 3.50 INFO:Pipeline: Time: 10:21:44, Epoch: 17, Step: 98900, Step Loss: 3.53 INFO:Pipeline: Time: 10:22:21, Epoch: 17, Step: 99000, Step Loss: 3.48 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.32 INFO:Pipeline: Time: 10:23:01, Epoch: 17, Step: 99100, Step Loss: 3.44 INFO:Pipeline: Time: 10:23:39, Epoch: 17, Step: 99200, Step Loss: 3.53 INFO:Pipeline: Time: 10:24:16, Epoch: 17, Step: 99300, Step Loss: 3.34 INFO:Pipeline: Time: 10:24:53, Epoch: 17, Step: 99400, Step Loss: 3.61 INFO:Pipeline: Time: 10:25:30, Epoch: 17, Step: 99500, Step Loss: 3.51 INFO:Pipeline: Time: 10:26:07, Epoch: 17, Step: 99600, Step Loss: 3.37 INFO:Pipeline: Time: 10:26:45, Epoch: 17, Step: 99700, Step Loss: 3.63 INFO:Pipeline: Time: 10:27:22, Epoch: 17, Step: 99800, Step Loss: 3.54 INFO:Pipeline: Time: 10:27:59, Epoch: 17, Step: 99900, Step Loss: 3.41 INFO:Pipeline: Time: 10:28:37, Epoch: 17, Step: 100000, Step Loss: 3.44 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.30 INFO:Pipeline: Time: 10:29:17, Epoch: 17, Step: 100100, Step Loss: 3.50 INFO:Pipeline: Time: 10:29:55, Epoch: 17, Step: 100200, Step Loss: 3.61 INFO:Pipeline: Time: 10:30:32, Epoch: 17, Step: 100300, Step Loss: 3.43 INFO:Pipeline: Time: 10:31:09, Epoch: 17, Step: 100400, Step Loss: 3.42 INFO:Pipeline: Time: 10:31:47, Epoch: 17, Step: 100500, Step Loss: 3.55 INFO:Pipeline: Time: 10:32:24, Epoch: 17, Step: 100600, Step Loss: 3.44 INFO:Pipeline: Time: 10:33:01, Epoch: 17, Step: 100700, Step Loss: 3.42 INFO:Pipeline: Time: 10:33:39, Epoch: 17, Step: 100800, Step Loss: 3.49 INFO:Pipeline: Time: 10:34:16, Epoch: 17, Step: 100900, Step Loss: 3.59 INFO:Pipeline: Time: 10:34:53, Epoch: 17, Step: 101000, Step Loss: 3.33 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.26 INFO:Pipeline: Time: 10:35:34, Epoch: 17, Step: 101100, Step Loss: 3.38 INFO:Pipeline: Time: 10:36:11, Epoch: 17, Step: 101200, Step Loss: 3.45 INFO:Pipeline: Time: 10:36:49, Epoch: 17, Step: 101300, Step Loss: 3.67 INFO:Pipeline: Time: 10:37:26, Epoch: 17, Step: 101400, Step Loss: 3.23 INFO:Pipeline: Time: 10:38:03, Epoch: 17, Step: 101500, Step Loss: 3.47 INFO:Pipeline: Time: 10:38:41, Epoch: 17, Step: 101600, Step Loss: 3.69 INFO:Pipeline: Time: 10:39:18, Epoch: 17, Step: 101700, Step Loss: 3.65 INFO:Pipeline: Time: 10:39:55, Epoch: 17, Step: 101800, Step Loss: 3.50 INFO:Pipeline: Time: 10:40:33, Epoch: 17, Step: 101900, Step Loss: 3.52 INFO:Pipeline: Time: 10:41:10, Epoch: 17, Step: 102000, Step Loss: 3.61 INFO:Pipeline: Eval Loss: 3.38, Perplexity: 29.23 INFO:Pipeline: Time: 10:41:50, Epoch: 17, Step: 102100, Step Loss: 3.32 INFO:Pipeline: Time: 10:42:28, Epoch: 17, Step: 102200, Step Loss: 3.48 INFO:Pipeline: Time: 10:43:05, Epoch: 17, Step: 102300, Step Loss: 3.44 INFO:Pipeline: Time: 10:43:43, Epoch: 17, Step: 102400, Step Loss: 3.47 INFO:Pipeline: Time: 10:44:20, Epoch: 17, Step: 102500, Step Loss: 3.43 INFO:Pipeline: Time: 10:44:57, Epoch: 17, Step: 102600, Step Loss: 3.44 INFO:Pipeline: Time: 10:45:35, Epoch: 17, Step: 102700, Step Loss: 3.53 INFO:Pipeline: Time: 10:46:12, Epoch: 17, Step: 102800, Step Loss: 3.45 INFO:Pipeline: Time: 10:46:49, Epoch: 17, Step: 102900, Step Loss: 3.47 INFO:Pipeline: Time: 10:47:27, Epoch: 17, Step: 103000, Step Loss: 3.54 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.20 INFO:Pipeline: Time: 10:48:07, Epoch: 17, Step: 103100, Step Loss: 3.43 INFO:Pipeline: Time: 10:48:44, Epoch: 17, Step: 103200, Step Loss: 3.51 INFO:Pipeline: Time: 10:49:22, Epoch: 17, Step: 103300, Step Loss: 3.69 INFO:Pipeline: Time: 10:49:59, Epoch: 17, Step: 103400, Step Loss: 3.48 INFO:Pipeline: Time: 10:50:36, Epoch: 17, Step: 103500, Step Loss: 3.44 INFO:Pipeline: Time: 10:51:14, Epoch: 17, Step: 103600, Step Loss: 3.57 INFO:Pipeline: Time: 10:51:51, Epoch: 17, Step: 103700, Step Loss: 3.62 INFO:Pipeline: Time: 10:52:28, Epoch: 18, Step: 103800, Step Loss: 3.43 INFO:Pipeline: Time: 10:53:06, Epoch: 18, Step: 103900, Step Loss: 3.61 INFO:Pipeline: Time: 10:53:43, Epoch: 18, Step: 104000, Step Loss: 3.42 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.18 INFO:Pipeline: Time: 10:54:23, Epoch: 18, Step: 104100, Step Loss: 3.42 INFO:Pipeline: Time: 10:55:01, Epoch: 18, Step: 104200, Step Loss: 3.48 INFO:Pipeline: Time: 10:55:38, Epoch: 18, Step: 104300, Step Loss: 3.54 INFO:Pipeline: Time: 10:56:15, Epoch: 18, Step: 104400, Step Loss: 3.50 INFO:Pipeline: Time: 10:56:53, Epoch: 18, Step: 104500, Step Loss: 3.29 INFO:Pipeline: Time: 10:57:30, Epoch: 18, Step: 104600, Step Loss: 3.29 INFO:Pipeline: Time: 10:58:07, Epoch: 18, Step: 104700, Step Loss: 3.76 INFO:Pipeline: Time: 10:58:45, Epoch: 18, Step: 104800, Step Loss: 3.52 INFO:Pipeline: Time: 10:59:22, Epoch: 18, Step: 104900, Step Loss: 3.47 INFO:Pipeline: Time: 10:59:59, Epoch: 18, Step: 105000, Step Loss: 3.53 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.16 INFO:Pipeline: Time: 11:00:40, Epoch: 18, Step: 105100, Step Loss: 3.54 INFO:Pipeline: Time: 11:01:17, Epoch: 18, Step: 105200, Step Loss: 3.60 INFO:Pipeline: Time: 11:01:55, Epoch: 18, Step: 105300, Step Loss: 3.41 INFO:Pipeline: Time: 11:02:32, Epoch: 18, Step: 105400, Step Loss: 3.60 INFO:Pipeline: Time: 11:03:09, Epoch: 18, Step: 105500, Step Loss: 3.48 INFO:Pipeline: Time: 11:03:47, Epoch: 18, Step: 105600, Step Loss: 3.64 INFO:Pipeline: Time: 11:04:24, Epoch: 18, Step: 105700, Step Loss: 3.30 INFO:Pipeline: Time: 11:05:01, Epoch: 18, Step: 105800, Step Loss: 3.43 INFO:Pipeline: Time: 11:05:39, Epoch: 18, Step: 105900, Step Loss: 3.44 INFO:Pipeline: Time: 11:06:16, Epoch: 18, Step: 106000, Step Loss: 3.63 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.14 INFO:Pipeline: Time: 11:06:57, Epoch: 18, Step: 106100, Step Loss: 3.43 INFO:Pipeline: Time: 11:07:34, Epoch: 18, Step: 106200, Step Loss: 3.62 INFO:Pipeline: Time: 11:08:11, Epoch: 18, Step: 106300, Step Loss: 3.32 INFO:Pipeline: Time: 11:08:49, Epoch: 18, Step: 106400, Step Loss: 3.40 INFO:Pipeline: Time: 11:09:26, Epoch: 18, Step: 106500, Step Loss: 3.57 INFO:Pipeline: Time: 11:10:03, Epoch: 18, Step: 106600, Step Loss: 3.53 INFO:Pipeline: Time: 11:10:41, Epoch: 18, Step: 106700, Step Loss: 3.49 INFO:Pipeline: Time: 11:11:18, Epoch: 18, Step: 106800, Step Loss: 3.48 INFO:Pipeline: Time: 11:11:55, Epoch: 18, Step: 106900, Step Loss: 3.34 INFO:Pipeline: Time: 11:12:33, Epoch: 18, Step: 107000, Step Loss: 3.37 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.13 INFO:Pipeline: Time: 11:13:13, Epoch: 18, Step: 107100, Step Loss: 3.57 INFO:Pipeline: Time: 11:13:50, Epoch: 18, Step: 107200, Step Loss: 3.34 INFO:Pipeline: Time: 11:14:28, Epoch: 18, Step: 107300, Step Loss: 3.46 INFO:Pipeline: Time: 11:15:05, Epoch: 18, Step: 107400, Step Loss: 3.46 INFO:Pipeline: Time: 11:15:42, Epoch: 18, Step: 107500, Step Loss: 3.48 INFO:Pipeline: Time: 11:16:20, Epoch: 18, Step: 107600, Step Loss: 3.64 INFO:Pipeline: Time: 11:16:57, Epoch: 18, Step: 107700, Step Loss: 3.53 INFO:Pipeline: Time: 11:17:34, Epoch: 18, Step: 107800, Step Loss: 3.58 INFO:Pipeline: Time: 11:18:12, Epoch: 18, Step: 107900, Step Loss: 3.19 INFO:Pipeline: Time: 11:18:49, Epoch: 18, Step: 108000, Step Loss: 3.28 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.11 INFO:Pipeline: Time: 11:19:29, Epoch: 18, Step: 108100, Step Loss: 3.34 INFO:Pipeline: Time: 11:20:07, Epoch: 18, Step: 108200, Step Loss: 3.54 INFO:Pipeline: Time: 11:20:44, Epoch: 18, Step: 108300, Step Loss: 3.36 INFO:Pipeline: Time: 11:21:21, Epoch: 18, Step: 108400, Step Loss: 3.44 INFO:Pipeline: Time: 11:21:59, Epoch: 18, Step: 108500, Step Loss: 3.56 INFO:Pipeline: Time: 11:22:36, Epoch: 18, Step: 108600, Step Loss: 3.43 INFO:Pipeline: Time: 11:23:14, Epoch: 18, Step: 108700, Step Loss: 3.54 INFO:Pipeline: Time: 11:23:51, Epoch: 18, Step: 108800, Step Loss: 3.46 INFO:Pipeline: Time: 11:24:28, Epoch: 18, Step: 108900, Step Loss: 3.40 INFO:Pipeline: Time: 11:25:06, Epoch: 18, Step: 109000, Step Loss: 3.62 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.10 INFO:Pipeline: Time: 11:25:46, Epoch: 18, Step: 109100, Step Loss: 3.36 INFO:Pipeline: Time: 11:26:23, Epoch: 18, Step: 109200, Step Loss: 3.43 INFO:Pipeline: Time: 11:27:00, Epoch: 18, Step: 109300, Step Loss: 3.36 INFO:Pipeline: Time: 11:27:38, Epoch: 18, Step: 109400, Step Loss: 3.52 INFO:Pipeline: Time: 11:28:15, Epoch: 18, Step: 109500, Step Loss: 3.60 INFO:Pipeline: Time: 11:28:52, Epoch: 18, Step: 109600, Step Loss: 3.49 INFO:Pipeline: Time: 11:29:30, Epoch: 18, Step: 109700, Step Loss: 3.57 INFO:Pipeline: Time: 11:30:07, Epoch: 18, Step: 109800, Step Loss: 3.35 INFO:Pipeline: Time: 11:30:44, Epoch: 19, Step: 109900, Step Loss: 3.45 INFO:Pipeline: Time: 11:31:21, Epoch: 19, Step: 110000, Step Loss: 3.61 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.11 INFO:Pipeline: Time: 11:32:01, Epoch: 19, Step: 110100, Step Loss: 3.52 INFO:Pipeline: Time: 11:32:38, Epoch: 19, Step: 110200, Step Loss: 3.38 INFO:Pipeline: Time: 11:33:15, Epoch: 19, Step: 110300, Step Loss: 3.40 INFO:Pipeline: Time: 11:33:52, Epoch: 19, Step: 110400, Step Loss: 3.42 INFO:Pipeline: Time: 11:34:30, Epoch: 19, Step: 110500, Step Loss: 3.51 INFO:Pipeline: Time: 11:35:07, Epoch: 19, Step: 110600, Step Loss: 3.38 INFO:Pipeline: Time: 11:35:44, Epoch: 19, Step: 110700, Step Loss: 3.36 INFO:Pipeline: Time: 11:36:22, Epoch: 19, Step: 110800, Step Loss: 3.55 INFO:Pipeline: Time: 11:36:59, Epoch: 19, Step: 110900, Step Loss: 3.35 INFO:Pipeline: Time: 11:37:36, Epoch: 19, Step: 111000, Step Loss: 3.50 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.08 INFO:Pipeline: Time: 11:38:17, Epoch: 19, Step: 111100, Step Loss: 3.57 INFO:Pipeline: Time: 11:38:54, Epoch: 19, Step: 111200, Step Loss: 3.48 INFO:Pipeline: Time: 11:39:31, Epoch: 19, Step: 111300, Step Loss: 3.45 INFO:Pipeline: Time: 11:40:09, Epoch: 19, Step: 111400, Step Loss: 3.54 INFO:Pipeline: Time: 11:40:46, Epoch: 19, Step: 111500, Step Loss: 3.42 INFO:Pipeline: Time: 11:41:23, Epoch: 19, Step: 111600, Step Loss: 3.61 INFO:Pipeline: Time: 11:42:01, Epoch: 19, Step: 111700, Step Loss: 3.49 INFO:Pipeline: Time: 11:42:38, Epoch: 19, Step: 111800, Step Loss: 3.59 INFO:Pipeline: Time: 11:43:15, Epoch: 19, Step: 111900, Step Loss: 3.43 INFO:Pipeline: Time: 11:43:53, Epoch: 19, Step: 112000, Step Loss: 3.55 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.08 INFO:Pipeline: Time: 11:44:33, Epoch: 19, Step: 112100, Step Loss: 3.28 INFO:Pipeline: Time: 11:45:10, Epoch: 19, Step: 112200, Step Loss: 3.48 INFO:Pipeline: Time: 11:45:47, Epoch: 19, Step: 112300, Step Loss: 3.55 INFO:Pipeline: Time: 11:46:25, Epoch: 19, Step: 112400, Step Loss: 3.71 INFO:Pipeline: Time: 11:47:02, Epoch: 19, Step: 112500, Step Loss: 3.43 INFO:Pipeline: Time: 11:47:39, Epoch: 19, Step: 112600, Step Loss: 3.29 INFO:Pipeline: Time: 11:48:17, Epoch: 19, Step: 112700, Step Loss: 3.48 INFO:Pipeline: Time: 11:48:54, Epoch: 19, Step: 112800, Step Loss: 3.46 INFO:Pipeline: Time: 11:49:31, Epoch: 19, Step: 112900, Step Loss: 3.43 INFO:Pipeline: Time: 11:50:08, Epoch: 19, Step: 113000, Step Loss: 3.40 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.06 INFO:Pipeline: Time: 11:50:49, Epoch: 19, Step: 113100, Step Loss: 3.58 INFO:Pipeline: Time: 11:51:26, Epoch: 19, Step: 113200, Step Loss: 3.32 INFO:Pipeline: Time: 11:52:04, Epoch: 19, Step: 113300, Step Loss: 3.64 INFO:Pipeline: Time: 11:52:41, Epoch: 19, Step: 113400, Step Loss: 3.67 INFO:Pipeline: Time: 11:53:18, Epoch: 19, Step: 113500, Step Loss: 3.64 INFO:Pipeline: Time: 11:53:56, Epoch: 19, Step: 113600, Step Loss: 3.44 INFO:Pipeline: Time: 11:54:33, Epoch: 19, Step: 113700, Step Loss: 3.46 INFO:Pipeline: Time: 11:55:10, Epoch: 19, Step: 113800, Step Loss: 3.57 INFO:Pipeline: Time: 11:55:48, Epoch: 19, Step: 113900, Step Loss: 3.54 INFO:Pipeline: Time: 11:56:25, Epoch: 19, Step: 114000, Step Loss: 3.54 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.06 INFO:Pipeline: Time: 11:57:04, Epoch: 19, Step: 114100, Step Loss: 3.33 INFO:Pipeline: Time: 11:57:42, Epoch: 19, Step: 114200, Step Loss: 3.30 INFO:Pipeline: Time: 11:58:19, Epoch: 19, Step: 114300, Step Loss: 3.46 INFO:Pipeline: Time: 11:58:56, Epoch: 19, Step: 114400, Step Loss: 3.34 INFO:Pipeline: Time: 11:59:34, Epoch: 19, Step: 114500, Step Loss: 3.74 INFO:Pipeline: Time: 12:00:11, Epoch: 19, Step: 114600, Step Loss: 3.56 INFO:Pipeline: Time: 12:00:48, Epoch: 19, Step: 114700, Step Loss: 3.37 INFO:Pipeline: Time: 12:01:26, Epoch: 19, Step: 114800, Step Loss: 3.62 INFO:Pipeline: Time: 12:02:03, Epoch: 19, Step: 114900, Step Loss: 3.58 INFO:Pipeline: Time: 12:02:40, Epoch: 19, Step: 115000, Step Loss: 3.47 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.06 INFO:Pipeline: Time: 12:03:21, Epoch: 19, Step: 115100, Step Loss: 3.53 INFO:Pipeline: Time: 12:03:58, Epoch: 19, Step: 115200, Step Loss: 3.73 INFO:Pipeline: Time: 12:04:35, Epoch: 19, Step: 115300, Step Loss: 3.55 INFO:Pipeline: Time: 12:05:13, Epoch: 19, Step: 115400, Step Loss: 3.46 INFO:Pipeline: Time: 12:05:50, Epoch: 19, Step: 115500, Step Loss: 3.36 INFO:Pipeline: Time: 12:06:27, Epoch: 19, Step: 115600, Step Loss: 3.54 INFO:Pipeline: Time: 12:07:05, Epoch: 19, Step: 115700, Step Loss: 3.42 INFO:Pipeline: Time: 12:07:42, Epoch: 19, Step: 115800, Step Loss: 3.31 INFO:Pipeline: Time: 12:08:20, Epoch: 19, Step: 115900, Step Loss: 3.48 INFO:Pipeline: Time: 12:08:57, Epoch: 20, Step: 116000, Step Loss: 3.45 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.06 INFO:Pipeline: Time: 12:09:37, Epoch: 20, Step: 116100, Step Loss: 3.36 INFO:Pipeline: Time: 12:10:14, Epoch: 20, Step: 116200, Step Loss: 3.52 INFO:Pipeline: Time: 12:10:52, Epoch: 20, Step: 116300, Step Loss: 3.58 INFO:Pipeline: Time: 12:11:29, Epoch: 20, Step: 116400, Step Loss: 3.51 INFO:Pipeline: Time: 12:12:06, Epoch: 20, Step: 116500, Step Loss: 3.37 INFO:Pipeline: Time: 12:12:44, Epoch: 20, Step: 116600, Step Loss: 3.43 INFO:Pipeline: Time: 12:13:21, Epoch: 20, Step: 116700, Step Loss: 3.41 INFO:Pipeline: Time: 12:13:58, Epoch: 20, Step: 116800, Step Loss: 3.53 INFO:Pipeline: Time: 12:14:36, Epoch: 20, Step: 116900, Step Loss: 3.35 INFO:Pipeline: Time: 12:15:13, Epoch: 20, Step: 117000, Step Loss: 3.71 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.05 INFO:Pipeline: Time: 12:15:53, Epoch: 20, Step: 117100, Step Loss: 3.46 INFO:Pipeline: Time: 12:16:31, Epoch: 20, Step: 117200, Step Loss: 3.59 INFO:Pipeline: Time: 12:17:08, Epoch: 20, Step: 117300, Step Loss: 3.33 INFO:Pipeline: Time: 12:17:45, Epoch: 20, Step: 117400, Step Loss: 3.75 INFO:Pipeline: Time: 12:18:22, Epoch: 20, Step: 117500, Step Loss: 3.48 INFO:Pipeline: Time: 12:19:00, Epoch: 20, Step: 117600, Step Loss: 3.37 INFO:Pipeline: Time: 12:19:37, Epoch: 20, Step: 117700, Step Loss: 3.56 INFO:Pipeline: Time: 12:20:14, Epoch: 20, Step: 117800, Step Loss: 3.55 INFO:Pipeline: Time: 12:20:51, Epoch: 20, Step: 117900, Step Loss: 3.59 INFO:Pipeline: Time: 12:21:29, Epoch: 20, Step: 118000, Step Loss: 3.48 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.05 INFO:Pipeline: Time: 12:22:08, Epoch: 20, Step: 118100, Step Loss: 3.47 INFO:Pipeline: Time: 12:22:45, Epoch: 20, Step: 118200, Step Loss: 3.32 INFO:Pipeline: Time: 12:23:23, Epoch: 20, Step: 118300, Step Loss: 3.33 INFO:Pipeline: Time: 12:24:00, Epoch: 20, Step: 118400, Step Loss: 3.37 INFO:Pipeline: Time: 12:24:37, Epoch: 20, Step: 118500, Step Loss: 3.40 INFO:Pipeline: Time: 12:25:14, Epoch: 20, Step: 118600, Step Loss: 3.45 INFO:Pipeline: Time: 12:25:52, Epoch: 20, Step: 118700, Step Loss: 3.52 INFO:Pipeline: Time: 12:26:29, Epoch: 20, Step: 118800, Step Loss: 3.42 INFO:Pipeline: Time: 12:27:06, Epoch: 20, Step: 118900, Step Loss: 3.36 INFO:Pipeline: Time: 12:27:44, Epoch: 20, Step: 119000, Step Loss: 3.46 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.05 INFO:Pipeline: Time: 12:28:23, Epoch: 20, Step: 119100, Step Loss: 3.47 INFO:Pipeline: Time: 12:29:00, Epoch: 20, Step: 119200, Step Loss: 3.61 INFO:Pipeline: Time: 12:29:37, Epoch: 20, Step: 119300, Step Loss: 3.38 INFO:Pipeline: Time: 12:30:15, Epoch: 20, Step: 119400, Step Loss: 3.34 INFO:Pipeline: Time: 12:30:52, Epoch: 20, Step: 119500, Step Loss: 3.46 INFO:Pipeline: Time: 12:31:29, Epoch: 20, Step: 119600, Step Loss: 3.30 INFO:Pipeline: Time: 12:32:07, Epoch: 20, Step: 119700, Step Loss: 3.45 INFO:Pipeline: Time: 12:32:44, Epoch: 20, Step: 119800, Step Loss: 3.50 INFO:Pipeline: Time: 12:33:21, Epoch: 20, Step: 119900, Step Loss: 3.56 INFO:Pipeline: Time: 12:33:58, Epoch: 20, Step: 120000, Step Loss: 3.32 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.05 INFO:Pipeline: Time: 12:34:39, Epoch: 20, Step: 120100, Step Loss: 3.57 INFO:Pipeline: Time: 12:35:16, Epoch: 20, Step: 120200, Step Loss: 3.46 INFO:Pipeline: Time: 12:35:53, Epoch: 20, Step: 120300, Step Loss: 3.34 INFO:Pipeline: Time: 12:36:31, Epoch: 20, Step: 120400, Step Loss: 3.51 INFO:Pipeline: Time: 12:37:08, Epoch: 20, Step: 120500, Step Loss: 3.47 INFO:Pipeline: Time: 12:37:45, Epoch: 20, Step: 120600, Step Loss: 3.55 INFO:Pipeline: Time: 12:38:22, Epoch: 20, Step: 120700, Step Loss: 3.61 INFO:Pipeline: Time: 12:39:00, Epoch: 20, Step: 120800, Step Loss: 3.40 INFO:Pipeline: Time: 12:39:37, Epoch: 20, Step: 120900, Step Loss: 3.37 INFO:Pipeline: Time: 12:40:14, Epoch: 20, Step: 121000, Step Loss: 3.51 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.05 INFO:Pipeline: Time: 12:40:54, Epoch: 20, Step: 121100, Step Loss: 3.34 INFO:Pipeline: Time: 12:41:32, Epoch: 20, Step: 121200, Step Loss: 3.37 INFO:Pipeline: Time: 12:42:09, Epoch: 20, Step: 121300, Step Loss: 3.50 INFO:Pipeline: Time: 12:42:46, Epoch: 20, Step: 121400, Step Loss: 3.36 INFO:Pipeline: Time: 12:43:23, Epoch: 20, Step: 121500, Step Loss: 3.51 INFO:Pipeline: Time: 12:44:01, Epoch: 20, Step: 121600, Step Loss: 3.61 INFO:Pipeline: Time: 12:44:38, Epoch: 20, Step: 121700, Step Loss: 3.49 INFO:Pipeline: Time: 12:45:15, Epoch: 20, Step: 121800, Step Loss: 3.44 INFO:Pipeline: Time: 12:45:53, Epoch: 20, Step: 121900, Step Loss: 3.31 INFO:Pipeline: Time: 12:46:30, Epoch: 20, Step: 122000, Step Loss: 3.68 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.05 INFO:Pipeline: Eval Loss: 3.37, Perplexity: 29.05