INFO:Pipeline: Preparing Repository... WARNING:huggingface_hub.repository:Cloning https://huggingface.co/benlipkin/gpt2_64_wikitext_100M_20_5cd4da41b7fe7e3d 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: 62435.97 INFO:Pipeline: Time: 00:00:49, Epoch: 1, Step: 100, Step Loss: 10.28 INFO:Pipeline: Time: 00:01:12, Epoch: 1, Step: 200, Step Loss: 9.57 INFO:Pipeline: Time: 00:01:35, Epoch: 1, Step: 300, Step Loss: 9.18 INFO:Pipeline: Time: 00:01:59, Epoch: 1, Step: 400, Step Loss: 8.71 INFO:Pipeline: Time: 00:02:22, Epoch: 1, Step: 500, Step Loss: 8.54 INFO:Pipeline: Time: 00:02:45, Epoch: 1, Step: 600, Step Loss: 8.40 INFO:Pipeline: Time: 00:03:09, Epoch: 1, Step: 700, Step Loss: 8.16 INFO:Pipeline: Time: 00:03:32, Epoch: 1, Step: 800, Step Loss: 7.95 INFO:Pipeline: Time: 00:03:56, Epoch: 1, Step: 900, Step Loss: 7.74 INFO:Pipeline: Time: 00:04:19, Epoch: 1, Step: 1000, Step Loss: 7.61 INFO:Pipeline: Eval Loss: 7.41, Perplexity: 1646.83 INFO:Pipeline: Time: 00:04:45, Epoch: 1, Step: 1100, Step Loss: 7.37 INFO:Pipeline: Time: 00:05:08, Epoch: 1, Step: 1200, Step Loss: 7.15 INFO:Pipeline: Time: 00:05:31, Epoch: 1, Step: 1300, Step Loss: 7.05 INFO:Pipeline: Time: 00:05:55, Epoch: 1, Step: 1400, Step Loss: 6.86 INFO:Pipeline: Time: 00:06:18, Epoch: 1, Step: 1500, Step Loss: 6.72 INFO:Pipeline: Time: 00:06:42, Epoch: 1, Step: 1600, Step Loss: 6.66 INFO:Pipeline: Time: 00:07:05, Epoch: 1, Step: 1700, Step Loss: 6.66 INFO:Pipeline: Time: 00:07:28, Epoch: 1, Step: 1800, Step Loss: 6.62 INFO:Pipeline: Time: 00:07:52, Epoch: 1, Step: 1900, Step Loss: 6.53 INFO:Pipeline: Time: 00:08:15, Epoch: 1, Step: 2000, Step Loss: 6.44 INFO:Pipeline: Eval Loss: 6.29, Perplexity: 537.39 INFO:Pipeline: Time: 00:08:41, Epoch: 1, Step: 2100, Step Loss: 6.37 INFO:Pipeline: Time: 00:09:04, Epoch: 1, Step: 2200, Step Loss: 6.32 INFO:Pipeline: Time: 00:09:28, Epoch: 1, Step: 2300, Step Loss: 6.30 INFO:Pipeline: Time: 00:09:51, Epoch: 1, Step: 2400, Step Loss: 6.14 INFO:Pipeline: Time: 00:10:15, Epoch: 1, Step: 2500, Step Loss: 6.23 INFO:Pipeline: Time: 00:10:38, Epoch: 1, Step: 2600, Step Loss: 6.19 INFO:Pipeline: Time: 00:11:01, Epoch: 1, Step: 2700, Step Loss: 6.15 INFO:Pipeline: Time: 00:11:25, Epoch: 1, Step: 2800, Step Loss: 6.07 INFO:Pipeline: Time: 00:11:48, Epoch: 1, Step: 2900, Step Loss: 6.03 INFO:Pipeline: Time: 00:12:12, Epoch: 1, Step: 3000, Step Loss: 6.00 INFO:Pipeline: Eval Loss: 5.85, Perplexity: 345.55 INFO:Pipeline: Time: 00:12:37, Epoch: 1, Step: 3100, Step Loss: 5.99 INFO:Pipeline: Time: 00:13:01, Epoch: 1, Step: 3200, Step Loss: 5.89 INFO:Pipeline: Time: 00:13:24, Epoch: 1, Step: 3300, Step Loss: 5.92 INFO:Pipeline: Time: 00:13:48, Epoch: 1, Step: 3400, Step Loss: 5.85 INFO:Pipeline: Time: 00:14:11, Epoch: 1, Step: 3500, Step Loss: 5.84 INFO:Pipeline: Time: 00:14:34, Epoch: 1, Step: 3600, Step Loss: 5.84 INFO:Pipeline: Time: 00:14:58, Epoch: 1, Step: 3700, Step Loss: 5.81 INFO:Pipeline: Time: 00:15:21, Epoch: 1, Step: 3800, Step Loss: 5.76 INFO:Pipeline: Time: 00:15:44, Epoch: 1, Step: 3900, Step Loss: 5.75 INFO:Pipeline: Time: 00:16:08, Epoch: 1, Step: 4000, Step Loss: 5.75 INFO:Pipeline: Eval Loss: 5.54, Perplexity: 254.65 INFO:Pipeline: Time: 00:16:34, Epoch: 1, Step: 4100, Step Loss: 5.67 INFO:Pipeline: Time: 00:16:57, Epoch: 1, Step: 4200, Step Loss: 5.69 INFO:Pipeline: Time: 00:17:20, Epoch: 1, Step: 4300, Step Loss: 5.59 INFO:Pipeline: Time: 00:17:44, Epoch: 1, Step: 4400, Step Loss: 5.55 INFO:Pipeline: Time: 00:18:07, Epoch: 1, Step: 4500, Step Loss: 5.60 INFO:Pipeline: Time: 00:18:31, Epoch: 1, Step: 4600, Step Loss: 5.63 INFO:Pipeline: Time: 00:18:54, Epoch: 1, Step: 4700, Step Loss: 5.55 INFO:Pipeline: Time: 00:19:17, Epoch: 1, Step: 4800, Step Loss: 5.53 INFO:Pipeline: Time: 00:19:41, Epoch: 1, Step: 4900, Step Loss: 5.52 INFO:Pipeline: Time: 00:20:04, Epoch: 1, Step: 5000, Step Loss: 5.48 INFO:Pipeline: Eval Loss: 5.32, Perplexity: 204.31 INFO:Pipeline: Time: 00:20:30, Epoch: 1, Step: 5100, Step Loss: 5.53 INFO:Pipeline: Time: 00:20:54, Epoch: 1, Step: 5200, Step Loss: 5.49 INFO:Pipeline: Time: 00:21:17, Epoch: 1, Step: 5300, Step Loss: 5.40 INFO:Pipeline: Time: 00:21:40, Epoch: 1, Step: 5400, Step Loss: 5.35 INFO:Pipeline: Time: 00:22:04, Epoch: 1, Step: 5500, Step Loss: 5.38 INFO:Pipeline: Time: 00:22:27, Epoch: 1, Step: 5600, Step Loss: 5.41 INFO:Pipeline: Time: 00:22:51, Epoch: 1, Step: 5700, Step Loss: 5.38 INFO:Pipeline: Time: 00:23:14, Epoch: 1, Step: 5800, Step Loss: 5.39 INFO:Pipeline: Time: 00:23:37, Epoch: 1, Step: 5900, Step Loss: 5.44 INFO:Pipeline: Time: 00:24:01, Epoch: 1, Step: 6000, Step Loss: 5.41 INFO:Pipeline: Eval Loss: 5.16, Perplexity: 173.54 INFO:Pipeline: Time: 00:24:27, Epoch: 1, Step: 6100, Step Loss: 5.28 INFO:Pipeline: Time: 00:24:50, Epoch: 2, Step: 6200, Step Loss: 5.28 INFO:Pipeline: Time: 00:25:14, Epoch: 2, Step: 6300, Step Loss: 5.37 INFO:Pipeline: Time: 00:25:37, Epoch: 2, Step: 6400, Step Loss: 5.26 INFO:Pipeline: Time: 00:26:00, Epoch: 2, Step: 6500, Step Loss: 5.30 INFO:Pipeline: Time: 00:26:24, Epoch: 2, Step: 6600, Step Loss: 5.33 INFO:Pipeline: Time: 00:26:47, Epoch: 2, Step: 6700, Step Loss: 5.19 INFO:Pipeline: Time: 00:27:11, Epoch: 2, Step: 6800, Step Loss: 5.22 INFO:Pipeline: Time: 00:27:34, Epoch: 2, Step: 6900, Step Loss: 5.18 INFO:Pipeline: Time: 00:27:58, Epoch: 2, Step: 7000, Step Loss: 5.17 INFO:Pipeline: Eval Loss: 5.02, Perplexity: 151.59 INFO:Pipeline: Time: 00:28:23, Epoch: 2, Step: 7100, Step Loss: 5.27 INFO:Pipeline: Time: 00:28:47, Epoch: 2, Step: 7200, Step Loss: 5.23 INFO:Pipeline: Time: 00:29:10, Epoch: 2, Step: 7300, Step Loss: 5.21 INFO:Pipeline: Time: 00:29:34, Epoch: 2, Step: 7400, Step Loss: 5.08 INFO:Pipeline: Time: 00:29:57, Epoch: 2, Step: 7500, Step Loss: 5.22 INFO:Pipeline: Time: 00:30:21, Epoch: 2, Step: 7600, Step Loss: 5.11 INFO:Pipeline: Time: 00:30:44, Epoch: 2, Step: 7700, Step Loss: 5.11 INFO:Pipeline: Time: 00:31:07, Epoch: 2, Step: 7800, Step Loss: 5.10 INFO:Pipeline: Time: 00:31:31, Epoch: 2, Step: 7900, Step Loss: 5.08 INFO:Pipeline: Time: 00:31:54, Epoch: 2, Step: 8000, Step Loss: 5.13 INFO:Pipeline: Eval Loss: 4.91, Perplexity: 135.46 INFO:Pipeline: Time: 00:32:20, Epoch: 2, Step: 8100, Step Loss: 5.15 INFO:Pipeline: Time: 00:32:43, Epoch: 2, Step: 8200, Step Loss: 5.09 INFO:Pipeline: Time: 00:33:07, Epoch: 2, Step: 8300, Step Loss: 5.12 INFO:Pipeline: Time: 00:33:30, Epoch: 2, Step: 8400, Step Loss: 5.08 INFO:Pipeline: Time: 00:33:54, Epoch: 2, Step: 8500, Step Loss: 5.02 INFO:Pipeline: Time: 00:34:17, Epoch: 2, Step: 8600, Step Loss: 5.05 INFO:Pipeline: Time: 00:34:40, Epoch: 2, Step: 8700, Step Loss: 5.04 INFO:Pipeline: Time: 00:35:04, Epoch: 2, Step: 8800, Step Loss: 4.98 INFO:Pipeline: Time: 00:35:27, Epoch: 2, Step: 8900, Step Loss: 5.09 INFO:Pipeline: Time: 00:35:51, Epoch: 2, Step: 9000, Step Loss: 4.97 INFO:Pipeline: Eval Loss: 4.81, Perplexity: 123.02 INFO:Pipeline: Time: 00:36:16, Epoch: 2, Step: 9100, Step Loss: 4.95 INFO:Pipeline: Time: 00:36:40, Epoch: 2, Step: 9200, Step Loss: 5.05 INFO:Pipeline: Time: 00:37:03, Epoch: 2, Step: 9300, Step Loss: 5.05 INFO:Pipeline: Time: 00:37:27, Epoch: 2, Step: 9400, Step Loss: 4.94 INFO:Pipeline: Time: 00:37:50, Epoch: 2, Step: 9500, Step Loss: 5.00 INFO:Pipeline: Time: 00:38:14, Epoch: 2, Step: 9600, Step Loss: 4.94 INFO:Pipeline: Time: 00:38:37, Epoch: 2, Step: 9700, Step Loss: 4.88 INFO:Pipeline: Time: 00:39:01, Epoch: 2, Step: 9800, Step Loss: 4.84 INFO:Pipeline: Time: 00:39:24, Epoch: 2, Step: 9900, Step Loss: 4.94 INFO:Pipeline: Time: 00:39:48, Epoch: 2, Step: 10000, Step Loss: 4.89 INFO:Pipeline: Eval Loss: 4.73, Perplexity: 113.39 INFO:Pipeline: Time: 00:40:13, Epoch: 2, Step: 10100, Step Loss: 4.96 INFO:Pipeline: Time: 00:40:37, Epoch: 2, Step: 10200, Step Loss: 4.95 INFO:Pipeline: Time: 00:41:00, Epoch: 2, Step: 10300, Step Loss: 4.92 INFO:Pipeline: Time: 00:41:23, Epoch: 2, Step: 10400, Step Loss: 4.88 INFO:Pipeline: Time: 00:41:47, Epoch: 2, Step: 10500, Step Loss: 4.88 INFO:Pipeline: Time: 00:42:10, Epoch: 2, Step: 10600, Step Loss: 4.91 INFO:Pipeline: Time: 00:42:34, Epoch: 2, Step: 10700, Step Loss: 4.93 INFO:Pipeline: Time: 00:42:57, Epoch: 2, Step: 10800, Step Loss: 4.92 INFO:Pipeline: Time: 00:43:21, Epoch: 2, Step: 10900, Step Loss: 4.87 INFO:Pipeline: Time: 00:43:44, Epoch: 2, Step: 11000, Step Loss: 4.88 INFO:Pipeline: Eval Loss: 4.66, Perplexity: 105.38 INFO:Pipeline: Time: 00:44:10, Epoch: 2, Step: 11100, Step Loss: 4.80 INFO:Pipeline: Time: 00:44:33, Epoch: 2, Step: 11200, Step Loss: 4.80 INFO:Pipeline: Time: 00:44:56, Epoch: 2, Step: 11300, Step Loss: 4.80 INFO:Pipeline: Time: 00:45:20, Epoch: 2, Step: 11400, Step Loss: 4.82 INFO:Pipeline: Time: 00:45:43, Epoch: 2, Step: 11500, Step Loss: 4.84 INFO:Pipeline: Time: 00:46:07, Epoch: 2, Step: 11600, Step Loss: 4.81 INFO:Pipeline: Time: 00:46:30, Epoch: 2, Step: 11700, Step Loss: 4.78 INFO:Pipeline: Time: 00:46:53, Epoch: 2, Step: 11800, Step Loss: 4.84 INFO:Pipeline: Time: 00:47:17, Epoch: 2, Step: 11900, Step Loss: 4.76 INFO:Pipeline: Time: 00:47:40, Epoch: 2, Step: 12000, Step Loss: 4.78 INFO:Pipeline: Eval Loss: 4.59, Perplexity: 98.83 INFO:Pipeline: Time: 00:48:06, Epoch: 2, Step: 12100, Step Loss: 4.81 INFO:Pipeline: Time: 00:48:29, Epoch: 2, Step: 12200, Step Loss: 4.74 INFO:Pipeline: Time: 00:48:53, Epoch: 3, Step: 12300, Step Loss: 4.70 INFO:Pipeline: Time: 00:49:16, Epoch: 3, Step: 12400, Step Loss: 4.76 INFO:Pipeline: Time: 00:49:39, Epoch: 3, Step: 12500, Step Loss: 4.73 INFO:Pipeline: Time: 00:50:03, Epoch: 3, Step: 12600, Step Loss: 4.71 INFO:Pipeline: Time: 00:50:26, Epoch: 3, Step: 12700, Step Loss: 4.81 INFO:Pipeline: Time: 00:50:50, Epoch: 3, Step: 12800, Step Loss: 4.75 INFO:Pipeline: Time: 00:51:13, Epoch: 3, Step: 12900, Step Loss: 4.77 INFO:Pipeline: Time: 00:51:36, Epoch: 3, Step: 13000, Step Loss: 4.70 INFO:Pipeline: Eval Loss: 4.54, Perplexity: 93.28 INFO:Pipeline: Time: 00:52:02, Epoch: 3, Step: 13100, Step Loss: 4.80 INFO:Pipeline: Time: 00:52:25, Epoch: 3, Step: 13200, Step Loss: 4.66 INFO:Pipeline: Time: 00:52:49, Epoch: 3, Step: 13300, Step Loss: 4.71 INFO:Pipeline: Time: 00:53:12, Epoch: 3, Step: 13400, Step Loss: 4.81 INFO:Pipeline: Time: 00:53:36, Epoch: 3, Step: 13500, Step Loss: 4.67 INFO:Pipeline: Time: 00:53:59, Epoch: 3, Step: 13600, Step Loss: 4.73 INFO:Pipeline: Time: 00:54:22, Epoch: 3, Step: 13700, Step Loss: 4.67 INFO:Pipeline: Time: 00:54:46, Epoch: 3, Step: 13800, Step Loss: 4.68 INFO:Pipeline: Time: 00:55:09, Epoch: 3, Step: 13900, Step Loss: 4.74 INFO:Pipeline: Time: 00:55:33, Epoch: 3, Step: 14000, Step Loss: 4.62 INFO:Pipeline: Eval Loss: 4.48, Perplexity: 88.58 INFO:Pipeline: Time: 00:55:58, Epoch: 3, Step: 14100, Step Loss: 4.71 INFO:Pipeline: Time: 00:56:22, Epoch: 3, Step: 14200, Step Loss: 4.65 INFO:Pipeline: Time: 00:56:45, Epoch: 3, Step: 14300, Step Loss: 4.70 INFO:Pipeline: Time: 00:57:09, Epoch: 3, Step: 14400, Step Loss: 4.71 INFO:Pipeline: Time: 00:57:32, Epoch: 3, Step: 14500, Step Loss: 4.65 INFO:Pipeline: Time: 00:57:55, Epoch: 3, Step: 14600, Step Loss: 4.57 INFO:Pipeline: Time: 00:58:19, Epoch: 3, Step: 14700, Step Loss: 4.62 INFO:Pipeline: Time: 00:58:42, Epoch: 3, Step: 14800, Step Loss: 4.67 INFO:Pipeline: Time: 00:59:06, Epoch: 3, Step: 14900, Step Loss: 4.72 INFO:Pipeline: Time: 00:59:29, Epoch: 3, Step: 15000, Step Loss: 4.67 INFO:Pipeline: Eval Loss: 4.44, Perplexity: 84.41 INFO:Pipeline: Time: 00:59:55, Epoch: 3, Step: 15100, Step Loss: 4.58 INFO:Pipeline: Time: 01:00:18, Epoch: 3, Step: 15200, Step Loss: 4.57 INFO:Pipeline: Time: 01:00:41, Epoch: 3, Step: 15300, Step Loss: 4.64 INFO:Pipeline: Time: 01:01:05, Epoch: 3, Step: 15400, Step Loss: 4.72 INFO:Pipeline: Time: 01:01:28, Epoch: 3, Step: 15500, Step Loss: 4.49 INFO:Pipeline: Time: 01:01:52, Epoch: 3, Step: 15600, Step Loss: 4.66 INFO:Pipeline: Time: 01:02:15, Epoch: 3, Step: 15700, Step Loss: 4.66 INFO:Pipeline: Time: 01:02:38, Epoch: 3, Step: 15800, Step Loss: 4.58 INFO:Pipeline: Time: 01:03:02, Epoch: 3, Step: 15900, Step Loss: 4.61 INFO:Pipeline: Time: 01:03:25, Epoch: 3, Step: 16000, Step Loss: 4.59 INFO:Pipeline: Eval Loss: 4.39, Perplexity: 80.75 INFO:Pipeline: Time: 01:03:51, Epoch: 3, Step: 16100, Step Loss: 4.54 INFO:Pipeline: Time: 01:04:14, Epoch: 3, Step: 16200, Step Loss: 4.54 INFO:Pipeline: Time: 01:04:38, Epoch: 3, Step: 16300, Step Loss: 4.55 INFO:Pipeline: Time: 01:05:01, Epoch: 3, Step: 16400, Step Loss: 4.61 INFO:Pipeline: Time: 01:05:25, Epoch: 3, Step: 16500, Step Loss: 4.58 INFO:Pipeline: Time: 01:05:48, Epoch: 3, Step: 16600, Step Loss: 4.51 INFO:Pipeline: Time: 01:06:11, Epoch: 3, Step: 16700, Step Loss: 4.59 INFO:Pipeline: Time: 01:06:35, Epoch: 3, Step: 16800, Step Loss: 4.60 INFO:Pipeline: Time: 01:06:58, Epoch: 3, Step: 16900, Step Loss: 4.50 INFO:Pipeline: Time: 01:07:22, Epoch: 3, Step: 17000, Step Loss: 4.63 INFO:Pipeline: Eval Loss: 4.35, Perplexity: 77.54 INFO:Pipeline: Time: 01:07:47, Epoch: 3, Step: 17100, Step Loss: 4.53 INFO:Pipeline: Time: 01:08:11, Epoch: 3, Step: 17200, Step Loss: 4.54 INFO:Pipeline: Time: 01:08:34, Epoch: 3, Step: 17300, Step Loss: 4.49 INFO:Pipeline: Time: 01:08:57, Epoch: 3, Step: 17400, Step Loss: 4.47 INFO:Pipeline: Time: 01:09:21, Epoch: 3, Step: 17500, Step Loss: 4.52 INFO:Pipeline: Time: 01:09:44, Epoch: 3, Step: 17600, Step Loss: 4.58 INFO:Pipeline: Time: 01:10:08, Epoch: 3, Step: 17700, Step Loss: 4.52 INFO:Pipeline: Time: 01:10:31, Epoch: 3, Step: 17800, Step Loss: 4.52 INFO:Pipeline: Time: 01:10:54, Epoch: 3, Step: 17900, Step Loss: 4.54 INFO:Pipeline: Time: 01:11:18, Epoch: 3, Step: 18000, Step Loss: 4.51 INFO:Pipeline: Eval Loss: 4.31, Perplexity: 74.64 INFO:Pipeline: Time: 01:11:44, Epoch: 3, Step: 18100, Step Loss: 4.57 INFO:Pipeline: Time: 01:12:08, Epoch: 3, Step: 18200, Step Loss: 4.44 INFO:Pipeline: Time: 01:12:31, Epoch: 3, Step: 18300, Step Loss: 4.56 INFO:Pipeline: Time: 01:12:55, Epoch: 4, Step: 18400, Step Loss: 4.40 INFO:Pipeline: Time: 01:13:18, Epoch: 4, Step: 18500, Step Loss: 4.55 INFO:Pipeline: Time: 01:13:41, Epoch: 4, Step: 18600, Step Loss: 4.44 INFO:Pipeline: Time: 01:14:05, Epoch: 4, Step: 18700, Step Loss: 4.44 INFO:Pipeline: Time: 01:14:28, Epoch: 4, Step: 18800, Step Loss: 4.41 INFO:Pipeline: Time: 01:14:52, Epoch: 4, Step: 18900, Step Loss: 4.52 INFO:Pipeline: Time: 01:15:15, Epoch: 4, Step: 19000, Step Loss: 4.47 INFO:Pipeline: Eval Loss: 4.28, Perplexity: 72.25 INFO:Pipeline: Time: 01:15:41, Epoch: 4, Step: 19100, Step Loss: 4.50 INFO:Pipeline: Time: 01:16:04, Epoch: 4, Step: 19200, Step Loss: 4.54 INFO:Pipeline: Time: 01:16:27, Epoch: 4, Step: 19300, Step Loss: 4.52 INFO:Pipeline: Time: 01:16:51, Epoch: 4, Step: 19400, Step Loss: 4.46 INFO:Pipeline: Time: 01:17:14, Epoch: 4, Step: 19500, Step Loss: 4.50 INFO:Pipeline: Time: 01:17:38, Epoch: 4, Step: 19600, Step Loss: 4.53 INFO:Pipeline: Time: 01:18:01, Epoch: 4, Step: 19700, Step Loss: 4.43 INFO:Pipeline: Time: 01:18:24, Epoch: 4, Step: 19800, Step Loss: 4.53 INFO:Pipeline: Time: 01:18:48, Epoch: 4, Step: 19900, Step Loss: 4.42 INFO:Pipeline: Time: 01:19:11, Epoch: 4, Step: 20000, Step Loss: 4.41 INFO:Pipeline: Eval Loss: 4.25, Perplexity: 69.89 INFO:Pipeline: Time: 01:19:37, Epoch: 4, Step: 20100, Step Loss: 4.46 INFO:Pipeline: Time: 01:20:00, Epoch: 4, Step: 20200, Step Loss: 4.39 INFO:Pipeline: Time: 01:20:24, Epoch: 4, Step: 20300, Step Loss: 4.41 INFO:Pipeline: Time: 01:20:47, Epoch: 4, Step: 20400, Step Loss: 4.39 INFO:Pipeline: Time: 01:21:11, Epoch: 4, Step: 20500, Step Loss: 4.38 INFO:Pipeline: Time: 01:21:34, Epoch: 4, Step: 20600, Step Loss: 4.42 INFO:Pipeline: Time: 01:21:58, Epoch: 4, Step: 20700, Step Loss: 4.36 INFO:Pipeline: Time: 01:22:21, Epoch: 4, Step: 20800, Step Loss: 4.42 INFO:Pipeline: Time: 01:22:44, Epoch: 4, Step: 20900, Step Loss: 4.38 INFO:Pipeline: Time: 01:23:08, Epoch: 4, Step: 21000, Step Loss: 4.38 INFO:Pipeline: Eval Loss: 4.22, Perplexity: 67.77 INFO:Pipeline: Time: 01:23:34, Epoch: 4, Step: 21100, Step Loss: 4.38 INFO:Pipeline: Time: 01:23:57, Epoch: 4, Step: 21200, Step Loss: 4.35 INFO:Pipeline: Time: 01:24:20, Epoch: 4, Step: 21300, Step Loss: 4.30 INFO:Pipeline: Time: 01:24:44, Epoch: 4, Step: 21400, Step Loss: 4.40 INFO:Pipeline: Time: 01:25:07, Epoch: 4, Step: 21500, Step Loss: 4.36 INFO:Pipeline: Time: 01:25:31, Epoch: 4, Step: 21600, Step Loss: 4.35 INFO:Pipeline: Time: 01:25:54, Epoch: 4, Step: 21700, Step Loss: 4.33 INFO:Pipeline: Time: 01:26:17, Epoch: 4, Step: 21800, Step Loss: 4.39 INFO:Pipeline: Time: 01:26:41, Epoch: 4, Step: 21900, Step Loss: 4.44 INFO:Pipeline: Time: 01:27:04, Epoch: 4, Step: 22000, Step Loss: 4.42 INFO:Pipeline: Eval Loss: 4.19, Perplexity: 65.86 INFO:Pipeline: Time: 01:27:30, Epoch: 4, Step: 22100, Step Loss: 4.37 INFO:Pipeline: Time: 01:27:54, Epoch: 4, Step: 22200, Step Loss: 4.37 INFO:Pipeline: Time: 01:28:17, Epoch: 4, Step: 22300, Step Loss: 4.31 INFO:Pipeline: Time: 01:28:40, Epoch: 4, Step: 22400, Step Loss: 4.40 INFO:Pipeline: Time: 01:29:04, Epoch: 4, Step: 22500, Step Loss: 4.35 INFO:Pipeline: Time: 01:29:27, Epoch: 4, Step: 22600, Step Loss: 4.40 INFO:Pipeline: Time: 01:29:51, Epoch: 4, Step: 22700, Step Loss: 4.37 INFO:Pipeline: Time: 01:30:14, Epoch: 4, Step: 22800, Step Loss: 4.34 INFO:Pipeline: Time: 01:30:38, Epoch: 4, Step: 22900, Step Loss: 4.40 INFO:Pipeline: Time: 01:31:01, Epoch: 4, Step: 23000, Step Loss: 4.40 INFO:Pipeline: Eval Loss: 4.16, Perplexity: 64.03 INFO:Pipeline: Time: 01:31:27, Epoch: 4, Step: 23100, Step Loss: 4.27 INFO:Pipeline: Time: 01:31:50, Epoch: 4, Step: 23200, Step Loss: 4.32 INFO:Pipeline: Time: 01:32:14, Epoch: 4, Step: 23300, Step Loss: 4.35 INFO:Pipeline: Time: 01:32:37, Epoch: 4, Step: 23400, Step Loss: 4.36 INFO:Pipeline: Time: 01:33:00, Epoch: 4, Step: 23500, Step Loss: 4.28 INFO:Pipeline: Time: 01:33:24, Epoch: 4, Step: 23600, Step Loss: 4.28 INFO:Pipeline: Time: 01:33:47, Epoch: 4, Step: 23700, Step Loss: 4.31 INFO:Pipeline: Time: 01:34:11, Epoch: 4, Step: 23800, Step Loss: 4.25 INFO:Pipeline: Time: 01:34:34, Epoch: 4, Step: 23900, Step Loss: 4.30 INFO:Pipeline: Time: 01:34:58, Epoch: 4, Step: 24000, Step Loss: 4.28 INFO:Pipeline: Eval Loss: 4.14, Perplexity: 62.50 INFO:Pipeline: Time: 01:35:23, Epoch: 4, Step: 24100, Step Loss: 4.33 INFO:Pipeline: Time: 01:35:47, Epoch: 4, Step: 24200, Step Loss: 4.32 INFO:Pipeline: Time: 01:36:10, Epoch: 4, Step: 24300, Step Loss: 4.35 INFO:Pipeline: Time: 01:36:34, Epoch: 4, Step: 24400, Step Loss: 4.28 INFO:Pipeline: Time: 01:36:57, Epoch: 5, Step: 24500, Step Loss: 4.28 INFO:Pipeline: Time: 01:37:20, Epoch: 5, Step: 24600, Step Loss: 4.29 INFO:Pipeline: Time: 01:37:44, Epoch: 5, Step: 24700, Step Loss: 4.28 INFO:Pipeline: Time: 01:38:07, Epoch: 5, Step: 24800, Step Loss: 4.33 INFO:Pipeline: Time: 01:38:31, Epoch: 5, Step: 24900, Step Loss: 4.30 INFO:Pipeline: Time: 01:38:54, Epoch: 5, Step: 25000, Step Loss: 4.35 INFO:Pipeline: Eval Loss: 4.11, Perplexity: 61.03 INFO:Pipeline: Time: 01:39:20, Epoch: 5, Step: 25100, Step Loss: 4.27 INFO:Pipeline: Time: 01:39:43, Epoch: 5, Step: 25200, Step Loss: 4.29 INFO:Pipeline: Time: 01:40:07, Epoch: 5, Step: 25300, Step Loss: 4.29 INFO:Pipeline: Time: 01:40:30, Epoch: 5, Step: 25400, Step Loss: 4.18 INFO:Pipeline: Time: 01:40:53, Epoch: 5, Step: 25500, Step Loss: 4.33 INFO:Pipeline: Time: 01:41:17, Epoch: 5, Step: 25600, Step Loss: 4.25 INFO:Pipeline: Time: 01:41:40, Epoch: 5, Step: 25700, Step Loss: 4.29 INFO:Pipeline: Time: 01:42:03, Epoch: 5, Step: 25800, Step Loss: 4.28 INFO:Pipeline: Time: 01:42:27, Epoch: 5, Step: 25900, Step Loss: 4.30 INFO:Pipeline: Time: 01:42:50, Epoch: 5, Step: 26000, Step Loss: 4.27 INFO:Pipeline: Eval Loss: 4.09, Perplexity: 59.62 INFO:Pipeline: Time: 01:43:16, Epoch: 5, Step: 26100, Step Loss: 4.26 INFO:Pipeline: Time: 01:43:39, Epoch: 5, Step: 26200, Step Loss: 4.25 INFO:Pipeline: Time: 01:44:03, Epoch: 5, Step: 26300, Step Loss: 4.29 INFO:Pipeline: Time: 01:44:26, Epoch: 5, Step: 26400, Step Loss: 4.27 INFO:Pipeline: Time: 01:44:49, Epoch: 5, Step: 26500, Step Loss: 4.20 INFO:Pipeline: Time: 01:45:13, Epoch: 5, Step: 26600, Step Loss: 4.26 INFO:Pipeline: Time: 01:45:36, Epoch: 5, Step: 26700, Step Loss: 4.32 INFO:Pipeline: Time: 01:46:00, Epoch: 5, Step: 26800, Step Loss: 4.26 INFO:Pipeline: Time: 01:46:23, Epoch: 5, Step: 26900, Step Loss: 4.25 INFO:Pipeline: Time: 01:46:47, Epoch: 5, Step: 27000, Step Loss: 4.28 INFO:Pipeline: Eval Loss: 4.07, Perplexity: 58.45 INFO:Pipeline: Time: 01:47:12, Epoch: 5, Step: 27100, Step Loss: 4.17 INFO:Pipeline: Time: 01:47:36, Epoch: 5, Step: 27200, Step Loss: 4.29 INFO:Pipeline: Time: 01:47:59, Epoch: 5, Step: 27300, Step Loss: 4.24 INFO:Pipeline: Time: 01:48:23, Epoch: 5, Step: 27400, Step Loss: 4.29 INFO:Pipeline: Time: 01:48:46, Epoch: 5, Step: 27500, Step Loss: 4.29 INFO:Pipeline: Time: 01:49:10, Epoch: 5, Step: 27600, Step Loss: 4.15 INFO:Pipeline: Time: 01:49:33, Epoch: 5, Step: 27700, Step Loss: 4.21 INFO:Pipeline: Time: 01:49:57, Epoch: 5, Step: 27800, Step Loss: 4.24 INFO:Pipeline: Time: 01:50:20, Epoch: 5, Step: 27900, Step Loss: 4.19 INFO:Pipeline: Time: 01:50:44, Epoch: 5, Step: 28000, Step Loss: 4.26 INFO:Pipeline: Eval Loss: 4.04, Perplexity: 57.11 INFO:Pipeline: Time: 01:51:09, Epoch: 5, Step: 28100, Step Loss: 4.29 INFO:Pipeline: Time: 01:51:33, Epoch: 5, Step: 28200, Step Loss: 4.15 INFO:Pipeline: Time: 01:51:56, Epoch: 5, Step: 28300, Step Loss: 4.24 INFO:Pipeline: Time: 01:52:20, Epoch: 5, Step: 28400, Step Loss: 4.22 INFO:Pipeline: Time: 01:52:43, Epoch: 5, Step: 28500, Step Loss: 4.24 INFO:Pipeline: Time: 01:53:07, Epoch: 5, Step: 28600, Step Loss: 4.20 INFO:Pipeline: Time: 01:53:30, Epoch: 5, Step: 28700, Step Loss: 4.23 INFO:Pipeline: Time: 01:53:54, Epoch: 5, Step: 28800, Step Loss: 4.13 INFO:Pipeline: Time: 01:54:17, Epoch: 5, Step: 28900, Step Loss: 4.23 INFO:Pipeline: Time: 01:54:41, Epoch: 5, Step: 29000, Step Loss: 4.11 INFO:Pipeline: Eval Loss: 4.03, Perplexity: 56.04 INFO:Pipeline: Time: 01:55:06, Epoch: 5, Step: 29100, Step Loss: 4.21 INFO:Pipeline: Time: 01:55:30, Epoch: 5, Step: 29200, Step Loss: 4.23 INFO:Pipeline: Time: 01:55:53, Epoch: 5, Step: 29300, Step Loss: 4.14 INFO:Pipeline: Time: 01:56:17, Epoch: 5, Step: 29400, Step Loss: 4.13 INFO:Pipeline: Time: 01:56:40, Epoch: 5, Step: 29500, Step Loss: 4.21 INFO:Pipeline: Time: 01:57:04, Epoch: 5, Step: 29600, Step Loss: 4.16 INFO:Pipeline: Time: 01:57:27, Epoch: 5, Step: 29700, Step Loss: 4.11 INFO:Pipeline: Time: 01:57:51, Epoch: 5, Step: 29800, Step Loss: 4.16 INFO:Pipeline: Time: 01:58:14, Epoch: 5, Step: 29900, Step Loss: 4.20 INFO:Pipeline: Time: 01:58:38, Epoch: 5, Step: 30000, Step Loss: 4.19 INFO:Pipeline: Eval Loss: 4.01, Perplexity: 55.03 INFO:Pipeline: Time: 01:59:03, Epoch: 5, Step: 30100, Step Loss: 4.10 INFO:Pipeline: Time: 01:59:27, Epoch: 5, Step: 30200, Step Loss: 4.18 INFO:Pipeline: Time: 01:59:50, Epoch: 5, Step: 30300, Step Loss: 4.23 INFO:Pipeline: Time: 02:00:14, Epoch: 5, Step: 30400, Step Loss: 4.19 INFO:Pipeline: Time: 02:00:37, Epoch: 5, Step: 30500, Step Loss: 4.22 INFO:Pipeline: Time: 02:01:00, Epoch: 6, Step: 30600, Step Loss: 4.14 INFO:Pipeline: Time: 02:01:24, Epoch: 6, Step: 30700, Step Loss: 4.28 INFO:Pipeline: Time: 02:01:47, Epoch: 6, Step: 30800, Step Loss: 4.07 INFO:Pipeline: Time: 02:02:11, Epoch: 6, Step: 30900, Step Loss: 4.17 INFO:Pipeline: Time: 02:02:34, Epoch: 6, Step: 31000, Step Loss: 4.23 INFO:Pipeline: Eval Loss: 3.99, Perplexity: 54.16 INFO:Pipeline: Time: 02:03:00, Epoch: 6, Step: 31100, Step Loss: 4.16 INFO:Pipeline: Time: 02:03:24, Epoch: 6, Step: 31200, Step Loss: 4.14 INFO:Pipeline: Time: 02:03:47, Epoch: 6, Step: 31300, Step Loss: 4.18 INFO:Pipeline: Time: 02:04:11, Epoch: 6, Step: 31400, Step Loss: 4.13 INFO:Pipeline: Time: 02:04:34, Epoch: 6, Step: 31500, Step Loss: 4.14 INFO:Pipeline: Time: 02:04:58, Epoch: 6, Step: 31600, Step Loss: 4.17 INFO:Pipeline: Time: 02:05:21, Epoch: 6, Step: 31700, Step Loss: 4.07 INFO:Pipeline: Time: 02:05:45, Epoch: 6, Step: 31800, Step Loss: 4.05 INFO:Pipeline: Time: 02:06:08, Epoch: 6, Step: 31900, Step Loss: 4.13 INFO:Pipeline: Time: 02:06:32, Epoch: 6, Step: 32000, Step Loss: 4.14 INFO:Pipeline: Eval Loss: 3.97, Perplexity: 53.22 INFO:Pipeline: Time: 02:06:58, Epoch: 6, Step: 32100, Step Loss: 4.10 INFO:Pipeline: Time: 02:07:21, Epoch: 6, Step: 32200, Step Loss: 4.15 INFO:Pipeline: Time: 02:07:45, Epoch: 6, Step: 32300, Step Loss: 4.05 INFO:Pipeline: Time: 02:08:08, Epoch: 6, Step: 32400, Step Loss: 4.19 INFO:Pipeline: Time: 02:08:32, Epoch: 6, Step: 32500, Step Loss: 4.18 INFO:Pipeline: Time: 02:08:55, Epoch: 6, Step: 32600, Step Loss: 4.24 INFO:Pipeline: Time: 02:09:19, Epoch: 6, Step: 32700, Step Loss: 4.08 INFO:Pipeline: Time: 02:09:42, Epoch: 6, Step: 32800, Step Loss: 4.04 INFO:Pipeline: Time: 02:10:06, Epoch: 6, Step: 32900, Step Loss: 4.02 INFO:Pipeline: Time: 02:10:29, Epoch: 6, Step: 33000, Step Loss: 4.12 INFO:Pipeline: Eval Loss: 3.96, Perplexity: 52.39 INFO:Pipeline: Time: 02:10:55, Epoch: 6, Step: 33100, Step Loss: 4.18 INFO:Pipeline: Time: 02:11:18, Epoch: 6, Step: 33200, Step Loss: 4.09 INFO:Pipeline: Time: 02:11:42, Epoch: 6, Step: 33300, Step Loss: 4.15 INFO:Pipeline: Time: 02:12:05, Epoch: 6, Step: 33400, Step Loss: 4.16 INFO:Pipeline: Time: 02:12:29, Epoch: 6, Step: 33500, Step Loss: 4.20 INFO:Pipeline: Time: 02:12:52, Epoch: 6, Step: 33600, Step Loss: 4.10 INFO:Pipeline: Time: 02:13:16, Epoch: 6, Step: 33700, Step Loss: 4.18 INFO:Pipeline: Time: 02:13:39, Epoch: 6, Step: 33800, Step Loss: 4.12 INFO:Pipeline: Time: 02:14:03, Epoch: 6, Step: 33900, Step Loss: 4.10 INFO:Pipeline: Time: 02:14:26, Epoch: 6, Step: 34000, Step Loss: 4.16 INFO:Pipeline: Eval Loss: 3.94, Perplexity: 51.66 INFO:Pipeline: Time: 02:14:52, Epoch: 6, Step: 34100, Step Loss: 4.08 INFO:Pipeline: Time: 02:15:15, Epoch: 6, Step: 34200, Step Loss: 4.06 INFO:Pipeline: Time: 02:15:39, Epoch: 6, Step: 34300, Step Loss: 4.06 INFO:Pipeline: Time: 02:16:02, Epoch: 6, Step: 34400, Step Loss: 4.11 INFO:Pipeline: Time: 02:16:26, Epoch: 6, Step: 34500, Step Loss: 4.11 INFO:Pipeline: Time: 02:16:49, Epoch: 6, Step: 34600, Step Loss: 4.08 INFO:Pipeline: Time: 02:17:13, Epoch: 6, Step: 34700, Step Loss: 4.01 INFO:Pipeline: Time: 02:17:36, Epoch: 6, Step: 34800, Step Loss: 4.11 INFO:Pipeline: Time: 02:18:00, Epoch: 6, Step: 34900, Step Loss: 4.09 INFO:Pipeline: Time: 02:18:23, Epoch: 6, Step: 35000, Step Loss: 4.10 INFO:Pipeline: Eval Loss: 3.93, Perplexity: 50.89 INFO:Pipeline: Time: 02:18:49, Epoch: 6, Step: 35100, Step Loss: 4.10 INFO:Pipeline: Time: 02:19:12, Epoch: 6, Step: 35200, Step Loss: 4.02 INFO:Pipeline: Time: 02:19:36, Epoch: 6, Step: 35300, Step Loss: 4.15 INFO:Pipeline: Time: 02:19:59, Epoch: 6, Step: 35400, Step Loss: 4.05 INFO:Pipeline: Time: 02:20:23, Epoch: 6, Step: 35500, Step Loss: 3.98 INFO:Pipeline: Time: 02:20:46, Epoch: 6, Step: 35600, Step Loss: 4.05 INFO:Pipeline: Time: 02:21:10, Epoch: 6, Step: 35700, Step Loss: 4.09 INFO:Pipeline: Time: 02:21:33, Epoch: 6, Step: 35800, Step Loss: 4.04 INFO:Pipeline: Time: 02:21:57, Epoch: 6, Step: 35900, Step Loss: 4.03 INFO:Pipeline: Time: 02:22:20, Epoch: 6, Step: 36000, Step Loss: 4.02 INFO:Pipeline: Eval Loss: 3.91, Perplexity: 50.09 INFO:Pipeline: Time: 02:22:46, Epoch: 6, Step: 36100, Step Loss: 3.96 INFO:Pipeline: Time: 02:23:10, Epoch: 6, Step: 36200, Step Loss: 4.02 INFO:Pipeline: Time: 02:23:33, Epoch: 6, Step: 36300, Step Loss: 4.01 INFO:Pipeline: Time: 02:23:56, Epoch: 6, Step: 36400, Step Loss: 4.06 INFO:Pipeline: Time: 02:24:20, Epoch: 6, Step: 36500, Step Loss: 3.98 INFO:Pipeline: Time: 02:24:43, Epoch: 6, Step: 36600, Step Loss: 4.17 INFO:Pipeline: Time: 02:25:07, Epoch: 7, Step: 36700, Step Loss: 4.06 INFO:Pipeline: Time: 02:25:30, Epoch: 7, Step: 36800, Step Loss: 4.03 INFO:Pipeline: Time: 02:25:54, Epoch: 7, Step: 36900, Step Loss: 4.14 INFO:Pipeline: Time: 02:26:17, Epoch: 7, Step: 37000, Step Loss: 4.07 INFO:Pipeline: Eval Loss: 3.90, Perplexity: 49.50 INFO:Pipeline: Time: 02:26:43, Epoch: 7, Step: 37100, Step Loss: 4.08 INFO:Pipeline: Time: 02:27:07, Epoch: 7, Step: 37200, Step Loss: 4.00 INFO:Pipeline: Time: 02:27:30, Epoch: 7, Step: 37300, Step Loss: 4.09 INFO:Pipeline: Time: 02:27:53, Epoch: 7, Step: 37400, Step Loss: 4.01 INFO:Pipeline: Time: 02:28:17, Epoch: 7, Step: 37500, Step Loss: 4.13 INFO:Pipeline: Time: 02:28:40, Epoch: 7, Step: 37600, Step Loss: 4.09 INFO:Pipeline: Time: 02:29:04, Epoch: 7, Step: 37700, Step Loss: 4.00 INFO:Pipeline: Time: 02:29:27, Epoch: 7, Step: 37800, Step Loss: 4.07 INFO:Pipeline: Time: 02:29:51, Epoch: 7, Step: 37900, Step Loss: 3.99 INFO:Pipeline: Time: 02:30:14, Epoch: 7, Step: 38000, Step Loss: 4.09 INFO:Pipeline: Eval Loss: 3.89, Perplexity: 48.81 INFO:Pipeline: Time: 02:30:40, Epoch: 7, Step: 38100, Step Loss: 4.12 INFO:Pipeline: Time: 02:31:03, Epoch: 7, Step: 38200, Step Loss: 3.94 INFO:Pipeline: Time: 02:31:27, Epoch: 7, Step: 38300, Step Loss: 4.04 INFO:Pipeline: Time: 02:31:50, Epoch: 7, Step: 38400, Step Loss: 4.03 INFO:Pipeline: Time: 02:32:14, Epoch: 7, Step: 38500, Step Loss: 4.03 INFO:Pipeline: Time: 02:32:37, Epoch: 7, Step: 38600, Step Loss: 4.04 INFO:Pipeline: Time: 02:33:01, Epoch: 7, Step: 38700, Step Loss: 4.05 INFO:Pipeline: Time: 02:33:24, Epoch: 7, Step: 38800, Step Loss: 4.07 INFO:Pipeline: Time: 02:33:48, Epoch: 7, Step: 38900, Step Loss: 3.99 INFO:Pipeline: Time: 02:34:11, Epoch: 7, Step: 39000, Step Loss: 4.08 INFO:Pipeline: Eval Loss: 3.88, Perplexity: 48.29 INFO:Pipeline: Time: 02:34:37, Epoch: 7, Step: 39100, Step Loss: 3.99 INFO:Pipeline: Time: 02:35:01, Epoch: 7, Step: 39200, Step Loss: 4.07 INFO:Pipeline: Time: 02:35:24, Epoch: 7, Step: 39300, Step Loss: 3.98 INFO:Pipeline: Time: 02:35:48, Epoch: 7, Step: 39400, Step Loss: 4.05 INFO:Pipeline: Time: 02:36:11, Epoch: 7, Step: 39500, Step Loss: 3.99 INFO:Pipeline: Time: 02:36:35, Epoch: 7, Step: 39600, Step Loss: 4.04 INFO:Pipeline: Time: 02:36:58, Epoch: 7, Step: 39700, Step Loss: 4.06 INFO:Pipeline: Time: 02:37:21, Epoch: 7, Step: 39800, Step Loss: 4.05 INFO:Pipeline: Time: 02:37:45, Epoch: 7, Step: 39900, Step Loss: 4.06 INFO:Pipeline: Time: 02:38:08, Epoch: 7, Step: 40000, Step Loss: 3.91 INFO:Pipeline: Eval Loss: 3.87, Perplexity: 47.73 INFO:Pipeline: Time: 02:38:34, Epoch: 7, Step: 40100, Step Loss: 3.95 INFO:Pipeline: Time: 02:38:58, Epoch: 7, Step: 40200, Step Loss: 3.99 INFO:Pipeline: Time: 02:39:21, Epoch: 7, Step: 40300, Step Loss: 4.04 INFO:Pipeline: Time: 02:39:44, Epoch: 7, Step: 40400, Step Loss: 3.99 INFO:Pipeline: Time: 02:40:08, Epoch: 7, Step: 40500, Step Loss: 3.99 INFO:Pipeline: Time: 02:40:31, Epoch: 7, Step: 40600, Step Loss: 4.01 INFO:Pipeline: Time: 02:40:55, Epoch: 7, Step: 40700, Step Loss: 4.01 INFO:Pipeline: Time: 02:41:18, Epoch: 7, Step: 40800, Step Loss: 3.98 INFO:Pipeline: Time: 02:41:42, Epoch: 7, Step: 40900, Step Loss: 3.97 INFO:Pipeline: Time: 02:42:05, Epoch: 7, Step: 41000, Step Loss: 3.96 INFO:Pipeline: Eval Loss: 3.85, Perplexity: 47.13 INFO:Pipeline: Time: 02:42:32, Epoch: 7, Step: 41100, Step Loss: 3.94 INFO:Pipeline: Time: 02:42:55, Epoch: 7, Step: 41200, Step Loss: 4.07 INFO:Pipeline: Time: 02:43:19, Epoch: 7, Step: 41300, Step Loss: 4.05 INFO:Pipeline: Time: 02:43:42, Epoch: 7, Step: 41400, Step Loss: 3.96 INFO:Pipeline: Time: 02:44:05, Epoch: 7, Step: 41500, Step Loss: 3.93 INFO:Pipeline: Time: 02:44:29, Epoch: 7, Step: 41600, Step Loss: 3.91 INFO:Pipeline: Time: 02:44:52, Epoch: 7, Step: 41700, Step Loss: 4.03 INFO:Pipeline: Time: 02:45:16, Epoch: 7, Step: 41800, Step Loss: 3.92 INFO:Pipeline: Time: 02:45:39, Epoch: 7, Step: 41900, Step Loss: 3.97 INFO:Pipeline: Time: 02:46:03, Epoch: 7, Step: 42000, Step Loss: 4.06 INFO:Pipeline: Eval Loss: 3.84, Perplexity: 46.69 INFO:Pipeline: Time: 02:46:28, Epoch: 7, Step: 42100, Step Loss: 4.00 INFO:Pipeline: Time: 02:46:52, Epoch: 7, Step: 42200, Step Loss: 3.98 INFO:Pipeline: Time: 02:47:15, Epoch: 7, Step: 42300, Step Loss: 4.02 INFO:Pipeline: Time: 02:47:39, Epoch: 7, Step: 42400, Step Loss: 3.98 INFO:Pipeline: Time: 02:48:02, Epoch: 7, Step: 42500, Step Loss: 4.04 INFO:Pipeline: Time: 02:48:26, Epoch: 7, Step: 42600, Step Loss: 3.91 INFO:Pipeline: Time: 02:48:49, Epoch: 7, Step: 42700, Step Loss: 3.97 INFO:Pipeline: Time: 02:49:12, Epoch: 8, Step: 42800, Step Loss: 3.97 INFO:Pipeline: Time: 02:49:36, Epoch: 8, Step: 42900, Step Loss: 4.01 INFO:Pipeline: Time: 02:49:59, Epoch: 8, Step: 43000, Step Loss: 3.98 INFO:Pipeline: Eval Loss: 3.83, Perplexity: 46.18 INFO:Pipeline: Time: 02:50:25, Epoch: 8, Step: 43100, Step Loss: 3.94 INFO:Pipeline: Time: 02:50:48, Epoch: 8, Step: 43200, Step Loss: 3.93 INFO:Pipeline: Time: 02:51:12, Epoch: 8, Step: 43300, Step Loss: 3.97 INFO:Pipeline: Time: 02:51:35, Epoch: 8, Step: 43400, Step Loss: 4.00 INFO:Pipeline: Time: 02:51:59, Epoch: 8, Step: 43500, Step Loss: 3.91 INFO:Pipeline: Time: 02:52:22, Epoch: 8, Step: 43600, Step Loss: 3.94 INFO:Pipeline: Time: 02:52:45, Epoch: 8, Step: 43700, Step Loss: 3.99 INFO:Pipeline: Time: 02:53:09, Epoch: 8, Step: 43800, Step Loss: 3.99 INFO:Pipeline: Time: 02:53:32, Epoch: 8, Step: 43900, Step Loss: 4.02 INFO:Pipeline: Time: 02:53:56, Epoch: 8, Step: 44000, Step Loss: 4.03 INFO:Pipeline: Eval Loss: 3.82, Perplexity: 45.75 INFO:Pipeline: Time: 02:54:22, Epoch: 8, Step: 44100, Step Loss: 3.97 INFO:Pipeline: Time: 02:54:45, Epoch: 8, Step: 44200, Step Loss: 4.02 INFO:Pipeline: Time: 02:55:08, Epoch: 8, Step: 44300, Step Loss: 3.93 INFO:Pipeline: Time: 02:55:32, Epoch: 8, Step: 44400, Step Loss: 3.98 INFO:Pipeline: Time: 02:55:55, Epoch: 8, Step: 44500, Step Loss: 3.96 INFO:Pipeline: Time: 02:56:19, Epoch: 8, Step: 44600, Step Loss: 3.92 INFO:Pipeline: Time: 02:56:42, Epoch: 8, Step: 44700, Step Loss: 4.01 INFO:Pipeline: Time: 02:57:06, Epoch: 8, Step: 44800, Step Loss: 3.93 INFO:Pipeline: Time: 02:57:29, Epoch: 8, Step: 44900, Step Loss: 3.98 INFO:Pipeline: Time: 02:57:53, Epoch: 8, Step: 45000, Step Loss: 4.00 INFO:Pipeline: Eval Loss: 3.81, Perplexity: 45.28 INFO:Pipeline: Time: 02:58:19, Epoch: 8, Step: 45100, Step Loss: 3.87 INFO:Pipeline: Time: 02:58:42, Epoch: 8, Step: 45200, Step Loss: 3.90 INFO:Pipeline: Time: 02:59:06, Epoch: 8, Step: 45300, Step Loss: 3.96 INFO:Pipeline: Time: 02:59:29, Epoch: 8, Step: 45400, Step Loss: 3.98 INFO:Pipeline: Time: 02:59:52, Epoch: 8, Step: 45500, Step Loss: 3.95 INFO:Pipeline: Time: 03:00:16, Epoch: 8, Step: 45600, Step Loss: 4.00 INFO:Pipeline: Time: 03:00:39, Epoch: 8, Step: 45700, Step Loss: 3.90 INFO:Pipeline: Time: 03:01:03, Epoch: 8, Step: 45800, Step Loss: 3.93 INFO:Pipeline: Time: 03:01:26, Epoch: 8, Step: 45900, Step Loss: 3.91 INFO:Pipeline: Time: 03:01:50, Epoch: 8, Step: 46000, Step Loss: 3.97 INFO:Pipeline: Eval Loss: 3.80, Perplexity: 44.86 INFO:Pipeline: Time: 03:02:15, Epoch: 8, Step: 46100, Step Loss: 3.91 INFO:Pipeline: Time: 03:02:39, Epoch: 8, Step: 46200, Step Loss: 3.92 INFO:Pipeline: Time: 03:03:02, Epoch: 8, Step: 46300, Step Loss: 3.97 INFO:Pipeline: Time: 03:03:26, Epoch: 8, Step: 46400, Step Loss: 3.87 INFO:Pipeline: Time: 03:03:49, Epoch: 8, Step: 46500, Step Loss: 3.96 INFO:Pipeline: Time: 03:04:13, Epoch: 8, Step: 46600, Step Loss: 3.90 INFO:Pipeline: Time: 03:04:36, Epoch: 8, Step: 46700, Step Loss: 3.86 INFO:Pipeline: Time: 03:05:00, Epoch: 8, Step: 46800, Step Loss: 4.03 INFO:Pipeline: Time: 03:05:23, Epoch: 8, Step: 46900, Step Loss: 3.98 INFO:Pipeline: Time: 03:05:47, Epoch: 8, Step: 47000, Step Loss: 3.98 INFO:Pipeline: Eval Loss: 3.79, Perplexity: 44.44 INFO:Pipeline: Time: 03:06:13, Epoch: 8, Step: 47100, Step Loss: 3.92 INFO:Pipeline: Time: 03:06:36, Epoch: 8, Step: 47200, Step Loss: 3.86 INFO:Pipeline: Time: 03:06:59, Epoch: 8, Step: 47300, Step Loss: 4.06 INFO:Pipeline: Time: 03:07:23, Epoch: 8, Step: 47400, Step Loss: 3.86 INFO:Pipeline: Time: 03:07:46, Epoch: 8, Step: 47500, Step Loss: 3.86 INFO:Pipeline: Time: 03:08:10, Epoch: 8, Step: 47600, Step Loss: 3.94 INFO:Pipeline: Time: 03:08:33, Epoch: 8, Step: 47700, Step Loss: 3.91 INFO:Pipeline: Time: 03:08:56, Epoch: 8, Step: 47800, Step Loss: 3.82 INFO:Pipeline: Time: 03:09:20, Epoch: 8, Step: 47900, Step Loss: 3.93 INFO:Pipeline: Time: 03:09:43, Epoch: 8, Step: 48000, Step Loss: 3.94 INFO:Pipeline: Eval Loss: 3.79, Perplexity: 44.10 INFO:Pipeline: Time: 03:10:09, Epoch: 8, Step: 48100, Step Loss: 3.82 INFO:Pipeline: Time: 03:10:32, Epoch: 8, Step: 48200, Step Loss: 3.98 INFO:Pipeline: Time: 03:10:56, Epoch: 8, Step: 48300, Step Loss: 3.94 INFO:Pipeline: Time: 03:11:19, Epoch: 8, Step: 48400, Step Loss: 3.86 INFO:Pipeline: Time: 03:11:42, Epoch: 8, Step: 48500, Step Loss: 3.93 INFO:Pipeline: Time: 03:12:06, Epoch: 8, Step: 48600, Step Loss: 3.94 INFO:Pipeline: Time: 03:12:29, Epoch: 8, Step: 48700, Step Loss: 3.91 INFO:Pipeline: Time: 03:12:53, Epoch: 8, Step: 48800, Step Loss: 3.90 INFO:Pipeline: Time: 03:13:16, Epoch: 9, Step: 48900, Step Loss: 3.94 INFO:Pipeline: Time: 03:13:39, Epoch: 9, Step: 49000, Step Loss: 3.89 INFO:Pipeline: Eval Loss: 3.78, Perplexity: 43.72 INFO:Pipeline: Time: 03:14:05, Epoch: 9, Step: 49100, Step Loss: 3.85 INFO:Pipeline: Time: 03:14:28, Epoch: 9, Step: 49200, Step Loss: 3.92 INFO:Pipeline: Time: 03:14:52, Epoch: 9, Step: 49300, Step Loss: 3.93 INFO:Pipeline: Time: 03:15:15, Epoch: 9, Step: 49400, Step Loss: 3.87 INFO:Pipeline: Time: 03:15:39, Epoch: 9, Step: 49500, Step Loss: 3.96 INFO:Pipeline: Time: 03:16:02, Epoch: 9, Step: 49600, Step Loss: 3.91 INFO:Pipeline: Time: 03:16:25, Epoch: 9, Step: 49700, Step Loss: 3.96 INFO:Pipeline: Time: 03:16:49, Epoch: 9, Step: 49800, Step Loss: 3.87 INFO:Pipeline: Time: 03:17:12, Epoch: 9, Step: 49900, Step Loss: 3.99 INFO:Pipeline: Time: 03:17:36, Epoch: 9, Step: 50000, Step Loss: 3.97 INFO:Pipeline: Eval Loss: 3.77, Perplexity: 43.35 INFO:Pipeline: Time: 03:18:01, Epoch: 9, Step: 50100, Step Loss: 3.88 INFO:Pipeline: Time: 03:18:25, Epoch: 9, Step: 50200, Step Loss: 3.95 INFO:Pipeline: Time: 03:18:48, Epoch: 9, Step: 50300, Step Loss: 3.88 INFO:Pipeline: Time: 03:19:11, Epoch: 9, Step: 50400, Step Loss: 3.90 INFO:Pipeline: Time: 03:19:35, Epoch: 9, Step: 50500, Step Loss: 3.80 INFO:Pipeline: Time: 03:19:58, Epoch: 9, Step: 50600, Step Loss: 3.91 INFO:Pipeline: Time: 03:20:21, Epoch: 9, Step: 50700, Step Loss: 3.93 INFO:Pipeline: Time: 03:20:45, Epoch: 9, Step: 50800, Step Loss: 3.91 INFO:Pipeline: Time: 03:21:08, Epoch: 9, Step: 50900, Step Loss: 3.96 INFO:Pipeline: Time: 03:21:32, Epoch: 9, Step: 51000, Step Loss: 3.89 INFO:Pipeline: Eval Loss: 3.76, Perplexity: 43.16 INFO:Pipeline: Time: 03:21:57, Epoch: 9, Step: 51100, Step Loss: 3.89 INFO:Pipeline: Time: 03:22:21, Epoch: 9, Step: 51200, Step Loss: 3.94 INFO:Pipeline: Time: 03:22:44, Epoch: 9, Step: 51300, Step Loss: 3.90 INFO:Pipeline: Time: 03:23:08, Epoch: 9, Step: 51400, Step Loss: 3.96 INFO:Pipeline: Time: 03:23:31, Epoch: 9, Step: 51500, Step Loss: 3.95 INFO:Pipeline: Time: 03:23:54, Epoch: 9, Step: 51600, Step Loss: 3.90 INFO:Pipeline: Time: 03:24:18, Epoch: 9, Step: 51700, Step Loss: 3.78 INFO:Pipeline: Time: 03:24:41, Epoch: 9, Step: 51800, Step Loss: 3.84 INFO:Pipeline: Time: 03:25:04, Epoch: 9, Step: 51900, Step Loss: 3.94 INFO:Pipeline: Time: 03:25:28, Epoch: 9, Step: 52000, Step Loss: 3.91 INFO:Pipeline: Eval Loss: 3.76, Perplexity: 42.75 INFO:Pipeline: Time: 03:25:54, Epoch: 9, Step: 52100, Step Loss: 3.87 INFO:Pipeline: Time: 03:26:17, Epoch: 9, Step: 52200, Step Loss: 3.95 INFO:Pipeline: Time: 03:26:40, Epoch: 9, Step: 52300, Step Loss: 3.98 INFO:Pipeline: Time: 03:27:04, Epoch: 9, Step: 52400, Step Loss: 3.90 INFO:Pipeline: Time: 03:27:27, Epoch: 9, Step: 52500, Step Loss: 3.90 INFO:Pipeline: Time: 03:27:51, Epoch: 9, Step: 52600, Step Loss: 3.81 INFO:Pipeline: Time: 03:28:14, Epoch: 9, Step: 52700, Step Loss: 3.88 INFO:Pipeline: Time: 03:28:37, Epoch: 9, Step: 52800, Step Loss: 3.88 INFO:Pipeline: Time: 03:29:01, Epoch: 9, Step: 52900, Step Loss: 3.88 INFO:Pipeline: Time: 03:29:24, Epoch: 9, Step: 53000, Step Loss: 3.84 INFO:Pipeline: Eval Loss: 3.75, Perplexity: 42.45 INFO:Pipeline: Time: 03:29:50, Epoch: 9, Step: 53100, Step Loss: 3.82 INFO:Pipeline: Time: 03:30:14, Epoch: 9, Step: 53200, Step Loss: 3.86 INFO:Pipeline: Time: 03:30:37, Epoch: 9, Step: 53300, Step Loss: 3.84 INFO:Pipeline: Time: 03:31:00, Epoch: 9, Step: 53400, Step Loss: 3.73 INFO:Pipeline: Time: 03:31:24, Epoch: 9, Step: 53500, Step Loss: 3.83 INFO:Pipeline: Time: 03:31:47, Epoch: 9, Step: 53600, Step Loss: 3.92 INFO:Pipeline: Time: 03:32:11, Epoch: 9, Step: 53700, Step Loss: 3.84 INFO:Pipeline: Time: 03:32:34, Epoch: 9, Step: 53800, Step Loss: 3.87 INFO:Pipeline: Time: 03:32:57, Epoch: 9, Step: 53900, Step Loss: 3.85 INFO:Pipeline: Time: 03:33:21, Epoch: 9, Step: 54000, Step Loss: 3.84 INFO:Pipeline: Eval Loss: 3.74, Perplexity: 42.19 INFO:Pipeline: Time: 03:33:47, Epoch: 9, Step: 54100, Step Loss: 3.88 INFO:Pipeline: Time: 03:34:10, Epoch: 9, Step: 54200, Step Loss: 3.91 INFO:Pipeline: Time: 03:34:33, Epoch: 9, Step: 54300, Step Loss: 3.83 INFO:Pipeline: Time: 03:34:57, Epoch: 9, Step: 54400, Step Loss: 3.92 INFO:Pipeline: Time: 03:35:20, Epoch: 9, Step: 54500, Step Loss: 3.85 INFO:Pipeline: Time: 03:35:44, Epoch: 9, Step: 54600, Step Loss: 3.90 INFO:Pipeline: Time: 03:36:07, Epoch: 9, Step: 54700, Step Loss: 3.80 INFO:Pipeline: Time: 03:36:30, Epoch: 9, Step: 54800, Step Loss: 3.89 INFO:Pipeline: Time: 03:36:54, Epoch: 9, Step: 54900, Step Loss: 3.85 INFO:Pipeline: Time: 03:37:17, Epoch: 10, Step: 55000, Step Loss: 3.86 INFO:Pipeline: Eval Loss: 3.73, Perplexity: 41.85 INFO:Pipeline: Time: 03:37:43, Epoch: 10, Step: 55100, Step Loss: 3.84 INFO:Pipeline: Time: 03:38:06, Epoch: 10, Step: 55200, Step Loss: 3.81 INFO:Pipeline: Time: 03:38:29, Epoch: 10, Step: 55300, Step Loss: 3.83 INFO:Pipeline: Time: 03:38:53, Epoch: 10, Step: 55400, Step Loss: 3.79 INFO:Pipeline: Time: 03:39:16, Epoch: 10, Step: 55500, Step Loss: 3.89 INFO:Pipeline: Time: 03:39:40, Epoch: 10, Step: 55600, Step Loss: 3.85 INFO:Pipeline: Time: 03:40:03, Epoch: 10, Step: 55700, Step Loss: 3.84 INFO:Pipeline: Time: 03:40:26, Epoch: 10, Step: 55800, Step Loss: 3.78 INFO:Pipeline: Time: 03:40:50, Epoch: 10, Step: 55900, Step Loss: 3.83 INFO:Pipeline: Time: 03:41:13, Epoch: 10, Step: 56000, Step Loss: 3.88 INFO:Pipeline: Eval Loss: 3.73, Perplexity: 41.66 INFO:Pipeline: Time: 03:41:39, Epoch: 10, Step: 56100, Step Loss: 3.89 INFO:Pipeline: Time: 03:42:02, Epoch: 10, Step: 56200, Step Loss: 3.85 INFO:Pipeline: Time: 03:42:26, Epoch: 10, Step: 56300, Step Loss: 3.82 INFO:Pipeline: Time: 03:42:49, Epoch: 10, Step: 56400, Step Loss: 3.83 INFO:Pipeline: Time: 03:43:12, Epoch: 10, Step: 56500, Step Loss: 3.87 INFO:Pipeline: Time: 03:43:36, Epoch: 10, Step: 56600, Step Loss: 3.85 INFO:Pipeline: Time: 03:43:59, Epoch: 10, Step: 56700, Step Loss: 3.84 INFO:Pipeline: Time: 03:44:23, Epoch: 10, Step: 56800, Step Loss: 3.91 INFO:Pipeline: Time: 03:44:46, Epoch: 10, Step: 56900, Step Loss: 3.82 INFO:Pipeline: Time: 03:45:10, Epoch: 10, Step: 57000, Step Loss: 3.86 INFO:Pipeline: Eval Loss: 3.72, Perplexity: 41.39 INFO:Pipeline: Time: 03:45:35, Epoch: 10, Step: 57100, Step Loss: 3.91 INFO:Pipeline: Time: 03:45:59, Epoch: 10, Step: 57200, Step Loss: 3.85 INFO:Pipeline: Time: 03:46:22, Epoch: 10, Step: 57300, Step Loss: 3.80 INFO:Pipeline: Time: 03:46:45, Epoch: 10, Step: 57400, Step Loss: 3.92 INFO:Pipeline: Time: 03:47:09, Epoch: 10, Step: 57500, Step Loss: 3.86 INFO:Pipeline: Time: 03:47:32, Epoch: 10, Step: 57600, Step Loss: 3.77 INFO:Pipeline: Time: 03:47:55, Epoch: 10, Step: 57700, Step Loss: 3.82 INFO:Pipeline: Time: 03:48:19, Epoch: 10, Step: 57800, Step Loss: 3.84 INFO:Pipeline: Time: 03:48:42, Epoch: 10, Step: 57900, Step Loss: 3.80 INFO:Pipeline: Time: 03:49:06, Epoch: 10, Step: 58000, Step Loss: 3.88 INFO:Pipeline: Eval Loss: 3.72, Perplexity: 41.13 INFO:Pipeline: Time: 03:49:31, Epoch: 10, Step: 58100, Step Loss: 3.93 INFO:Pipeline: Time: 03:49:55, Epoch: 10, Step: 58200, Step Loss: 3.78 INFO:Pipeline: Time: 03:50:18, Epoch: 10, Step: 58300, Step Loss: 3.86 INFO:Pipeline: Time: 03:50:41, Epoch: 10, Step: 58400, Step Loss: 3.86 INFO:Pipeline: Time: 03:51:05, Epoch: 10, Step: 58500, Step Loss: 3.84 INFO:Pipeline: Time: 03:51:28, Epoch: 10, Step: 58600, Step Loss: 3.83 INFO:Pipeline: Time: 03:51:51, Epoch: 10, Step: 58700, Step Loss: 3.85 INFO:Pipeline: Time: 03:52:15, Epoch: 10, Step: 58800, Step Loss: 3.84 INFO:Pipeline: Time: 03:52:38, Epoch: 10, Step: 58900, Step Loss: 3.86 INFO:Pipeline: Time: 03:53:02, Epoch: 10, Step: 59000, Step Loss: 3.90 INFO:Pipeline: Eval Loss: 3.71, Perplexity: 40.88 INFO:Pipeline: Time: 03:53:27, Epoch: 10, Step: 59100, Step Loss: 3.83 INFO:Pipeline: Time: 03:53:50, Epoch: 10, Step: 59200, Step Loss: 3.86 INFO:Pipeline: Time: 03:54:14, Epoch: 10, Step: 59300, Step Loss: 3.90 INFO:Pipeline: Time: 03:54:37, Epoch: 10, Step: 59400, Step Loss: 3.84 INFO:Pipeline: Time: 03:55:00, Epoch: 10, Step: 59500, Step Loss: 3.79 INFO:Pipeline: Time: 03:55:24, Epoch: 10, Step: 59600, Step Loss: 3.82 INFO:Pipeline: Time: 03:55:47, Epoch: 10, Step: 59700, Step Loss: 3.85 INFO:Pipeline: Time: 03:56:11, Epoch: 10, Step: 59800, Step Loss: 3.88 INFO:Pipeline: Time: 03:56:34, Epoch: 10, Step: 59900, Step Loss: 3.85 INFO:Pipeline: Time: 03:56:57, Epoch: 10, Step: 60000, Step Loss: 3.75 INFO:Pipeline: Eval Loss: 3.71, Perplexity: 40.69 INFO:Pipeline: Time: 03:57:23, Epoch: 10, Step: 60100, Step Loss: 3.84 INFO:Pipeline: Time: 03:57:46, Epoch: 10, Step: 60200, Step Loss: 3.83 INFO:Pipeline: Time: 03:58:10, Epoch: 10, Step: 60300, Step Loss: 3.75 INFO:Pipeline: Time: 03:58:33, Epoch: 10, Step: 60400, Step Loss: 3.85 INFO:Pipeline: Time: 03:58:57, Epoch: 10, Step: 60500, Step Loss: 3.83 INFO:Pipeline: Time: 03:59:20, Epoch: 10, Step: 60600, Step Loss: 3.83 INFO:Pipeline: Time: 03:59:43, Epoch: 10, Step: 60700, Step Loss: 3.74 INFO:Pipeline: Time: 04:00:07, Epoch: 10, Step: 60800, Step Loss: 3.85 INFO:Pipeline: Time: 04:00:30, Epoch: 10, Step: 60900, Step Loss: 3.78 INFO:Pipeline: Time: 04:00:54, Epoch: 10, Step: 61000, Step Loss: 3.79 INFO:Pipeline: Eval Loss: 3.70, Perplexity: 40.52 INFO:Pipeline: Time: 04:01:19, Epoch: 11, Step: 61100, Step Loss: 3.80 INFO:Pipeline: Time: 04:01:43, Epoch: 11, Step: 61200, Step Loss: 3.82 INFO:Pipeline: Time: 04:02:06, Epoch: 11, Step: 61300, Step Loss: 3.82 INFO:Pipeline: Time: 04:02:29, Epoch: 11, Step: 61400, Step Loss: 3.85 INFO:Pipeline: Time: 04:02:53, Epoch: 11, Step: 61500, Step Loss: 3.86 INFO:Pipeline: Time: 04:03:16, Epoch: 11, Step: 61600, Step Loss: 3.78 INFO:Pipeline: Time: 04:03:39, Epoch: 11, Step: 61700, Step Loss: 3.85 INFO:Pipeline: Time: 04:04:03, Epoch: 11, Step: 61800, Step Loss: 3.81 INFO:Pipeline: Time: 04:04:26, Epoch: 11, Step: 61900, Step Loss: 3.79 INFO:Pipeline: Time: 04:04:49, Epoch: 11, Step: 62000, Step Loss: 3.83 INFO:Pipeline: Eval Loss: 3.70, Perplexity: 40.25 INFO:Pipeline: Time: 04:05:15, Epoch: 11, Step: 62100, Step Loss: 3.87 INFO:Pipeline: Time: 04:05:39, Epoch: 11, Step: 62200, Step Loss: 3.78 INFO:Pipeline: Time: 04:06:02, Epoch: 11, Step: 62300, Step Loss: 3.81 INFO:Pipeline: Time: 04:06:25, Epoch: 11, Step: 62400, Step Loss: 3.80 INFO:Pipeline: Time: 04:06:49, Epoch: 11, Step: 62500, Step Loss: 3.78 INFO:Pipeline: Time: 04:07:12, Epoch: 11, Step: 62600, Step Loss: 3.86 INFO:Pipeline: Time: 04:07:36, Epoch: 11, Step: 62700, Step Loss: 3.83 INFO:Pipeline: Time: 04:07:59, Epoch: 11, Step: 62800, Step Loss: 3.77 INFO:Pipeline: Time: 04:08:22, Epoch: 11, Step: 62900, Step Loss: 3.79 INFO:Pipeline: Time: 04:08:46, Epoch: 11, Step: 63000, Step Loss: 3.83 INFO:Pipeline: Eval Loss: 3.69, Perplexity: 40.09 INFO:Pipeline: Time: 04:09:11, Epoch: 11, Step: 63100, Step Loss: 3.86 INFO:Pipeline: Time: 04:09:35, Epoch: 11, Step: 63200, Step Loss: 3.83 INFO:Pipeline: Time: 04:09:58, Epoch: 11, Step: 63300, Step Loss: 3.87 INFO:Pipeline: Time: 04:10:22, Epoch: 11, Step: 63400, Step Loss: 3.82 INFO:Pipeline: Time: 04:10:45, Epoch: 11, Step: 63500, Step Loss: 3.76 INFO:Pipeline: Time: 04:11:08, Epoch: 11, Step: 63600, Step Loss: 3.83 INFO:Pipeline: Time: 04:11:32, Epoch: 11, Step: 63700, Step Loss: 3.82 INFO:Pipeline: Time: 04:11:55, Epoch: 11, Step: 63800, Step Loss: 3.76 INFO:Pipeline: Time: 04:12:18, Epoch: 11, Step: 63900, Step Loss: 3.77 INFO:Pipeline: Time: 04:12:42, Epoch: 11, Step: 64000, Step Loss: 3.80 INFO:Pipeline: Eval Loss: 3.69, Perplexity: 39.90 INFO:Pipeline: Time: 04:13:08, Epoch: 11, Step: 64100, Step Loss: 3.73 INFO:Pipeline: Time: 04:13:31, Epoch: 11, Step: 64200, Step Loss: 3.85 INFO:Pipeline: Time: 04:13:54, Epoch: 11, Step: 64300, Step Loss: 3.74 INFO:Pipeline: Time: 04:14:18, Epoch: 11, Step: 64400, Step Loss: 3.74 INFO:Pipeline: Time: 04:14:41, Epoch: 11, Step: 64500, Step Loss: 3.78 INFO:Pipeline: Time: 04:15:05, Epoch: 11, Step: 64600, Step Loss: 3.86 INFO:Pipeline: Time: 04:15:28, Epoch: 11, Step: 64700, Step Loss: 3.80 INFO:Pipeline: Time: 04:15:52, Epoch: 11, Step: 64800, Step Loss: 3.81 INFO:Pipeline: Time: 04:16:15, Epoch: 11, Step: 64900, Step Loss: 3.73 INFO:Pipeline: Time: 04:16:38, Epoch: 11, Step: 65000, Step Loss: 3.75 INFO:Pipeline: Eval Loss: 3.68, Perplexity: 39.70 INFO:Pipeline: Time: 04:17:04, Epoch: 11, Step: 65100, Step Loss: 3.85 INFO:Pipeline: Time: 04:17:28, Epoch: 11, Step: 65200, Step Loss: 3.70 INFO:Pipeline: Time: 04:17:51, Epoch: 11, Step: 65300, Step Loss: 3.80 INFO:Pipeline: Time: 04:18:15, Epoch: 11, Step: 65400, Step Loss: 3.85 INFO:Pipeline: Time: 04:18:38, Epoch: 11, Step: 65500, Step Loss: 3.78 INFO:Pipeline: Time: 04:19:02, Epoch: 11, Step: 65600, Step Loss: 3.81 INFO:Pipeline: Time: 04:19:25, Epoch: 11, Step: 65700, Step Loss: 3.75 INFO:Pipeline: Time: 04:19:48, Epoch: 11, Step: 65800, Step Loss: 3.79 INFO:Pipeline: Time: 04:20:12, Epoch: 11, Step: 65900, Step Loss: 3.82 INFO:Pipeline: Time: 04:20:35, Epoch: 11, Step: 66000, Step Loss: 3.79 INFO:Pipeline: Eval Loss: 3.68, Perplexity: 39.51 INFO:Pipeline: Time: 04:21:01, Epoch: 11, Step: 66100, Step Loss: 3.71 INFO:Pipeline: Time: 04:21:24, Epoch: 11, Step: 66200, Step Loss: 3.80 INFO:Pipeline: Time: 04:21:48, Epoch: 11, Step: 66300, Step Loss: 3.85 INFO:Pipeline: Time: 04:22:11, Epoch: 11, Step: 66400, Step Loss: 3.87 INFO:Pipeline: Time: 04:22:35, Epoch: 11, Step: 66500, Step Loss: 3.84 INFO:Pipeline: Time: 04:22:58, Epoch: 11, Step: 66600, Step Loss: 3.76 INFO:Pipeline: Time: 04:23:21, Epoch: 11, Step: 66700, Step Loss: 3.82 INFO:Pipeline: Time: 04:23:45, Epoch: 11, Step: 66800, Step Loss: 3.71 INFO:Pipeline: Time: 04:24:08, Epoch: 11, Step: 66900, Step Loss: 3.79 INFO:Pipeline: Time: 04:24:32, Epoch: 11, Step: 67000, Step Loss: 3.76 INFO:Pipeline: Eval Loss: 3.67, Perplexity: 39.41 INFO:Pipeline: Time: 04:24:57, Epoch: 11, Step: 67100, Step Loss: 3.68 INFO:Pipeline: Time: 04:25:21, Epoch: 12, Step: 67200, Step Loss: 3.84 INFO:Pipeline: Time: 04:25:44, Epoch: 12, Step: 67300, Step Loss: 3.84 INFO:Pipeline: Time: 04:26:08, Epoch: 12, Step: 67400, Step Loss: 3.77 INFO:Pipeline: Time: 04:26:31, Epoch: 12, Step: 67500, Step Loss: 3.76 INFO:Pipeline: Time: 04:26:54, Epoch: 12, Step: 67600, Step Loss: 3.79 INFO:Pipeline: Time: 04:27:18, Epoch: 12, Step: 67700, Step Loss: 3.77 INFO:Pipeline: Time: 04:27:41, Epoch: 12, Step: 67800, Step Loss: 3.83 INFO:Pipeline: Time: 04:28:05, Epoch: 12, Step: 67900, Step Loss: 3.74 INFO:Pipeline: Time: 04:28:28, Epoch: 12, Step: 68000, Step Loss: 3.83 INFO:Pipeline: Eval Loss: 3.67, Perplexity: 39.20 INFO:Pipeline: Time: 04:28:54, Epoch: 12, Step: 68100, Step Loss: 3.76 INFO:Pipeline: Time: 04:29:17, Epoch: 12, Step: 68200, Step Loss: 3.75 INFO:Pipeline: Time: 04:29:40, Epoch: 12, Step: 68300, Step Loss: 3.85 INFO:Pipeline: Time: 04:30:04, Epoch: 12, Step: 68400, Step Loss: 3.81 INFO:Pipeline: Time: 04:30:27, Epoch: 12, Step: 68500, Step Loss: 3.76 INFO:Pipeline: Time: 04:30:51, Epoch: 12, Step: 68600, Step Loss: 3.81 INFO:Pipeline: Time: 04:31:14, Epoch: 12, Step: 68700, Step Loss: 3.75 INFO:Pipeline: Time: 04:31:37, Epoch: 12, Step: 68800, Step Loss: 3.79 INFO:Pipeline: Time: 04:32:01, Epoch: 12, Step: 68900, Step Loss: 3.80 INFO:Pipeline: Time: 04:32:24, Epoch: 12, Step: 69000, Step Loss: 3.73 INFO:Pipeline: Eval Loss: 3.66, Perplexity: 39.04 INFO:Pipeline: Time: 04:32:50, Epoch: 12, Step: 69100, Step Loss: 3.83 INFO:Pipeline: Time: 04:33:13, Epoch: 12, Step: 69200, Step Loss: 3.73 INFO:Pipeline: Time: 04:33:37, Epoch: 12, Step: 69300, Step Loss: 3.84 INFO:Pipeline: Time: 04:34:00, Epoch: 12, Step: 69400, Step Loss: 3.76 INFO:Pipeline: Time: 04:34:24, Epoch: 12, Step: 69500, Step Loss: 3.78 INFO:Pipeline: Time: 04:34:47, Epoch: 12, Step: 69600, Step Loss: 3.75 INFO:Pipeline: Time: 04:35:10, Epoch: 12, Step: 69700, Step Loss: 3.77 INFO:Pipeline: Time: 04:35:34, Epoch: 12, Step: 69800, Step Loss: 3.76 INFO:Pipeline: Time: 04:35:57, Epoch: 12, Step: 69900, Step Loss: 3.79 INFO:Pipeline: Time: 04:36:20, Epoch: 12, Step: 70000, Step Loss: 3.80 INFO:Pipeline: Eval Loss: 3.66, Perplexity: 38.90 INFO:Pipeline: Time: 04:36:46, Epoch: 12, Step: 70100, Step Loss: 3.70 INFO:Pipeline: Time: 04:37:10, Epoch: 12, Step: 70200, Step Loss: 3.74 INFO:Pipeline: Time: 04:37:33, Epoch: 12, Step: 70300, Step Loss: 3.76 INFO:Pipeline: Time: 04:37:56, Epoch: 12, Step: 70400, Step Loss: 3.85 INFO:Pipeline: Time: 04:38:20, Epoch: 12, Step: 70500, Step Loss: 3.83 INFO:Pipeline: Time: 04:38:43, Epoch: 12, Step: 70600, Step Loss: 3.81 INFO:Pipeline: Time: 04:39:06, Epoch: 12, Step: 70700, Step Loss: 3.76 INFO:Pipeline: Time: 04:39:30, Epoch: 12, Step: 70800, Step Loss: 3.77 INFO:Pipeline: Time: 04:39:53, Epoch: 12, Step: 70900, Step Loss: 3.75 INFO:Pipeline: Time: 04:40:17, Epoch: 12, Step: 71000, Step Loss: 3.80 INFO:Pipeline: Eval Loss: 3.66, Perplexity: 38.81 INFO:Pipeline: Time: 04:40:43, Epoch: 12, Step: 71100, Step Loss: 3.77 INFO:Pipeline: Time: 04:41:06, Epoch: 12, Step: 71200, Step Loss: 3.73 INFO:Pipeline: Time: 04:41:30, Epoch: 12, Step: 71300, Step Loss: 3.79 INFO:Pipeline: Time: 04:41:53, Epoch: 12, Step: 71400, Step Loss: 3.74 INFO:Pipeline: Time: 04:42:16, Epoch: 12, Step: 71500, Step Loss: 3.73 INFO:Pipeline: Time: 04:42:40, Epoch: 12, Step: 71600, Step Loss: 3.80 INFO:Pipeline: Time: 04:43:03, Epoch: 12, Step: 71700, Step Loss: 3.79 INFO:Pipeline: Time: 04:43:27, Epoch: 12, Step: 71800, Step Loss: 3.75 INFO:Pipeline: Time: 04:43:50, Epoch: 12, Step: 71900, Step Loss: 3.78 INFO:Pipeline: Time: 04:44:13, Epoch: 12, Step: 72000, Step Loss: 3.88 INFO:Pipeline: Eval Loss: 3.65, Perplexity: 38.66 INFO:Pipeline: Time: 04:44:40, Epoch: 12, Step: 72100, Step Loss: 3.70 INFO:Pipeline: Time: 04:45:03, Epoch: 12, Step: 72200, Step Loss: 3.75 INFO:Pipeline: Time: 04:45:27, Epoch: 12, Step: 72300, Step Loss: 3.77 INFO:Pipeline: Time: 04:45:50, Epoch: 12, Step: 72400, Step Loss: 3.79 INFO:Pipeline: Time: 04:46:13, Epoch: 12, Step: 72500, Step Loss: 3.75 INFO:Pipeline: Time: 04:46:37, Epoch: 12, Step: 72600, Step Loss: 3.81 INFO:Pipeline: Time: 04:47:00, Epoch: 12, Step: 72700, Step Loss: 3.71 INFO:Pipeline: Time: 04:47:24, Epoch: 12, Step: 72800, Step Loss: 3.68 INFO:Pipeline: Time: 04:47:47, Epoch: 12, Step: 72900, Step Loss: 3.80 INFO:Pipeline: Time: 04:48:10, Epoch: 12, Step: 73000, Step Loss: 3.73 INFO:Pipeline: Eval Loss: 3.65, Perplexity: 38.55 INFO:Pipeline: Time: 04:48:36, Epoch: 12, Step: 73100, Step Loss: 3.69 INFO:Pipeline: Time: 04:48:59, Epoch: 12, Step: 73200, Step Loss: 3.72 INFO:Pipeline: Time: 04:49:23, Epoch: 13, Step: 73300, Step Loss: 3.83 INFO:Pipeline: Time: 04:49:46, Epoch: 13, Step: 73400, Step Loss: 3.77 INFO:Pipeline: Time: 04:50:10, Epoch: 13, Step: 73500, Step Loss: 3.77 INFO:Pipeline: Time: 04:50:33, Epoch: 13, Step: 73600, Step Loss: 3.81 INFO:Pipeline: Time: 04:50:56, Epoch: 13, Step: 73700, Step Loss: 3.76 INFO:Pipeline: Time: 04:51:20, Epoch: 13, Step: 73800, Step Loss: 3.87 INFO:Pipeline: Time: 04:51:43, Epoch: 13, Step: 73900, Step Loss: 3.78 INFO:Pipeline: Time: 04:52:06, Epoch: 13, Step: 74000, Step Loss: 3.78 INFO:Pipeline: Eval Loss: 3.65, Perplexity: 38.40 INFO:Pipeline: Time: 04:52:32, Epoch: 13, Step: 74100, Step Loss: 3.83 INFO:Pipeline: Time: 04:52:56, Epoch: 13, Step: 74200, Step Loss: 3.77 INFO:Pipeline: Time: 04:53:19, Epoch: 13, Step: 74300, Step Loss: 3.77 INFO:Pipeline: Time: 04:53:42, Epoch: 13, Step: 74400, Step Loss: 3.80 INFO:Pipeline: Time: 04:54:06, Epoch: 13, Step: 74500, Step Loss: 3.82 INFO:Pipeline: Time: 04:54:29, Epoch: 13, Step: 74600, Step Loss: 3.81 INFO:Pipeline: Time: 04:54:53, Epoch: 13, Step: 74700, Step Loss: 3.77 INFO:Pipeline: Time: 04:55:16, Epoch: 13, Step: 74800, Step Loss: 3.73 INFO:Pipeline: Time: 04:55:39, Epoch: 13, Step: 74900, Step Loss: 3.70 INFO:Pipeline: Time: 04:56:03, Epoch: 13, Step: 75000, Step Loss: 3.77 INFO:Pipeline: Eval Loss: 3.64, Perplexity: 38.26 INFO:Pipeline: Time: 04:56:28, Epoch: 13, Step: 75100, Step Loss: 3.74 INFO:Pipeline: Time: 04:56:52, Epoch: 13, Step: 75200, Step Loss: 3.71 INFO:Pipeline: Time: 04:57:15, Epoch: 13, Step: 75300, Step Loss: 3.78 INFO:Pipeline: Time: 04:57:39, Epoch: 13, Step: 75400, Step Loss: 3.78 INFO:Pipeline: Time: 04:58:02, Epoch: 13, Step: 75500, Step Loss: 3.81 INFO:Pipeline: Time: 04:58:26, Epoch: 13, Step: 75600, Step Loss: 3.74 INFO:Pipeline: Time: 04:58:49, Epoch: 13, Step: 75700, Step Loss: 3.63 INFO:Pipeline: Time: 04:59:12, Epoch: 13, Step: 75800, Step Loss: 3.76 INFO:Pipeline: Time: 04:59:36, Epoch: 13, Step: 75900, Step Loss: 3.76 INFO:Pipeline: Time: 04:59:59, Epoch: 13, Step: 76000, Step Loss: 3.78 INFO:Pipeline: Eval Loss: 3.64, Perplexity: 38.19 INFO:Pipeline: Time: 05:00:25, Epoch: 13, Step: 76100, Step Loss: 3.74 INFO:Pipeline: Time: 05:00:48, Epoch: 13, Step: 76200, Step Loss: 3.82 INFO:Pipeline: Time: 05:01:12, Epoch: 13, Step: 76300, Step Loss: 3.74 INFO:Pipeline: Time: 05:01:35, Epoch: 13, Step: 76400, Step Loss: 3.63 INFO:Pipeline: Time: 05:01:58, Epoch: 13, Step: 76500, Step Loss: 3.81 INFO:Pipeline: Time: 05:02:22, Epoch: 13, Step: 76600, Step Loss: 3.76 INFO:Pipeline: Time: 05:02:45, Epoch: 13, Step: 76700, Step Loss: 3.68 INFO:Pipeline: Time: 05:03:08, Epoch: 13, Step: 76800, Step Loss: 3.67 INFO:Pipeline: Time: 05:03:31, Epoch: 13, Step: 76900, Step Loss: 3.64 INFO:Pipeline: Time: 05:03:55, Epoch: 13, Step: 77000, Step Loss: 3.78 INFO:Pipeline: Eval Loss: 3.64, Perplexity: 38.07 INFO:Pipeline: Time: 05:04:21, Epoch: 13, Step: 77100, Step Loss: 3.70 INFO:Pipeline: Time: 05:04:44, Epoch: 13, Step: 77200, Step Loss: 3.71 INFO:Pipeline: Time: 05:05:07, Epoch: 13, Step: 77300, Step Loss: 3.69 INFO:Pipeline: Time: 05:05:31, Epoch: 13, Step: 77400, Step Loss: 3.67 INFO:Pipeline: Time: 05:05:54, Epoch: 13, Step: 77500, Step Loss: 3.74 INFO:Pipeline: Time: 05:06:17, Epoch: 13, Step: 77600, Step Loss: 3.77 INFO:Pipeline: Time: 05:06:41, Epoch: 13, Step: 77700, Step Loss: 3.73 INFO:Pipeline: Time: 05:07:04, Epoch: 13, Step: 77800, Step Loss: 3.74 INFO:Pipeline: Time: 05:07:27, Epoch: 13, Step: 77900, Step Loss: 3.72 INFO:Pipeline: Time: 05:07:51, Epoch: 13, Step: 78000, Step Loss: 3.78 INFO:Pipeline: Eval Loss: 3.64, Perplexity: 37.95 INFO:Pipeline: Time: 05:08:16, Epoch: 13, Step: 78100, Step Loss: 3.76 INFO:Pipeline: Time: 05:08:40, Epoch: 13, Step: 78200, Step Loss: 3.67 INFO:Pipeline: Time: 05:09:03, Epoch: 13, Step: 78300, Step Loss: 3.77 INFO:Pipeline: Time: 05:09:26, Epoch: 13, Step: 78400, Step Loss: 3.76 INFO:Pipeline: Time: 05:09:50, Epoch: 13, Step: 78500, Step Loss: 3.78 INFO:Pipeline: Time: 05:10:13, Epoch: 13, Step: 78600, Step Loss: 3.76 INFO:Pipeline: Time: 05:10:36, Epoch: 13, Step: 78700, Step Loss: 3.75 INFO:Pipeline: Time: 05:11:00, Epoch: 13, Step: 78800, Step Loss: 3.81 INFO:Pipeline: Time: 05:11:23, Epoch: 13, Step: 78900, Step Loss: 3.75 INFO:Pipeline: Time: 05:11:47, Epoch: 13, Step: 79000, Step Loss: 3.77 INFO:Pipeline: Eval Loss: 3.63, Perplexity: 37.85 INFO:Pipeline: Time: 05:12:12, Epoch: 13, Step: 79100, Step Loss: 3.76 INFO:Pipeline: Time: 05:12:36, Epoch: 13, Step: 79200, Step Loss: 3.75 INFO:Pipeline: Time: 05:12:59, Epoch: 13, Step: 79300, Step Loss: 3.68 INFO:Pipeline: Time: 05:13:23, Epoch: 14, Step: 79400, Step Loss: 3.65 INFO:Pipeline: Time: 05:13:46, Epoch: 14, Step: 79500, Step Loss: 3.80 INFO:Pipeline: Time: 05:14:09, Epoch: 14, Step: 79600, Step Loss: 3.82 INFO:Pipeline: Time: 05:14:33, Epoch: 14, Step: 79700, Step Loss: 3.76 INFO:Pipeline: Time: 05:14:56, Epoch: 14, Step: 79800, Step Loss: 3.75 INFO:Pipeline: Time: 05:15:20, Epoch: 14, Step: 79900, Step Loss: 3.68 INFO:Pipeline: Time: 05:15:43, Epoch: 14, Step: 80000, Step Loss: 3.71 INFO:Pipeline: Eval Loss: 3.63, Perplexity: 37.78 INFO:Pipeline: Time: 05:16:09, Epoch: 14, Step: 80100, Step Loss: 3.80 INFO:Pipeline: Time: 05:16:32, Epoch: 14, Step: 80200, Step Loss: 3.73 INFO:Pipeline: Time: 05:16:55, Epoch: 14, Step: 80300, Step Loss: 3.74 INFO:Pipeline: Time: 05:17:19, Epoch: 14, Step: 80400, Step Loss: 3.70 INFO:Pipeline: Time: 05:17:42, Epoch: 14, Step: 80500, Step Loss: 3.73 INFO:Pipeline: Time: 05:18:05, Epoch: 14, Step: 80600, Step Loss: 3.70 INFO:Pipeline: Time: 05:18:29, Epoch: 14, Step: 80700, Step Loss: 3.71 INFO:Pipeline: Time: 05:18:52, Epoch: 14, Step: 80800, Step Loss: 3.78 INFO:Pipeline: Time: 05:19:15, Epoch: 14, Step: 80900, Step Loss: 3.70 INFO:Pipeline: Time: 05:19:39, Epoch: 14, Step: 81000, Step Loss: 3.68 INFO:Pipeline: Eval Loss: 3.63, Perplexity: 37.63 INFO:Pipeline: Time: 05:20:05, Epoch: 14, Step: 81100, Step Loss: 3.74 INFO:Pipeline: Time: 05:20:28, Epoch: 14, Step: 81200, Step Loss: 3.71 INFO:Pipeline: Time: 05:20:51, Epoch: 14, Step: 81300, Step Loss: 3.73 INFO:Pipeline: Time: 05:21:15, Epoch: 14, Step: 81400, Step Loss: 3.81 INFO:Pipeline: Time: 05:21:38, Epoch: 14, Step: 81500, Step Loss: 3.74 INFO:Pipeline: Time: 05:22:01, Epoch: 14, Step: 81600, Step Loss: 3.77 INFO:Pipeline: Time: 05:22:25, Epoch: 14, Step: 81700, Step Loss: 3.79 INFO:Pipeline: Time: 05:22:48, Epoch: 14, Step: 81800, Step Loss: 3.73 INFO:Pipeline: Time: 05:23:11, Epoch: 14, Step: 81900, Step Loss: 3.81 INFO:Pipeline: Time: 05:23:35, Epoch: 14, Step: 82000, Step Loss: 3.75 INFO:Pipeline: Eval Loss: 3.63, Perplexity: 37.60 INFO:Pipeline: Time: 05:24:00, Epoch: 14, Step: 82100, Step Loss: 3.66 INFO:Pipeline: Time: 05:24:24, Epoch: 14, Step: 82200, Step Loss: 3.71 INFO:Pipeline: Time: 05:24:47, Epoch: 14, Step: 82300, Step Loss: 3.67 INFO:Pipeline: Time: 05:25:10, Epoch: 14, Step: 82400, Step Loss: 3.71 INFO:Pipeline: Time: 05:25:34, Epoch: 14, Step: 82500, Step Loss: 3.73 INFO:Pipeline: Time: 05:25:57, Epoch: 14, Step: 82600, Step Loss: 3.79 INFO:Pipeline: Time: 05:26:20, Epoch: 14, Step: 82700, Step Loss: 3.70 INFO:Pipeline: Time: 05:26:44, Epoch: 14, Step: 82800, Step Loss: 3.78 INFO:Pipeline: Time: 05:27:07, Epoch: 14, Step: 82900, Step Loss: 3.81 INFO:Pipeline: Time: 05:27:30, Epoch: 14, Step: 83000, Step Loss: 3.74 INFO:Pipeline: Eval Loss: 3.62, Perplexity: 37.51 INFO:Pipeline: Time: 05:27:56, Epoch: 14, Step: 83100, Step Loss: 3.76 INFO:Pipeline: Time: 05:28:20, Epoch: 14, Step: 83200, Step Loss: 3.71 INFO:Pipeline: Time: 05:28:43, Epoch: 14, Step: 83300, Step Loss: 3.72 INFO:Pipeline: Time: 05:29:06, Epoch: 14, Step: 83400, Step Loss: 3.73 INFO:Pipeline: Time: 05:29:30, Epoch: 14, Step: 83500, Step Loss: 3.73 INFO:Pipeline: Time: 05:29:53, Epoch: 14, Step: 83600, Step Loss: 3.67 INFO:Pipeline: Time: 05:30:16, Epoch: 14, Step: 83700, Step Loss: 3.68 INFO:Pipeline: Time: 05:30:40, Epoch: 14, Step: 83800, Step Loss: 3.71 INFO:Pipeline: Time: 05:31:03, Epoch: 14, Step: 83900, Step Loss: 3.78 INFO:Pipeline: Time: 05:31:26, Epoch: 14, Step: 84000, Step Loss: 3.74 INFO:Pipeline: Eval Loss: 3.62, Perplexity: 37.44 INFO:Pipeline: Time: 05:31:52, Epoch: 14, Step: 84100, Step Loss: 3.63 INFO:Pipeline: Time: 05:32:16, Epoch: 14, Step: 84200, Step Loss: 3.80 INFO:Pipeline: Time: 05:32:39, Epoch: 14, Step: 84300, Step Loss: 3.73 INFO:Pipeline: Time: 05:33:02, Epoch: 14, Step: 84400, Step Loss: 3.67 INFO:Pipeline: Time: 05:33:26, Epoch: 14, Step: 84500, Step Loss: 3.67 INFO:Pipeline: Time: 05:33:49, Epoch: 14, Step: 84600, Step Loss: 3.75 INFO:Pipeline: Time: 05:34:13, Epoch: 14, Step: 84700, Step Loss: 3.69 INFO:Pipeline: Time: 05:34:36, Epoch: 14, Step: 84800, Step Loss: 3.73 INFO:Pipeline: Time: 05:34:59, Epoch: 14, Step: 84900, Step Loss: 3.71 INFO:Pipeline: Time: 05:35:23, Epoch: 14, Step: 85000, Step Loss: 3.72 INFO:Pipeline: Eval Loss: 3.62, Perplexity: 37.38 INFO:Pipeline: Time: 05:35:49, Epoch: 14, Step: 85100, Step Loss: 3.78 INFO:Pipeline: Time: 05:36:12, Epoch: 14, Step: 85200, Step Loss: 3.74 INFO:Pipeline: Time: 05:36:36, Epoch: 14, Step: 85300, Step Loss: 3.65 INFO:Pipeline: Time: 05:36:59, Epoch: 14, Step: 85400, Step Loss: 3.66 INFO:Pipeline: Time: 05:37:22, Epoch: 15, Step: 85500, Step Loss: 3.80 INFO:Pipeline: Time: 05:37:46, Epoch: 15, Step: 85600, Step Loss: 3.67 INFO:Pipeline: Time: 05:38:09, Epoch: 15, Step: 85700, Step Loss: 3.70 INFO:Pipeline: Time: 05:38:33, Epoch: 15, Step: 85800, Step Loss: 3.71 INFO:Pipeline: Time: 05:38:56, Epoch: 15, Step: 85900, Step Loss: 3.70 INFO:Pipeline: Time: 05:39:19, Epoch: 15, Step: 86000, Step Loss: 3.68 INFO:Pipeline: Eval Loss: 3.62, Perplexity: 37.31 INFO:Pipeline: Time: 05:39:45, Epoch: 15, Step: 86100, Step Loss: 3.73 INFO:Pipeline: Time: 05:40:08, Epoch: 15, Step: 86200, Step Loss: 3.67 INFO:Pipeline: Time: 05:40:32, Epoch: 15, Step: 86300, Step Loss: 3.64 INFO:Pipeline: Time: 05:40:55, Epoch: 15, Step: 86400, Step Loss: 3.65 INFO:Pipeline: Time: 05:41:18, Epoch: 15, Step: 86500, Step Loss: 3.70 INFO:Pipeline: Time: 05:41:42, Epoch: 15, Step: 86600, Step Loss: 3.76 INFO:Pipeline: Time: 05:42:05, Epoch: 15, Step: 86700, Step Loss: 3.73 INFO:Pipeline: Time: 05:42:29, Epoch: 15, Step: 86800, Step Loss: 3.80 INFO:Pipeline: Time: 05:42:52, Epoch: 15, Step: 86900, Step Loss: 3.73 INFO:Pipeline: Time: 05:43:15, Epoch: 15, Step: 87000, Step Loss: 3.71 INFO:Pipeline: Eval Loss: 3.62, Perplexity: 37.22 INFO:Pipeline: Time: 05:43:41, Epoch: 15, Step: 87100, Step Loss: 3.79 INFO:Pipeline: Time: 05:44:04, Epoch: 15, Step: 87200, Step Loss: 3.65 INFO:Pipeline: Time: 05:44:28, Epoch: 15, Step: 87300, Step Loss: 3.69 INFO:Pipeline: Time: 05:44:51, Epoch: 15, Step: 87400, Step Loss: 3.72 INFO:Pipeline: Time: 05:45:14, Epoch: 15, Step: 87500, Step Loss: 3.73 INFO:Pipeline: Time: 05:45:38, Epoch: 15, Step: 87600, Step Loss: 3.66 INFO:Pipeline: Time: 05:46:01, Epoch: 15, Step: 87700, Step Loss: 3.70 INFO:Pipeline: Time: 05:46:24, Epoch: 15, Step: 87800, Step Loss: 3.69 INFO:Pipeline: Time: 05:46:48, Epoch: 15, Step: 87900, Step Loss: 3.75 INFO:Pipeline: Time: 05:47:11, Epoch: 15, Step: 88000, Step Loss: 3.76 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 37.14 INFO:Pipeline: Time: 05:47:37, Epoch: 15, Step: 88100, Step Loss: 3.73 INFO:Pipeline: Time: 05:48:00, Epoch: 15, Step: 88200, Step Loss: 3.74 INFO:Pipeline: Time: 05:48:24, Epoch: 15, Step: 88300, Step Loss: 3.70 INFO:Pipeline: Time: 05:48:47, Epoch: 15, Step: 88400, Step Loss: 3.68 INFO:Pipeline: Time: 05:49:10, Epoch: 15, Step: 88500, Step Loss: 3.75 INFO:Pipeline: Time: 05:49:34, Epoch: 15, Step: 88600, Step Loss: 3.77 INFO:Pipeline: Time: 05:49:57, Epoch: 15, Step: 88700, Step Loss: 3.74 INFO:Pipeline: Time: 05:50:20, Epoch: 15, Step: 88800, Step Loss: 3.62 INFO:Pipeline: Time: 05:50:44, Epoch: 15, Step: 88900, Step Loss: 3.69 INFO:Pipeline: Time: 05:51:07, Epoch: 15, Step: 89000, Step Loss: 3.76 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 37.11 INFO:Pipeline: Time: 05:51:33, Epoch: 15, Step: 89100, Step Loss: 3.69 INFO:Pipeline: Time: 05:51:56, Epoch: 15, Step: 89200, Step Loss: 3.81 INFO:Pipeline: Time: 05:52:19, Epoch: 15, Step: 89300, Step Loss: 3.74 INFO:Pipeline: Time: 05:52:43, Epoch: 15, Step: 89400, Step Loss: 3.73 INFO:Pipeline: Time: 05:53:06, Epoch: 15, Step: 89500, Step Loss: 3.70 INFO:Pipeline: Time: 05:53:30, Epoch: 15, Step: 89600, Step Loss: 3.67 INFO:Pipeline: Time: 05:53:53, Epoch: 15, Step: 89700, Step Loss: 3.63 INFO:Pipeline: Time: 05:54:16, Epoch: 15, Step: 89800, Step Loss: 3.70 INFO:Pipeline: Time: 05:54:40, Epoch: 15, Step: 89900, Step Loss: 3.79 INFO:Pipeline: Time: 05:55:03, Epoch: 15, Step: 90000, Step Loss: 3.67 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 37.04 INFO:Pipeline: Time: 05:55:29, Epoch: 15, Step: 90100, Step Loss: 3.65 INFO:Pipeline: Time: 05:55:52, Epoch: 15, Step: 90200, Step Loss: 3.64 INFO:Pipeline: Time: 05:56:16, Epoch: 15, Step: 90300, Step Loss: 3.73 INFO:Pipeline: Time: 05:56:39, Epoch: 15, Step: 90400, Step Loss: 3.73 INFO:Pipeline: Time: 05:57:02, Epoch: 15, Step: 90500, Step Loss: 3.69 INFO:Pipeline: Time: 05:57:26, Epoch: 15, Step: 90600, Step Loss: 3.70 INFO:Pipeline: Time: 05:57:49, Epoch: 15, Step: 90700, Step Loss: 3.65 INFO:Pipeline: Time: 05:58:12, Epoch: 15, Step: 90800, Step Loss: 3.76 INFO:Pipeline: Time: 05:58:36, Epoch: 15, Step: 90900, Step Loss: 3.69 INFO:Pipeline: Time: 05:58:59, Epoch: 15, Step: 91000, Step Loss: 3.75 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 37.00 INFO:Pipeline: Time: 05:59:25, Epoch: 15, Step: 91100, Step Loss: 3.66 INFO:Pipeline: Time: 05:59:48, Epoch: 15, Step: 91200, Step Loss: 3.76 INFO:Pipeline: Time: 06:00:11, Epoch: 15, Step: 91300, Step Loss: 3.73 INFO:Pipeline: Time: 06:00:35, Epoch: 15, Step: 91400, Step Loss: 3.73 INFO:Pipeline: Time: 06:00:58, Epoch: 15, Step: 91500, Step Loss: 3.73 INFO:Pipeline: Time: 06:01:21, Epoch: 16, Step: 91600, Step Loss: 3.73 INFO:Pipeline: Time: 06:01:45, Epoch: 16, Step: 91700, Step Loss: 3.72 INFO:Pipeline: Time: 06:02:08, Epoch: 16, Step: 91800, Step Loss: 3.85 INFO:Pipeline: Time: 06:02:31, Epoch: 16, Step: 91900, Step Loss: 3.76 INFO:Pipeline: Time: 06:02:55, Epoch: 16, Step: 92000, Step Loss: 3.77 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 36.93 INFO:Pipeline: Time: 06:03:20, Epoch: 16, Step: 92100, Step Loss: 3.73 INFO:Pipeline: Time: 06:03:44, Epoch: 16, Step: 92200, Step Loss: 3.70 INFO:Pipeline: Time: 06:04:07, Epoch: 16, Step: 92300, Step Loss: 3.72 INFO:Pipeline: Time: 06:04:30, Epoch: 16, Step: 92400, Step Loss: 3.81 INFO:Pipeline: Time: 06:04:54, Epoch: 16, Step: 92500, Step Loss: 3.69 INFO:Pipeline: Time: 06:05:17, Epoch: 16, Step: 92600, Step Loss: 3.68 INFO:Pipeline: Time: 06:05:40, Epoch: 16, Step: 92700, Step Loss: 3.63 INFO:Pipeline: Time: 06:06:03, Epoch: 16, Step: 92800, Step Loss: 3.67 INFO:Pipeline: Time: 06:06:27, Epoch: 16, Step: 92900, Step Loss: 3.74 INFO:Pipeline: Time: 06:06:50, Epoch: 16, Step: 93000, Step Loss: 3.73 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 36.87 INFO:Pipeline: Time: 06:07:16, Epoch: 16, Step: 93100, Step Loss: 3.67 INFO:Pipeline: Time: 06:07:39, Epoch: 16, Step: 93200, Step Loss: 3.75 INFO:Pipeline: Time: 06:08:02, Epoch: 16, Step: 93300, Step Loss: 3.65 INFO:Pipeline: Time: 06:08:25, Epoch: 16, Step: 93400, Step Loss: 3.64 INFO:Pipeline: Time: 06:08:49, Epoch: 16, Step: 93500, Step Loss: 3.62 INFO:Pipeline: Time: 06:09:12, Epoch: 16, Step: 93600, Step Loss: 3.65 INFO:Pipeline: Time: 06:09:35, Epoch: 16, Step: 93700, Step Loss: 3.70 INFO:Pipeline: Time: 06:09:59, Epoch: 16, Step: 93800, Step Loss: 3.62 INFO:Pipeline: Time: 06:10:22, Epoch: 16, Step: 93900, Step Loss: 3.68 INFO:Pipeline: Time: 06:10:45, Epoch: 16, Step: 94000, Step Loss: 3.64 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 36.83 INFO:Pipeline: Time: 06:11:11, Epoch: 16, Step: 94100, Step Loss: 3.76 INFO:Pipeline: Time: 06:11:34, Epoch: 16, Step: 94200, Step Loss: 3.76 INFO:Pipeline: Time: 06:11:57, Epoch: 16, Step: 94300, Step Loss: 3.74 INFO:Pipeline: Time: 06:12:21, Epoch: 16, Step: 94400, Step Loss: 3.68 INFO:Pipeline: Time: 06:12:44, Epoch: 16, Step: 94500, Step Loss: 3.68 INFO:Pipeline: Time: 06:13:07, Epoch: 16, Step: 94600, Step Loss: 3.68 INFO:Pipeline: Time: 06:13:31, Epoch: 16, Step: 94700, Step Loss: 3.69 INFO:Pipeline: Time: 06:13:54, Epoch: 16, Step: 94800, Step Loss: 3.77 INFO:Pipeline: Time: 06:14:17, Epoch: 16, Step: 94900, Step Loss: 3.71 INFO:Pipeline: Time: 06:14:41, Epoch: 16, Step: 95000, Step Loss: 3.68 INFO:Pipeline: Eval Loss: 3.61, Perplexity: 36.78 INFO:Pipeline: Time: 06:15:06, Epoch: 16, Step: 95100, Step Loss: 3.77 INFO:Pipeline: Time: 06:15:29, Epoch: 16, Step: 95200, Step Loss: 3.64 INFO:Pipeline: Time: 06:15:53, Epoch: 16, Step: 95300, Step Loss: 3.62 INFO:Pipeline: Time: 06:16:16, Epoch: 16, Step: 95400, Step Loss: 3.68 INFO:Pipeline: Time: 06:16:39, Epoch: 16, Step: 95500, Step Loss: 3.68 INFO:Pipeline: Time: 06:17:02, Epoch: 16, Step: 95600, Step Loss: 3.65 INFO:Pipeline: Time: 06:17:26, Epoch: 16, Step: 95700, Step Loss: 3.66 INFO:Pipeline: Time: 06:17:49, Epoch: 16, Step: 95800, Step Loss: 3.71 INFO:Pipeline: Time: 06:18:12, Epoch: 16, Step: 95900, Step Loss: 3.72 INFO:Pipeline: Time: 06:18:35, Epoch: 16, Step: 96000, Step Loss: 3.67 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.75 INFO:Pipeline: Time: 06:19:01, Epoch: 16, Step: 96100, Step Loss: 3.73 INFO:Pipeline: Time: 06:19:24, Epoch: 16, Step: 96200, Step Loss: 3.73 INFO:Pipeline: Time: 06:19:48, Epoch: 16, Step: 96300, Step Loss: 3.69 INFO:Pipeline: Time: 06:20:11, Epoch: 16, Step: 96400, Step Loss: 3.64 INFO:Pipeline: Time: 06:20:34, Epoch: 16, Step: 96500, Step Loss: 3.70 INFO:Pipeline: Time: 06:20:57, Epoch: 16, Step: 96600, Step Loss: 3.69 INFO:Pipeline: Time: 06:21:21, Epoch: 16, Step: 96700, Step Loss: 3.71 INFO:Pipeline: Time: 06:21:44, Epoch: 16, Step: 96800, Step Loss: 3.76 INFO:Pipeline: Time: 06:22:07, Epoch: 16, Step: 96900, Step Loss: 3.77 INFO:Pipeline: Time: 06:22:31, Epoch: 16, Step: 97000, Step Loss: 3.70 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.70 INFO:Pipeline: Time: 06:22:56, Epoch: 16, Step: 97100, Step Loss: 3.71 INFO:Pipeline: Time: 06:23:19, Epoch: 16, Step: 97200, Step Loss: 3.66 INFO:Pipeline: Time: 06:23:43, Epoch: 16, Step: 97300, Step Loss: 3.71 INFO:Pipeline: Time: 06:24:06, Epoch: 16, Step: 97400, Step Loss: 3.75 INFO:Pipeline: Time: 06:24:29, Epoch: 16, Step: 97500, Step Loss: 3.68 INFO:Pipeline: Time: 06:24:53, Epoch: 16, Step: 97600, Step Loss: 3.63 INFO:Pipeline: Time: 06:25:16, Epoch: 17, Step: 97700, Step Loss: 3.67 INFO:Pipeline: Time: 06:25:39, Epoch: 17, Step: 97800, Step Loss: 3.70 INFO:Pipeline: Time: 06:26:02, Epoch: 17, Step: 97900, Step Loss: 3.77 INFO:Pipeline: Time: 06:26:26, Epoch: 17, Step: 98000, Step Loss: 3.69 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.68 INFO:Pipeline: Time: 06:26:51, Epoch: 17, Step: 98100, Step Loss: 3.66 INFO:Pipeline: Time: 06:27:15, Epoch: 17, Step: 98200, Step Loss: 3.66 INFO:Pipeline: Time: 06:27:38, Epoch: 17, Step: 98300, Step Loss: 3.74 INFO:Pipeline: Time: 06:28:01, Epoch: 17, Step: 98400, Step Loss: 3.74 INFO:Pipeline: Time: 06:28:24, Epoch: 17, Step: 98500, Step Loss: 3.66 INFO:Pipeline: Time: 06:28:48, Epoch: 17, Step: 98600, Step Loss: 3.66 INFO:Pipeline: Time: 06:29:11, Epoch: 17, Step: 98700, Step Loss: 3.69 INFO:Pipeline: Time: 06:29:34, Epoch: 17, Step: 98800, Step Loss: 3.80 INFO:Pipeline: Time: 06:29:58, Epoch: 17, Step: 98900, Step Loss: 3.67 INFO:Pipeline: Time: 06:30:21, Epoch: 17, Step: 99000, Step Loss: 3.64 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.62 INFO:Pipeline: Time: 06:30:46, Epoch: 17, Step: 99100, Step Loss: 3.68 INFO:Pipeline: Time: 06:31:10, Epoch: 17, Step: 99200, Step Loss: 3.71 INFO:Pipeline: Time: 06:31:33, Epoch: 17, Step: 99300, Step Loss: 3.68 INFO:Pipeline: Time: 06:31:56, Epoch: 17, Step: 99400, Step Loss: 3.65 INFO:Pipeline: Time: 06:32:20, Epoch: 17, Step: 99500, Step Loss: 3.63 INFO:Pipeline: Time: 06:32:43, Epoch: 17, Step: 99600, Step Loss: 3.73 INFO:Pipeline: Time: 06:33:06, Epoch: 17, Step: 99700, Step Loss: 3.79 INFO:Pipeline: Time: 06:33:30, Epoch: 17, Step: 99800, Step Loss: 3.69 INFO:Pipeline: Time: 06:33:53, Epoch: 17, Step: 99900, Step Loss: 3.71 INFO:Pipeline: Time: 06:34:16, Epoch: 17, Step: 100000, Step Loss: 3.72 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.60 INFO:Pipeline: Time: 06:34:42, Epoch: 17, Step: 100100, Step Loss: 3.68 INFO:Pipeline: Time: 06:35:05, Epoch: 17, Step: 100200, Step Loss: 3.68 INFO:Pipeline: Time: 06:35:28, Epoch: 17, Step: 100300, Step Loss: 3.70 INFO:Pipeline: Time: 06:35:52, Epoch: 17, Step: 100400, Step Loss: 3.66 INFO:Pipeline: Time: 06:36:15, Epoch: 17, Step: 100500, Step Loss: 3.67 INFO:Pipeline: Time: 06:36:38, Epoch: 17, Step: 100600, Step Loss: 3.65 INFO:Pipeline: Time: 06:37:01, Epoch: 17, Step: 100700, Step Loss: 3.68 INFO:Pipeline: Time: 06:37:25, Epoch: 17, Step: 100800, Step Loss: 3.62 INFO:Pipeline: Time: 06:37:48, Epoch: 17, Step: 100900, Step Loss: 3.77 INFO:Pipeline: Time: 06:38:11, Epoch: 17, Step: 101000, Step Loss: 3.72 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.57 INFO:Pipeline: Time: 06:38:38, Epoch: 17, Step: 101100, Step Loss: 3.73 INFO:Pipeline: Time: 06:39:01, Epoch: 17, Step: 101200, Step Loss: 3.71 INFO:Pipeline: Time: 06:39:25, Epoch: 17, Step: 101300, Step Loss: 3.71 INFO:Pipeline: Time: 06:39:48, Epoch: 17, Step: 101400, Step Loss: 3.66 INFO:Pipeline: Time: 06:40:11, Epoch: 17, Step: 101500, Step Loss: 3.73 INFO:Pipeline: Time: 06:40:34, Epoch: 17, Step: 101600, Step Loss: 3.70 INFO:Pipeline: Time: 06:40:58, Epoch: 17, Step: 101700, Step Loss: 3.72 INFO:Pipeline: Time: 06:41:21, Epoch: 17, Step: 101800, Step Loss: 3.65 INFO:Pipeline: Time: 06:41:44, Epoch: 17, Step: 101900, Step Loss: 3.82 INFO:Pipeline: Time: 06:42:07, Epoch: 17, Step: 102000, Step Loss: 3.70 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.56 INFO:Pipeline: Time: 06:42:33, Epoch: 17, Step: 102100, Step Loss: 3.68 INFO:Pipeline: Time: 06:42:56, Epoch: 17, Step: 102200, Step Loss: 3.70 INFO:Pipeline: Time: 06:43:19, Epoch: 17, Step: 102300, Step Loss: 3.70 INFO:Pipeline: Time: 06:43:43, Epoch: 17, Step: 102400, Step Loss: 3.65 INFO:Pipeline: Time: 06:44:06, Epoch: 17, Step: 102500, Step Loss: 3.71 INFO:Pipeline: Time: 06:44:29, Epoch: 17, Step: 102600, Step Loss: 3.68 INFO:Pipeline: Time: 06:44:53, Epoch: 17, Step: 102700, Step Loss: 3.68 INFO:Pipeline: Time: 06:45:16, Epoch: 17, Step: 102800, Step Loss: 3.72 INFO:Pipeline: Time: 06:45:39, Epoch: 17, Step: 102900, Step Loss: 3.72 INFO:Pipeline: Time: 06:46:03, Epoch: 17, Step: 103000, Step Loss: 3.71 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.52 INFO:Pipeline: Time: 06:46:28, Epoch: 17, Step: 103100, Step Loss: 3.70 INFO:Pipeline: Time: 06:46:51, Epoch: 17, Step: 103200, Step Loss: 3.73 INFO:Pipeline: Time: 06:47:15, Epoch: 17, Step: 103300, Step Loss: 3.70 INFO:Pipeline: Time: 06:47:38, Epoch: 17, Step: 103400, Step Loss: 3.73 INFO:Pipeline: Time: 06:48:01, Epoch: 17, Step: 103500, Step Loss: 3.76 INFO:Pipeline: Time: 06:48:25, Epoch: 17, Step: 103600, Step Loss: 3.66 INFO:Pipeline: Time: 06:48:48, Epoch: 17, Step: 103700, Step Loss: 3.71 INFO:Pipeline: Time: 06:49:11, Epoch: 18, Step: 103800, Step Loss: 3.77 INFO:Pipeline: Time: 06:49:34, Epoch: 18, Step: 103900, Step Loss: 3.71 INFO:Pipeline: Time: 06:49:58, Epoch: 18, Step: 104000, Step Loss: 3.69 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.51 INFO:Pipeline: Time: 06:50:23, Epoch: 18, Step: 104100, Step Loss: 3.62 INFO:Pipeline: Time: 06:50:47, Epoch: 18, Step: 104200, Step Loss: 3.63 INFO:Pipeline: Time: 06:51:10, Epoch: 18, Step: 104300, Step Loss: 3.66 INFO:Pipeline: Time: 06:51:33, Epoch: 18, Step: 104400, Step Loss: 3.74 INFO:Pipeline: Time: 06:51:57, Epoch: 18, Step: 104500, Step Loss: 3.61 INFO:Pipeline: Time: 06:52:20, Epoch: 18, Step: 104600, Step Loss: 3.72 INFO:Pipeline: Time: 06:52:43, Epoch: 18, Step: 104700, Step Loss: 3.67 INFO:Pipeline: Time: 06:53:07, Epoch: 18, Step: 104800, Step Loss: 3.62 INFO:Pipeline: Time: 06:53:30, Epoch: 18, Step: 104900, Step Loss: 3.76 INFO:Pipeline: Time: 06:53:53, Epoch: 18, Step: 105000, Step Loss: 3.74 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.50 INFO:Pipeline: Time: 06:54:19, Epoch: 18, Step: 105100, Step Loss: 3.65 INFO:Pipeline: Time: 06:54:42, Epoch: 18, Step: 105200, Step Loss: 3.76 INFO:Pipeline: Time: 06:55:05, Epoch: 18, Step: 105300, Step Loss: 3.70 INFO:Pipeline: Time: 06:55:29, Epoch: 18, Step: 105400, Step Loss: 3.67 INFO:Pipeline: Time: 06:55:52, Epoch: 18, Step: 105500, Step Loss: 3.56 INFO:Pipeline: Time: 06:56:15, Epoch: 18, Step: 105600, Step Loss: 3.73 INFO:Pipeline: Time: 06:56:38, Epoch: 18, Step: 105700, Step Loss: 3.72 INFO:Pipeline: Time: 06:57:02, Epoch: 18, Step: 105800, Step Loss: 3.70 INFO:Pipeline: Time: 06:57:25, Epoch: 18, Step: 105900, Step Loss: 3.68 INFO:Pipeline: Time: 06:57:48, Epoch: 18, Step: 106000, Step Loss: 3.61 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.47 INFO:Pipeline: Time: 06:58:14, Epoch: 18, Step: 106100, Step Loss: 3.72 INFO:Pipeline: Time: 06:58:37, Epoch: 18, Step: 106200, Step Loss: 3.75 INFO:Pipeline: Time: 06:59:00, Epoch: 18, Step: 106300, Step Loss: 3.71 INFO:Pipeline: Time: 06:59:24, Epoch: 18, Step: 106400, Step Loss: 3.73 INFO:Pipeline: Time: 06:59:47, Epoch: 18, Step: 106500, Step Loss: 3.66 INFO:Pipeline: Time: 07:00:10, Epoch: 18, Step: 106600, Step Loss: 3.71 INFO:Pipeline: Time: 07:00:34, Epoch: 18, Step: 106700, Step Loss: 3.69 INFO:Pipeline: Time: 07:00:57, Epoch: 18, Step: 106800, Step Loss: 3.62 INFO:Pipeline: Time: 07:01:20, Epoch: 18, Step: 106900, Step Loss: 3.68 INFO:Pipeline: Time: 07:01:44, Epoch: 18, Step: 107000, Step Loss: 3.76 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.45 INFO:Pipeline: Time: 07:02:09, Epoch: 18, Step: 107100, Step Loss: 3.66 INFO:Pipeline: Time: 07:02:32, Epoch: 18, Step: 107200, Step Loss: 3.62 INFO:Pipeline: Time: 07:02:56, Epoch: 18, Step: 107300, Step Loss: 3.67 INFO:Pipeline: Time: 07:03:19, Epoch: 18, Step: 107400, Step Loss: 3.66 INFO:Pipeline: Time: 07:03:42, Epoch: 18, Step: 107500, Step Loss: 3.66 INFO:Pipeline: Time: 07:04:06, Epoch: 18, Step: 107600, Step Loss: 3.64 INFO:Pipeline: Time: 07:04:29, Epoch: 18, Step: 107700, Step Loss: 3.59 INFO:Pipeline: Time: 07:04:52, Epoch: 18, Step: 107800, Step Loss: 3.69 INFO:Pipeline: Time: 07:05:15, Epoch: 18, Step: 107900, Step Loss: 3.64 INFO:Pipeline: Time: 07:05:39, Epoch: 18, Step: 108000, Step Loss: 3.72 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.45 INFO:Pipeline: Time: 07:06:03, Epoch: 18, Step: 108100, Step Loss: 3.67 INFO:Pipeline: Time: 07:06:27, Epoch: 18, Step: 108200, Step Loss: 3.71 INFO:Pipeline: Time: 07:06:50, Epoch: 18, Step: 108300, Step Loss: 3.75 INFO:Pipeline: Time: 07:07:13, Epoch: 18, Step: 108400, Step Loss: 3.66 INFO:Pipeline: Time: 07:07:37, Epoch: 18, Step: 108500, Step Loss: 3.73 INFO:Pipeline: Time: 07:08:00, Epoch: 18, Step: 108600, Step Loss: 3.62 INFO:Pipeline: Time: 07:08:23, Epoch: 18, Step: 108700, Step Loss: 3.64 INFO:Pipeline: Time: 07:08:47, Epoch: 18, Step: 108800, Step Loss: 3.67 INFO:Pipeline: Time: 07:09:10, Epoch: 18, Step: 108900, Step Loss: 3.70 INFO:Pipeline: Time: 07:09:33, Epoch: 18, Step: 109000, Step Loss: 3.67 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.43 INFO:Pipeline: Time: 07:09:59, Epoch: 18, Step: 109100, Step Loss: 3.66 INFO:Pipeline: Time: 07:10:22, Epoch: 18, Step: 109200, Step Loss: 3.68 INFO:Pipeline: Time: 07:10:45, Epoch: 18, Step: 109300, Step Loss: 3.61 INFO:Pipeline: Time: 07:11:09, Epoch: 18, Step: 109400, Step Loss: 3.70 INFO:Pipeline: Time: 07:11:32, Epoch: 18, Step: 109500, Step Loss: 3.70 INFO:Pipeline: Time: 07:11:56, Epoch: 18, Step: 109600, Step Loss: 3.68 INFO:Pipeline: Time: 07:12:19, Epoch: 18, Step: 109700, Step Loss: 3.70 INFO:Pipeline: Time: 07:12:42, Epoch: 18, Step: 109800, Step Loss: 3.66 INFO:Pipeline: Time: 07:13:06, Epoch: 19, Step: 109900, Step Loss: 3.69 INFO:Pipeline: Time: 07:13:29, Epoch: 19, Step: 110000, Step Loss: 3.70 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.42 INFO:Pipeline: Time: 07:13:54, Epoch: 19, Step: 110100, Step Loss: 3.73 INFO:Pipeline: Time: 07:14:18, Epoch: 19, Step: 110200, Step Loss: 3.70 INFO:Pipeline: Time: 07:14:41, Epoch: 19, Step: 110300, Step Loss: 3.73 INFO:Pipeline: Time: 07:15:04, Epoch: 19, Step: 110400, Step Loss: 3.69 INFO:Pipeline: Time: 07:15:28, Epoch: 19, Step: 110500, Step Loss: 3.74 INFO:Pipeline: Time: 07:15:51, Epoch: 19, Step: 110600, Step Loss: 3.64 INFO:Pipeline: Time: 07:16:14, Epoch: 19, Step: 110700, Step Loss: 3.68 INFO:Pipeline: Time: 07:16:38, Epoch: 19, Step: 110800, Step Loss: 3.72 INFO:Pipeline: Time: 07:17:01, Epoch: 19, Step: 110900, Step Loss: 3.73 INFO:Pipeline: Time: 07:17:24, Epoch: 19, Step: 111000, Step Loss: 3.73 INFO:Pipeline: Eval Loss: 3.60, Perplexity: 36.43 INFO:Pipeline: Time: 07:17:49, Epoch: 19, Step: 111100, Step Loss: 3.62 INFO:Pipeline: Time: 07:18:12, Epoch: 19, Step: 111200, Step Loss: 3.69 INFO:Pipeline: Time: 07:18:35, Epoch: 19, Step: 111300, Step Loss: 3.72 INFO:Pipeline: Time: 07:18:59, Epoch: 19, Step: 111400, Step Loss: 3.76 INFO:Pipeline: Time: 07:19:22, Epoch: 19, Step: 111500, Step Loss: 3.54 INFO:Pipeline: Time: 07:19:45, Epoch: 19, Step: 111600, Step Loss: 3.63 INFO:Pipeline: Time: 07:20:09, Epoch: 19, Step: 111700, Step Loss: 3.67 INFO:Pipeline: Time: 07:20:32, Epoch: 19, Step: 111800, Step Loss: 3.70 INFO:Pipeline: Time: 07:20:55, Epoch: 19, Step: 111900, Step Loss: 3.68 INFO:Pipeline: Time: 07:21:19, Epoch: 19, Step: 112000, Step Loss: 3.73 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.40 INFO:Pipeline: Time: 07:21:44, Epoch: 19, Step: 112100, Step Loss: 3.60 INFO:Pipeline: Time: 07:22:08, Epoch: 19, Step: 112200, Step Loss: 3.61 INFO:Pipeline: Time: 07:22:31, Epoch: 19, Step: 112300, Step Loss: 3.66 INFO:Pipeline: Time: 07:22:54, Epoch: 19, Step: 112400, Step Loss: 3.74 INFO:Pipeline: Time: 07:23:18, Epoch: 19, Step: 112500, Step Loss: 3.74 INFO:Pipeline: Time: 07:23:41, Epoch: 19, Step: 112600, Step Loss: 3.70 INFO:Pipeline: Time: 07:24:04, Epoch: 19, Step: 112700, Step Loss: 3.66 INFO:Pipeline: Time: 07:24:28, Epoch: 19, Step: 112800, Step Loss: 3.75 INFO:Pipeline: Time: 07:24:51, Epoch: 19, Step: 112900, Step Loss: 3.71 INFO:Pipeline: Time: 07:25:14, Epoch: 19, Step: 113000, Step Loss: 3.64 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.40 INFO:Pipeline: Time: 07:25:40, Epoch: 19, Step: 113100, Step Loss: 3.75 INFO:Pipeline: Time: 07:26:03, Epoch: 19, Step: 113200, Step Loss: 3.70 INFO:Pipeline: Time: 07:26:27, Epoch: 19, Step: 113300, Step Loss: 3.71 INFO:Pipeline: Time: 07:26:50, Epoch: 19, Step: 113400, Step Loss: 3.74 INFO:Pipeline: Time: 07:27:14, Epoch: 19, Step: 113500, Step Loss: 3.63 INFO:Pipeline: Time: 07:27:37, Epoch: 19, Step: 113600, Step Loss: 3.72 INFO:Pipeline: Time: 07:28:00, Epoch: 19, Step: 113700, Step Loss: 3.69 INFO:Pipeline: Time: 07:28:24, Epoch: 19, Step: 113800, Step Loss: 3.62 INFO:Pipeline: Time: 07:28:47, Epoch: 19, Step: 113900, Step Loss: 3.62 INFO:Pipeline: Time: 07:29:10, Epoch: 19, Step: 114000, Step Loss: 3.76 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.39 INFO:Pipeline: Time: 07:29:36, Epoch: 19, Step: 114100, Step Loss: 3.74 INFO:Pipeline: Time: 07:29:59, Epoch: 19, Step: 114200, Step Loss: 3.68 INFO:Pipeline: Time: 07:30:23, Epoch: 19, Step: 114300, Step Loss: 3.65 INFO:Pipeline: Time: 07:30:46, Epoch: 19, Step: 114400, Step Loss: 3.77 INFO:Pipeline: Time: 07:31:09, Epoch: 19, Step: 114500, Step Loss: 3.73 INFO:Pipeline: Time: 07:31:33, Epoch: 19, Step: 114600, Step Loss: 3.65 INFO:Pipeline: Time: 07:31:56, Epoch: 19, Step: 114700, Step Loss: 3.73 INFO:Pipeline: Time: 07:32:19, Epoch: 19, Step: 114800, Step Loss: 3.72 INFO:Pipeline: Time: 07:32:43, Epoch: 19, Step: 114900, Step Loss: 3.67 INFO:Pipeline: Time: 07:33:06, Epoch: 19, Step: 115000, Step Loss: 3.71 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.39 INFO:Pipeline: Time: 07:33:32, Epoch: 19, Step: 115100, Step Loss: 3.67 INFO:Pipeline: Time: 07:33:55, Epoch: 19, Step: 115200, Step Loss: 3.67 INFO:Pipeline: Time: 07:34:18, Epoch: 19, Step: 115300, Step Loss: 3.57 INFO:Pipeline: Time: 07:34:42, Epoch: 19, Step: 115400, Step Loss: 3.67 INFO:Pipeline: Time: 07:35:05, Epoch: 19, Step: 115500, Step Loss: 3.61 INFO:Pipeline: Time: 07:35:28, Epoch: 19, Step: 115600, Step Loss: 3.60 INFO:Pipeline: Time: 07:35:51, Epoch: 19, Step: 115700, Step Loss: 3.69 INFO:Pipeline: Time: 07:36:15, Epoch: 19, Step: 115800, Step Loss: 3.71 INFO:Pipeline: Time: 07:36:38, Epoch: 19, Step: 115900, Step Loss: 3.65 INFO:Pipeline: Time: 07:37:01, Epoch: 20, Step: 116000, Step Loss: 3.71 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.39 INFO:Pipeline: Time: 07:37:26, Epoch: 20, Step: 116100, Step Loss: 3.70 INFO:Pipeline: Time: 07:37:49, Epoch: 20, Step: 116200, Step Loss: 3.69 INFO:Pipeline: Time: 07:38:12, Epoch: 20, Step: 116300, Step Loss: 3.71 INFO:Pipeline: Time: 07:38:36, Epoch: 20, Step: 116400, Step Loss: 3.66 INFO:Pipeline: Time: 07:38:59, Epoch: 20, Step: 116500, Step Loss: 3.69 INFO:Pipeline: Time: 07:39:22, Epoch: 20, Step: 116600, Step Loss: 3.72 INFO:Pipeline: Time: 07:39:46, Epoch: 20, Step: 116700, Step Loss: 3.63 INFO:Pipeline: Time: 07:40:09, Epoch: 20, Step: 116800, Step Loss: 3.71 INFO:Pipeline: Time: 07:40:32, Epoch: 20, Step: 116900, Step Loss: 3.66 INFO:Pipeline: Time: 07:40:55, Epoch: 20, Step: 117000, Step Loss: 3.63 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.38 INFO:Pipeline: Time: 07:41:21, Epoch: 20, Step: 117100, Step Loss: 3.69 INFO:Pipeline: Time: 07:41:44, Epoch: 20, Step: 117200, Step Loss: 3.64 INFO:Pipeline: Time: 07:42:07, Epoch: 20, Step: 117300, Step Loss: 3.74 INFO:Pipeline: Time: 07:42:31, Epoch: 20, Step: 117400, Step Loss: 3.69 INFO:Pipeline: Time: 07:42:54, Epoch: 20, Step: 117500, Step Loss: 3.70 INFO:Pipeline: Time: 07:43:17, Epoch: 20, Step: 117600, Step Loss: 3.64 INFO:Pipeline: Time: 07:43:40, Epoch: 20, Step: 117700, Step Loss: 3.67 INFO:Pipeline: Time: 07:44:04, Epoch: 20, Step: 117800, Step Loss: 3.65 INFO:Pipeline: Time: 07:44:27, Epoch: 20, Step: 117900, Step Loss: 3.66 INFO:Pipeline: Time: 07:44:50, Epoch: 20, Step: 118000, Step Loss: 3.63 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.38 INFO:Pipeline: Time: 07:45:16, Epoch: 20, Step: 118100, Step Loss: 3.69 INFO:Pipeline: Time: 07:45:39, Epoch: 20, Step: 118200, Step Loss: 3.73 INFO:Pipeline: Time: 07:46:02, Epoch: 20, Step: 118300, Step Loss: 3.70 INFO:Pipeline: Time: 07:46:25, Epoch: 20, Step: 118400, Step Loss: 3.63 INFO:Pipeline: Time: 07:46:49, Epoch: 20, Step: 118500, Step Loss: 3.68 INFO:Pipeline: Time: 07:47:12, Epoch: 20, Step: 118600, Step Loss: 3.69 INFO:Pipeline: Time: 07:47:35, Epoch: 20, Step: 118700, Step Loss: 3.67 INFO:Pipeline: Time: 07:47:59, Epoch: 20, Step: 118800, Step Loss: 3.69 INFO:Pipeline: Time: 07:48:22, Epoch: 20, Step: 118900, Step Loss: 3.68 INFO:Pipeline: Time: 07:48:45, Epoch: 20, Step: 119000, Step Loss: 3.67 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.38 INFO:Pipeline: Time: 07:49:11, Epoch: 20, Step: 119100, Step Loss: 3.62 INFO:Pipeline: Time: 07:49:34, Epoch: 20, Step: 119200, Step Loss: 3.61 INFO:Pipeline: Time: 07:49:57, Epoch: 20, Step: 119300, Step Loss: 3.67 INFO:Pipeline: Time: 07:50:20, Epoch: 20, Step: 119400, Step Loss: 3.68 INFO:Pipeline: Time: 07:50:44, Epoch: 20, Step: 119500, Step Loss: 3.64 INFO:Pipeline: Time: 07:51:07, Epoch: 20, Step: 119600, Step Loss: 3.67 INFO:Pipeline: Time: 07:51:30, Epoch: 20, Step: 119700, Step Loss: 3.71 INFO:Pipeline: Time: 07:51:53, Epoch: 20, Step: 119800, Step Loss: 3.60 INFO:Pipeline: Time: 07:52:17, Epoch: 20, Step: 119900, Step Loss: 3.67 INFO:Pipeline: Time: 07:52:40, Epoch: 20, Step: 120000, Step Loss: 3.64 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.38 INFO:Pipeline: Time: 07:53:06, Epoch: 20, Step: 120100, Step Loss: 3.61 INFO:Pipeline: Time: 07:53:29, Epoch: 20, Step: 120200, Step Loss: 3.67 INFO:Pipeline: Time: 07:53:52, Epoch: 20, Step: 120300, Step Loss: 3.72 INFO:Pipeline: Time: 07:54:15, Epoch: 20, Step: 120400, Step Loss: 3.68 INFO:Pipeline: Time: 07:54:39, Epoch: 20, Step: 120500, Step Loss: 3.72 INFO:Pipeline: Time: 07:55:02, Epoch: 20, Step: 120600, Step Loss: 3.69 INFO:Pipeline: Time: 07:55:25, Epoch: 20, Step: 120700, Step Loss: 3.71 INFO:Pipeline: Time: 07:55:49, Epoch: 20, Step: 120800, Step Loss: 3.69 INFO:Pipeline: Time: 07:56:12, Epoch: 20, Step: 120900, Step Loss: 3.71 INFO:Pipeline: Time: 07:56:35, Epoch: 20, Step: 121000, Step Loss: 3.65 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.38 INFO:Pipeline: Time: 07:57:00, Epoch: 20, Step: 121100, Step Loss: 3.63 INFO:Pipeline: Time: 07:57:23, Epoch: 20, Step: 121200, Step Loss: 3.62 INFO:Pipeline: Time: 07:57:47, Epoch: 20, Step: 121300, Step Loss: 3.72 INFO:Pipeline: Time: 07:58:10, Epoch: 20, Step: 121400, Step Loss: 3.70 INFO:Pipeline: Time: 07:58:33, Epoch: 20, Step: 121500, Step Loss: 3.65 INFO:Pipeline: Time: 07:58:56, Epoch: 20, Step: 121600, Step Loss: 3.71 INFO:Pipeline: Time: 07:59:20, Epoch: 20, Step: 121700, Step Loss: 3.66 INFO:Pipeline: Time: 07:59:43, Epoch: 20, Step: 121800, Step Loss: 3.70 INFO:Pipeline: Time: 08:00:06, Epoch: 20, Step: 121900, Step Loss: 3.68 INFO:Pipeline: Time: 08:00:29, Epoch: 20, Step: 122000, Step Loss: 3.68 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.38 INFO:Pipeline: Eval Loss: 3.59, Perplexity: 36.38